The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)

Jointly held with

The 5th International Symposium on Conservation and Management of Tropical Lakes

PROCEEDINGS
The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
The 5th International Symposium on Conservation and Management of Tropical Lakes

RCChE-2020

Co-organized by:
Organized by:

Supported by:

February 04-05, 2021

Phnom Penh, Cambodia

INSIGHTS AND CHALLENGES TOWARD

ACHIEVING SDGS

Institute of Technology of Cambodia, Phnom Penh, Cambodia

February 04-05, 2021
 Contents
Chemical Engineering Session ................................................................................................. 1
Environment Session ................................................................................................................ 74
Lake Environment Session ....................................................................................................... 143
Food Technology and Microbiology Session ...........................................................................206
Session 1: Chemical Engineering

The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)

Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Rare Earth Elements and Yittrium (REYs) Precipitation From Solid Waste Coal Fly Ash
Using Oxalic Acid: Study on Response Surface Methodology

Aulia Nur Rahmawati1, Hotden Manurung1,2*, I Made Bendiyasa1,2, Himawan Tri Bayu Murti Petrus1,2,*

1
Departemen Teknik Kimia (Advanced Materials and Sustainable Mineral Processing Research Group), Fakultas Teknik,
Universitas Gadjah Mada. Jalan Grafika No. 2 Daerah Istimewa Yogyakarta 55281, Indonesia
2
Unconventional Geo-resources Research Center, Faculty of Engineering. Universitas Gadjah Mada, Jl. Grafika No. 2,
Yogyakarta, Indonesia
*bayupetrus@ugm.ac.id
Abstract

Rare earth elements and Yttrium called as REY become an important issue nowadays. REY is listed as a essential material in
many industries such as catalyst, batteries, metal alloy, ceramics, electronic devices, magnet, glass and military defence system.
In last decade, the supply and demand of REY is in disequilibrium due to the increasing demand annually. Currently, China
is the largest producer of REY in the world with control over 95% from total world’s production. Regarding to the Chinese
goverment regulation about export restriction, including REY have created constrains in REY supply in many countries
including Indonesia. Due to this condition, many researchers from various countries have been looking for the alternative
source of REE such as red mud, zircon sand, spent catalyst, spent battries, coal and coal fly ash. Coal fly ash is listed as one
of promising alternative sources of REY. In this presents study, coal fly ash as a raw material. REY precipitation is the last
step for REY extraction from coal fly ash after conducting many steps such as physical separation (enrichment), alkaline
leaching and acid leaching. In order to optimize the REY precipitation, the varible study were as temperature, pH levels and
volume-to-volume ratio. Currently, several nano particle have been succesfully created after multilevel precipitation but still
waiting for analysis using EDX and ICP-MS to indentify the REY concentration on the nano particle.

Keywords: acid leaching, coal fly ash, precipitation, rare earth elements

I. Introduction Plant (PLTU) industry.

In Indonesia, coal is still one of the main energy source According to Seredin and Finkelman (2008), coal
to supply the ever-increasing demand for electricity. Based combustion products (such as fly ash) are one of the sources
on data from the Ministry of Energy and Mineral Resources, of Rare earth elements and Yttrium (REY). REY is
almost 50% of the Steam Power Plant (PLTU) in Indonesia becoming an important issue and listed as an essential
uses coal as its main fuel. This has resulted in the amount of material in many industries such as catalyst, batteries, metal
coal demand continue to increase every year until it reached alloy, ceramics, electronic devices, magnet, glass and
almost 50 million tons in 2016. In addition to its use as fuel, military defense system. Rare earth elements and Yttrium
coal also produces solid waste in the form of fly ash. The (REY) is an element whose demand continues to increase
production of fly ash continues to rise every year due to vast due to its use in various involvement in industrial sectors.
consumption of coal as a raw material in the Steam Power REY contained in fly ash have economic value besides being

1
used in the construction sector. However, currently there are and the filtrate was dried and then analyzed with EDX. The
not many industrial-scale technologies that can take final stage is the precipitation of rare earth metals from the
advantage of REY in flying ash. Therefore, It is necessary to leaching solution pH 5 using oxalic acid (H2C2O4) with
do more in-depth research on the extraction of REY from fly several variations, namely process temperature,
ash waste to separate REY from fly ash waste using concentration of oxalic acid and the ratio of the volume of
precipitation method. the leaching solution to oxalic acid. REY sediment was then
This study aims to obtain optimal conditions in the dried at 100 ºC for 3 hours and calcination at 1000 ºC for 4
separation of REY so that the highest levels of REY can be hours.
obtained from fly ash waste. This is the background of this
research, so that in addition to being used as a source of REY 2.3. Analytical methods
collection, it is also a step in reducing environmental Fly ash from coal combustion solid waste, was analyzed
pollution. using ICP-MS. The resulting solids were then analyzed by
EDX to determine the percentage of REY that had been
II. Materials and Methods precipitated. The results of precipitation were then analyzed
using the Response Surface Method (RSM). RSM functions
2.1. Sampling sites
is to develop, improve, and optimize the process of
Raw material for fly ash was obtained from PT.
determining the optimum formulation for REY precipitation
Pembangkit Listrik Tanjung Awar-Awar Tuban Indonesia
[3].
Corporation. This raw material is a type of sub-bituminous
coal originating from Kalimantan.
III. Results and Discussion
3.1. Fly Ash Raw Material Elements
2.2. Experimental set-up
Fly ash, which is the raw material for obtaining REY,
Coal fly ash originating from the PLTU Tanjung Awar-
was sieved first to smooth the granules until it passes 400
Awar Tuban were pulverized to a size of 400 mesh and
mesh. The Resulted Granules was then analyzed with ICP-
analyzed using ICP-MS and XRD. Furthermore, the
MS. The results of the analysis of REY content with ICP can
resulting components of this process become non-magnetic
be seen in Table 1.
components. This component is obtained through separation
Tabel 1. Elemental compositions of industrial coal fly ash.
process using a wet magnetic separation device with a strong
(Concentrasions of major oxide given in %, REY in ppm).
current of 2A. Magnetic wet separation aims to separate
Element Tuban
magnetic and non-magnetic fly ash using tools from the
Mineral and Coal Technology Research and Development SiO2 50.9
Center, Bandung. Next step is digest process the non- Al2O3 28.7
magnetic fly ash using 8 M concentration of NaOH, Fe2O3 9.0
CaO 3.3
temperature 90 ºC. for 120 minutes.
MgO 2.2
The mixture was then filtered and washed using distilled Na2O 0.8
water as much as 60 times the weight of the dry sample. K2O 0.9
Furthermore, the washing results were filtered and dried in TiO2 1.0
the oven for 4 hours. The resulting fly ash was then leached La 42.3
with 3M concentration of HCl with a solid / liquid (S / L) Ce 84.9
Pr 9.7
ratio of 1:10, temperature 90 º C for 4 hours. The leaching
Nd 37.4
results were then filtered to separate the leaching solution Sm 7.7
from the filtrate. The next step is the addition of CaCO3 until Eu 1.7
the pH of the mixture reaches 3.5 ± 0.2 with 2 hours stirring Gd 7.7
so that the mixture is stable, then filtered as a pH 3.5 Tb 1.3
leaching solution and the filtrate is dried. The pH 3.5 Dy 7.8
leaching solution is added with Ca (OH) 2 until the pH of Ho 1.8
Er 5.1
the mixture reaches 5 ± 0.2 with 2 hours stirring so that the
Tm 0.8
mixture is stable, then filtered as a pH 5 leaching solution

2
 Yb 5.0 3.4. Precipitation Results
Lu 0.8
In this deposition process, the purification process is first
Y 46.4
carried out by increasing the pH of the leaching solution to
Source: Rosita et. all [2] 5 ± 0.2 with the addition of 10% b/b 𝐶𝑎𝐶𝑂3 and
Based on the results of the characterization, a Coutlock 𝐶𝑎(𝑂𝐻)3 . During this process the mixture is stirred to keep
[4] calculation can be performed to determine the feasibility the mixture suspended for 2 hours and then stored for 24
of REY extraction from coal fly ash at PLTU Paiton Tanjung hours. The mixture was then filtered to separate the leaching
Awar-Awar. The REY concentration value is shown in solution and the residue was dried, calcined and analyzed
Equation 1. using EDX. Purification purposes using 𝐶𝑎𝐶𝑂3 and
((𝑁𝑑+𝐸𝑢+𝑇𝑏+𝐷𝑦+𝐸𝑟+𝑌)/𝛴𝑅𝐸𝑌)
𝐶𝑜𝑢𝑡𝑙𝑜𝑜𝑘 = (P1) 𝐶𝑎(𝑂𝐻)3 was to be able to remove or reduce the levels of
((𝐶𝑒+𝐻𝑝+𝑇𝑚+𝑌𝑏+𝐿𝑢)/𝛴𝑅𝐸𝑌)
((37.4 + 1.7 + 1.3 + 7.8 + 5.1 + 46.4)/𝛴𝑅𝐸𝑌) impurities in the REY solution such as Fe, P, Al and Th in
= the liquid 𝑅𝐸𝐸2 (𝐶𝑙)3(𝑙𝑖𝑞) (Silva et.all, 2018).
((84.9 + 0 + 0.8 + 5.0 + 0.8)/𝛴𝑅𝐸𝑌)
= 1.091(> 0.7) The precipitation process is carried out using 𝐶2 𝐻2 𝑂4
From the result of Equation 1, the Coutlook value is with variations in process temperature, variations in the
1,091 (> 0.7) so it can be concluded that the coal fly ash from concentration of 𝐶2 𝐻2 𝑂4 and and variations in the volume
PLTU Tanjung Awar-Awar Tuban is suitable for extraction ratio (v / v) of fluid 𝑅𝐸𝐸2 (𝐶𝑙)3(𝑙𝑖𝑞) : 𝐶2 𝐻2 𝑂4 . The variation
as a source of REY. of process temperature used is 25°C, 60 °C and 80 °C. The
variation in the concentration of 𝐶2 𝐻2 𝑂4 was 5% and 10% .
3.2. Hasil Pemisahan Wet Magnetik For variations in the volume ratio (v/v) 𝑅𝐸𝐸2 (𝐶𝑙)3(𝑙𝑖𝑞) :
Furthermore, fly ash was separated by a wet magnetic 𝐶2 𝐻2 𝑂4 was 2:1, 1:1 and 1:2. REE solution with a volume
process, to obtain non-magnetic (NM) fly ash, with the according to the variation then adjusted the temperature
results in accordance with Figure 1. according to the variation of the 𝐶2 𝐻2 𝑂4 slowly
(2,5mL/minute) to prevent the formation of poor quality
crystals. The EDX analysis results of precipitation with
variations in temperature treatment can be seen in Table 2.

Table 2. Elemental compositions of industrial coal fly ash

after precipitation (Oxide and REO given in %).

Figure 1. REY concentrations in raw materials, magnetic

phase and non-magnetic phase.

Based on Figure 4.2, it can be seen that after separation

with a magnetic separator, the REY elements experience
enrichment in the non-magnetic phase.

3.3. Leaching Results

Based on Table 2, it can be seen that the highest REY
REY leaching of coal fly ash was carried out with 3 M
deposition efficiency level is in the conditions of a volume
HCl under operating conditions of 90 ° C for 240 minutes or
ratio of 1:1, with 10% concentration of oxalic acid and an
4 hours. According to Wang dkK [6] this condition is the
operating temperature of 25° C. amounting to 70.13%. There
optimum condition for REY leaching. During the leaching
are several operating conditions where REY deposition does
process, the stirring was carried out continuously for 4 hours.
not occur, such as in the 2:1 volume ratio and 5% oxalic acid

3
concentration. This may occur due to excess oxalic acid Department (Advanced Materials and Sustainable Mineral
which can cause the deposition of calcium and aluminum Processing Research Group), Faculty of Engineering,
oxalate [7]. Gadjah Mada University and Unconventional Geo-
3.4. Analysis of Response Surface Methode Resources Research Center, Faculty of Engineering. Gadjah
From the P-Value of Figure 3. The three variables do not Mada University.
have a significant effect on the percent yield of REO
precipitation (P> 0.5). References

[1] Seredin, V. V and Finkelman, R. B., 2018. International

Journal of Coal Geology Metalliferous coals : A review
of the main genetic and geochemical types, International
Journal of Coal Geology,76(4), pp. 253–289.
https://doi.org/10.1016/j.coal.2008.07.016.
[2] Rosita, Widya., Bendiyasa, I Made., Perdana, Indra.,
Anggara, Ferian., 2020. Sequential particle-size and
magnetic separation for enrichmen of rare-earth
elements and yttrium in Indonesia coal fly ash. Journal
of Environmental Chemical Engineering, 2213-3437
https://doi.org/ 10.1016/j.jecel.2019.103575.
Figure 3. Analysis of Variance [3] Trihaditia, Riza., Syamsiah, Melissa., Awaliyah, Aliya.,
Figure 4 shows the contour plot where each color 2018. Penentuan formulasi optimum pembuatan cookies
shows the range of responses generated for each REY dari bekatul padi pandanwangi dengan penambahan
element. The maximum conditions for the plots are dark tepung terigu menggunakan metode RSM (Response
green with a value for REO of 50 %. Surface Method), Agroscience Vol 8 No. 2 Tahun 2018,
1979-4661.
[4] Seredin, V V (2010): A New Method for Primary
Evaluation of the Outlook for Rare Earth Element Ores,
Geology of Ore Deposits, 52(5), 5-6.
https://doi.org/10.1134/S1075701510050077
[5] Blissett, R. S., Samlley, N., and Rowson, N.A. (2014):
An investigation into six coal fly ashes from United
Kingdom and Poland to evaluate rare earth element
content, FUEL, 119, 236-239.
https://doi.org/10.1016/j.fuel.2013.11.053
Figure 4. Contout plot of % REO vs Oxalic acid (%), [6] Wang, J. et al., 2017. Recovery of rare earths and
Temperature aluminum from FCC waste slag by acid leaching and
selective precipitation, Journal of Rare Earths. Elsevier
IV. Conclusion Ltd, 35(11), pp. 1141–1148. https://doi.org/
10.1016/j.jre.2017.05.011.
From the results of precipitation using EDX, the largest [7] Silva, R., Morais, C., Teixeira, L., & Oliveira, E.
percentage yield of precipitation is 70.13% REO. It can be (2018). Selective removal of impurities from rare
seen that the impurity content is still high, so it must be earth sulfuric liquor using different reagents. Miner
separated first so that the resulting REO has a higher level Engineering, 238-246.
of purity.

Acknowledgement
We are thankful to the Chemical Engineering

4
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Adsorption Characteristics of Methylene Blue from Aqueous Solution by Amorphous

Silica

Naing Min Tun1*, Wai Hnin Phyu Phyu2, Aung Than Htwe3

1,3
Department of Chemistry, University of Yangon,
University Avenue Road., P.O. 11041, Yangon, Myanmar
2
Department of Chemistry, Monywa University,
Kyaukka Road., Monywa, Myanmar
*nainglay20@gmail.com

Abstract

This research was conducted to find the removal potential of amorphous silica from the source of agricultural by-product
rice husk. Rice husk abundant agricultural by-product in Myanmar, it was used to produce amorphous silica in water rinsed
and different acid-treated (HCl, H2SO4, H3PO4) by the sol-gel method. The as-prepared silica was characterized by energy
dispersive X-ray fluorescence (EDXRF) analysis, X-ray differential (XRD) analysis, and Fourier transform infrared (FT IR)
spectroscopy analysis. From the obtained results, all kinds of samples composed of amorphous silica and the silica content
above 99 % by quantitatively. Continuously, the as-prepared silica was used as an adsorbent for the removal of methylene
blue (MB) from aqueous solution. In the experimental investigations were carried out the effect of contact time, the effect of
adsorbent dosage, and the initial dye concentration. The equilibrium data were fitted to the linear form of Langmuir and
Freundlich isotherm. Langmuir isotherm model was the best R2 value than the Freundlich isotherm. The maximum monolayer
adsorption capacity of MB on silica was found in 17.5438 mg/g for Si-4 with the correlation coefficients R2 value of 0.9987.

Keywords: Adsorbent, amorphous silica, Langmuir isotherm, Freundlich isotherm, rice husk

I. Introduction reaction systems [5]. The reaction variables have studied

depend on the time, temperature, and composition of the
The high silica content of rice husk has attracted
reaction mixture. Furthermore, the rice husk has been used
increasing uses as a source of commercial production of
as a source of microbial nutrients for single-cell protein
silica. Recently, many researchers have reported the
production [6], as a raw material for the production of
production of silica from rice husk by various methods,
ethanol and reducing sugar [7].
optimization by the synthesis of highly purified silica in the
The removal of methylene blue from wastewater was
amorphous form [1], amorphous form of spherical silica
investigated by the potential use of pretreated rice husk and
nanoparticles by different acids treatment [2], the production
rice husk ash [8]. Adsorption is one of the most effective
of amorphous silica by hydrochloric acid leaching [3], and
techniques to remove heavy metals, organic and inorganic
high surface area nanosilica extraction with hydrochloric
pollutants, and non-biodegradable pollutants (including
treatment [4]. Sodium silicate produced by reacting the rice
dyes) from wastewater due to the simplicity as well as the
husk ash and aqueous sodium hydroxide in open and closed
availability to the choice of a wide range of adsorbents. The

5
purpose of this study is to remove the methylene blue from oxide form.
aqueous solution using a high purity of amorphous silica
from rice husk as agricultural waste. 3.1 XRD Analysis
Fig. 1 shows the X-ray diffraction profiles of SiO2
II. Materials and Methods particles from rice husk Si-1 untreated, and different acid-
treated samples Si-2, Si-3, and Si-4. As the X-ray
2.1. Preparation of silica particles diffractograms of all profiles are not having any sharp peaks,
The preparation of silica particles was carried out the following samples were amorphous in nature. In the
according to the described method [1]. In this research, the diffractograms of all samples, a broad single peak appeared
procedure was modified the sonication of sodium silicate by at the range of 20-23º (2θ), it was indicated that the presence
addition of 5 mL of ethanol after refluxing with HCl. The as- of only amorphous silica by [1, 9]. According to Fig. 1, all
prepared samples of SiO2 were nominated for untreated or samples are observed the same amorphous nature pattern as
distilled water rinsed (Si-1), with hydrochloric acid-treated each other.
(Si-2), for sulphuric acid-treated (Si-3), and for phosphoric
25000
acid-treated sample (Si-4). The extraction process and
Si-1
purification of the silica occurred according to equations 1 20000

Intensity, cps
Si-2
and 2, respectively [1]. 15000
Si-3
SiO2(Ash) + 2NaOH → Na2SiO3 + H2O (Eq.1) 10000 Si-4
Na2SiO3 + H2SO4 → SiO2 + Na2SO4 + H2O (Eq.2) 5000
2.2. Experimental 0

10
14
18
21
25
29
33
36
40
44
48
51
55
59
63
66
2.2.1 Characterization of silica particles
The as-prepared silica powder was used for XRD 2θ (deg)
analysis (Rigaku X-ray Diffractometer). The elemental Fig 1. The XRD pattern of silica (Si-1, Si-2, Si-3, and Si-4)
composition of the obtained silica samples was determined
by (EDXRF) (Shimadzu EDX 7000, EDXRF Spectrometer), 3.2 FT IR analysis
and The binding of functional groups of silica was identified
by FT IR analysis. The FT IR analysis of silica particle samples was carried
out in the wave number 400-4000 cm-1. Fig. 2 shows the FT
2.2.2 Removal of methylene blue (MB) by using
IR profiles untreated (Si-1) and acid-treated (Si-2, Si-3, and
amorphous silica
Si-4) silica particles. Normally, the FT IR spectra of all silica
About 0.03 g of silica was mixed with 60 mL of
samples detected the typical bands of O-Si-O stretching at
methylene blue with a known concentration. The mixture
1,065-1075 cm-1 and 797-799 cm-1 and the bending
was shaken with a linear shaker at 300 rpm at a specific time.
vibrations at 450-454 cm-1 [1].
The mixture was allowed to settle at definite times and was
centrifuged at about 5000 rpm for 10 min. The separated
phase of the aliquot was analyzed by PD-303
spectrophotometer with a wavelength of 665 nm.

III. Results and Discussion

The elemental composition of samples was identified by

EDXRF there was found SiO2 is the most abundant element
in all samples. The silica content in all samples (Si-1, Si-2,
Fig 2. The FTIR profile of silica (Si-1, Si-2, Si-3, and Si-4)
Si-3, and Si-4) was found above 99 %. The minimum
3.3 Study on adsorption isotherm
amount of silica content was found in Si-1 untreated sample.
In addition, the presence of silica the other elements such as
The nature of the adsorption will depend on not only the
Fe, K, Ca, Cr, Cu, Zn, and Mn are also composed of the
physical or chemical characteristics of the adsorbent system

6
but also the taken times and the amount of dosage. Fig. 3 energy term of the Langmuir equation varies as a function of
shows the adsorption capacity of silica samples Si-1, Si-2, the surface coverage [12]. The empirical equation of
Si-3, and Si-4 depend on adsorption times with the same Freundlich isotherm is given as follows [13]:
concentration of MB 10 mg/L at 32 °C. This figure shows q𝑒 = K𝑓 (𝐶𝑒 )1⁄𝑛 (Eq.7)
that the adsorption efficiency of MB by Si-1, Si-2, Si-3, and The Freundlich equation can be linearized in logarithmic
Si-4 gradually increases by the time until 7 h. For the first form described as following:
time, the adsorption capacity was reached at a minimum of 1
log q𝑒 = log K 𝑓 + log 𝐶𝑒 (Eq.8)
𝑛
7 mg/g for Si-2 to a maximum of 16 mg/g for Si-4 at 20 min.
25 Where Ce is the equilibrium concentration of the adsorbate
(mg/L), qe is the amount of adsorbate adsorbed per unit mass
20 of adsorbent (mg/g), Kf and n are the Freundlich isotherm
qe (mg/g)

15 constants.
Si-1 The value of 1/n possible ranging between 0 and 1, if the
10 Si-2
Si-3
value is closer to zero become more favourable the
5 heterogeneous system [14]. The Freundlich isotherm model
Si-4
0 data fitting of Si-1, Si-2, Si-3, and Si-4 are shown in Fig. 5.
0 1 2 3 4 5 6 7 0.5
t (h) Si-1
Fig 3. The variation of adsorption capacity with time
0.4 Si-2
0.3 Si-3
3.4 Adsorption isotherm Si-4
Ce/qe
0.2

Adsorption equilibrium isotherms are important to 0.1

describe the mechanism and interaction of methylene blue 0
onto the adsorbent surface. In this experiment, the Langmuir 0 1 2 3 4 5 6
Ce (mg/L)
and Freundlich isotherms were used for the methylene blue
adsorbed onto silica particles of Si-1, Si-2, Si-3, and Si-4. Fig. 4. Langmuir adsorption isotherm of methylene blue
The amount of removal efficiency (R, %) and the amount of onto silica samples (Si-1, Si-2, Si-3, and Si-4).
qe (mg/g) at adsorption equilibrium was calculated by the 1.4
following equations [10]. 1.2
(𝐶0 − 𝐶𝑒)𝑉 1
q𝑒 = (Eq.3)
𝑚 0.8 Si-1
log qe

(𝐶0 − 𝐶𝑒)
𝑅= × 100 (Eq.4) 0.6 Si-2
𝐶0
The generalized Langmuir isotherm can be represented
0.4 Si-3
0.2 Si-4
by the following equation [11]:
0
𝑚 𝑋 𝐾𝐿 𝐶𝑒
q𝑒 = [ 1+ 𝐾𝐿𝐶𝑒
] (Eq.5) -1 -0.5 0 0.5 1
log Ce
Its linear form can be expressed as:
C𝑒 1 1 Fig. 5. Freundlich adsorption isotherm of methylene blue
=𝑋 + 𝑋 𝐶𝑒 (Eq.6)
q𝑒 𝑚 𝐾𝐿 𝑚 onto silica particles (Si-1, Si-2, Si-3, and Si-4)
Where Ce is the equilibrium concentration (mg L-1) of MB
in the solution after adsorption, Xm is the amount of MB The isotherm parameters and correlation coefficients (R2)
adsorption capacity on the adsorbent (mg g-1) and KL is the were obtained from linear fitting. Tables 3 and 4 show the
equilibrium constant related to the energy of adsorption (L maximum amount of Xm of MB onto sample Si-4 were
mg-1). The Langmuir isotherm model of Si-1, Si-2, Si-3, and conducted about 17.5438 mg g-1 and for Si-2 was found
13.8696 mg g-1. According to the obtained results, the
Si-4 are shown in Fig. 4.
correlation coefficients of Langmuir isotherm are more
On the other hand, the Freundlich isotherm assumes to relevant than Freundlich isotherm.
describe the equilibrium of heterogeneous surfaces and
hence does not assume monolayer capacity. In which the

7
Table 3. Langmuir isotherm constants and correlation Production of silica nanoparticles from rice husk as
coefficients for adsorption of methylene blue onto silica agricultural waste by environmental friendly technique.
Sample Langmuir isotherm parameters Environmental Studies of Persian Gulf 2(1), 56-65.
Xm (mg g-1) KL (L g-1) R2 RL [3] Chakraverty, A., Kaleemullah, S., 1991. Conver-sion of
Si-1 16.2601 3.2190 0.9885 0.9688 rice husk into amorphous silica and combu-stible gas.
Si-2 13.8696 1.8115 0.9971 0.9822 Energy Convers. Mgmt. 32, 565-570.
Si-3 16.6944 1.9966 0.9991 0.9804 [4] Liou, T. -H., Yang C. -C., 2011. Synthesis and surface
Si-4 17.5438 2.4152 0.9987 0.9764 characteristics of nanosilica produced from alkali-
extracted rice husk ash. Materials Science and
Table 4. Freundlich isotherm constants and correlation Engineering 176(7), 521-529.
coefficients for adsorption of methylene blue onto silica [5] Foletto, E. L., Gratieri, E., Oliveira, L. H. de, Jahn, S.
Sample Freundlich isotherm parameters L., 2006. Conversion of rice hull ash into soluble
Kf n R2 sodium silicate. Materials Research 9(3), 335–338.
Si-1 10.6316 3.0312 0.8991 [6] Hussein, A. M., El-Saied, H., Yasin, M. H., 1992.
Si-2 8.0612 3.3749 0.9551 Bioconversion of hemicelluloses of rice hull black
Si-3 9.7836 3.0184 0.9594 liquor into single-cell protein. Journal of Chemical
Si-4 10.8118 3.0075 0.9368 Technology and Biotechnology 53, 147-152.
[7] Singh, A., Das, K., Sharma, D. K., 1984. Produ-ction of
reducing sugars from bagasse and rice husk by acid
IV. Conclusions hydrolysis. Agricultural Wastes 9, 131-145.
The characterization of silica by EDXRF, XRD, and FT [8] Sharma, P., Kaur, R., Baskar, C., Chung, W.-J.,
IR analysis was confirmed the quantitative presence of SiO2 2010. Removal of methylene blue from aqueous waste
above 99 % and the amorphous silica. The adsorption of MB using rice husk and rice husk ash. Desalination 259(1-
onto silica was investigated favorable and represented by 3), 249–257.
Langmuir and Freundlich isotherm. Langmuir isotherm [9] Hassan, A. F., Abdelghny, A. M., Elhadidy H., Youssef.
showed that the better fitting the experimental data than A. M., 2014. Synthesis and character-rization of high
Freundlich isotherm. The linear correlation coefficients R2 surface area nanosilica from rice husk ash by surfactant-
of Langmuir isotherm showed above the range of 0.9885 to free sol-gel method. Journal of Sol-Gel Science and
0.9991experimental data and for the Freundlich isotherm Technology 69, 465-472.
from 0.8991 to 0.9594 varying depending on the sample [10] Ge, X., Liu, J., Song, X., Wang, G., Zhang, H., Zhang,
nature. Y., Zhao, H., 2016. Hierarchical iron containing γ-MnO
Acknowledgments 2 hollow microspheres: A facile one-step synthesis and
I would like to thanks Dr Ni Ni Than, Professor (Head), effective removal of As(III) via oxidation and
Department of Chemistry, University of Yangon, for her adsorption. Chemical Engineering Journal 301, 139–
suggestion and comments. I would like to also express my 148.
deep gratitude to organizers of RCChE2020 and [11] Langmuir, L., 1918. Adsoption of gases on plane
AUN/SEED-Net for submitting of research paper to present surfaces of glass, mica and platinum. Journal of
at “The 13th AUN/SEED-Net Regional Conference on American Chemical Society 40, 1361–1403.
Chemical Engineering 2020”. [12] Weber, T. W., Chakkravorti, R. K., 1974. Pore and
solid diffusion models for fixed-bed adsorbers, AlChE
References Journal 20(2), 228-238.
[1] Rafiee, E., Shahebrahimi, S., Feyzi, M., Shaterzadeh, [13] Freundlich, H. M. F., 1906. Over the adsorption in
M., 2012. Optimization of synthesis and solution. The Journal of Physical Chemistry 57A, 385–
characterization of nanosilica produced from rice husk 471.
(a common waste material). International Nano Letters [14] Haghseresht, F., Lu, G., 1998. Adsorption
2(1), 1-8. characteristics of phenolic compounds onto coal-reject-
[2] Ghorbani, F., Sanati, A. M., Maleki, M., 2015. derived adsorbents. Energy Fuels 12, 1100–1107.

8
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Removal of Cr (VI) from Wastewater Using Chitosan-Activated Carbon Composite as

Adsorbent: Study of the Adsorption Equilibrium

Faridatuz Zuhroh *, Agus Prasetya 2 and Wahyudi Budi Sediawan 2

1
Department of Chemical Engineering, Faculty of Engineering, Gadjah Mada University,
55381 Sleman, Yogyakarta, Indonesia

* profesifarida@gmail.com

Abstract

One of alternative treatment that mostly used to reduce the level of wastewater problem is adsorption. This research focuses
on the finding of the capacity of adsorption which represent maximum removal of Cr from water. The used adsorbent was a
composite of chitosan and activated carbon. Started with 20 grams of activated carbon poured into 1000 ml of chitosan
solution. Medium heating and constant stirring make this solution changed its viscosity and gradually looked like gel. Bead
formation had been obtained by slowly pumped the gel into the solution of NaOH 0,7 M. Finally, the adsorbent needed to be
neutralized and dried at temperature of 50oC. The adsorption experiment used a Chromium solution 20 ppm with variation of
adsorbent mass arranged as 0.5gram, 1 gram, and 1.5 gram at temperature of 298 K, 308 K, and 318 K. The concentration of
Cr in wastewater solution were measured every 10, 20, 40, 60, 120, 180, and 210 minutes. The adsorbent characteristics were
analysed using XRD, BET, SEM, and FTIR for its crystallinity, surface area, surface morphology and functional groups. XRD
graphic showed the adsorbent was an amorphous solid. Currently, the adsorbent characteristics were being analysed. BET
result will informs whether the surface area on its pores were increased or not than its original material. Data of Cr (VI) that
dispersed onto the bead composite could be determined by SEM or EDX analysis. Then, FTIR data analysis will gives
information about its functional group, it will confirm whether chitosan is really coated on the activated carbon’s surface or
not. Generally, by gathering all the experiment data will inform the result of the adsorption process in term of its adsorption
equilibrium.

Keywords: Activated Carbon, Adsorption, Bead Composite, Chitosan and Cr (VI) ion

I. Introduction for the removal of heavy metal ions from industrial effluents.
Among these processes, the most promising efficient
Heavy metals present in the water as contaminants is a
technique has been identified as adsorption with a suitable
problem and a threat to human and ecosystem in various
adsorbent.[1] This is really a serious environmental issue
countries. Many considerable methods and efforts are being
related to wastewater treatment.
made to decrease the concentration of heavy metals in the
Reported that Cr is used on a large scale in many
effluent wastewater to permissible discharge levels set by
different industries as plating, alloying, tanning of animal
pollution control and regulatory authorities of each country.
hides, inhibition of water corrosion, textile dyes and
A wide range of processes such as chemical precipitation,
mordants, pigments, ceramic glazes, refractory bricks, and
ion exchange, membrane filtration, electrolytic method,
pressure-treated lumber.[2] If the effluent wastewater from
coagulation, reverse osmosis and adsorption are being used
these industries did not met the standard authorized by

9
pollution control and regulatory authorities, some pollution 2.2. Preparation of Adsorbent
problem may occurred as the negative impact. The preparation of adsorbent was started with preparing
Actually, chromium has several oxidation states ranging the materials. The raw material used were active carbon and
from Cr(−II) to Cr(+VI). The trivalent and hexavalent states chitosan. Each material needs some pre-treatment in order to
are the most stable, although Cr with valences of I, II, IV and make a good composite.
V have also been shown to exist in a number of Particle size is an important factor to make a good
compounds.[3] composite. The active carbon used in this experiment was
Generally the trivalent Cr is considered not harmful. It is the active carbon with its particle size was 100mesh. And for
needed by human for metabolism, but only in a little amount. the chitosan, it was 80mesh.
But in contrast, the hexavalent Cr is considered the most After the particle size was determined, about 40gram of
toxic forms of chromium, as it presents high oxidizing active carbon need to be soaked in oxalic acid 250ml 0.2M
potential, high solubility, and mobility across the for 4 hours. This step was done for keeping the activation of
membranes in living organisms and in the environment. the carbon. After neutralized with distilled water, the
In recent years, contamination of the environment by activated carbons need to be dried to eliminate the water
Chromium (Cr), especially hexavalent Cr, has become a inside.
major area of concern. Because of that, constant monitoring Same solution for the chitosan, oxalic acid 1000ml 0.2M
and analysis of chromium becomes a necessity. [3,4] was also needed to dissolve 20gram of chitosan. This step
In the aquatic environment, the toxicity of Cr(VI) has was done for making the chitosan viscous gel. The form of
been shown to be greater than Cr(III).[4] Other problem is gel can be obtained by heating and stirring the solution in
that Cr(III) can be oxidized and changed became Cr(VI). the same time.
Cr(III) in the forms of oxides, hydroxides, and sulphates The dried activated carbon need to be poured into the
is less toxic as it is relatively insoluble in water, presents chitosan solution after its viscosity became higher. Constant
lower mobility, and is mainly bound to organic matter in soil stirring still need to be maintained. The mixture was also
and aquatic environments. Moreover, Cr(III) forms tend to heated at 40-50ºC to facilitate mixing. It was important to
form hydroxide precipitates with Fe at typical ground water the homogeneity of the compositing process between
pH values. At high concentrations of oxygen or Mn oxides, activated carbon and chitosan.
Cr(III) can be oxidized to Cr(VI).[5,6] Oxidation of Cr(III) For the best condition, it was used the magnetic stirrer
to Cr(VI) represents a significant environmental hazard tools that had a temperature controller on its side because the
because a relatively nontoxic species is transformed into a heating and stirring were supposed to be held in the same
more toxic one.[2] time.
Chitosan is good adsorbent because of its active group The next step was slowly pumping the gel into some
and widely used in the removal process of heavy metal in solution to gain the bead formation. The needed solution was
wastewater. And activated carbons are unique and versatile NaOH 0,7M. Soon after the pumped gel dropped and
adsorbents because of the availability of extensive surface contacted with the solution, its form was changed into some
area, its micro porous structure and high adsorption spherical bead of solid that moved to the ground of the
capacity.[1] solution. It took some time to let the bead composite
The objective of this study was to demonstrate the formation was stabilized.
adsorption performance where the adsorbent is a bead The beads were still needed to be neutralized with distil-
composite for the removal of Cr (VI) ion in water. The bead led water and then dried at the temperature of 50oC so the
composite was a modified of activated carbon and chitosan. beads were ready to be characterized and used as the
adsorbent in the adsorption process.
II. Materials and Methods
2.3. Experimental set-up
2.1. Materials
Adsorption experiments were carried out in the batch
Active Carbon was purchased from Alfa Kimia
contact method. The used media was a water bath shaker
(Yogyakarta, Indonesia) and for the Chitosan was from
where some erlenmeyers can be arranged in it.
Monodon Group (Lampung, Indonesia).
The adsorption experiment used a Chromium soluti

10
on 20 ppm with the independent variables were adsorb other words, the adsorption rate and the desorption rate are
ent mass, time and temperature. equal.
Variation of adsorbent mass arranged as 0.5gram, 1 Equation of equilibrium in adsorption is usually
gram, and 1.5 gram at temperature of 298 K, 308 K, performed as adsorption isotherm. This equation defines the
and 318 K. The concentration of Cr in wastewater sol connection between the concentration of ion in a liquid
ution were measured every 10, 20, 40, 60, 120, 180, phase with the concentration that adsorbed in the surface of
and 210 minutes. adsorbent. In aqueous system, there are Langmuir Isotherm
and Freundlich Isotherm.
2.4. Analytical methods Langmuir Isotherm is usually used for the adsorption
model which the adsorbate molecule is adsorbed in one layer
X-Ray Diffraction (XRD) analysis (monolayer) in the surface of adsorbent. It means that all the
X-Ray Diffraction analysis were performed in order to adsorbed molecule does not have interaction each other and
define the change of microstructure in the adsorbent material. all the surface area of the adsorbent have the same affinity
XRD result analysis also contain of information on level to the adsorbate. Contrarily, Freundlich Isotherm is
amorphous and crystalline nature of particles. used when the adsorption model was multilayer. It means
that there is a possibility that each surface of the adsorbent
Fourier transforms infrared (FTIR) spectroscopy may have different functional group.
The prepared samples were prepared from the bead
composite and bead composite after adsorption. The FTIR 𝑞𝑚 𝐾𝑙 𝐶𝑒
𝑞𝑒 = (Eq.1)
analysis were performed to compare the functional group. 1 + 𝐾𝑙 𝐶𝑒
So the result will confirm whether chitosan is really
coated on the activated carbon’s surface or not. 𝐶𝑒 1 1
= + 𝐶 (Eq.2)
𝑞𝑚 𝑞𝑚 𝐾𝑙 𝑞𝑚 𝑒
BET analysis
1
The prepared samples were tested and recorded on a ( ) (Eq.3)
𝑞𝑒 = 𝐾𝑓 𝐶𝑒 𝑛
Quantachrome NovaWin - Data Acquisition and Reduction
for NOVA instruments 1994-2013, Quantachrome
1
Instruments version 11.03. 𝐿𝑜𝑔𝑞𝑒 = 𝐿𝑜𝑔𝐾𝑓 + 𝐿𝑜𝑔𝐶𝑒 (Eq.4)
𝑛
Samples for the BET analysis were prepared from the
raw material (chitosan) and the bead composite adsorbent. where 𝑞𝑒 is amount of ion that adsorbed when equilibrium,
BET analysis was performed to compare the surface area 𝑞𝑚 is the maximum adsorption capacity of ion, 𝐾𝑙 is the
and pore volume.
constant of Langmuir Isotherm, 𝐾𝑓 is the constant of
Freundlich Isotherm, 𝐶𝑒 is the concentration of solution
SEM analysis
when equilibrium.
A scanning electron microscope inter-phased with an
The equations above are used to determine the
electron dispersive X-ray spectrometer was used to study the
favorability of the adsorption system. (Eq.1) and (Eq.2) for
surface morphologies and elemental analysis of the bead
the Langmuir isotherm, (Eq.3) and (Eq.4) for the Freundlich
composite before and after adsorption. The result of SEM
isotherm.
analysis will show the data of Cr (VI) that dispersed on-
to the bead composite. The result also can be confirm
I. Results and Discussion
ed with conduct the EDX analysis.
Result of the X-Ray Diffraction (XRD) analysis for the
Adsorption Equilibrium bead composite adsorbent is shown in Table 1. As shown in
Adsorption equilibrium is obtained when the solid Table 1 and imagined in Fig. 1, the strongest peak of
adsorbent has contact with the adsorbate molecule and then adsorbent samples were 2θ = 25.9800°, 27.2443°, and
there is movement from the solution to the solid adsorbent 28.1800°.
make both concentration in solution and solid is equal. In

11
Table 1. Strongest Peak X-Ray Diffraction Analysis surface area. The form of bead composite is actually a stable
form of modification composite, it is easy to mobile in the
No Peak Number 2Theta (deg) solution and easy to be investigated in the process of
1 29 25.9800 removal of Cr(VI) in water. But actually the initial diameter
2 31 27.2443 or size of the spherical bead should be made in bigger size,
because after the drying treatment, the size will decreased as
3 32 28.1800
the water inside the bead is eliminated. The size of spherical
bead composite will influence the amount of adsorbent uses
XRD bead composite
in every adsorption experiment.
600

500 Acknowledgement
We are thankful to the Energy Conservation and
intensity (cps)

400

300
Pollution Prevention Lab (KEPP) and Instrumental Analysis
Lab (ANINS) for their laboratory equipment and analysis
200
support.
100

0
0 10 20 30 40 50 60 70 80 90
References
2 theta (deg)

Journal Papers

Fig. 1. X-Ray Diffraction Graphic Analysis

[1] T. Anirudhan, S. Sreekumari., 2011. Adsorptive
removal of heavy metal ions from industrial effluents
using activated carbon derived from waste coconut
Table 2. BET Analysis buttons. Journal of Environmental Sciences. 1989-
1998.
Ads. Surface Pore Volume Pore Radius [2] A. Zayed, N. Terry., 2003. Chromium in The
Area (𝒄𝒄/𝒈) (𝑨° ) Environment: Factors Affecting Biological
(𝒎𝟐 /𝒈)
Remediation. Journal of Plant and Soil. 139-156.
1 9.491 0.017 26.966 [3] S. Avudainayagam, M. Megharaj, G. Owens, R. S.
2 12.047 0.037 26.895 Kookana, D. Chittleborough, R. Naidu., 2003.
3 12.314 0.038 26.742 Chemistry of chromium in soils with emphasis on
4 18.392 0.040 26.970 tannery waste sites. Reviews of Environmental
Contamination and Toxicology. 53–91.
As shown in Table 2, the BET result informs that there [4] R. Rakhunde, L. Deshpande, H. Juneja., 2012.
is an increasement in the surface area on its pores than its Chemical Speciation of Chromium in Water: A
original material. Review. Journal of Critical Reviews in
Currently, the adsorbent characteristics were still being Environmental Science and Technology. 776-810.
analyzed in lab for the rest analysis, so there is not any data [5] T. Becquer, C. Quantin, M.Sicot et al., 2003.
result of FTIR and SEM analysis yet. Chromium Availibility in Ultramafic Soils from New
Caledonia. Journal of Science of the Total
Environment. 251-261.
IV. Conclusion
[6] J. Peralta-Videa, M.Lopez, M.Narayan et al., 2009.
It is very possible to combine the chitosan adsorbent with The Biochemistry of Environmental Heavy Metal
activated adsorbent to make a composite adsorbent because Uptake by Plants: Implications for The Food Chain.
of each good properties. Chitosan is good in its mechanical Journal of Biochemistry and Cell Biology. 1665-
properties but lack in acid solution, and it is covered with 1677.
activated carbon which is good in its pore structure and

12
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

The Extraction of lignin from Sugarcane Bagasse Using

Urea Based Hydrotropic Process
Ulfia Al Rahma1, Muslikhin Hidayat1, Wahyudi Budi Sediawan1*, Indah Hartati2

1
Chemical Engineering Department, Faculty of Engineering, Gadjah Mada University, jln.Grafika No.2,
Yogyakarta 55281, Indonesia
2
Department of Chemical Engineering, Faculty of Engineering, Universitas Wahid Hasyim, 50236, Indonesia

*wbsediawan@ugm.ac.id

Abstract
Lignin is one of the largest reservoirs of aromatic compound that can be used in various applications while
hydrotropes have been explored as new green chemical for lignin separation. The objective of this study was to recover
lignin from filtrate of sugarcane bagasse microwave-assisted hydrotropic extraction process. The lignin extraction were
performed at a fixed solid and liquid ratio of 1:20, agitation speed of 900 rpm, urea concentration 30%, while temperature
and extraction duration were varied between 70 -90 oC, and 30-150 minutes, respectively. The soluble lignin was
precipitated from the hydrotropic extraction filtrate by adding 5 times of water and bringing the pH of the solution to pH 4-5.
The research showed that the higher the hydrotrope concentration applied, the higher the recovery of lignin would be. The
profile of the lignin recovery of hydrotropic extraction conducted on higher temperatures also showed a similar trend as the
one the application of higher hydrotrope concentration. The optimum condition obtained from the research was at a
temperature of 90oC, 30% of concentration urea hydrotropes, and in the time of 120 minutes, the % lignin extracted 13.45 %.
Keywords: Lignin, hydrotropes, extraction, urea

I. Introduction biomass. Kraft, sulfate, organosolv, soda, and hydrothermal

are common processes applied in biomass pretreatment [5].
Bagasse is a solid residue obtained from sugarcane
Lignin from Kraft, sulfate, and soda process are still
processing in sugar and ethanol industry [1]. Generally,
contains chemicals. Lignin from organosolv and
280 kg of bagasse is produced from the processing of 1 ton
hydrothermal processes are having good quality as it has
of sugarcane [2]. Lignocellulosic materials consist of three
low molecular weight and good solubility in organic
main polymers which are cellulose, lignin, and
solvents. However, both of organosolv and hydrothermal
hemicellulose [3]. The cellulose, hemicellulose, and lignin
processes are expensive. The utilization of harsh chemicals
content of sugarcane bagasse are 41-55%, 20-22.5%, and
and operation condition are the other back draws of the
18-26.3%, respectively [4].
above processes [5].
Lignin, the most abundant natural aromatic polymer on
The application of friendly extraction process including
earth, has been regarded as a potential raw material for
the utilization of green chemicals and application of mild
various higher values of fuels and chemicals [3]. Type of
operation condition will be beneficial. Hydrotropic
biomass sources, extraction methods, and extraction
extraction is considered as an environmentally friendly
severity are some factors affecting the lignin isolation
water-based method with a simple recovery of hydrotropic
efficiency. Lignin extraction process is intended to separate
solutions to obtain several products. Several hydrotropes
and recover both cellulose and lignin from lignocellulose

13
have been applied in lignin and cellulose separation from 42. The precipitated lignin obtained was then heated in an
various biomasses. Devendra and Pandey [6] reported that oven at 100oC until its constant weight. The lignin
sodium cumene sulfonate was able to remove up to 50% of extraction % on solute was calculated based on its initial
lignin from rice straw. Urea is one of potential hydrotrope lignin content of raw material sugarcane bagasse [8].
agent since it is safe and low-price product [6].
In this work we reported the application of urea as the 2.3. Analytical methods
hydrotropic agent in microwave assisted lignin extraction The lignin content was determined by using Chesson
from sugarcane bagasse. method. One gram of dry sample (a) was refluxed for 2
hours with 150 ml H2O at 100oC. The residue was filtered,
II. Materials and Methods washed with 300 mL of hot water and oven-dried to
2.1. Material constant weight (b). The residual powder obtained was then
Bagasse was acquired from sugar factory Geneng, added with 150 mL of 0.5 M H2SO4 and then refluxed for 2
Ngawi regency. Sugarcane bagasse was air dried, milled hours at 100 oC. The residue was filtered and washed with
and then sieved with a sieve size of 70 mesh (210 µm). The 300 ml of water until neutral condition. The residue was
cellulose, hemicellulose and lignin content of sugarcane dried to a constant weight (c). The dry residue was added
bagasse was tabulated on Table 1. with 10 mL of 72% (v/v) H2SO4 and soaked at room
temperature for 4 hours. 150 mL of 0.5 M H 2SO4 was
Table 1. Chemical composition of sugarcane bagasse added and refluxed at 100 oC for 2 hours. The residue was
Component Values (%) filtered and washed with 400 mL of distilled water until
Cellulose 24.317 neutral condition. The residue obtained was weighted (d).
Hemicellulose 45.588 The solid residue was then ashed in a furnace, and the ash
Lignin 18.249 was weighted (e). The lignin content was calculated by:
d e
Lignin (%) = x100% (Eq.1)
Urea was from PT. Kujang Indonesia and sulfuric acid a
in which used to control the pH of the reaction mixture was The lignin solids analysis was carried out using FTIR at
from Merck. room temperature of 298 K in region 4000-400 cm -1 on
Thermo Scientific Nicolet iS10.
2.2. Lignin Extraction Procedure
20 grams of bagasse and 400 ml of 30% urea III. Results and Discussion
hydrotropes solution were put into glass tube. The reaction During urea based hydrotropic extraction of lignin from
mixture was subjected to microwave heating (70-900C) for sugarcane bagasse, the color of the filtrate solution became
30-150 minutes at 900 rpm. After completing the extraction dark (Figure 1a). The darkening of the hydrotropic solution
process, the suspension was filtered and rinsed using 500 during pretreatment of biomass was also reported in alkyl
ml of hot water (90oC) to remove the remaining benzene sulfonates based hydrotropic processing of cotton
hydrotropes attached to the solid residue. After filtration, stalk Karthyani et al., [9]. It was believed that the
the solid residue was dried in an oven at 90oC until darkening solution indicates that biomass components were
constant weight and the lignin-containing filtrate was solubilized into the hydrotropic solution. Karthyani et al.,
treated for further processing. Series of experiments were [9] mentioned that due to hydrophobic interactions of the
carried out in same way to determine the effect of aromatic ring of phenolic lignin with the aromatic ring of
temperature, hydrotropes concentration and time on lignin the hydrotrope, lignin solublization was more preferential
extraction. than cellulose and hemicelluloses [8].
The dissolved lignin contained in the hydrotropic The hydrotropic lignin was precipitated from the
filtrate was recovered by water precipitation. 200 ml of the filtrate by water dilution under acidic condition and the
filtrate was added with distilled water up to 4x the original obtained lignin was brown colored powder (Figure 1b).
volume. The pH of the solution was set to pH 4-5 by Brown colored powder of sugarcane bagasse lignin was
utilized 5 N of sulphuric acid [7]. The precipitated lignin also obtained from alkyl benzene sulfonate based
was then vacuum-filtered by using Whatman filter paper no. hydrotropic treatment. Ansari et al., [10] as mentioned by

14
Jingjing [11] that most isolated lignin are brown powder.
The optimum temperature and time of the hydrotropic
lignin extraction were achieved at 90oC and time of 120
minutes. This increasing of time, the % extracted lignin
also increases at various temperatures. Based on the graph
above, the optimum % extracted lignin at 120 minutes
reached 13.45%, while at 150 minutes there was a decrease,
(a) (b)
where the percentage of lignin extracted was only
Figure 1. Filtrate solution of hydrotropic delignification
10.86%.The decreased lignin extraction yield at 150
and (b) hydrotropic lignin
minutes was due to the inhibition of mass transfer of the
hydrotrope solution which entered the sugarcane baggage
3.1. Effect of Temperature and Time
matrix due to the degradation of urea as hydrotropes.
Lignin extraction from sugarcane bagasse was carried
out by utilized urea solution (30%) in temperature range of
3.1.2. Lignin Characterization
70-90 oC and time range of 30-150 minutes with time
FTIR analysis was performed on lignin powder
interval of 30 minutes. The profile of lignin extraction
obtained from hydrotropic extraction conducted at
percentage was illustrated on Figure 2. It was shown that
temperature of 90 oC and time of 120 minutes.
temperature is significantly affecting the dissolution of
lignin in hydrotropic extraction. The lignin extracted was
increase with the increase of temperature in lignin
extraction from 30 minutes to 120 minutes. The lignin
extraction performed at 120 minutes and at temperature of
70, 80 and 900C gave lignin extraction percentage of 9.70,
12.05 and 13.45%, respectively. Gabov [8] stated that at
higher temperatures, hydrotropic molecules aggregate and
the hydrogen bonds between water molecules become
unstable resultin in excess hydration and in increased Fig.3. FTIR Lignin
surface activity. The rise of temperature can cause a
significant change in the structure of the aggregate, thereby Lignin is a complex aromatic heteropolymer that is
causing more solute to dissolve in the hydrotrope solution. composed of three main monolignols, namely, pcoumaryl,
Increasing temperature can raise the amount of aggregation coniferyl, and synapyl alcohol, with different degrees of
and thus create more hydrotrope aggregates, each of which methoxylation. When incorporated into lignin polymers,
can interact with solute molecules [8]. monolignol produces p-hydroxyphenyl (H), guaiacyl (G),
and syringyl (S), each of which has a structural variation in
the polymer. Based on the literature for soft-stemmed
plants such as bagasse, the most dominant lignin
components are Guacyl and p-hydroxyphenyl minor [12].
Based on the FTIR results presented in Figure 3, it shows
that the wavenumber 1161.35 cm-1 indicates the presence
of a p-hydroxyphenyl group which is one of the
components of lignin.

IV. Conclusion
Urea based hydrotropic solution was proved as a good
hydrotropic agent for lignin extraction. Temperature an
Figure 2. Lignin extraction percentage of hydrotropic time lignin, the higher the temperature of the extracted
extraction of lignin from sugarcane bagasse lignin also increases, the optimum temperature is at 90 oC.
Meanwhile, the longer the time used for extraction, the

15
decreasing of the extraction results, the best optimum time [9] Karthyani, S., Pandey, Ashok., Devendra, P Leena.,
to obtain lignin extraction is 120 minutes. 2017. Delignification of Cotton Stalk Using Sodium
Cumene Sulfonat for Bioethanol Production.
Acknowledgement Biofuels, 1759-7269.
The authors are grateful for Author greatly [10] Ansari, Khursheed B., Gaikar, Vilas G. 2013., Green
acknowledge The ministry of Research, Technology and hydrotropic Extraction Technology for
Higher Education of Republic of Indonesia which support Delignification of Sugarcane Bagasse by Using
Alkybenzene Sulfonates As Hydrotropes. Chemical
this work through PDUPT research grant of 2020, with
Engineering Science, 1-10.
contract number of 2845/UN1/DITLIT/DIT-LIT/2020.
[11] Jingjing, Li., 2011. Isolation Lignin From Wood.
Saima University of Applied Science, Imatra.
References
[12] Van Maris, Antonius., Abbott, A Derek., Bellismi,
[1] Chandel, Anuj K.da Silva., Silvio S. Carvalho.,
Eleonara., deen Brink, Van Joost., Kuyper, Marko., H
Walter Singh., Om V., 2012. Sugarcane Bagasse and
Luttik, A Marijke., Wisselink, Woufer H., Scheffers,
leaves: Foreseeable Biomass of Biofuel and Bio-
Alexander W., Van Dijken, Johanes P., 2006.
Products. Journal of Chemical Technology and
Alcoholic Fermentation of Carbon Source in
Biotechnology 87, 11-20.
Biomasss Hydrolysates by Saccaromyces
[2] Spiridon, Iuliana. 2018., Biological and
Cerevisiae, Current Status. Springer Science 90,
Pharmaceutical Applications of Lignin and Its
391-418.
Derivatives: a Mini-Review. CELLULOSE
CHEMISTRY AND TECHNOLOGY Cellulose Chem.
Technol 52, 543-550.
[3] Yunpu, Wang, Leilei., D A I, Liangliang., F A
NShaoqi., Shan Yuhuan, L I U., 2016. Journal of
Analytical and Applied Pyrolysis Review of
Microwave-Assisted lignin Conversion for
Renewable Fuels and Chemicals. Journal of
Analytical and Applied Pyrolysis 119, 104-103.
[4] Mokhena, C Teboho., Mochane, J Mokgotsa.,
Moutang, E Tshwafo., Linganiso, Z Linda., Thekisoe,
M Oriel., Songca, P Sandile., 2018. Sugarcane
Bagasse and Cellulose Polymer Composites.
Intechopen, 71497.
[5] Chung, Hoyong.,Washburn, Newell, R., 2016.
Extraction and Types of Lignin. Lignin in Polymer
Composites, 13-25.
[6] Devendra, Leena P., Pandey, Ashok., 2015.
Hydrotropic pretreatment on rice straw for
bioethanol production. Renewable Energy 98, 2-8.
[7] Watkins, Dereca., Nurudin, Md., Hosur, Mahesh.,
Tcherbi-Narteh, Alfred., Jaelani, Saek., 2014.,
Exctraction and Characterization of Lignin from
different biomass Resources. Journal of Materials
Research and Technology, 4(1): 26-32.
[8] Gabov, Konstantin. 2018., Hydrotropic Process for
Green Biorefinery Applications applications.
Finlad, United state.

16
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Bleaching of Rice Straw Hydrotropic Pulp with Hydrogen Peroxide

Rara Ayu Lestary 1, Moh. Fahrurrozi 1, Wahyudi Budi Sediawan1* and Indah Hartati1, 2
1
Departement of Chemical Engineering, Faculty of Engineering, Universitas Gadjah Mada,
Jl. Grafika No. 2, Yogyakarta 55281, Indonesia
2
Departement of Chemical Engineering, Faculty of Engineering, Universitas Wahid Hasyim,
Jl. Menoreh Tengah X No 22 Semarang, Indonesia
* wbsediawan@ugm.ac.id

Abstract
One of the potential lignocellulose resources from agriculture biomass residue is rice straw. Cellulose, the main
component of lignocellulose can be processes into various products. Hydrotropic treatment and peroxide bleaching
can be considered as an environmentally friendly pretreatment for lignocellulose. Bleaching plays an important role
in removing both hemicellulose and lignin resulting in producing high cellulose purity and brightness. This research
aimed to study the rice straw hydrotropic pulp bleaching with hydrogen peroxide. Hydrotropic treatment of rice straw
was conducted with 30% w/v of urea solution at 80oC for 60 minutes and the chelation stage was conducted with
EDTA at 70oC for 60 minutes. Bleaching of rice straw hydrotropic pulp was investigated under various time and
concentration of hydrogen peroxide (2%, 4%, and 6%) at 90oC with microwave heating. Lignocellulose content,
chemical composition, brightness, and yield of bleached pulp were evaluated. The results showed that the maximum
value of bleached pulp brightness was obtained with 4% hydrogen peroxide concentration at temperature 90 oC for
120 minute reaction time with the cellulose content was 52.41% and the lignin content was 14.45%.

Keywords: Bleaching, Hydrotropic Pulp, Hydrogen Peroxide, Rice Straw

I. Introduction Hydrotropic treatment has several attractive

features that make it a remarkable alternative for
Rice straw, one of the potential lignocellulose biomass fractionation. Hydrogen peroxide is
resources from agriculture biomass residue, is usually considered as an environmentally friendly bleaching
being disposed of by burning it in an open field, agent because it decomposes into oxygen and water
plowing it into the field, or using it as animal feed [2]. only. The effectiveness of hydrogen peroxide as a
One of the main components in rice straw cell walls is bleaching agent is greatly influenced by the presence
cellulose, a versatile polymer, in which utilized for of metal ions in pulp. Chelation stage prior the
fibre, films, membranes, excipient, and composites bleaching process is useful in removing pulp metal
[3]. ions and in increasing the bleaching efficiency [5].
Cellulose and its derivatives can be obtained from Bleaching plays an important role to remove
lignocellulose treatment, fractionation and conversion remaining hemicellulose and lignin, so cellulose is
processes. Pretreatment of lignocellulose materials is expected to have high cellulose purity and high
an essential step for hemicelluloses and lignin removal brightness. In the present work, rice straw hydrotropic
because cellulose is embedded in hemicelluloses and pulp has been used to investigate the effect of
lignin matrix [4]. Hydrotropic treatment followed by concentration and time of bleaching on brightness,
peroxide bleaching are considered as an yield, and chemical compositional of unbleached pulp.
environmentally friendly lignocellulose pretreatment.

17
 hours to quantify the cellulose content (d). The sample
II. Materials and Methods was burned in a furnace and lignin content was
measured as the difference between the sample weight
2.1. Materials before and the sample weight after burning (e). The
Rice straw was obtained from a local rice field in percentage of hemicellulose, cellulose, and lignin can
Blitar, East Java. Urea was produced and supplied by then be calculated by Eq. 1, 2, and 3.
PT Pupuk Kujang, Hydrogen Peroxide H2O2 30% b-c
Solution (Merck), Sodium Hydroxide NaOH (Merck), Hemicellulose % = (Eq.1)
a
Sulfuric Acid H2SO4 95-97% (Merck), EDTA c-d
cellulose % = (Eq.2)
(Merck), and distilled water. a
d-e
Lignin % = (Eq.3)
2.2. Experimental set-up a
2.2.1 Hydrotropic Treatment and Chelation Stage
Rice straw was chopped, grounded into powder 2.2.4 Chemical analysis
and 60 mesh sieved. Rice straw powder was inserted Residual H2O2 was determined by iodometric
in glass jar followed by the addition of urea solution method. The filtrate sample from the bleaching stage
30% w/v. Hydrotropic treatment was conducted in a was inserted into Erlenmeyer flask and added with 50
glass jar with microwave heating at 80oC for 60 mL distilled water, 3 drops of ammonium molybdate
minutes and agitating speed was set at 300 rpm. The solution, 4 mL of 4N H2SO4 solution, and 10 mL of
pulp was separated by filtration process and oven- 10% KI solution. The mixture was titrated with 0.1 N
dried at 105oC. Na2S2O3 solution until the brown tri iodide color has
The rice straw hydrotropic pulp was then chelated been reduced to a light straw color then a few drops of
with EDTA 2% w/w at adjusted pH 4-5, temperature the starch solution was added to the mixture.
of 70oC for 60 minutes. The chelation residue was Subsequently, the titration was performed until the
separated by a filtration process then it was oven-dried color of the solution changes to colorless. The
at 105oC. concentration of H2O2 in the sample can then be
calculated by equation 4, where A is the titration
2.2.2 Bleaching stage volume for sample, B is titration volume for blank, and
The bleaching process was performed at various N is normality of Na2S2O3.
reaction times and H2O2 concentration with 1.5%
NaOH in a glass jar with microwave heating. The (A-B)(N)(1.7007)
H2 O2 % w/w= (Eq.4)
microwave temperature was set to 90oC and the Sample weight
agitation speed was set at 300 rpm. The bleached pulp
was separated, washed with distilled water and oven 2.2.5 Colour Measurement
dried at 105oC. Colour measurement of sample was conducted
with Chroma meter Minolta. The L, a, and b parameter
2.2.3 Compositional analysis value were measured. The brightness can then be
Cellulose and lignin content in unbleached and calculated by Eq. 5.
bleached pulp were determined by Chesson-Datta
Analysis [6]. 1 gr of sample (a) and 150 ml distilled Brightness % = 100-[(100-L)2+(a2+b2)]0.5 (Eq.5)
water were refluxed at temperature of 100oC for 2
hours. The sample was then oven-dried at 105oC (b). III. Results and Discussion
The dried sample and 150 mL of 1 N H2SO4 solution
Rice straw hydrotropic pulp obtained by
were refluxed at temperature of 100oC for 2 hours to
hydrotropic pretreatment and chelation stage. The
quantify the hemicellulose content. The sample was
cellulose, hemicellulose, and lignin composition of
then oven-dried at 105oC (c). Furthermore, the sample
rice straw before hydrotropic pretreatment was
was immersed in 10 mL of 72% H2SO4 solution at
31.14%, 28.90%, and 21.10% respectively. The
room temperature then it was refluxed using 150 mL
chemical composition of rice straw after the chelation
of 1 N H2SO4 solution at temperature of 100oC for 2

18
stage was 37.42%, 25.80%, 18.06% respectively. The hydrogen peroxide concentration for 120 minute
hydrotropic pretreatment and chelation stage were reaction time. The brightness of the unbleached pulp
able to increase the cellulose content of rice straw by was 54,23%, while the brightness of the bleached pulp
up to 6.28% from the initial composition. obtained from bleaching process with 4% hydrogen
Rice straw hydrotropic pulp was subjected to peroxide concentration for 40, 80, and 120 minutes
peroxide bleaching. The effect of increased bleaching were 76.29%, 77.28%, and 80.78% respectively.
reaction time and concentration of hydrogen peroxide
on chemical composition in rice straw is shown in 20.00
Fig.1 and Fig.2. Fig.1 shows that the application of
18.00
higher concentration of hydrogen peroxide results in

Lignin %
the increase of the cellulose content, in which the 16.00
highest cellulose content was found in 6% hydrogen
14.00 40 min
peroxide concentration (53.33%). The partial
elimination of hemicellulose and lignin causes 12.00 80 min
120 min
cellulose content to increase. Bleaching with hydrogen 10.00
peroxide under alkaline conditions will produce 0 2 4 6 8
perhydroxyl anion, which is believed to be the main H2O2 Concentration, %
active species in the elimination of chromophore in
Fig 2. Lignin % in Rice Straw Hydrotropic Pulp
lignin, especially the conjugated carbonyl structures
that react easily with perhydroxyl anion. This reaction 85.00
will cause the lignin dissolution so that the amount of 80.00
Brightness %
lignin in the pulp decrease [5]. This phenomenon is
characterized by decreasing lignin content with 75.00
increasing reaction time. Similar results were obtained 70.00
for cellulose extraction with alkaline peroxide 40 min
treatment by Fitriana et al (2020). The higher 65.00 80 min
120 min
hydrogen peroxide concentration used, the higher 60.00
cellulose content obtained [7]. 2 0 4 6 8
H2O2 Concentration, %
Fig 3. Brightness % Rice Straw Hydrotropic Pulp

Table 1. Yield and H2O2 Consumption

H2O2
H2O2 Time Yield
No. Consumption
(%) (min) (%)
(%)
1 2 40 61.66 73.13
2 2 80 57.79 88.77
3 2 120 56.15 87.20
4 4 40 60.04 70.16
5 4 80 58.54 79.63
6 4 120 58.47 90.25
Fig 1. Cellulose % in Rice Straw Hydrotropic Pulp
7 6 40 65.88 84.21
8 6 80 60.73 88.02
The hydrotropic treatments, chelation, and
9 6 120 59.10 92.57
bleaching stage not only impact on the changes of the
The brightness of the rice straw hydrotropic pulp
chemical composition of rice straw but also affect the
at various times of the bleaching process increased as
optical properties. The effects of the increased reaction
the hydrogen peroxide concentration increased from
time and concentration of hydrogen peroxide on the
2% to 4%. The increase in brightness of rice straw
brightness of the unbleached pulp are shown in Fig.3.
hydrotropic pulp is caused by the high formation of
Pulp brightness reached the maximum value at 4%

19
perhydroxyl anion, which causes lignin to dissolved
[8]. Similar results were obtained for alkaline peroxide References
bleaching by Tutus et al (2014), the brightness of the
pulp increases with increasing hydrogen peroxide [1] Abraham, A., Mathew, A.K., Sindhu, R.,
concentration [9]. Pandey, A. and Binod, P., 2016. Potential of
rice straw for bio-refining: An overview.
The brightness of the rice straw pulp at 6% Bioresource Technology, 215, pp.29-36.,
hydrogen peroxide concentration is lower than 4%. It [2] Matsumura, Y., Minowa, T. and Yamamoto,
occurs caused by the use of excess hydrogen peroxide H., 2005. Amount, availability, and potential
so that some hydrogen peroxide does not react with ion use of rice straw (agricultural residue) biomass
hydroxide to produce perhydroxyl anion but as an energy resource in Japan. Biomass and
decomposes to oxygen and water [8]. It can be seen in Bioenergy, 29(5), pp.347-354.
Table 1, the consumption of hydrogen peroxide is [3] Ibrahim, M.M., El-Zawawy, W.K., Jüttke, Y.,
expressed as the percentage of the residual hydrogen Koschella, A. and Heinze, T., 2013. Cellulose
peroxide to the initial amount of hydrogen peroxide and microcrystalline cellulose from rice straw
added. Consumption of hydrogen peroxide at a and banana plant waste: preparation and
concentration of 6% is greater than 4% meanwhile the characterization. Cellulose, 20(5), pp.2403-
2416.
lignin content of 6% is higher than 4%.
[4] Ramos, L.P., 2003. The chemistry involved in
IV. Conclusion the steam treatment of lignocellulosic
materials. Química Nova, 26(6), pp.863-871.
Hydrogen peroxide bleaching of rice straw
[5] Dence, C.W. and Reeve, D.W., Pulp
hydrotropic pulp was able to produce bleached pulp Bleaching: Principles and Practice,(1996).
brightness value of 67-80%. Hydrogen peroxide
[6] Datta, R., 1981. Acidogenic fermentation of
concentration and bleaching time give significant lignocellulose-acid yield and conversion of
effect on cellulose content, lignin content and components. Biotechnol. Bioeng.;(United
brightness of the bleached pulp. The maximum value States), 23(9).
of bleached pulp brightness was obtained from the [7] Fitriana, N.E., Suwanto, A., Jatmiko, T.H.,
bleaching process with 4% hydrogen peroxide Mursiti, S. and Prasetyo, D.J., 2020. Cellulose
concentration at temperature 90oC for 120 minute extraction from sugar palm (Arenga pinnata)
reaction time with the cellulose content was 52.41% fibre by alkaline and peroxide treatments.
and the lignin content was 14.45%. E&ES, 462(1), p.012053.
[8] Fuadi, A.M. and Sediawan, P.P.I.W.B., 2009.
Acknowledgement Pemakaian hidrogen peroksida sebagai bahan
pemutih pulp (Doctoral dissertation,
The authors are thankful to The Ministry of
[Yogyakarta]: Universitas Gadjah Mada).
Research, Technology, and Higher Education of
[9] Tutus, A., 2004. Bleaching of rice straw pulps
Republic of Indonesia for their support this work
with hydrogen peroxide. Pakistan Journal of
through PDUPT research grant of 2020 with contract
Biological Sciences, 7(8), pp.1327-1329.
number of 2845/UN1/DITLIT/DIT-LIT/2020.

20
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Different Effectiveness of disinfectants on Virus and Bacteria Removal on Surface

Chakriya Kong1 , Dariya Sek 1, Monychot Tepy Chanto1,2, Sokly Siev1,3 , Chanthol Peng 1,3*, Reasmey Tan2 , Thavarith
Chunhieng1 , Romny Om4, Phen Sieng1, San Penh1

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Water and Environment Research Unit, Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Department of Electrical and Energy Engineering,
*peng@itc.edu.kh

Abstract

The infectious diseases caused by virus and bacteria have posed serious threats to public health worldwide. The rapid
community spread of infection and the lack of sanitization could pose major human health problem. The diseases are often
transmitted by close human-to-human contact or by contacting contaminated surface. In prevention, the disinfection of surface
is essential to reduce the transmission. In this study, different disinfectant such as ethanol (75% v/v and 90% v/v), hydrogen
peroxide (3%) and calcium hypochlorite (2 ppm and 4 ppm) were evaluated their effectiveness under various application
conditions to disinfect virus and bacteria on the surface. The virus, bacteriophage T4 and the bacteria, Escherichia coli K-12
were used as a model strains in this study. The disinfectants were applied on the virus and bacteria laying on the surface by
spraying method for 5 and 10 times with 30s and 60s of contact time. The effectiveness of those disinfectants was evaluated
based on the survival of virus and bacteria through plaque assay and spread plate method. As a result, hydrogen peroxide
showed the most promising effective disinfectant compared to other. For instance, at spraying for 10 times in 60s, the bacteria
are completely reduced and the highest virus reduction was 90%. While ethanol at concentration of 90% showed 85%
reduction for virus and around 98% for bacteria, the rest showed reduction less than 50%. This implies that the effectiveness
of disinfectant depends on the type, contact time and the concentration.

Keywords: Disinfectant, Bacteriophage, E. coli, Contacting Time

I. Introduction used in different application. Calcium hypochlorite is used

The health center and community acquired infection as disinfection of water treatment, surface area, etc. Its
caused by virus and bacteria have become a majority activity was reported to be effective against Staphylococcus.
concern. In prevention, the chemical disinfectants are used aureus, Salmonella choleraesuis, and Pseudomonas.
to eliminate the contaminated surface. There are many types aeruginosa with just contact time less than10 minutes [1].
of disinfectants such as calcium hypochlorite (𝐶𝑎(𝑂𝐶𝐿)2 ), Ethanol is used for hospital pagers, scissors, and
ethanol (𝐶2 𝐻5 𝑂𝐻) and hydrogen peroxide (H2O2) which are stethoscopes, surface and hand sanitation. Ethanol has

21
antimicrobial activity against with P. aeruginosa, Serratia ml of Luria- Bertani (LB) liquid media at 37℃ for 18-20h
marcescens, Escherichia coli and Salmonella typhosa,
using BR-21UM Bio-shaker at the speed of 120 rpm [6].
rotavirus, echovirus, and astrovirus [2]. Hydrogen peroxide
The overnight culture of E. coli K-12 was probably
used to remove pollutants from waste water and from air and
1.00E+09 CFU/ml.
treat pollutions that can be easily oxidized. Hydrogen
2.2. T4 phage Preparation
peroxide is active against E. coli, Streptococcus species, and
To prepare this virus, a single plaque of T4 from mother
Pseudomonas species [1]. However, it is essential to
stock was used for propagation with host cell, E. coli K-12
understand their effectiveness including the safety and
using the double layer agar method and incubated at 37 oC
killing time in order to choose the right disinfectant to
overnight. T4 phage purification was conducted by repeated
inactivate bacteria and virus for disease prevention.
plating and picking of single plaques, followed by plate
In review, E. coli is a gram negative bacteria which the
lysate and the polyethylene glycol (PEG) # 6000 – NaCl
peptidoglycan layer is relatively thin and it is found at the
precipitation method [7].The concentration of daughter
inside of the outer membrane. E. coli is found in the intestine
phage was determined by using plaque essay method.
of men and animals and it is released into the environment
through fecal material [3]. One of the E. coli strain K-12
2.3. Disinfectant Preparation
which is generally used in experimental and considered as
The stock solution of calcium hypochlorite 𝐶𝑎(𝑂𝐶𝐿)2
nonpathogenic.
was prepared from Ca(OCL)2 powder by dissolving 0.1g of
The bacteriophage T4 which is one of the strain
𝐶𝑎(𝑂𝐶𝐿)2 in 100ml distilled water. The stock solution was
of bacteriophages that infects E. coli K-12 was used as a
stirred on the magnetic stirrer until they are completely
model virus in this research. T4 phage infection of E. coli
dissolved. The concentration of 𝐶𝑎(𝑂𝐶𝐿)2 were identified
has been one of the most thoroughly studied model systems
using Colorimeter C401. The used concentrations of ethanol
in molecular biology and microbiology for over 60 years due
were archived by the dilution of absolute ethanol solution
to harmless for human [4]. In order to multiply their quantity,
99,9%(v/v) with distilled water. The 3% hydrogen peroxide
they need to get into a bacterium where they break the
is purchased from the market. The disinfectant conditions
bacterial cell to release the new viruses[5].
were summarized in Error! Reference source not found. .
Most of studies focus on the activity of disinfection
suspension but little is known about their activity when
spraying on the surface application. Recently, the use of 2.4 Disinfectant Spray Treatment Method
various disinfectant on the cabin spray has been emerged. Spreading method is a popular procedure to identify the
Yet, the study of the effectives of the spray application of amount colony of bacteria. In this method, 0.1ml of E. coli
K-12 overnight culture was pipetted to spread on the layer
disinfectant is unknown. Therefore, the aims of this study is
of Chromocult agar in plastic plate. Then, the plates were
to investigate the effectiveness of different disinfections on
transferred to incubator at 37 °C for overnight before
model strain of bacteria and virus by spray application.
counting [8].
Calcium hypochlorite at 2ppm and 4 ppm, ethanol
Plaque assay is one of the widespread approaches used
concentration at 75% and 90%, hydrogen peroxide
to determine the amount of infectious virus in a sample. The
concentration 3% were carried out to identify their
effectiveness against on the bacteria (E. coli K-12) and virus virus T4 phage stock was vortexed and pipetted 110𝜇 to the
(T4 phage) with different contacting time, 30 and 60 second 0.5ml eppendorf. After that, 110 l of host cell E. coli K-12
on surface. were added. Then, 200 µl of the solution were pipetted and
transferred into the 3 ml of LB soft agar which was the
II. Materials and Methods complex mixture of tryptone, yeast extraction, Sodium
Chloride, Magnesium Sulphate (1M, 0.5% (w/v)) and
2.1 Escherichia coli K-12 preparation Calcium Chloride (1M, 0.5% (w/v)). Next, it was poured on
To use E. coli K-12 as a model bacterium, the host cell the top layer of LB agar. The solution was winded to
solution was prepared from pure culture E. coli K-12’s homogenize and left until it was dried. The culture of
mother stock. The preparation of bacteria was started by bacteriophage was incubated at 37 °C for overnight before
selecting a single colony from mother stock to cultivate in 2 counting [9]. The removal effectiveness of disinfectant on

22
bacteria and virus were calculated by equation below: highest removal efficiency is hydrogen peroxide spaying 10
(𝑁𝑖 − 𝑁𝑓 ) × 100 (E.q.1) times which was able to remove 90% and 87% of T4 phage
𝑅𝑒𝑚𝑜𝑣𝑎𝑙 =
𝑁𝑖 with 60s and 30s contacting time respectively. On the other
hand, ethanol concentration 90% spraying for 10 times is
Where 𝑁𝑖 is the number of total colony or plaque without
permitted to reduce 85% in 60s and 70% in 30s. Calcium
spraying disinfectant, and 𝑁𝑓 is the number of colony or
Hypochlorite 2ppm and 4ppm achieved the lowest
plaque after spraying with disinfectant. effectiveness against bacteria as its efficiency was around
Table 1: The spaying condition of disinfectant 50%.
Disinfectant Concentration Spraying Contact time According to Error! Reference source not found. and
(time) (second ) Error! Reference source not found., the effectiveness of
𝐻2 𝑂2 3% 5, 10 30, 60 disinfectants on bacteria were higher than the effectiveness
of disinfectant on virus. It can be assumed that hydrogen
𝐶5 𝐻5 𝑂6 75%, 90% 5, 10 30, 60
peroxide (3%) and ethanol 90% were the most effective
𝐶𝑎(𝑂𝐶𝑙)2 2ppm,4ppm 5, 10 30, 60 disinfectants against microorganism growth while calcium
hypochlorite seemed to be the least effective one compared
The spraying treatment on bacteria E. coli K-12 was to other disinfectants.
performed duplicate by dropping 5μl of E. coli K-12
overnight culture solution (≈1.00E+06 CFU/ml) on the 120

sterilized glass plate. Then the disinfectant was sprayed in 5 100

time and 10 time on the bacteria surface within contacting
Removal (%)
80
time 30 second and 60 second. After that, 5ml of PBS were
60
added and shaken until the solution was homogenous. Then,
40
0.1ml of that solution was pipetted to conduct spread plate
method to determine the bacteria colony. 20
The spraying treatment on bacteriophage was 0
performed triplicate by dropping 5μl of T4 phage H₂O₂ (Ca(ClO)₂) (Ca(ClO)₂) Ethanol Ethanol
Comercial 2ppm 4ppm 75% 90%
(≈1.00E+06 PFU/ml) on the sterilized glass plate. Then, 5
Disinfection
times and 10 times of disinfectant were sprayed on the T4
5t 30s 5t 60s 10t 30s 10t 60s
phage within contacting time 30 second and 60 second. After
that, 5ml of SM buffer were added and shaken until Fig. 1: The removal efficiency of disinfectants on E. coli
homogenous. Then, plaque essay method was used to K-12. 5t, 10t: 5, 10 times spry of disinfectants.
identify the plaque of virus.
120

III. Results and Discussion 100

Removal (%)

The Bacteria (E. coli K-12) was removed by the 80

effectiveness of disinfectant under various application 60
conditions. In spraying test, the most effective disinfectants
40
were hydrogen peroxide (3%), ethanol (75% and 90%)
20
spaying 10 times in 60s which the bacteria were completely
removed. Furthermore, spraying 10 times in 30s of the same 0
H₂O₂ (Ca(ClO)₂) (Ca(ClO)₂) Ethanol Ethanol
disinfectants revealed the reduction around 98%, while the Comercial 2ppm 4ppm 75% 90%
highest effectiveness of calcium hypochlorite (4ppm) Disinfection
showed only 92% of reduction within spraying 10 times in 5t 30s 5t 60s 10t 30s 10t 60s
60s as have showed in Fig. 1. Fig. 2: The removal efficiency of disinfectants on T4
In spray treatment test, the effectiveness of T4 phage phage
plaque reduction of disinfectant was noticeable. As have
shown in Fig. 2, the most effective disinfectant with the Base on the duration and number of spray, it was shown

23
that the removal efficiency of bacteria and virus were IV. Conclusion
decrease as the number of spray and contacting time were The most effective disinfectants for removing bacteria
reduced while increasing the contacting time and number of and virus were 𝐻2 𝑂2 concentration 3% and ethanol 90% at
spray lead to high removal results. Additionally, ethanol 90% 10times 60s which showed 98% removal for bacteria and 90%
and hydrogen peroxide (3%) seemed to be effective for for virus reduction while the effectiveness of Ca (OCl)2 on
surface disinfection while Calcium Hypochlorite (2ppm) bacteria and virus were less than 50%. However, the
and (4ppm) did not work effectively in this experiment. effectiveness of disinfectants includes calcium hypochlorite,
The safety level of disinfectants was limited to avoid any ethanol, and hydrogen peroxide against bacteria and virus
unexpected health hazard. The permitted level of hydrogen depended on the concentration of disinfectants, contacting
peroxide is between 6-3% which is commonly used to time and number of spraying.
remove surface contaminated, pollutants from wastewater Although some tested disinfectants showed a promising
and, teeth, hair and produced food [1]. Commercially antibacterial and antiviral on the model strain, in term of
available 3% hydrogen peroxide is a stable and effective dose and safety in spray application is not yet known. The
disinfection for using on inanimate surfaces. Hydrogen further study is necessary to investigate on health toxicity
peroxide is poorly absorbed by skin; however, it is mildly associated with cabin and disinfectant spray system.
irritating to the skin and mucous membranes in the
concentration more than 3% [10]. Acknowledgement
The USA Food and Drug Administration (FDA)
We are thankful to the Institute of Technology of
classified Ethanol 60-95% was safe and effective active
Cambodia for the financial support of this study.
agent to use in antiseptic and hygiene hand wash products
[11]. Ethanol is widely used in all kinds of products with
direct exposure to the human skin. The higher concentration References
of Ethanol have been describe to be a higher removal [1] Rutal,W.A., Weber. D. J. (2019). Guidline for
efficiency [12]. Disinfectant and Sterilization in Healthcare
The toxic effects of calcium hypochlorite are primarily Fecilities, 2008. Center for Disease Control and
due to the corrosive properties of hypochlorite. Ingesting a prevention. University of North Carolina Health
small amount of household bleaches (3-6% hypochlorite) Care System pp 40
can cause the gastrointestinal irritation and if ingesting more [2] Kampf, G., Grotheer,D.,Steinmann, J., 2019.
concentrated commercial bleach 10% or higher hypochlorite Efficacy of three ethanol-based hand rubs against
may suffer severe corrosive injuries to the mouth, throat, feline calicivirus , a surrogate virus for norovirus,
esophagus and stomach with bleeding, perforation, and pp. 144–149.
eventually death. Contacting of strong hypochlorite [3] Siva,B.,Storms,J.,Sauvageau,Z.,and
solutions with your skin may cause burning pain, Dominic,S.,2016. Host receptors for bacteriophage
inflammation, and blisters [13]. adsorbtion.FEMS Microbiology letters,363,1-11.
Surface disinfectant products are used for surface [4] Bryan,D., El-shibiny,A., Hobbs,Z.,Porter,J., Kutter,
disinfection, which refers to the application of chemical E.M., 2016. Bacteriophage T4 Infection of
disinfectants to surfaces in order to sanities them. This Stationary Phase E . coli. Life after Log from a
disinfection differs from aerosol/dry-fogging and water Phage Perspective, vol. 7, no. September, pp. 1–12.
disinfection in terms of the requirements for the disinfectant [5] Gutiérrez,D., Fernádez, L., Maetínez, B.,
as well as the mode of application. [14]. Clean and disinfect Rodríguez, A.,García, P., 2016. Bacteriophage. The
high-touch surfaces daily in household common areas (e.g. Enemies of Bad Bacteria Are Our Friends.
tables, hard-backed chairs, doorknobs, light switches, Frontiers for Young Minds,vol.4.
phones, tablets, touch screens, remote controls, keyboards, [6] Karen,L.,Roger,B.,(2018) Growth of E.coli in
handles, desks, toilets, sinks) this type of disinfection liquid media. Current protocol in molecules
primarily involves treating easy-to-clean, washable and biology vol 125, pp81
chemical-resistant surfaces [15]. [7] Synnott. A. J., Kuang.Y., Kurimoto. M.,Yamamichi.
K., Iwano. H., Tanji.Y. (2009). Isolation from

24
 sewage influent and characterization of novel
Staphylococcus aureus bacteriophages with wide
host ranges and potent lytic capabilities. Appled
Environ Microbiology
[8] Spread plate technique; principles,procedure and
advance,(2019).Microbe Noted.
https//microbenotes .com/spread-plate-technique/.
Assesed 05 December 2020
[9] Acharya,T.,(2018). Medical Micrology Guid.
Bacteriophage plaque Assay Principle, Procedure
and Results. https://microbeonline.com/phage-
plaque-assay-principle-procedure-results/.Assesed
05 December 2020.
[10] Agency for Toxic Substances and Disease Registry
4770 Buford Hwy NE Atlanta, GA30341: Toxic
Substance Portal-Hydrogen Peroxide
[11] Food and Drug Administration: Tentative final
monograph for healthcare antiseptic drug products;
proposed rule. Federal Register. 1994, 59: 31441-
31452.
[12] Kramer A, Rudolph P, Kampf G,Pittet D: Limited
efficacy of alcohol-based hand gels. Lancet. 2002,
359: 1489-1490. 10.1016/S0140-6736(02)08426-
X.
[13] Agency for Toxic Substances and Disease Registry
(ATSDR). 2002. Managing Hazardous Materials
Incidents. Volume III – Medical Management
Guidelines for Acute Chemical Exposures:
Calcium Hypochlorite/Sodium Hypochlorite
Atlanta, GA: U.S. Department of Health and
Human Services, Public Health Service.
[14] Eichtalstrasse 49 CH-8634 Homebrechtikon., 2020
Sanosil LTD. All Rights Reserved.
[15] CDC Home Care Guidance for people pets.,
Centers for Disease control and prevention., July10,
2020: Detail Disinfection Guidan.

25
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Identification and Quantitative Analysis of Alternative Diesel from Waste Plastic Pyrolysis
Zin Thu Aung 1, Chinda Charoenphonphanich1*, Hidenori Kosaka 2, Pop-Paul Ewphun2, Prathan Srichai3

1
School of Engineering, Department of Mechanical Engineering, King Mongkut’s of Institute of Technology Ladkrabang,
Soi chalongkrung1, Ladkrabang, Bangkok 10520 Thailand
2
School of Engineering, Department of System and Control Engineering, Tokyo Institute of Technology,
Ishikawadai 6th Building, Room 323-2-12-1 Ookayama, Meguro-ku, Tokyo, 152-8552, Japan
Department of Mechanical Engineering, Princess of Naradhiwas University, Naradhiwas 96000 Thailand
*Corresponding author: kmitl.chinda@gmail.com

Abstract

Plastic is a synthetic material made from a wide range of organic polymers, facilitating the human life and help ing
promote the global economy. The use of plastics, however, has been associated with significant environmental problems
due to their accumulation in landfills, as plastic waste does not degrade or degrades at very low pace. Nowadays, fast
pyrolysis of waste plastic into valuable fuels is main platform method in the waste dispo sal but also could be used as
alternative fuel for internal combustion engines. The purpose of this study was to identify, quantify and compare the
composition of waste plastic diesel (WPD) with the commercial diesel (CD) of Thailand. Simulated distillation (GC-
FID) and n-d-M method were used for finding the composition of both fuels. Results indicated that the content of
naphtha, kerosene, diesel, and heavy oil were determined quantitatively and also identified the paraffin, n aphthenes,
and aromatic contents for both fuels. Naphtha and heavy oil contents of WPD were 9.2 and 8.9wt% higher than those
of CD but kerosene and diesel contents were 0.7and 17.4wt% less than those of commercial diesel. After that, paraffin,
naphthenes and aromatic contents of WPD from hydrocarbon type analysis were 80.42, 14.54 and 5.04wt% and these
hydrocarbon contents of CD were 60.61, 25.91 and 13.48wt% respectively. By knowing them, the appropriate method
can be determined for fuel upgrading and interpret correctly the combustion and emissions results.

Keywords: Fast pyrolysis, Simulated distillation, n-d-M method, Waste plastic diesel, Commercial diesel

I. Introduction batch by becoming yellowish and brittle [3].Unlike

recycling, pyrolysis does not require a keen sorting of
Plastics are essential materials due to their numerous
different plastics. Therefore, fast pyrolysis of waste plastic
applications in daily life. Consequently, a huge number of
into valuable fuels is main platform method the waste
plastic products accumulate as waste in the environment.
disposal but also could be used as alternative fuel for internal
Plastic waste is a big issue in the world including Thailand,
combustion engines.
because the amount of recycled plastic remains low due to
Previous studies found that individual type of waste
recycling problems [1]. One of the recycling problems is
plastics or mixed waste plastic, which were used to produce
economy as they need to be collected separately or sorted
alternative fuel, identify the chemical compounds and
before the process can begin [2]. Most plastics are not
carbon number, and investigate the physical properties [4-6].
compatible with each other and hence they cannot be
There are no or few studies to quantify the boiling fractions
processed together during recycling.
and hydrocarbon type of waste plastic fuel.
For instance, a polyvinyl chloride (PVC) bottle in
Therefore, the purpose of this study is to identify and
polyethylene terephthalate (PET) recycle can ruin the entire

26
quantify the composition of waste plastic diesel (WPD) from Table 2. Experimental conditions for ASTM D2887
real mixed waste plastic pyrolysis, and compare with Column DB-10,10mx0.53mm,2.65µm
commercial diesel (CD). Column temperature 40℃ to 350℃
Carrier gas flow rate 13.989 L/min (helium)
II. Materials and Methods Injection temperature 350℃
FID temperature 375℃
2.1. Materials Gas flow rate
Nitrogen (makeup) 45 mL/min
The waste plastic diesel utilized in this study was derived Hydrogen 40 mL/min
from catalytic fast pyrolysis of real mixed waste plastic Air flow 450 mL/min
(mostly PVC) and commercial diesel (B10) were purchased Injection volume 0.1 µL
from PTT fuel station in Thailand. Table 1 shows the
comparison of physical properties of WPD and CD. 2.3. n-d-M Method (PNA composition Analysis)
This method requires three physical properties of
Table 1. Physical properties of fuels refractive index (n20), density (d20), and molecular weight
Property ASTM CD WPD (M). For this reason, the method is called n-d-M method.
Method
The method is included in the ASTM manual under ASTM
D3238. It calculates the distribution of carbon in paraffin
Density@15℃ D4052 824 805
(kg/m3) (%CP), naphthenes (%CN), and aromatics (%CA) using
Viscosity@40℃ D445 3.24 2.9 equations 1 to 5. The refractive index and density at 20℃
(Cst) and molecular weight are used as input data, which are
Cetane Index D976 56.43 67.93 estimated from correlations that are adopted in API-TDB [7].
Energy Content D240 45.86 46.29
(MJ/kg)
𝑣 = 2.51(𝑛 − 1.475) − (𝑑 − 0.851) (Eq.1)
Sulfur Content D5453 0.003 0.014
(wt.%) 𝑎 = 430 𝑖𝑓 𝑣 > 0 𝑎𝑛𝑑 670 𝑖𝑓 𝑣 < 0
%𝐶𝐴 = 𝑎𝑣 + 3660/𝑀 (Eq.2)
2.2. Simulated Distillation Method
The identification and quantitative analysis of WPD and 𝑤 = (𝑑 − 0.851) − 1.11(𝑛 − 1.475) (Eq.3)
10600 (Eq.4)
CD were done by simulated distillation (ASTM D2887). %𝐶𝑅 = 1440𝑤 − 3%𝑆 + 𝑖𝑓 𝑤 < 0
Simulated distillation is a Gas Chromatographic technique 𝑀
%𝐶𝑅 = %𝐶𝑁 + %𝐶𝐴 (Eq.5)
for determining the boiling point distribution of fuels by
%𝐶𝑃 = 100 − %𝐶𝑅 (Eq.6)
Flame Ionization Detection (GC-FID). Two standard
solutions were used for quantification of waste plastic diesel
and commercial diesel: normal alkanes ranging from n-C5 to III. Results and Discussion
n-C10 and n-C10 to n-C40. Table 2 shows the testing condition ` The simulated distillation curve represents the boiling
of GC-FID. In simulated distillation method, the analyse points of compounds in a fuel mixture at atmospheric
retention times are directly related to the boiling points of pressure. Simulated distillation curve is presented in term of
various hydrocarbons. A calibration curve is generated from boiling point versus wt% of mixture vaporized because
standard solution, where the retention time of each n-alkane composition is measured in terms of wt% or weight fractions
is plotted against its boiling point. Calculating the area of in gas chromatography. Simulated distillation curve is very
each time interval allows the proportion of elution weight close to actual or true boiling point curve. The distillation
(%) in each time interval. The elution weight (%) in each curve obtained applying the ASTM D2887 method allow the
boiling point range can be determined from calibration curve quantification of fuel composition such as naphtha, kerosene,
and used to obtain the relationship between the elution diesel and heavy oil as shown in Fig. 1. The composition of
weight (%) and boiling point to create the distillation curve CD and WPD are listed in Table 2.
of fuel samples.

27
 100
500
Heavy Oil
Weight percent vaporized (%)

90
450 421℃
80
400

Temperature (℃)
70
350
60
50 Diesel 300

40 250
30 200
20 Kerosene 150
CD
10 94%
Naphtha 100 WPD
0
100 150 200 250 300 350 400 450 50
0 10 20 30 40 50 60 70 80 90 100
Boiling cut point (℃)
Weight percent vaporized (%)
Fig. 1. Distillation curve with cut points
Fig. 2. Comparison distillation curves of CD and WPD
Table 2. Fuel compositions of WPD and CD
Composition Cut point CD WPD Table 3. Boiling points of WPD and CD
range wt.% wt.%
Naphtha IBP-200℃ 9.6 18.8 Distillation ASTM CD WPD
Kerosene 200-250℃ 15.4 14.7 Wt.% (℃) (℃)
Diesel 250-370℃ 65.2 47.8 IBP:0.5% 126 100
Heavy Oil 370-FBP 9.8 18.7 10% 201 172
20% 236 208
It can be seen that naphtha and heavy oil contents of 30% 260 236
WPD were 9.2 and 8.9 wt% higher than those of CD but 50% 298 289
kerosene and diesel contents were 0.7 and 17.4% wt% less 60% 2887 315 316
than those of commercial diesel. 70% 332 342
A comparison of the distillation curve between CD and 80% 350 369
WPD is shown in Fig. 2 and distillation temperatures of 90% 370 402
WPD and CD are provided in Table 3. In Fig. 2, initial 95% 389 426
boiling range of WPD is almost 60% lower and final boiling FBP:99.5% 421 474
range is almost 40% higher than those of CD. Therefore, the
WPD can be called as wide distillation fuel (WDF) because For low temperature combustion of modern diesel
it is included lighter and heavier compounds than those of engine, desirable fuel characteristics are low aromatic and
CD. In fact, the lighter and heavier compounds can be high cetane index (high normal paraffin) but high cetane
removed to match the initial boiling point and final boiling index and end boiling point can lead to high smoke for hot
point of WPD and CD fuels by distillation. Nowadays, temperature combustion of normal diesel. Table 5 shows
gasoline-diesel blended or wide distillation fuels have that all detected carbon number concentrations of WPD are
potential to reduce soot emission and increase thermal greater than that of CD. That is why WPD has high cetane
efficiency but HC emission of these fuels are slightly higher index and paraffin contents. The C28-C40 are not detected for
than that of diesel because of low cetane number [8]. both fuels but three peaks are detected between C24 and C28
Remarkably, cetane index of WPD is much higher than that in WPD. Therefore, end fractions 6% of WPD should be
of CD, although it contains almost 60% of lighter removed to achieve the same end point as shown in Fig. 2
compounds. Paraffin and aromatic ratio of WPD is higher and Table 5.
than that CD as shown in Table 4 of WPD

28
Table 4. PNA concentration of fuels detected for both fuels but three peaks are detected between
C24 and C28 in WPD. Therefore, end fractions 6% of WPD
Carbon Content CD WPD
should be removed to achieve the same end point.
Paraffin (%CP) 60.61 80.41
Naphthenes (%CN) 25.91 14.54
Acknowledgement
Aromatic (%CA) 13.48 5.05
This work was supported by ASEAN University
Table 5. Constituents (area%) identified by GC-FID Network/Southeast Asia Engineering Education
Carbon Content CD WPD Development Network (AUN/SEED-Net). The authors
would like to thank Assoc. Prof. Dr Kanit Wattanavichien,
C6 2.1457 3.38784 Center of Fuel and Energy from Biomass (Chulalongkorn
C7 Not detected 1.71151 University) for his contribution in this work.
C8 Not detected 2.35746
C10 Not detected 1.41769 References
C11 1.38791 3.71853
[1] W Khatha., S Ekarong., M Somkiat., S
C14 2.84199 4.32735
Jiraphon.,2020. Fuel properties, performance and
C15 1.7203 4.77871
emission of alternative fuel from pyrolysis of waste
C16 1.07963 5.20978
plastic. IOP Conf. Series: Materials Science and
C17 2.42153 5.19759
Engineering 717.
C18 3.68398 6.00175
[2] Fazal Mabood., M.R.Jan Jasmin Shah., Farah
C20 3.59628 5.52902
Jabeen., 2012. Catalytic pyrolysis of waste plastic
C24 1.47851 3.10487
and tyres. LAP LAMBERT academic publishing,
- Not detected 2.52215
U.S.A.
- Not detected 2.06772
[3] Anandhu, V., Jilse, S., 2018. Pyrolysis process to
- Not detected 1.61383
produce fuel from different types of plastic- a review.
C28 Not detected Not detected
IOP Conf. Series: Materials Science and Engineering
C32 Not detected Not detected
396.
C36 Not detected Not detected
[4] Brajendra, K.S., Bryan, R.M., Karl, E.V., Kenneth,
C40 Not detected Not detected
M.D., Nandakishore, R., 2014. Production,
characterization and fuel properties of alternative
IV. Conclusion diesel fuel from pyrolysis of waste plastic grocery
In this work, the composition of WPD are identified and bags. Fuel processing technology 122, 79-90.
quantified using simulated distillation and n-d-M methods. [5] A.M. Motawie., Hala. B.I., Hasabo, M.A., Sahar,
Some conclusions can be drawn as follows. M.A., R.M, Abualsoud., 2016. Fractional distillation
1. Naphtha and heavy oil contents of WPD were 9.2 a of fuel from mixed plastic waste. Conference paper.
nd 8.9wt% higher than those of CD but kerosene and [6] Z.T, Aung., C, Charoenphonphanich., H, Kosaka., P,
diesel contents were 0.7and 17.4wt% less than those Ewphum., P, Srichai., 2019. Investigation on physical
of commercial diesel. properties and measurement of bulk modulus of
2. Initial boiling range of WPD is almost 60% lower and waste plastic diesel. The 10th AUN/SEED-NET
final boiling range is almost 40% higher than those of CD. RCMEManuE,129-132.
Therefore, the WPD can be called as wide distillation fuel [7] M. R, Riazi., 2005. Characterization and properties
(WDF) because it is included lighter and heavier compounds of petroleum fractions. 1st ed. ASTM, U.S.A.
than those of CD. [8] J, Wang., Z, Wang., H, Liu., 2015. Combustion and
3 All detected carbon number concentration of WPD are emission characteristics of direct injection
greater than that of CD. That is why WPD has high cetane compression ignition engine fueled with full
index and paraffin contents. Then, the C28-C40 are not distillation fuel. Journal of fuel 140, 561-567

29
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Hydrothermal Technology for Wastewater Treatment Plants Sludge Treatment

Into Potential Organic Liquid Fertilizer

Farida Crisnaningtyas1*, Mohammad Fahrurrozi1, Arif Sosiawan2, Chandra Wahyu Purnomo1,3

1
Department of Chemical Engineering, Faculty of Engineering, Universitas Gadjah Mada Jl. Grafika No. 2
Sleman, Yogyakarta 55381, Indonesia
2
PT. Sarihusada Generasi Mahardhika, Yogyakarta Factory, Yogyakarta Indonesia
3
Agrotechnology Innovation Center, PIAT-UGM, Sleman, Yogyakarta, Indonesia
*farida2018@mail.ugm.ac.id

Abstract

Wastewater Treatment Plants (WWTP) sludge can be converted into liquid fertilizer using hydrothermal treatment (HT).
This conversion technology utilizes high temperature and pressure with the presence of water as a solvent as well as reactant.
Treatment temperature of 180oC, 200 oC, 220 oC, with water-sludge ratio of 1:3, 1:5, 1:7, and a residence time of 30 minutes
were used in this research. The sludge was mixed with zeolite as catalyst, with ratio of 20% (w/w) of the sludge. Liquid
product was analyzed for the content of macronutrients (i.e. N, P, K) using the Kjeldahl method, Phosphate High Range
Portable Photometer, and Pottasium High Range Portable Photometer, respectively. The content of micronutrients was
analyzed by using Inductively Coupled Plasma (ICP). The results showed that the liquid product has concentration of 4000-
16000 ppm of total N, 300-800 ppm of P, and 3000-5000 ppm of K. Result shows that there was a potential application to
utilize the liquid as organic fertilizer. The solid product was become more stable and safer to be disposed.

Keywords: hydrothermal, sludge, liquid fertilizer, waste treatment

I. Introduction e stable in the environment. Fermentation and compostin

g methods take a longer time, more than one day or eve
Dairy industry generates a large amount of solid waste n weeks to get the results [2]. Meanwhile, pyrolysis me
which is easily to decomposed that causes an unpleasant thod requires pretreatment before being put into the reac
odor in the environment [1]. Several methods to treat solid tor. On the other hand, hydrothermal offers effective and
waste including sludge from wastewater treatment plants economical sludge treatment by thermal method with
(WWTP) are composting, fermentation, fortification, py rapid process. It is require high temperature and pressure
olysis, hydrothermal, and so on [2]. The method used is surrounded by water during the reaction time. In
focused on decreasing the growth rate of bacteria, there hydrothermal process, the solid material is bounded by water
by inhibiting further digestion of sludge, making it mor during reaction, which is kept in liquid state by allowing the
pressure to rise with the steam pressure in high-pressure

30
reactors [3]. After hydrothermal process, waste sludge Pressure sensor
becomes more stable. If the raw materials contain organic
materials, the characteristic of the hydrothermal product Pressure valve Stirrer

would slightly different, but with more stable and safe

properties, so we can use it into something more valuable
such as fertilizer, both solid and liquid compounds such as Reactor Controller
N,P,K, and micronutrients.
Research on hydrothermal has been done a lot. Yuan et Motor
al. used hydrothermal technology with an operating
temperature of 110oC to 200oC; reaction time 0, 10, 30, 60
minutes; stirring 60 rpm to study the fertilizer potential of
liquid product from hydrothermal treatment of swine Figure 1. Scematic view of small scale hydrothermal
manure [4]. Jambaldorj states that the concentrations of N reactor
and K decrease when the temperature is increased, but that
the concentration of P increases when the temperature is
The sludge was mixed with distilled water with the ratio of
increased [5]. Shao et al, said about the advantage of
1:3, 1:5, and 1:7. Natural zeolite was added to the reactor
hydrochar from the hydrothermal process and it can be used
with 20% of sludge. Reactor was sealed, then heated with
as an adsorbent material and solid fuel by using an operating
the average heating rate of 6°C/min to targeted temperatures
temperature of 130-190oC, holding time 1 hour, and the ratio
(180, 200, and 220 °C) and kept for 30 min (the holding
of liquid and solid is 8:1 [6].
time). Once the holding time was completed, the reactor was
Publications on the processing of dairy waste using the
cooled down (< 60°C) and depressurized. The treated
hydrothermal method are still rare. Research on this shows
mixture was taken out. Solid and liquid were separated. The
that organic waste from the dairy industry can be used to
solid phase was oven-dried (105°C for 4-8 hours)
recover nutrient elements which can be used as raw material
meanwhile liquid phase was filtered through sterile
for fertilizer production.
analytical filter units (Whatman 42). A schematic diagram of
the experimental setup is shown in Figure 2.

II. Materials and Methods

water Analysis of
2.1. Materials micro nutrient

Sludge was collected from a wastewater treatment plant sludge Hydro Liquid Analysis of
of a dairy milk factory located in Jogjakarta. Sludge has thermal pro N, P, K
(180, 200,
moisture content of 89,03% and pH of 6,94. Composition of 220oC) Solid
N, P and K was 1.96, 0.71, 0.14 respectively, meanwhile, for 30 min, Proximate
analysis
Ca, Mg, Fe, Mn, Cu, and Zn were 0.51, 0.09, 0.77, 0.09.
0.01, 0.05 respectively. Natural zeolite (3 mesh) for the Figure 2. Schematic diagram of experimental
experiment was taken from Klaten, Central Java.
2.3. Analyses
2.2. Experimental set-up
The macro and micronutrients were analyzed after
Hydrothermal treatments were carried out in a 2 L batch product separation. The total-N was measured using the
reactor. The reactor was equipped with a stirrer, a pressure Kjeldahl method while potassium (K) and Phosphate (P)
sensor, and a temperature controller as shown in Figure 1. were analyzed with HANNA portable high range
spectrophotometer (HI96717, HANNA, USA). The
micronutrient and other heavy metals were determined using
the ICP emission spectroscopy (8300, Perkin Elmer, USA).
.

31
III. Results and Discussion Effects of temperature on the nutrient components
The concentration of nutrients in the liquid product of
The hydrothermal treatment produces three products i.e
hydrothermal was shown in Table 3 and Table 4. The
liquid, solid and gas products. The distribution of those
macronutrients in liquid product were Nitrogen (N),
products was shown in Table 2. Experiments showed when
phosphate (P), and potassium (K).
temperature increased, the yield of liquid product also
increased. When the temperature raised, the solid production Table 3. The concentration of macronutrients in the liquid
decreased gradually and the liquid product increased [8]. In product of hydrothermal
Table 2, the amount of liquid product has increased as well
as temperature increase. The highest amount of liquid Ratio Temp, (℃) N (mg/L) P (mg/L) K (mg/L)
produced at 220℃. This happened because biomass 1:3 180 16223 680 4200
conversion needed the role of temperature to provide heat to 200 12236 810 5400
disintegrate the fragmentation of biomass bonds [8]. 220 9032 860 6800
Temperature is the most influential parameter for bio-oil 1:5 180 9339 550 4000
from hydrothermal liquefaction process with raw materials 200 740 5200
8357
for several types of biomass [9]. High temperature causes a 220 810 5400
7073
higher reaction speed and affects the number of hydrolyzed 1:7 480 3400
180 5360
compounds [10]. The same results were observed for the
200 5115 550 4200
reaction of cellulose in water at 190°C in the absence of
220 4322 700 4800
pressurized H2, suggesting that the solubilization ratio
appears to be temperature-dependent only [11].
The effects of temperature on the nutrient component
produced from hydrothermal treatment were shown in table
Table 2. Distribution of solid and liquid product after
hydrothermal process 3. The higher the temperature, the higher the P and K
obtained. The result shows that the highest concentration of
Ratio Temp, (℃) Liquid (ml) Solid (g) P and K was achieved at 220oC. On the other hand, the
1:3 180 500 174 concentration of N decreased when the temperature
200 504 130 increased. The reason is that the steam supply increased as
220 505 131 well as the temperature increase and make the dilution
1:5 180 700 169 happened.
But among the other compounds that were produced in
200 705 121
this experiment, Nitrogen was found to be in the highest
220 750 97
concentration. This merely because the initial concentration
1:7 180 800 119
of Nitrogen in the sludge’s raw material was also high.
200 860 67
Before treatment, nitrogen was present in the form of
220 865 68 macromolecular organic nitrogen in the solid phase of the
sludge.
Beside the macronutrient, there was micronutrient
Besides temperature, the water-sludge ratio had an
content from the sludge that we want to gain as the potential
important effect on the production of liquid products from
materials for fertilizer. Table 4 shows that the contents of
hydrothermal process. The higher ratio of water/sludge, the
micronutrient in the liquid product from 180oC to 200oC,
higher quantity of liquid product was generated. Water
most of the elements were increased as well as the
becomes the solvent of hydrothermal process. It has an
temperature increase, but the micronutrients decrease when
important role to enhance the stability and solubility of
the temperature increase from 200oC to 220oC.
decomposition compounds. As reported previously, the
increase of water in the ratio, enhance the yield of liquid and
gas [12]. On the other hand, the highest amount of solid
product was in the lowest temperature.

32
 Table 4. The concentration of micronutrients in the mising technology for producing potential liquid fertilize
liquid product of hydrothermal r from the sludge waste.

Temp Fe Al Mn Ca Na Si
Ratio References
(℃) (%) (%) (%) (%) (%) (%)
[1] B. S. S. and N. P. Shinkar, “Dairy Industry
1:3 180 0.006 0.001 6.9E-05 0.195 0.023 0.013
Wastewater Sources , Characteristics & its Effects
200 0.007 0.005 9.41E-05 0.190 0.027 0.014 on Environment,” Int. J. Curr. Eng. Technol., vol.
220 0.006 0.004 5.76E-05 0.198 0.027 0.016 Vol.3, No., pp. 1611–1615, 2013.
1:5 180 0.004 0.001 9.05E-05 0.143 0.018 0.019 [2] S. Rana, P. Mishra, R. Gupta, and L. Singh, solid-
200 0.004 0.004 4.48E-05 0.099 0.018 0.020 wastes to useful products. Elsevier B.V., 2020.
220 0.003 0.003 5.96E-05 0.121 0.016 0.021 [3] J. A. Libra et al., “Hydrothermal carbonization of
1:7 180 0.002 0.001 3.08E-05 0.100 0.011 0.016 biomass residuals: a comparative review of the
chemistry, processes and applications of wet and dry
200 0.003 0.002 4.64E-05 0.110 0.013 0.018
pyrolysis,” Biofuels, vol. 2, no. 1, pp. 71–106, 2010.
220 0.002 0.002 2.47E-05 0.084 0.013 0.017
[4] T. Yuan et al., “Fertilizer potential of liquid product
from hydrothermal treatment of swine manure,”
After separated the liquid and solid product of Waste Manag., vol. 77, pp. 166–171, 2018.
hydrothermal treatment, and dried the solid product, the [5] G. Jambaldorj, M. Takahashi, and K. Yoshikawa,
proximate analysis was carried out by using several “Liquid Fertilizer Production from Sewage Sludge
standards: ASTM D3173 for water content, ASTM D3174 by Hydrothermal Treatment,” Proc. Int. Symp.
for ash content, ASTM D3175 for volatile content, and EcoTopia Sci. 2007, vol. 07, pp. 605–608, 2007.
[6] Y. Shao, Y. Long, H. Wang, D. Liu, D. Shen, and T.
ASTM D3172 for fixed carbon content. The result of the
Chen, “Hydrochar derived from green waste by
proximate analysis is shown in Table 5. microwave hydrothermal carbonization,” Renew.
Energy, 2018.
Table 5. Proximate analysis of solid product [7] Q. Mo, J. Liao, L. Chang, A. L. Cha, and W. Bao,
“Transformation behaviors of C , H , O , N and S in
volatile fixed lignite during hydrothermal dewatering process,”
Temp Moisture Ash matter carbon vol. 236, no. August 2018, pp. 228–235, 2019.
(oC) (%) (%) (%) (%) [8] S. Nizamuddin et al., “An overview of effect of
Raw 9.62 52.00 31.38 7.00 process parameters on hydrothermal carbonization
180 5.31 46.59 28.69 19.41 of biomass,” vol. 73, no. December 2016, pp. 1289–
200 4.95 44.90 37.77 12.38 1299, 2017.
220 6.93 41.18 33.66 17.24 [9] B. De Caprariis, P. De Filippis, A. Petrullo, and M.
Scarsella, “Hydrothermal liquefaction of biomass :
Table 5 indicates that the carbon content of sludge Influence of temperature and biomass composition
increase after hydrothermal treatment. The fixed carbon on the bio-oil production,” Fuel, vol. 208, pp. 618–
content is influenced by moisture, ash, and volatile matter. 625, 2017.
If the ash and volatile content increase, the combustion [10] Z. Robbiani, “Hydrothermal carbonization of
efficiency will reduce [13]. biowaste / fecal sludge Conception and construction
of a HTC prototype,” ETH Zurich, 2013.
[11] M. T. Reza, M. H. Uddin, J. G. Lynam, S. K.
IV. Conclusion Hoekman, and C. J. Coronella, “Hydrothermal
The liquid product of hydrothermal treatment from carbonization of loblolly pine : reaction chemistry
wastewater treatment plant sludge contained the main and water balance,” Biomass Convers. Biorefinery,
nutrient elements (N, P, K) and the micronutrient needed by vol. 4, pp. 1–11, 2014.
plants. The results showed that liquid products had conc [12] M. K. et al Jindal, “Effect Of Process Conditions On
entrations of 4000-16000 ppm Nitrogen, 400-800 ppm P Hydrothermal Liquefaction Of Biomass 1*,” IJCBS
Res. Pap., vol. 2, no. 8, pp. 8–18, 2015.
hosphate, and 3000-6000 ppm Potassium. The optimum
[13] A. A. Khan, W. De Jong, P. J. Jansens, and H.
concentration of macronutrients (N, P, K) is produced at Spliethoff, “Biomass combustion in fluidized bed
180oC and 1:3 ratio of sludge/water. Considering the res boilers : Potential problems and remedies,” Fuel
ult from the liquid product, hydrothermal could be a pro Process. Technol., vol. 90, no. 1, pp. 21–50, 2008.

33
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Fate and Transport of Antibiotics from Pig Farms along the Bang Pakong River, Thailand

Rathborey Chan 1,2,3*, Ratboren Chan1,2, Sirinthrar Wandee2, Manna Wang3, Wilai Chiemchaisri2, Chart
Chiemchaisri2 and Chihiro Yoshimura3

1
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Department of Environmental Engineering, Faculty of Engineering, Kasetsart University,
Bangkhen, Bangkok, 10900 Thailand
3
Department of Civil and Environmental Engineering, Tokyo Institute of Technology, Ookayama,
Meguro-ku, Tokyo 152-8552, Japan
*Corresponding author: chanrathborey@itc.edu.kh

Abstract

This study was investigated the impact of different season (dry and rainy seasons) in 2018 on the fate and transport of
antibiotics drained from pig farms along the Bang Pakong River, Thailand. Twenty-one residual and antibiotics were detected
in pig farm effluent, drainages, and the river in both seasons. Amoxicillin, tetracycline, chlortetracycline, and tiamulin were
detected as dominant compounds in pig farm effluent, showing detection ratio of 83–100%. Maximum residual concentrations
of those antibiotics were 0.12–5.20 in the dry season and 0.12–16.10 µg/L in the rainy season. Tetracycline (15.1 µg/L) and
chlortetracycline (16.1 µg/L) were found at the highest concentration in the rainy season. We the applied multiple linear
regression and random forest to residual antibiotics from pig farm effluents and the river in both seasons in order to model
dilution and removal processes of antibiotics drained in river. Beside dilution, removal processes of antibiotics were defined
by their partition coefficients, solubility, biodegradability, and volatility. Compared to multiple linear regression, random
forest models better helped us to identifying the removal processes of pig farm antibiotics drained to river (R2: 0.77 in the dry
season, 0.35 in the rainy season).

Keywords: Residual Antibiotics, Pig Farm, Random Forest, Molecular Property

I. Introduction pattern of those compounds are varied across regions and

Pig farms in Thailand were developed in the 1960s countries worldwide. Due to low adsorption of antibiotics in
when the first commercial pig feeds were imported from the pig’s gut, 30%–90% of antibiotics were estimated to excrete
United Kingdom and the United Stage. Antimicrobial drugs to the environment through manure and urine. Then, those
for prevention pig diseases and growth promoter was take compounds consisting in wastes could then enter the stream
into consideration in this development. Common antibiotics or river via runoff or direct pig farm effluent discharged.
uses in pig farms were in classes of β-lactam, tetracyclines, Additionally, residual antibiotics in the surface water, even
sulfonamides, lincosamides, macrolides, and quinolones1. at the low concentration, could also exert selective pressure
Those antibiotics are given to pigs through feeds, drinking to the bacterial population to acquire antibiotic resistant,
water, and others as injection. But dosages and consumption which is more harmful to human and animal.

34
 When antibiotics reached river from pig farms and other
sources, their removal processes during transport along the
river depended on dilution and their molecular properties 2.
The rate disappearance of antibiotics through dilution was
clearly shown to associate with seasonal variation (i.e., dry
and rainy season) and movement of antibiotics drained along
the river. In contrast, the information about all significant
molecular properties that could affect removal processes of
antibiotics drained along the river have never mentioned in
one study area. Moreover, statistical methods that can make
more evident the relationship between one response variable
and two or more explanatory variables are rarely used. Since
it is currently not possible to quantitatively describe removal
processes of antibiotics, we aim to study impact of seasonal Fig. 1. The location of sampling sites along the Bang
variation on fate and transport of pig farm antibiotics drained Pakong River (a) and the flow directions of pig
long the river. Multiple linear regressing and random forest farm effluent, drainages and the river (b)
models were applied to residual antibiotics from the pig farm
effluent and river in both seasons to model the dilution and 2.3. Modelling fate and transport of antibiotics
their molecular properties, which possibly effect removal The study area was divided into four zones (Fig.1b).
processes of antibiotics drained in the river. Multiple linear regressing and random forest models were
applied to residual antibiotics from the pig farm effluent and
II. Materials and Methods the river in both seasons to model the dilution and their
molecular properties i.e., biodegradability (Biowin1 and 2),
2.1. Study area and sample collection partition coefficient (LogKow), solubility (LogSol), and
The site investigation was conducted at Chachoengsao volatility (LogVol), which possibly effect removal processes
Province, which is in the eastern Thailand and known as one of antibiotics drained in the river. All calculation methods
of intensive pig production areas. Water consumed in most and modelling processes were detailed in our current paper
pig farms is commonly Bang Pakong River water. Typical publication 4.
wastewater treatment used in pig farms is lagoon treatment
in series, but additional anaerobic digestion is mostly found III. Results and Discussion
at large farms. Fig.1a shows all sampling locations and flow
direction of six farm (F1-F6), nine drainage water sampling 3.1. Residual antibiotics
location (D1-D9), and nine river sampling locations (R1-R9). In small farms, the major residual compounds in pig farm
effluent were TC, CTC, AMX, TIA, and FCC. In addition,
Pig farm F2, F4, and F6 are classified into small farms while
those dominant compounds were also detected in effluent
F1, F3, and F5 are medium farms. from medium farms while FCC was found at the highest
2.2. Antibiotic determination concentration with 272.5 µg/L at Farm F5. The maximum
Pig farm effluents, drainage water, river water, and pig concentration of AMX was 32.5 µg/L while maximum
feed samples were analyzed for twenty-one antibiotics i.e., concentrations of TE and CTC were 20.4 µg/L, 11.2 µg/L,
penicillin G (PEN G), ampicillin (AMP), amoxicillin respectively. Furthermore, the maximum concentration of
(AMX), neomycin (NEO), gentamicin (GEN), tetracycline TIA was 11.7 µg/L while others were detected at the level
lower than 1.9 µg/L. TC, CTC and AMX were confirmed in
(TE), chlortetracycline (CTC), doxycycline (DOX),
pig feeds with maximum concentrations of 1.7 mg/kg, 8.3
oxytetracycline (OXY), erythromycin (ERY), tylosin (TYL), mg/kg, and 0.53 mg/kg. Other antibiotics (i.e., TMP, FCC,
timicosin (TIL), norfloxacin (NOR), enrofloxacin (ENR), OXY, NEO, TIA, NOR, DOX, and SDZ) also presented in
sulfadiazine (SDZ), trimethoprim (TMP), sulfamethoxazole the pig feed, showing residual concentrations of 0.0003–
(SXZ), colistin (CLS), lincomycin (LIN), tiamulin (TIA), 0.033 mg/kg. The results revealed that most antibiotics in
and florfenicol (FCC). The extraction method was followed the effluent were residual compounds from the pig feed.
the EPA 20073, whereas their concentration were detected by The concentrations of antibiotics detected in drainages
and the river in dry and rainy seasons are shown in Fig. 2. In
using LC-MS/MS.

35
drainages, AMX was detected at most of the sampling sites Overall, AMX, TC, CTC and TIA were the dominant
with the maximum concentration of 0.12 µg/L in the dry antibiotics in pig farm effluents, which were also dominant
season while it was detected only at the site D3 and D9 in in drainages and the river except for TMP and SDZ. Thus,
the rainy season, showing the maximum concentration of the antibiotic contamination in the aquatic environment is
0.71 µg/L. TE and CTC were also a dominant compounds in likely caused by the residual antibiotics from the pig farm.
drainage water. The maximum concentration TE was 0.063 Furthermore, the application of the pig manure to farmland
µg/L in the dry season and 15.1 µg/L in the rainy season as organic fertilizer should be another source for residual
while CTC was 5.2 µg/L and 16.1 µg/L. TIA presented at all antibiotics in the aquatic environment 5. Residual antibiotics
sampling sites, but its maximum concentration decreased in effluent may be firstly disposed to the drainage system
from 0.23 µg/L in the dry season to 0.12 µg/L in the rainy and then further distributed to the river. Likewise, SDZ and
season. Furthermore, TMP and SDZ were detected at most TMP in this study were investigated as dominant antibiotics
of the sampling sites in the dry season (max. 0.14 µg/L), but used in urban areas. However, the rainy season and surface
it only detected at the site D1, D3 and D9 in the rainy season runoff in October (176 mm rainfall/month) should dilute
(max.4.42 µg/L). Other antibiotic compounds were detected residual antibiotics much more than dry season (54 mm
of lower than 0.05 µg/L except for LIN at the site D9 in the rainfall/month), higher concentrations of tetracyclines and
dry season (max. 2.3 µg/L). All dominant antibiotics in sulfonamides in drainages and river water were still detected
drainages also presented at most of sampling sites in the in rainy season, indicating the higher release of antibiotics
river. AMX was detected at most of sampling sites in the dry from wastewater sources including pig farms at that period.
season with the maximum concentration of 0.12 µg/L, but it
was only detected at the site R3 in the rainy season (max. 3.2. Modelling fate and transport of antibiotics
0.02 µg/L). TE and CTC were detected at higher levels in In the dry season, dilution factor (D) in Zone 1 was
the rainy season with maximum concentrations of 4.7 µg/L estimated to be 217.9 while D factors in Zone 2 were 8.8 in
and 1.1 µg/L, compared to those of 0.14 µg/L and 0.18 µg/L the west flow and 1.4 in the south flow. In addition, D factor
in the dry season. TIA was detected at all sampling sites in Zone 3 was 0.51, whereas Zone 4 showed a negative D
(max. 0.003 µg/L). Moreover, maximum concentration of factor. The negative D factor in zone, where is located at the
sulfonamides (TMP and SDZ) increase form 0.087 µg/L in urban areas, might be caused by the release of wastewater
the dry season to 10.32 µg/L in the rainy season. Other contamination from many sources including pig farms to the
antibiotics were also detected at some sampling sites, river. In addition, the hydraulic retention time of the river
showing their maximum concentration of 0.17 µg/L in the water was short in this season. Thus, those processes were
dry season and 0.0036 µg/L in the rainy season. the possible cause for the high fluctuation of EC (954 µS/cm)
at Site R7. In the rainy season, D factor in all targeted zones
varied from 5.83 to 187, respectively. Most D factors were
higher in most targeted zones in rainy season than those in
the dry season except for the Zone 1 due to rich rainfall
precipitation in the rainy season.
For the MLR model in the dry season, Biowin2 showed
the highest t-value of 2.8, followed by LogVol (t-value:
1.58). The percentage of variance explained by this model
was 19.2% (overall R2: 0.17, residual standard error: 22.4 on
DF of 69, and F-statistic: 7.1). Likewise, the p-value of
Biowin2 was lower than 0.05 while the p-value of LogVol
was only 0.1. Identically, Biowin2 was the most relevant in
the model, corresponding to 10.7% of total contribution. For
the MLR model in the rainy season, t-values of all variables
were in acceptable range (t-value>1.6). The percentage of
variance explained by this model was 19.8% (overall R2: 0.2,
residual standard error: 10.1 on DF 74, and F-statistic: 3.6).
Variables shown percentage of relative important ranging
from 1.5% to 6.1%.
For the RF model in the dry season, the percentage of
variance explained by the model was 92.5% with the mean
Fig. 2. The concentrations of antibiotic classes detected at squared residuals of 16.6. The highest percentage decrease
each sampling site in the drainage and the river dry accuracy (%IncMSE) was presented by Biwon2 following
and rainy seasons by LogKow, LogVol, LogSol, and Biowin1. In the rainy

36
season, the percentage of variance explained by the model 20th anniversary KWEF commemorative research project.
was 92.4% with the mean squared residuals of 3.6. In both The study was also carried out under the scheme of
seasons, all variables presented significance in the models, collaborative research (CR) and PhD Sandwich Program of
showing %IncMSE of 3.8%–24.7%. AUN/SEED-Net that is supported by the Japan International
Based on AIC, RF models in both seasons indicated the Cooperation Agency (JICA).
lower prediction error than MLR models. Fig. 3 shows the
plots of actual and modelled concentrations in dry and rainy
seasons. The coefficient of determination of MLR and RF
models accounted for 0.06 and 0.14, respectively in the dry
season while they were 0.43 and 0.51 in the rainy season. RF
models showed better fitness between modelled and actual
concentrations than MLR models. Lower %IncMSEs of all
variables in the RF model in the dry season may imply the
low effluent flux from pig farms due to no hydraulic
overload in the retention lagoon. Moreover, due to the longer
hydraulic retention in the drainages system in this season,
residual antibiotics were not transported well from drainages
to the river.
Overall, major removal processes of residual antibiotics
in this area during dry and rainy seasons were determined by
their partition coefficients, solubilities, biodegradability and Fig. 3. Comparison of multiple linear regression (MLR) and
volatility. Partition coefficient and solubilities of antibiotics random forest (RF) models performance on test data
could define their adsorption affinity to suspend particles or
in dry and rainy seasons
sediment in the river6. Antibiotics with low LogKow and high
solubility are defined to low adsorption affinity while those
with the high LogKow and the low solubility are defined to References
high adsorption affinity. Likewise, antibiotics with Biowin1 [1] Lekagul, A., Tangcharoensathien, V., Mills, A., Rushton,
and Biowin2 values of higher than 0.5 are defined as high J. and Yeung, S., 2020. How antibiotics are used in pig
biodegradation in surface water. In this case, AMX may be farming: a mixed-methods study of pig farmers, feed
highly removed in the river water by biodegradation because mills and veterinarians in Thailand. BMJ Global Health,
its Biowin1 and Biowin 2 values are above 0.5. Likewise, 5(2), p.e001918.
substantial removal of TIA drained along the river may be [2] Selvam, A., Kwok, K., Chen, Y., Cheung, A., Leung,
occurred by adsorption because its LogKow and solubility are K.S. and Wong, J.W., 2017. Influence of livestock
4.75 and 0.6 mg/L. TE, CTC, SDZ, and TMP, were still high activities on residue antibiotic levels of rivers in Hong
in drainages and the river because both antibiotics are more Kong. Environmental Science and Pollution Research,
stable and persistence in biotransformation. Moreover, TE 24(10), pp.9058-9066.
and CTC persistence in aquatic environment may be also [3] EPA., 2007. Method 1694 : Pharmaceuticals and
associated with low volatility in water. Personal Care Products in Water , Soil , Sediment , and
Biosolids by HPLC / MS / MS. EPA Method.
IV. Conclusion [4] Chan, R., Wandee, S., Wang, M., Chiemchaisri, W.,
We investigated the fate and transport of antibiotics Chiemchaisri, C. and Yoshimura, C., 2020. Fate,
along Bang Pakong River and found that residual antibiotics transport and ecological risk of antibiotics from pig
in the river are discharged from pig farms especially in the farms along the bang pakong River, Thailand.
rainy season. In this target area, TE, CTC and TIA were Agriculture, Ecosystems & Environment, 304,
dominant antibiotics in the pig farm effluent and were p.1071235.
determined as major antibiotics in the drainages and the river. [5] Kemper, N., 2008. Veterinary antibiotics in the aquatic
In addition, the removal processes of antibiotics during such and terrestrial environment. Ecological indicators, 8(1),
transportation to the river were modeled as functions of their pp.1-13.
partition coefficients, their solubilities, biodegradability and [6] Boy-Roura, M., Mas-Pla, J., Petrovic, M., Gros, M.,
volatility. Soler, D., Brusi, D. and Menció, A., 2018. Towards the
understanding of antibiotic occurrence and transport in
Acknowledgement groundwater: Findings from the Baix Fluvià alluvial
aquifer (NE Catalonia, Spain). Science of the total
Authors are grateful to Kurita Water and Environment
environment, 612, pp.1387-1406.
Foundation (KWEF), Japan for funding this research under

37
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Microcrystalline Cellulose Production from Acid Hydrolysis

of Rice Straw’s Hydrotropic Pulp

Frita Dewi Damayanti 1, Wahyudi Budi Sediawan 1,*, Muhammad Mufti Azis 1, Indah Hartati 1, 2

1
Department of Chemical Engineering, Faculty of Engineering, Universitas Gadjah Mada, 55281 Indonesia
2
Department of Chemical Engineering, Faculty of Engineering, Universitas Wahid Hasyim, 50236, Indonesia
*Corresponding author: wbsediawan@ugm.ac.id

Abstract
As one of the largest rice producers in Asia, Indonesia has huge of biomass resources such as rice straw. The objective of
this work was to find the highest cellulose content from hydrolysis of rice straw’s hydrotropic pulp with hydrochloric acid
and sulfuric acid. The rice straw was successively pretreated with hydrotropic delignification and bleaching before hydrolysis.
The hydrotropic delignification and bleaching were able to remove significant amount of lignin, hemicellulose, and other
extractive substances. In this study, the hydrotropic delignification was performed at 80 oC for 60 minutes with urea solution
30% and bleaching was performed at 70oC for 60 minutes with hydrogen peroxide. The hydrolysis of rice straw hydrotropic
pulp was carried out in a reactor under microwave irradiation, where the hydrolysis reaction times (10-50 minutes) were
examined. The result shows that the highest cellulose content found at the optimum of observed conditions (2 M, 80oC, 30
minutes) for hydrochloric acid and sulfuric acid hydrolysis were 64.44% and 56.93% respectively.

Keywords: hydrolysis, hydrotropic pulp, microcrystal cellulose, rice straw

I. Introduction hemicellulose, and lignin which is contains 35-50%, 20-35%,

and 10-25% respectively [2] .Conversion of rice straw is a
Nowadays, the research of sustainable and
very complicated process due to the presence of complex
environmentally friendly processes involves the use of
structure of lignin and hemicelluloses with cellulose. A
alternatives lignocellulosic materials. Indonesia is an
pretreatment process is required as it can remove significant
agricultural country where most of the population has a
amount of lignin, hemicellulose, and other extractive
livelihood in agriculture or farming. As one of the largest
substances [3].
rice producers in Asia, Indonesia has huge amount of
There are three important processes for MCC production,
biomass resources such as rice straw. Rice straw is the
namely the delignification, bleaching, and hydrolysis
residue of rice processing that has not been fully utilized.
process. Pretreatments are required for increasing
Most of the rice straw was burnt on site, generating large
accessibility of surface area of cellulose for reducing sugar
amounts of heavy smoke as well as carbon dioxide and
production by removing components such as hemicellulose
resulting in environmental pollution [1]. Lignocellulosic
and lignin from the complex structure. Many researches
materials are abundant, cheap, and renewable and may be
have been carried out to develop an environmentally
used for the production of microcrystalline cellulose (MCC).
friendly chemical pretreatment process such as the use of
MCC has various applications in pharmaceutical, chemical,
hydrotropic solution on delignification process. Urea
food industries, etc. It is meaningful to use rice straw as raw
solution is considered as a green chemical for lignin removal
material for MCC production. Rice straw is mainly
due to its non-corrosiveness, its lower risk to human health,
composed of three major polymers like cellulose,

38
and less impact on other lignocellulose component [4]. rice straw powder and 1200 ml 30% urea solution were put
Bleaching is a process using chemical substances to remove into 1500 ml flask. The mixture was then heated at 80oC for
stains or dirt attached to the material. One of the commonly 60 minutes in a microwave heater. The mixture was then
used as a bleaching agent is hydrogen peroxide. Hydrogen filtered. The solid residue collected was washed thoroughly
peroxide is an oxidizing agent that can be used as an by distilled water until became neutral and then dried at 80oC
environmentally friendly pulp bleach [5]. for 6 h. Finally, the rice straw’s hydrotropic pulp was
The hydrolysis reaction in dilute acid is very complex, weighed to determine the yield.
mainly because the substrate is in a solid phase and the Bleaching of Rice Straw’s Hydrotropic Pulp. Twenty
catalyst in a liquid phase. Various factors (particle size, grams of rice straw’s hydrotropic pulp was put into a flask,
liquid to solid ratio, type and concentration of acid used, and mixed with hydrogen peroxide (5% concentration) and
temperature, and reaction time) influence monomer yield [6]. distilled water to obtain a solid to liquid ratio of 1:10. The
Hydrochloric and sulfuric acids have been used to catalyst bleaching process is carried out in alkaline condition (pH =
the hydrolysis of rice straw. For other raw materials 11) by addition of aqueous sodium hydroxide (2%). The
hydrofluoric and acetic acids were also employed as a mixture was then heated in a microwave at 70oC for 60
catalyst. minutes. After that, the mixture was filtered and washed
Microwave irradiation has been widely used in many using distilled water until the filtrate became neutral. The
areas because of its high heating efficiency and easy solid obtained was then dried in an oven until constant
operation [7]. Microwave heats the target object directly by weight.
applying an electromagnetic field to dielectric molecules as Acid Hydrolysis of Rice Straw’s Hydrotropic Pulp.
compared to conduction or convection heating which is Samples of 10 g bleached pulp were mixed with 200 ml
based on intramolecular heat transfer. It can significantly solutions of acid (sulfuric and hydrochloric acid). For this
increase the hydrolysis of rice straw. Several researcher have study, both sulfuric and hydrochloric acid concentrations of
used microwave as a potential method for treatment of 2 M were applied. The hydrolysis was carried out at 80oC in
various lignocellulosic materials and to damage the a microwave oven with temperature control. To prevent
recalcitrant lignin [3]. water loss by evaporation, the reactor was equipped with a
The objective of this work is to study the hydrolysis of reflux system. Mild agitation was also applied. On
rice straw’s hydrotropic pulp with hydrochloric acid and completion of the hydrolysis reaction (10, 20, 30, 40, 50
sulfuric acid to find the highest cellulose content. minutes), the reactor contents were filtered and washed
thoroughly by distilled water until the residue became
II. Materials and Methods neutral. The samples were dried at 80oC for 6 h before
further analysis.
2.1. Raw Materials
Rice straw was obtained from local farm around Table 1. Chemical composition of rice straw using Chesson
Yogyakarta city (Yogyakarta, Indonesia). Other reagents Method
were purchased from Merck Chemicals and Life Sciences
(Jakarta, Indonesia); they were analytically pure and applied Number Components Values (%)
without pretreatment. 1 Cellulose 37.56
The rice straw was cut into pieces of size of 1-2 cm and 2 Hemicellulose 35.37
washed with water to remove dirty and aqueous soluble 3 Lignin 14.45
substances. The washed rice straw was then dried under 4 Ash 11.29
sunlight for 2 days. The rice straw was then crushed until it
becomes a powder. The powder was sieved using a 60 Mesh 2.3. Analysis
sieve, and the undersize was dried. The chemical The contents of cellulose, hemicellulose, lignin, and ash
compositions of rice straw before hydrolysis are presented were determined by using Chesson analysis method [8]. One
in Table 1. g of dry sample (a) was added with 150 ml of aquadest, then
refluxed at 100oC in water bath for 1 h. The resultant
2.2. Isolation of Cellulose from Rice Straw solution was filtered, and the residue was washed with hot
Hydrotropic Treatment of Rice Straw. Forty grams of water (300 ml). The residue was then dried in an oven until

39
constant weight (b). After that, the residue was added with increasing reaction time in the range from 10 to 50 minutes.
150 ml of 1 N H2SO4 then refluxed in a water bath at the This is probably due to the excessive hydrolysis of cellulose
temperature of 100oC for 1 h. The obtained solids were to glucose.
filtered and washed with aquadest (300 ml) to neutral Table 2 shows that rice straw before hydrolysis
condition, dried and weighed (c). The dried residue was then contained 47.42% of cellulose and 10.89% of lignin.
added with 10 ml of 72% H2SO4 and soaked at room Sulfuric acid hydrolysis for 10 minutes was the lowest for
temperature for 4 h. The residue was then mixed with 150 cellulose content and the highest of lignin content among
ml of 1 N H2SO4 and refluxed on a water bath for 1 h. After others. Hydrolysis for 10 minutes has been able to increase
that, the residue was filtered and washed with aquadest (400 cellulose by 13.01% and to decrease the lignin content by
ml) until neutral condition and then heated in an oven at 2.38%. Meanwhile, for 30 minutes of hydrolysis, the
temperature of 105oC until constant weight. The solid cellulose became 56.93% or increased 20.05% and lignin
obtained was weighed (d). Finally, the solid residue was decreased by 20.57%. Hydrochloric acid hydrolysis for 10
ashes in a furnace, and the ash was weighed (e). minutes was the lowest for cellulose content and the highest
The composition of the pulp was calculated by: lignin content among others. Hydrolysis for 10 minutes has
𝑏−𝑐 been able to increase cellulose by 14.15% and to decrease
𝐻𝑒𝑚𝑖𝑐𝑒𝑙𝑙𝑢𝑙𝑜𝑠𝑒 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (%) = × 100 (1)
𝑎 the lignin content by 3.86%. Meanwhile, for 30 minutes of
𝑐−𝑑 hydrolysis, the cellulose became 64.44% or increased 35.89
𝐶𝑒𝑙𝑙𝑢𝑙𝑜𝑠𝑒 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (%) = × 100 (2)
𝑎 % and lignin decreased by 20.47%. For 40 – 50 minutes,
𝑑−𝑒 both sulfuric and hydrochloric acid, the cellulose content
𝐿𝑖𝑔𝑛𝑖𝑛 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (%) = × 100 (3)
𝑎 decreased with reaction time. From this data, it can be seen
𝑒
𝐴𝑠ℎ 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (%) = × 100 (4) that if the cellulose content increased higher, the lignin
𝑎
The yields of MCC from rice straw were calculated content will also decreased more. This may be due to
according to equation (Eq. 5) as follows: production of sugar. The level and composition of the sugar
𝑚′ released depend on the type of the acid in the reaction
𝑌𝑖𝑒𝑙𝑑 (%) = × 100 (5) mixture. Acid treatment with sulfuric acid mainly releases
𝑚𝑜
where 𝑚′ and 𝑚𝑜 are the mass of product and raw xylose [9]. It can also be seen that the using of hydrochloric
material of the hydrolysis process, respectively. acid was more effective to increase cellulose content than
the using of sulfuric acids for hydrolysis of rice straw.
This study was compared to the traditional heating
III. Results and Discussion
method by Fan, G [7], especially sharp decrease in treatment
For acid hydrolysis, both sulfuric and hydrochloric acid time. The time of delignification and acid treatment
concentrations were 2 M, the temperature of reaction was decreased from 12 h to 60 minutes and 5 h to 30 minutes,
80oC, and reaction time was varied from 10 to 50 minutes. respectively. Thus it is reasonable to conclude that
The cellulose and lignin content obtained from the solid microwave assist is an effective methodology to prepare
phase at various reaction time are presented in Table 2. MCC from rice straw as raw material.
Based on Table 2, the yield of MCC slightly decreased Characterizations of rice straw before and after
from 88.60 to 82.25% for sulfuric acid hydrolysis and hydrolysis have been examined by applied X-ray Diffraction
decreased from 82.96 to 78.36% for hydrochloric acid (XRD) analysis technique. The result is presented in Fig.1.
hydrolysis. The MCC yield decreased steadily with The XRD pattern of rice straw before hydrolysis, after

Table 2. Acid Hydrolysis of Rice Straw

Sulfuric Acid Hydrochloric Acid

Time Cellulose content Lignin content Yield Cellulose content Lignin content Yield
Entry
(min) (wt %) (wt %) (%) (wt %) (wt %) (%)
1 0 47.42 10.89 100 47.42 10.89 100
2 10 53.59 10.63 88.60 54.13 10.47 82.96
3 20 54.22 10.61 88.31 61.08 9.70 81.42
4 30 56.93 8.65 85.95 64.44 8.66 80.36
5 40 54.29 8.32 84.11 63.95 8.55 79.93
6 50 53.35 7.28 82.25 63.42 7.26 78.36
40
hydrolysis with sulfuric acid, and after hydrolysis with which support this work through PDUPT research grant of
hydrochloric acid are shown in Fig. 1. The peak around 2020, with contract number of 2845/UN1/DITLIT/DIT-
22.3o is contributed to the typical crystal lattice of cellulose LIT/LT/2020.
Iβ [10]. The shoulder peak at 15.8o is also assigned to the
typical crystal lattice of cellulose Iα [11]. The diffraction References
peak at 22.3o is wide and round in the curve of rice straw [1] J. Domínguez-Robles, R. Sánchez, P. Díaz-Carrasco, E.
before hydrolysis (Fig. 1-C) but becomes relatively sharper Espinosa, M. T. García-Domínguez, and A. Rodríguez,
and narrower after hydrolysis (Fig. 1-D1 and 1-D2), 2017. Isolation and characterization of lignins from
indicating the removal of amorphous components such as wheat straw: Application as binder in lithium batteries.
lignin, hemicellulose, and silica. It is well known that Int. J. Biol. Macromol., vol. 104, 909–918.
[2] B. C. Saha, 2004. Lignocellulose Biodegradation and
cellulose consist of both crystalline and amorphous regions.
Applications in Biotechnology. U.S Gov. Work. Am.
In biomass, hemicellulose and lignin are amorphous in Chem. Soc., 2–34.
nature while cellulose is crystalline [12]. The sharper [3] R. Singh, S. Tiwari, M. Srivastava, and A. Shukla, 2014.
diffraction peak indicates higher crystallinity degree. The Microwave Assisted Alkali Pretreatment of Rice Straw
crystallinity increases from 53.75% to 65.32% in hydrolysis for Enhancing Enzymatic Digestibility, vol. 2014.
with sulfuric acid and 63.46% in hydrolysis with [4] M. Pulcer, 2006. Displacement Washing of Pulp with
hydrochloric acid. The result showed that removal of Urea Solutions, vol. 60, 365–370.
[5] A. Tutus, 2004. Bleaching Rice straw with hydrogen
amorphous parts of rice straw, that is lignin and
Peroxide, 1327–1329.
hemicellulose, was more in sample treated with sulfuric acid [6] M. J. Taherzadeh, C. Niklasson, and G. Lidn, 1997.
than hydrochloric acid. Acetic acid friend or foe in anaerobic batch conversion
30 of glucose to ethanol by Saccharomyces cerevisiae, vol.
22.3 C 52, no. 15.
25
D1 [7] S. Zhu, Y. Wu, Z. Yu, J. Liao, and Y. Zhang, 2005.
Intensity (counts)

D2 Pretreatment by microwave/alkali of rice straw and its

20
enzymic hydrolysis, Process Biochem., vol. 40, no. 9,
15 3082–3086.
15.8 [8] A. Fermentation and L. Yield, 1981. Acidogenic
10
Fermentation of Lignocellulose-Acid Yield and
5
Conversion of Components, vol. XXIII, 2167–2170.
[9] C. J. Israilides, G. A. Grant, and Y. W. Han, 1978. Sugar
0 level, fermentability, and acceptability of straw treated
10 20 30 40 50 60 70 with different acids, Appl. Environ. Microbiol., vol. 36,
2-theta (deg) no. 1, 43–46.
[10] M. L. Nelson and R. T. O’Connor, 1964. Relation of
Fig. 1 XRD pattern of rice straw before hydrolysis (C), Certain Infrared Bands to Cellulose Crystallinity and
after hydrolysis with sulfuric acid (D1), and after Crystal Lattice Type . Part II . A New Infrared Ratio for
hydrolysis with hydrochloric acid (D2) Estimation of Crystallinity in Celluloses I and II *, J.
Appl. Polym. Sci., vol. 8, 1325–1341.
[11] H. S. Barud et al., 2008. Thermal behavior of cellulose
V. Conclusion
acetate produced from homogeneous acetylation of
Optimum result for hydrolysis of rice straw was obtained bacterial cellulose, Thermochim. Acta, vol. 471, no. 1–
by using hydrochloric acid of 2 M for 30 minutes. The 2, 61–69.
[12] J. P. O’Dwyer, L. Zhu, C. B. Granda, and M. T.
percentage of cellulose content obtained after hydrolysis
Holtzapple, 2007. Enzymatic hydrolysis of lime-
using hydrochloric acid is higher (64.44%) than the one pretreated corn stover and investigation of the HCH-1
using sulfuric acid (56.93%). Model: Inhibition pattern, degree of inhibition, validity
of simplified HCH-1 Model, Bioresour. Technol., vol.
Acknowledgement 98, no. 16, 2969–2977.

The authors are grateful to The Ministry of Research,

Technology and Higher Education of Republic of Indonesia

41
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Effect of Temperature and Airflow Rates on Design Parameters for Drying of Natural
Tannin Extracts From Rhizophora apiculata Bark Waste

Edia Rahayuningsih 1*, Aswati Mindaryani 1, Mukmin Sapto Pamungkas 1, Dwi Tyaningsih Adriyanti 2, Chandra
Wahyu Purnomo 1, and Joshua Bagaskara 1

1
Department of Chemical Engineering, Faculty of Engineering, Universitas Gadjah Mada,
55381 Sleman, Yogyakarta, Indonesia
2
Faculty of Forestry, Universitas Gadjah Mada, Yogyakarta, Indonesia
*edia_rahayu@ugm.ac.id

Abstract

PT Bintuni Utama Murni Wood Industries (PT BUMWI) produces bark waste for the Rhizophora apiculata type in the
form of chips (small pieces) reaching 250-500 tons per month. Based on previous research results, the bark has a natural dye
content of 0.2101 grams per gram of dry solid with a tannin content of up to 0.0531 grams per gram of solid. In this study, the
bark waste's concentrated extract was dried in an open system using a hotplate on a metal plate with airflow. The drying
temperature was varied at 60, 70, and 80 oC, and the airflow rates was adjusted to variations of 1, 2, and 3 m/s. Drying was
carried out for the extract thickness of 1 mm. The mass of the material was weighed every 2 minutes. The optimum drying
was found by using response surface methodology (RSM) and obtained at a temperature of 80 ℃, airflow rates 1.75 m/s with
drying design parameters were mass transfer coefficient (Kya) at 65.77 kg H2O m-2 hour -1∆y-1, a constant drying rate (Rc) at
9.01 kg H2O hour-1 m-2, and critical time (tc) for 0.5 hours. This optimization can be used to design appropriate technology for
processing bark waste as a source of natural dyes on a more massive scale.

Keywords: Rhizophora apiculata, natural dye, drying design parameters

I. Introduction evaporation, and drying processes [2].

This research focuses on the drying process of the
PT Bintuni Utama Murni Wood (PT BUMWI) produces
concentrated extract. The drying system commonly used is
bark waste from the Rhizophora apiculata type in chips
energy from sunlight. Drying with this method is done by
(small pieces) reaching 250-500 tons per month. In addition,
contacting the material directly under the sun or indirectly.
based on the results of previous identification, PT BUMWI's
The resulting water vapor is removed from the drying area
Rhizophora apiculata bark has a natural dye content of
through an airstream by free or forced convection [3].
0.2101 grams per gram of solid with a tannin content of up
The drying process needs to be studied to obtain a dye
to 0.0531 grams per gram of solid. Tannin is a natural dye in
powder that is more applicable and more efficient. The
textiles that can give a yellow-brown colour [1]. The tannin
experimental design is carried out with an appropriate
content in bark waste is relatively high and can be used as a
technology reference frame to suit conditions in a remote
potential source of mordant dyes. Tannin powder can be
area. Therefore, this present research was conducted to
produced from the bark by extraction, followed by

42
obtain the optimum drying process by varying airflow rates Where, Rc is constant drying rate in kg H2O hour-1 m-2,
and heating temperatures. Ls is mass of the sample in kg, A is effective drying surface
area in m2, xi+1 is water content at t = t+1, xi is water content
II. Materials and Methods at t = t, ∆t is the time interval in an hour, Kya is overall mass
transfer coefficient in kg H2O m-2 hour -1∆y-1, y’s is moisture
2.1. Materials
saturates on the surface of the material at wet bulb
The material used in this research was concentrated
temperature, y is drying gas humidity in kg H2O per kg dry
extract (ρ = 1.084 g/mL). This material was obtained from
solids, tc is constant drying time in an hour, and ∆x is change
the extraction of Rhizophora apiculata bark powder from PT
in water content in solids.
BUMWI then evaporated up to 8x concentrations.
Optimization of the drying process from each
independent variable (X; T and v) on the dependent variable
2.2. Methods
(Y; Kya and Rc) was carried out using Response Surface
The drying tray with dimensions of 10 x 10 cm was dried
Methodology in Minitab 16 Statistical Software. The
in a 105 oC oven for 15 minutes. The tray was kept in the
experimental design model is a second-order polynomial
desiccator for 5 minutes. The empty tray was weighed on a
model, presented in the following equation [4].
digital analytical balance. A total volume of 10 mL of
k k
concentrated extract was poured into the drying tray. The Yi = b0 + å bi X i + å bii X 2 + åå bij X i X j + e (Eq.5)
tray containing concentrated extract was weighed initially. i =1 i =1 i< j
Drying begins by placing the tray on the hotplate with
airflow through on it. The tray was weighed every 2 minutes III. Results and Discussion
for 1 hour. Optimization of the drying process was carried
out using a 2k factorial design shown in Table 1. In designing the appropriate technology, drying aims to
extend shelf life and facilitate handling during sales [5].
Table 1. Experimental Range and Level Design Dyes products in powder form will be easy to distribute and
easy to apply in their use. After conducting research on the
effect of variations in airflow rates and drying temperature
Independent Range and Level
Variable -1 0 +1 in the hotplate system, different results were obtained.
Temperature [T], oC 60 70 80 From Fig. 1 it can be seen that the higher the airflow
Airflow rate [v], m/s 1 2 3 rates, at the same time, the higher the sample moisture
content will be obtained. This result shows that the drying
2.3. Analytical methods process was affected by airflow rates. Fig. 1(b) and (c)
shows that drying at a temperature of 70 ºC and 80 oC
Unsteady state mass balance was applied to the drying
!" produces a moisture content that is relatively the same with
process and the rate of change in moisture content ( ) speed variations. The difference can be observed in the
!#
follows (Eq.1) gradient of the decrease in mass of the sample. The change
𝑑𝑥 (𝑥$%& − 𝑥$ ) in the sample's water content at a temperature of 80 ºC with
= (Eq.1)
𝑑𝑡 ∆𝑡 a speed of 2 m/s is the fastest.

The design parameters of the dryer were found by using

the numerical method was performed to obtain the constant
drying rate (Rc), mass transfer coefficient (Kya), and
constant drying time (tc) using the following equation:
𝐿𝑠 (𝑥$%& − 𝑥$ )
𝑅𝑐$→$%& = (Eq.2)
𝐴 ∆𝑡
𝑅𝑐
𝐾( 𝑎 = (Eq.3)
(𝑦 𝑠 − 𝑦′)
)

𝐿𝑠 ∆𝑥
𝑡𝑐 = (Eq.4)
𝐴 𝑅𝑐

43
 Fig. 2. Drying Rate vs. Time
Fig. 1. Sample Mass vs. Time
Based on Fig. 2, The drying rate profile has increased
The drying rate (-dx/dt) was determined by processing from 60 to 80 oC. It can be seen that the lowest water content
the sample's water content against time using the calculus produced at the end of drying occurs at drying 70 oC, an
method, dividing the graph of moisture content against time airflow rate of 1 m/s with a water content of 6.2 %. Airflow
into three parts. Part 1, which is the part that forms a straight rates affect the temperature distribution at the time of drying.
line, is carried out by linear regression until the maximum The drying rate increases for a speed of 1 m/s to 2 m/s.
correlation coefficient (R2) is obtained. Part 2 is the part that However, at the highest airflow rates, heat transferred to the
includes the curved line and is subjected to polynomial hotplate decrease.
regression. Part 3 is the one that forms a horizontal line, Airflow rates at 3 m/s cause the surface temperature to
which indicates that the drying process has stopped. The be the same as air flowed temperature so that the heat
relationship curve between the rate of reduction in water transfer for drying is reduced. The drying operation at a
content with time is presented in Fig. 2. temperature of 80 ℃ with a 2 m/s is the highest drying rate
achieved, which is 0.8858 grams/minute in 4 minutes.
A summary of the design parameters for drying is
presented in Table 2. It can be concluded that the mass
transfer coefficient and rate of drying value will increase
with increasing drying temperature at the same airflow rates.
High temperatures will provide more heat energy. When
viewed for variations in airflow rates, an increase in airflow
rates will encourage the fluid's tendency to flow turbulently.
With the turbulent flow, the film layer formed during the
mass transfer process will become thinner so that the mass

44
 transfer process (diffusion) between water compounds takes (Eq.6) and (Eq.7) with the coefficient of determination R2
place more intensely. However, in the drying process with a value at 97.51% and 90.98%.
hotplate system flowed by air, it will cause a decrease in the 𝐾( 𝑎 = 183.1 − 6.62 𝑇 + 75.3𝑣 + 0.06𝑇 * (Eq.6)
temperature on the hotplate surface at high speeds so that the −12.89 𝑣 * − 0.38 𝑇. 𝑣
heat energy for drying and Kya value will decrease. 𝑅𝑐 = 47.35 − 1.69𝑇 + 11.8𝑣 + 0.014𝑇 *
(Eq.7)
−2.11 𝑣 * − 0.051 𝑇. 𝑣
Table 2. Drying Design Parameter The optimum results were obtained at a temperature of
80 oC with airflow rates at 1.75 m/s. The maximum value of
[T], OC [v], m/s Rc Kya tc, hours Kya and Rc were 65.770 kg H2O m-2 hour -1∆y-1 and 9.013 kg
60 1 3.71 37.27 1.20 H2O hour-1 m-2. Critical time (tc) was obtained at 0.5 hours.
70 1 2.78 38.23 0.83
IV. Conclusion
80 1 7.97 59.40 0.60
60 2 4.81 48.32 1.20 The optimum condition for drying of concentrated
extract from Rhizophora apiculata bark occurred at a
70 2 6.69 52.28 0.50
temperature of 80 oC with an airflow rates of 1.75 m/s. The
80 2 8.79 65.46 0.50 drying design parameters were constant drying rate at 9.01
60 3 3.72 37.37 1.20 kg H2O hour-1 m-2, mass transfer coefficient (Kya) at 65.77
70 3 3.79 38.13 0.93 kg H2O m-2 hour -1∆y-1, and critical time (tc) for 0.5 hours.
80 3 5.96 44.37 0.50
Acknowledgment
Rc (kg H2O hour m ); Kya (kg H2O m hour ∆y-1)
-1 -2 -2 -1
We would like to thank The Direktorat Penelitian UGM
Optimization was carried out by Response Surface for their financial support in this research.
Methodology (RSM). Contour plot and optimization of Kya
and Rc value are shown in Fig. 1. References
[1] Park, C., 2007, A Dictionary of Environment and
Conservation 1st, Oxford University Press, New York.
[2] Rahayuningsih, E., Mindaryani, A., Adriyanti, D.T.,
Prihastiwi, L.P., Adina, P., Ayu, E. D., 2020, Conceptual
Design of a Process Plant for the Production of Natural
Dye from Merbau (Intsia bijuga) Bark, Regional
Symposium on Chemical Engineering (RSCE) 26th,
(a) Kuala Lumpur
[3] Earle, R. L., 2004, Unit Operations in Food Processing
(b)
Web. Edition, The New Zealand Institute of Food
Science & Technology (Inc.), Terrace End.
[4] Montgomery, D. C., Myers, R.H., Anderson, C.M.,
2012, Response Surface Methodology: Process and
Product Optimization Using Designed Experiments,
New York: John Wiley & Sons.
[5] Chowdury, K., Khan, S., Karim, R., Hasan, A., and
Obaid, M., 2011, Effect of Moisture, Water Activity and
Fig. 1. (a) Contour Plot (b) Optimization Curve Showing Packaging Materials on Quality and Shelf Life of Some
Locally Packed Chanachur, Bangladesh Journal of
The Effects of Independent Variables on Kya and Rc Value Scientific and Industrial Research 46, 33-40.

The model equations for Kya and Rc are presented in

45
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

The Assessment of Concentration of Particulate Matter Emitted from Vehicles in Phnom

Penh City

Rany You1, Chanto Chea1, Tomoki Nakayama2, Rithy Kan3, Chanmoly Or3, Mitsuhiko Hata4,
Masami Furuuchi4, Chanreaksmey Taing1,5, *

1
Faculty of Food and Chemical Engineering, Institute of Technology of Cambodia
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Environmental Science, Nagasaki University
1-14 Bunkyomachi, Nagasaki, 852-8521, Japan
3
Research and Innovation Center, Institute of Technology of Cambodia
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, Japan
Kakumamachi, Kanazawa, Ishikawa 920-1164, Japan
5
Research Unit Water and Environment, Institute of Technology of Cambodia
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* tsmey16@gmail.com
Abstract

Ambient air pollution is a mix of chemicals, particulate matter, and biological materials that react with each
other to form tiny hazardous particles. It causes breathing problems, chronic diseases, increased hospitalization,
and premature mortality. One of the air pollutant generators is traffic activities. Therefore, the purposes of this
study are to assess the concentration of Total Suspended Particulate (TSP), correlation of TSP with the number of
vehicles, and Particulate Matter (PM) fraction emitted from traffic along with the road sites in Phnom Penh city to
the atmosphere. The experiment was conducted in 4 sites including Kim Il Sung Blvd (KISB), Chroy Changvar
Cross-intersection (CCV), Stung Mean Chey Cross-intersection (SMC), and Russian Federation Blvd (RFB).
High-Volume air sampler was used for TSP measurement, Nano sampler for PM fraction, and Low-cost sensor for
PM2.5 detection. As the result, at four sampling sites the concentrations of TSP and PM2.5 reached the maximum
at 1072µg/m3 and 41.35±1.37µg/m3, 234.09µg/m3 and 28.33±1.79µg/m3, 151.27µg/m3 and 15.88±1.57µg/m3, and
161.64µg/m3 and 14.24±1.65µg/m3, respectively at 5-7 pm. The correlation between traffic and TSP at SMC has
r2=0.82, shows that traffic was the main source that contributes to the concentration of TSP. However, the traffic
movement was not merely a source of the pollutants at RFB (r2=0.03) while the other two sites, KISB and CCV
had the moderate effect with r2=0.56 and 0.64, respectively. The concentration of PM fraction was found the
highest in coarse particles size at KISB, SMC, and RFB while the concentration of PM0.1 reached the
maximum,116.79µg/m3 at CCV. Thus, this study indicates that the atmospheric particle concentration in Phnom
Penh city is affected by the air pollutants emitted from vehicles.

Keywords: Total suspended particulate, Particulate Matter, Low-cost sensor, PM2.5

46
I. Introduction together and multiplied by 6 to represent for 1 hour. TSP was
measured every 2-hour so the filter used with a high-volume
The development of Phnom Penh, the capital city of
air sampler (HV) model (SHIBATA HV-500F) must be
Cambodia, is particularly important. On the other hand, this
changed every 2-hour. Whereas Nano sampler (NS) model
urbanization invigorates people to immigrate from rural
KANOMAX 3180 is used for measuring the concentration
areas to urban. Due to this, the number of vehicles has been
of PM fraction. Low-Cost Sensor (LCS) was used to detect
increased and the congestion happened almost every part of
the amount of PM2.5.
the city [1]. Ambient air pollution is a mix of chemicals,
particulate matter, and biological materials that react with
2.3. Analytical methods
each other to form tiny hazardous particles called Particulate
The concentration of TSP and PM were calculated by the
Matter (PM), a mixture of solid, liquid, or solid and liquid
following formula:
particles suspended in the air [2]. It causes breathing
problems, chronic diseases, increased hospitalization, and 𝑚𝑓 − 𝑚𝑖
premature mortality. To simplify the assessment of PM 𝐶𝑇𝑆𝑃 = × 103 (Eq.1)
𝑉
levels and facilitate the implementation of PM pollution 𝑚𝑓 − 𝑚𝑖 (Eq.2)
𝐶𝑃𝑀 = × 106
control policies, it is common to categorize PM levels by the 𝑡×𝑞
aerodynamic diameter such as total suspended particulate Where 𝐶𝑇𝑆𝑃 , 𝐶𝑃𝑀 are the concentration of TSP and PM
(TSP) (PM<30μm), PM10 (PM<10μm), PM2.5 particularly (μg/m3), 𝑚𝑖 , 𝑚𝑓 are the mass filter before an
(PM<2.5μm), coarse particles (2.5μm<PM<10μm), and d after sampling (mg), 𝑉 is the total volume of an air s
ultrafine particles (UF) (PM<0.1μm) [3]. However, traffic ample shown on HV, (m 3), 𝑡 is sampling period (min),
movement produces road dust and air turbulence that can re- and 𝑞 is the flow rate (L/min).
entrain road dust [4], which are the dominant sources of air III. Results and Discussion
pollution. Therefore, the purposes of this study are to assess
the concentration of TSP, correlation of TSP with the number 3.1. Pollutants Concentration
of vehicles, and PM fraction emitted from traffic along with At KISB in the rush hours, 5-7pm, the mass
the road sites in Phnom Penh city to the atmosphere. concentration of TSP and PM2.5 rose to the peak at
II. Materials and Methods 1072μg/m3 and 41.35±1.37μg/m3, respectively as shown in
Fig. 2(a). Also, during 5-7 pm has released by far the
2.1. Sampling sites greatest mass concentration of TSP, approximately greater
The experiment of this project was conducted in Phnom than twice as much as during 9-11 am, which is the time that
Penh, the capital city of Cambodia. The four main cross- received the next largest amount of TSP concentration.
intersection such as Kim Il Sung Boulevard (KISB), Chroy Whereas Fig. 2(b) clearly shows that at CCV, the amount of
Changvar cross-intersection (CCV), Stung Mean Chey PM2.5 reached the peak of 22.79±1.28μg/m3 and
cross-intersection (SMC), and Russian Federation 28.32±1.79μg/m3 in rush hours, 7-9am and 5-7pm,
Boulevard (RFB) were selected as shown in Fig. 1. respectively while it tended to the minimum about
14.59±0.97μg/m3 during 1-3 pm. As represented in Fig. 2(c),
2.2. Method of Sampling the peaks of TSP concentration reached 151.27μg/m3 and
In each sampling site, the data has been collected for 2 148.89μg/m3 in the rush hours during 7-9am and 5-7pm,
days that must be included one weekday and one weekend individually whereas between 9am to 5pm, the
day. However, the duration of the experiment was 12 hours concentration of TSP seems a bit fluctuated. Moreover, the
per day for all instruments. Throughout the monitoring concentration of PM2.5 in the rush hours, 7-9 am and 5-7
period, traffic volume at every site described above has been pm, also reached the peak at 15.88±1.30μg/m3 and
recorded for 12 hours during the experiment. Since there 12.30±1.57μg/m3, respectively while the others remained in
were many vehicles, we cannot count directly while they a low concentration. Lastly, at RFB, during rush hours the
travel on the road, then we use the camera to record for concentration of TSP reached the highest at 161.64μg/m3
counting at another time. We recorded 5 minutes at the while PM2.5 also reached a maximum of 14.24±1.65μg/m3
beginning of time and 5 minutes of the end of time, then we as seen in Fig. 2(d).
added the number of vehicles from these two periods

47
 Figure1.Study area in Phnom Penh city.

TSP PM2.5 TSP PM2.5

160 1200
Mass concentration
Mass concentration

140 1000
120
( µg/m³)

800
( µg/m³)

100
80 600
60 400
40
200
20
0 0
7-9am 9-11am 11-1pm 1-3pm 3-5pm 5-7pm 7-9am 9-11am 11-1pm 1-3pm 3-5pm 5-7pm
Time Time

Fig. 2(a). The concentration of TSP and PM at KISB Fig. 2(b). The concentration of TSP and PM at CCV

TSP PM2.5 TSP PM2.5

250 200
Mass concentration

Mass concentration

200 150
( µg/m³)

( µg/m³)

150
100
100
50 50

0 0
7-9am 9-11am 11-1pm 1-3pm 3-5pm 5-7pm 7-9am 9-11am 11-1pm 1-3pm 3-5pm 5-7pm

Time Time

Fig. 2(c). The concentration of TSP and PM at SMC Fig. 2(d). The concentration of TSP and PM at RFB

48
3.2. Correlation between TSP and number of Vehicles CCV the concentration of PM<0.1 µm (UF) and
At KISB, there is a clear correlation according to the PM>10μm were the highest, 116.79μg/m3 and
coefficient of determination R-squared (R2) was 0.567 116.48μg/m3 if compared to the total concentration.
which means that there is about 56.7% possibility of TSP
link to the amount of traffic while at CCV reveals that IV. Conclusion
traffic count explained 64.41%. However, the traffic
To sum up, the concentrations of TSP measured by HV
responses 82.76% of the variation in TSP concentration.
and PM2.5 reached the maximum at all sites during 5-7
This can be assumed that it is the main source that
pm. Moreover, the highest concentration of TSP was
contributes to the mass concentration of TSP. At RFB, the
found at KISB. The correlation between TSP and the
traffic counted explained 3.8% of the variation in the TSP
number of vehicles shows that the number of vehicles can
level meaning that traffic was not merely a source of TSP
be the main source that contributes to the mass
concentration as shown in Fig. 3.
concentration of TSP at SMC while it was not merely a
source at RFB. The UF was found the highest at CCV
20000 while the other 3 sampling sites were found in coarse
R² = 0.6441 particles. Thus, this study exactly indicates that the
atmospheric particle concentration in Phnom Penh city is
Number of Vehicles

15000
affected by the air pollutants emitted from traffic along
with the road sites.
R² = 0.8276
10000 The chemical composition of PM such as
carbonaceous, heavy metal, and water-soluble ionic
5000 R² = 0.0383 R² = 0.567 compounds should be analyzed for the future study to
determine the emission sources.

0
0 500 1000 1500
Acknowledgement
TSP Concentration (μg/m3) The authors would like to thank Dr. OR Chanmoly and
KISB CCV
Mr. KAN Rithy for giving us the permission to use the
SMC RFB
Linear (KISB) Linear (CCV) Biomass Energy Laboratory for the sample storage and
Linear (SMC) Linear (RFB) analysis. We also thank the sampling sites’ owners for
allowing us do the experiment.
Fig. 3. The relation between Number of Vehicles and TSP

3.3. Particulate Matter Fraction Concentration References

Table 1. The concentration of PM fraction. [1] Ung P, Sroy S. Effect of traffic and construction
3 on air quality of Phnom Penh City, Cambodia.
Particle size The Concentration of PM (µg/m )
(µm) 2016;(September):0–2.
KISB CCV SMC RFB
[2] Brunekreef B, Holgate ST. Air pollution and
>10 90.46 116.48 152.60 49.18 health. 2002;360:1233–42.
2.5-10 100.17 113.50 160.98 63.89
[3] Englert N. Fine particles and human health - A
1-2.5 69.98 101.34 152.00 44.59
0.5-1 69.98 102.14 145.08 42.77 review of epidemiological studies. Toxicol Lett.
0.1-0.5 14.07 3.95 4.00 7.16 2004;149(1–3):235–42.
<0.1 66.59 116.79 160.45 48.84 [4] WHO. Particulate Matter. Encycl Met.
2013;(1):1663–1663.
From Table.1 the concentration of PM fraction at [5] Srivastava A, Gupta S, Jain VK. Source
KISB, SMC, and RFB were found in coarse particles, 2.5- apportionment of total suspended particulate
10µm, with the amount of 100.17μg/m3, 160.98μg/m3, and matter in coarse and fine size ranges over delhi.
63.89μg/m3, respectively. Coarse particles come from road Aerosol Air Qual Res. 2008;8(2):188–200.
dust, construction, and traffic activities[5]. However, at

49
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Synthesis of Ni-CaO/Activated Carbon (AC) Catalyst for Cooking Oil Conversion to

Green Diesel

Desy Septriana 1, Joko Wintoko 1, Muhammad Mufti Azis1,*

1
Department of Chemical Engineering, Faculty of Engineering, Universitas Gadjah Mada, 55281 Indonesia
*muhammad.azis@ugm.ac.id

Abstract

Herein, we report the synthesis procedure of promising Ni-CaO/Activated Carbon (AC) catalyst for cooking oil
conversion to green diesel. The Ni-CaO/AC catalyst was synthesized using wetness impregnation method over commercial
AC. The Ni and Ca precursors were 20% of Ni(NO3)·6H2O and 15% Ca(NO3)·4H2O. After impregnation step, the solid
catalyst was calcined in a horizontal tubular furnace under N 2 flow at 700°C. To detect the presence of Ni and Ca, we have
characterized the catalyst with energy dispersive X-ray spectroscopy (EDX). The result showed that the Ni and Ca content
were 64 and 34%, respectively. Surface area of Ni-CaO/AC is 19.129 m2/g. Then, pore volume of Ni-CaO/AC is 5.61 .10-2
cc/g. The catalyst powder was then modified to form a catalyst pellet by mixing it with phenolix resin and plasticizer. Our
result shows that the use of 40 wt% of phenolix resin and 6% of plasticizer could form a stable catalyst pellet. Evaluation of
catalyst performance in decarboxylation reaction was conducted in a batch autoclave with variation in operation temperature
of 300, 330, 350 and 370 °C. The results showed that the largest percentage removal of carboxylic acid was obtained at 350
O
C with estimated conversion of 96.95%. However, the target product of decarboxylation reaction over Ni-CaO/AC pellet
was still low and hence requires further improvement.

Keywords: cooking oil, decarboxylation, impregnation, Ni-CaO/AC, pellet

I. Introduction Decarboxylation is one of promising route for green

diesel production. Even though the decarboxylation process
Green Diesel / renewable diesel is an important does not require H2, it still requires inert gases with
transportation fuel as it may replace diesel fuel derived from atmospheric conditions such as Ar, N2, and He, because
fossil. For large scale production green diesel is produced these inert gases can spontaneously change the route of the
by converting fatty acids or fatty oils from abundant source carboxyl groups in oil or fatty acids, for example it can
of biomass. Green Diesel has similar properties with diesel prevent dehydration. Among four gases as a deoxygenation
fuel which contains low oxygenate component. It is reaction medium, N2 is commonly used as it is easy to
important as oxygenate component could lower the handle and to analyze [1].
chemical stability and energy content in fuel as commonly The oxygen group on the carbon surface determines
found in biodiesel. For this reason, it is necessary to remove the hydrophilic / hydrophobic properties of the carbon
the oxygenate component in fatty acids or fatty oil from support thereby making it alkaline, acidic or neutral from
biomass to become Green Diesel. The process for removing the surface of the catalyst. The oxygen group affects the
oxygen components is called a deoxygenation process acidity of carbon surface, thereby reducing the hydrophobic
which includes: hydrodeoxygenation, decarboxylation and properties of the carbon. This surface group plays an
decarbonylation. important role in the dispersion of the active phase, and in
catalytic activity [2]. The acidity of the support gives a total

50
number of acid sites. The number of acid sites of activated Table 1. Fatty Acid Composition of Cooking Oil
carbon is very high. The amount of acid on the support site
can be calculated by the NH3 adsorbs method. The number Components Values (%)
of acid sites of activated carbon is higher than ZrO2 and H- Myristic Acid 2.41
ZSM-5 [2]. The acidity of activated carbon comes from the Palmitic Acid 32.74
acidity or functional group of surface oxygen such as Stearic Acid 3.41
carboxylates and phenols. A high acid site number is Oleic Acid 61.43
preferred in the decarboxylation reaction.
The incorporation of acid-base metals to give acid-
2.2. Catalyst Synthesis
base properties can promote a decarboxylation reaction with
The catalyst for decarboxylation of Cooking Oil is Ni-
low coke formation. The effectiveness of metal promoters
CaO/AC catalyst. The catalyst was prepared by wet
can be increased by the addition of different functional
impregnation method. Two precursor solutions, 20% Ni
groups incorporated with a metal promoter which is
(NO3)2.6H2O and 15% Ca(NO3)2⋅4H2O were used.
preferably more highly reducible. Metal-supported catalysts
can be made in the hydrogen-free state. The acid-base Furthermore, two these solutions are impregnated into
property can enhance oxygenate transfer by decarboxylation activated carbon continuously for 5 hours under vacuum
and decarbonylation reactions. The use of a single active pressure. Subsequently, the impregnation solution was left
site that is alkaline is not favored by the deoxygenation overnight, then followed by evaporation, filtering and
reaction because there is a breakdown of carbon-carbon drying at 105 oC to obtain a dry powder [5]. Later, the
bonds which will reduce the fraction of diesel range resulting dry impregnation solution is called a Ni-CaO / AC
hydrocarbons [3]. Because of that, Ca metals was used as catalyst which was calcined at temperature of 700 oC for 4
promotor and give basic properties. CaO facilitates oxygen hours with N2 gas flowing at atmospheric pressure in a
removal by absorbing more CO2 in gas phase. Transition tubular furnace [6]. Then, the resulting powder of Ni-
metal oxides (TMOs) is a good promotor in catalyst CaO/AC was modified to pellets with the aid of 40 wt% of
synthesis, because they can alternate noble metals such as phenolix resin and 6% of plasticizer.
Pt and Pd in the field of catalytic cracking activity [4].
The objective of present work is to synthesize Ni- 2.3. Catalyst Characterization
CaO/AC catalyst to convert conventional palm based Volume/size analysis of the catalyst was carried out
cooking oil to green diesel. The reaction was conducted in a with a NOVA 2000e surface area and pore size analyzer
batch reactor and conversion of palmitic and oleic acid will (Quantachrome Instruments). The samples were degassed at
be evaluated under different operation temperatures. 120 °C with a continuous flow of nitrogen and helium at 10
psi overnight prior to the analysis. Liquid nitrogen was used
as an adsorbent during the measurements. The Multi-Point
II. Materials and Methods
Brunauer−Emmett−Teller (BET) method was used to
2.1. Materials
determine the surface area and the BarretJoyner-Halenda
The feedstock of this work, palm oil-Cooking Oil was
(BJH) method was used as the most common and accurate
obatained from mini market namely Bimoli.
method to compute the average pore volume and pore size
Ni(NO3)2·6H2O (>98.0% purity), (Ca(NO3)2·4H2O) with
of the samples.
99.0% purity were obtained from Merck and N-Hexane (GC
Energy dispersive X-ray was carried out to determine
grade) with purity > 98% was used for dilution that
the presence of metals on the activated carbon that might
obatained from Merck. The N2 gas, with 99.9% purity, was
contribute to the hydrogenation/deoxygenation catalysis.
supplied by Samator gas industri. Powdered activated
charcoal Darco were purchased from Merck. Deionized
water was prepared in our laboratory. Composition of used 2.4. Experimental set-up
cooking oil available in table 1. The decarboxylation tests were performed in a
customized autoclave reactor (250 ml) operating in a batch
mode as displayed in Fig.1. A multi-blade impeller was
equipped for mixing of liquid reactant and solid catalyst.
The temperature was measured with a K-type

51
thermocouple. In a typical batch experiment, 90 g of
cooking oil as a model compound of vegetable oils, and 4.5 Table 2. Characterisation of Ni-CaO/AC Catalyst
g of catalysts were placed in the reactor. After loading
Cooking Oil and catalyst, the reactor was flushed with Parameter Value
nitrogen in order to remove the remaining oxygen. Then, Surface Area (m2/g) BET 19.129
the reactor was heated to 300OC and maintained for 105 Pore Volume (cc/g) BJH 5.61× 10-2
minutes at that temperature. The stirring speed was
maintained to 100 rpm during reaction. The reactor was Both Surface area and Pore Volume of Ni-CaO/AC
subsequently cooled down to room temperature. The liquid catalyst in pellets were relatively small. This is probably
products were collected after filtering solid phase catalyst. due to the strong influence of phenolix resin and plasticizer
The final products was further analysed using GC–MS. This which was added to the catalyst.
experiment was repeated for the other temperatures at 330,
350, 370 OC. Table 3. Elemental Analysis of Ni-CaO/AC Catalyst with
EDX
Analyte Composition (%)
Ni 64.314
Ca 34.783
Si 0.683
K 0.195
S 0.024

The EDX mapping of the catalyst is presented in Table

Fig.1. Schematic of Batch Reactor Setup: 3. It is important to note that carbon was not included as it
(1) N2 Tube, (2) regulator, (3) mixer, (4) reactor, (5) is undetectable with the EDX. As seen here, the
pressure sensor, (6) purge valve, (7) motor, (8) temperature impregnation of Ni and Ca has been recorded well with the
controller Ni and Ca content of 64.314 and 34.783 %, respectively.

2.5. Analytical methods Catalyst evaluation in a batch reactor

Liquid products were analyzed by a gas The result from catalyst evaluation of Ni-CaO/AC
chromatograph–mass spectrometer system (Shimadzu GC- Catalyst in decarboxylation of cooking oil is presented in
MS QP-2010) equipped with Rtx-5MS column. The carrier Fig. 2.
gas was helium and the flow rate was 1.0 ml/min. The
column temperature was programmed to increase from 80 98.00
to 300 °C at 5 °C/min. The final temperature was 300 °C 97.00 Palmitic Acid
Oleic Acid
% Removal of-COOH group

and held for 12 min. The sample was diluted by hexane 96.00
(Degree of decarboxylation

95.00
before injection in GC-MS. For data analysis, percentage
94.00
removal of carboxylic acid can be computed as: 93.00
92.00
91.00
90.00
Where ni is initial peak area of –COOH group in Cooking 89.00
Oil, np is peak area of –COOH group in the liquid product. 88.00
87.00
300 330 350 370
III. Results and Discussion Temperature ( ° C)

Catalyst characterization Fig.2. Degree of decarboxylation of Palmitic Acid and

The surface area as well as the pore volume of Ni- Oleic Acid over Ni-CaO/AC Catalysts
CaO/AC catalyst is presented in Table 2.

52
As seen in Figure 2, it showed that in general the catalyst palmitic and oleic acid, above 90% area based conversion.
facilitates high conversion of Palmitic and Oleic Acid with The apparent optimum condition was 350 OC with 96.95%
area based conversion higher than 90%. The percentage conversion. Inspection on green diesel products from GC-
removal of carboxylic acid was obtained at 350OC with MS results did not give any significant fraction of C14-C22
conversion of 96.95%. Hence, it can be concluded that the compounds. This might be due to the reduced surface area
optimum temperature in this experiment was 350 OC. of the catalyst as a result of binder addition to form pellet.
Hence, further investigation is still needed to improve the
selectivity of cooking oil to green diesel.

Acknowledgement
We are thankful to Gadjah Mada University for the
financial support through research grant of Department of
Chemical Engineering.

References
[1] S.Ying, P. Lv, C. Zhao, M. Li, L. Yang, Z. Wang, Y.
Chen, and S. Liu, “Solvent-free catalytic
deoxygenation of oleic acid via nano-Ni/HZSM-5:
Effect of reaction medium and coke characterization”,
Fig.3. Peak Area of GCMS result for Decarbocylation Fuel Processing Technology. Elsevier, 179(June), pp.
Product at 370 OC 324–333, 2018
The desired product like pentadecane (green diesel ) [2] N. Hongloi, P. Prapainainar, A. Seubsai, K. Sudsakorn,
was only detected at 370OC with less than 1% w/w as be and C. Prapainainar, “Nickel catalyst with different
seen in Fig.3. Evaluation of GC-MS products revealed that supports for green diesel production”, Energy.
the target product of green diesel which lies in the range of Elsevier Ltd, 182, pp. 306–320, 2019
paraffin C14-C22 was not fully detected. Most of [3] G.A.Alsultan, N.A. Mijan, H.V. Lee, U. Rashid, A.
deoygenated product was still carboxylic group, while a Islam, and Y.H. Taufiq-Yap, “Review on thermal
small compositions were in the form of ketones, alcohols conversion of plant oil (Edible and inedible) into
and alkanes. The low selectivity to green diesel is probably green fuel using carbon-based nanocatalyst”, Catalysts,
due to the reduced active surface area and pore volume in 9(4), pp. 1–25, 2019
the presence of phenolic resin and plasticizer. [4] N.A. Mijan, H.V. Lee, J.C. Juan, A.R. Noorsaadah,
Existence of acid sites support oxygen removal via and Y.H.Taufiq-Yap, “Catalytic deoxygenation of
hydrolysis of carboxylic ester to acid and then dehydration triglycerides to green diesel over modified CaO-based
pathway. Both oxygen and hydrogen content can be catalyst ” , The Royal Society of Chemistry, pp. 46445-
removed in the reaction pathway.. The Ni-CaO/AC pellets 4640, 2017
is also fragile under the environment used in this reaction. [5] Md. Z. Hossain, M.B.I. Chowdhury, A.K. Jhawar, W.Z.
From literature, the catalytic deoxygenation of triolein to Xu, “Continuous low pressure decarboxylation of
green diesel over modified CaO-based catalyst powder fatty acids to fuel-range hydrocarbons with in situ
showed optimum deoxygenation conditions of 5 wt% of hydrogen production”, Fuel. Elsevier, 212 (June 2017),
catalyst, 340OC within 105 minutes. Ni-CaO catalyst can pp. 470–478, 2017
give a hydrocarbon compound yield (C8-C20) of about 90% [6] G.A. Alsultan, N.A. Mijan, H.V. Lee, A.S. Albazzaz,
[4]. and Y.H. Taufiq-Yap, “Deoxygenation of waste
cooking to renewable diesel over walnut shell-derived
IV. Conclusion nanorode activated carbon supported CaO-La2O3
Decarboxylation of cooking oil over Ni-CaO/AC catalyst”, Energy Conversion and Management,
catalyst has been investigated in the present work using 151(May), pp. 311–323, 2017
batch reactor at various temperature of 300, 330, 350 and
370 OC. In general, the catalyst gave high conversion of

53
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Packing of NiO Nanoparticles into the SBA-15 Support by Ethylene Glycol Assisted
Impregnation Method

Nway Nay Hlaing1,*, Tay Zar Lin2, Aye Mya Thu1, Osamu Nakagoe3, Shuji Tanabe3

1
Chemical Engineering Department, Yangon Technological University, Yangon 095-11011, Myanmar
2
Mechanical Engineering Department, Technological University (Thanlyin), Thanlyin 095-11291, Myanmar
3
Division of Chemistry and Material Science, Graduate School of Engineering, Nagasaki University, 1-14 Bunkyo-machi,
Nagasaki 852-8521, Japan
* nwaynayhlaing@ytu.edu.mm

Abstract

Production of hydrogen from biomass by gasification is a practical way for energy substitution. Tar formation
is a primary problem in the gasification process and catalytic tar steam reforming is an effective process due to
100% tar conversion. Almost all reported tar steam reforming catalysts are prepared by the conventional method
that the active metal’s particle size is usually large. In this work, the catalysts were prepared by the ethylene glycol
(EG) assisted impregnation method. Firstly, large pore SBA-15 materials were synthesized using a P123 surfactant
and Hexane. The structural properties were accessed by N2 physisorption. The material with a BET surface area,
BJH pore volume, and BJH desorption pore size of 666 m2/g, 1.27 m3/g, and 9 nm were obtained at the aging
temperature of 333 K and aging time of 2 days. 10Ni/SBA-15 catalysts were synthesized by EG assisted
impregnation method. The characteristics were analyzed by N2 physisorption, X-ray diffraction, and TEM
observation. The catalyst prepared by 1:10 molar ratio of EG to metal specie showed favorable properties. The
particle sizes of the NiO in the synthesized catalyst were smaller than 5 nm and the particles are highly dispersed
in the mesoporous framework of SBA-15 support. The particle sizes in the conventional catalyst showed greater
than 14 nm. The activity of the catalysts will examine in the catalytic tar reforming.

Keywords: 10Ni/SBA-15 catalysts, ethylene glycol assisted impregnation, NiO nanoparticle, SBA-15 material, tar steam
reforming catalyst

I. Introduction gasification, pyrolysis, and liquification. The product

components produced by the gasification or pyrolysis of
Hydrogen is an environmentally friendly fuel and the
biomass include the gaseous mixture, char, and tar. Removal
generation of hydrogen from biomass can be accomplished
of tar from the gas stream is required and fabrication of
by biological or thermochemical process [1, 2].
downstream components raises the cost. Therefore,
Thermochemical conversions have higher efficiency than
conventional gasification processes are used to operate at
biological processes and the process options in the
high temperatures (> 1073 K) to remove tar from the product
thermochemical conversion of biomass are combustion,

54
gas stream [3]. The process can operate at a lower 1:0.0168:4.02:0.295:4.42:133 molar composition of
temperature (< 1073 K) by the utility of catalyst and chemicals (P123:TEOS:C6H14:NH4F:HCl: H2O) was used in
catalytic steam reforming of tar was an effective way to this synthesis. P123 and NH4F have dissolved in HCl (conc.
carry out the process at low temperature and also improved 35%) solution and stirred at room temperature until the
carbon efficiency by reforming of tar into synthesis gas [4]. solution becomes clear. The solution was then transferred to
Accordingly, the development of tar steam reforming a water batch, set at 288±1 K, and stirred for 1 h. A mixture
catalyst is an impressive research topic for scientists and of TEOS and Hexane was added and continued stirring at
researchers. 288±1 K for 24 h. Follow by aging were performed at 333
Ni-based catalyst is the most promising due to its high K or 353 K for 2 days or 5 days. The product was collected
catalytic activity and low cost compared to novel metal by centrifugation and washing with water and dried in an
catalysts [2]. It had already been known that a catalyst with oven at 343 K overnight. Large pore SBA-15 was obtained
a high surface area could provide the more available active after calcination in air at 773 K with a heating rate of 1
sites. Therefore, the choice of catalyst support with a high K/min for 6 h.
surface area is important to disperse active metal species into The preparation of Ni/SBA-15 catalyst was performed
it. Mesoporous structure SBA-15 material was recognized to by an EG assisted impregnation technique and the metal
be favorable catalyst support due to their great advantages loading was set 10 wt %. The molar ratio of metal species to
of large surface area and pore volumes, tunable mesoporous EG was used 1 or 10. In a typical catalyst preparation, EG
size, and highly ordered mesostructure which allows and Ni(NO3)2·6H2O were dissolved in ultra-pure and mixed
diffusion and adsorption of large molecules [5, 6]. M. Kruk with SBA-15 support material. The resultant slurry was pre-
and L. Cao synthesized large pore SBA-15 in the presence stirred for 2 h or 12 h. The mixture was impregnated at 353
of Hexane and they reported that tailorable pore size was as K in a water bath with intermittent stirring and kept at this
large as 16.5 nm [7]. temperature until all the water was evaporated. The catalyst
Supported Ni catalysts were commonly prepared by precursor powder was then dried at 373 K in an oven
impregnation or precipitation method [8, 9]. By using the overnight and calcined at 723 K for 2 h in a furnace.
conventional techniques, NiO particles were decomposed on
the outer surface of the support, resulting in large particle 2.2. Characterization of Catalyst
size by agglomeration during calcination. Fortunately, it was The specific surface area, pore diameter and pore volume
reported that the double-solvent approach can deliver metal of the SBA-15 and 10Ni/SBA-15 catalysts were measured
particles into the channel of the porous support [10-13]. S. using a Micromeritics Tristar 3000 by N2 adsorption at 77 K.
Qiu et. al. recently reported that EG can assist to pack the The crystalline structure and crystallite size of the catalysts
metal particles into the porous structure of the SBA-15 were accessed using Miniflex 600, Rigaku with Cu Kα
support [14]. radiation (wavelength = 1.54ºA). TEM images of the
In this work, firstly, large pore SBA-15 materials were catalysts were taken by JEOL, JEM-2010.
synthesized by the soft template procedure. Structural
properties of SBA-15 materials were analyzed by N2 III. Results and Discussion
physisorption. 10wt% Ni loaded SBA-15 supported
N2 adsorption-desorption isotherms, and BJH desorption
catalysts were synthesized by EG assisted impregnation
pore size distributions of the large pore SBA-15 support and
method. NiO particle size and the characteristics of the
10Ni/SBA-15 catalysts are shown in Fig. 1 and 2. BET
10Ni/SBA-15 catalysts were accessed by XRD, TEM, and
surface area (SBET), BJH pore volume (VP), and pore size
N2 physisorption measurement.
(DBJH) are listed in Table 1.
The N2 adsorption isotherms of the SBA-15 materials are
II. Materials and Methods typical type IV isotherm with H1 hysteresis loops,
2.1. Synthesis of large pore SBA-15 and 10Ni/SBA-15 representing the characteristic for mesoporous materials
catalyst according to IUPAC classification [16]. The isotherms of the
Large pore mesoporous structure SBA-15 materials were SBA-15 synthesized at 333 K (SBA-15;333/2D and SBA-
synthesized by following the procedure of Sun et al. [15]. 15;333/5D) showed a thinner and narrow hysteresis loop
than that of the SBA-15 synthesized at 353 K. Consequently,

55
 1000 P o re S ize D is tr b
i u to
i n (PSD )
SBA -15 ; 353 /2D Table 1. Characteristic of the SBA-15 materials and
Am oun tA d so rb ed (cm 3 STD g -1)

03
.
-1
) SBA -15 ; 333 /5D
10Ni/SBA-15 catalysts
800 PSD (cm 3g -1nm
02
. SBET VP VMP DBJH DNiO^/*
SBA -15 ; 333 /2D Material
600 01
.
m2/g cm3/g cm3/g nm nm
S; 353/2D 545 0.935 0.036 3.4/6.7 -
0
400 0 4 8 12 16 20 S; 333/5D 670 1.41 0.053 9.1 -
P o re D iam e te r (nm ) S; 333/2D 666 1.27 0.045 9 -
200 15^ /
10Ni/S;W 478 0.889 0.028 7.4/3.6
20*
0 3.2^ /
0 0 .2 0 .4 0 .6 0 .8 1 10Ni/S;1:1 596 0.885 0.047 5.3/3.6
4*
R e la tiv e P re ssu re (P /P 0)
4.8^ /
10Ni/S;1:10 629 0.817 0.042 9.1
Fig. 1. Structural properties of the large pore SBA-15. 4.5*
S: SBA-15, SBET: BET specific surface area, VP: total pore
1000 P ore S izeD is tr b
i u t ion (PSD ) volume, VMP: micro-pore volume, DBJH: BJH pore diameter
V o lum eA d so rb ed (cm 3 STD g -1)

04
. 10N iS
/ BA -15 W
;
(taken as a maximum point), DNiO^ crystallite size of NiO
)

10N iS
/ BA -15 ; 1 :1
-1

800 03
.
P SD (cm 3g -1nm

02
.
10N iS
/ BA -15 ; 1 :10 (calculated from XRD result),DNiO^ average particle size of NiO
600 01
.
(calculated from 70-100 particles of TEM images)
0
400 0 4 8 12 16 20
P oreD iam e te r (nm ) SBET and VP of the SBA-15;333 were higher than SBA-
200
15;353. Both SBA-15;333 materials exhibited high SBET,
large VP, and favorable pore diameter. Consideration from
0 the energy usage, SBA-15; 333/2D required shorter aging
0 02
. 04
. 06
. 08
. 1
time that it was chosen as a support for the synthesizing of
R e la tiv e P re ssu re (P /P 0)
the 10Ni/SBA-15 catalysts.
Fig. 2. Structural properties of the 10Ni/SBA-15
SBET and VP of the SBA-15 support were lost by Ni
catalysts.
impregnation. SBET lose in 10Ni/SBA-15;W, 10Ni/SBA-15;
1:1 and 10Ni/SBA-15;1:10 catalysts were 28%, 11% and 6%.
pore size distributions in SBA-15;333/2D and SBA- In the same way, VP losses were 30%, 30%, and 45%,
15;333/5D catalysts were narrow pore diameter range (6-12 respectively. Among them, 10Ni/SBA-15;1:10 catalysts
nm) and pore diameter at a maximum point were 9 nm and have the smallest loss in SBET and VP.
9.1 nm, respectively. In contrast, SBA-15;353/2D catalyst X-ray diffraction patterns and TEM images can be seen
had dual pore size distribution in which the new pore size in Fig. 3 and 4. The particle size of the NiO (DNiO) of the
region appeared at 3-6 nm. These small pore sizes were 10Ni/SBA-15 catalysts is presented in Table 1.
probably due to the evaporation of hexane during aging at In 10Ni/SBA-15;W catalyst, the sharp diffraction peaks
353 K. were identified at 2θ of 37.18, 43.21, and 62.7, attributed to
The hysteresis loop of the 10Ni/SBA-15;W catalyst (Fig. the crystalline NiO phase. In 10Ni/SBA-15;1:1 and
2) become wider significantly and resulted in dual pore size 10Ni/SBA-15;1:10 catalysts, the diffraction peaks of NiO
distribution with 3-5 nm and 5-12 nm by Ni impregnation. are weak and diffused, associating to small NiO
In the EG assisted 10Ni/SBA-15;1:1 catalyst, the hysteresis nanoparticles with high dispersion. The crystallite sizes of
loop enlarged and the pore size distribution range moved to the NiO were calculated from 2θ of 43.21 degrees and it was
a lower pore size region with a dual pore size distribution 14.86 nm in the 10Ni/SBA-15;W catalyst while the
range of 3-4 nm and 4-9 nm. Fortunately, significant crystallite sizes were smaller than 5 nm in the 10Ni/SBA-
deformation was not observed in the 10Ni/SBA-15;1:10 15;1:1 and10Ni/SBA-15;1:10 catalysts.
catalyst by Ni impregnation. Pore size distribution range The dispersion of NiO particles in the SBA-15 support
also similar to the pure SBA-15 support. However, the can be seen in TEM images. In 10Ni/SBA-15;W catalyst
amount of pore volume was less than the pure SBA-15 (Fig. 4 (a)), large NiO particles are deposited on the surface
support, associated with the placement of Ni particles into of the SBA-15 support. In 10Ni/SBA-15;1:1 and 10Ni/SBA-
the porous structure of the SBA-15 support. 15;1:10 catalysts (Fig. 4 (b) and (c)), very small NiO single

56
 N Oi (a )10N iS
/ B A -15 W
;
that NiO nanoparticles are delivered into the porous
1000
N Oi
N Oi (b )10N iS
/ B A -15 ;1 1
: structure of SBA-15 support in 10Ni/SBA-15;1:10 catalyst.
800 (a ) (c )10N iS
/ B A -15 ;1 1
:0 Therefore, 10Ni/SBA-15; 1:10 catalyst will the most
t ns ity (coun ts )

possible candidate for tar steam reforming catalyst, and

600
(b ) catalytic activity will be confirmed by the catalytic steam
400 reforming experiments in our future study.
nI e

(c )
200
References
0
10 20 30 40 50 60 70 80 [1] L. Wanga, L. Wellerb Curtis, D. Jonesb David, A.
2 The a ( u K )
t C Hannab Milford, Biomass and Bioenergy 2008, 32:
Fig. 3. XRD patterns of the 10Ni/SBA-15 catalysts. 573-581.
[2] Behdad Moghtaderi, Fuel, 2007, 86: 2422–2430.
(a) [3] D. Li, Y. Nakagawa and K. Tomishige, Chin. J. Catal.,
2012, 33: 583–594.
[4] M. Yung Matthew, S. Jablonski Whitney, and A.
Magrini-Bair Kimberly, Energy & Fuels, 2009, 23:
1874–1887.
[5] E. Ahmadi, N. Dehghannejad, S. Hashemikia,
M.Ghasemnejad, and H. Tabebordbar, Drug Deliv,
(b) (c)
2014, 21(3): 164–172.
[6] N. Rahmat, A. Z. Abdullah, A.R. Mohamed, Am. J.
Applied Sci., 2010, 7(12): 1579-1586.
[7] M. Kruk and L. Cao, Langmuir, 2007, 23, 7247-7254.
[8] M. Campanati, G. Fornasari and A. Vaccari, Catalysis
Today, 2003, 77: 299–314.
[9] P. van Beurden, ECN-I--04-003, December 2004.
[10] Q. Zhu and Q. Xu, Chem 1, 2016, 220-245.
[11] Q. Zhu, J. Li, and Q. Xu, J. Am. Chem. Soc., 2013,
Fig. 4. TEM images of the catalysts (a) 10Ni/SBA-15;W,
135:10210−10213.
(b) 10Ni/SBA-15;1:1 and (c) 10Ni/SBA-15;1:10.
[12] A. Aijaz, A. Karkamkar, Y. J. Choi, N. Tsumori, E. R.
Nnebro, T. Autrey, H. Shioyama, and Q. Xu, J. Am.
particles are highly dispersed in the mesoporous SBA-15 Chem. Soc. 2012, 134:13926−13929.
support. 10Ni/SBA-15; 1:10 catalysts showed better [13] Arshad Aijaz and Qiang Xu, J. Phys. Chem. Lett.,
dispersion than 10Ni/SBA-15;1:1 catalyst. According to 2014, 5: 1400−1411.
TEM results, the particle size of NiO in 10Ni/SBA-15;W [14] S. Qiu, Q. Zhang, W. Lv, T. Wang, Q. Zhang and L.
catalyst was 20 nm while that in 10Ni/SBA-15;1:1 Ma, RSC Adv., 2017, 7: 24551–24560.
and10Ni/SBA-15;1:10 catalysts were 4 and 4.5 nm. [15] M. Kruk and L. Cao, Langmuir, 2007, 23:7247-7254.
[16] K. S. W. SING, Pure & Appl.Chem., 1982, 54: 2201-
IV. Conclusion 22l8.
Mesoporous SBA-15 material with a narrow pore size
distribution, 6-12 nm, with pore size (at maximum volume
point), 9 nm, was obtained by aging at 333 K and aging time
for 2 days using soft template method. The catalyst with
small NiO nanoparticles (< 5nm) was obtained by
synthesizing EG assisted impregnation method while the
conventional method resulted in large NiO particles (19.9
nm). As determined by N2 physisorption, it can be expected

57
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Effect of Temperature and pH on Lanthanum Precipitation from Spent Catalysts Using

Oxalic Acid
Fellicia Kartika Sari 1, Himawan Tri Bayu Murti Petrus 1* and Widi Astuti 2

1
Department of Chemical Engineering (Sustainable Mineral Processing Research Group), Universitas Gadjah Mada
Jl. Grafika 2, Yogyakarta 55281, Indonesia
2
Research Division for Mineral Technology, Indonesian Institute of Sciences (LIPI) Jl. Ir. Sutami Km. 15, Tanjung Bintang,
Lampung Selatan, Indonesia
*E-mail : bayupetrus@ugm.ac.id

Abstract

Currently, REE is categorized as a critical mineral where there are few underground sediment reserves and occurs in small
concentrations in many locations. The increasing use of REE globally has a vital function in the high technology industry to
produce superior products. Indirectly unwanted by-products are also produced as a waste of spent catalysts. Interestingly,
spent catalysts as secondary resources are mostly done to extract valuable minerals using reagents. This research is the basis
for studying the precipitation of one of the REE minerals (i.e., lanthanum) from spent catalysts using oxalic acid at various
temperatures and pH. Begins with a leaching process of 125 grams of used lanthanum catalyst using 500 mL of 1 M citric
acid, followed by precipitation using 0.5 M oxalic acid by varying the temperature (60; 80; 90) and pH (1.0; 1.5; 2.0). Based
on the analyzed results using the response surface method (RSM), to find the optimal value of several independent variables
that affect a response variable on precipitation characteristics. A second-order polynomial equation was used to correlate
response and independent variables. The coefficient of determination (R²) shows a satisfactory result of 98.86%, and the
standard probability plot ensures the model’s adequacy.

Keywords: lanthanum, oxalic acid, precipitation, REE, RSM (Response Surface Method), spent catalysts

I. Introduction ontained in the liquid into solids.

Technological advances and industrial growth will produce The choice of acid greatly influences the precipitation yield.
desired products, but unwanted by products are also In previous studies, most of the precipitation process used
produced as waste from spent catalysts with different types, oxalic acid, which in Indonesia always increases every year.
namely liquid, gas, and solid waste[1]. Currently, Indonesia is still importing oxalic acid from
Minimizing waste not only demonstrates an ability to tre abroad to fulfil some of the domestic needs for oxalic acid.
at waste from an economic perspective but is also an im Oxalic acid is widely used as a neutralizing or acidifying
portant part of environmental management responsibility. agent, a material for celluloid and rayon, explosives,
Many of the wasteby-products of spent catalysts, which glycerol purifiers, stearin, tanneries, dyestuffs and for
management methods are widely used by industry, are t laboratory purposes[3].
he disposal of land (land disposal)[2]. Oxalic acid is an example of an organic acid that will be used
Precipitation reaction is a method of separating rare eart in the precipitation process. Meanwhile, the spent catalyst
h elements which are widely used precipitating substanc lanthanum used was the result of leaching that had been
e in a solution due to a chemical reaction. This is by ad carried out in previous studies. The use of oxalic acid is to
ding certain chemicals that can change the compounds c obtain high levels of lanthanum from spent catalysts and

58
minimize waste of spent catalysts. time. The precipitate is then filtered using filter paper and
dried in an oven at 900C. After drying, the solids are then
II. Materials and Methods weighed. The filtered liquid (filtrate) was then analyzed
using ICP-OES. The weighed solid was analyzed using XRF
2.1. Raw material
and XRD. Table 2 is the analysis result using ICP-OES
The initial raw material for precipitation is a solution of
solution of precipitation spent catalyst lanthanum using
leaching spent catalyst lanthanum with citric acid (C₆H₈O₇)
oxalic acid.
0.1 M. For the leaching process using a hot plate stirrer
equipped with an Erlenmeyer, magnetic stirrer, thermometer, Table 2. ICP-OES Result for Lantahnum Precipitation
and laboratory meters for pH. The leaching solution was
heated to a temperature of 90oC and stirred at a speed of 400 Temperature Concentrate
rpm for 8 hours in pH 2. Table 1 and Fig. 1 are the analysis pH
(0C) (ppm)
of the leaching solution using XRF and XRD 1 0.071
60 1.5 0.222
Table 1. XRF result of spent catalyst 2 0.735
Constituents %Mass 1 0.075
Al 24.109 80 1.5 0.221
Si 57.464 2 0.742
Ni 8.569 1 0.066
Fe 2.351 90 1.5 0.236
Ti 2.246 2 0.453
La 1.912
P 1.001 2.3. Recovery Analysis of Precipitate
Ca 0.625 The result of the precipitation process will produce
Mn 0.451 sediment in the form of wet solids and a filtrate solution. The
Co 0.353 wet solids are dried and weighed, then analyzed using XRF
K 0.22 to determine the elemental composition in these solids and
Zn 0.151 XRD to identify the crystalline phase in the material.
In the filtrate solution, using ICP-OES analysis to
determine the concentration that is reduced during the
precipitation process. Therefore, the recovery calculation
must be done using a constant denominator, not the total
Mass. It can be seen in the following formula :
𝑀0 − 𝑀𝑙
%𝑅 = (Eq.1)
𝑀0
Where % R is the recovery percentage, M0 and Ml are the
weight of lanthanum after precipitation and after the
leaching process, where the Mass after leaching is 1.0703 in
gr.

Fig. 1. XRD pattern of the spent catalyst

2.4. Design Response Surface Method of Experiment
This study aims to understand the role of the response
2.2. Precipitation process surface method in determining the value of the independent
The precipitation process is carried out using a hot plate variables and response variable. The selection of a suitable
stirrer. The immediate solution (leaching results) of 100 ml experimental design and analysis for the response surface is
was put into Erlenmeyer, then added 0.5 M oxalic acid essential. A full three-level factorial design was used with
(C2H2O4) until it reached pH 1. The process was carried out two independent variables, pH and temperature. The
for 2 hours at 600C, 800C, and 900C. The experiment was response variable must be the precipitation of lanthanum[4].
repeated at pH 1.5 and pH 2 with the same temperature and

59
The design is shown in Table 3 : The estimation results of the model equation coefficient from
Table 3. Experimental range and level of independent the Minitab 18 output are shown in the following equation:
variables Precipitation = 0,728 + 0,3418 X1 + 0,00380 X2
(gr) - 0,1067 X12 - 0,000014 X22
Range Level - 0,001493 X1X2
Independent variables
+1 0 -1
pH (X1) 1 1.5 2 The data satisfactoriness and the model’s fitness were
Temperature, 0C (X2) 60 80 90 assessed by regression and analysis of variance (ANOVA).
It can be concluded that the second-order model for this case
With the second-order polynomial equation, written as is the right model. The results can be seen in Table 5. This
following [5] : can be seen from the P-value for the model is greater than α
= 5% while for the second-order model, the P-value is less
𝑘 𝑘
than α. The P-value = 0.001 is smaller than the degree of
𝑦 = 𝛽0 + ∑ 𝛽𝑖 𝑥𝑖 + ∑ 𝛽𝑖𝑖 𝑥𝑖2 + ∑ ∑ 𝛽𝑖𝑗 𝑥𝑖 𝑥𝑗 + 𝜀 (Eq.2) significance α = 5%, and this means that the independent
𝑖=0 𝑖=1
variables xi make a significant contribution to the model. As
Where x1 and x2 represent the coded value of the input factor it can be seen from the table, the P-value for all linear
that determines the response (y). In addition, 𝛽0 , 𝛽𝑖 , 𝛽𝑖𝑖 , independent variables were practically smaller than the
and 𝛽𝑖𝑗 represents the regression equation, and the last a square or 2-way interaction mode.
symbol is ε is the error[6]. The model’s RSM accuracy was
Table 5. Analysis of Variance from Minitab 18
analyzed using the coefficient of determination (R2). The
value R2 has ranged from 0 to 1. If R2 is close 1, it shows
that the model is highly accurate[7]. Source DF Adj SS Adj MS F-Value P-Value
Model 5 0,015191 0,003038 52,05 0,004
III. Results and Discussion Linear 2 0,012352 0,006176 105,82 0,002
X1 1 0,011984 0,011984 205,33 0,001
3.1. Analysis of Variance (ANOVA)
X2 1 0,000368 0,000368 6,31 0,087
Based on research conducted using ICP-OES analysis
Square 2 0,001437 0,000719 12,31 0,036
and recalculating to get the grams contained in the filtrate
X1* X1 1 0,001422 0,001422 24,37 0,016
solution from the precipitated results. The results are in X2* X2 1 0,000015 0,000015 0,25 0,648
Table 4. It can be seen that at a temperature of 60 oC with a 2-Way 1 0,000520 0,000520 8,91 0,058
pH of 2, the result of lanthanum recovery is 98.78%. This Interaction
calculation uses the formula previously described and X1* X2 1 0,000520 0,000520 8,91 0,058
precipitate which is only produced at a temperature of 60 oC Error 3 0,000175 0,000058
with a pH of 2. Total 8 0,015366

Table 4. The Filtrate Solution from The precipitated For the model summary, from the result using Minitab 18,
R2 square in the precipitation resul 0,98 and the prediction
0,84. The results can be seen in Table 6
Temperature Precipitate La Recovery
pH
(0C) (gr) % Table 6. Model Summary Precipitation vs
1 1.056 98.67
60 1.5 1.055 98.60 pH;Temperature
2 1.057 98.78
1 1.037 96.89 S R-sq R-sq(adj) R-sq(pred)
80 1.5 1.037 96.90
0,0076398 98,86% 96,96% 84,61%
2 1.035 96.69
1 0.986 92.10
90 1.5 0.964 90.04
2 0.940 87.83

60
IV. Conclusion
In this study, from the experimental results it can be
concluded that oxalic acid (C2H2O4) can be used as a
depositing agent for a solution of precipitation spent catalyst
lanthanum and the maximum presentation of lanthanum is
obtained at 60 ° C at pH 2 with a large percentage of
recovery of 98.67%.

Acknowledgement
The authors are very thankful to the Department of
Chemical Engineering (Sustainable Mineral Processing
Research Group), Universitas Gadjah Mada and Research
Division for Mineral Technology, Indonesian Institute of
Sciences (LIPI) Lampung support.

References
[1] O. Sidjabat, “Effect of the Element Lanthanum (La)
on Fe-Zeolite Catalyst in the Treatment of Liquid
Waste Containing Phenol,” vol. 41, no. 3, pp. 43–51,
2007.
[2] P. Studi and T. Lingkungan, “Utilization of Waste as
Oil Processing Catalyst,” pp. 182–188, 2008.
[3] N. Nurul, S. Chadijah, and K. Ramadani, “Optimal
Time And Temperature In The Production Of Oxalic
Acid (H2c2o4) From HVS Waste With Alkali
Melting Method,” Al-Kimia, vol. 5, no. 1, pp. 39–47,
2017, DOI: 10.24252/al-Kimia.v5i1.2847.
[4] H. Tri et al., “Study on Temperature and Molarity
Effect on Lanthanum Extraction From Spent
Catalyst Using,” no. May, pp. 2–5, 2018.
[5] R. Faulina, S. Andari, and D. Anggraeni, “Response
surface methodology (RSM) dan aplikasinya,”
Magister Stat. Its, pp. 152–175, 2011.
[6] A. Yulandra, I. Trisnawati, I. M. Bendiyasa, W.
Rachmi pusparini, and H. T. B. M. Petrus, “Optimasi
Presipitasi Logam Tanah Jarang dari Campuran
Konsentrat Logam Tanah Jarang dengan
Metode ”Response Surface Methodology”,” Met.
Indones., vol. 42, no. 1, p. 28, 2020, doi:
10.32423/jmi.2020.v42.28-34.
[7] N. Vedaraman et al., “Ultrasonic extraction of
natural dye from Rubia Cordifolia, optimization
using response surface methodology (RSM) &
comparison with artificial neural network (ANN)
model and its dyeing properties on different
substrates,” Chem. Eng. Process. - Process Intensif.,
vol. 114, pp. 46–54, 2017, doi:
10.1016/j.cep.2017.01.008.

61
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Synthesis Zeolit X and Zeolit A from Geothermal Sludge

Reny Oktavianti 1*, Chandra Wahyu Purnomo 1 and Rochmadi 1

1
Department of Chemical Engineering (Sustainable Mineral Processing Research Group), Universitas Gadjah Mada
Jl. Grafika 2, Yogyakarta 55281, Indonesia
*E-mail : reny.oktavianti@mail.ugm.ac.id

Abstract

Geothermal sludge was a waste from Geothermal Power Plant Dieng, produced about 165 tons per month. An alternative
way to reduce waste accumulation is to utilize it for zeolite synthesis. Based on EDX analysis carried of day geothermal
sludge it contains 98% silica. Geothermal sludge has the potential to be processed into zeolite base material which has high
economic value. Zeolite synthesis was carried out by hydrothermal at various operating temperatures 100 0C with a processing
time of 7 hours, and variations molar ratio of Si/Al was 1, 1.5, 2. The characterization of synthetic products was carried out
using XRD. Zeolite X and zeolite A synthesis have the highest crystallinity at ratio Si/Al of 1.5 at 77.62 % and 22.38 %
respectively. Crystallite size of zeolite X molar ratio Si/Al 1, 1.5, and 2 is a 137.120 nm, 122.898 nm, and 146.914 nm
respectively.

Keywords: synthesis, geothermal sludge, hydrothermal, zeolite A, zeolite X

I. Introduction Zeolite is a porous material widely used in research and

Indonesia has the potential source of silica from industry and has high commercial value. Zeolite has the
geothermal solid waste. Geothermal waste, a by-product of form of alumina -silicate crystals that are very regular with
steam power plants that use underground geothermal cavities interconnected in all directions which cause the
resources. The use of geothermal energy as a Geothermal zeolite surface area to be quite large. The formation of
Power Plant positively impacts national development zeolite is influenced by temperature, time, hydroxide
because geothermal energy is abundant [1]. However, the concentration, Si/Al ratio, reaction pressure, ageing, and
negative impact of using geothermal energy as an energy stirring [4]. In this research, zeolite will be synthesized
source is that geothermal energy production produces waste. hydrothermally with the primary material of geothermal
For example, the deposition column's resulting sludge at sludge obtained from power plant Dieng owned by PT. Geo
power plant Dieng-Wonosobo reaches around 165 tons per Dipa Energy. The final result of this research aims to get
month [2]. zeolite X and A in the zeolite synthesis.
One alternative to prevent environmental pollution from
waste accumulation is to process the geothermal sludge II. Materials and Methods
waste at the power plant into synthetic zeolite raw materials. 2.1. Raw material
Geothermal dry contains approximately 98% silica after This study's raw materials were geothermal solid waste
energy dispersive X-ray analysis. From previous research, obtained from PT GEO DIPA, Dieng-Wonosobo, Central
experiments with a geothermal waste composition of silica Java, Indonesia (98% silica), Sodium Hydroxide and
sand elements from power plant Dieng were mostly SiO2 of Aluminium Hydroxide powder were purchased from Merck
77.7748% [3]. (UN 1283), and demineralized water.

62
2.2. Zeolite Synthesis specific surface area, and morphology. The crystallite was
Geothermal sludge is cleaned of impurities and then determined by using X-ray Diffraction (XRD).
Component Element %
Before Furnace After Furnace III. Results and Discussion
SiO2 94.727 98.138 3.1. Characteristics of Geothermal Powder Raw
SO3 2.648 0.732 Materials
K2 O 1.033 0.509
The raw material can use in geothermal powder. The
Fe2O3 0.825 0.416
CaO 0.601 0.129 composition of geothermal sludge can be seen in Table 1.
Sb2O3 0.057 0.025 Where the silica content is the largest component compared
As2O3 0.040 0.020 to other components.
CuO 0.038 0.009
MnO 0.001 0.007 Table 1. Composition Geothermal Sludge
Br 0.013 0.005
Rb2O 0.009 0.005
ZnO 0.007 0.001
dried using an oven until it forms a powder and sieving Based on the analysis of the EDX results in Table 1, Si
into 100 mesh. The geothermal powder is then calcined (silica) 's elemental content in the geothermal powder before
by igniting in a furnace at a temperature of 650ºC for 6 calcination was 94.727% and after calcining was 98.138%.
hours. X-Ray Diffraction analyzed geothermal solid This shows that after calcined geothermal powder at a
waste to determine the chemical and its mineralogical. temperature of 6500C for 6 hours can reduce the metal
The geothermal powder (50 gram) and mixed with content in it to increase the percentage of silica element
NaOH solution 3.5M then heated followed by stirring content in geothermal powder.
(300 rpm) at 90˚C for 60 minutes until homogenous was
obtained. Table 2. ICP-OES Result for Synthesis Zeolite
Based on the analysis of the EDX results in Table
Component Element (ppm)
1, Si (silica) 's elemental content in the geothermal Furnace Method Leaching Method
Si 7,098 20,371
powder before calcination was 94.727% and after Al 209 209
Na 1,205 1,602
calcining was 98.138%. This shows that after
calcined geothermal powder at a temperature of In table 4.2, the results of the ICP (Inductively Coupled
Plasma) analysis from silica extraction using the tube
6500C for 6 hours can reduce the metal content
furnace method are shown, the components of silica,
in it to increase the percentage of silica element alumina, sodium are 7,098 ppm; 209 ppm; 1,205 ppm and
the leaching method each obtained 20,371 ppm; 209 ppm;
content in geothermal powder.
1,602 ppm. Of the two approaches, the most extensive silica
Table 2 is the analysis result using ICP solution of extract is the leaching method. This is because the silica
extraction silica. leaching method has high solubility. Compared to using the
The zeolite formation is carried out by adding Al powder tube furnace method, the solubility of silica is lower because
(in grams) to the sodium silicate solution with the calculated it is carried out in the solid-solid phase with the addition of
Si/Al ratio (1; 1.5; 2). The solution was put into an autoclave NaOH as a solvent.
reactor and heated at 100 °C for 7 hours. The hydrothermal Any factor that increases diffusivity and solubility in the
output of the zeolite was then washed by using above steps will facilitate extraction. Extraction of silica
demineralized water. from geothermal powders yields a percent yield of 90-95%
[17]. And taking silica from geothermal powders using the
2.3. Catalyst characterization tube furnace method yields a percent yield of 70-75%. The
The catalysts were analyzed in term of crystallinity, percent yield also strengthens that the leaching method of

63
silica removal results in a relatively large silica solubility Nap (AlO2) p (SiO2) p. h H2O (gel)
compared to the tube furnace method.
The final stage of the hydrothermal sol-gel method is
drying (drying) to remove unwanted water and liquid in
zeolite X, to expand zeolite X's surface.

Product Component (%)

Zeolit Zeolit A Zeolit X
Ratio 1 28.17 71.83
Ratio 1.5 22.38 77.62
Ratio 2 27.32 72.68
3.3. Charactesization of synthetic zeolite, Si/Al ratio 1;
1.5; 2 with X-Ray Diffraction (XRD)
Figure 1. Zeolite X and A Molar Ratio Si / Al diffractogram

Zeolite X is the most dominant phase formed in each

synthetic zeolite ratio. The number of zeolite X peaks started
decreases with increasing Si/Al molar ratio. The number of
3.2. Synthesis of Zeolite X zeolite peaks X at the rate 1; 1.5; and 2, respectively 12, 10,
Zeolite synthesis is carried out by dissolving the and 8. The height of zeolite A also decreases in intensity as
ingredients with distilled water. The material dissolved with the Si/Al molar ratio increases. However, the ratio of 1.5 has
water is then mixed with NaOH solution to form a sodium the least amount of zeolite A mixture, so that the purest
aluminate and sodium silicate solution. NaOH functions in zeolite X is obtained at a ratio of 1.5.
zeolite synthesis as an activator during smelting to form a The crystallinity of zeolite X was higher with increasing
water-soluble solution of sodium aluminate and sodium Si/Al molar ratio. However, zeolite X's highest crystallinity
silicate, then mixed and stirred using a stirrer for 60 minutes occurs at a ratio of 1.5 so that the crystallinity of zeolite X is
until the mixture is homogeneous. 1.5 > ratio 2 > ratio 1. This difference in crystallinity occurs
The next stage of the sol-gel method is polycondensation; because the number of crystalline fields produced in each
at this stage, the transition process from sol to gel occurs. To ratio is different. Samples that can reflect more rays will
produce the maximum gel, a curing/ripening stage of the gel produce high intensity so that the crystallinity of zeolite X
is required, carried out for 30 minutes. In the ageing stage, a increases [12].
gel tissue formation is stiff, strong and shrinks in solution. The qualitative analysis of synthetic zeolites is related to
This stage plays an essential role in zeolite synthesis because the number of zeolite peaks formed. The more zeolite X
it includes a gel formation process that is the beginning of peaks formed and the less zeolite A peaks, the greater the
nucleation and crystal growth [14]. In mixing these materials, purity of zeolite X. Meanwhile, quantitative analysis was
the following reactions occur [14]. The reaction that occurs carried out to determine the synthetic zeolite's percentage
in silica removal: composition. Based on the XRD results, it was found that all
synthetic products were formed of two types of zeolites,
SiO2 (s) + 2NaOH (aq) Na2SiO3 (aq) + H2O (l) namely zeolites X and A. The quantitative analysis of the
composition of synthetic zeolites is shown in Table 3.
The reaction of making aluminate: Synthesis Zeolite Products Synthesis Zeolite Composition
Al2O3 (s) + 2NaOH (aq) Na2Al2O4 (aq) + H2O (l) (%)

Overall reaction: Table 3. Results of Quantitative Analysis

NaOH (aq) + x Al2O3. ySiO2 Na2SiO3 (aq) + Na2Al2O4 (aq)
NaOH (aq) + Na2Al2 (OH) 4 (aq) + Na2SiO3 (aq) Based on Table 3. that the highest percentage of zeolite
X purity was found at a ratio of 1.5 compared to synthetic
Nax (AlO2) y (SiO2) p (SiO2) z.NaOH.H2O (gel)

64
zeolite with other ratios of 77.62%. In general, the With the most extensive zeolite synthesis X and zeolite A
percentage of zeolite X purity decreases with increasing composition, the Si/Al molar ratio of 1.5 reached 77.62%
Si/Al ratio. This is due to the decreasing number of zeolite and 22.38%. Zeolite crystal size X Si/Al molar ratio 1; 1.5
X peaks formed. However, at a ratio of 1.5, the highest and 2 are 137.120 nm; 122.495 nm and 146.914 nm.
percentage of zeolite X is found at a ratio of 1.5.
Acknowledgement
Table 4. Distance between Crystal
Zeolit Distance between Crystal (Å)
Table 4. shows that the distance between crystal planes Si/Al 1 3.984
from the smallest is zeolite X ratio 1.5 < zeolite X ratio 2 < Si/Al 1.5 3.165
zeolite X ratio 1. The smaller the distance between the Si/Al 2 4.848
crystal planes, the denser and more stable the crystal The authors are very thankful to the Department of
structure is. This causes the crystal structure formed to have Chemical Engineering (Sustainable Mineral Processing
high crystallinity. The highest crystallinity is zeolite X at a Research Group), Universitas Gadjah Mada.
ratio of 1.5.
References
Table 5. Size of Synthesized Zeolite X Crystals
[1] Saptadji, N.M. (2009). “Teknik Panas Bumi”.
Zeolit Crystal Size (nm) Departemen Teknik Perminyakan Fakultas Ilmu
Si/Al 1 137.120
Kebumian dan Teknologi Mineral Institut Teknologi
Si/Al 1.5 122.495
Bandung, Bandung.
Si/Al 2 146.914
[2] Suprapto, S. J. (2009). “Panas Bumi Sebagai Sumber
The XRD analysis data can also be used to determine the Energi dan Penghasil Emas”. Warta Geologi Volume 4
crystal size. The crystal size of synthetic zeolite X based on No. 2, Bandung.
calculations using the Debye Scherrer equation is presented [3] Syakur, A., Tumiran, Berahim, H., Rochmadi, 2011,
in Table 5. Based on these results, it is known that the crystal “Pengujian Karakteristik Limbah Pasir PLTP Dieng
size of synthetic zeolite X is in the range of 100-200 nm. Sebagai Bahan Pengisi Isolator Resin Epoksi Silane”,
The distance between particles in zeolite crystals is Jurnal Rekayasa Elektrika Vol. 9, No. 4..
directly proportional to the crystal size [6]. The small crystal [4] Payra, Pramatha and Dutta, Prabir K. (2003). "Handbook
size causes the distance between the particles to be short so of Zeolite Science and Technology". The Ohio State
that the crystal structure formed becomes even tighter and University Columbus.
more regular. The denser and more stable crystal structure [5] Hamdan, H., 1992, Introduction to Zeolites: Synthesis,
causes the degree of crystallinity to be high, therefore characterization, and modifications, University
according to the XRD results from the highest crystallinity Teknologi Malaysia, Kuala Lumpur.
of zeolite X, it is at a ratio of 1.5 the smallest crystal size [6] Ramirez, M., Garcia, T., Camacho, R., Badillo, L.,
compared to zeolite X ratios 1 and 2, which is 122.495 nm. Mojica, M., Gomez, M. (2020). Simple Synthesis of
Hierarchically Structured X Zeolit from Geothermal
IV. Conclusion Nanosilika and Its Evaluation in the Dehydration of
Aqueous Solutions of Ethanol.
Synthesis of zeolite X and A using the hydrothermal [7] Nikmah, R. A. Syukuri., Widiastuti, N., dan Fansuri, H.
autoclave method produces crystalline zeolites, and the oven 2008. Pengaruh Waktu dan Perbandingan Si/Al Terhadap
hydrothermal method has amorphous zeolites. According to Pembentukan Zeolit A dari Abu Dasar Bebas Karbon dari
the XRD analysis shown. Geothermal waste can be used as PLTU PT. IPMOMI dengan Metode Hidrotermal.
a raw material for making zeolite to increase the economic Journal of Indonesia Zeolites. Vol. 7 No. 1. Mei 2008
value of the waste. ISSN: 1411-6723.
Analysis using XRD shows that at each molar ratio Si /
Al 1; 1.5 and 2 produce a mixture of zeolites X and A. The
difference in Si/Al ratio affects the crystal structure formed.

65
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
The 5th International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Studying of Particulate Matter Concentrations in the Ambient Air during National

Holidays and Special Events

Pothiphimean Chhheang 1, Chinda Haing 1, Chanreaksmey Taing 1,2,*, Rithy Kan 3, Chanmoly Or 3,
Mitsuhiko Hata 4, Masami Furuuchi 4

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Research Unit Water and Environment, Institute of Technology of Cambodia
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Graduate School of Natural Science and Technology, Kanazawa University,
Kakuma-machi, Kanazawa, Ishikawa, 920-1192, Japan
* tsmey16@gmail.com

Abstract

Cambodian has many public holidays and special events. Annual movement of population in urban area during holidays
affects the ambient air quality in Phnom Penh. Due to these reasons, the aim of this study is to determine the particulate
matter (PM) concentrations during four special events, including Pchum Ben's Day, Water Festival, Year-end event,
and Chinese New Year's Day and the influence of each event on the ambient air quality in Phnom Penh. The sampling
point was conducted at the rooftop of Institute of Technology of Cambodia, building H (10 meter above the ground),
three times a month and during studying occasions. The PM and Total Suspended Particles (TSP) mass concentrations
were measured by gravimetric method, whilst Carbonaceous Aerosol (CA) were measured by IMPROVE-TOR protocol.
According to the results, PM and TSP was generally higher during short-term events like year-end compared to long
events like Chinese New Year. The average mass concentration of total carbon (TC) accounted for about 20% of TSP
atmosphere during all the events. Meanwhile, PM concentrations were all lower than the WHO Air Quality Guideline
(PM2.5: 25 µg/m3 PM10: 50 µg/m3 ) during all the events, except on the year-end celebration with the concentration of
55.694 µg/m3 for PM10 and 37.282 µg/m3 for PM2.5 . According to the result, the ambient air quality in Phnom Penh was
mainly polluted on year-end, moderately effect on the Pchum Ben's Day and Water Festival. Surprisingly, the ambient
air has not been severely polluted during Chinese New Year's Day as expected.

Keywords: Anthropogenic sources, carbonaceous aerosols, particulate matter, public holidays, total suspended particles.

I. Introduction During the last decade, as a developing country,

Cambodia undergoes through rapid economic and
Population growth, industrialization, and socioeconomic
population growth. As a result, the numbers of vehicles and
development are the factors contributing to air pollution.
constructions have been significantly increased which leads
The pollution may come from both natural and man-made
to major environmental issues, including air, water, soil and
sources. Fuel combustion is considered as an important
solid wastes pollution. According to [2] , vehicle emission is
source in lower-income countries, whilst motor vehicles in
considered as the critical source of air pollution in Phnom
higher-income countries [1].

66
Penh. The level of pollution from vehicles may vary from 24 hours a day and only one sample was collected. At the
day by day, especially during special events and national same time, the sampling was conducted periodically 3 times
holidays. Interestingly, during the holidays and religious- a month or every ten days and were used as control.
related event, the human activities and population flow in
and out of the city fluctuate. On some occasions, people 2.3. Analytical methods
leave the city and go back to their homeland, while people
from other provinces come to visit Phnom Penh during the TSP and PM mass concentration
events. For instance, during Water Festival, people from The mass concentration of Total Suspended Particle and
other provinces come to celebrate this event in Phnom Penh. Particulate Matter in samples were determined by the
Meanwhile, during Pchum Ben's day, people left the city and following formula.
back to their hometown. It is interesting to discover the
pollution level during these events on ambient air quality in (*+ , *-)
𝑇𝑆𝑃 𝑜𝑟 𝑃𝑀 = /0
× 104 (Eq. 1)
Phnom Penh. Therefore, the objective of this study was to
evaluate the impact of the events on the ambient air quality The TSP and PM were expressed as µg/m3 where; Mf was
by determining PM, TSP, and CA concentration on the the final mass of filter and Mi was the blank filter mass (mg),
before, during and after the event in comparison to monthly and VT was the sampling volume (m3).
concentration.
Carbonaceous aerosol mass concentration
II. Materials and Methods
The carbonaceous aerosol (CA) in the collected samples
2.1. Sampling sites were determined via the protocol of IMPROVE TOR [4], [5]
The sampling point was located at the rooftop of building by using Carbon Thermo-Optical Analyzer. Firstly, the filter
H of Institute of Technology of Cambodia, which is situated is cut into 0.5cm2 circular segment. Then this segment is
on the Russian Confederation Boulevard, Phnom Penh, insert into a heating zone where temperature is continuously
Cambodia. The equipment was set at the geographical increased in stepwise under two conditions, non-oxidizing
longitude of 104°53’48.19’’ and latitude of 11°34’14.18’’. (100% helium) and oxidizing (98%, 2% of helium and
The sampling point is approximately 10 meters above the oxygen, respectively). During this heating stage, the
ground level. There is no big building around to block the carbonaceous aerosols on the sample is volatized, pyrolyzed,
air circulation and it is surrounded by resident building, and combusted into gas-phase compounds which is oxidized
laboratories, and canteen. The high-volume air sampler into CO2 as they pass through an oxidizer (MnO2) at and
(SHIBATA HV-500F) was used to collect the TSP. The reduced to CH4 as they pass through nickel catalyst at
sampler was operated at the constant flow rate of 500 L/min. approximately The CH4 concentration is then determined by
Meanwhile for PM, Nano sampler designed by [3] was used a flame-ionization detector (FID).
with the constant flow rate of 40L/min.
For 100% helium condition, the temperature is 120°C
2.2. Sampling period (OC1), 250°C (OC2), 450°C (OC3), and 550°C (OC4). As
The Table 2.1. represented the summary of sampling temperature increases to the next level, the response from
periods during the events. The sampling was conducted for. FID might return to baseline or remain constant for 30 sec.
The residence time is varied according to the sample loads,
Table 1. Summary of sampling periods the longer time it takes for more heavily loaded samples. As
for 98% helium and 2% oxygen condition, temperature
Events/holidays Start date Stop date
plateaus are 550°C (EC1), 700°C (EC2), and 800°C (EC3).
Pchum Ben (PB) 07/10/18 11/10/18 OC, EC, and TC are determined from eight carbon fractions
as describe below [6]:
Water Festival (WF) 20/11/18 24/11/18
OC = OC1 + OC2 + OC3 + OC4 + OP (Eq. 2)
Year-end (YE) 30/12/18 01/01/19
EC = EC1 + EC2 + EC3 - OP (Eq. 3)
Chinese New Year (CNY) 03/02/19 08/02/19 TC = OC + EC (Eq. 4)

67
 Table 2. Summary of TSP and Carbonaceous mass concentration

TC TSP TC/TSP OC EC Char-EC Soot-EC Char-EC

Event OC/EC
(µg/m3) (µg/m3) (%) (µg/m3) (µg/m3) (µg/m3) (µg/m3) /Soot-EC
PB 18.7±4.0 106.4±20.6 17.5±1.8 13.4±2.7 5.2±2.0 2.8±0.8 2.6±1.2 2.6±0.9 1.0±0.2
Oct 17.8±7.3 106.0±27.2 16.8±3.4 12.2±4.2 5.6±3.1 2.4±0.7 3.1±0.7 2.5±0.5 1.1±0.2
WF 20.3±5.9 137.8±39.9 15.5±4.8 14.5±4.1 5.8±2.0 2.6±0.5 3.1±1.6 2.7±0.5 1.0±0.5
Nov 19.9±10.1 122.0±69.8 16.9±2.6 13.9±6.9 6.0±3.3 2.4±0.1 2.9±1.0 2.7±0.3 1.9±1.0
YE 28.0±2.2 149.4±8.3 17.8±5.0 21.9±1.3 6.1±0.8 3.6±0.3 3.0±0.1 3.0±1.0 2.6±1.1
Dec 21.2±3.3 111.9±21.9 19.1±1.2 15.2±2.5 6.0±0.8 2.5±0.1 1.5±0.8 3.0±0.2 2.1±0.8
CNY 25.2±3.8 149.7±16.3 16.5±1.9 19.2±2.8 6.0±1.2 3.2±0.3 3.9±1.0 2.2±0.6 1.9±0.8
Feb 31.8±8.4 152.8±18.2 17.2±1.5 23.9±6.4 7.9±2.0 3.0±0.2 2.2±2.0 3.0±02 1.6±0.7

Char-EC = EC1 – OP (Eq. 5) %, 15.5%, 17.8%, 16.5% of average mass concentra

Soot-EC = EC2 + EC3 (Eq. 6) tion in TSP in the atmosphere for PB, WF, YE, an
d CNY, respectively. The ratio of OC/EC in all the event
III. Results and Discussion exceeded 2.0,2.8±0.8, 2.6±0.5, 3.6±0.3, 3.2±0.3, orderly.
High OC/EC ratio might be the contribution from OC-
The concentration of TSP, TC, OC, EC, Char-EC and Soot- rich source emission to long-range transportation [8]. S
EC were summarized in Table 1. As shown in Table 1, all ome studies suggested the particles generated from the
the parameters, except TSP and Char-EC, shared the same exhaust emission of diesel and gasoline motor vehicles
trend, the highest concentration of pollutants was found on when the ratios between 1.0 and 4.2 [9].The result is
YE followed by CNY, WF and PB. As for TSP, the highest parallel to the report by [10] The ratios of OC/EC lar
concentration was found during CNY, following by YE, 14 ger than 2.0 has been used as an indicator presence o
9.7±16.3 µg/m3 and 149.4±8.3 µg/m3, respectively. The TSP f SOC formation [11]. The OC fraction was found the
level continuously rose from 106.4±20.6 µg/m3 in PB and highest concentration during YE (21.9±1.3 µg/m 3 ) and
reached the highest o n CNY with 149.7±16.3 µg/m3. moderately greater than Dec (15.2±2.5 µg/m3). Generally,
During December to early February, this period is people celebrate this event heavily, especially in Phno
considered as the coldest month of the year. Normally, m Penh. At the midnight, there is also firework crack
people, especially in rural and suburb areas, practice open er and firework burning to welcome new year. All the
burning to against the cold weather. This might be the factor temporary activities contribute to air pollution. As evid
contributing to higher concentration during the last two ent by [12], firework burning is considered as a short-
events. Despite the rise of TSP level during studying period, term anthropogenic sources of air pollution since it em
the concentrations were still under the standard (330 µg/m3) its a large amount of particulates and other toxic gase
according to the Sub-decree on The Control of Air Pollution s.
and Noise Disturbance by the Royal Government of
Cambodia [7]. Overall, EC are made up of approximately 50% Char-EC
and 50% of Soot-EC. It was reported by [4] that the ratio (<1)
For CA, the average mass concentration of TC accounted for indicated motor vehicular exhaust, a ratio of 1.31 indicated
18.7±4.0 µg/m3, 20.3±5.9 µg/m3, 28.0±2.2 µg/m3, 25.2±3. coal combustion and ratio of 22.6 indicated biomass
8 µg/m3, individually. OC fractions were almost three- combustion. In this study, the highest ratio was found on
fold higher than EC fractions in all the events. Amon CNY (3.8±1.5), suggesting that a great contribution from
g all events, carbon composition made up of 15-20% coal combustion process. The lowest concentration was
of total TSP mass concentration. The average mass c observed on PB (1.9±0.4), indicating a greater generation
oncentration of total carbon (TC) accounted for 17.5 of EC from the exhaust of vehicular emissions.

68
 The Table 3 described the PM mass concentration in [1] J. Cramer, “Population growth and local air
comparison to WHO standard. Both PM¬10 (55.7±5.8 pollution: methods, models, and results,” Popul. Dev.
µg/m3) and PM2.5 (37.3±5.6 µg/m3) concentration on YE Rev., vol. 28, pp. 22–52, 2002.
[2] P. Ung and S. Sroy, “Effect of traffic and
exceeded the WHO standard, 50 µg/m3 for PM10 and 25
construction on air quality of Phnom Penh City,
µg/m3 for PM2.5. As for PB and WF, the PM10 was still in Cambodia,” no. September, pp. 0–2, 2016.
allowance range, while PM2.5 was over the standard. [3] M. Furuuchi et al., “Development and performance
Surprisingly, both PM10 and PM2.5 were under the standard evaluation of air sampler with inertial filter for
on CNY. According to the comparison between the mean nanoparticle sampling,” Aerosol Air Qual. Res., vol.
concentration of individual event with the standard by one 10, no. 2, pp. 185–192, 2010.
sample t-test using SPSS program, all the mean [4] J. C. Chow, J. G. Watson, L. W. A. Chen, W. P.
Arnott, H. Moosmüller, and K. Fung, “Equivalence
concentrations were not statistically significant from the
of elemental carbon by thermal/optical reflectance
standard, except the PM10 on PB and CNY. This result and transmittance with different temperature
indicated that YE was the most polluted event among all the protocols,” Environ. Sci. Technol., vol. 38, no. 16,
studying events, following by WF, PB and CNY. pp. 4414–4422, 2004.
[5] J. G. Watson, J. C. Chow, and L.-W. A. Chen,
Table 3. Comparison of PM mass concentration to WHO Summary of Organic and Elemental Carbon/Black
Standard Carbon Analysis Methods and Intercomparisons,
vol. 5, no. 1. 2005.
Event PM10 PM2.5 [6] J. C. Chow et al., “The IMPROVE_A temperature
protocol for thermal/optical carbon analysis:
PB 24.74±9.06 17.29±6.90 Maintaining consistency with a long-term database,”
WF 41.51±11.94 26.46±5.91 J. Air Waste Manag. Assoc., vol. 57, no. 9, pp. 1014–
1023, 2007.
YE 55.44±6.83 37.12±6.77 [7] Sub-Decree No 42, 2000. “ANUKRET on The
Control of Air Pollution and Noise Disturbance.
CNY 35.21±9.86 18.32±2.97
https://static1.squarespace.com/static/593a250a15d
Standard 50.00 ±0.00 25.00±0.00 5dbd460e153ad/t/59784a59bf629a80c566ddce/150
1055578125/The+Control+of+Air+Pollution+and+
Noise+Disturbances+%282000%29.pdf Accessed
IV. Conclusion on on 15 November 2020.
The TSP, PM and carbonaceous aerosol mass concentrations [8] W. Phairuang, M. Inerb, M. Furuuchi, M. Hata, S.
Tekasakul, and P. Tekasakul, “Size-fractionated
on PB, WF, YE and CNY were studied. For further studies,
carbonaceous aerosols down to PM0.1 in southern
it is important to focus on temporal trends of PM, TSP and Thailand: Local and long-range transport effects,”
carbonaceous aerosol during Water Festival and year-end. In Environ. Pollut., vol. 260, p. 114031, 2020.
these two events, there is firework burning which is known [9] J. J. Schauer, M. J. Kleeman, G. R. Cass, and B. R.
to temporarily degrade air quality as it emits a large amount T. Simoneit, “Measurement of emissions from air
of air pollutants. In addition, it is also essential to extend pollution sources. 5. C1 - C32 organic compounds
sampling periods on pre-event and post-event. The evidence from gasoline-powered motor vehicles,” Environ.
Sci. Technol., vol. 36, no. 6, pp. 1169–1180, 2002.
from this study is not strong enough to evaluate the effect of
[10] EANET, “MYANMAR: Policies and Practices
holidays and events on ambient air quality in PP by Concerning Acid Deposition,” no. 2, pp. 40–43,
collecting the samples a day before and a day after event. 2019.
Moreover, gases pollutants like NOx, O3 and other heavy [11] R. Zhang et al., “Characterization of atmospheric
metals determination should be considered. organic and elemental carbon of PM2.5 in a typical
semi-arid area of northeastern China,” Aerosol Air
Qual. Res., vol. 12, no. 5, pp. 792–802, 2012.
Acknowledgement
[12] S. F. Kong et al., “The impacts of firework burning
This work was supported by the Atmospheric Environment at the Chinese Spring Festival on air quality:
and Air Pollution Control at Kanazawa University. Insights of tracers, source evolution and aging
processes,” Atmos. Chem. Phys., vol. 15, no. 4, pp.
References 2167–2184, 2015.

69
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Identification and Quantitative Analysis of Alternative Diesel from Waste Plastic Pyrolysis
Zin Thu Aung 1, Chinda Charoenphonphanich1*, Hidenori Kosaka 2, Pop-Paul Ewphun2, Prathan Srichai3

1
School of Engineering,Department of Mechanical Engineering, King Mongkut’s of Institute of Technology Ladkrabang,
Soi chalongkrung1, Ladkrabang, Bangkok 10520 Thailand
2
School of Engineering, Department of System and Control Engineering, Tokyo Institute of Technology,
Ishikawadai 6th Building, Room 323-2-12-1 Ookayama, Meguro-ku, Tokyo, 152-8552, Japan
Department of Mechanical Engineering, Princess of Naradhiwas University, Naradhiwas 96000 Thailand
*kmitl.chinda@gmail.com

Abstract

Plastics are an essential part of the human life and the global economy. However, the use of plastics has
been associated with significant environmental problems due to their accumulation in landfills, as plastic
waste does not degrade or degrades at very low pace. Nowadays, fast pyrolysis of waste plastic into valuable
fuels is main platform method in minimizing not only the waste disposal but also could be used as an
alternative fuel for internal combustion engines. The purpose of this study was to identi fy, quantify and
compare the composition of waste plastic diesel (WPD) with the commercial diesel (CD) of Thailand.
Simulated distillation (GC-FID) and n-d-M method were used to find the composition of both fuels. Results
indicated that the content of naphtha, kerosene, diesel, and long residue were determined quantitatively and
also identified the paraffin, naphthenes, and aromatic contents for both fuels. Naphtha and heavy oil contents
of WPD were 9.2 and 8.9wt% higher than that of CD but kerosene and diesel contents were 0.7and 17.4wt%
less than that of commercial diesel. After that, paraffin, naphthenes and aromatic contents of WPD from PNA
analysis were 80.42, 14.54 and 5.04wt% and these hydrocarbon contents of CD were 60.61, 25.91 and
13.48wt% respectively. By knowing them, the appropriate method can be determined for fuel upgrading and
interpret correctly of combustion and emissions results.

Keywords: Fast pyrolysis, Simulated distillation, n-d-M method, Waste plastic diesel, Commercial diesel

I. Introduction For instance, a PVC bottle in PET recycle can ruin the entire
batch by becoming yellowish and brittle [3]. Unlike
Plastic have essential materials due to their numerous recycling, pyrolysis does not require a keen sorting of
applications in normal life. Consequently, a huge number of different plastics. Therefore, fast pyrolysis of waste plastic
plastic products accumulate as waste in the environment. into valuable fuels is main platform method in minimizing
Plastic waste is a big issue in Thailand, because the amount not only the waste disposal but also could be used as an
of recycled plastic remains low due to recycling problems alternative fuel for internal combustion engines.
[1]. One of the recycling problems is economy as they need Previous studies found that individual type of waste
to be collected separately or sorted before the process can plastics or mixed waste plastic, which were used to produce
begin [2]. Most plastics are not compatible with each other alternative fuel, identify the chemical compounds and
and hence cannot be processed together during recycling. carbon number, and investigate on physical properties

70
[4][5][6]. There are no or few studies to quantity the boiling the relationship between the elution weight (%) and boiling
fractions and hydrocarbon type of waste plastic fuel. point, that is, to create the distillation curve of fuel samples.
Therefore, the purpose of this study is to identify and
quantify the composition of waste plastic diesel (WPD) Table 2. Experimental conditions for ASTM D2887
from real mixed waste plastic pyrolysis, and compare with Column DB-10,10mx0.53mm,2.65µm
commercial diesel (CD). Column temperature 40℃ to 350℃
Carrier gas flow rate 13.989 L/min (helium)
II. Materials and Methods Injection temperature 350℃
FID temperature 375℃
2.1. Materials Gas flow rate
The waste plastic diesel utilized in this study were derived Nitrogen (makeup) 45 mL/min
from catalytic fast pyrolysis of real mixed waste plastic Hydrogen 40 mL/min
Air flow 450 mL/min
(mostly PVC) and commercial diesel (B10) were purchased
Injection volume 0.1 µL
from PTT fuel station in Thailand. Table 1 shows the
comparison of physical properties of WPD and CD. 2.3. n-d-M Method (PNA composition Analysis)
This method requires three physical properties of
Table 1. Physical properties of fuels refractive index(n20), density(d20), and molecular weight(M).
Property ASTM CD WPD For this reason, the method is called n-d-M method. The
Method
method is included in the ASTM manual under ASTM
D3238. It calculates the distribution of carbon in
Density@15℃ D4052 824 805
(kg/m3) paraffins(%Cp), naphthenes(%CN), and aromatics(%CA)
Viscosity@40℃ D445 3.24 2.9 using equation 1 to 5. The refractive index and density at
(Cst) 20℃ and molecular weight are used as input data, which are
Cetane Index D976 56.43 67.93 estimated from correlations that are adopted in API-TDB [7].
Energy Content D240 45.86 46.29
(MJ/kg)
𝑣 = 2.51(𝑛 − 1.475) − (𝑑 − 0.851) (Eq.1)
Sulfur Content D5453 0.003 0.014
(wt.%) 𝑎 = 430 𝑖𝑓 𝑣 > 0 𝑎𝑛𝑑 670 𝑖𝑓 𝑣 < 0
%𝐶𝐴 = 𝑎𝑣 + 3660/𝑀 (Eq.2)
2.2. Simulated Distillation Method
The identification and quantitative analysis of WPD and 𝑤 = (𝑑 − 0.851) − 1.11(𝑛 − 1.475) (Eq.3)
CD were done by simulated distillation (ASTM D2887). 10600 (Eq.4)
Simulated distillation is a Gas Chromatographic technique %𝐶𝑅 = 1440𝑤 − 3%𝑆 + 𝑖𝑓 𝑤 < 0
𝑀
for determining the boiling point distribution of fuels by %𝐶𝑅 = %𝐶𝑁 + %𝐶𝐴 (Eq.5)
Flame Ionization Detection (GC-FID). Two standard %𝐶𝑃 = 100 − %𝐶𝑅 (Eq.6)
solutions were used for quantification of waste plastic diesel
and commercial diesel: normal alkanes ranging n-C5 to n-C10 III. Results and Discussion
and n-C10 to n-C40. Table 2 shows the testing condition of ` The simulated distillation curve represents the boiling
GC-FID. In simulated distillation method, the analyte points of compounds in a fuel mixture at atmospheric
retention times are directly related to the boiling points of pressure. Simulated distillation curve is presented in term of
the various hydrocarbons. A calibration curve is generated boiling point versus wt% of mixture vaporized because
from standard solution, where the retention time of each n- composition is measured in terms of wt% or weight fractions
alkane is plotted against its boiling point. And calculating in gas chromatography. And simulated distillation curve is
the area of each time interval which is divided by total area very close to actual or true boiling point curve. The
allows the proportion of elution weight (%) in each time distillation curve obtained applying the ASTM D2887
interval. The elution weight (%) in each boiling point range method allow the quantification of fuel composition such as
can be determine from calibration curve and used to obtain

71
naphtha, kerosene, diesel and heavy oil as shown in Fig. 1. although it contains almost 60% of lighter compounds. And
And the composition of CD and WPD are listed in Table 2. paraffin and aromatic ratio of WPD is higher than that CD
100 as shown in Table 4.
Heavy Oil
Weight percent vaporized (%)

90 500
80
450 421℃
70
400

Temperature (℃)
60
350
50 Diesel
300
40
30 250

20 Kerosene 200
10 150
Naphtha CD
0 100 94%
100 150 200 250 300 350 400 450 WPD

Boiling cut point (℃) 50

0 10 20 30 40 50 60 70 80 90 100
Fig. 1. Distillation curve with cut points Weight percent vaporized (%)

Table 2. Fuel compositions of WPD and CD Fig. 2. Comparison distillation curves of CD and WPD
Composition Cut point CD WPD Table 3. Boiling points of WPD and CD
range wt.% wt.%
Distillation ASTM CD WPD
Naphtha IBP-200℃ 9.6 18.8 Wt.% (℃) (℃)
Kerosene 200-250℃ 15.4 14.7
Diesel 250-370℃ 65.2 47.8
Heavy Oil 370-FBP 9.8 18.7 IBP:0.5% 126 100
10% 201 172
It can be seen that naphtha and heavy oil contents of 30% 260 236
WPD were 9.2 and 8.9 wt% higher than those of CD 50% 2887 298 289
but kerosene and diesel contents were 0.7and 17.4 wt 70% 331 342
% less than those of commercial diesel. 90% 370 402
A comparison of the distillation curve between CD and 95% 389 426
WPD is shown in Fig. 2 and distillation temperature of WPD FBP:99.5% 421 474
and CD are provided in Table 3. In Fig. 2, initial boiling
range of WPD is almost 60% lower and final boiling range For low temperature combustion of modern diesel
is almost 40% higher than those of CD. Therefore, the WPD engine, desirable fuel characteristics are low aromatic and
can be called as wide distillation fuel (WDF) because it is high cetane index (high normal paraffin) but high cetane
included lighter and heavier compounds than those of CD. index and end boiling point can lead to high smoke for hot
In fact, the lighter and heavier compounds can be removed temperature combustion of normal diesel. Table 5 can be
to match the initial boiling point and final boiling point of shown that all detected carbon number concentration of
WPD and CD fuels by distillation. Nowadays, gasoline- WPD are greater than that of CD. That’s why WPD have
diesel blended or wide distillation fuels have potential to high cetane index and paraffin contents. The C28-C40 are not
reduce soot emission and to increase thermal efficiency but detected for both fuels but three peaks are detected between
HC emission of these fuels are slightly higher than that of C24 and C28 in WPD. Therefore, end fractions 6% of WPD
diesel because of low cetane number [8]. Interestingly, should be removed to achieve the same end point as shown
cetane index of WPD is much higher than that of CD, in Fig. 2 and Table 5.

72
Table 4. PNA concentration of fuels detected for both fuels but three peaks are detected between
Carbon Content CD WPD C24 and C28 in WPD. Therefore, end fractions 6% of WPD
should be removed to achieve the same end point
Paraffin (%Cp) 60.61 80.41
Naphthenes (%CN) 25.91 14.54
Aromatic (%CA) 13.48 5.05
Acknowledgement
This work was supported by ASEAN University
Table 5. Constituents(area%) identified by GC-FID
Network/Southeast Asia Engineering Education
Carbon Content CD WPD Development Network (AUN/SEED-Net). The authors
C6 2.1457 3.38784 would like to thank Asso Prf: Dr Kanit Wattanavichien,
C7 Not detected 1.71151 Center of Fuel and Energy from Biomass (Chulalongkorn
C8 Not detected 2.35746 University) for his contribution in this work.
C10 Not detected 1.41769
C11 1.38791 3.71853 References
C14 2.84199 4.32735
C15 1.7203 4.77871 [1] W Khatha., S Ekarong., M Somkiat., S Jiraphon.,2020.
C16 1.07963 5.20978 Fuel properties, performance and emission of
C17 2.42153 5.19759 alternative fuel from pyrolysis of waste plastics. IOP
C18 3.68398 6.00175 Conf. Series: Materials Science and Engineering 717.
C20 3.59628 5.52902 [2] Fazal Mabood., M.R.Jan Jasmin Shah., Farah Jabeen.,
C24 1.47851 3.10487 2012. Catalytic pyrolysis of waste plastic and tyres.
- Not detected 2.52215 LAP LAMBERT academic publishing, U.S.A.
- Not detected 2.06772 [3] Anandhu, V., Jilse, S., 2018. Pyrolysis process to
- Not detected 1.61383 produce fuel from different types of plastic- a review.
C28 Not detected Not detected IOP Conf. Series: Materials Science and Engineering
C32 Not detected Not detected 396.
C36 Not detected Not detected [4] Brajendra, K.S., Bryan, R.M., Karl, E.V., Kenneth,
C40 Not detected Not detected M.D., Nandakishore, R., 2014. Production,
characterization and fuel properties of alternative diesel
IV. Conclusion fuel from pyrolysis of waste plastic grocery bags. Fuel
In this work, the composition of WPD are identified and processing technology 122, 79-90.
quantified using simulated distillation and n-d-M methods. [5] A.M. Motawie., Hala. B.I., Hasabo, M.A., Sahar, M.A.,
Some conclusions can be drawn as follows. R.M, Abualsoud., 2016. Fractional distillation of fuel
1. Naphtha and heavy oil contents of WPD were 9.2 a from mixed plastic waste. Conference paper.
nd 8.9wt% higher than those of CD but kerosene and [6] Z.T, Aung., C, Charoenphonphanich., H, Kosaka., P,
diesel contents were 0.7and 17.4wt% less than those Ewphum., P, Srichai., 2019. Investigation on physical
of commercial diesel. properties and measurement of bulk modulus of waste
2. Initial boiling range of WPD is almost 60% lower and plastic diesel. The 10th AUN/SEED-NET
final boiling range is almost 40% higher than those of CD. RCMEManuE,129-132.
Therefore, the WPD can be called as wide distillation fuel [7] M.R, Riazi., 2005. Characterization and properties of
(WDF) because it is included lighter and heavier compounds petroleum fractions. 1st ed. ASTM, U.S.A.
than those of CD. [8] J, Wang., Z, Wang., H, Liu., 2015. Combustion and
3 All detected carbon number concentration of WPD are emission characteristics of direct injection compression
greater than that of CD. That’s why WPD have high cetane ignition engine fueled with full distillation fuel. Journal
index and paraffin contents. Then, the C28-C40 are not of fuel 140, 561-567.

73
Session 2: Environment

3th AUN/SEED
The 13 D‐Net Region nal Conferen
nce on Chem mical Engineeering 2020 (R
RCChE‐2020)
Jointlly held with
th
Thee 5 Internattional Sympoosium on Con nservation and Managem ment of Troppical Lakes
“Insightss and Challen
nges toward A Achieving SD
DGs”

Asssessment of
o Water Quality and Soil Saliniity in the C
Coastal Areea of Camb
bodia

Kongkeea Phan1, 2, *, Huy

H Sieng1, 3 Sotha Chek1,, 3, Kyoung W
Woong Kim4, C
Chheng Y Senng2, Sophanitth Hoeng2

1
Cambodian Chemical Socciety, Street 5598, Phnom P Penh, Cambodiia
2
Faculty of
o Science and Technology, In nternational Un
University, Phnoom Penh 121011, Cambodia
c
Royal Acaddemy of Cambo odia, Russian BBlvd, Phnom Peenh, Cambodiaa
d
School
S of Earth
h Sciences and d Environmentaal Engineering,
g, Gwangju Insttitute of Sciencce and Technollogy,
Gwangju 500-712,
5 Repubblic of Korea
*
Coorresponding author:
a kongkeeaphan@gmaiil.com
Abstract

In the present
p study, we
w investigate water quality anda soil salinitty in the coastaal area of Cam mbodia. A total number of 59
water samp ples were colllected from Kampot;
K 23 waater samples w were collectedd from Kep annd 50 Water ssample of are
collected from
fr Koh Kong g provinces. On-site
O measureements were coonducted for ppH, ORP, EC, T TDS, salinity aand DO using
respective Hanna field in nstruments. Co oncurrently, agrricultural soilss in Koh Kongg are collected from each sitee using a grab
sampling technique. Soill salinity is determined by so oil/water extraact method folllowing a standdard protocol. All chemical
measuremeents were perfo ormed using Sp pectrophotomeetry methods. A Analytical resuults reveal thatt 28.6% of tubee well, 28.6%
of dug well and 40% of ponds
p are salin
ne in Kep and Kampot.
K Chem mical analysis rreveals that 2.66 % of tube weell and 17.9%
of dug well have As > 10 0 ppb, exceeded d the WHO's Drinking
D Waterr Quality Guiddeline. Approxiimately, 29.3% % of tube well,
28.6% and 60% of pond have Fe > 0.3 mg L-1 whereaas 47.5% of tuube well, 45.7% % of dug well and 20% of poond has Mn >
0.1 mg L-1. Moreover, 2.5 h F- > 1.5 mg
5 % tube well has g L-1; 5.7% of ttube well and 14.3% of dug w well have NO3- > 50 mg L-1.
However, 12.5%
1 w and 6.2% of dug well haave F- > 0.8 mgg L-1, Japanesee Drinking Watter Quality Stanndard. In Koh
of tube well
Kong, anallytical results reveal
r that app
proximately 122.5% of tube w well, 4% of dugg well and 20% % pond water are saline. Fe
and Mn aree the common contaminants for most of the water sourcees in Koh Konng coastal area.. Our current ddata show that
most of agricultural soilss in Koh Kong coastal area area non-salinityy, but the agriccultural fields w which are closse to coastline
and/or in the
t estuarine are
a considered d as high and server salinityy. This study ssuggests that aappropriate waater treatment
technologiees are needed tot provide safee drinking wateer to people ressiding in Cambbodian coastal areas. Moreovver, adaptation
actions shoould be further promoted in orrder to cope with the potentiaal impacts of thhe climate channge in this coaastal zone.

Keywords: Water quality; climate chaange; salinity; coastal area; C

Cambodia

I. Introdu
uction Sihhanoukville andd Kampot Province [2]. Cem ment factories
are located in Kampot annd breweriess, handicraft
The coastaal shoreline off Cambodia is approximately y 435
mannufacturing, peetrol storage, local and internnational ports,
km along the Gulf of Thailand
T coverring four prov
vinces
hottels and restaurrants are beingg increasingly developed in
namely Ko oh Kong, Sihaanoukville, Kam mpot and Kep p with
Sihhanoukville [2]. The coastal w water is considdered to be of
an estimateed area betweeen 17,791km² and 18,477km m² [1].
fairrly good quallity in terms of total suspended solids,
Urbanizatioon and industrrialization in the coastal zoone of
disssolved oxygeen, biochemical oxygen ddemand, total
Cambodia are of a relatiively small scaale. Most urbaan and
nitrrogen and tottal phosphoroous [3]. However, various
industrial developments are located along the coaast of
polllution sources such as uuntreated wasttewater from

74
manufacturring and minin ng, service andd tourism induustries, 2.2.. Field sampliing
solid and liquid wastes discharged from slaughterh houses
and livestock farms, as well
w as transporrtation by waterways, Waater sample waas collected froom tube well, dug well and
have a majajor impact on water quality y in the countrry [3]. ponnd in Kampot, Kep and Koh Kong provincees (Fig. 1). In
For instancce, the domesttic and municiipal waste from the totaal, 59 water sammples (tube waater n = 31, duug well n = 23
provincial town of Kam mpot has been discharged th hrough andd pond n = 5)) were collectted from Kam mpot province
the poor draining
d systeem to open manmade
m canaals or whiile 23 water saamples (tube w well n = 11 andd dug well n =
natural strreams and theen to the coaastal water withoutw 12)) were collecteed from Kep province. Watter sample of
treatment [4].
[ The impaccts of the disccharge of sewaage in tub e well (n = 80)), dug well (n = 25), canal (nn = 7), pond (n
costal wateer could increase in oligotro ophic, coastal water = 55) and rain watter (n = 5) weere collected accross the Koh
pollution, eutrophication n, and public health deterio oration Konng coastal aarea of Cam mbodia. Grounndwater was
[2]. All thee rivers that fllow into the Gulf
G of Thailan nd are colllected tube weell after pumpping out aboutt 5-10 min to
relatively small
s and veryy short with varying
v water levels flussh the standingg water out of the tube. Wateer sample was
depending upon rainfall. They have th heir sources in
n hills thenn filled into two bottles for different purposes of
about 500 to 600 m in alltitude which are a located 15 to 20 anaalyses. One w was filled innto a 500ml acid-cleaned
km from the sea [4]. The T ecologicaal communities and polyyethylene botttle which was used to analyzze cations and
living orgaanisms in receiiving water arre affected by direct anioons concentrattions. Anotherr was filled innto a 250 ml
discharge ofo the effluentts from variou us industries [5].
[ In acidd-cleaned polyyethylene bottlee and acidifiedd to pH < 2 for
addition, trace
t metals in n drinking water
w and food
dstuffs meaasuring arsennic and heaavy metal cooncentrations.
cultivated in the contam minated soil canc pose signiificant Conncurrently, onn-site measureement of phyysicochemical
health effeect to their co onsumers [6-7 7]. There is a little prooperties was cconducted at each site durring the field
known ab bout environm mental pollutiion in Camb bodian sammpling. pH andd ORP was meeasured by Hannna HI 98191
coastal areeas. The marin ne water and seediment qualitty and pH//ORP meter (Hanna, Italy)). Conductivitty, TDS and
biological samples of Cambodian coastal c zones were saliinity were m measured by HI 98192 E EC/TDS/NaCl/
documenteed in 2003 [4]. There is no a regular monitoring Ressistivity meterr (Hanna, Italyy) while DO w was measured
program which
w has been n well establiished in the coastal
c by HI9147 DO m meter (Hanna, Italy). All fieeld equipment
areas of Cambodia,
C except Sihanouk kville [8]. It is of andd instruments were checkedd and calibraated with the
significancce to investig gate water qu uality and saalinity resppective standarrd solutions prrior to each fieeld trip. Water
susceptibiliity in the coasttal areas of Cam
mbodia. fromm dug well w was collectedd and treated in the same
mannner as grounndwater. One was filled intto the 500ml
II. Materials and Metthods acidd-cleaned polyyethylene bottlee and another w was filled into
the 250ml acid-clleaned polyethyylene bottle aft fter which was
2.1. Samplling sites
aciddified with HN NO3 (65%) to pH less than 22. Pond water
werre collected ussing grab samppling method aat depth about
0-3 0 cm. A compposite sample w was filled into tthe respective
polyyethylene bott ttles. All wateer samples weere kept in a
coooler after field and transferreed to a fridge at laboratory
wheere they were sstored at 4 0C uuntil analysis.

2.3.. Sample prep

paration and aanalysis

All chemical m measurements were perform med at Food

Cheemistry Laborratory of thee Faculty of Science and
Tecchnology, Inteernational Uniiversity, Phnom m Penh. All
watter samples weere taken out oof the fridge annd allowed to
reacch to a room ttemperature beefore analysis.. Arsenic was
Fig. 1 Map
p of sampling sites
s testted with an arsenic test kit (Hach, USA). The
conncentrations off Cr, Cu, F, Fe and Mn weree measured by

75
Spectropho otometer DR 1900 (Hach, US SA). Hach DR R 1900 (ECC1:5) following a method desccribed by Harddie and Doyle
was also used to measure nitrate, nitrite, sulfatee, fee [9].. In brief, 1: 5 soil/water suspension is prepared by
chlorine annd total chlorinne. Al, Ca, K and phosphatee were weiighing 20.0 g air-dry soil innto a vial and add 100 mL
analyzed byb Hanna HI 83099 (Hann na, Italy) usin
ng the deioonized water w which is then mechanically shaken (end-
respective reagents obtaained from Hanna company y. All oveer-end preferreed) at 25°C in a closed systeem for 30 min
analytical methods
m and procedures
p of Hach are com mpliant to ddissolve solublle salts. It is theen allowed aboout 15 min for
with U.S.E EPA. Likewisse, as soon as a agricultural soil the soil to settle. The measurem ment of EC is performed by
samples arre delivered to a laboratory, the samples arre air- Hannna HI 981922 dipping the conductivity cell into the
dried in thiin layer and co
ontinuously turrned over and mixed
m suppernatant, movving it up annd down sligghtly without
to avoid fungal
f developpment. After that, the dried d soil distturbing the setttled soil.
samples arre ground with h a mortar and a pestle and passed
p
through 0.2 2 mm (80 messhes) sieve. In n the laboratory
y, soil III . Results and
d Discussion
salinity is measured
m as th
he electrical conductivity
c at 25°C
from an unfiltered
u 1:5 soil : distilled
d water suspeension

Table 1 Summary of cheemical measurements of wateer sources in K

Kep province
Tube Well (n
n = 11) Dug Welll (n = 12) WHO's
DWQ Q CDWQS
Parameters Mean Median SD Min Maax Mean Median S
SD Min Max Guideliine
As (ppb) 4.00 0.00 9.66 n.d 30
0.00 4.444 0.00 110.14 n.d 30.00 10 50
Cu (µg L-1) 227.55 141.33 289.62 n.d 930
0.00 147.755 49.17 1889.89 1.33 570.00 2,0000 1,000
Fe (mg L-1) 0.30 0.10 0.50 n.d 1.71 0.466 0.05 0.94 0.01 3.30 - 0.3
Mn (mg L-1) 0.19 0.08 0.25 n.d 0.73
0 0.600 0.31 0.73 n.d 2.07 - 0.1
Cr (mg L-1) 0.01 0.01 0.01 n.d 0.02
0 0.01 0.01 0.01 n.d 0.03 0.05 0.05
Al (mg L-1) 0.04 0.02 0.05 n.d 0.15
0 0.022 0.02 0.01 0.01 0.03 - 0.2
Ca (mg L-1) 4.60 0.01 15.06 n.d 50
0.00 17.244 0.01 550.94 n.d 176.66 - -
K+ (mg L-1) 22.78 20.00 6.31 18.33 30
0.00 26.677 25.84 7.07 20.00 35.00 - -
F- (mg L-1) 0.39 0.05 0.60 n.d 1.68 0.122 0.10 0.14 n.d 0.44 1.5 1.5
NO3- (mg L-1- ) 0.65 0.29 0.78 n.d 1.93 0.733 0.00 1.04 n.d 2.50 50 50
NO2- (mg L-1) 0.00 0.00 0.00 n.d 0.01
0 0.000 0.00 0.01 n.d 0.02 3 3
PO43-(mg L-1) 0.61 0.67 0.44 n.d 1.27 0.333 0.17 0.40 n.d 1.07 - -
SO42- (mg L-1) 13.09 4.00 18.14 n.d 58
8.33 27.422 9.50 229.13 n.d 68.33 - 250
Cl2 (mg L-1) 0.30 0.01 0.92 n.d 3.07 0.022 0.01 0.02 n.d 0.07 - -
Tot Cl2(mg L-1) 0.06 0.05 0.05 0.01 0.15
0 0.233 0.05 0.48 0.01 1.40 5 0.2-0.5
DWQ, Drink king water quality
y; CDWQS, Camb
bodian drinking water
w quality standdard; SD, Standardd deviation; Min, Minimum; Max, M
Maximum; n.d,
not detectable

andd 40% of pond are saline andd 45.2% of tube well, 45.7%
of dug well andd 100% of ppond are turbbid. Chemical
anaalysis reveals tthat 2.6 % of ttube well and 17.9% of dug
welll have As > 10 ppb, exceeeded the WH HO's Drinking
Waater Quality G Guidelines. Cooncurrently, 299.3% of tube
welll, 28.6% and 660% of pond hhave Fe greateer than 0.3 mg
L-1 whereas 47.5% % of tube welll, 45.7% of dugg well and 20% %
of ppond has Mn > 0.1 mg L-1. Concurrently, 5.7% of tube
welll and 14.3% oof dug well havve nitrate greateer than 50 mg
L-1, exceeded CD DWQS and W WHO's DWQ G Guidelines. In
Kohh Kong, analyttical results revveal that approximately 12.5%
%
of tube well, 4% % of dug welll and 20% poond water are
Fig. 2 Wateer salinity level from shorelin
ne saliine. Fe and Mnn are the comm mon contaminnants for most
of tthe water sourrces in Koh Koong coastal areea. The water
saliinity decreasess while the ddistance from the shoreline
The Summmary of chemiccal measuremen nts of water so
ources incrreases (Fig. 2).
in Kep pro
ovince is preseented in Tablee 1. In Kampo ot and
Kep, appro
oximately 28.6%% of tube welll, 28.6% of dugg well

76
 Table 2 The summary of pH and salinity (µS cm-1) of considered non-salinity, but some areas which are close to
soil in Koh Kong coastal area coastline or in the estuarine is considered high and server
KMP(n=15) KRS(n = 13) BSK (n = 17) salinity, respectively. This study suggests that appropriate
Statistics EC pH EC pH EC pH treatment technologies are necessary for residents in the
Mean 40.43 5.18 73.58 5.78 130.01 4.83 coastal area to access to clean water and minimize their
Median 39.36 5.23 31.11 5.95 32.81 4.49 potential health risks. Moreover, adaptation actions should
SD 18.82 0.86 153.81 1.05 356.69 0.73 be further promoted in order to cope with the potential
Min 13.44 4.10 6.65 4.25 22.70 3.79 impacts of the climate change in this coastal area.
Max 72.53 6.83 583.57 7.44 1510.67 6.16
KMP, Khemarak Pumin; KRS, Kirisakor; BSK, Botumsakor Acknowledgements
SD, Standard deviation; Min, Minimum; Max, Maximum; Salinity is This work was supported by GIST Research Institute (GRI)
measured by 1:5 dilution EC1:5 method grant funded by the Gwangju Institute of Science and
Technology (GIST) in 2020. The authors thank students in
The summary of pH and salinity (µS cm-1) of soil in Koh Food Chemistry Lab, Faculty of Science and Technology,
Kong coastal area is presented in Table 2. The mean IU for their field and lab assistance.
salinity of soil in Khemarak Pumin, Kirisakor and
Botimsakor are less than 150 µS cm-1 which is considered References
non-salinity by Australian soil salinity classification for
[1] Johnsen, S., Munford, G. (2012). European Union Delegation
sandy soil (Table 3). to Cambodia. “Country Environment Profile.”
Table 3 Australian salinity class based on saturated paste https://eeas.europa.eu/archives/delegations/cambodia/d
ocuments/publications/country_env_profile_cam_april_
equivalent with corresponding 1:5 dilution value
2012_en.pdf Accessed: 15 December 2020.
Saturated [2] Rizvi, A.R. and Singer, U. (2011). Cambodia Coastal Situation
paste ECeq 1:5 Dilution EC1:5 Analysis, Gland, Switzerland: IUCN. 58 pp
or ECsp [3] Water Environment Partnership in Asia (WEPA). 2015. WEPA
Salinity All soils Sand Loam Clay Outlook on Water Environmental Management in Asia,
class (dS m-1) (dS m-1) (dS m-1) (dS m-1) Ministry of the Environment, Japan
Non- [4] Water Environment Partnership in Asia (WEPA). 2020. The
0-2 0-0.14 0-0.18 0-0.25
salinity state of water environment, Cambodia. http://www.wepa-
Low 2.0-4.0 0.15-0.28 0.19-0.36 0.26-0.50 db.net/policies/state/cambodia/seaarea.htm Accessed: 15
Moderate 4.0-8.0 0.29-0.57 0.37-0.72 0.51-1.00 December 2020
High 8.0-16.0 0.58-1.14 0.73-1.45 1.01-2.00 [5] Krishna, A.K., Satyanarayanan, M., Govil, P.K., 2009.
Severe 16.0-32.0 1.15-2.28 1.46-2.90 2.01-4.00 Assessment of heavy metal pollution in water using
Extreme > 32.0 > 2.28 > 2.90 > 4.00 multivariate statistical techniques in an industrial area: A case
Adapted from Hardie and Doyle [9]; 1 dS m-1 = 1,000 µS cm-1 study from Patancheru, Medak District, Andhra Pradesh, India.
Journal of Hazardous Materials 167, 366-373.
Statistical data analysis indicates that soil salinity in [6] Kavcar, P., Sofuoglu, A., Sofuoglu, S.C., 2009. A health risk
Khemarak Pumin does not significantly differ from those in assessment for exposure to trace metals via drinking water
Kirisakor (t-test, p > 0.05) and Botumsakor (t-test, p > ingestion pathway. International Journal of Hygiene and
Environmental Health 212, 216-227.
0.05). Likewise, there is no significant difference in soil
[7] Huang, M.L., Zhou, S.L., Sun, B., Zhao, Q.G., 2008. Heavy
salinity in Kirisakor and Botumsakor (t-test, p > 0.05). metals in wheat grain: Assessment of potential health risk for
However, upper range of soil salinity in Kirisakor and inhabitants in Kunshan, China. Science of the Total
Botumsakor are considered as high and server, respectively, Environment 405, 54-61.
for sandy soil (Table 3). This could reflect seawater [8] Vorng, S. 2016. Water and Wastewater Management
intrusion in Koh Kong coastal area, especially the Cambodia. Conference on Watershed Management for
agricultural fields close to coastline or in the estuarine. Controlling Municipal Wastewater in South East Asia, 28 July
2016, Nagoya, Japan
IV. Conclusion [9] Hardie, M., Doyle, R. 2012. Measuring Soil Salinity: In
Sergey Shabala, S and Cuin, T.A. (Edit). Plant Salt Tolerance,
The present study reveals that water sources near shoreline Methods and Protocols, Springer New York Heidelberg
are saline. Water salinity decrease as distance from Dordrecht London
shoreline increases. Fe and Mn are the common
contaminants for most of the water sources in Cambodian
coastal area. Soil pH in Koh Kong coastal area is acidic;
most of the agricultural soil in the Koh Kong coastal area is

77
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Application of Statistical Downscaling for Seasonal Rainfall Forecasts in Cambodia: A

Comparison between Constructed Analogue and Bias Correction Methods

Thean THOEURN 1, 2, *, Tri Wahyu Hadi 1, Rattana CHHIN 2, 3

1
Earth Sciences Study Program, Faculty of Earth Sciences and Technology, Bandung Institute of Technology,
Jalan Ganesa No. 10 Bandung 40132, Jawa Barat Indonesia
2
Faculty of Water Resources and Hydrology, Institute of Technology of Cambodia, Russian Federation Blvd.,
P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Water and Environment Unit, Research and Innovation Center, Institute of Technology of Cambodia, Russian
Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* thoeurn.thean@gmail.com

Abstract

Seasonal rainfall prediction is very important product for various purposes, especially agriculture application. The forecast
of the rain in Cambodia contains a high value of uncertainty due to high rainfall variability and complex geography. The level
of uncertainty can be delivered quantitatively using probability. This study aims (1) to apply and evaluate two statistical
downscaling methods; Constructed Analogue (CA) and Bias Correction (BC) Methods based on the Climate Forecast System
version 2 (CFSv2) output for seasonal rainfall forecasts and (2) to assess the probabilistic forecasting skills of the forecasted
rainfall over Cambodia. The predictors used for CA method include stream function (ψ), velocity potential (χ) and geopotential
height at 850 hPa level (z850), whereas the predictand is rainfall quantity. Two statistical indicators are used to evaluate the
performance of each method in downscaling rainfall in Cambodia, namely Correlation Coefficient and Brier Score. Based on
the Correlation Coefficient, the predicted rainfall of CA method shows higher correlation with observation compared to BC
method. However, BS used to evaluate the seasonal extreme showed that the CA method was not able to capture the extreme
rainfall events. The area-averaged BS over Cambodia is 0.26 and 0.33 for CA and BC method (lower BS, better prediction
skill), respectively. Thus, CA method is recommended for climate downscaling in Cambodia.

Keywords: Bias Correction, Brier Score, CFSv2, Constructed Analogue

I. Introduction methods for the prediction of rainfall but two statistical

methods, namely Constructed Analogue (CA) and Bias
Seasonal forecast of rainfall in Cambodia is important
Correction (BC), are considered to be potentially applicable
for various purposes, especially agricultural management
for Cambodia. The CA method has been shown to have
and practices. Seasonal rainfall forecast is available globally
significant skill in reproducing the variability of daily
from National Center of Environmental Prediction (NCEP)
precipitation over the United States (US), especially in the
and Climate Forecast System version 2 (CFSv2); however,
western coast, while the BC method provides downscaling
downscaling is needed to obtain forecast data with higher
capabilities comparable to other statistical and dynamical
spatial resolution that is necessary for climate impact
methods in the context of hydrologic impacts [1]. However,
assessment. There are a number of statistical downscaling
performance of these methods for application in Cambodia

78
have not been thoroughly assessed. Moreover, CFSv2 downscaling which is stimulated by analog weather
provides ensemble rainfall prediction that needs to be forecasting. CA should be basically implemented in two
processed and delivered as probabilistic forecast products. steps: diagnosis, and prognostic processes. Diagnosis
Specific methods are needed to assess the skill of the process is performed to find relationships between patterns
probabilistic rainfall forecast in quantifying the uncertainty of predictor that are similar in two different times (usually
of spatially varying seasonal rainfall patterns across the past and future). The measure of similarity of analogue
Cambodia due to complex topography. The objectives of this for two vectors 𝑎⃗(𝑢) and 𝑎⃗(𝑡) can be identified by using,
study are as the following: cosine similarity as given by the following expression:
1. To apply and evaluate the performance of two 𝑎⃗(𝑢). 𝑎⃗(𝑡)
𝑆(𝑢) = (Eq.1)
statistical methods, CA and BC, applied for downscaling |𝑎⃗(𝑢)||𝑎⃗(𝑡)|
CFSv2 output for seasonal rainfall forecasts in Cambodia
2. To assess the probabilistic forecasting skills of In the diagnosis process, both vectors represent same
seasonal rainfall forecast in Cambodia. predictor variable; in this case, one of previously mentioned
three variables 𝜓, 𝜒, or z at 850hPa. For target time t, an
II. Materials and Methods analog predictor in the database is searched and selected
based on the similarity value, 𝑆(𝑢) which is valid for a past
2.1. Dataset
time u. Different from original analogue method, where
Rainfall data with long records are needed to implement
only single analogue is searched, 30 best analogues are
the CA method but gauge observations that meet such
selected by the rank of 𝑆(𝑢) in this CA implementation.
specification are not obtainable for this study. Therefore,
The relationship between those 30 analogues and the
Asian Precipitation—Highly Resolved Observational Data
predictor value at target time t can be established by several
Integration Towards Evaluation of Water Resources
methods like linear regression or weighted averaging.
(APHRODITE) gridded secondary dataset is used because it
Herein, a weighted averaged model is employed to get the
covers Cambodia and available from 1951 to present.
constructed value by:
APHRODITE dataset covers Asian continent and contains
daily rain gauge data from a dense network derived from the
Global Telecommunication System (GTS) and consists of 𝑍𝐶𝐴 (𝑡) = ∑ 𝑊𝑖 𝑃𝑖 (Eq.2)
gridded rainfall data with 0.25o×0.25o spatial resolution [2]. 𝑖=1

APHRODITE is regarded as a proper dataset for climate Where 𝑃𝑖 is a database predictor of each 30 best analogue
study over Indochina Region [3]. and 𝑍𝐶𝐴 is a target predictor value at target time t.
Other most important dataset for this study is the The prognosis process aims to form one analogue rainfall
database of global model output, for which the Climate (constructed analogue) at every target time t, based on the
Forecast System Reanalysis (CFSR) dataset with best analog group (subset) obtained from the process
1979−2010-time span is used. In addition, archives of diagnosis. In this study we used a weighted averaging
CFSv2 operational predictions model output from 2012 to method in the prognosis process. The weight (W) of each
2015 are also collected. These databases actually contain member of the analog subset of predictors was determined
many variables but for the purpose of this study, only based on the correlation coefficient and RMSE on target
horizontal wind vector and geo-potential height z at 850 hPa predictors.
level from the CFSR dataset are selected. The wind vectors This weight is used in the formation of constructed
are then transformed into two scalar variables namely stream analogues at the target time t:
function 𝜓 and velocity potential 𝜒. These two variables are 𝑅𝐶𝐴 (𝑡) = ∑ 𝑊𝑖 𝑅𝑖 (Eq.3)
already available in the CFSv2 data, so further calculations 𝑖=1
are not needed. More detailed discussions about the use of Where 𝑅𝑖 is a predictand analog subset that pairs with an
the variables in CA method implementation is given in the analog subset of predictors and RCA is a predicted rainfall of
following sections. CA. In this study, the training period starts from 1979 to
2010, while the target period starts from 2012 to 2015.
2.2. Constructed Analogue Method
Constructed analogue (CA) is a method of statistical

79
2.3. Bias Correction Method Where OBS is the value of observed rainfall, and RF is
rainfall forecast. CC is ranged from -1 (poorest skill) to 1
The first step of statistical Bias Correction is to find the
(highest skill).
ratio for each quintile (𝛼𝑞 ) between data rainfall observation
(in this research, APHRODITE data used as secondary data)
2.4. Probabilistic Forecast Verification
with database of rainfall estimation based on CFSv2. 𝛼𝑞 is
Brier Score (BS) is used to evaluate the accuracy of
(a) (b)

Fig. 1. Correlation between observed rainfall and predicted rainfall of Constructed Analogue (a) and Bias Correction (b)
with each grid cell over Cambodia region. The correlation is significant at 99% confident interval based on student t-
test. The cross-hatched and the yellow contours indicate that the test is failed

considered a correction factor for each quantile and a probabilistic predictions. BS is a measure of mean square
multiplier for estimated values CFSv2 in the same quantile error of probability of prediction for two categories such as
( 𝐶𝐹𝑆𝑉2_𝑝𝑟𝑒𝑑𝑞 ) to obtain corrected rainfall ( 𝑃_𝑝𝑟𝑒𝑑𝑞 ). rain event and no rain event. BS is calculated as follows;
𝑁
The correction factor for each quantile is as follows. 1
𝐵𝑆 = ∑(𝑝𝑖 − 𝑎𝑖 )2 ; 0<𝐵𝑆 < 1 (Eq.7)
𝑁
𝑑𝑎𝑡𝑎𝑏𝑎𝑠𝑒_𝐴𝑝ℎ𝑟𝑜𝑑𝑖𝑡𝑒𝑞 𝑖=1
𝛼𝑞 = (Eq.4) Where; 𝑁 is the number of samples 𝑝𝑖 is the probability
𝑑𝑎𝑡𝑎𝑏𝑎𝑠𝑒_𝐶𝐹𝑆𝑣2𝑞
of prediction and 𝑎𝑖 is that of observation (𝑎𝑖 is 1 if the
event occurs and 0 if the event does not occur). BS ranged
After obtaining correction factor of each quantile, the
from 0 (highest skill) to 1 (poorest skill).
correction factor will be used to correct rainfall in the testing
period with the equation as following.
III. Results and Discussion
𝑃_𝑝𝑟𝑒𝑑𝑞 = 𝛼𝑞 × 𝐶𝐹𝑆𝑉2_𝑝𝑟𝑒𝑑𝑞 (Eq.5) 3.1. A Comparison of Statistical Performance between
the CA and BC Methods
2.3. Deterministic Forecast Verification Figure 1 illustrates the correlation of daily rainfall of
ensemble mean between the downscaled rainfalls with
Correlation Coefficient (CC) is a measure of linear
observation. Based on the result, it can be seen that the CA
correlation of two variables. It indicates how well
(Fig. 1a) method showed higher correlation compared to BC
observation and forecast value fit a line. This can be
Method (Fig. 1b). This indicates that BC method is not well
estimated by:
𝑛 ∑𝑛s=1 RF𝑠 OBS𝑠 − ∑𝑛s=1 RF𝑠 ∑𝑛s=1 OBS𝑠
perform compared to CA method. The CA method exhibits
CC= (Eq.6)
√[𝑛 ∑𝑛s=1 RF𝑠2 − (∑𝑛s=1 RF𝑠 )2][𝑛 ∑𝑛s=1 OBS𝑠2 − (∑𝑛s=1 OBS𝑠 )2] slightly higher correlations over certain regions such as the
western part of Cambodia, while BC method show limited

80
skill over some regions, i.e., hatched areas in Fig. 1b IV. Conclusion
(correlation is not significant). Based on statistical performance analysis, Bias
Correction (BC) method shows lower correlation compared
3.2. Verification of the Probabilistic Seasonal Rainfall to Constructed Analogue (CA) method. Brier Score (BS)
Forecast using to evaluate probabilistic forecast of the seasonal
rainfall above normal showed that the CA method was not
In this study, the probability of seasonal rainfall which is
able to capture the extreme event. The area-averaged BS
more than 66 percent is used to define seasonal rainfall
over Cambodia is 0.26 and 0.33 for CA and BC method
above normal. BS is one of the probabilistic verification
(lower BS, better prediction skill), respectively. Thus, CA
methods used for verifying the seasonal prediction skill of

Fig. 2. Brier Score of seasonal above normal of Constructed Analogue (a) and Bias Correction method (b) for May-July (MJJ).
The correlation is significant at 99% confident interval based on student t-test. The cross-hatched and the yellow contours indicate
that the test is failed
climate model result. Technically, the lower BS indicates method is recommended for climate downscaling in
that the probability of the forecast is close to the observed Cambodia.
event and vice versa for the higher BS. The result of BS for References
seasonal rainfall above normal is illustrated in Fig. 2. The [1] Wood, A.W., Leung, L.R., Sridhar, V., Lettenmaier,
area-averaged BS over Cambodia is 0.26 and 0.33 for CA D.P., 2004. Hydrologic implications of dynamical
and BC method, respectively. It is noted that the BS as a and statistical approaches to downscaling climate
result of the CA method over the south-western part of model outputs. Climatic Change 62, 189.
Cambodia is higher than that of BS method. This is because [2] Yatagai, A., Kamiguchi, K., Arakawa, O., Hamada,
this region is the coastline area which frequently A., et al., 2012. Aphrodite constructing a long-term
experienced higher rainfall. Moreover, the prediction of the daily gridded precipitation dataset for Asia based on
CA method was not able to capture extremely low and high a dense network of rain gauges. Bulletin of the
observed rainfall values. For the BC method, certain parts of American Meteorological Society 93, 1401–1415.
Cambodia have higher BS due to a large uncertainty [3] Chhin, R., Bui, H.-H., Yoden, S., 2017.
produced by the BC method. However, the BC method Characterization of monthly precipitation over
shows higher skill than the CA method in several locations Indochina region to evaluate CMIP5 historical runs.
such as the north-eastern part of Cambodia. DPRI Annuals 60B, 502–522.

81
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

A Survey of Household Water Use and Groundwater Quality Index Assessment in a Rural
Community of Cambodia for Studying Potential Water Treatment Plant

Sreymao Sek 1,2, Borin Heang 1,2, Pisut Painmanakul 3, Chantha Oeurng 2 and Saret Bun 2,*

1
Water and Environmental Engineering, Graduate School,
Institute of Technology of Cambodia, Phnom Penh 12156, Cambodia
2
Faculty of Hydrology and Water Resources Engineering,
Institute of Technology of Cambodia, Phnom Penh 12156, Cambodia
3
Department of Environmental Engineering, Faculty of Engineering,
Chulalongkorn University, Bangkok 10330, Thailand
* Corresponding author: saret@itc.edu.kh

Abstract

The objective of this assessment is to estimate the household domestic water use and groundwater quality index in a small
rural community of Preyveng province. About household water use, 100 respondents in Preal commune were randomly
selected as the participants for field survey questionnaire. Average daily water consumption in Preal is about 71 liters per
capita. From this value, it was estimated that more than 80% of the households in study area is facing the water scarcity in
terms of water quantity, which may consider as a serious concern for health promoting in the study area. All households in
Preal heavily relies on groundwater wells for household domestic water use. The household used raw groundwater as drinking
water was found up to 56% without a proposer treatment system. Hence, their perception about groundwater was studied in
order to acknowledge their knowledge on quality of water. As the result, more than haft of the household indicated that their
current water use is safe while other 19% considered not safe and still directly accessed to that raw water in a reason of no
choice. About water price investigation, the results demonstrated that the price of domestic water use in Preal commune is
about 0.228 USD/m3, which is more expensive than clean water supply of Phnom Penh Water Supply Authority. In terms of
water quality, about 75% of groundwater wells in Preal commune presented in poor conditions and unsuitable for drinking
purpose. Arsenic, fluoride, and iron were defined as the main associated contaminants in groundwater on water quality index.
Consequently, the presence of these three contaminants has been considered as one of the major challenges associating to
water scarcity in Preal commune as well as many rural communities of Southeast Asia countries.

Keywords: Cambodia, Groundwater quality index, Household water use, Survey questionnaire

I. Introduction accessibility in community level, where faces high water

scarcity. It advances as the key resource of drinking water.
Over past few decades, water scarcity has been
Withdrawal of groundwater was estimated about 982 km 3
experienced as the serious problem than ever from local to
annually as the most extracted raw material in the world [1].
global scale. Groundwater can be considered as the suitable
More than half has been used for domestic water supply in
alternative water source due to its quantity, quality, and

82
many counties and generally provides more than 45% of for household potable water and other 77 wells used as
drinking water in the world, especially in the small towns irrigation or agriculture wells (about 65 families). It is
and rural communities where rely for domestic supplies divided into three types of groundwater well including pump
[2,3]. Groundwater is an optional resource where pipe water or mixed, ring, and unprotected dug wells.
supply is not accessible.
In Cambodia, groundwater roles as the main source for
drinking water supply, which used up to 53% of Cambodian
households in dry season. It is currently used for a small
community water supply and expected to be trend more for
industrial and agricultural irrigation use. Main issues and
challenges, groundwater contents high level of arsenic, iron,
manganese, fluorides, and salt in Mekong and Tonle Sap
river basin, particularly along the river and some areas [4].
Not only Cambodia, the neighboring countries including
Laos, Thailand, and Vietnam have reported about the serious
arsenic, iron, fluoride, etc. contamination in groundwater.
To propose an efficient system for addressing the water
scarcity, the information on water consumption and
groundwater contamination related to the water quantity and Fig. 1. Study area of Preal commune in Preyveng province
quality are essential for estimating the optimal preparation
of the comprehensive water system including groundwater 2.2. Data collection and analysis on water use
treatment processes design and supply system. Such type of A survey was conducted at the individual household
investigation assumes importance for satisfying the growing level in the study area. Basic information of sample size
needs of the people in the community. This study aimed to selecting for data collection is crucial importance related to
estimate the household domestic water consumption and resource and time consuming as well as result accuracy.
groundwater quality index in a small rural community of Determination sample size mainly depend on three
Preyveng province, Cambodia, for proposing a quantitative parameters: level of precision (e), level of confidence or risk,
information for designing groundwater treatment processes and degree of variability in attribute [6]. In this study, sample
and supply system. Moreover, the results and findings of the size of the respondents was determined amount of family
present study with respect to Preal commune in Kanhchriech need to conduct interview with prepared questionnaires. The
district of Preyveng province would be practical, beneficial 95% of confidence level, 50% of variability degree, and 5%
and meaningful and will act as a model for the rural sector of precision level was selected to estimate the sample size.
water resource managers in rural or semi-urban area of Maximum sample size for infinite population is 100. Hence,
Cambodia as well as in other parts of the globe. 100 respondents were collected for conducting an analysis.
Methodology of water use analysis is based on an
II. Materials and Methods interdisciplinary, descriptive, intergraded, and cross-
sectional analysis approaches. Survey questionnaires
2.1. Study area contained both closed and open-ended question included the
This study has been carried out in a small scale of questions on domestic and drinking water consumption and
Preyveng province in Preal commune of Kanhchriech source, respondent perception on water quality, and level of
district (see Fig. 1). Preal is one of the communes among 8 awareness about safe water accessibility. It was conducted
in Kanhchriech district, where located at the south of Phnom through direct face-to-face interview at the household of
Penh capital about 45 kilometers, along the coordinate 11º respondents. Various types and sizes of vessels used in each
39’N latitude and 105º 37’E longitude. Coverage area of household were measured the volume. Amount of household
Preal commune is about 66.5 km2 to group 16 villages. domestic and drinking water use was asked in terms of usage
Based on commune database online (CDB) of national duration per each vessel in which convenient for respondent
committee for sub-national democratic development and data accuracy. It was therefore used to determine daily
(NCDD) [5], more than 1,100 groundwater wells were used

83
water consumption for overall household and drinking only. Table 1. Weight (wi) and relative weight (Wi) of each
water quality parameter
2.3. Socio-economic classification
Annual household income was selected for socio- Parameter Unit WHO Std. wi Wi
pH - 6.5 - 8.5 4 0.125
economic classification in this study since it is easy to
Iron mg/L 0.3 4 0.125
operate and understand, as indicated by Singh and Turkiya Arsenic µg/L 10 5 0.156
(2013) [7]. A collected annual income data from each Fluoride mg/L 1.5 4 0.125
surveyed household was classified into five socio-economic Hardness mg/L* 100 2 0.063
groups. These five groups are Group 1 (< USD 500), Group Chloride mg/L 250 3 0.094
2 (USD 500 - 1,000), Group 3 (USD 1,000 - 2,000), Group Manganese mg/L 0.05 4 0.125
4 (USD 2,000 - 4,000), and Group 5 (> USD 4,000). Nitrate mg/L 50 5 0.156
Turbidity NTU 1 1 0.031
*as CaCO3
2.4. Groundwater quality index method
Groundwater quality was mainly based on the secondary
Estimated values of WQI were classified into five levels
data source. Available groundwater quality data of a total 21
described water condition, as presented in Table 2.
accessible wells (see Fig. 1.) sampled and analyzed between
2007 and 2008 was collected (MRD-Cambodia, 2010). Table 2. Water quality classification based on WQI
Water quality index (WQI) is the method for investigating
the influence of individual water quality parameter on the WQI range Water quality
overall water quality. Procedure for WQI calculation was < 50 Excellent water
divided into three steps. First, all nine parameters were 50 – 100 Good water
assigned as weight based on their perceived effects on 100 – 200 Poor water
primary health. Assigned weight of each parameter was 200 – 300 Very poor water
> 300 Unsuitable water for drinking purpose
referred to the previous studies of Ramakrishnaiah,
Sadashivaiah, and Ranganna (2009) [8] and Sadat-Noori,
III. Results and Discussion
Ebrahimi, and Liaghat (2014) [9], as shown in Table 1.
Secondly, a relative weight (Wi) of each parameter was 3.1. Daily water consumption
determined by using Eq. (1), where wi is the weight each Total domestic water consumption in daily of Preal
parameter, n is number of parameters investigated, Wi is the commune were estimated about 31,303 liters. Maximum
relative weight. WHO standard level, weight (wi), and domestic water was taken by Group 4 households followed
relative weight (Wi) were provided in Table 1. by Group 5 and 3 households. These three socio-economic
groups constituted more than 70% of total consumption. For
wi maximum water consumption by Group 4 represented the
Wi = n (Eq.1)
∑i=1 wi socio-economic group with the annual income between USD
2,000 and 4,000 is in a good agreement with the finding of
Ci Singh and Turkiya (2013) [7] that maximum water
qi = × 100 (Eq.2)
Si consumption was observed at a socio-economic class with
annual income USD 2,000 – 4,000.
SIi = Wi × qi (Eq.3)
Table 3. Daily water consumption (litter per capita)
WQI = ∑ SIi (Eq.4)
Socio-economic Mean value SD
Last step, a quality rating scale (qi) was determined using Group 1 76.8 69.0
Group 2 65.8 45.2
Eq. (2), where qi is quality ranking, Ci is concentration of
Group 3 68.6 46.4
each parameter (mg/L), and Si is WHO standard level Group 4 76.0 64.1
(mg/L). Value of qi was used for determining sub-index (SIi) Group 5 66.2 44.3
before calculating WQI using Eq. (3), and (4), respectively. Mean 70.7 53.8

84
 consumption possibly due to the higher cost required.
Table 3 presented the domestic water consumption per
capita in Preal commune. Average daily water consumption 3.3. Perception of water quality
per household in the study area is 313.0 litters whereas per Perception of water users on current water use quality is
capita is around 70.7 litters. It is quite low, almost two times another important information should be acknowledged
compared to the minimum water quantity (150 L/day.capita) even there is a water quality standard set, especially, for a
recommended by WHO. More than 80% of the households rural community of developing country. Approximate 40%
in Preal commune is facing water scarcity in terms of water of household in Preal commune responded that they did not
quantity. The value found is such as a serious concern for concern about their current water use quality. Moreover, it
health promoting in the study area which instantly required was noticed during the field survey that few people
a suitable and sustainable solution. simplified that their water was traditionally used long time
ago without any serious problems observed while other few
3.2. Sources of water use people mentioned that raw groundwater is more favorite and
Information of water sources for domestic water was also tastier compared to boiled or filtrated or bottled waters.
investigated during field survey. It showed that 93% of the More than haft of household in Preal commune believe that
households heavily relies on groundwater wells only, while their current water use is safe while other 19% considered
other 6% are using both groundwater wells and rain water not safe, but still directly accessed to that raw water. From
storage. There is only 1% of the households depend on this result, the water quality assessment becomes more and
groundwater wells with surface water. more important in order to clarify the safety level of their
current water use.

3.4. Estimated current price of water use

The price of current water used in Preal commune was
estimated for both total domestic water and drinking water
across socio-economic groups. It should be noted that the
price of water was determined from the cost of the related
payment for water supply, for example, the cost of petroleum
or electric city used for pump, maintenance cost of the pump,
cost of the heating energy for boiling water, maintenance
cost of household filtration unit, cost of bottled water, etc.
Summary and calculation were done for this price of water
Fig. 2. Number of households in different drinking waters
estimation and the result was illustrated in Fig. 3.

Based on Fig. 2, 49% of households in Preal commune 0.5

directly drank raw groundwater without any treatment Domestic water Drinking water

processes and other 37% used groundwater as the source for 0.4 Average water price for drinking water
Water price [USD/m3]

drinking water with additional processes including boiling 0.286 USD/m3

Average water price for domestic water
(5%), water filter unit (27%), and both (5%). Moreover, only 0.3 0.228 USD/m3
1% of residents was found using only bottled water as the
drinking water while other 8% used bottled water with 0.2
boiled groundwater (5%), with raw groundwater (1%), with
filtrated groundwater (1%), and with raw and boiled 0.1
groundwater (1%). It indicated that a majority of households
in Preal commune consumed raw groundwater directly 0
without treatment processes (49%), followed by filtrated Group 1 Group 2 Group 3 Group 4 Group 5
groundwater using household filtration unit (27%), and Socio-economic groups
mixed four types of drinking water (13%). Boiled Fig. 3. Estimated water price of each water use
groundwater (5%) and bottled water (1%) were found in low

85
 as well as to find out the main associated contaminants on
The results demonstrated that the price of domestic water water quality index. Water quality index was determined in
use in Preal commune is about 0.228 USD/m3 whereas order to check if groundwaters in Preal commune is suitable
drinking water is about 0.286 USD/m3. This result is very for drinking purpose or not. Water quality index of each
important for future estimation on water use development as groundwater well was estimated and the result showed that
well as water economic analysis in the study area. By the there is no groundwater can be categorized as ‘Excellent
way, assuming that a single household monthly consumed water’ while only 5 wells are classed in ‘Good water’ and
about 10 m3 of water, the price of water use in Preal the rest fall below this range. It means that 75% of the
commune (0.228 USD/m3) is more expensive than that of groundwater wells in study area presents in poor conditions
clean water supply of Phnom Penh Water Supply Authority. and unsuitable for drinking purpose. Moreover, water
It is expected the people living condition as well as local quality index was mainly distributed by four contaminants
GDP should be better after a qualified water was supplied including arsenic, iron, fluoride, and manganese. It indicated
with a reasonable price due to the payment on water sector, that the presence of these contaminants is the barer for
consequently resulted on health concern as well as human accessing to the qualified water use in daily life of the people
productivity, etc. in Preal commune and may distribute the nearby
communities as well.
3.5. Groundwater quality assessment
Based on the result from previous part, all households in IV. Conclusion
Preal commune mainly relies on groundwater wells for their
This study aimed to estimate the household domestic
domestic activities and more than haft of households directly
water consumption and assess the groundwater quality in a
consumed raw groundwater as drinking water without any
small rural community of Preyveng province, Cambodia. In
treatments. Therefore, this groundwater quality assessment
average, daily water consumption in Preal commune is about
is very important in order to clarify the potential effect of
71 liters per capita. Related to daily water use, all residents
contaminants in groundwater as well as on human health.
in this community heavily relies only on groundwater wells
This part aims to assess the quality of groundwater in Preal
for household domestic water use and the household used
commune mainly based on the secondary data source.
raw groundwater as drinking water was found up to 56%
Table 4. Statistical analysis of groundwater quality without treatment process. Due to the importance of
parameters in Preal commune groundwater used in Preal commune and their accessibility
without treatment, the assessment of its water quality is very
Parameter Unit Min Mean Max SD necessary for a protection and solution preparation. Based
pH - 5.30 6.80 7.42 0.47 on water quality index method, about 75% of groundwater
Iron mg/L 0.00 2.70 9.00 3.01 wells in Preal commune presented in poor conditions and
Arsenic µg/L 0.00 6.00 25.00 8.00 unsuitable for drinking purpose. Arsenic, fluoride, and iron
Fluoride mg/L 0.35 2.46 5.33 2.23
were defined as the main associated contaminants in
Hardness mg/L* 18.0 131.0 198.0 54.0
Chloride mg/L 0.29 4.32 14.00 3.68 groundwater on water quality index. Based on the result of
Manganese mg/L 0.01 0.10 0.15 0.03 this study, a removal of iron, arsenic, and fluoride from
Nitrate mg/L 0.00 1.38 10.54 2.87 groundwater is very necessary for addressing water scarcity
Turbidity NTU 0.42 10.82 106.0 23.84 through groundwater use.
*as CaCO3
Acknowledgement
Statistical analysis results were presented in Table 4.
Based on the results, it indicated that groundwater in Preal We are thankful to Project for Strengthening
commune mainly contaminated by iron, arsenic, fluoride, Engineering Education and Research for Industrial
and manganese, which are mainly associated on human Development in Cambodia of JICA through LBE Research
health effect from daily consumption. Additionally, water Grant for financial support.
quality index should be estimated in order to understand the
water condition of each groundwater wells in the study area

86
References
[1] Margat, J., Van der Gun, J., 2013. Groundwater
around the world: a geographic synopsis. CRC Press.
[2] National Groundwater Association, 2016. Facts
about global groundwater usage.
[3] UNESCO, 2009. The united nations world water
development report 3–water in a changing world.
United Nations Educational Scientific and Cultural
Organization, Paris.
[4] Ha, K., Ngoc, N. T. M., Lee, E., Jayakumar, R., 2015.
Current Status and Issue of Groundwater in the
Mekong River Basin. KIGAM, CCOP & UNESCO,
Bangkok.
[5] CDB Online, 2010. Commune Database Online.
[6] Miaoulis, G., Michener, R. D, 1976. An introduction
to sampling, Kendall.
[7] Singh, O., Turkiya, S. 2013. A survey of household
domestic water consumption patterns in rural semi-
arid village, India. GeoJournal 78(5), 777-790.
[8] Ramakrishnaiah, C., Sadashivaiah, C., Ranganna, G.,
2009. Assessment of water quality index for the
groundwater in Tumkur Taluk, Karnataka State, India.
Journal of Chemistry 6(2) (2009), 523-530.
[9] Sadat-Noori, S., Ebrahimi, K., Liaghat, A., 2014.
Groundwater quality assessment using the Water
Quality Index and GIS in Saveh-Nobaran aquifer,
Iran. Environmental Earth Sciences 71(9), 3827-
3843.

87
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Calcination of Raw Ferralsols for Enhancing Its Adsorptive Removal of

Phosphorus from Aqueous Solutions

Thi An Hang Nguyen 1*, Thi Van Le 2, Keisuke Sato 1,2, and Ngoc Duy Vu 3

1
Vietnam Japan University, Vietnam National University, Hanoi, Luu Huu Phuoc St.,
Nam Tu Liem Dist., Hanoi 101000, Vietnam
2
Graduate School of Science and Engineering, Ritsumeikan University,
Biwako-Kusatsu Campus, Shiga 525-8577, Japan
3
University of Science (VNU-HUS), Vietnam National University, Hanoi, 334 Nguyen Trai Rd.,
Thanh Xuan Dist., Hanoi 120106, Vietnam
* Corresponding author : nta.hang@vju.ac.vn

Abstract

It is well recognized that high levels of phosphorus (P) in water bodies can cause eutrophication, adversely influencing the
aquatic life and ecosystem sustainability. Thus, the decontamination of P-rich wastewater before discharging it into the
surrounding environment is essential. This article investigates the feasibility of calcination of natural ferralsols to intensify
its adsorptive removal of P from aqueous solutions. First, the best calcination temperature was determined. Next, a
comparative study between the natural ferralsols (NF) and ferralsols calcined at the selected temperature (CF500) in terms
of both physicochemical properties and adsorptive behaviors was performed. Then, SEM, XRD, XRF, and FTIR analyses
were performed to clarify the effects of calcination. It was found that among three examined calcination temperatures,
500oC resulted in the highest P sorption capacity and most neutral pH value in the post-adsorption solution. CF500
produced by lab-scale furnace and commercial scale one were quite similar with respect to P binding capacity. CF500
exhibited better P sorption capacity and rate (19.38 mg/g and 0.23 min -1) than NF (12.09 mg/g and 0.12 min-1). Additionally,
CF500 possessed higher porosity (51%), more neutral pH H2O (6.18), lower organic matter (< 0.08%), lower available P (0.86
mg P2O5/100 g soil), higher Feoxalate (2.296 mg/g) and Aloxalate (2.167 mg/g) as compared to NF, which were considered to
favor the P retention ability of CF500. These findings were consistent with the changes in morphology, mineral and metal
oxide composition as well as the functional groups of NF after calcination. Due to simple procedure, significant
improvement in P treatment performance, and no secondary pollution, calcination can be a promising method for fabrication
of the P adsorbent from naturally occurring and locally abundant ferralsols. The favorable characteristics of CF500 enable
its actual application as the potential filter materials in constructed wetlands for purification of P-rich wastewater.

Keywords: Adsorption, Calcination, Ferralsols, Phosphorus removal, Phosphorus rich wastewater

I. Introduction adversely impact the aquatic life. Thus, the

decontamination of P-rich wastewater is urgent. There is an
It is well recognized that excessive levels of phosphorus
increasing trend to use naturally occurring materials as P
in the aquatic medium can cause eutrophication and
adsorbents. However, their wide application is usually

88
restricted by low P sorption capacity, thus requiring the where Ci is the initial P concentration (mg/L), Ce is the
modification. Among several methods used for this purpose, equilibrium P concentration (mg/L).
thermal treatment also known as calcination is preferred
because of its simplicity and lack of secondary pollution. Ci - Ce
This study aims at improving the P sorption capacity of qe = xV (Eq.2)
m
natural ferralsols by calcination by (i) selection the best
calcination temperature, (ii) characterization of NF and where qe is adsorption capacity at the equilibrium (mg/g);
CF500, and (iii) investigation of adsorptive behaviors of Ci is the initial P concentration (mg/L); Ce is the
NF and CF500. equilibrium P concentration (mg/L); and m is the mass of
the adsorbent (g).
II. Materials and Methods
2.1. Sampling sites III. Results and Discussion
Collection of NF in Dak Nong province in Vietnam.
3.1. Calcination
2.2. Experimental set-up
Calcination of NF was first done with lab-scale furnace Table 1. Comparison of ferralsols before and after
at 3 types of temperatures (300, 500, and 700°C) to select calcination at different temperatures
the best one, which will be applied to the commercial
Materials P adsorption pH of post-
furnace.
capacity (mg/g) adsorption
NF and CF500 were investigated in aspects to
solution
physicochemical properties and adsorptive behaviors
NF 2.58 6.05
according to the standard methods.
CF300 3.83 6.67
2.3. Analytical methods CF500 4.68 6.95
CF700 4.25 6.76
NF and CF500 was characterized using analyses of
SEM (JSM-IT100, Jeol, Japan), XRD (Empyrean, 3.2. Characterization
PANalytical, Netherlands), XRF (S4 Pioneer, BrukerE
AXS, Germany), and FTIR (FT/IR-4600typeA, Jasco,
(a) (b)
Germany) at the VNU Key Laboratory of Advanced
Materials for Green Growth.
Physicochemical properties of NF and CF500 were
analyzed at the Lab of Pedology and Soil Environment,
Faculty of Environmental Sciences (FES), VNU University
of Science (HUS) using the respective analysis methods of
TCVN 5979:2007, TCVN 8941:2011, TCVN 8941:2011,
TCVN 5256:2009, and oxalate extraction. Fig. 1. SEM observation of a) NF and b) CF500
The concentrations of ortho-P in aqueous solutions
were measured according to Method 365.3 (EPA) using It is demonstrated in Fig. 1 that calcination of NF
UV/Vis Diode Array Spectrophotometer (S2100 UV, Unico, resulted in the porous structure on the surface of CF500
USA) at the wavelength number of 710 nm. The pH values with much greater pore diameters, thus improving the P
were monitored with SevenCompact pH/Ion meter (S220- sorption on CF500.
Kit, Mettler Toledo, China).
Table 2. Mineral composition of NF and CF500
2.3. Calculation
Materials NF (%) CF500 (%)
(Ci - Ce)
Removal efficiency (%) = x 100% (Eq.1) Kaolinite 37 81
Ci Al2O3.2SiO2.2H2O

89
 Hematite, Fe2O3 5 13 % limon 46.8 17.28
Gibbsite, Al(OH)3 50 0 % clay 43.92 8.16
Goethite, FeO(OH) 8 6 Total P %P2O5 0.09 0.08
Mg P2O5/
Available P 3.44 0.86
It can be seen from Table 2 that after calcination, the 100 g soil
Gibbsite and Goethite in NF were mostly converted into FeOxalate g/kg 0.728 2.296
Kaolinite and Hematite, which were proven to favor P AlOxalate g/kg 0.848 2.167
sorption [1].
Table 4 indicates that CF 500 resulted in the near
Table 3. Metal oxide composition of NF and CF500 neutral pH values, had extremely low contents of organic
matter, total P, available P but high levels of Feoxalate and
Composition NF (%) CF500 (%) Aloxalate. These enable its P elimination ability.
Al2O3 30.54 35.44
Fe2O3 21.83 22.92 3.2. Adsorption tests
TiO2 4.15 4.40 Influential factors
Na2O <0.01 <0.01
K 2O 0.03 0.03
CaO 0.02 0.02
MnO 0.09 0.10

Fig. 3. Effect of pH on P adsorption of a) NF and b) CF500

Effect of pH on P sorption of NF and CF500 is

displayed in Fig. 3. As shown in Fig. 3, the optimal pH
values for both NF and CF500 were 3.

Fig. 2. FTIR spectra of NF and CF500

It can be seen from Fig. 2 that some negative functional

groups (e.g. -OH-) disappeared on the surface of CF500.
Consequently, the binding of PO43- anions on CF500 was
intensified due to the electrostatic attraction [2].

Physicochemical properties
Fig. 4. Effect of adsorbent dose on P sorption of a) NF and
b) CF500
Table 4. Physicochemical properties of NF and CF500
In Fig. 4, the P adsorption capacity of NF and CF500
Parameters Unit NF CF500
reached the highest values at adsorbent dose of 1g/75 mL.
pHH2O - 5.14 6.18
Therefore, this will be used as the optimal adsorbent dose
Organic matter % 0.62 <0.08
for next experiments.
Density g/cm3 2.79 3.56
Soil texture % sand 9.28 74.56
Isotherm and kinetic studies

90
Table 5. Isotherm parameters for P sorption of NF and It is evident from Table 7 that the P sorption by CF500
CF500 was spontaneous and feasible (△G <0), had endothermic
nature (△H >0) with high affinity (△S >0) [4,5].
Isotherm Isotherm NF CF500
model parameters
Langmuir qm (mg/g) 12.09 19.38
IV. Conclusion
KL 0.005 0.008 Calcination of raw ferralsols favored its P retention
R2 0.910 0.982 capacity. This was resulted from the changes in the surface
Freundlich n 1.91 2.35 morphology, functional groups, and mineral composition of
Kf 0.34 1.10 the material under high temperature. The optimal
R2 0.980 0.980 calcination temperature was 500°C. The P sorption on
CF500 was characterized by mono-layer and chemisorption.
Table 5 shows that CF500 (19.38 mg/g) exhibited a
significantly higher qmax value as compared to NF (12.09
Acknowledgement
mg/g), suggesting that calcination was efficient in
enhancing P binding ability of NF. Langmuir model was This research was funded by Vietnam National
better in describing experimental isotherm data of CF500, Foundation for Science and Technology Development
indicating that the P sorption by CF500 was mono-layer.
(NAFOSTED) [grant number 105.99-2018.13, 2018]. We
Table 6. Kinetic parameters for P sorption of NF and are thankful to the Japan International Cooperation Agency
CF500 (JICA) for the financial support to attend the Conference.

Kinetic Kinetic parameters NF CF500 References

model
Pseudo- qe (mg/g) 1.75 1.96 [1] Almasri, D. A., Saleh, N. B., Atieh, M. A., McKay, G.,
first-order k1 (min-1) 0.12 0.25 Ahzi, S., 2019. Adsorption of phosphate on iron
R2 0.840 0.932 oxide doped halloysite nanotubes. Scientific reports,
Pseudo- qe (mg/g) 1.87 1.99 9(1), 1-13.
second- k2 (g/mg.min) 0.025 0.164 [2] Wu, B., Lo, I. M., 2020. Surface Functional Group
order R2 0.880 0.949 Engineering of CeO2 Particles for Enhanced
Phosphate Adsorption. Environmental Science &
It can be observed from Table 6 that the kinetic data of Technology, 54(7), 4601-4608.
both NF and CF500 matched the Pseudo-second-order [3] Kumar, P. S., Vincent, C., Kirthika, K., Kumar, K. S.,
kinetic model better than the Pseudo-first-order model, 2010. Kinetics and equilibrium studies of Pb2+ in
implying that the main pathway for P removal by NF and removal from aqueous solutions by use of nano-
CF500 was via chemisorption [3]. Also, the P sorption rate silversol-coated activated carbon. Brazilian Journal
of CF500 (k2 = 0.164 min-1) was substantially higher than of Chemical Engineering, 27(2), 339-346.
that of NF (k2 = 0.025 min-1). [4] Peng, L., Qin, P., Lei, M., Zeng, Q., Song, H., Yang, J.,
… Gu, J., 2012. Modifying Fe3O4 nanoparticles
Table 7. Thermodynamic parameters for P sorption of with humic acid for removal of Rhodamine B in
CF500 water. Journal of Hazardous Materials, 209-210,
193–198.
Temperature (K) 303 313 323
[5] Mezenner, N. Y., & Bensmaili, A., 2009. Kinetics and
qm (mg/g) 19.58 21.29 23.51
thermodynamic study of phosphate adsorption on
Kd 1.017 1.227 1.553
iron hydroxide-eggshell waste. Chemical
△G (J/mol) -41.47 -533.20 -1181.40
Engineering Journal, 147(2-3), 87-96.
△H (J/mol) 20642
△S (J/mol/K) 51

91
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Oxidation-Precipitation of Iron (II) in Groundwater using Modified Airlift Reactor:

Kinetics and Influence of Process Conditions

Saret Bun 1, Fumiyuki Nakajima 2, Tomohiro Tobino 3 and Pisut Painmanakul 4,*

1
Water and Environmental Engineering, Faculty of Hydrology and Water Resources Engineering,
Institute of Technology of Cambodia, Phnom Penh 12156, Cambodia
2
Environmental Science Center, The University of Tokyo, Tokyo 113-0033, Japan
3
Department of Urban Engineering, Graduate School of Engineering,
The University of Tokyo, Tokyo 113-8656, Japan
4
Department of Environmental Engineering, Faculty of Engineering,
Chulalongkorn University, Bangkok 10330, Thailand
* Corresponding author: saret@itc.edu.kh

Abstract

The objective of this work is to study the influence of operation parameters on ferrous iron oxidation kinetic under batch
conditions using enhanced aeration efficiency reactor, modified airlift reactor (MALR). It presented as a rectangular internal-
loop column for containing 6.6 litter total working volume. The influence of process conditions, i.e., initial pH (between 6.5
and 8.0), gas flow rate (between 0.5 and 2 L/min), initial ferrous iron concentration (3 and 20 mg/L), and insoluble ferric
hydroxide particle (between 0 and 20 mg/L) were investigated. As the result, iron oxidation is strongly dependent on pH
which emphasizes a good ability to maintain a constant pH of gas-liquid reactor such as MALR. Around neutral pH should
be considered for controlling iron oxidation reaction. For the study of different initial concentrations of ferrous iron, it
indicates that higher ferrous initial concentration (>10 mg/L), the kinetic constant was influenced by heterogeneous reaction.
Therefore, the kinetic constant values estimated are possibly exceed than the actual value as it is the combination of
homogeneous and heterogeneous reaction kinetic constants. It also found that dissolved oxygen supplied to an aeration system
has a pronounced effect on ferrous iron oxidation performance when it was provided in low condition, however, it does not
influent if excess amount of oxygen was provided or oxygen concentration reached the saturation. However, air flow rate may
not only significant impact only on dissolved oxygen level in diffused aerator, but also the hydrodynamic characteristics which
resulted in heterogeneous mechanism based on ferrous adsorption on ferric hydroxide particles. The study showed that the
rate of ferrous iron oxidation was significantly affected by the presence of ferric hydroxide and proportionally increased with
its concentration. Ferric hydroxide acts as an effective catalyst for the oxygenation. Heterogenous reaction occurred through
adsorption mechanism highlights the good mixing ability in multi-phase reactor including MALR. In practical aspect, this
effective catalyst can be enhanced by recirculating the ferric hydroxide particles from ferrous oxidation byproducts in a slurry
phase, consequently, possible saving on reactor size through the decreasing residence time.

Keywords: Ferrous iron, Groundwater, Kinetic, Oxidation-precipitation

92
I. Introduction II. Materials and Methods
Groundwater is a key resource of drinking water. In 2.1. Experimental set-up
Cambodia, more than haft of the population relies heavily MALR presented as a rectangular internal-loop column
on groundwater resources to deal with water shortages in dry for containing 6.6 litter total working volume was used in
season when surface water is not available. Even it is looked this study. Compressed air pump and nitrogen gas tank were
so clear and clean, some contaminants, i.e., iron, arsenic, connected to rigid stone air spargers at the riser bottom of
fluoride, etc., are commonly found. Since it is often mildly the reactor controlled by the mass flow meter, as shown in
acidic and devoid of dissolved oxygen, iron presents in Fig. 1. A DO probe meter was installed in the middle of
groundwater mostly in a form of soluble ferrous iron [1]. Its downcomer compartment for measuring dissolved oxygen
presence results an unpleasantness of taste in mouth and concentration in aquatic solution. A pH meter was used to
anesthetic red or brown stains on clothes and sanitary measure pH and temperature values in aquatic solution as
facilities or even foods. Therefore, the presence of iron the gas separation zone. The samples were sampled from the
likely the most common water problem faced by consumers gas separation zone of the reactor before preparing for
and water treatment operators [1]. A recommendation of analysis. All experiments were conducted at the room
World Health Organization for maximum iron in drinking temperature.
water is 0.3 mg/L as guideline in order to avoid aesthetic and
organoleptic problems. However, a range between 0.7 and
80 mg/L have been reported as a ferrous iron concentration
in groundwater [2]. Seriously, iron concentration in
groundwater of Cambodia up to 100 mg/L was documented
by RDI-Cambodia (2016) [3]. Consequently, removal of
iron from water is really necessary.
Iron removal from water can be done through
physicochemical or biological process of a conventional
water treatment. Oxidation-precipitation process followed
by sedimentation or/and filtration is the most common
treatment pathway. Soluble iron (Fe2+) will be oxidized to
insoluble iron (Fe3+). Among all oxidation processes,
aeration process is usually considered for ferrous iron
oxidation in water with high concentration (higher than 5
mg/L) and it also could avoid costs on chemicals used.
Based on technological application, water can be
saturated by oxygen through various ways, i.e., diffused
Fig. 1. Experimental set-up of MALR for ferrous oxidation
aerator [4], orifice spay reactor [5], and open-air cascades or
trays [6]. In the last decade, gas-liquid reactors including
2.2. Experimental procedure
bubble column and airlift reactors have been used as
Experiment was carried out in the 6.6 liter working
diffused aerators for ferrous iron oxidation [1,2] and
volume of MALR under semi-batch conditions at room
manganese oxidation [7] due to its benefits such as good
temperature. The influence of process conditions, i.e., initial
mixing, low energy consumption, gas-liquid mass transfer,
pH (between 6.5 and 8.0), gas flow rate (between 0.5 and 2
and pH control properties.
L/min), initial ferrous iron concentration (3 and 20 mg/L),
The objective of this work is to study the influence of
and insoluble ferric hydroxide particle (between 0 and 20
operation parameters, i.e., ferrous initial concentration,
mg/L). Values of pH, dissolved oxygen, and ferrous
initial pH, supplied gas flow, and additional ferric hydroxide
concentration were measured along the time.
particles on ferrous iron oxidation kinetic under batch
The synthetic waters were prepared by deoxygenated tap
conditions using Modified Airlift Reactor (MALR).
water with ferrous sulphate. Nitrogen gas was used for
reducing the initial oxygen concentration in tap water to less

93
than 1 mg/L and for mixing during the synthetic and pH d [Fe2+ ]
- ≈ k [Fe2+ ] (Eq.1)
adjustment. Diluted acid and base were prepared for pH dt
adjustment in aquatic solution. For the effective study of
additional ferric hydroxide particles on ferrous oxidation, Fitting experimental result into Eq. (1), k values for each
ferric particle in sludge appearance was used. Ferrous condition can be estimated from the slopes of Fig. 2. The k
samples were recovered from the gas separator zone within values for initial pH 6.5, 7.0, and 8.0 are 0.013, 0.049, and
every 5 minutes for analysis. After sampling immediately, it 0.291 min-1, respectively. In conclusion as practical aspect,
was mixed with HCl in order to limit future ferrous iron oxidation is strongly dependent on pH which
oxidation; then, it was added 1,10-phenanthroline to react emphasizes a good ability to maintain a constant pH of gas-
with ferrous iron and resulted in a color. Last, its liquid reactor such as MALR. Last, around neutral pH
concentration was measured by UV spectrophotometer. should be considered for controlling iron oxidation reaction.

III. Results and Discussion 3.2. Effect of ferrous initial concentration

A slightly drop of aquatic pH approximate 0.2 was
3.1. Effect of initial pH observed for low initial concentration of ferrous iron (≤ 10
Ferrous oxidation performance in different initial pH mg/L) within 15 minutes before subsequent increase at the
was examined in semi-batch experiments using 10 mg/L of end (data not shown). However, a significant decrease of pH
ferrous initial concentration and gas flow 1 L/min without value was observed for the highest ferrous initial
pH constant maintaining. The result showed that 47% concentration from for an hour aeration time. Experimental
removal efficiency was obtained at the lowest initial pH results obtained in semi-batch experiments for an initial
within 45 minutes aeration. Time required to achieve more ferrous concentration equal 3, 10, and 20 mg/L, keeping all
than 90% removal efficiency for initial pH 7.0 and 8.0 are other experimental conditions identical with initial pH 7.0
40 and 20 minutes, respectively. Fig. 2 presents the are reported in Fig 3. Result showed ferrous concentration
evolution of ferrous iron with time for different initial pH decreases along an aeration and its oxidation is much slower
values. The result of ferrous iron followed an exponential when its initial concentration increases. It can be explained
decrease demonstrated a first-order removal mechanism. It based on the strong pH-dependency of ferrous iron oxidation
can be therefore written as Eq. (1) below, where k is pseudo- by air. Only initial pH adjustment before starting the
first-order constant presented in min-1 unit. experiment without pH constant controlling was performed
in this experiment.
1
1
y = 0.992e-0.013x
R² = 0.974 y = 0.98e-0.03x
R² = 0.99
[Fe2+]/[Fe2+]0

[Fe2+Z]/[Fe2+]0

0.1
y = 1.028e-0.049x
0.1
R² = 0.993
y = 0.98e-0.08x
Initial pH = 6.5 R² = 0.99
y = 1.03e-0.05x
y = 1.061e-0.291x Initial pH = 7.0 R² = 0.99
R² = 0.997 Initial pH = 8.0 [Fe(III)]o ~ 3 mg/L
0.01 [Fe(II)]o ~ 10 mg/L
0 20 40 60 [Fe(II)]o ~ 20 mg/L
0.01
Time [min] 0 20 40 60
Fig. 2. Semi-logarithmic plot of ferrous iron concentration Time [min]
with aeration time in function of initial pH value Fig. 3. Semi-logarithmic plot of ferrous iron concentration
with aeration time in function of initial concentration

94
 Consequently, higher concentration of ferrous iron must In order to investigate the other constants of Eq. (2), the
particularly drop aquatic pH value than lower ones along the additional experiments will be examined for validating the
same aeration time, followed ferrous oxidation proposed expression. A study of additional ferric hydroxide
stoichiometry equation. Lower aquatic pH surely resulted in particles dispersed in the system will be analyzed in order to
lower oxidation performance as higher pH could faster clarify heterogeneous reaction. Moreover, since the effect of
oxidize ferrous iron as aforementioned. Even increasing ferric iron particles was occurred through adsorption of
initial concentration of ferrous iron could not enhance its ferrous iron on surface of iron particle, liquid flow pattern as
conversion yield due to the strong effect of aquatic pH, it well as recirculating ferric particles should be considered.
was confirmed in many researches that iron oxidation Therefore, different gas flow rates will be investigated in
byproducts, ferric iron has a contribution on ferrous order to study not only the effect of oxygen distribution on
oxidation rate through catalytic effect as a heterogeneous ferrous oxidation through homogeneous reaction, but also
mechanism which allows ferrous iron adsorption on ferric effect mixing condition on ferrous removal through
hydroxide particles together with homogeneous mechanism. heterogeneous reaction.

0.1 3.3. Effect of supplied gas flow rate

Evolution of ferrous iron removal as a function of
3 mg/L
0.08 supplied gas flow rate was studied in semi-batch
experiments in order to evaluate the effect of gas flow with
0.06 10 mg/L ferrous initial concentration and initial pH 7.0.
k [min-1]

10 mg/L Effect of supplied air flow on ferrous oxidation was

0.04 demonstrated in Fig. 5. The plots were started until 35-50
20 mg/L minutes aeration. The experimental result indicates that
0.02 This study: higher air flow increases ferrous oxidation yield. Aeration
k = 0.0941 1014 [OH-]2
time to achieve at least 90% removal efficiency for gas flow
0 0.5, 1.0, and 2.0 L/min are 60, 40, and 35 minutes,
-0.2 0.3 0.8 respectively.
1014 [OH-]2 [mol/L]
Fig. 4. Comparison of aquatic pH presented in [OH-]2 on 1

kinetic constant values

y = 1.0372e-0.0393x
R² = 0.9953
A comparative plot of the aquatic pH and aquatic pH
[Fe2+]/[Fe2+]0

represented for different ferrous initial concentrations

influence on k value were illustrated in Fig. 4. If there is only 0.1 y = 1.0973e-0.0663x
R² = 0.9804
a homogeneous reaction in the system, experimental data y = 1.0284e-0.0488x
R² = 0.9932
sets must follow the model line constructed from the
previous section results. Based on the results, it clearly Qg ~ 0.5 L/min
indicates that at higher ferrous initial concentration (>10 Qg ~ 1 L/min
mg/L), the kinetic constant was influenced by heterogeneous Qg ~ 2 L/min
0.01
reaction. Therefore, the k values estimated in Fig. 3 are
0 15 30 45 60
possibly exceed than the actual value as it is the combination Time [min]
of kinetic constant between homogeneous and
Fig. 5. Semi-logarithmic plot of ferrous iron concentration
heterogeneous reactions. Kinetic equation can be derived to
with aeration time in function of supplied gas flow rate
Eq. 2, where k1 and k2 represented kinetic constant of
homogeneous and heterogeneous reactions, respectively.
There are two possible reasons caused a better ferrous
oxidization after increasing air flow rate supplied to diffused
d [Fe2+ ]
- = k [Fe2+ ] = (k1 + k2 [Fe3+ ]) [Fe2+ ] (Eq.2) aerator. Based on the pattern of dissolved oxygen at each gas
dt

95
flow, an extensive difference can be remarkable at the first 1
[Fe(OH)3] ~ 0 mg/L
10-30 minutes after aerating from a very low concentration. [Fe(OH)3] ~ 10 mg/L
This time step is the log yield period for ferrous oxidation as [Fe(OH)3] ~ 20 mg/L
well. Between this time range, the dissolved oxygen might

[Fe2+] [mg/L]
not yet sufficient for oxidizing ferrous iron, consequently,
higher air flow resulted in faster oxidation yield compared y = 0.8964e-0.0749x
0.1 R² = 0.9934
to the lower ones. This finding is in a good confirmation with
M. Ghosh (1962) [8], who found that dissolved oxygen
supplied for aeration does have a significant influence on y = 0.7429e-0.1269x
R² = 0.9675
ferrous oxidation reaction kinetics at certain amount. Other y = 0.8116e-0.0996x
R² = 0.9810
possible pathway is influence of mixing behaviors on
heterogeneous reaction. Using initial concentration of 0.01
0 10 20 30 40 50
ferrous iron 10 mg/L may result in a catalytic effect of ferric
Time [min]
hydroxide particles roleplay as an adsorbent for ferrous iron
Fig. 6. Semi-logarithmic plot of ferrous iron concentration
adsorption as found in previous section. Therefore, higher
with aeration time in function of ferric hydroxide addition
slurry mixing characteristic could advance better adsorption
performance resulted a faster ferrous removal.
For aquatic pH value 6.76 ± 0.18, the k cat values at
different additional ferric hydroxide particles were
3.4. Effect of additional Fe(OH)3 particles
estimated. The k1 and k2 constants could be calculated for a
Prepared ferric hydroxide concentration between 0 and
certain process condition, as written in Eq. (4).
20 mg/L were added into the ferrous iron water sample. In
order to obtain a constant concentration of ferric hydroxide
in the process, 5 mg/L of ferrous initial concentration was d [Fe2+ ]
- = (0.0745 + 0.0026 [Fe3+ ]) [Fe2+ ] (Eq.4)
selected as this level was expected that there is no significant dt
influence of autocatalytic effect caused by ferric iron
precipitated from ferrous oxidation [1]. Therefore, kinetic IV. Conclusion
equation can be written as Eq. (3). The present part aims to study the influence of process
conditions on ferrous iron oxidation kinetic under batch
d [Fe2+ ] conditions of MALR. Iron oxidation is strongly dependent
- = (k1 + k2 [Fe3+ ]) [Fe2+ ] = kcat [Fe2+ ] (Eq.3) on pH which emphasizes a good ability to maintain a
dt
constant pH of gas-liquid reactor such as MALR. Around
Experimental results from semi-batch condition with gas neutral pH should be considered for controlling iron
flow 2 L/min were reported in Fig. 6. Result indicates that oxidation reaction. For different initial concentrations of
higher additional concentration of ferric hydroxide increases ferrous iron, it indicates that higher ferrous initial
ferrous oxidation yield. Aeration time to achieve more than concentration (>10 mg/L), kinetic constant was influenced
90% removal efficiency when adding ferric hydroxide by heterogeneous reaction. Therefore, the kinetic constant
concentration 0, 10, and 20 mg/L are 30, 22, and 17 minutes, values estimated are possibly exceed than the actual value as
respectively. To remove ferrous iron from 0.3 mg/L to < 0.3 it is the combination of homogeneous and heterogeneous
mg/L, adding 20 mg/L of ferric hydroxide could decrease reaction kinetic constants. It also found that dissolved
aeration time almost two times, from 40 to 22 minutes oxygen supplied to an aeration system has a pronounced
compared to non-additional condition. Aquatic pH effect on ferrous iron oxidation performance when it was
decreased while increase additional ferric hydroxide, provided in low condition, however, it does not influent if
however, the kinetic constant increased with ferric excess amount of oxygen was provided or oxygen
hydroxide additional. It means that the rate of ferrous iron concentration reached the saturation. However, air flow rate
oxidation was significantly affected by the presence of ferric may not only significant impact only on dissolved oxygen
hydroxide and increased with its concentration. level in diffused aerator, but also the hydrodynamic
characteristics which resulted in heterogeneous mechanism

96
based on ferrous adsorption on ferric hydroxide particles. [8] Ghosh, M. 1962. A study of the rate of oxidation of
The study showed that the rate of ferrous iron oxidation iron in aerated ground waters. Sanitary Engineering
was significantly affected by the presence of ferric Series 012.
hydroxide and proportionally increased with its
concentration. Ferric hydroxide acts as an effective catalyst
for the oxygenation. Heterogenous reaction occurred
through adsorption mechanism highlights the good mixing
ability in multi-phase reactor including MALR. In practical
aspect, this effective catalyst can be enhanced by
recirculating the ferric hydroxide particles from ferrous
oxidation byproducts in slurry phase, consequently, possible
saving on reactor size through decreasing residence time.

Acknowledgement
This research has been supported by AUN/SEED-Net
Program of JICA through Collaborative Research. Authors
also acknowledge Project for Strengthening Engineering
Education and Research for Industrial Development in
Cambodia of JICA through LBE Research Grant.

References
[1] El Azher, N., Gourich, B., Vial, C., Soulami, M. B.,
and Ziyad, M. 2008. Study of ferrous iron oxidation
in Morocco drinking water in an airlift reactor.
Chemical Engineering and Processing: Process
Intensification 47(9-10), 1877-1886.
[2] Stiriba, Y., Gourich, B., and Vial, C. 2017.
Numerical modeling of ferrous iron oxidation in a
split-rectangular airlift reactor. Chemical
Engineering Science 170, 705-719.
[3] RDI-Cambodia. (2016). Summary of Groundwater
Data. http://www.rdic.org. Accessed 2017.
[4] Roustan, M. 2003. Transferts gaz-liquide dans les
procédés de traitement des eaux et des effluents
gazeux. TEC&DOC.
[5] Klein, D. B., and Neufeld, R. D. 2005. Use of a
multiple orifice spray reactor to accelerate ferrous
iron oxidation in acidic mine water. Mine Water and
the Environment 24(4), 186-193.
[6] Rathinakumar, V., Dhinakaran, G., and Suribabu, C.
2014. Assessment of aeration capacity of stepped
cascade system for selected geometry. International
Journal of ChemTech Research 6(1), 254-262.
[7] Kouzbour, S., et al. 2017. Removal of manganese (II)
from drinking water by aeration process using an
airlift reactor. Journal of water process engineering
16, 233-239.

97
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Evaluation of Wastewater Treatment Efficiency Utilizing Coconut Fiber as Filter Media

Chenda Lai1*, Thary Vorn2, Boreborey Ty3

1
Water and Environmental Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
2
SUDrain, Impact Hub Phnom Penh, St 306 #17, Phnom Penh, Cambodia
3
Faculty of Chemical and Food Engineering,
Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, Phnom Penh, Cambodia.
*Correspoding author: chendalai7@gmail.com

Abstract

The system using coconut fiber as filter media (CFFM) was developed for wastewater treatment in this study. The aim of
this experiment is overall to estimate the efficiency of CFFM with various media ratios and hydraulic retention time (HRT)
for wastewater pollutant reduction. Two reactor tanks were designed with 14cm3 and complied with low media ratio (LMR
V/V=15%) and high media ratio (HMR V/V =30%). All parameters were measured after treated 24 hours, 3 days, 1 week, 2
weeks, and 4 weeks called Phase-1, and the process was repeated in Phase-2 using the old media from Phase-1. Nutrient
compositions such as NO3-, NH4+, SO42-, and PO43- were determined by IC while heavy metals were determined by ASS. The
results showed that HMR had no significant effect on the removal while the volume increased 2 times (15%-30%). It is noticed
that LMR of V/V=15% with 2 x 2 cm and 0. 19g/cm3 density was a good condition for this wastewater pollutant load. The
percentage removals of COD, Cu, Cd, NO3-, NH4+, andSO42- were only slightly increased between 1 day and 3 days of HRT
but it showed a remarkable change at 1week of HRT with 7.2% (SO42-) minimum and 25 % maximum (COD) of increasing.
The maximum removal of COD in Phase-1 was 78.01% and 93.68% in Phase-2 while Cu was considered as the highest (about
80%) among heavy metals. Besides that, SO42- was the best removal (81.25%) of nutrient compositions, whereas Fe was the
lowest one among all studied parameters. It was proved that the removals were related to the HRT; however, 1week of HRT
is the better choice for pollutants reduction in terms of efficiency, time, and space-saving.

Keywords: Coconut fiber, Heavy metals, Media ratio, Nutrient compositions

I. Introduction compound (3%), water soluble(5%), ash and other

compound (7.8%) [3]. Sato (2017) did the microsomes
Poor environmental conditions and the lack of pollution
experiment with the bundle of coconut fiber 0.2m length.
control lead to disease and poor sanitation contributes to
The maximum removals in synthetic sewage were (30.5%)
environmental degradation [1]. Coconut fiber is noticeable
of BOD, 10.2% of COD, 30.8% of TN, and 21.8% of TP
to be one of the potential filtering media for pollutant
while in synthetic leachate, the maximum removal were
reduction [2]. The chemical components of coconut fiber are
45.2% of BOD, 46.5% of COD, 41.2% of TP, and 32.9% of
lignin (about 45.8%), cellulose (43.4%), pectin and related
TN [4].

98
 24h, 3 days, 1 week,
The aim of this study is to evaluate the efficiency of HRT -
2 weeks, 4 weeks
coconut fiber for wastewater pollutant reduction, whereas
3
two specific objectives are taken into account: (1) designing, Flow rate cm /h 24h HRT: 500
and (2) optimizing lab-scale of wastewater treatment (WWT)
using biological filter from recycling coconut waste as a 2.3. Analytical methods
filter media and evaluating the efficient removal of 2.3.1. pH, Temperature, EC, and Turbidity
wastewater treatment utilizing coconut fiber in different pH and EC were measured using pH meter HM-30P and
media ratios (MRs) and hydraulic retention times (HRTs). EC meter(hold) respectively while Turbidity was measured
with Turbidity meter model HI98703.
II. Materials and Methods
2.1. Sampling Sites 2.3.2. Total Suspended Solid
Wastewater was collected from the wastewater Cellulose Nitrate Filter Papers with pore size 0.45µm
treatment system at Phnom Penh Special Economic Zone. were used to separate liquid and solid following the method
The wastes are mainly from the small medium enterprises 2540 of Eaton et al. (2005). The formula for TSS calculation
(SME) and factories in that zone. was followed as in Eq. 1 below

2.2. Experimental set-up (F-I)(g) x 1000mg/L

TSS mg/L = (Eq. 1)
In order to set up the experiment, two acrylic rectangular sample volume in L
tanks shaped with an operation volume of 14cm3 were
designed and complied with the coconut fiber media ratio of Where
low media ration. The specification set-up of the coconut F= final dried weight of the filter (mg)
fiber filter media (CFFM) system is shown in Table 1.
I = Initial weight of the filter (mg)

2.3.3. Chemical Oxygen Demand (COD)

There are COD low range (LR) capacity with 0-
150mg/L, and COD high range (HR) capacity with 20-
1500mg/L following the method of 5220-COD/SM. They
were added with 2mL of the reagent test tube, heated in
HATCH DRB-200 for 2h and 150 oC, and cool down it
before analysis with HATCH DR 1000 by COD LR and
COD HR method.

2.3.4. Heavy Metal Analysis

Fig. 1. Experimental set-up The samples were filtered with 0.45µm filer paper using
auto air pumps. The standard condition of each metal was
Table 1. Specification set-up of coconut fiber prepared by Fe (0ppm, 0.1ppm, 0.5ppm, 1ppm, and 2ppm),
Pb (0ppm, 0.5ppm, 1ppm, 2ppm, and 5ppm), Cd (0 ppm,
Modeling Unit Condition 0.05ppm, 0.1ppm, 0.2ppm, and 0.5ppm), and Cu (0ppm,
0.1ppm, 0.5ppm, 1ppm and 2ppm). The samples were
Rectangular
Media Shape and analyzed with AAS model AA-7000 SHIMADZU.
cm Surface: 2 x 2
Size
Height: 1 2.3.5. Nutrient Composition Analysis
3 Nutrients composition in terms of the cation (NH4+) and
Density g/cm 0.19
anion (SO42-, PO43, and NO3-,) were determined by Ion
Chromatography (IC) model IC CTO-20A. The samples

99
 100
need to be filtered to remove particles larger than 0.45µm
Blank-1
90
before analysis (Eaton, 2017). LMR-1
80 HMR-1
70

% Removal of COD
III. Results and Discussion
60

50

Influent wastewater was analyzed in terms of physical 40

and chemical parameters and its characteristic is shown in 30

Table 2. 20

10

Table 2. The characteristic of influent wastewater 0

1 day 3 days 1 week 2 weeks 3 weeks 4 weeks
HRT

MoE. Std Fig.2. % Removal of COD in Phase-1

Parameters Unit Value
(1) The media provided a good treatment of SO42- since the
pH - 7.88 ± 0.16 5-9 first HRT (67.43% and 69.40% for LMR and HMR). As
oC shown in 8, there was no significant difference in removal
Temperature 27.82 ± 0.21 <45
efficiency between LMR and HMR. The removal involves
TSS mg/L 404.33 ± 4.24 <50 with the biofilm and H2S transform in anaerobic condition.
Turbidity NTU 85.03 ± 2.78 None 100
Blank-1
90
EC mS/cm 912.33 ± 130 None LMR-1
80 HMR-1
TDS ppm 582.33 ± 8.48 <1000
70
% Removal of SO42-

COD mg/L 425.50 ± 15.32 <50 60

Cd ppm 2.61 ± 0.31 <0.1 50

40
Cu ppm 2.90 ± 0.14 <0.2
30
Fe ppm 0.13 ± 0 .033 <1.0
20
Pb ppm 2.66 ± 0.32 <0.1 10

NO3 - ppm 11.58 ± 1.52 <10 0

1 day 3 days 1 week 2 weeks 3 weeks 4 weeks

NH4+ ppm 24.22 ± 3.36 None
HRT
SO42- ppm 341.50 ± 28 <300
PO43- ppm 140.01 ± 43.36 <3.0 Fig.3. % Removal of Fe in Phase-1

3.1. Evaluation of Treatment Performance in Phase-1 3.2. Summary of All Percentage Removals and
and Effect of Media Ratio Parameters Comparatives
The removal efficiency of COD (Fig.5.) indicates that Turbidity in Phase-1 provided the best removal in terms
in 1 day of treatment the percentage removal started from of the physical parameters in LMR-1 condition which
40.18% of LMR and 41.63% of HMR. It is not a significant showed the maximum reduction of about 98% at 4 weeks
difference between LMR and HMR. Dharmarathne et al. HRT. The performance was the same in LMR-2 condition.
(2013b) also reported that HFD (high fiber density) and LFD However, in terms of chemical parameters, Cu was the best
(low fiber density) treatment tank showed a very parallel parameter in LMR-1, whereas COD was the best one for
trend in the COD removal [5]. The removal of COD is LMR-2. Irons provided the lowest removal among all heavy
related to the biofilm which grows on the surface of media. metals in both phases while PO43- was the lowest one in
terms of nutrient compositions.

100
 100 Turb removal of COD in Phase-1 was 78.01% and 93.68% in
TSS
90 Cu
Phase-2, whereas Fe was the lowest removal among all
80 COD studied parameters. This study is beneficial for the further
% Removal for LMR-1

SO42-
70
Cd
research as it provided the data of treatment by using
60 NO3- coconut fiber which is the low-cost materials as the filter
50 Pb media.
NH4+
40
PO43-
30 Fe ACKNOWLEDGMENTS
20
10
This research was financially supported by AFD-EU;
0
laboratory support by SATREPS lab, Ministry of
1 day 3 days 1 week 2 weeks 3 weeks 4 weeks Environment lab, Phnom Penh Special Economic Zone.
HRT

Fig.4. % Removal in all condition for LMR-1 References

TSS
100
Cu
[1] Tyagi, S., Garg, N., & Paudel, R. (2014).
90
Turb Environmental Degradation: Causes and
COD
80
Consequences. European Researcher, 81, 1491.
% Removal of HMR-1

Cu
70
SO42- [2] Islam, T., Islam, S., Saifullah, I., Datta, D., &
60 Cd
Ahmed, A. (2017). Suitability of Recycled Coconut
50 Pb
NH4+ Fiber as Filter Media for the Treatment of
40
30
PO43- Wastewater. Proceedings of 5th International
Fe
20
Conference on Solid Waste Management in the
10
Developing Countries, Wastesafe-2017, Khulna,
0 Bangladesh.
1 day 3 days 1 week 2 weeks 3 weeks 4 weeks
[3] Sagar, S. S., Chavan, R. P., Patil, C. L., Shinde, D.
HRT N., & Kekane, S. S. (2015). Physico-chemical
parameters for testing of water-A
Fig.5. % Removal in all condition for HMR-1 review. International Journal of Chemical
Studies, 3(4): 24-28.
The summary of percentage removals in both phases of [4] Sato, N. (2017). Microcosm Experiments on a
HMRs were shown in Fig. 5. Cu was the best element Coconut-Fibre Biofilm Treatment System to
among chemical parameters in HMR-1, whereas COD Evaluate Waste Water Treatment Efficiencies. S.,
became the best for HMR-2 with maximum removal of Garg, N., & Paudel, R. (2014). Environmental
91.68%. It was also observed that there was no difference Degradation: Causes and Consequences. European
between HMR and LMR on the iron performances which Researcher, 81, 1491.
provided the lowest removal among heavy metals [5] Dharmarathne, N., Sato, N., Kawamoto, K., Sato,
parameters while PO43- was the lowest one among nutrient H., & Tanaka, N. (2013, November 1). Evaluation
compositions. of Wastewater Treatment Efficiency Using Coconut
Fiber Biofilm Reactor System with Synthetic
IV. CONCLUSIONS Leachate.

Based on the results from the CFFM system, the finding

of this study can be summarized as follows: The CFFM
could be successfully operated by LMR (V/V=15%) with
this type of wastewater. The higher MR did not show better
results while the volume increased 2 times. The maximum

101
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Current Presence and Possible Repercussions of UV Filters in Coral Reef in Okinawa

Prefecture

Jorge Garcia-Hernandez*, Rajendra Khanal and Chihiro Yoshimura

Department of Civil and Environmental Engineering

School of Environment and Society
Tokyo Institute of Technology
2-12-1-M1-4, Ookayama, Meguro-ku, Tokyo, 152-8552, Japan
* garcia.j.aa@m.titech.ac.jp
Abstract

Ultra-violet (UV) filters are pollutants of arising concern due to its persistence in water environment. In previous
researches, UV filters have been reported in pristine coral environment in Okinawa, Japan. The main objective of this research
is to review the spatiotemporal variation of 12 of the 16 UV filters, categorized as 3 groups by the Food and Drug
Administration (FDA) on the Act to Modernize the Regulation of Sunscreen Products in the United States proposed in 2019.
The target area is the Japanese coral reef environment in Ryukyu Trench, Ryukyu Islands, and Okinawa trough, for
highlighting its toxicological and bioaccumulating impacts on coral reefs ecosystems. The web of science database from 2000
to 2020 was searched with the key words: anthropogenic activities, persistent organic pollutants, sunscreens, UV filters, UV
protection, sunblock, UV stabilizer, coral, Japan, and Okinawa. Most of the studies agreed that organic UV filters such as
oxybenzone, octinoxate or octocrylene significantly degrade the water quality, which may have potential risk to coral
ecosystems. In addition, the wide spread of UV filters has shown a perceivable presence in some Okinawa beaches, containing
around 1.4 µg/L oxybenzone during summer season. Nevertheless, oxybenzone at some reef sites have been found to be
around 0.01 µg/L which is almost 500 times lower than the LC50 range of 5.4 to 14.5 µg/ for Acropora cervicornis larvae.
More toxicological and bioaccumulate studies on coral reef bleaching by UV filters have to be performed to support this
statement and re-evaluate their ecological risks in coral reef ecosystems at different season and complex environmental
exposure.

Keywords: UV filters, Okinawa, UV protection, coral, Oxybenzone

I. Introduction possible coral bleaching. Nevertheless, a few studies have

focused on the Japanese coral reefs impact due to the use of
In recent years, the concern about micropollutants and
sunscreen chemicals.
their impact in the environment have increased. Sunscreens
Japanese coral reefs are located along the boundary of
and other Persistent Organic Pollutants (PPPs) have received
the Philippine sea plate and is composed by the Ryukyu
particular attention from several researchers around the
Trench, Ryukyu Islands and Okinawa Trough in the
world due to the concern about their potential impact in the
Okinawa prefecture [1]. Each reef is characterized by the
environment. Several studies have focused on the
property of the island, such as the soil type, land-use land
ecotoxicological effects of sunscreen pollutants in marine
cover, and precipitation regime. The climatic conditions and
biodiversity, potential bioaccumulation problems and
the crystalline water create a perfect place to receive

102
 thousands of tourists each year with Okinawa islands being These are substances with almost null absorption of visible
the most visited one. Coral reefs around the main island in radiation but important light absorption in the UVA (315-
Okinawa is located a few meters off the coast. 400nm) and UVB (280-315nm) [7], moreover, regarding
Several activities such as scuba diving and snorkeling are their mechanism of action, can be classified into organic
popular and it attracts tourists to discover the beauty of the (chemical) absorbers and inorganic (physical) blockers.
Japanese coral reefs. According to the tourism minister of A detailed sunscreen classification has been done by the
Japan, in 2017 over 9 million people visited Okinawa and FDA (Food and Drugs Administration) in 2019 in the United
the trend tends to increase year by year possibly to 12 States, where sunscreen products for over-the-counter
million people by 2021 [2]. However, there are several human use are classified in three main categories called
environmental issues to face as the tourism is expanding GRASE (Generally Recognized as Safe and Effective).
such as coastal development and construction, land The main objective of this research is to review the
reclamation and degradation of coral reefs. spatiotemporal variation of 12 of the 16 UV filters
(Avobenzone, Cinoxate, Dioxybenzone, Ensulizole,
Homosalate, Meradimate, Octinoxate, Octisalate,
Octocrylene, Oxybenzone, Padimate O, Sulisobenzone),
categorized as 3 groups by the FDA on the Act to Modernize
the Regulation of Sunscreen Products in the United States
proposed in 2019.

II. Methodology
We conducted this review using the web of science,
filtering data from 2000 to 2020 using the keywords:
anthropogenic activities, persistent organic pollutants,
sunscreens, UV filters, UV protection, sunblock, UV
Fig. 1. Japanese coral reefs. Provided by Marine Biotic stabilizer, coral, Japan, and Okinawa.
Environment Survey. From this screening process, 40 different articles were
identified to highlight the presence of persistent organic
Coral reefs declining has been reported in the 1980’s
pollutants in Japanese marine environment. Nevertheless,
where approximately 40% of the coral bleached and 10%
just 2 of them reported concentrations and affectations in
died in Sesoko and Okinawa Island [1]. The main stressors
Japanese coral reefs by sunscreen UV filters [5,6].
that influence in coral bleaching are solar irradiance, global • UV filters, UV
warming and anthropogenic activities like municipal and Data base protection,
sunblock, UV
industrial wastewater effluent and micro-pollution. searching stabilizer,
key words coral, Japan, • 40 articles
Micropollutants directly and indirectly enter the aquatic and Okinawa about the
presence of
environment via industrial or municipal wastewater persistent
organic
effluents, stormwater runoff, or recreational activities. Screening pollutants in
process Japanese
Several studies have focused on the ecotoxicological effects marine
environment
• 2 articles just
addressed
of sunscreen pollutants in marine biodiversity, proving the . the
problematic
presence of Oxybenzone (BP-3) in Scrobicularia plana, an Final about the
sunscreen
indicator species for assessing the health statues of estuarine screening pollutants in
Japanese
and coastal ecosystems [3]. Furthermore, in situ and coral reefs

laboratory experiments in several tropical regions have Fig. 2. Screening process of literature review.
shown coral bleaching of communities, such as
Zooxanthella, where the organic ultraviolet filters induce the III. Results
lytic viral cycle with latent infections [4]. 3.1. UV filters presence in coral reefs in Okinawa
UV-filters are widely popular to offer protection to Prefecture
human skin against deleterious effects of UV radiation. High level of organic pollutants (Organo-tin compounds

103
(OTCs), organochlorine pesticides (OCPs) and Oxybenzone were detected in coral tissue from three
polychlorinated biphenyls (PCBs) were detected in surface important tourist sports at Oahu island (Ka’a’awa, Waikiki
water and sediments of some rivers and coastal waters of beach and Kaneohe Bay). The detection frequencies of the
Okinawa [6]. OTCs were found at considerable four chemicals were 100%, 100%, 96% and 82%, showing
concentrations in sediments that could be a possible threat a the presence of those compounds in Hawaiian coral reefs.
to marine life including coral communities around Okinawa In this same study [10], eight different UV filters were
Island. This study refers to a large number of chemical detected in marine environments, especially in Kaneohe bay,
groups and non-sunscreen chemical was detected or where the concentration of BP-3 was higher before water-
consider in this study. Nevertheless, this study set the ball users arrived at this site, than later in the morning after
rolling for further investigations and monitoring of coral reef people arrived. The first sample at 8am contained 6.6 ± 0.7
ecosystem to elucidate the effects of toxic organic pollutants ng/L BP-3 and 2.8 ± 0.1 ng/L of OC, then around 10 am and
on the subtropical marine life around Okinawa Island. with more than 120 people around the coral reefs another
Yutaka Tashiro performed a constant sampling once a sample was taken showing an increased to 27.0± 18.4 ng/L
month in Okinawa main Island for analyzing the occurrence and 23.6 ± 19.0 ng/L, respectively [10], assuming that
of UVFs and UVLs in seawater [5]. The results varied due sunscreen primary source in coastal water is due to the
to the fact that the distribution of the chemical in the beach increased of swimmers around the area, causing different
water is not uniform and is strongly affected by water fluctuations in the final exposure in coral reefs.
movement. The maximum concentration of BP-3 (1.34 µg/L) The two research articles from the Okinawa prefecture
detected at Sunset beach in July was one fifth of its PNEC are considered the pioneers for addressing the environmental
value (6 µg/L). In fact, the concentration of UV chemicals at impacts of sunscreen pollutants this field in Japan. More
the reefs were much lower that the estimated lethal research has to be performed based on the UV filters are the
concentration, however it was suggested that more studies most consumed in the Japanese sunscreen market. It can be
over the range of concentrations detected and exposure time inferred that, according the investigations addressed in this
are strongly necessary to prevent impact in coral ecosystems. review paper, the main impacts of chemical filters are
In this same investigation, the wide spread of UV filters has bleaching of hard corals, damage of coral larvae, and
shown a perceivable presence in some Okinawa most visited possible biomagnification and bioaccumulation issues.
beaches, containing around 1.4 µg/L oxybenzone during Coral bleaching in Okinawa prefecture have occurred
summer season. Nevertheless, oxybenzone at some reef sites before in 1998, 2001, 2008 and recently 2016 [12]. The
have been found to be around 0.01 µg/L which is almost 500 Ministry of the Environment Government of Japan have
times lower than the LC50 range of 5.4 µg/L to 14.5 µg/L reported that in 2016 in Sekisei Lagoon around 90% of the
for Acropora cervicornis larvae. reef presented damage from coral beaching, including 70%
that died by the bleaching phenomenon” [12]. Several
IV. Discussion investigations on Japanese coral reefs damage have been
4.1. Importance of UV filters shown that temperature rising and ocean acidification are the
Based on the production and consumption of cosmetic main factors leading to the stress and decline of the coral
sun products, commonly people applied around 65-130mg communities. Nevertheless, Persistent Organic Pollutants
of sunscreen each time. In Germany, annually 7,900 tons of such as sunscreen UV filters in Japanese coral reefs can
sunscreen are consumed assuming that 20,000 tons are contribute to the bleaching problem.
released in the northern Mediterranean [9]. According to As the consumption of this products increases, the
some estimations, 10% of the sunscreens produced in the concern about the possible repercussions of UV filters in the
world are normally used in tropical areas with coral reefs [9]. environment raises up. Japanese cultural background about
Nevertheless, this statement was done in 2004, and based on skin protection is an influential factor to consider in the
the gradual increase of sunscreen products sales, the amount consumption, especially when the skin beauty is important
of these chemicals entering to the environment is unknown. for some sectors of the society. The cultural background and
A novel report [10] was released in 2019 in Hawaii, the desire for a white and spot free skin is coming from the
where UV-filters were reported in coral tissue in Hawaiian Bihaku or skin whitening culture which is a common
coral reefs. Octinoxate, Homosalate, Octocrylene and practice by the majority of the Japanese woman. Bihaku

104
products sales have been increased gradually year by year clam Scrobicularia plana. Science of The Total
generated revenue of around 62 billion ¥ in 2018 [14]. Environment, 733, 139102.
Pharmaceutical industries find the Japanese skin care market [4] Danovaro, R., Bongiorni, L., Corinaldesi, C.,
as reliable and demanding. Giovannelli, D., Damiani, E., & Astolfi, P. et al., 2008.
Sunscreens Cause Coral Bleaching by Promoting
V. Conclusion Viral Infections. Environmental Health Perspectives,
Although Oxybenzone was found in Okinawa corals 116(4), 441-447.
with a concentration of 1.4 µg/L, the potential harm for [5] Tashiro, Y., & Kameda, Y. 2013., Concentration of
larvae species such as Acropora cervicornis is lower than the organic sun-blocking agents in seawater of beaches
LC50 range. However, further investigations should be done and coral reefs of Okinawa Island, Japan. Marine
to prove potential ecotoxicological risks by prolonged Pollution Bulletin, 77(1-2), 333-340.
exposure of UV filters under real conditions, to refuse or [6] Imo, S., Sheikh, M., Sawano, K., Fujimura, H., &
affirm, any affectation to the coral ecosystem. The main Oomori, T. 2008,. Distribution and Possible Impacts
motivation to elaborate this review paper is to highlight of Toxic Organic Pollutants on Coral Reef
investigations and actions other countries are taking already Ecosystems around Okinawa Island, Japan. Pacific
to face coral declining, and review the presence of 12 Science, 62(3), 317-326.
sunscreen chemicals stated in the FDA list published in 2019. [7] Giokas, D., Salvador, A., & Chisvert, A. 2007., UV
Probable data of the presence of sunscreen chemicals in filters: From sunscreens to human body and the
Japanese coral reefs will bring more investigations to this environment. Trac Trends In Analytical Chemistry,
field and consequently find alternatives to mitigate any 26(5), 360-374.
potential risk. Spatial-temporal variation of UV filters in [8] Tsui, M., Leung, H., Wai, T., Yamashita, N., Taniyasu,
Japanese marine environment and ecotoxicological studies S., & Liu, W. et al. 2014., Occurrence, distribution
in coral reef biota, are necessary to know the affectation of and ecological risk assessment of multiple classes of
sunscreen chemicals in coral ecosystems. UV filters in surface waters from different countries.
Water Research, 67, 55-65.
Acknowledgement [9] Danovaro, R., & Corinaldesi, C. 2003., Sunscreen
Products Increase Virus Production Through
Authors would like to express gratitude to Asia-Pacific
Prophage Induction in Marine Bacterioplankton.
Network for Global Change Research for funding this Microbial Ecology, 45(2), 109-118.
project - "Collaborative Research Platform to Manage Risk [10] Mitchelmore, C., He, K., Gonsior, M., Hain, E.,
and Enhance Resilience of Coral Reef in Southeast Asia, Heyes, A., & Clark, C. et al. (2019). Occurrence and
CRRP2019-08MY-Khanal". distribution of UV-filters and other anthropogenic
contaminants in coastal surface water, sediment, and
References coral tissue from Hawaii. Science Of The Total
[1] Nakano Y (2004) Index map for Chapter 6. In: Environment, 670, 398-410.
Tsuchiya M, Nadaoka K, Kayanne H, Yamano H (eds) [11] Schaap, I., & Slijkerman, D. 2018., An
Coral reefs of Japan. Ministry of the Environment, environmental risk assessment of three organic UV-
Tokyo, pp 155. filters at Lac Bay, Bonaire, Southern Caribbean.
[2] The Japan times., 2018. Okinawa Tourist Numbers Marine Pollution Bulletin, 135, 490-495.
Top Those of Hawaii For First Time: [12] Kayanne, H., Suzuki, R., Liu, G. 2017., Bleaching in
https://www.japantimes.co.jp/news/2018/02/09/nati the Ryukyu Islands in 2016 and associated Degree
onal/okinawa-tourist-numbers-top-hawaii-first-time/. Heating Week threshold., Journal Of Coral Reef
Accessed 16 December 2020. Studies, 19(1), 17-18.
[3] O'Donovan, S., Mestre, N., Abel, S., Fonseca, T., [13] Tokyoesque (2017)., Beauty, Grooming & Skincare
Carteny, C., Willems, T., Prinsen, E., Cormier, B., Market in Japan. https://tokyoesque.com/industries-
Keiter, S. and Bebianno, M., 2020. Effects of the UV japan-market/skincare-market-in-japan/. Accessed
filter, oxybenzone, adsorbed to microplastics in the 24 November 2020.

105
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Prediction of Future Land Use Change in the basin of Tonle Sap Lake Using Scenario
Analysis in CLUMondo

Sreykeo Puok*, Kong Chhuon

Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia

,Russian Ferderation Blvd., P.O. Box 86, Phnom Penh, Cambodia.

*Corresponding author: sreykeo.puok12@gmail.com

Abstract

Land-Use and Land-Cover (LULC) is an important factor for maintaining healthy ecosystem. The change
in LULC responses to different socio-ecological system and regime shifts. Due to the significant change of LULC in Tonle
Sap Lake (TSL) of Cambodia in last decade, it puts more concern on the consequence impacts to the surrounding environment
while it lacks of scientific study on how LULC in this area will be change in the future. The aim of this study is to assess the
future land use change in the basin of TSL using CLUMondo model over the period 2015-2045 based on the recent land use
update. Four scenarios were simulated based on the government policy and commitment: The Economic Land Concession
scenario, the Moderate Conservation scenario, the Conservation scenario, and the SDG-2030 scenario. All scenarios show an
increase in the urban area, annual crop, wood plantation, and agricultural land while the flooded forest remain constant. The
economic land concession or business as usual scenario cause highly change of LULC. Particularly, until 2045, the forest
cover in the basin of TSL will change differently from each scenario which are -10%, -8%, -6%, -6% for ELC, Moderate
Conservation, Conservation and SDG-2030 respectively. The shifting cultivation declines dramatically and will all disappear
by 2030 for all scenarios while grassland will be completely transferred to annual crop, wood plantation, urban area, and
agricultural land by 2045 to compensate the need of land for food production in the scenario of conservation and
SDG2030. The result of this study can guide and support future land use planning, management, and policy formulation in
the basin of TSL.

Keywords: CLUMondo model, Tonle Sap Lake, Land-Use, Scenario,

I. Introduction where the fluvial water flows in two directions during the
Land use and land cover change is an important factor in year as a result of the rain season. This interesting catchment
the study of environmental change [1]. Land use is thus a is influenced by the surrounding environment where land
central component of biophysical, social, and economic use changes impact the hydrological processes such as
systems acting across various scales. Therefore, the study of surface runoff generation, groundwater recharge and soil
land use and land cover change becomes very important to erosion. However, some studies have assessed the land cover
address the environmental issues. but not the prediction for the future change. The study on the
The basin of the Tonlé Sap Lake (TSL) is a unique basin drivers of deforestation and agricultural transformations in

106
the Northwestern uplands of Cambodia by analyzing wood production, subsistence crop, and built-up area. The
historical rejection of land use/ cover changes using a driving factor indicate the factors that can affect land use
chronological series of Landsat data and the investigated change. In this report, these driving factors were
of land cover and land use Change in Stung Chrey Bak characterized into 4 categories included, Climate,
Catchment [3,4]. Recently, there are many studies have been Socioeconomic, Soil type, and Topographic. We estimate the
conduct on the impact of land use change on water supply conversion order base on the realities of land use services
for the Angkor temple and the surrounding population and with the land use types. The conversion order is a technical
the impact of land use and climate change on water resources parameter that tells the model how to change land use to
and remand in Srepok River Basin [5,6]. The objective of fulfill demand while, the values that do not supply a good or
this paper is to systematically study of land use change with service get a 0, and all land-uses that do supply a good or
different changing scenarios using CLUMondo model over service get a value from 1 onwards.
Tonle Sap basin. This study is intended to integrate together
ArcMap (GIS) and CLUMondo model approach to identify
a change in land use pattern in the basin of Tonle Sap Lake
during 2015 to 2045.
II. Materials and Methods
2.1. Study area
The area of the basin of Tonle Sap Lake is located in the
northwest of Cambodia ( 102°00’-106°00’ E, 11°30’-14°30’
N), with a total land area of 81,372 Km2. The basin consists
of the Tonle Sap lake and 12 sub-basin which covered on 10
provinces [7]. The average annual rainfall is from 1000-
1500mm with, the daily mean temperatures between 21-
25 °C (minimum) and 30-35 °C (maximum). The average
humidity varies from 70% (March) to 85% (September) [8].
Fig. 1.Flowchart of land system simulation
The main land system divided into five classifications:
Forest (48.5%), Agricultural land (25.6%), Barren land
(13.3%), Aquatic ecosystems (12.7%), and Urban area The conversion resistance factor for each land system
(0.1%) [7]. estimated base on the costs of conversion, information on
2.2. Data Sources local policies on land use, and scenarios setup. In the model
The existing data included the land use/cover (2015) we have assigned each land-use type a dimensionless factor
were obtained from the National Institute Statistic (NIS). that represents the relative elasticity to conversion, ranging
These land use datasets separated into 9 categories (water, from 0 (easy conversion) to 1 (irreversible change).
urban area, annual crop, flooded forest, wood plantation, The conversion matrix was estimated by the model
grass land, forest, agricultural, and shifting cultivation). The which indicates what types of land use conversions are
climate and topographic got from Mekong river commission allowed. We have assigned each land-use type based on the
(MRC) and SRTM, the socioeconomic and soil data were reality of the land policy and scenario setup which 0 (no
obtained from NIS. conversion allowed) and 1 (conversion is allowed).
2.3. Model implementation 2.4. Scenario setting
To simulate the land system, change until 2045, we To capture land system change under different levels of
applied CLUMondo model with the spatial resolution of 2 × socioeconomic development and environmental
2 km to simulate the future scenarios land system in TLS. conservation, four scenarios were developed according to
This model determines the global spatial allocation of land the historical trend of land system change, the land use
system through the allocated the land system change based planning from 2015 to 2045, and the biodiversity
on the demand of goods, services for difference land conservation targets. These differences reflect the
system[1]. The overall flowchart was described in Fig. 1. importance given to different land system functions under
The demands considered in this study included cash crop, different scenarios. The first scenario Economic Land

107
Concessions (ELC) is a scenario based on only demand, grassland(0%), shifting cultivation(0%), and forest land
population, and economic. it's long-term to clear land in (21%) in Conservation and SDG 2030 scenario. For both
order to develop industrial scale agriculture and can be scenario Economic land concessions and Moderate
granted for various activities. Second, The Moderate conservation, the shifting cultivation land decrease to 0%
Conservation (MCO) is a scenario based on demand, since 2030, when the grassland remains less than 4.5% and
increasing rate of population and economic, with moderate 19% for forest land in 2045.
conservation of forest. It is balancing between demand and The model simulation presented the land system change
supply. Third, Strong Conservation (SCO) is a scenario that is a major driver of change in the spatial pattern and
based on the limit conservation of forest. The limit must be overall provision of ecosystem service. Four scenarios show
high to preserve the biodiversity or to put in the conservation that, while the demand for cash crop, wood product,
area. The last one, Sustainable Development Goals 2030 subsistence crop, and built-up area can be provided by both
(SDG-2030) is a scenario that based on the restoration to smallholders and large-scale land acquisition (LSLAs), but
achieve the 60% of forest cover over Cambodia in 2030. as the result “Water” and “Flooded forest” area would
III. Results and Discussion remain constant for an area of 353,200 ha (4.34%) and
The results from the four scenarios are presented in terms 277,200 ha (3.41%). Even though, the urban area, annual
of the simulated land system changes in TLS basin. crop, wood plantation, and agricultural land are similar to
Fig. 2 shows the resulting land system maps in 2045 increase until 2045, but other types of land system are very
under the four future scenarios which three quite different different changing. In the ELC and Moderate Conservation
land system changes, even though the demands for most scenario, most of the increasing of the urban area and annual
land system commodities and services are similar across all crop land take a place of forest cover and shifting cultivation.
scenarios. The percentage of land system change between However, we see the increasing of the urban area and annual
2015-2045 under different scenario shows in Fig. 3. In this crop take a place of grassland and shifting cultivation than
case study, the water and flooded forest remained unchanged the forest in the Conservation and SDG-2030 scenario. As
for all the scenarios, when the other land system is strongly can be seen, all the changes of each land use type are almost
changing. The total amount of agricultural land, urban area, linear except the type of water and flooded forest that
annual crop land, and wood plantation approximately remains stable.
increase to 30.5%, 8.75%, 28.85%, and 3.1% in 2045.
However, three types of land system will be takeover in

Fig. 2. Land system changes in case study under the four scenarios

108
 28.988

30.541
30.33
28.80

28.83

30.48
30.60
28.67
35.00

27.45
26.68
30.00

20.798
21.03
18.74
25.00

17.00

16.94
13.62
20.00
%

8.765
15.00

8.77
8.76

8.75
4.341

3.407
10.00

3.156
4.34
4.34
4.34
4.34

4.15
3.41
3.41
3.41
3.41
3.13

3.12
3.12

3.11

2.35
2.33

2.05

0.005
0.000

0.20
0.00

0.00
0.00
5.00
0.00
Water Urban Area Annual Crop Flooded Forest Wood Grass land Forest Cover Agricultural Shifting
Plantaion lands Cultivation
Land system 2015 Economic Land Concessions Moderate Conservation Conservation SDG-2030

Fig. 3 The percentage of land system change from 2015 to 2045

The main reason that produces the results performing 3648–3667, 2013, doi: 10.1111/gcb.12331.
like this is that all data of scenario including cash crop [2] E. F. Lambin, H. J. Geist, and E. Lepers, “ D
production, wood production, subsistence crops production, Ynamics of L and -U Se and L and -C Over C Hange
in T Ropical R Egions ,” Annu. Rev. Environ.
and built-up land were defined annually by a linear increase.
Resour., vol. 28, no. 1, pp. 205–241, 2003, doi:
IV. Conclusion 10.1146/annurev.energy.28.050302.105459.
[3] R. Kong et al., “Understanding the drivers of
LULC change is closely related to the socio-economic
deforestation and agricultural transformations in the
development of an area and local planning and policy. In this Northwestern uplands of Cambodia,” Appl. Geogr.,
study four scenarios, which considered multiple demands vol. 102, no. April 2018, pp. 84–98, 2019, doi:
for commodities and services, representing different 10.1016/j.apgeog.2018.12.006.
pathways of managing TLS's land resources, were using [4] S. Chann, N. Wales, and T. Frewer, “An
CLUMondo simulation demonstrated that land-use change Investigation of Land Cover and Land Use Change
between 2015 and 2045. The result from the analysis in in Stung Chrey Bak Catchment, Cambodia,” no. 53,
2011.
CLUMondo model can serve as an indicator of the direction
[5] K. Chim, J. Tunnicliffe, A. Shamseldin, and T. Ota,
and magnitude of change in the future. CLUMondo tool can “Land use change detection and prediction in upper
conveniently explore the possible patterns of multiple land- Siem Reap River, Cambodia,” Hydrology, vol. 6, no.
use changes under the influence of both human and natural 3, 2019, doi: 10.3390/hydrology6030064.
effects. These results of land-use models are important to [6] T. Van Ty, K. Sunada, Y. Ichikawa, and S. Oishi,
evaluate policy options and assess the impact of land-use “Scenario-based Impact Assessment of Land
change on natural and socio-economic conditions. Use/Cover and Climate Changes on Water
Resources and Demand: A Case Study in the Srepok
Acknowledgement River Basin, Vietnam-Cambodia,” Water Resour.
This work was supported by the Science and Technology Manag., vol. 26, no. 5, pp. 1387–1407, 2012, doi:
10.1007/s11269-011-9964-1.
Research Partnership for Sustainable Development
[7] C. O. Project, M. Committee, M. Committee, T. H.
(SATREPS), the Japan Science and Technology Agency E. M. Project, O. F. Mekong, and R. Commission,
(JST)/Japan International Cooperation Agency (JICA). “Learning from the lakes : IWRM implementation
in Tonle Sap Lake of Cambodia and Songkhla Lake
Basin of Thailand,” no. August, pp. 1–89, 2013.
References
[8] C. Oeurng, T. A. Cochrane, S. Chung, M. G.
[1] S. Van Asselen and P. H. Verburg, “Land cover Kondolf, T. Piman, and M. E. Arias, “Assessing
change or land-use intensification: Simulating land climate change impacts on river flows in the Tonle
system change with a global-scale land change Sap Lake Basin, Cambodia,” Water (Switzerland),
model,” Glob. Chang. Biol., vol. 19, no. 12, pp. vol. 11, no. 3, 2019, doi: 10.3390/w11030618.

109
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

_________________________________________________________________________

Water Quality Mapping Using High Resolution Satellite Image Sentinel-2

Vattanakvichea NHEM1, Sokly SIEV*1,3,7, Rattana CHHIN2, Porsry UNG1,4, Hideto FUJII5,
Chihiro YOSHIMURA6

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia Russian Federation Blvd., P.O.
Box 86, 12156 Phnom Penh, Cambodia,
2
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia, Russian Federation
Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Water and Environmental Research Unit, Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Department of Science, Technology & Innovation Promotion and development, National Institute of Science,
Technology & Innovation, Ministry of industry, Science, Technology & Innovation
5
Yamagata University, Japan
6
Department of Civil and Environmental Engineering, Tokyo Institute of Technology, Tokyo, Japan
7
Department of Science, Technology & Innovation Policy, General Department of Science, Technology & Innovation,
Ministry of industry, Science, Technology & Innovation

*sievsokly@yahoo.com

Abstract

Water quality of Tonle Sap Lake have been investigated for many years. However, the applicability of high-
resolution satellite images for monitoring spatial variations of water quality has not been studied. Therefore, this study aimed
to explore the spatial variation based on field observation and the remote sensing as a mean of water quality mapping. Water
quality including temperature, electrical conductivity (EC), total dissolved solids (TDS), salinity, and total bacteria
concentration were sampled during the dry season (March 2018) at 19 points in Chhuk Tru area and statistically fitted by the
reflectance of Sentinel-2’s bands with the central wavelength between 665 nm to 1610 nm. As the result, the coefficient of
determination (R2) of temperature and Band 8A at 865nm was 0.83. EC (R2= 0.80), TDS (R2=0.79), and Salinity (R2=0.79)
were best fitted with Band 11 at 1610 nm. Total bacteria concentration in water and sediment (R2= 0.87 and R2= 0.55) were
fitted with Band 11 at 1610 nm and band 4 at 665 nm, respectively. Those correlations are possibly due to light absorption
and reflecting properties of each water quality parameter. Those results showed potential applicability of Sentinel-2 for
monitoring the water quality, especially in the large area (e.g., floating village and Tonle Sap Lake) using fitted equations and
reflectance bands. The output of this research may provide benefits to decision makers for water quality management in
floating villages.

Keywords: High-resolution image, Sentinel-2, Tonle Sap Lake, Water quality

110
I. Introduction

Water quality is a general descriptor of water

properties in terms of physical, chemical, and biological
characteristics. Water quality parameters can be measured
by direct sampling of water which consuming and is the only
representative of a limited spatial and temporal
domain [1]. Since it is difficult to define a single water quality
standard to satisfy all kinds of uses and users, monitoring
water quality at a local to global scale is needed [2]. In the
late 1970s, satellite remote sensing for monitoring water Figure 1. The Location of Chhuk Tru Area
quality was set back because of a lack of appropriate sensors
such as a lack of a sufficient number of spectral bands as 2.2. Water quality and bacteria monitoring
well as relatively low radiometric sensitivity and low spatial
and temporal resolution. However, with the availability A lake water sample was collected near the rear of
of new satellites with a higher spatial and temporal the floating house and a storage water sample was collected
resolution like Sentinel-2, water quality retrieval and from the storage tank of a floating house. The concentration
mapping from the orbit has become more reachable. of the bacteria on both water and sediment was measured
Researchers showed that Sentinel-2 is not only can improve following the method of previous work [6]. Water depth
global inland water mapping but can offer useful rage and other water quality such as temperature, EC, TDS,
information for monitoring certain water quality Salinity and bacteria concentration of the lake was
indicators [3]. Like, Toming et al. (2016) [4] have shown the conducted in-situ measurement using YSI EXO Water
suitability of Sentinel-2 data to map different water quality Quality Sonders [7]. Descriptive statistics have been applied
such as Chl-a, Water color, CDOM, and DOC for small to show the central to delineate the spatial distribution of
inland water. But there are not many applications of high- important basic water quality parameters in order to
resolution satellite image for monitoring water quality has tendency and variation of water quality parameters for
been studied yet, especially at Tonle Sap Lake. Hence, the highlighting the specific characteristics.
spatial variation base of field observation and remote
sensing as a means of water quality mapping has been 2.3. Relating reflectance to water quality
used to determine the best-fitted equations
and reflectance bands. It is an effective tool for synoptic Image for water quality measuring were selected
water and water level monitoring, water demand modelling, from a dry season (March 2018). Since it was the best period
and water quality monitoring. for remote sensing of water quality in Chhuk Tru area.
Sentinel-2 image were available on 07 March 2018. The 19
sampling points was conducted surrounding the Chhuk Tru
The purpose of this study is to test the suitability
floating villages. The cloud coverage can cause misleading
of Sentinel-2 data for mapping different water quality
results in the analyses of surface and atmospheric parameter
parameters, namely temperature, electrical conductivity [9]
. The downloaded Sentinel-2 image from was treated to
(EC), total dissolved solids (TDS), salinity, and bacteria
mask the cloud cover by thresholding the reflectance band.
concentration on both water and
In order to extract and map the water line, the Normalized
sediment with the reflectance of Sentinel-2’s bands with
Difference Water Index (NDWI) was applied on the image,
a central wavelength between 665nm and 1610nm.
then water bodies can be mapped. The correlation and
regression analysis technique have been used in order to
II. Materials and Methods investigate and study the relationship between water quality
parameters and the bands of Sentinel-2 by light absorption
2.1. Study Area and reflectance of each parameters. The regression analysis
develops the mathematical equation which express each
Chhuk Tru, the selected area for this study, is one relation. The R2 equation can be expressed as the following:
of the biggest villages on the Tonle Sap Lake that has ∑ (𝑦𝑦𝑖𝑖 −ŷ𝑖𝑖 )2
abundant in natural resource, so the local have more 𝑅𝑅2 = 1 − ∑ 𝑖𝑖 2 (Eq 1.)
𝑖𝑖(𝑦𝑦𝑖𝑖 −ȳ 𝑖𝑖 )
opportunities to develop a stable life style [5]. In this study, Where yi is actual observation, ȳi is mean of actual
water samples were collected during dry season (March 7, observation and ŷi is predicted observation.
2018) at 19 points in Chhuk areas as shown in Figure 1. The height of the reflectance peak around 820nm
has been used for capturing the temperature of the water
from 30℃ to 35℃ [8]. Therefore, Band 8A was chosen to
capture the image for estimating the temperature of the water
in Chhnuk Tru areas. In the other hand, Band 11 was used

111
for capture the image for EC, TDS, and Salinity at 1610nm. 3.2. Conductivity, Total dissolved solid and
The image of bacteria concentration in water was capture by Salinity
band 11 which is the same as other parameters. But, for
bacteria concentration in sediment with the high spatial Electrical conductivity is the measure of water
resolution 10m and wavelength 665nm, band 4 was used to capacity to conduct electrical current. It illustrated
mature the image for monitoring bacteria concentration in significant correlation with TDS and Salinity. Band 11 was
sediment. By reflectance measurement and in-situ used for analyzing the data of EC, TDS and salinity. The
measurement, it gives the suitable values from water quality spatial distribution of water that is used to estimate those
parameter and its reflectance. parameters are generally increase from solid surface to water
surface. As a result, EC value was around 100µS/cm, TDS
III. Results and Discussion was around 50mg/L and Salinity was 0.025psu with R2=
0.80, R2=0.79 and R2=0.79 (Table 1), respectively. Band 11,
3.1. Temperature with the wavelength 1610nm (SWIR), give the best potential
for salinity detection and discrimination [10] the same for
Daily temperature at TSL about 20℃ and 36℃ in electrical conductivity and total dissolve solid. Since the
dry season that the weather is influence by monsoon. In this bands found in SWIR were associate with information
study, the temperature of water was around 30 to 31℃ related to salt content differences.
(Figure 2) with the coefficient of determination R2= 0.83
(Figure 3), while the highest temperature from those 19 3.3. Bacteria concentration
points was around 34℃.
For bacteria concentration, two different bands
were used to determine the bacteria concentration. Band 11
was used to analyze bacteria concentration in water and
Band 4 was used for estimating sediment. The mean value
of bacteria concentration in water and sediment was around
70.37 CFU and 1075 CFU with R2= 0.87 and R2= 0.55,
respectively as shown in Table 1. bacteria concentration in
sediment, Band 4 (Visible) is needed for monitoring it
because the sediment in suspension present in the bodies of
water are visibly seen by their coloring, which are varies
Cloud
depending on the concentration of sediments from dark
brown to yellowish green. The correlation of between
Figure 2. Map of Temperature sedimentary bacteria and B-4 reflectance is very low due to
the cause of turbidity in the low water depth. Since turbidity
35 is a measured of the cloudiness in water.
y = 25.398x + 28.679
Temperature [oC]

34 R² = 0.8312
Table 1. Parameter of water quality, types of bands value
33 range and R2

32 Parameter Types of Value range R2

Bands
31 Temperature B8A 31-34℃ 0.83
30 EC B11 47-735 µS/cm 0.80
0.05 0.1 0.15 0.2 0.25 TDS B11 27-408 mg/L 0.79
Salinity B11 0.02-0.3 psu 0.79
Band-8A [reflectance] Bac-wat B11 5.19-467-86 0.87
Bac-sed B4 0-0.0041 0.55
Figure 3. Correlation between Temperature and B8A
reflectance The results of this study were suitable compare to
other study, in term of correlation from the reflecting
properties of each parameters. except the bacteria
concentration in sediment. due to the reflecting value is
bigger than other parameters. It is also give the possible to
measuring another water quality parameter such as turbidity
an Chlorophyll a via Sentinel-2 since it is also related to the
water quality parameters that we already study.

112
 References

[1] Salama, M., Radwan, M. and van der Velde, R.,

2012. A hydro-optical model for deriving water
quality variables from satellite images (HydroSat):
A case study of the Nile River demonstrating the
future Sentinel-2 capabilities. Physics and
Chemistry of the Earth, 50(52), pp.224-232.
[2] Ritchie, J., Zimba, P. and Everitt, J., 2003.
Remote Sensing Techniques to Assess Water
Quality. Photogrammetric Engineering & Remote
Sensing, 69(6), pp.695-704.
[3] Saberioon, M., Brom, J., Nedbal, V., Soucĕk, P.
and Císar, P., 2020. Chlorophyll-a and total
suspended solids retrieval and mapping using
Figure 4. The coefficient of correlation (R) between water Sentinel-2A and machine learning for inland
quality parameters waters. Ecological Indicators, 113, pp.1-11.
[4] Toming, K., Kutser, T., Laas, A., Sepp, M.,
The corresponding goodness of fit statistical Paavel, B. and Nõges, T., 2016. First Experiences
parameters of the water quality is shown in Figure 4 The in Mapping Lake Water Quality Parameters with
correspondence between Temperature and other parameters Sentinel-2 MSI Imagery. Remote Sensing, 8(8),
is not good with R between 0.3 to 0.47 since the difference p.640.
band of Sentinel-2 was used. But it gives the best fit between [5]. Weber, E., 2020. MEKONG RIVER
EC, TDS, Salinity, and Bac-wat because these parameters EXPERIENCE. [online] Travel-inspirations.lu.
using the same Sentinel-2 bands. While the coefficient of Available at: https://travel-
correlation (R) between the Bac-sed and other parameters is inspirations.lu/fr/inspiration/60. Accessed 26
too low which is close to zero. October 2020.
[6] Ung, P., Un, S., Chheun, S., Aun, S., Penh, S.,
IV. Conclusion Sann, S., Tan, R., Miyanaga, K. and Tanji, Y.,
2018. Analysis of Total Bacterial Concentration
Satellite sensing technology has been developed and Microbial Community in Waters Used by
over the past few decades that make it possible for experts Floating Villagers, Tonle Sap Lake.
in monitoring the large area on the earth. Those results have In: ENHANCING SUSTAINABILITY AND
shown that the mean value from the raster data gave a better RESILIENCE UNDER ANTHROPOGENIC
correlation with the reflectance from the bands, but there are PRESSURE AND CLIMATE CHANGE. Phnom
some errors from reflectance due to the cloudiness around Penh: Porsry Ung, pp.320-324.
that area during the study. Furthermore, besides the [7] Kaing, V., Sok, T., Ich, I., Oeurng, C., Song, L.,
promising technical feature of sentinel-2, it is worth Siev, S., Uk, S., Mong, M., Hak, D., Khanal, R. and
monitoring the public availability of its imagery. It has a Chihiro, Y., 2018. In: ENHANCING
wide range of useful application in monitoring water quality, SUSTAINABILITY AND RESILIENCE UNDER
yet still there is room for improvement. ANTHROPOGENIC PRESSURE AND CLIMATE
CHANGE. Phnom Penh: Vinhteang Kaing, pp.303-
Acknowledgement 306.
[8]. Rizwan Salwwm, M., Honkanen, S. and Turunen,
We are thankful to the Science and Technology J., 2012. Partially athermalized waveguide
Research Partnership for Sustainable Development gratings. Proceedings of SPIE, 8428(17), pp.1-9.
(SATREPS), the Japan Science and Technology Agency [9]. Coluzzi, R., Imbrenda, V., Lanfredi, M. and
(JST)/Japan International Cooperation Agency (JICA) for Simoniello, T., 2018. A first assessment of the
their financial support. (Grant No: JPMJSA1503) Sentinel-2 Level 1-C cloud mask product to
support informed surface analyses. Remote
Sensing of Environment, 217, pp.426-443.
[10]. Bannari, A., El-Battay, A., Bannari, R. and
Rhinane, H., 2018. Sentinel-MSI VNIR and SWIR
Bands Sensitivity Analysis for Soil Salinity
Discrimination in an Arid Landscape. Remote
Sensing, 10(6), p.855

113
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Optimization of Electrocoagulation Reactor Integrated Sedimentation for Turbidity and

Color Removal from Industrial Wastewater

Sreynich Pang 1, Sreyla Vet 1, Penghour Hong 1, Pisut Painmanakul 2 and Saret Bun 1,*

1
Water and Environmental Engineering, Faculty of Hydrology and Water Resources Engineering,
Institute of Technology of Cambodia, Phnom Penh 12156, Cambodia
2
Department of Environmental Engineering, Faculty of Engineering,
Chulalongkorn University, Bangkok 10330, Thailand
* Corresponding author: saret@itc.edu.kh

Abstract

This study aimed to optimize electrode configuration and operation condition of electrocoagulation reactor for removing
color and turbidity from synthetic textile wastewater. The experiment was firstly conducted in the small batch column for
investigating the performance of different electrode types, arrangements, gaps, and current density. The result showed that the
arrangement of electrodes as monopolar and bipolar with 1.5 cm inner gap and 1.5 mA/cm2 current density provided the
optimal performance in terms of gas flow and electrode loss. Monopolar was found as optimum level in terms of energy
consumption while comparable treatment efficiency was observed. Hence, the experiments for color and turbidity removal
kinetic under optimal condition were conducted for defining the sufficient hydraulic retention time and overflow rate of
continues electrocoagulation reactor integrated sedimentation. Bentonite and reactive dye were used for preparing turbidity
and color wastewater, respectively. To archive 95% removal efficiency of both pollutants, the detention time about 25 to 30
minutes using 4.5 mA/cm2 current density are required. Plus, overflow rate 2.1 m/hr was found as the effective condition for
gravity settling to design sedimentation compartment. Therefore, integrated system in continues operation was constructed as
semi-industrial scale, 141 litters, for performance evaluation study. Approximate 95% and 97% removal efficiency of both
pollutants was obtained under the optimal condition found (13.5 mA/cm 2 current density) with liquid flow rate 3 and 1 L/min,
respectively. However, 2 L/min was found as the optimum liquid flow for simultaneous removal of both pollutants.

Keywords: Color, Electrocoagulation, Industrial wastewater, Turbidity

I. Introduction of the ecosystem. Therefore, removal of turbidity and color

from wastewater are occasionally required.
As the increasing of industrial sector, amount of
Various technologies have been employed for dye
wastewater enter to the water body has risen. Most of the
effluent treatment including conventional biological
industrial wastewater effluence contains high amount of
processes (aerobic and anaerobic), coagulation, and
color and turbidity. Discharge untreated wastewaters from
adsorption [1]. Since the effluence contains toxic dyes in
industries are the main source of adding substantial
nature as mentioned, it may also impact on the bacteria
coloration to water. A highly colored water could not sustain
development in biological process. Furthermore, chemical
aquatic life, which could lead to the long-term impairment

114
coagulation requires an addition of chemicals, which leads turbidity, bentonite was rapidly mixed with water at 300 rpm
to generate huge quantity of sludge. for 5 minutes, followed by slow mixing at 40 rpm for 30
Electrocoagulation (EC) process has been interested for minutes, and let it settle down 35 minutes. Moreover,
studying with different types of effluents from the industries. synthetic color wastewater was prepared by dissolving the
EC has been employed successfully for the reduction dyes reactive dye in tap water for 5 minutes with 300 rpm.
using a direct current source between metal or aluminum
electrodes immersed in polluted water. It has the capacity to 2.3. Electrode optimization
treat wastewater more efficient comparing to conventional Configuration and operation condition of electrodes
coagulant, about 10 – 15% [2]. This study aims to develop were initially investigated to define the optimum condition
and evaluate the electrocoagulation reactor combining both in the batch column experiment in terms of generated gas
EC and sedimentation units in terms of design criteria and flow and electrode corrosive loss ratio (Qg/Eloss). Generated
operation condition in batch and continue modes for color gas flow (Qg) was directly measured using soap film meter
and turbidity removal from wastewater of textile industry. captured from the reactor cap covering (see Fig. 1). It can be
estimated by using Eq. (1), where Qg is gas flow rate [mL/s],
II. Materials and Methods ΔV = V2 - V1 is the different volume of gas (mL), and Δt =
t2 - t1 is the different time of gas moved (sec.). Electrode loss
2.1. Experimental set-up
(Eloss) presented in percentage was defined the different
Batch column was constructed from clear acrylic
weights of electrode used before and after the experimental
material in cylinder shape with 3 cm-diameter for containing
work, as expressed in Eq. (2). Electrode configuration and
4 litters of water sample, as illustrated in Fig. 1. At the
operation condition parameters included electrode gap (1 –
bottom, it was connected to the drainage port for discharging
2 cm), electrode arrangement (monopolar and bipolar), and
the sludge and sample after conducting the experiments. The
current density (1.5 – 2.5 mA/cm2). The initial turbidity and
electrode plates (20 cm × 5 cm × 0.2 cm) were installed in
color concentration in this part were 250 NTU and 6,000
the reactor and connected to the direct current (DC)
ADMI, respectively.
generator for supplying the current. Aluminum electrode
plates were used in this study. At the top, an acrylic cap was ΔV
covered for collecting the generated gas from the system to Qg = (Eq.1)
Δt
the soap film meter. Tap water was firstly used as liquid
phase at room temperature (25 ± 3°C) for optimization study initial weight - final weight
of EC configuration and operation condition. Eloss = (Eq.2)
initial weight

2.4. Settling column test

Setting test was examined in batch column to determine
the over flow rate (OFR) as the design criteria of settling
performance for design the sedimentation tank as the
separation process after the EC treatment. This
sedimentation experiment was conducted within 100
(1) Column test
(2) Gas collection pipe minutes with 4 ports, i.e., 13, 18, 23, and 28 cm, from the
(3) Soap film meter water surface for water sampling. The initial concentration
(4) Electrode plates
(5) Electrical wire of turbidity and color prepared for this part were 250 NTU
(6) DC generator
(7) Sampling port and 6,000 ADMI, respectively.
(8) Waste drainage port

Fig. 1. Experimental set-up in batch column reactor III. Results and Discussion
3.1. Optimize EC configuration and operation condition
2.2. Synthetic wastewater This part aimed to optimize the electrocoagulation
To synthesize turbidity and color wastewater, bentonite condition including electrode configuration, arrangement,
and reactive dye were prepared with tap water. For synthetic

115
and current density. Electrode configuration, i.e., gap, arrangement were comparable, other evaluation factor,
arrangement, and current density, was initially investigated treatment cost, was used to select the optimum one. Based
in terms of Qg/Eloss. It was investigated between monopolar on the energy consumption estimation, bipolar required
and bipolar, inner electrode gap (1.0 – 2.5 cm), and current approximate $US 0.31 to remove one kilogram of dye from
density (1.5 – 2.5 mA/cm2). It was found that the optimum 1 m3 of wastewater, while $US 0.05 was paid by using
electrode in monopolar and bipolar arrangement were 1.5 monopolar electrode. Hence, aluminum electrode in
mA/cm2 of current density with 1.5 cm and 2 cm of electrode monopolar arrangement with 1.5 cm gap was selected as the
gap, respectively (data not shown). These conditions were economize condition in terms of turbidity and color removal.
therefore used to conduct under synthetic wastewater of 250 Therefore, it will be designed for investigating under
NTU turbidity and 6,000 ADMI color. different current density.
Current density plays an important role in EC process. In
a 100 this study, it was then varied in 5 levels, i.e., 1.5, 2.5, 3.5,
4.5, and 5.5 mA/cm2 for turbidity and color removal from
Removal efficiency [%]

80 wastewater. The result showed that the current density of 4.5

mA/cm2 was the optimal condition and the operation time of
60
steady stage was 30 minutes for both turbidity and color
removal (data not shown).
40

20 Bipolar a 100

Removal efficiency [%]

Monopolar 80
0
0 30 60 90
Time [min] 60
100
b 40
80
Removal efficiency [%]

Cₒ= 50 NTU
20
Cₒ= 250 NTU
60 Cₒ= 500 NTU
0
0 10 20 30
40
Time [min]

20 Bipolar b 100
Monopolar
Removal efficiency [%]

80
0
0 30 60 90
Time [min] 60

Fig. 2. Removal efficiency under optimum condition and

40
electrode arrangements for: (a) turbidity, and (b) color
Cₒ =2,500 ADMI
20
The result was shown in Fig. 2. Bipolar connection Cₒ =6,000 ADMI
Cₒ =15,000 ADMI
consumed high voltage than monopolar one to increase 0
temperature for promoting electrode plate to generate more 0 10 20 30
coagulants and gas bubble at the early stage. However, a Time [min]
similar performance was observed at the stable stage. At 35 Fig. 3. Removal kinetic at different initial concentration of:
minutes treatment time, monopolar could remove turbidity (a) turbidity, and (b) color
and color up to 95 % and 89%, respectively; while bipolar
could treat up to 96% and 93% for respective turbidity and 3.2. Turbidity and color removal kinetic
dye. Since the treatment performance of both electrode Effect of different initial concentrations on the treatment

116
performance was kinetically studied in this part. The initial that after decreasing the liquid flow to 2 L/min, the removal
turbidity and color concentrations were examined in range efficiency increased due to the higher hydraulic retention
of 50 to 500 NTU, and 2,500 to 15,000 ADMI, respectively. time allowing the coagulant and particles attached to each
The result was shown in Fig. 3. Initial concentration of 50 other to form floc for settle down.
NTU needed electrolysis time about 25 minutes to dissolve
aluminum ion to improve the lag stage until reaching the a 300 100
stable stage, which provided removal efficiency around 75%.

Residual turbidity [NTU]

For 250 NTU and 500 NTU, it was required electrolysis time
75
only 20 minutes with removal efficiency 96% and 98%,

% Removal
200
respectively (see Fig. 3 (a)). Low concentration of color
consumed less electrolysis time and provided higher 50
treatment efficiency than the high concentration at lag stage.
100
At stable stage, removal efficiency of three initial
25
concentrations were very similar, 98% ± 1% after 20 StD: 20 NTU
minutes (see Fig. 3 (b)).
0 0
3.3. Settling performance 0 50 100 150
Settling test was studied in a batch column containing 4- b 6000 100
litter sample with 33-cm height. Four sampling ports, 13, 18,
23, and 28 cm from water surface, were designed for
Residual color [ADMI]
75
samples collection. It was investigated after operating EC of

% Removal
4000
optimum condition previously found with 250 NTU and
6,000 ADMI as initial concentrations. The results showed 50
that 2.1 and 1.7 m/hr of overflow rate (OFR) can separate
2000
turbidity (~ 87%) and color (~ 90%), respectively (data not
25
shown). This finding will be used as the design criteria of
StD: 300 ADMI
the settling separation process.
0 0
3.4. Performance of ECR integrated sedimentation 0 50 100 150
Time [min]
Based on the result separately found, each design criteria
were used for constructing the new ECR, which was Fig. 4. Treatment performance in continuous EC integrated
included EC (4 units of EC) and separation compartments. sedimentation (2 L/min) for: (a) turbidity, and (b) color
The experiment was conducted for individual pollutants
with 13.5 mA/cm2. Under initial concentration 250 NTU and IV. Conclusion
6,000 ADMI, turbidity and color can be removed to less than The objective of the present study is to optimize the
20 NTU and 300 ADMI after using liquid flow 3 and 1 electrode configuration and operation condition of
L/min, respectively (data not shown). electrocoagulation reactor integrated sedimentation for
Then, a simultaneous removal of both pollutants was removing color and turbidity from synthetic textile
conducted to define the optimum liquid flow. Under 3 L/min wastewater. In order to confirm the effect of the new reactor
flow, color cannot be completely removed to the target level, design, several parameters in batch and continue reactor was
300 ADMI (data not shown). The simultaneous removal evaluated. It was firstly examined in a small batch column
under the flow 2 L/min was conducted and the result was for different electrode types, arrangements, gaps, and current
shown in Fig. 4. Removal efficiency was reached to 90% for density. Monopolar with 1.5 cm inner gap and 1.5 mA/cm 2
both turbidity and color after 60 minutes operation as the current density was found as the optimum condition in terms
stable stage. At 150 minutes, color removal reaches 97%, of Qg/Eloss ratio and energy consumption, while monopolar
(residual color 213 ADMI), and 94% removal efficiency for and bipolar resulted comparable treatment performance.
turbidity (residual turbidity 14.2 NTU). It can be explained Therefore, color and turbidity removal kinetic under optimal

117
condition were conducted for defining the sufficient
hydraulic retention time and overflow rate of continues ECR
integrated sedimentation. Approximate 95% removal
efficiency was obtained under the detention time about 25 to
30 minutes using 4.5 mA/cm2 current density. Moreover, 2.1
m/hr of overflow rate was found as the effective condition
for gravity settling to design sedimentation compartment.
Integrated system in continues operation was constructed to
evaluate treatment performance. A simultaneous removal of
turbidity and color can be completely removed to be lower
than effluent wastewater standard under the optimum
condition found with 13.5 mA/cm2 current density with 2
L/min liquid flow.

Acknowledgement
Authors acknowledges Project for Strengthening
Engineering Education and Research for Industrial
Development in Cambodia of JICA through LBE Research
Grant for financial support.

References
[1] Verma, A., Bhunia, P., and Dash, R., 2014.
Reclamation of wastewater using composite
coagulants: A sustainable solution to the textile
industries. Chem. Eng. Trans. 42, 175-180.
[2] Zaleschi, L., Teodosiu, C., Cretescu, I., Rodrigo, M.
A., 2012. A comparative study of electrocoagulation
and chemical coagulation processes applied for
wastewater treatment. Environmental Engineering &
Management Journal. 11(8).

118
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
The 5th International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Biodegradation of Perfluorooctanesulfonate (PFOS) and Perfluorooctanoic Acid (PFOA)

by Acclimated Sludge

Sovannlaksmy Sorn 1, Hiroe Hara-Yamamura 2 and Ryo Honda 2,*

1
School of Natural Science and Technology, Kanazawa University, Kakuma-machi, Kanazawa City,
Ishikawa Prefecture 920-1192, Japan
2
Faculty of Geosciences and Civil Engineering, Kanazawa University, Kakuma-machi, Kanazawa City,
Ishikawa Prefecture 920-1192, Japan
* Corresponding author: rhonda@se.kanazawa-u.ac.jp

Abstract

Perfluorooctanesulfonate (PFOS) and Perfluorooctanoic acid (PFOA) have been strictly regulated due to their high
bioaccumulation and potential toxicity to human health. PFOS and PFOA are commonly found in the environment; however,
knowledge of biodegradation of PFOS and PFOA is still limited. The aim of this research is to investigate PFOS and PFOA
biodegradability using environmental microbes from acclimated sludges. The return sludge from a local wastewater treatment
plant was cultured in synthetic wastewater with a stepwise increase of PFOS or PFOA dose at 0, 5, 10, 20 µg/L in every 5
days under aerobic condition. After 20-days, acclimated sludge with each concentration of PFOS or PFOA were utilized as
inoculum in Kirk liquid media for batch biodegradation experiment. The mixture of 20 µg/L PFOS and 20 µg/L PFOA was
spiked in glass vials with a rubber stopper and seal crimp in triplication. These vials were then incubated at 30°C under orbital
shaking at 170-180 rpm and their headspaces were replaced with ambient air every 2-4 days to supply oxygen. Liquid samples
were taken after 6 hours, and days 1, 2, 7. PFOS and PFOA were measured by LC-MS/MS after solid phase extraction. With
PFOS-acclimated sludge, PFOS was decomposed approximately 40% after 7 days of incubation. On the contrary, significant
biodegradation was not observed for PFOA with PFOA- nor PFOS-acclimated sludges. Thus, this study could contribute to
the biodegradation of PFOS using microbial communities derived from wastewater sludge.

Keywords: Bioremediation, LC-MS/MS, Perfluoroalkyl and Polyfluoroalkyl substances (PFASs), Perfluorooctanoic acid
(PFOA), Perfluorooctanesulfonate (PFOS)

I. Introduction Possessed the stable chemistry, PFASs are very persistent

and hard to biodegrade in environment. PFASs have
Perfluoroalkyl and Polyfluoroalkyl substances (PFASs)
potential toxic to human health after exposure to
are a large group of synthetic substances with more than
contaminated water, agricultural and aquacultural products.
5000 subclasses. PFASs have a great ability to repel fire,
They accumulated long term in human bodies and frequently
water, oil, and stains. Therefore, they are widely used in
detected in blood serum of human, wildlife nearby
variety of products such as the stain and water resistance
contaminated sites. Therefore, PFASs were regulated in
fabrics, nonstick pans, protecting sprays, firefighting foams,
several countries. The US Environmental Protection Agency
food packaging materials, cleaning products and so on.

119
(EPA) has set the safety consumption dose in drinking water samples from Aqueous film forming foam utilized sites were
up to 70 ppt and banned PFASs production since 2000s [1]. employed as seeding for fungi isolation by American
However, the legacy of PFASs still existed in environment research group to degrade 6:2 FTOH [11]. PFOS degradable
these days. Pseudomonas aeruginosa strain was enriched from activated
PFASs have been commonly detected as high sludge of PFCs contaminated municipal wastewater
concentration in wastewater treatment plants, industrial treatment plants in Korea [12]. These studies suggested the
producing or using PFASs, and firefighter training sites [2]. potential of enrichment and isolation PFASs degradable
PFASs are hardly to remove using existed treatment microbes from historically PFASs contaminated sites and
technologies, therefore; residue of PFASs have still found in effectiveness of acclimation processes. It also reflected the
downstream even at low concentrations. The US reported development on bioremediation technologies that are
the median of total 17 PFASs in source of water is 21.4 ng/L, expected to increase the efficiency of treatment PFAS in real
and 19.5 ng/L in treated drinking water [3]. PFASs were also polluted environment. The aim of this research is to
reported in drinking water and bottle waters at range of 116- investigate PFOS and PFOA biodegradability using
140ng/L according to an assessment of PFASs in Brazil, environmental microbes from acclimated sludges.
France, and Spain [4]. Regarding the numbers of survey by
French Overseas Territories, PFOS and PFOA were founded II. Materials and Methods
as the most frequently existed compounds in surface water
2.1. Sludge acclimatation
[5].
The return sludge was taken from a local wastewater
Previous studies have shown their great efforts on PFASs
treatment plant and used as inoculum. It was cultured in
remediation using various technologies in both physical and
synthetic wastewater by ratio 1:2 which prepared according
chemical approaches such as absorption, photocatalytic,
to OECD recipe [13]. New synthetic wastewater has been
electrochemical oxidation, and reverse osmosis [6]. These
replacing every 2-3 days with stepwise increase of PFOS or
treatment technologies have limitations for their application
PFOA dose at 0, 5, 10, 20 µg/L in every 5 days under aerobic
on high operational cost and may also produce
condition.
perfluorocarbons wastes (PFCs). The PFCs waste are
potentially more harmful to health and even more
2.2. Batch biodegradation experiment
persistence in future.
After 20-days, sludge acclimatized with each of PFOS
Microbial approach for remediation is a key to eliminate
and PFOA, and non-acclimated sludge were utilized as
the harmful pollutants from environment. It has gained the
inoculum in Kirk liquid media as described by Ramírez and
interest of researchers to combat PFASs in the last decades.
co-workers [14] for batch biodegradation experiment.
PFASs degradable microbes enriched from environmental
Deactivated sludge by autoclaving non-acclimated sludge
samples have been applied as one of affordable
has also been examined to reveal the sludge absorption of
bioremediation treatment technologies. Previously, loam
PFOS and PFOA. A mixture of 20 µg/L PFOS and 20 µg/L
soil from an agronomy farm was used as microcosms on 8:2
PFOA was spiked in 50 mL glass vials with a rubber stopper
Fluorotelomer Stearate Monoester biodegradation [7],
and seal crimp in triplication. These vials were then
aerobic and anaerobic river sediment also performed for 6:2
incubated at 30°C under orbital shaking at 170-180 rpm and
fluorotelomer sulfonate treatment [8], and marine sediment
their headspaces were replaced with ambient air every 2-4
from False Creek used for N-Ethyl Perfluorooctane
days to supply oxygen.
Sulfonamido Ethanol (EtFOSE) and EtFOSE-Based
Mix liquor in each glass vial was transferred into a sterile
Phosphate biodegradation [9]. Lab scale acclimation has
50 mL polypropylene (PP) centrifuge tubes after 6 hours and
also been performed to enrich biodegrading bacteria strain
days 1, 2, 7. The samples were centrifuged at 13,000 ×g for
from soil samples, and Pseudomonas parafulva was reported
5 min to separate supernatant and pallet. Then, the used glass
as PFOA-degrading bacteria strain [10]. Some studies draw
vials, and sludge have been rinsed using Phosphate-buffered
their attention on pure culture isolation which were isolated
saline to collect all attachment PFOS and PFOA. After
from PFASs contaminated environment that commonly
rinsing, all liquid samples were collected at final volume 40
known as potential inoculum sources for stronger
degradability microbial cultivation. Groundwater and soil mL and stored at 4°C till extraction.

120
2.4. Analytical determination of PFOS and PFOA
Each liquid sample was spiked with 50 µL of isotopically 23 Activated sludge
labeled 1 mg/L 13C4-PFOS and 1 mg/L 13C4-PFOA standards. Deactivated sludge

Concentration (µg/L)
21 PFOS acclimated sludge
PFOS and PFOA were enriched on Oasis WAX (150mg, 6cc)
PFOA acclimated sludge
cartridge (Waters, U.S.A.) according to manufacturer’s 19
instruction [15]. Briefly, sample was the cartridges were
preconditioned by 4 mL of 0.1% ammonia/methanol, 4 mL 17
of methanol and 4 mL of Milli Q water. Samples were loaded
15
at the frow rate of 5 ml/min or slower, and washed with 4
mL of ammonium acetate buffer. Subsequently, the 13
cartridges were dried under vacuum for 2 min before elution 0 1 2 3 4 5 6 7
into 15 mL PP conical tubes using 4 mL of methanol and 4 Degradation time (days)
mL of 0.1% ammonia/methanol. The eluent was evaporated Fig. 1. Biodegradation of PFOS using non acclimated sludge,
to dryness under gentle nitrogen stream at temperature 40°C. deactivated sludge, PFOS-acclimated sludge and PFOA
The extracts were reconstituted in 0.5 mL of 50% methanol 3.2. PFOA biodegradation acclimated sludge.
and Milli Q water before being transferred to PP vial for LC-
MS/MS analysis. Results of the experiments set up under four different
PFOS and PFOA were analyzed on a Nexera X2 HPLC types of sludge conditions are presented in Fig.2. None of
(Shimadzu, Japan) using a 2.1×100 mm Inertsil ODS-4 the tested conditions showed a significant decrease in the
column (3µm) (GL Science, Japan) coupled to a Sciex 3200 concentration of PFOA during experiment 7 days. Thus,
triple quadruple spectrometer in multiple reaction microbial communities in PFOA-AS are likely unable to
monitoring and negative electrospray ionization mode. 5 µL degrade PFOA nor PFOS. These results supported the highly
of extracts were injected via an autosampler and eluted with tolerant nature of PFOA in microbial communities derived
50% of 10 mM ammonium acetate and 50 % of acetonitrile from wastewater sludge. At the same time, it indicated that
as mobile phase. The flow rare was set at 0.2 mL/min. microbial profile in PFOS-AS might differ to PFOA-AS.

III. Results and Discussion 23

Concentration (µg/L)

3.1. PFOS biodegradation 21

The trend of the PFOS biodegradation experiments
performed with non-acclimated sludge (AS), deactivated 19
sludge (AS), PFOS acclimated sludge (PFOS-AS), and Activated sludge
17
PFOA acclimated sludge (PFOA-AS) is shown in Fig.1. Deactivated sludge
PFOS concentration is gradually decreased after incubation 15 PFOS acclimated sludge
2 days about 20%, and successively up to 40% after 7 days PFOA acclimated sludge
in PFOS-AS and followed by AS, however, no significant 13
0 1 2 3 4 5 6 7
degradation observed under PFOA-AS condition. This
result suggested that PFOS is degradable using microbial Degradation time (days)
communities in PFOS-AS or AS even at low concentration, Fig. 2. Biodegradation of PFOA using non acclimated sludge,
but not PFOA-AS. Moreover, there is no significant deactivated sludge, PFOS-acclimated sludge and PFOA
decrease observed under DS tested. The complete acclimated sludge.
sterilization of the control could be responsible for the
conclusive biodegradation in PFOS-acclimated test. Thus, it IV. Conclusion
indicated that PFOS biodegradation using acclimated sludge
can be effective at low dose and short degradation time, This study showed the biodegradability of PFOS by
compared to previous studies [7,8,9]. PFOS-acclimated sludge. It was found that the
decomposition was approximately 40% after 7 days of
incubation. On the contrary, significant biodegradation was

121
not observed for PFOA with PFOA-AS nor PFOS-AS. Thus, [8] Zhang, S., Lu, X., Wang, N., & Buck, R. C., 2016.
this study could contribute to the biodegradation of PFOS Biotransformation potential of 6:2 fluorotelomer
using microbial communities derived from wastewater sulfonate (6:2 FTSA) in aerobic and anaerobic
sludge. Further study should be focused on microbial sediment. Chemosphere, 154, 224–230.
community with high degradation capacity acclimated [9] Benskin, J. P., Ikonomou, M. G., Gobas, F. A. P. C.,
sludge by 16S rRNA sequencing. Begley, T. H., Woudneh, M. B., & Cosgrove, J. R.,
2013. Biodegradation of N-ethyl perfluorooctane
Acknowledgement sulfonamido ethanol (EtFOSE) and EtFOSE-based
phosphate diester (SAmPAP diester) in marine
We are thankful to Prof. Shaily Mahendra of Department
sediments. Environmental Science and Technology,
of Civil and Environmental Engineering, University of
47(3), 1381–1389.
California in Los Angeles for her experiences on this field
[10] Yi, L. B., Chai, L. Y., Xie, Y., Peng, Q. J., & Peng, Q.
of research on bioremediation and documentary supports. Z., 2016. Isolation, identification, and degradation
performance of a PFOA-degrading strain. Genetics
References and Molecular Research, 15(2), 1–12.
[1] EPA, 2017. Technical Fact Sheet- Perfluorooctane [11] Merino, N., Wang, M., Ambrocio, R., Mak, K.,
Sulfonate (PFOS) and Perfluorooctanoic acid O’Connor, E., Gao, A., Hawley, E. L., Deeb, R. A.,
(PFOA). Tseng, L. Y., & Mahendra, S., 2018. Fungal
[2] Dauchy, X., 2019. Per- and polyfluoroalkyl biotransformation of 6:2 fluorotelomer alcohol.
substances (PFASs) in drinking water: Current state Remediation, 28(2), 59–70.
of the science. Current Opinion in Environmental [12] Kwon, B. G., Lim, H. J., Na, S. H., Choi, B. I., Shin,
Science & Health, 7, 8–12. D. S., & Chung, S. Y., 2014. Biodegradation of
[3] Boone, J. S., Vigo, C., Boone, T., Byrne, C., Ferrario, perfluorooctanesulfonate (PFOS) as an emerging
J., Benson, R., Donohue, J., Simmons, J. E., Kolpin, contaminant. Chemosphere, 109, 221–225.
D. W., Furlong, E. T., & Glassmeyer, S. T., 2019. Per- [13] OECD, 2010. OECD Guidelines for the Testing of
and polyfluoroalkyl substances in source and treated Chemicals (Technical Report 209). Activated Sludge,
drinking waters of the United States. Science of the Respiration Inhibition Test (Carbon and Ammonium
Total Environment, 653, 359–369. Oxidation. Guideline 16.
[4] Schwanz, T. G., Llorca, M., Farré, M., & Barceló, D., [14] Ramírez, D. A., Muñoz, S.V., Atehortua, L., Michel,
2016. Perfluoroalkyl substances assessment in F. C., Jr., 2010. Effects of different wavelengths of
drinking waters from Brazil, France and Spain. light on lignin peroxidase production by the white-
Science of the Total Environment, 539, 143–152. rot fungi Phanerochaete chrysosporium grown in
[5] Munoz, G., Labadie, P., Botta, F., Lestremau, F., submerged cultures. Bioresource Technology, 101
Lopez, B., Geneste, E., Pardon, P., Dévier, M. H., & (23), 9213-9220.
Budzinski, H., 2017. Occurrence survey and spatial [15] Waters. (2009). PFOS/PFOA extraction method
distribution of perfluoroalkyl and polyfluoroalkyl using Oasis WAX.
surfactants in groundwater, surface water, and https://www.waters.com/waters/promotionDetail.ht
sediments from tropical environments. Science of the m?id=10108600&locale=145. Accessed 26 Oct 2020.
Total Environment, 607–608, 243–252.
[6] Kucharzyk, K.H., Darlington, R., Benotti, M., Deeb,
R., Hawley, E., 2017. Novel treatment technologies
for PFAS compounds: A critical review.
Environmental Management, 204 (2),757-764.
[7] Dasu, K., Liu, J., & Lee, L. S., 2012. Aerobic soil
biodegradation of 8:2 fluorotelomer stearate
monoester. Environmental Science and Technology,
46(7), 3831–3836.

122
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

.
Effect of Crop Growth through Improving Drip Irrigation Uniformity

Channtola SOT1*, Vouchleang SREANG1, Mengheak PHOL1, Sophanith THA1, Makara SOY1, Somnang
SRIM1, Seum MAO1, Chanthan HEL2, Chantha OEURNG1, Pinnara KET1

1
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia.
2
Faculty of Telecommunication, Electronic, and Network Engineering, Institute of Technology of Cambodia, Russian Fed
eration Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
.
* Corresponding author: tolasot25@gmail.com

Abstract

Crop production experiments of three crops, such as, Red Oakleaf, Red Lettuce and Bok Choy with different system
uniformities and field scales controlled by driplines were conducted in a greenhouse in 2020. The Christiansen uniformity
coefficients were evaluated. The uniformity of systems was established with nominal emitter discharge rates along a
dripline. For all of the system uniformities tested, dry matter above ground uptake displayed high uniformity coefficients
throughout the entire growing season. The effects of system uniformity on the crop growth was evaluated. The crop
growth experiment of 3 plants, such as, Red Oakleaf (R1), Red Lettuce (R2) and Bok Choy (R3) in the new system
installed presented the excellent crop growth with the uniformity of 70-90%. The yields obtained R1 7000kg/ha, R2
6500kg/ha, R3 14000kg/ha. However, the crop growth uniformity of Bok Choy was affected by insects so that lead to
lower uniformity despites the irrigation uniformity.

Keywords: Drip Irrigation, irrigation uniformity, crop growth uniformity

123
I. Introduction Table 1: Date of each growing step
Planting date Transplanting Harvesting
Water use efficiency (WUE) in irrigation is the factor
date date
that saves water to increase crop yields. WUE can be
25.09.2020 17.10.2020 18.11.2020
enhanced through improving water distribution
uniformity in irrigation system (Phengphaengsy and
Soil properties were measured as indicated in the Table
Okudaira 2008; Li 1998).
2
Actually, in drip irrigation system, the uniformity is
Table 2. Soil properties
considered as a key parameter to design a proper system.
Parameters Value
Improper determination of the parameter, it leads to
Soil Texture Loam
nonuniform and lower yields (Bralts and Kesner 1983),
(Zhao et al. 2012). Some previous scientific studies pH 6.63
prove that a good design of the water pressurize by using
the software to increase the system performance and
water efficiency obtained energy saving was (34.23 ; V(A) V(B) V(C)
29.54%) with slope 2% by using lateral length 30m for
drip irrigation(Attia et al. 2019).

This knowledge of irrigation uniformity is important

for Cambodian farmers. Currently, there is emerge and Figure 1. Three selected crops planted in the
increase of using drip irrigation for vegetable production greenhouse.
in Cambodia. Therefore, there is a need to prove this A) Red lettuce, B) Red Oakleaf C) Bok Choy
knowledge to the farmers.
The purpose of this study was to investigate the II.2. Experimental set-up
effect of drip irrigation uniformity to crop growth
uniformity of lettuce growth in greenhouse. There were two events of installation of the drip system.
The first event (F1) presented the improper uniformity
II. Materials and Methods of the drip system installed in August. The second event
II.1. Sampling sites (F2) presented the proper uniformity of the drip system
Field experiment was conducted in greenhouse after consideration of quality of the drip line, emitters
at Institute of Technology of Cambodia, Phnom Penh, and soil moisture distribution. There were different
Cambodia. The area of greenhouse equaled to 10mx8m. materials used in the both events.
There are total amount of 16 raised bed boxes. Only 10
boxes were used for testing. Dimension of each box II.3. Data collection
was 1.3mx1.9m =2.47m2. In the first event, there are 24 emitters with 4 laterals
Growing the plants started in rainy season with for testing, space of emitter was 317mm and space of
temperature around 30 ℃ at day and 27 ℃ at night. lateral was 325mm. The pressure of pump was 2.5bar
Planting was spent 43 days with three types of crops, (Figure 2).
Red Oak leaf, Red Lettuce and Bok Choy. They were
selected to transplant in the greenhouse on 17 October
2020. The crops were planted in the raised bed boxes as
indicated in Figure 1. Worm compost was selected for
nutrient and not allow other chemical fertilizers for
nutrient, so food safety was achieved from this condition.
In a box, we cropped 12 plants and irrigated the same
amount of water, nutrient, temperature at the same time.
Planting and transplanting dates are presented in
Table 1.

124
 Figure 2. Schema of drip system at the first event
Where 𝒙𝒊 parameter of discharge each emitter
After measured wet area was designed the spacing (ml/min), 𝒙 is the average of all discharge in the emitter,
emitters and laterals in the same shape of box which had n is the total emitter.
48 emitters with 6 laterals in order to get the water (Zhao et al. 2012) suggests a design Christiansen
uniformly cover the soil and the root length will get uniformity coefficient (CU) got greater than 80%.
water enough for growing according to the graph shown However,
(Figure 3). Using the same principle of Christiansen uniformity
to calculate the uniformity of crop growth
∑𝒏𝒊=𝟏|𝒃𝒊 − 𝒃|
𝑪𝑼𝒄𝒓𝒐𝒑 = 𝟏𝟎𝟎 × (𝟏 − )
𝒏𝒃
Where 𝒃𝒊 parameter of biomass of each plant
(g/plant), 𝒃 is the average of weight’s plant, n is the total
emitter.

III. Results and discussion

III.1. Improvement of drip irrigation uniformity

Figure 3. The schema of drip irrigation installed at

the second event.
Emitters flow rates measured in the boxes using plastic
glass as indicated in the Figure 4. The flow rates were
measured for a minute

A(A) (((B)

Figure 5. Emitter flow rates of the 2 installation

events

Figure 4. Emitter flow rate measurement in the box Figure 5. shows the flow rate variation in the
of soil. growing boxes. The first event of drip installation
A) first event of drip system installation, B) Second presented the large variation within the box. This
event of drip system installation.
means that there were nonuniformity of the
Biomass of plants were collected for 3-weeks and those
irrigation water supply in the box that impacted the
plants were dried in oven at 70℃ in 48h. In each plant
type, the three-difference types of biomasses were crop growth. The new event of installation
collected. presented the small graduation of the flow rate in
the box.
II.4. Analytical methods

The Christiansen uniformity coefficient (CU)

(Chinese
Standard, 1995) uses for determine quality
uniformity of discharge and yield of planting.
∑𝒏𝒊=𝟏|𝒙𝒊 − 𝒙|
𝑪𝑼 = 𝟏𝟎𝟎 × (𝟏 − )
𝒏𝒙

125
 Table 3: CU Analysis
Flow rate of old Flow rate of new
emitter (F1) emitter (F2)
(ml/min) (ml/min)

Average 36.73 27.75

CU 54% 91%

For F1 CU reached 54% but CU F2 reached 91% (Table

1). input discharge from pump, pressure and method of
measurement was similar to F1 CU level 54% it’s was
nonuniformity so planting wouldn’t get water enough to Figure 6. Variation of crop yields planting in the
profit themselves then it’s made low the yield, but level greenhouse after improving the irrigation uniformity
CU of F2 91% its uniformity so we can control the flow
rate to planting and it’s an important to saving the water III.3. Crop growth uniformity
and it’s a part to get the high yield.
The Figure 7 illustrated the crop growth uniformity
Drip F1 and F2 was design the similar physical so from transplanting until harvesting.
Figure 3 can describe F1 and F2. For F1, the The results shown that both lettuce crops presenting
specification of embitter and lateral are 2.4L/h, whereas the high uniformity of the corp growth within 70 to 90%.
F2 is 2.2L/h. we used 24 emitters with 4 laterals, space However, the crop growth of Bok Choy presented lowest
of each emitter was 317mm and space of lateral was uniformity due to the fact that it is sensitive to
325mm. we used pump that has pressure 2.5bar. destruction from insects such as worms.
Comparison between old drip system with new drip
system. Acording to measuring in greenhouse, The old
water system had different discharge and pressure that
caused drip irrigation nonuniformly. In addition, using
unproper excavating emmitter made different flow rate
in each lateral and provided nonuniform water supply.
Wheareas, installing new drip, we calibrated by
measuring flow rate. Following the graphic above, the
result is uniform compared to old drip

III.2. Crop growth

Figure 7: Crop growth uniformity during the plantation
Figure 6. Presents the crop growth during the second
event, F2. Getting uniformity of water, we can see the IV. Conclusion
varied growth of difference plants. Between 7 to 18 days
The study has shown the differences between old drip
of those plants weigh were similar, yet the weight started
and new drip experiment effect crop yield. After
extremely changes after 29 days because these various
installing a new irrigation system, there are two main
plants absorbed different fertilizer and amount of water.
factors: first, the selection of a new emitter and the use
After 3 months, the yields obtained R1 7000kg/ha, R2
of the CU Method to calculate the results.
6500kg/ha, R3 14000kg/ha. Each plant was healthier
1. The Christiansen uniformity coefficient (CU): For
and got higher yield than general types of plant because
old drip, F1 CU reach 54%, the crop yield is lower flow
some condition like uniformity of water, temperature is
rate. For new drip set up, level of CU reaches 91%. The
enough for plant growing, soil type of loam mix with
new drip system setup presented the successful of
worm compost and in a greenhouse no longer allow bad
irrigation uniformity.
inset to destroy.
2. The crop growth experiment of 3 plants, such as,
Red Oakleaf (R1), Red Lettuce (R2) and Bok Choy (R3)
in the new system installed presented the excellent crop

126
growth with the uniformity of 70-90%. The yields Zhao, Weixia, Jiusheng Li, Yanfeng Li, and Jianfe
obtained R1 7000kg/ha, R2 6500kg/ha, R3 14000kg/ha. ng Yin. 2012. “Effects of Drip System Unifo
However, the crop growth uniformity of Bok Choy is rmity on Yield and Quality of Chinese Cabb
age Heads.” Agricultural Water Management
affected by insects so that lead to lower uniformity
110:118–28. doi: 10.1016/j.agwat.2012.04.00
despites the irrigation uniformity. 7.

Acknowledgement
We are thankful to ARES-CCD for financial
support for this research.

References
Attia, Samir S., Abdel Ghany M. El-Gindy, Hani
A. Mansour, Soha E. Kalil, and Yasser E. A
rafa. 2019. “Performance Analysis of Pressur
ized Irrigation Systems Using Simulation Mo
del Technique.” Plant Archives 19:721–31.
Bralts, Vincent F., and Charles D. Kesner. 1983.
“Drip Irrigation Field Uniformity Estimation.
” Transactions of the American Society of A
gricultural Engineers 26(5):1369–74. doi: 10.
13031/2013.34134.
Li, Jiusheng. 1998. “Modeling Crop Yield as Affe
cted by Uniformity of Sprinkler Irrigation Sy
stem.” 38:135–46.
Phengphaengsy, Fongsamuth, and Hiroshi Okudair
a. 2008. “Assessment of Irrigation Efficienci
es and Water Productivity in Paddy Fields in
the Lower Mekong River Basin.” Paddy an
d Water Environment 6(1):105–14. doi: 10.1
007/s10333-008-0108-z.
Zhao, Weixia, Jiusheng Li, Yanfeng Li, and Jianfe
ng Yin. 2012. “Effects of Drip System Unifo
rmity on Yield and Quality of Chinese Cabb
age Heads.” Agricultural Water Management
110:118–28. doi: 10.1016/j.agwat.2012.04.00
7.
(Phengphaengsy and Okudaira 2008)
Attia, Samir S., Abdel Ghany M. El-Gindy, Hani
A. Mansour, Soha E. Kalil, and Yasser E. A
rafa. 2019. “Performance Analysis of Pressur
ized Irrigation Systems Using Simulation Mo
del Technique.” Plant Archives 19:721–31.
Bralts, Vincent F., and Charles D. Kesner. 1983.
“Drip Irrigation Field Uniformity Estimation.
” Transactions of the American Society of A
gricultural Engineers 26(5):1369–74. doi: 10.
13031/2013.34134.
Li, Jiusheng. 1998. “Modeling Crop Yield as Affe
cted by Uniformity of Sprinkler Irrigation Sy
stem.” 38:135–46.
Phengphaengsy, Fongsamuth, and Hiroshi Okudair
a. 2008. “Assessment of Irrigation Efficienci
es and Water Productivity in Paddy Fields in
the Lower Mekong River Basin.” Paddy an
d Water Environment 6(1):105–14. doi: 10.1
007/s10333-008-0108-z.

127
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Groundwater Quality Assessment in the coastal area of Preah Sihanouk and Kampot
province, Cambodia

Sovandy Sem1, Ech Chin2, Thida Khoeun2, Ratha Doung2, Kong Chhuon2, Sambo Lun2, Sytharith Pen2, Ratino
Sith2, Sylvain Massuel3, Khy Eam Eang2*
1
Master of UWE, Graduate School, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86,
Phnom Penh, Cambodia
2
Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia, Russian Federation B
lvd., P.O. Box 86, Phnom Penh, Cambodia
3
UMR G-EAU, IRD, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, Phnom Penh, Cambodia

*Corresponding author: khyeam_eang@yahoo.com

Abstract
An attempt has been made to assess the water quality within the fast-growing coastal zone of Sihanoukville and Kampot
area. Groundwater samples were collected from the coastal areas of Sihanouk Ville and Kampot and analyzed for major
physicochemical parameters. Experimental results appear that the well water in Kampot was strong acidic to slightly
alkaline. In comparison to Sihanouk Ville, the pH was slightly acidic to neutral in the range of 4.83 to 7.1, respectively. The
ORP value higher, and then in comparison the ORP value in Kampot from -42 to 211.3 mV higher than in Sihanouk Ville
from -31.4 to 173.4 mV. Otherwise, the concentrations of manganese probably released from the manganese-bearing rock
into the wells due to having ORP negative. The value of TDS in Kampot (660.66 mg/L) is higher than in Sihanouk Ville
(295.33 mg/L). The concentration of TDS in this study site slightly fluctuated between Sihanouk Ville and the Kampot location.
For the result of alkalinity from Sihanouk Ville, water samples contained higher alkalinity may be affected by the presence of
bicarbonate and carbonate. The alkalinity of water samples from Sihanouk Ville province was mean and dramatically
fluctuated gradually varied. Heavy metals distributed in all water samples from Sihanouk Ville within two wells, F3W13 and
F3W18, the value over standard, and the other samples resulted in exceeding the DWS and while significantly lowest the
WHO guideline. For the Fe, almost well water samples were found acceptable with the standard of drinking water and few
samples in Sihanouk Ville had Fe much lower than the standard drinking water. So, related to drinking water quality evaluation
resulted that almost all water samples from Sihanouk Ville and Kampot were observed unsuitable for drinking purpose in
terms of pH, turbidity, alkalinity, TDS, ORP, and Mn.
Keywords: Groundwater quality, Drinking water
I. Introduction industrial developments are found along the coast of
Sihanoukville and Kampot Province with the fastest-
Groundwater resources are beneath seriously growing regions and are powering the economic growth.
anthropogenic activities and the constant risk of Moreover, the coastal aquifer at Sihanoukville is stressed by
contamination. Human activities such as agriculture, households increasingly depending on tube wells to
urbanization, and industry have caused irreversible counterbalance the undersized dispersion network [2]. So far,
degradation of groundwater quality; in this manner, in a coastal area, no environmental observation networks
protection is the foremost suitable technique within the exist in these areas and barely any data are available
battle against groundwater contamination [1]. Cambodia’s preventing the assessment of the situation and its potential
coastal regions inferring from resource-dependent activities, evolution, especially for groundwater. The quality of water
there are advanced pressures caused by urbanization and is basically affected by pollution from different sources.
industrialization along the coastline. Most urban and Sectoral approaches partition the total accessible water,

128
leading to a shortage. Within the coastal zone, the major atomic absorption spectrophotometer (AAS, Shimadzu AA-
cause of groundwater pollution is seawater intrusion. Over- 7000).
exploitation may be an extreme issue that affects the
2.3. Analytical methods
potability of water. Hydrogeochemical processes are
controlling components of water chemistry [3]. The present In this study, the groundwater was taken from the well by
study was carried out in the coastal area (Sihanoukville) in pumping or drawing water with a bucket/container in case
order to analyze water quality. The seawater intrusion was of no pumping machine. At the same time, four parameters
reported from this area by international organizations [2]. such as pH, ORP, conductivity, and temperature were
Essentially, a few groundwaters actually contain measured at situ by EXO Sondes YSI.
constituents of well-being concern, and a few well waters at The alkalinity was analyzed only with water samples that
Sihanouk Ville which is used for the household is sometimes had pH greater than 4.5, while water with pH lower than 4.5
not known whether its quality is fulfilled with the standard was assumed that no alkalinity in the water [4]. The
of drinking water. Subsequently, understanding the effect of measurement was divided into 4 steps including: Sulfuric
groundwater on public health is exceptionally valuable. It is acid preparation, calibration of pH meter, alkalinity
exceptionally useful to be known all these specified issues. measurement, and alkalinity calculation.
So, the present study was conducted with an objective to + AAS analysis for the heavy metals: Each water sample
evaluate groundwater chemistry and water quality in the was filtered 0.45 μm membrane after adding 0.05 ml of nitric
coastal area of Preah Sihanouk and Kampot province which acid (65% of 𝐻𝑁𝑂3). The accuracy of the AAS result was
focuses on the current status of the integration of confirmed by maintaining the 𝑟2 value of the calibration
groundwater with seawater intrusion and the suitability of curve at least 0.99.
drinking water quality compared to WHO standard.
II. Materials and Methods
2.1. Study Area
The study was conducted along the Cambodia’s coastal
aquifers (Fig. 1) which include Preah Sihanouk and Kampot
province. There are 51 groundwater samples were collected
from tube wells and dug wells with several sources like
household, community, and rural supply well by 39
locations in Sihanouk Ville and 12 locations in Kampot. Fig. 2. AAS machine used for Fe, Mn, and As analyses
Straightway, to survey drinking water this study has utilized
the current information to compare with Cambodia Drinking
Water Quality Standard (DWS) and World Health
Organization (WHO) as shown in Table 1.
III. Results and Discussion
3.1. Current status of water quality
In-situ measurement of the groundwater quality
conducted by EXO shows that pH value ranged out of the
allowable limit of DWS and WHO of drinking water (Fig. 3
and Table 2). Well water in Kampot was strongly acidic to
slightly alkaline with the pH. In comparison to Sihanouk
Fig. 1. Map of study area Ville, the pH was slightly acidic to neutral in the pH range
of 4.83 to 7.1, respectively. According to Appelo and Postma
2.2. Experimental set-up (2005), the pH ranges <6.3 resulted in abundant carbonic
Physicochemical properties include TDS, ORP, Temp, acid (H2CO3), indicating that wells from Sihanouk Ville and
pH, Turbidity and Conductivity were analyzed in-situ using Kampot contained H2CO3. In comparison, dug/tube wells
YSI EXO2 multi-parameter instrument. Then, well water from Sihanouk Ville obtained higher HCO3- due to pH
sample were analyzed in SATREPS laboratory of Institute ranged from 4.83 to 7.1. The concentration of Fe and Mn
of Technology of Cambodia. Alkalinity was determined by perhaps elevated in well water from Kampot due to acidic
0.01M sulfuric acid titration, harmful substances [Arsenic water [5].
(As), Iron (Fe), and Manganese (Mn)] were analyzed by Electrical Conductivity: The value of each well water
sample in both areas slightly fluctuated. Also, the twelve

129
sampling in Sihanouk Ville and Kampot location have high Total Dissolved Solids: The TDS distribution in Sihanouk
EC especially found in well F2W2, F3W2, F1W12, F1W15, Ville exhibited of a minimum 44 mg/L, a maximum of 1390
F3W14, and F3W20. The other sampling was exceeded the mg/L and a mean of 295.33 mg/L. The value of TDS in
desirable limit in drinking water of WHO, 250 μS/cm. Kampot ranged from 17 mg/L to 3212 mg/L with a mean of
Oxidation-reduction potential (ORP): In Sihanouk 660.66 mg/L. In Kampot, the TDS values were higher than
Ville, the ORP value ranged from -49 to 207.2 mV with a in Sihanouk Ville. The concentration of TDS in this study
mean is 95.83 mV. In Kampot, the ORP varied from -42 to site slightly fluctuated between Sihanouk Ville and the
211.3 mV with average is 112.76mV. The ORP value the Kampot location.
higher than in Sihanouk Ville. Perhaps, the concentrations Alkalinity: The wells contain low alkalinity with low pH,
of manganese probably released from the manganese rock indicating corrosive concern. Fig.5. illustrated the detailed
into the wells due to having ORP negative [6]. So, other well result of alkalinity from Sihanouk Ville. Water samples
numbers may obtain more manganese concentration than contained higher alkalinity may be affected by the presence
other wells. of bicarbonate and carbonate.
3.2. Heavy metals
Preah Sihanouk
8
7 Manganese (Mn): The manganese for drinking purposes
6 was recommended by the Cambodia standard and WHO, 0.1
5 mg/L and 0.5 mg/L, respectively. Fig.6. illustrated the
4
3 concentration of Mn distributed in all water samples from
pH

2 Sihanouk Ville. In this study, two wells the value over

1 standard as F3W13 and F3W18, and the other samples
0
resulted in exceeding the DWS and while significantly
lowest the WHO guideline.
Water samples Iron (Fe): The iron for drinking water was recommended by
the WHO and DWS in Cambodia, 0.3 mg/l almost all well
Kampot water samples were found in the standard of drinking water
8
7 except few samples in Sihanouk Ville that were not in the
6
5 standard of drinking water. So, the result of Fe in Sihanouk
pH

4 is good for household use. In addition, the amount of iron in

3
2 water causes a stain on laundry, materials, and vegetables.
1
0 Table 1. Drinking Water Quality Standard [7, 8]
Parameters Unit Cambodia WHO
pH - 6.5-8.5 6.5-8.5
Water samples
Turbidity NTU 5 5
Fig. 3. The result of pH at Sihanouk and Kampot collection TDS mg/L 800 -
dash-red line means maximum value of Cambodia EC µS/cm - 250
drinking Standard Hardness mg/L - 300
Turbidity: The value of turbidity in Sihanouk Ville higher Alkalinity mg/L - -
than in Kampot. Fig.4, referring to the result in Kampot, five Arsenic (As) mg/L 0.05 0.01
sampling points such as F2W6, F1W4, and F3W3. In Iron (Fe) mg/L 0.3 -
Sihanouk Ville, there were six locations such as Manganese (Mn) mg/L 0.1 0.5
F3W14, F3W15, F3W16, F3W18, F3W19 and F3W23 were -: not mentioned
accounted for turbidity seriously higher than standard.

Table 2. Ranges of physicochemical parameters and their comparison with MIME and WHO standards for drinking water
Site Temp (C) Cond (µS/cm) TDS (mg/L) pH ORP (mV) Turbidity (FUN)
Preah Min 27.5 16.1 11 3.82 -49 -4.69
Sihanouk Mean 29.2 304.85 187.52 5.94 95.83 61.35
Max 40.1 2294.4 1390 7.61 207.2 1202.4
Kampot Min 28.8 28.1 17 3.89 -42 -3.65
Mean 29.6 294.12 660.66 5.42 112.76 2.99
Max 30.7 1390.5 3212 7.23 211.3 24.73
MIME/WHO - 250 800 6.5-8.5 - 5
-: not mentioned

130
 Kampot Preah Sihanouk
30 75
25 65
Turbidity (FNU)

Turbidity (FNU)
20 55
15 45
35
10
25
5 15
0 5
-5 -5

Water samples Water samples

Fig. 4. Result of turbidity in Kampot and Preah Sihanouk, dash-red line means maximum value of Cambodia drinking standard

standard especially pH and EC. There are only 2 wells in

2 Preah Sihanouk Sihanouk Ville have Mn above the drinking standard.
Akalinity (mg/L)

1.5
Acknowledgement
1
0.5 We are thankful to the Science and Technology Research
0 We are thankful to the 4-C water project for the financial
support under the management of Institut de Recherche pour
-0.5
le Développement (IRD).
References
Water samples
[1] Kazakis, N., Voudouris, K. S. (2015). Groundwater
Fig. 5. Result of alkalinity at Sihanouk vulnerability and pollution risk assessment of porous
aquifers to nitrate: Modifying the DRASTIC method
Preah Sihanok using quantitative parameters. Journal of Hydrology,
0.5 525, 13-25.
[2] Rizvi, A. R., Singer, U. (2011). Cambodia Coastal
Mn (ml/l)

0.3 Situation Analysis. Education, 2(2), 3.

[3] Kumar, P. S., Elango, L., James, E. J. (2014).
0.1 Assessment of hydrochemistry and groundwater
quality in the coastal area of South Chennai, India.
-0.1 Arabian Journal of Geosciences, 7(7), 2641-2653.
[4] Appelo, C. A. J., Postma, D. (2005). Geochemistry,
Groundwater and Pollution, 2nd edition. Boca Raton
Water samples London New York. CRC Press, Taylor & Francis Group.
Fig. 6. The result of Mn dash-red line means maximum [5] Abdelshafy, M., Saber, M., Abdelhaleem, A., Abdel-
value of Cambodia drinking razek, S. M., Seleem, E. M. (2019). Hydrogeochemical
processes and evaluation of groundwater aquifer at
IV. Conclusion Sohag City, Egypt. Scientific African, 6, e00196.
[6] Bouteldjaoui, F., Kettab, A., Bessenasse, M. (2017).
Water from the dug/tube well is an important source for
Identification of the Hydrogeochemical Process in
various uses; however, related to drinking water quality,
Zahrez Basin, Algeria. Algerian Journal of
assessment resulted that almost all water samples from
Environmental Science and Technology, 3(1).
Sihanouk Ville and Kampot were observed unsuitable for
[7] MIME. (2004). Kingdom of Cambodia Drinking Water
drinking purpose in terms of pH, turbidity, alkalinity, TDS, Quality Standards. In Drinking Water Quality Standards.
ORP, and Mn. The result of Fe and temperature from [8] World Health Organization (2011). Guidelines for
Sihanouk Ville is lower than the standard. Kampot has only drinking-water quality, 4th ed. Geneva: World Health
4 wells and Sihanouk Ville have 9 wells that stay in the water Organization.
quality. Parameters in the wells are exceeding the limited

131
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

_________________________________________________________________________

Comparison Of Modelling of Soil Water Dynamic In Two Distinct Soils

Kim Sreang DY 1*, Pinnara KET2 Seum MAO1, Channtola SOT1, Mengheak PHOL1, Sophanith
THA1, Makara SOY1, Somnang SRIM1, Vouchleang SREANG1, Chantha OEURNG1
1
Faculty of Water and Environmental Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia, Russian
Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia

* dykimsreang168@gmail.com

Abstract

Understanding the performance of soil water dynamic is important for agricultural and water resources
management. Dealing with the infiltration process and its prediction is an important topic for current
research. The objective of the study is to characterize water dynamic in two distinct soils in urban garden by
using HYDRUS 1D. Infiltration experiment loam and sandy soils was conducted at Soil laboratory at Institute of
Technology of Cambodia. Mini-tension infiltrometer was used to measure the soil infiltration process. The data
from experiment was used to test the Hydrus 1D model. The result of simulation of infiltration shown that
Hydrus 1D has capability to simulate the infiltration well.

Key words: Soil degradation, soil water flow, HYDRUS-1D

I. Introduction
conditions flexibly (Simunek et al. 2008). It has
In Cambodia, irrigation is a significant water- been used worldwide and achieved a great deal
related activity, especially in urban area where of valuable research results.
agriculture cannot exist without adequate There is still the need of understa
irrigation. To manage efficient water use irrigation, nding of infiltration process from different soil
infiltration is crucial part to be defined. For especially in urban garden.
studying water movement Characteristics in In this research, we conducted a field
infiltration process, numerical simulation has infiltration experiment of sandy soil and loam
become an important technique. The HYDRUS soil of greenhouse garden at soil laboratory of
model can simulate water infiltration in the Institute of technology of Cambodia. The ability
variable saturated zone with the term source and of applying Hydrus 1D to simulate the
sink considered and manage different boundary infiltration process is our research question.

132
 II. Methodology

2.1 Study site and experimental set up

The infiltration experiment was conducted at

Institute of Technology in October 2020 (Figure
1&2). There were two types of soil, such as loam
soil (for growing vegetables) and sand soil, which
were used for this experiment. The data was
gathered from 10 crop cultivation boxes, which
have 1.3m ×1.9m in one box. And with a 317mm
emitter, the lateral space is 325mm. a b
) )
Figure 2. Infiltration on loam soil at greenhouse

2.3 HYDRUS-1D Model Setup

The HYDRUS-1D code was based on the

one-dimensional Richards equation to simulate
water movement in variably saturated media, and
the equation was solved by numerical method (Sim
a ˚unek et al., 2005). The basic water movement
) equation was described as:

The soil physical properties were estimated

using the equation of van Genuchten et. al 1991.

b = h>0
)
Figure 1. Experimental design
h≤0
Where h is the soil water pressure head (cm), is the
2.2 Mini-disc infiltrometer water content (cm3/cm3), r and s are the residual
and saturated water contents (cm3/cm3),
Mini-disc infiltrometer is an device used to respectively, ˛, m and n are empirical parameters
measure the rate of water infiltration into soil or and m=1 − 1/n.
other porous media.
= ά h>1

133
 result needs to be calibrated by the oven dry to
Where h is the soil water pressure head, θ is confirm this properties. It is reasonable that Ks
the volumetric water content, t is time, z is the suggested by model proposed the small value from
vertical coordinate with the origin at the soil loam soil and high value from sand soil with Ks f
surface (positive upward), and K(h) is the 0.0029 mm/s and 0.0825 mm/s respectively.
unsaturated hydraulic conductivity. For the Where α’ is an empirical parameter (1/cm) and it is
experiment studied, the initial condition and upper the reciprocal of ha, ha is often referred to as the
boundary condition were: air entry value (cm), and is the pore-size
h(z, 0) = hi(z) distribution parameter affecting the slope of the
h(0, t) = h0 retention function.
Where hi(z) is the initial soil water pressure The unsaturated hydraulic conductivity of each soil
head through the soil column, and h0 is the soil layer can be expressed as (van Genuchten et al.,
water potential at soil surface. 1991):
The free drainage was to be considered as
lower boundary condition: ∂h/∂z=0.
The soil profile was 20 cm depth. Ks{1 − ( h)mn [1 + ( h)n]−m}2
K(h) =
α’h ≤ 1 [1 + ( h)n] m

contents of the two different soils, e.g. loam and

sand. This is the first proposition. However, this

III. Results

1. Physical hydraulic properties simulated

The results of estimation of hydraulic properties from the experimentation were shown in the Table 2. The model
proposed the same soil moisture contents of the two different soils, e.g. loam and sand. This is the first
proposition. However, this result needs to be calibrated by the oven dry to confirm this properties. It is
reasonable that Ks suggested by model proposed the small value from loam soil and high value from sand soil
with Ks f 0.0029 mm/s and 0.0825 mm/s respectively.

Table 2: Water Flow of Sandy and Loam Soil

Soil types θr θs α (1/mm) n Ks (mm/sec) l

Sand 0.045 0.43 0.0145 2.68 0.0825 0.5
Loam 0.078 0.43 0.0036 1.56 0.0029 0.5

Values of soil hydraulic parameters as obtained from the HYDRUS-1D (θr is the residual water content,
θs s the saturated water content, α, n, and l are empirical parameters, and Ks is the saturated soil hydraulic
conductivity).

134
 1. Infiltration process simulated

- Loam Soil
Bottom Flux
Cumulative infiltration increased from 0sec
to <250 (figure 3). 0.00000005
The Loam Soil Pressure Head
0.00000000
concentration was drop down slowly from
h > 75 mm to nearly 250 sec. And the data -0.00000005

for Loam Flux Loam Soil decreased from -0.00000010

>0.00 mm/sec at time 0 sec to >- -0.00000015
0.00000025 mm/sec at time > 250 sec
-0.00000020
Cum.(figure
Infiltration
4&5).
-0.00000025
60
-0.00000030
50 0 50 100 150 200 250 300
Time [sec]
40
Fig.5. Bottom Flux of loam Soil
30
- Sandy Soil
20
Cum. Infiltration of Sand Soil was increased fatly
10
from 0 mm to <22mm at time from <0 sec to 60 sec
0 50 100 150 200 250 300
Tim e [sec]
(figure 6). However, at the time of 0 sec >30 sec,
the Bottom Flux of Sand Soil was decreed that
Figure 3. Cum. Infiltration of Loam soil 0.00000001mm/sec to >-0.00000008mm/sec, but at
Surface Pressure Head
the time of 30 sec to 60 sec, the bottom flux was
75 slowly distressed (figure 7).
For Surface Pressure Head of was drop down from
70
the h > 75 mm at almost 60sec (figure7).
65 Cum. Infiltration

60 22

20
55
18
50 16
0 50 100 150 200 250 300
Time [sec] 14

12
Figure 4. Pressure Head of Loam Soil
10
0 10 20 30 40 50 60
Tim e [sec]

Figure.6 Cum. Infiltration of Sand Soil

135
 further work is to calibrate and validate the
This saturation persisted less than 60mm. Very performance of the model with statistical analysis.
slight changes in soil water pressure head during
storms effected relatively large changes in both Acknowledgement
magnitude and direction of water flux.
We are thankful to the Science and Technology
The results of this experiment differed significantly
Bottom Flux Research Partnership for Sustainable Development
between Sand soil and Loam Soil.
(SATREPS), the Japan Science and Technology
0.00000000
Agency (JST)/Japan International Cooperation
-0.00000001
Agency (JICA) for their financial support.
-0.00000002
-0.00000003
References
-0.00000004
-0.00000005
Bouwer, H., 1986. Intake rate: cylinder
-0.00000006 infiltrometer. Methods of Soil Analysis: Part 1
-0.00000007 Physical and Mineralogical Methods, 5, pp.825-
-0.00000008 844
0 10 20 30 40 50 60
http://www.fao.org/3/r4082e/r4082e03.htm#2.2.
Time [sec]
Fig8: Bottom 1%20the%20infiltration%20process
Surface Pressure Head Flux of Sand Soil https://www.sciencedirect.com/topics/earth-and-
planetary-sciences/infiltrometer.
80
Šimůnek, J., M. Šejna, H. Saito, M. Sakai, and
70 M. Th. van Genuchten, The HYDRUS-1D software
60 package for simulating the one-dimensional
50
movement of water, heat, and multiple solutes in
variably-saturated media, Version 4.0, Hydrus
40
Series 3, Department of Environmental Sciences,
30 University of California Riverside, Riverside, CA,
20 USA, 2008.
Šimůnek, J., M. Th. van Genuchten, and M.
10
Šejna, Development and applications of the
0 10 20 30 40 50 60
HYDRUS and STANMOD software packages, and
Time [sec]
Fig.7. Surface Pressure Head of Sand Soil related codes, Vadose Zone Journal,
doi:10.2136/VZJ2007.0077, Special Issue “Vadose
Zone Modeling”, 7(2), 587-600, 2008.
IV. Conclusion Van Genuchten, M.V., Leij, F.J. and Yates,
S.R., 1991. The RETC code for quantifying the
In this study, the infiltration process has been hydraulic functions of unsaturated soils.
conducted at institute of technology of Cambodia.
Hydrus 1D has been used to simulate the water
dynamic of the process. The result proposed that
saturated conductivity simulated for sandy soil was
28 time of the loam soil with a value of 0.0825 and
0.002 respectively.
The sandy soil pressure head decreased sharply
with time compared to the loam soil. However, our

136
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Comparison of Coral Bleaching Hotspot Mapping in Southeast Asia (Thailand, Cambodia

and Vietnam) based on Sea Surface Temperature Modelling by National Oceanic and
Atmospheric Administration Coral Reef Watch before and during Covid-19 Pandemic

Rajendra Khanal 1*, Pham Quy Giang 2, Binaya Kumar Mishra 3, Ratino Sith 4, Soly Siev 5, Varinthorn
Boonyaroj 6. Vannak Ann 7, Vengsong Khov 8, Jorge Garcia-Hernandez 1

Department of Civil and Environmental Engineering, School of Environment and Society, Tokyo Institute of T
1

echnology, 2-12-1-M1-4, Ookayama, Meguro-ku, Tokyo, 152-8552, Japan; 2 Faculty of Environment, Ha Long
University, Vietnam; 3 Faculty of Science and Technology, Pokhara University, Nepal; 4 Faculty of Hydrology a
nd Water Resources Engineering, Institute of Technology of Cambodia, Phnom Penh, Cambodia; 5 Department
of Science, Technology & Innovation Policy, Ministry of Industry, Science, Technology and Innovation, Cambo
dia; 6 Faculty of Science and Technology, Rajamangala University of Technology, Bangkok, Thailand; 7 Water
and Environment Unit, Institute of Technology of Cambodia, Phnom Penh, Cambodia;8 Tonle Sap Authority, Mi
nistry of Water Resources and Meteorology, Cambodia

* Corresponding author: khanal.r.aa@m.titech.ac.jp

Abstract
Due to Covid-19 pandemic in the year 2020, international travel and hence the tourism was down by more than 95%
globally. Influx of tourism has been regarded as one of the reasons for coral bleaching. In this study, attempt has been made
to compare the coral bleaching hotspot (CBH) mapping in the year 2019 (pre-pandemic) and 2020 (during pandemic) based
on sea surface temperature (SST) modelled data by National Oceanic and Atmospheric Administration Coral Reef Watch
(NOAA-CRF). The NOAA 7-day maximum SST database is compared for the Coral Triangle region – with a focus on gulf
of Thailand and south China sea covering Thailand, Cambodia and Vietnam in southeast Asia – for the first day of the month
in Jan, Mar, June, Sep, and Nov both in 2019 and 2020. The CBH is taken as the measure of the difference between observed
SST and monthly maximum mean temperature, and was measured in the range 0 to 5 OC. No visual (< 0 OC) CBH was
observed in the region for the month January and March pre-pandemic and during pandemic. In the month of May, CBH (1 -
2 OC) was dominant in the gulf of Thailand covering mostly Thailand and Cambodia, and was higher during pre-pandemic
period. During July, CBH (1 - 2 OC) was dominant in the south China sea covering Vietnam, and was also higher during pre-
pandemic period. Surprisingly, CBH was higher during pandemic along the Vietnam in south China sea, the reason of which
is still a matter of further investigation. In general, with an exception in September 2020, CBH was higher during pre-
pandemic period. A detailed study covering daily, and monthly average SST would provide better understanding of impact of
covid-19 pandemic on CBH.

Keywords: Coral bleaching hotspot, covid-19 pandemic, sea surface temperature, National Oceanic and Atmospheric
Administration, Coral Reef Watch, gulf of Thailand, south China sea

Acknowledgement
Authors would like to express gratitude to Asia-Pacific Network for Global Change Research for funding this project -
"Collaborative Research Platform to Manage Risk and Enhance Resilience of Coral Reef in Southeast Asia, CRRP2019-
08MY-Khanal". Acknowledgements also goes to National Oceanic and Atmospheric Administration, Coral Reef Watch for
the data

137
 No data 0 1 2 3 4 5 OC

2019 (pre-pandemic) 2020 (during covid-19 pandemic)

80 100 120 140 80 100 120 140

20
1 November

20

0
0

-20
-20

20 20
1 September

0 0

-20 -20

20 20
1 July

0
0

-20
-20

20 20
1 May

0 0

-20 -20

20 20
1 March

0 0

-20 -20

20 20
1 January

0
0

-20 -20
80 100 120 140 80 100 120 140

Fig. 1. Coral bleaching hotspot mapping in the year 2019 (pre-pandemic) and 2020 (during pandemic) based on sea surface
temperature modelled data by National Oceanic and Atmospheric Administration Coral Reef Watch.

138
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Paradox of over-tourism, income opportunities and coral degradation: A case of Maya

bay, Thailand

Rajendra Khanal 1*, Varinthorn Boonyaroj 2, Jorge Garcia-Hernandez 1

Department of Civil and Environmental Engineering, School of Environment and Society

1

Tokyo Institute of Technology, 2-12-1-M1-4, Ookayama, Meguro-ku, Tokyo, 152-8552, Japan

2
Department of Environmental Science and Technology, Faculty of Science and Technology, Rajamangala University o
f Technology Phra Nakhon, 1381 Pracharat 1 Road, Wongsawang, Bang Sue, Bangkok, Thailand 10800

* Corresponding author: khanal.r.aa@m.titech.ac.jp

Abstract
Tourism plays an important role in the economic development of the country. Tourism contributes to as much as 7% of
Thailand national GDP. Pristine beaches in Thailand attracts millions of tourists every year. On one hand, local economy is
booted with tourism. Contract to that, over-tourism may lead to stress on the local environment. One of the classical examples
of impact of over tourism on the environment is the degradation of pristine beach and coral communities in the Maya bay, Phi
Phi Leh, southern Thailand. This paper, aims to analyze the paradox of over-tourism, income opportunities and the impact on
coral community in Maya bay, based on the literatures. Ever since, Maya bay was known to the world in early 2000’s, the
number of tourists visiting there every day increased by close to 3000-fold in the last 20 years. Though, tourism helped to
increase the local economy dramatically, later, due to impact of probably over exploitation of corals from snorkeling and
diving, and wash-off of the toxic UV-filters led to bleaching of the corals. Most of the corals were dead in 20 years of tourism
exploitation by 2017. Government of Thailand came up with the strategy to ban Maya bay as a tourist hotspot, and promoted
coral recovery. Because of the solid policy, and action plan of the multiple stakeholders, in the last three years, corals has been
significantly recovered. This paper discusses about the strength, weakness, challenges, opportunities and threat of the action
plan in restoring coral community in Maya bay.

Keywords: Maya bay, UV-filters, coral planting, bleaching, over-tourism, stakeholder management

I. Introduction long-term. Over-tourism is a shared responsibility, and can

Over-tourism is defined as a stage whereby the excessive be managed by sustainable tourism. Prioritization of the
visit by the tourists in a particular confined space leads to welfare of the local diversity, income opportunities, and
overcrowding and stress on the ecosystem services. A consideration to the restoration of ecosystem services are the
number of tourist hotspots has reported over-tourism in important framework for sustainable tourism.
recent years, which includes, Barcelona - Sagrada Familia, Tourism serves as the significant backbone of Thailand
Indonesia - Bali, India - Taj Mahal, Nepal – Everest, etc. economy, contributing to as much as 10% of national GDP.
Over-tourism occurs as a result of certain boom in unique Coral reefs along the gulf of Thailand and Andaman sea are
specialty of the particular area, so called a viral status in one the pristine corals in the world. These corals are the
today’s digital world. Over-tourism can be short term, or marine invertebrates, which vary in shape, size, softness,

139
and fluorescence. The unique fluorescence of the corals, and touch, BBC, 20 February 2019)
the huge diversity of aquatic flora and fauna are depended
on the reef build by the corals. Some of these corals grow at The representative figure for the over-tourism in Maya
a very slow rate, about 5mm per year, and it takes hundreds bay is shown in Fig. 1. Prior to 2000s, Maya bay was pristine.
and thousands of years to build the reefs. These reefs made During early 2000s, by 2008, around 30% of the coral were
by the corals supports ecosystem both in land and water. bleached. The number of tourists in early 2000 used to be
With an urbanization, unplanned growth, discharge of less than 10, it rose to around 170 by 2008, and by 2017,
municipal solid waste and wastewater, marine plastics, number of tourists rose to as much as 4000. The beach in
coupled with climate change has brought tremendous stress itself is around 200m. As much as 18 million tourists were
to the coral community reported to visit Maya bay, Phi Phi Island in 2017 [3]. This
One of the most popular tourism spots in Thailand is the huge number of people in such a confined area exerted
pristine white sandy beaches. May bay in Phi Phi island lies tremendous pressure on the coral environment. The increase
in the Krabi Province, southern part of Thailand lies at 7° 40' in tourism, though significantly led to increase in local
44.0004'' N and 98° 45' 53.9964'' E. The economic value of economy, however also led to some serious problems like,
coral tourism in Phi Phi island amounts to as much as USD disposal of solid waste at the beach, degradation of water
300 mil (approximate value in 2020, based on 110 mil in quality, and coral bleaching. The rate of coral bleaching, and
2003) [1]. haphazard waste along the Maya bay beach continuously
The main factors that led to destruction of corals in Maya increased, as a result, close to 90% of the corals were
bay are, i) pollutants from wash-off of UV-filters used to degraded by early 2010. Still, due to popularity of the beach,
prevent sunburn, ii) leakage of the boat gasoline, iii) led to increase in number of tourists. By 2015, almost all
recreational activities including jetski, boat, yachts, corals were bleached. This send an alarming message, as a
speedboats, and iv) mechanical destruction of corals during result, Government of Thailand proposed a policy to
diving, anchors, and propellers. completely banned tourism in Maya bay in 2018. Slowly and
The main objective of this paper is to review the policies steadily corals are being restored. However, recovery is a
behind promotion of tourism, over-tourism impact on slow process.
environment and the economy, and critical analysis of
restoration policies in conservation of corals in Maya bay. 2.2. Economic aspects of tourism
According to the data of Ministry of Tourism and Sports,
II. Discussions Thailand, the number of tourists visiting Thailand has
increased dramatically in the last 50 years. It used to be less
2.1. Over-tourism in Maya bay than 100,000 in 1960, and has increased to 10 million by
2008 2017 2000. Furthermore, with intensive tourism marketing,
Thailand welcomed close to 40 million tourists in 2018. This
makes Thailand one of the most popular tourist spots in the
world. Most of the tourists do visit the white sandy Thai
beach - railway beach, Koh Phi Phi, Tonsai beach, etc. At
Maya bay, local government charges ~ 14 USD per person
to visit the beach, and with close to 4000 people visiting
Maya bay every day, the amount of revenue generated per
day is quite huge ~ 56,000 USD. Tourism can have multiple
impact on environment, economic, local community,
infrastructure including hotel, transportations, transportation,
ecosystem services, all of which are again interrelated and
interdependent.

2.3. Coral restoration and conservation strategy

Fig. 1. Representation of the number of people visiting May Thai government was aware about the impact of over-
bay (Adapted from [2], Saira Asher, the beach nobody can

140
tourism on ecosystem services in Maya bay. The steps taken sustainable tourism. Qualitatively, it can be said as the
by the Thai government in preserving corals in Maya bay maximum number of tourists that can be handled by
includes, initiation of fine by the local environmental operating sites, without hampering the aesthetic tourism
authorities of ~ 18 USD to boat operators who throw anchors quality. Koh and Fakfare (2019) recently proposed the
into the water, waste collection bin, banning on fishing and unique strategy for the stakeholder management for the
construction. It was however difficult to follow the restoration of Maya bay – the BRAVE framework consisting
implementation, due to lack of manpower on monitoring, of business, residents, authorities, visitors and
and also ignorance from the tour operators. Even, some tour environmentalists [6]. The interests of different stakeholders
operators, who used to bring lots of tourist, continued to pay are different. Since, a beach area is quite open. It is nearly
fine and breached the directives. Following a series of impossible to restrict the number of people entering such
campaigns by the environmentalist, Thai government kind of open area, it all depends on concept of “responsible
banned a visit to Maya bay by judicatory order in June 2018, tourism”, whereby the BRAVE framework proposes the
firstly for a period of 6 month, and later from Oct 1, 2018 integrated action plan to limit the number of visitors at a
has been closed indefinitely [3]. particular site, perhaps alternate spots of tourism sites,
proper waste collection strategy, and promoting the use of
2.3.1 Engineering aspects of coral restoration non-toxic UV-filters.
A number of reports are available on the coral restoration
strategies, among them a study by Boström-Einarsson et al., Residents
(2020) seems quite comprehensive [4]. One of the biggest
problems in coral restoration is that, each restoration is seen
as a project lasting 6 to 24 months. A restoration of up to 12
years has been done in Caribbean coral reefs, which was
though successful but was not a solution to restore coral
community. The restoration methods can be divided into Societal -
transformation
physical, biological and structural restoration. Some of the
restoration strategies as proposed by Boström-Einarsson et
al., (2020) includes direct transplantation of coral, coral Visitors
gardening, micro-fragmentation, genetic diversity in asexual
propagation, larval enhancement, artificial reefs, and
substratum stabilization. Fig. 2. Conceptual framework for achieving sustainable
tourism (Adapted from [6], Koh and Fakfare (2019).
2.3.2 Japan experience on coral restoration (Unscaled / qualitative figure)
In Japan, the action plan to conserve coral reef ecosystem [5]
has been focused on three priority issues, i) measures against The framework for the sustainable tourism is shown in
negative impact from runoff, ii) promotion of regional Fig. 2. The concept is adapted from BRAVE framework
economy through sustainable tourism, and iii) promotion of proposed by Koh and Fakfare, 2019. The green circle
harmonic relationship between community and ecosystem. represents the interdependency of the BRAVE stakeholders.
The basic action to realize above three priority issues The tourism business can be checked by the authorities by
includes, information sharing, awareness training, the legislation. The aim of this legislation should not be to
communication outreach, capacity development, restoration grip business, but provide the competitive access for
projects, planned urbanization and coastal reconstruction, business opportunities to all the interested one. The three
water quality conservation, surveillance and monitoring, other players, residents, visitors and environmentalists sole
treatment of effluents, sustainable use of marine ecosystem depend on the societal-transformation strategy, which means
services, and follow up and review. to take responsibility for the conservation of environment,
without unduly compromising on one’s wish, to promote
2.4. Sustainable tourism responsible tourism.
It is really quite complex to define the indicator for

141
2.5. SWOT analysis the beaches affected by over tourism. Conservation strategy
The SWOT analysis of the conservation strategy taken should be extended to all the beaches, in the form of
by the Thai authorities is given in Table 1. The strong sustainable and responsible tourism.
government approach in conservation, and availability of
plenty of alternative tourist spots are the strength and Acknowledgement
opportunities of the project. Training of the tour operators,
Authors would like to express gratitude to Asia-Pacific
and maintaining database on spatio-temporal variation of
Network for Global Change Research for funding this
tourist visit, frequency, time period is a challenge and a
project - "Collaborative Research Platform to Manage Risk
weakness. However, definition of sustainable tourism in
terms of quantitative variables is really complicated. The and Enhance Resilience of Coral Reef in Southeast Asia,
proper balance between SWOT will lead in success of the CRRP2019-08MY-Khanal".
conservation strategy.
References
Table 1. SWOT analysis of coral restoration strategy in
Maya bay, Thailand [1] Christiernsson, A. 2003. An Economic valuation
Strength Weakness of the coral reefs at Phi Phi Island, master’s thesis,
• Strong government and • Quantitative Department of business administration and
local people support parameters to define spocial sciences, Lulea university of Technology,
• Funding opportunities for sustainable tourism [2] Saira Asher, the beach nobody can touch, BBC,
the restoration, • Qualitative and
20 February 2019) assessed December 10, 2020
• Alternate tourism quantitative measure to
opportunities define pollution, and and available at
economy https://www.bbc.co.uk/news/resources/idt-
Opportunities Threat sh/the_beach_nobody_can_touch
• As a model site for • Poacher tourism [3] SeATM, 2018. South east asia tourism monitor,
studying case of over- • Continuity of the briefing on tourism, development and
tourism preventive measures environment, Vol. 9, No.3, May – June 2018
• Multiple stakeholder • Change in policy due [4] Boström-Einarsson L, Babcock RC, Bayraktarov
engagement to socio-economic
E, Ceccarelli D, Cook N, Ferse SCA, et al. (2020)
• Consensus between pressure
government, tourism • Lack of trainings on Coral restoration – A systematic review of current
operators, and local people sustainable tourism methods, successes, failures and future directions.
PLoS ONE 15(1): e0226631
III. Conclusion [5] MOE, 2017. The Action Plan to Conserve Coral
Reef Ecosystems in Japan 2016 – 2020.
There are 11 strategies proposed by the United Nations Biodiversity Policy Division, Nature
World Tourism Organization to manage over-tourism, some Conservation Bureau, Ministry of the
of which includes, promotion and dispersion on a time and Environment, Japan
location basics, visitor segmentation, review of the policies, [6] Koh, E., & Fakfare, P. (2019). Overcoming “over-
and establishment of the surveillance and monitoring tourism”: The closure of Maya Bay. International
mechanism [7]. Journal of Tourism Cities, 6, 2, pp. 279-296
Impact of restoration policies are promising, and [7] World Tourism Organization (UNWTO);
significant improvement in coral community and water Centre of Expertise Leisure, Tourism &
quality has been observed in two years in Maya bay. Hospitality; NHTV Breda University of Applied
However, it is still early days of coral restoration, which is a Sciences; and NHL Stenden University of Applied
time-consuming process. Hence, all the stakeholders should Sciences (2018), ‘Overtourism’? – Understanding
be patient, and the decision to resume tourism in Maya bay and Managing Urban Tourism Growth beyond
should be supported by the quantitative evidence-based Perceptions, Executive Summary, UNWTO,
policies, and quantitative indicators. Maya bay is just one of Madrid

142
Session 3: Lake Environment
The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
The 5th International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Development of a Package Containing PAC and Ca(OCl )2 for Drinking Water

Treatment of Lake Water

Sereyvathana SOK1, Phary THACH1, Kazuhiko MIYANAGA2 and Reasmey TAN1,3,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
School of Life Science and Technology, Tokyo Institute of Technology, 4259 J3-8
Nagatsuta-cho, Midori-ku, Yokohama, 226-8501, Japan
3
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: rtan@itc.edu.kh

Abstract

The conventional treatment practices by villagers of usage alum can cause overdosing of concentration of alum
since it can cause the water taste to be sour and is considered as the health risk in drinking water such as vomiting,
diarrhea, mouth and skin ulcers, and so on. Therefore, the objective of this study was conducted by comparing the
effects of yellow PAC and white PAC in removing turbidity, changing pH value, and settling time and the effects
of using Ca(OCl)2 in removing bacteria in lake water to develop a package containing PAC and Ca(OCl)2 . To
achieve these objectives, the lake water sample was treated with the yellow PAC and white PAC. Two methods
were used in bacteria removal, sole using of Ca(OCl)2 and dual using of Ca(OCl)2 with PAC. As a result, the
highest efficiency of turbidity removal was within 97.66–99.89% and 91.55–99.88% for yellow and white PAC,
respectively between the settling time of 60–120 mins. The changed of pH for yellow PAC complied with the
standard for drinking water which is 6.5–8.5, while the pH for white PAC did not comply with the standard. For
the results of bacteria removal, 100 % of E. coli was removed in both methods and 98.8–99.4% for the first method
and 96.3–99.4% for the second method of total coliform removal were achieved. Overall, yellow PAC showed the
better performance in removing turbidity, subsequently the second method of bacteria removal showed the best
settling time is 60–120 mins. Moreover, yellow PAC has gradually changed in pH and adjusting pH was not need.
Finally, It was conducted the net contents of one package are 12.75g which is contained yellow PAC and
Ca(OCl)2 for treating 100 L of lake water.

Keywords: Bacteria removal, E. coli, Settling time, Total coliform, Turbidity removal.

I. Introduction It is used as an indicator of water property that is based on

purify and total suspended solids contained in the water [2].
Turbidity is a degree measurement of water clarify [1].
The high amount of these small solid particles that is existed

143
in the water, the greater turbidity of the water will be high. solution was contained 10 mg, therefore, the added 1 %
The familiar pollutant to turbidity is organic matter which is solution to 1 L of water is 10 mg/L.
a major cause of the quality of surface water. Turbidity is 2.2.2. Calcium hypochorite (Ca(OCl)2)
generally source from the small discrete mass of solids in the 0.1 g of calcium hypochlorite granule was suspended in
water which is connected with muddiness and color. 100 mL of distilled water to produce Ca(OCl)2 stock
Turbidity is commonly indicated by using turbidity meter. solution and stirred at room temperature until it
Turbidity is described in the united that called a homogenizes. 1 mL of stock was contained 1 mg. 1 ppm of
Nephelometric Turbidity Unit (NTU) or a Jackson Turbidity free chlorine was obtained by diluted 5 mg (5 mL) of
Unit (JTU) [3]. Ca(OCl)2 stock in 1 L of water (5 mg/L). This value was
Coagulation and flocculation are utilized to remove the obtained by an actual measurement using the N, N-diethyl-
organic matter from the water [4]. Coagulation and ρ-phenylenediamine (DPD) method using the Free/Total
flocculation are taken place in serial steps that permit Chlorine Colorimeter.
particle collision and growth into floc, then followed by 2.3. Experimental procedures
sedimentation. The serial steps are connected, if coagulation 2.3.1. Coagulation and flocculation experiments
is incomplete, the following steps are also incomplete and The lake water sample was priority treated with
will be unsuccessful. coagulants (yellow PAC and white PAC) to find the optimal
Total coliform bacteria refer to a broad span of dose in removing turbidity. High turbidity with 943 NTU of
facultatively anaerobic and aerobic, a large group of Gram- lake water sample was diluted with distilled water to
negative, non-spore-forming bacilli have an ability for obtained new initial turbidity 753 NTU, 558 NTU, and 341
developing in the presence of a comparatively high NTU randomly. 500 mL of the sample of each condition of
concentration of bile salt with the process of fermenting of initial turbidity was filled in 3 beakers and were treated with
lactose and a product of acid/ aldehyde in the limit of 24 two types of PAC products such as yellow and white to
hours at 35-37 °C [5]. Total coliform was presented in nature indicate one suitable product to make the package for lake
such as soil, vegetation, and water. Normally, they are water treatment.
located in fecal-polluted water and usually connected with a 2.3.2. Disinfection experiments
disease outbreak. Escherichia coli is a type of gram-negative 2.3.2.1. Sole using of Ca(OCl)2
bacilli of family Enterobacteriaceae [6]. In most water, the The 500 mL of lake water sample was primarily treated
greater in number genus is Escherichia, and yet some species with 120 mg/L of yellow PAC for 60 mins. 250 mL of water
of Citrobacter, Klebsiella, and Enterobacter are also sample after treatment was taken into a new bottle and
thermotolerant. Escherichia coli can be distinguished from treated with three different doses of Ca(OCl)2. The sample
the other thermotolerant coliforms by the capability to create was stirred for 1 min within 200 rpm and set for the contact
indole from tryptophan or the creation of the enzyme β- time for 60 mins.
glucuronidase [5]. 2.3.2.2. Dual using Ca(OCl)2 with PAC
500 mL of lake water sample was filled in 3 sterilized
II. Materials and Methods bottles and treated with 120 mg/L of yellow PAC and stirred
for 1min within 200rpm and then three different
2.1. Sampling sites
concentrations of Ca(OCl)2 were added to the samples and
Lake water was collected from Boeung Choeung Ek with
continuously stirred for 2 min at the same stirring speed and
coordinate 11.466076, 104.914543 which is located Sangkat
set for the contact time for 60 min.
Choeung Ek, Khan Dangkor, Phnom Penh during the rainy
2.4. Analytical methods
season. The sample was collected in plastic bottles of 1.5 L.
2.4.1. Turbidity measurement
After sampling, the sample was transported immediately to
In each settling time, 10 mL of sample were collected at
ITC laboratory to analyze within a day.
3 cm below the surface level. The turbidity measurement of
2.2. Preparation of stock solution
treated water was measured by using HANNA-HI 98703
2.2.1. Polyaluminum chloride (PAC)
Portable Turbidimeter in three replicates for accuracy. The
1 g of PAC powder was suspended in 99 mL of distilled
turbidity removal efficiency was calculated by using Eq.1.
water to produce a PAC stock solution of 1% and stirred at
room temperature until it homogenizes. 1 mL of 1% stock

144
 (𝑇4 − 𝑇6 ) after treated for 120 min. For each figure, it was clearly
𝑇𝑢𝑟𝑏𝑖𝑑𝑖𝑡𝑦 𝑟𝑒𝑚𝑜𝑣𝑎𝑙(%) = × 100 (Eq.1)
𝑇4 shown that each dose of yellow PAC is more effective than
where Ti and Tf are the initial and final turbidity, respectively the dose of white PAC. Therefore, yellow PAC was chosen
2.4.2. pH measurement due to its high effectiveness in turbidity removal comparing
At the end of the experiment, the sample was taken to to white PAC. Alternately, the comparison of all
measure the change of pH in using yellow and white PAC. concentration dosages of yellow PAC was significant, thus
The measurement was done by using WM-32EP EC/pH the amount of yellow PAC between 60-120 mg/L can
meter. completely remove the turbidity with a range from 300-1000
2.4.3. Microbiological analysis NTU to below 5 NTU.
After each performed treatment, the samples were taken (a) At initial turbidity 943 NTU (b) At initial turbidity 753 NTU
for total coliform and E. coli count. 100 mL of sample was 100 100

filtered by using Whatman® sterile membrane filters without 99 99

Turbidity removal (%)

absorbent pads with diameter 47 mm and pore size 0.45 µm

Turbidity removal (%)

98 98
and was cultured on each Chromocult® Coliform agar plate
which was then incubated at 36 ± 2 ℃ for 21-24 h. 97 97

The number of colony-forming unit (CFU) of 100 mL water 96 96

sample was expressed in the formula:
number of colony counted 95 95
CFU/100mL= ;sample volume filtered in mL< ×100 (Eq.2)
94 94
The bacteria removal efficiency was calculated by : 100 120 150 80 100 120
Baci -Bacf Concentration (mg/L) Concentration (mg/L)
Bacteria removal (%)= ×100 (Eq.3)
Baci
where Baci and Bacf are the initial and final bacteria that (c) At initial turbidity 558 NTU (d) At initial turbidity 341 NTU
were counted. 100 100
2.5. Statistical analysis 99 99
Turbidity removal (%)

All data were subjected to analysis of variance

Turbidity removal (%)

98 (b) At initial turbidity 753 NTU 98
(ANOVA). P-value lower than 0.05 indicates that the data 100

are considered to be statistically significant. 97

Turbidity removal (%)

99 97
98
96 96
97
III. Results and Discussion 95 95
96
The physicochemical characteristics of the lake water 95
94 94
sample such as turbidity, pH, and bacteria (total coliform and 60
94 80 100 40 60 80
Concentration
80 (mg/L) 100 120
Concentration (mg/L)
E. coli) were analyzed. After the measurement, the Concentration (mg/L)
parameter of turbidity was 943 NTU and pH was 7.72, while
Yellow PAC White PAC
the parameter of bacteria was 4.03´103 of total coliform and
2.37´102 of E. coli as illustrating in Table 1.
Figure 1. Effect of yellow PAC and white PAC dosage
on turbidity removal in varies condition of turbidity after
Table 1. Physicochemical characteristics of lake water
treated for 120 mins
Parameter Value
3.2. Effect of PAC on changing pH value
Turbidity (NTU) 943 As a result, in each initial turbidity, the change of
Total coliform (CFU/100ml) 4.03´103 pH in using each dose of yellow PAC was fallen into
E. coli (CFU/100ml) 2.37´102
pH the standard for drinking water 6.5-8.5. In contrast, th
7.72
e change of pH when using each dose of white PAC
was fallen out from standard that was allowed for dri
3.1. Effect of PCA on turbidity removal
nking water except some doses of 100 mg/L, 80 mg/L,
Figure 1 expresses in the effect of yellow PAC and white
PAC on turbidity removal in various conditions of turbidity 60 mg/L, and 40 mg/L. Therefore, using yellow PAC

145
has gradually changed in pH and it did need to adjust IV. Conclusion
pH before or after treatment and it is suitable for the The development of a package containing yellow PAC
small scale of drinking water treatment. and Ca(OCl)2 was made after this study. The package was
3.3. Effect of settling time in turbidity removal forming where the net contents of one package are 12.75g
the percentage of turbidity removal of using yellow and capable for treating 100L of water. Additionally, yellow
PAC and white PAC gradually increased from low to PAC displayed the better performance comparing to white
high dose from 15 min to 120 min of settling time. at PAC in removing turbidity with more than 99% owing to the
the settling time 60-120 min, the average of high re various raw materials and producing methods. Thus, this
moval of turbidity is more than 99% in using yellow made yellow PAC more suitable for drinking water process.
PAC and 98% of using white PAC. Also, the two methods of using Ca(OCl)2 showed a slight
3.4. Effect of Ca(OCl)2 on bacteria removal difference in removing bacteria with more than 98% in
The comparison of both conditions of removing total coliform and 100% of E. coli.
using Ca(OCl)2 was clearly shown that the first method of
using only Ca(OCl)2 is more effective than the second Acknowledgement
method of using Ca(OCl)2 with PAC. However, the second
We are thankful to the Science and Technology Research
method of using Ca(OCl)2 with PAC with a dosage of 7.5
Partnership for Sustainable Development (SATREPS), the
mg/L had the same effect in reducing bacteria to the first
Japan Science and Technology Agency (JST)/Japan
method where 100
99.4% in removing total coliform, and 100%
in removing E. 99
coli. International Cooperation Agency (JICA) for their financial
(a) (b) support.
98
100 100
97
99 References
Bacteria removal (%)

99

96 98
98
97
95 [1] Environmental Protection Agency, 2012. 5.5 Turbidity.
Bacteria removal (%)

Bacteria removal (%)

97
96

96
94
95
InWater:Monitoring&Assessment,https://archive.epa.
95 93 94
gov/ water/archive/web/html/vms55.html (Consulted
92
93 on April 01, 2020).
94
92 [2] Perlman, H., 2014. Turbidity. In the USGS Water
93 91
91 ScienceSchool,http://ky.gov/nrepc/water/ramp/rmtss.h
92 90
Control 3.75 5 7.5
Control 3.75
90
5
Control 3.75 7.5 5 7.5 tm (Consulted on April 01, 2020).
Concentration of Ca(OCl)2Concentration
(mg/L) Concentration of Ca(OCl)2 (mg/L)
of Ca(OCl)2 (mg/L) [3] Fondriest Environmental, Inc, 2014. “Turbidity, Total
Total Coliform E. coli Total Coliform E. Coli
Total Coliform E. Coli Suspended Solids and Water Clarity.” Fundamentals of
Fig. 2. Effect of Ca(OCl)2 on bacteria removal at the EnvironmentalMeasurements,https://www.fondriest.c
contact time for 60 mins:(a). The first method of using only om/environmentalmeasurements/parameters/water-
Ca(OCl)2 and (b).the second method using Ca(OCl)2 with quality/turbidity-total-suspended-solids-water-clarity/
PAC. (Consulted on March 29, 2020).
[4] Prakash, N.B., 2015. Waste Water Treatment by
3.5. The packages of yellow PAC with Ca(OCl)2 Coagulation and Flocculation. International Science
and Innovative Technology (IJESIT), 3, 479.
[5] World Health Organization, 2011. Guidelines for
drinking-water quality-4th ed. WHO Library
Cataloguing-in-Publication Data, 1–564.
[6] Bhavsar, S., and Krilov, L., 2015. Escherichia coli
Infections. Pediatrics in review/American Academy of
Pediatrics, 36, 167–71.

Fig. 3. Package containing yellow PAC and Ca(OCl)2

146
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Assessment of Particle Size Fraction Distribution of Surface Sediment of Tonle Sap Lake,
Cambodia: A Case Study in Chhnok Tru.

Vibol CHEM1, Rina HEU3,4,*, May Phue WAI1 , Sochetra SEN2,

Kimheang THAI2, Khy Eam EANG3 and Sokly SIEV2, 4

1
Graduate School of Water and Environmental Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. B
ox 86, 12156 Phnom Penh, Cambodia
3
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia, Russian Federation Blv
d., P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Water and Environmental Research Unit, Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
*heu.rina@itc.edu.kh

Abstract

Tonle Sap Lake (TSL), located in the center of Camboida, is the largest productive freshwater lake in Southeast Asia
regions. TSL gives a huge benefit to Cambodians, especially for riparian settlements along the lake. However, changes in
hydrology and water depth in the lake appear to be the main factors affecting the variability in the particle size distribution in
sediment and its properties. Thus, this study aims to evaluate the D50 of particle size distribution and understand the
compostioins of surface sediment as an ecological function of the lake. The extensive sampling survey was conducted to
collect samples of surface sediment from 18 sampling sites around Chhnok Tru. The particle size distribution (PSD) was
analyzed by using Laser Diffraction Particle Size Analyzer. A synthetic analysis showed that particle size fraction of surface
sediment was a sensitive and effective mechanism to understand climatic and environmental changes of the lake around
Chhnok Tru. Netherless, changes in PSD between different varve types were relatively small indicated that they have a similar
source for the material deposition. The result also showed that the median fraction (D50) of surface sediment was ranged from
164μm to 694μm consisted of 83.32% sand, 1.22% silt, and 15.46% clay. Therefore, the smaller sediment fractions and
variation may attribute to the location of the flow affected by the current in the turbulence area. Besides, smaller fractions
appear to carry out most of the photosynthesis around Chhnok Tru, while larger fractions are responsible for seasonal trends
in sedimentation. Overall, PSD showed from poor to very poor sorting, and generally sorting deteriorated with increasing
mean grain-size. The output information about particle size distribution is essential for the characterization of mineral
components of biogenic lake sediments leading to understanding the changes in physical and biological trends in the lake
around Chhnok Tru.

Keywords: Assessment, Chhnok Tru, Particle size distribution, Surface sediment, and Tonle Sap Lake

147
I. Introduction II. Materials and Methods
Tole Sap Lake (TSL) is the largest productive lake in 2.1. Sampling sites
South-East Asia in which it covers land areas of The sampling points of the study were taken randomly
approximately 3,000 km2 in the dry season to more than based on geological features around Chhnok Tru of Tonle
15,000 km2 in the rainy season [1]. For the water depth of Sap Lake; Kampong Chhnang province, by applying the
TSL, it is varied from 0.8 to 1 meters of water level in the mapping waypoints of coordinated system. In addition, there
dry season, while in the wet season, the water level is varied were 18 samples were collected on 5th March 2020 during
from 10 to 12 meters [2]. TSL has provided a huge benefit to dry season as shown in Fig. 2.1.
Cambodian people especially for the riparian settlements
along the lake. Unfortunately, scientific information about
its productivities as natural resources is not well understood.
In an aquatic environment, different compounds are
incorporated within or absorbed on mineral matter
depending on the physical, chemical, and biological
processes that may change substantially the sediment texture
[3]
.
Sediments in lakes play an important role in elemental
cycling in the aquatic environment, and they are responsible
for transporting a significant proportion of many nutrients
and contaminants [4]. The sediments also mediate their
uptake, storage, release, and transfer between environmental
compartments. Most sediment on the surface of the bottom
of the lake derives from surface erosion and comprises a
mineral component, arising from the erosion of bedrock, and
an organic component arising during soil-forming processes
included biological and microbiological production and Fig.2.1: Mapping of sampling sites in TSL
decomposition, and the additional organic component may
be added by biological activity within the water body [5]. The grab sampler was used to sample the surface
Besides, the grain-size analysis is useful for more than the sediment of the lake, and the sediment samples were
characterization of the mineral component of biogenic lake transferred into plastic zip-locks then kept in an ice-box to
sediments, and particle size distribution is important for maintain the quality of samples and transported to the
understanding the physical and chemical properties of a laboratory for analysis. Critically, we assumed that the
material [6]. The median of 50% particle size fraction in constant temperature in an ice-box is 4oC which prevents
sediment is the practical understanding of the distribution of bacterial activity that could modify their properties [8].
particle; based on size fraction. On top of that, the
importance of the D50 is to understand the dynamics and 2.2. Laser Diffraction Particle Size Analyzer
other geomorphological behaviors of the lake. Particle size Laser diffraction particle size analyzer (SALD-2300,
and mineralogy are directly related because individual SHMADZU, Japan) was used to measure the size fraction of
minerals tended to form within characteristic size ranges. particles of 18 samples. The particle size was calculated by
Sediments may thus be described in terms of discrete measuring the angle of light scattered by the particles as they
compositional fractions, the overall characteristics of which pass through a laser beam through the dispersion of particles
are due to the variation in the proportions of these fractions, [9]
. The light that is scattered at various angles by different
and the consequent changes in particle size [7]. However, particle sizes was measured by a multi-element detector, and
there is much not known or understood about the productive the magnitude of scattered light intensities was recorded [10].
ecological systems of the aquatic ecosystem of TSL, and Moreover, sediment samples were analyzed for particle size
how they interact between surface sediment and the fractions using the light intensity of cumulative percentages
productive systems.

148
of the ranging size of particles in micrometer. The D50 of by clay of 12.08, while silt was the lowest of 1.88 as shown
particle size fraction were studied to understand 50% in Fig.3.1. Therefore, the higher value of SD the higher error
distribution of particle fractions of samples. of sand, clay and clay, accordingly.

2.3. Analytical methods Mean percentage of particle fraction

The categorization of particle size fraction was based on 100
the diameter of the particle in micrometer (Table 2.1) [5,11]. 90 SD =13.12
Thus, the size fraction was summarized into types of 80
sediment such as sand fraction, silt fraction, and clay 70
fraction which was in ranges of their diameters from mean

Mean(%)
60
of 18 points, based on percentages. 50
40
SD =12.08
Table 2.1: Sediment types based on particle size diameter 30
20
Sediment types Diameter Unit 10 SD =1.88
0
Sand smaller than 2000 μm
Sand Silt Clay
Silt smaller than 63 μm
Clay smaller than 20 μm Fig.3.1: Mean of each particle fraction

Overall, the median of 50% particle size fraction was The analytical measurement of PSD by particle size
studied to understand the distribution of surface sediment analyzer showed the quantitative value of particle size in
around Chhnok Tru, and its critical associations to physical range with its cumulative percentage which is based on light
and biological changes of the lake. intensity as shown in Fig. 3.2. In addition to the PSD curve,
The PSD of sediment samples was found in arithmetic the result showed about D10, D50 and D90 of sediment
means of cumulative percentage of each sampling site. distribution. Thus, the 50% of median fraction (D50) of
Furthermore, they was plotted by means of its cumulative surface sediment was ranged from 164μm to 694μm with its
percentages with the particle diameter in micrometer. For mean of 300μm from each sampling site. This was meant
graphical analysis, Origin Pro-2021 program was used. that the 50% of particle size was sand fraction in which it
showed the biggest diameters among other fractions.
III. Results and Discussion
The sediment types and their mean percentages from 18
sampling points as a summary of surface sediment samples
around Chhnok Tru were shown in Table 3.1.

Table 3.1: Mean composition from 18 sampling sites

Sediment Diameter Percentage (%)

Sand < 2000 μm 83.32
Silt < 63 μm 1.22
Clay < 20 μm 15.46

In Table 3.1, it revealed that the particle size fraction of

sand fraction showed the highest composition in surface
sediment samples of Chhnok Tru with 83.23%, while it was Fig.3.2: PSD curve of surface sediment of TLS
followed by clay fraction and silt fraction of 15. 46% and
1.22%, respectively. Moreover, the standard deviation (SD) Therefore, smaller sediment fractions and variation may
of sand fraction showed in higher value of 13.12 followed attribute to the location of the flow affected by the current in

149
the turbulence area. Besides, smaller fractions appear to Estonian Journal of Ecology, 57(4), 231.
carry out most of the photosynthesis in the lake, while larger https://doi.org/10.3176/eco.2008.4.01
fractions are responsible for seasonal trends in [4] Uk, S., Yoshimura, C., Siev, S., Sophal, T., Yang, H.,
sedimentation. Also, there was a study that showed a similar Oeurng, C., Li, S., & Hul, S. (2018). Tonle Sap Lake:
result that sediments from lake expressed that biological and Current status and important research directions for
chemical content accumulated on the particles from 100 to environmental management. Lakes & Reservoirs:
1000 µm in which there was a tendency associated with Research & Management, 23.
increasing basic content as far as decreasing particle size [12]. https://doi.org/10.1111/lre.12222
[5] Bartram. (1996). (PDF) Water quality monitoring: A
IV. Conclusion practical guide to the design and implementation of
freshwater quality studies and monitoring programmes.
Tonle Sap Lake (TSL) is the most productive lake in
ResearchGate.
South-East Asia. Thus, understanding the pattern of the
https://www.researchgate.net/publication/253953121_
ecosystem of the lake around Chhnok Tur is the crucial
Water_quality_monitoring_a_practical_guide_to_the_
benefit to estimate the interaction between sediment and its
design_and_implementation_of_freshwater_quality_st
potential influents on the physical and biological condition
udies_and_monitoring_programmes
of the lake. According to the result of the study, the surface
[6] Żarczyński, M., Szmańda, J., & Tylmann, W. (2019).
sediment around Chhnok Tru was a sand fraction with the
Grain-Size Distribution and Structural Characteristics
highest composition of 83.23%. In addition, the mean of D50
of Varved Sediments from Lake Żabińskie
of PSD was also a sand fraction which came up with the
(Northeastern Poland). Quaternary, 2(1), 8.
mean of D50 of 300μm. Therefore, it could be concluded that
https://doi.org/10.3390/quat2010008
the smaller sediment fractions, and the variation may
[7] Chapman, D. V., Organization, W. H., Unesco, &
attribute to the location of the flow affected by the current in
Programme, U. N. E. (1996). Water quality assessments:
the turbulence area. Moreover, PSD of surface sediment
A guide to the use of biota, sediments and water in
around Chhnok Tru might strongly interact with
environmental monitoring. London : E & FN Spon.
productivities in the lake as macronutrients for the growth
https://apps.who.int/iris/handle/10665/41850
for higher productivities.
[8] Eadie, B J, and S J Lozano. (1999). Grain Size
Distribution of the Surface Sediments Collected during
Acknowledgment the Lake Michigan Mass Balance and Environmental
We are thankful to the Science and Technology Research Mapping and Assessment Programs. 42.
Partnership for Sustainable Development (SATREPS, Grant [9] Hackley, V. A., Gintautas, L., & Lum, L.-S. (2004).
No: JPMJSA1503), the Japan Science and Technology (PDF) Particle Size Analysis by Laser Diffraction
Agency (JST)/Japan International Cooperation Agency Spectrometry: Application to Cementitious Powders.
(JICA), and French Developmet Agency (AFD), Grant ResearchGate.
No:CKH 1236 02P) for their financial support. https://www.researchgate.net/publication/238669156_
Particle_Size_Analysis_by_Laser_Diffraction_Spectro
References metry_Application_to_Cementitious_Powders
[10] Lawrence, J. (2017). Particle Size and Distribution
[1] Kite, G. (2001). Modelling the Mekong: Analysis. Chemical Engineering.
Hydrological simulation for environmental impact [11] Blair, T. C., & McPherson, J. (1999). Grain-size and
studies. Journal of Hydrology, 253(1), 1–13. textural classification of coarse sedimentary particles.
https://doi.org/10.1016/S0022-1694(01)00396-1 Journal of Sedimentary Research, 69, 6–19.
[2] Somony, T., & Schmidt, U. (2004). Aquatic https://doi.org/10.2110/jsr.69.6
Resources Management: The Tonle Sap Great Lake, [12] Maslennikova, S., Larina, N., & Larin, S. (2012). The
Cambodia. 19. effect of sediment grain size on heavy metal content.
[3] Vaasma, T. (2008). Grain-size analysis of lacustrine Lakes Reservoirs and Ponds, 6, 43–54.
sediments: A comparison of pre-treatment methods.

150
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Occurrence, Transportation, Regulation and Treatment Methods of Organic Pollutants in

Surface Water: A Review on Case of Tonle Sap Lake, Cambodia

Mardi Meas 1, Rina Heu 2,4,*, Laty Ma 1, KhyEam Eang2, and Sokly Siev3,4

1
Graduate School of Water and Environmental Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia, Russian Federation
Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd., P.O.
Box 86, 12156 Phnom Penh, Cambodia
4
Water and Environmental Research Unit, Research and Innovation Centre, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
*heu.rina@itc.edu.kh

Abstract
Tonle Sap Lake (TSL), the largest freshwater lake in Southeast Asia, plays an important role in the economy of
Cambodia. It supports the population around it through fishery productivity, water supply and variety of biodiversity.
Currently, regional residents around TSL purchase clean water for their daily consuming from private water supply facility
that use raw water from TSL. However, there are growing concerns related to degradation of the quality of lake water. Even
though, TSL is invaluable by its ecosystems, lake water was polluted by organic pollutants that have been discharged by
various disposal through agriculture, aquaculture and human activities into the lake, and threaten local people’s drinking
water supply. Hence, this research aims to review on researches of occurrence, source, regulation and treatment methods of
organic pollutants in TSL water. Some organic pollutants including aliphatic hydrocarbons, ether, ketone, phenol,
phthalate, fatty acid ester, other oxygenated compounds, benzene and polycyclic hydrocarbons, nitrogenous, sulfur-
containing, phosphorus-containing, pharmaceuticals and personal care product and pesticide were found in lake’s surface
water because of agriculture, aquaculture and human activities . Among them, plasticizers and pesticides are over allowable
standard for drinking water and also the most toxic compounds. The biological method, adsorption, advanced oxidation
process, membrane filtration, and membrane bioreactor are effective whereas coagulation and flocculation are ineffective to
remove organic pollutant. The result of this study will provide the people around TSL the awareness about present of
pollutants in lake water and its quality. Furthermore, it will be used as a recommendation and reference for TSL residents
about using treatment method for pollutant removal.
Keywords: Organic pollutants, occurrence, transportation, regulation, treatment methods, Tonle Sap lake
I. Introduction enlarges during the flood phases to about 250 km long and
Tonle Sap Lake (TSL) is the largest freshwater water 100 km wide with an area of about 17,500 km2, and the
body in Cambodia (Figure 1), and its watershed extending depth reaches 8-10 m [29]. The climate in TSL region is
over 43% (approximately 85,786 km2) of the national land control by two monsoon periods characterized by two
[17]. Lake’s surface area varies between the wet and the distinct seasons, namely the rainy season (May-October)
dry season [5]. During the dry season, the lake is about 120 and dry season (December-April) [17]; and the water
km long and 35 km wide with an area of about 2500 temperature of lake varies between 28 and 33˚C [22].
km2,and maximum depth of about 3.3 m. Lake surface TSL provides various ecosystem services consider as a
basic environmental key for nurturing the local culture

151
associated with it. It is estimated that more than one million investigated [19, 25]. The studies found chemical
people living on and around TSL. Their living is mainly contaminants in TLS such as aliphatic hydrocarbons, ether,
depended on water resources for domestic use, irrigation ketone, phenol, phthalate, fatty acid ester, other oxygenated
and industry, adjustment of the local climate, fish compounds, benzene and polycyclic hydrocarbons,
production, aquaculture, transportation and tourism [17]. nitrogenous, sulfur-containing, phosphorus-containing,
Lake water supports regional population through water pharmaceuticals and personal care product (PPCPs) and
supply [26]. Even though, TSL is significant by its value pesticide such as insecticides, herbicide (atrazine-i.e.) and
but the lake and its ecosystems are under threat due to disinfectant (Table 1).
anthropogenic activities occurring inside and outside its Table 1: Concentration of compounds of organic
basin. Currently, the lake water was polluted by chemical pollutants in surface water of TSL during rainy season
disposal through agricultural production, aquaculture and and dry season [19, 25]
human activities. Water pollution in TSL threatens the
drinking water supply of local people, and lead to the 2.2. Transportation of Pollutants
explosive development of harmful invasive plant species, The TSL ecosystem has experienced major changes
such as water hyacinth, and impact biodiversity in and over the past few decades as a result of anthropogenic
around the lakes, as well as food quality and finally human
health [41]. Therefore, this study aims to review occurrence Category Rainy Season Dry Season
of organic pollutants in lake water, understanding its Min Max Min Max
sources and transportation, regulation and technologies to (μg/L) (μg/L) (μg/L) (μg/L)
treat pollutants. The result from this review will be used to Aliphatic
0.0003 6.203 - -
hydrocarbons
aware local people living around TSL about the lake water
Benzene 0.0029 0.8703 - -
quality and give recommendation to its resident about
Polycyclic
pollutants treatment method in order to obtain clean water 0.0021 0.3036 - -
hydrocarbons
that ensure safe consuming. Ether 0.181 1.854 - -
Ketone 0.0265 3.830 - -
Phenol 0.001 3.800 - -
Plasticizer
0.020 74.340 - -
(Phthalates)
Fatty acid ester 0.0037 0.7308 - -
Other compounds 0.01 27.100 - -
Nitrogenous 0.0015 5.570 - -
Sulfur 0.0287 1.730 - -
Phosphorus 0.0161 8.423 - -
PPCPs 0.0065 0.4766 - -
Pesticide 0.001 1.0600 0.0510 2.3168
activities both inside and outside of its basin (e.g., water
infrastructure development; land use change, agriculture,
aquaculture) [17]. Also, TSL is surrounded by the most
densely populated region of Cambodia [39] with
Figure 1: Tonle Sap Lake and its floodplains [9] approximately 5 million people living around the lake [43,
2. Occurrence and Transportation of Pollutants 17]. The average annual population growth was 1.4%
between 1998 and 2008 [40, 17]. There is extraordinary
2.1. Occurrence of Pollutants
growth in the city of Siem Reap, the city from which
Degradation of water quality in TSL has been growing
tourists visit Angkor. The annual increases in hotel rooms is
concerns [6], which water quality is indicated as a key
112% between 2003 and 2004 [5]. The city at present has
factor to determine the environmental health and quality of
the ecosystem [6]. To Understand the importance of water no waste treatment facilities, so hotels build their own
quality role of lake, many researchers conducted study on package treatment plants or septic systems, so there is
contaminants occurred in the lake water. Thus, the increasing concern about contamination of the Siem Reap
chemical contamination, which obtained from intensive River which drains to the lake [5]. Deforestation and land
agricultural production, aquaculture and human activities use changes in the TSL basin and its inundation area are
within catchment area in Tonle Sap Lake of Cambodia was substantial. The area affected by deforestation in the TSL

152
basin is equal to 15% of the basin total land area, indicating Table 2: Human-Toxicological maximum permissible risk
a higher sediment load into TSL and its floodplain is level evaluation [1,3,16,30,31,33,34,35,36,37,38,45,48,49]
anticipated because of erosion in its catchments. Then, the
Category Toxicity Effect
sediment could carry the organic pollutant and nutrients
from upland into the lake. Aliphatic 100 Liver, kidney and hematologic
hydrocarbons (μg/kgbw/day) effects
The pollution from agricultural chemicals is also
potentially substantial in the TSL basins [17]. A study in six 4.3 Decreased lymphocyte and
Benzene
(μg/kg bw/day) Group A (human carcinogen)
provinces surrounding TSL indicated a high level of trade
of prohibited pesticides still exists, despite the 1998 Sub- Polycyclic 6.3
Carcinogenicity
hydrocarbons (μg/kg bw/day)
Decree on Standards and Management of Agricultural
Materials [50]. In TSL basins, there are about 67% of Ether 23.6 (μg/day) Increase in liver weight
farmers in the TSL region used pesticides for agricultural Ketone 100 Skin irritation and transient,
activities [32, 17] and the households around TSL also (μg/kg bw/day) hypnotic or sedative effects,
40
applied pesticides, herbicides and fungicides which was Phenol Diarrhea, salivation, urine, and
(μg/kg bw/day) blood and liver effects
dominant in the plains zone [44]. After raining, the runoff
Plasticizer 4 Cancer, or reproductive harm,
from those agriculture and household was connected to the
(Phthalates) (μg/kg bw/day) disrupt the endocrine
rivers and could potentially contaminate Tonle Sap Lake
Nitrogenous 10 Carcinogenicity, neural tube
[19]. It has been found that large communities within (μg/kg/day) defects, and diabetes
floating villages relied mainly on fishing, rice cultivation Gastrointestinal, kidneys, liver,
and trading while their economic opportunity depended on Phosphorus 0.02
cardiovascular system, and
(White) (μg/kg bw/day)
zones they are living, e.g., land-base, water-base or semi- nervous system effect
land-water base [13, 19]. The existence of communities Rat study: tendency of
PPCPs 40 transiently suppressed body
living at floating village on the lake further induced (mg/kg/bw/da) weight and transiently
eutrophication, small-scale-aquaculture, boat transportation decreased motor activity
within the area and solid waste/wastewater contamination 0.1-50
Carcinogenic, oncogenic,
not only from household, but also small factories, e.g., ice- Pesticide genotoxic and teratogenic
(μg/kg bw/day)
effect
making factory or fish processing [19].
3. Toxicity and Standard Regulation of Organic
Table 3: Standard regulation of drinking water
Pollutants in Drinking Water
MIME Others
Most chemicals arising in drinking water are health Pollutants
(μg/L) (μg/L)
concern due to their extended exposure of years, rather Polychlorinated biphenyls 0.5 -
than months [28]. The organic pollutants cause toxic, Trihalomethanes 250 -
teratogenic, mutagenic and carcinogenic effects in aquatic Aliphatic hydrocarbons - -
environment as summarized in Table 2. To ensure a safe Benzene 10 10a
Polycyclic hydrocarbons - 0.1b
drinking water for public health, the Royal Government of
Ether (Methyl terrt-butyl ether) - 13c
Cambodia has established a comprehensive policy on Ketone ( 2-butanone (methyl ethyl
- 16d
National Water Supply and Sanitation and set the Drinking ketone))
Water Quality Standard (DWS) for Cambodia (Table 3) Phenol - 50b
Drinking Water Quality Standards were prepared based on Plasticizer (Phthalates) - 3 -6b-8b
b

Fatty acid ester(methyl ester of

standard of the latest WHO drinking water quality - 42b
fatty acid)
guidelines (2003) [27]. These standards are developed by Nitrogenous (Nitrite) - 100a
an inter-ministerial process initiated by Ministry of Sulfur (sulfate) - 250000a
Industry, Mines and Energy (MIME) and concerned Phosphorus - 0.1b
PPCPs - 0.05a
ministries with support from the World Health
Individual Pesticide 0.2-30 0.1e
Organization (WHO) (Table 3). Total Pesticides - 0.5e

a
: World Health organization(WHO)
b
: Environmental Protection Agency(EPA)
c
: State Water Resource Control Board(SWRCB)
d
: Indiana Department of Environmental Management(IDEM)
e
: European Council Directive(ECD)

153
4. Treatment Methods of Organic Pollutants 5. Conclusion and Future Work Need
Currently, there are some methods to removal those Organic micro-pollutants were found in Tonle Sap Lake
organic pollutants from water such as coagulation and such as aliphatic hydrocarbons, ether, ketone, phenol,
flocculation, biological methods, Chlorination, Activated phthalate (plasticizer), fatty acid ester, other oxygenated
carbon adsorption, ozonation and advanced oxidation compounds; benzene and polycyclic hydrocarbons;
process (AOPs), membrane process and membrane nitrogenous, sulphur-containing, phosphorous-containing,
bioreactor (MBR) (Table 4). Coagulation and flocculation, pharmaceuticals and personal care product (PPCPs) and
biological methods, chlorination are ineffective process for pesticide. Their sources are derived from various disposal
organic micropollutant (OMP) removal. Advanced through agriculture, aquaculture and human activities. Base
treatment processes including adsorption, advanced on study of chemical contaminants, plasticizer (bis (2-
oxidation process, membrane process and membrane ethylhexyl) phthalate) and pesticide (DDT, Mefenoxam,
bioreactors can achieve higher and more consistent OMP Triadimefon and Chloroneb) are over allowable of drinking
removal [12]. water quality standard of EPA. Benzene, polycyclic
Based on our interview with the Chhnok Tru Water hydrocarbons, plasticizer (phthalates), Nitrogenous and
Suppl (CTWS) in November 2020, CTWS is utilizing the pesticide, are significant toxic as their toxicities screen
conventional water treatment method through addition of values are small but could cause human carcinogenicity.
Polyaluminum Chloride or Alumn in coagulation, The biological method, adsorption, AOPs, membrane
flocculation, sedimentation, sand filtration and disinfection filtration, and MBR are effective to remove OMP whereas
by chlorination. Base on this review on OMP removal coagulation and flocculation are ineffective for OMP
methods, this conventional water treatment method removal. Thus, water treatment methods are currently used
(coagulation-flocculation and chlorination) are ineffective at TSL region such as CTWS, is ineffective for OMPs
to remove OMPs. removal. This study will be used as recommendation and
reference for TSL resident to assess the water quality and
Table 4: Treatment methods for OMP removal removal method of OMPs. However, further investigation
[12,15,18,20,46] is necessary to monitor the chemical pollutants during dry
Treatment season.
Advantages Disadvantages
methods
Ineffective MP/OMP removal,
Coagulation Process Acknowledgement
Large amount of sludge,
– simplicity, Low The authors would like to acknowledge financial
Introduction of coagulant salts in the
flocculation cost
aqueous phase support from French Agency for Development –European
Low cost, Union (AFD-EU scholarship) with research grant number
Biological
Significant Slow process, low biodegradation
methods CKH 1236 02P.
removal
Ineffective removal at lower
Significant chemical dose, High dose of References
Chlorination
removal chemical Requirement, toxic by- [1] Ahmed, M., Rauf, M., Mukhtar, Z., & Saeed, N. A.
product (2017). Excessive use of nitrogenous fertilizers:
Relatively high financial costs, An unawareness causing serious threats to
Lower efficiency in the presence
Effective environment and human health. Environmental
Adsorption of natural organic matters (NOMs),
removal
Need for regeneration, Disposal of
Science and Pollution Research, 24(35), 26983–
used adsorbent 26987. https://doi.org/10.1007/s11356-017-0589-7
Effective High energy consumption, [2] Bonheur, N., & Lane, B. D. (2002). Natural resources
Advanced
removal, High Formation of by products, management for human security in Cambodia’s
Oxidation Tonle Sap Biosphere Reserve. Environmental
degradation Interference of radical scavengers
Removal Science & Policy, 5(1), 33–41.
Membrane
efficiency High energy demand, Membrane https://doi.org/10.1016/ S1462-9011(02)00024-2
processes
process fouling, Disposal of concentrate [3] Burdock, G. A., & Ford, R. A. (1992). Safety evaluation
Effective
Moderately high energy of dibenzyl ether. Food and Chemical Toxicology,
consumption, Inconsistent removal 30(7), 559–566. https://doi.org/10.1016/0278-
Membrane removal, Less
of polar and resistant compounds, 6915(92)90189-R
Bioreactor sludge, Small
Membrane fouling, Less sorption of
scale
on the aged MBR sludge

154
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Bioaccumulation of heavy metals and trace elements in six fish species from Tonle Sap
Lake, Cambodia

Sokneang IN 1, 2*, Sovannmony NGET 1, 2, Soukim HENG 1,2, Dung Viet PHAM3, MasateruNISHIYAMA3,
Hasika MITH 1, 2 and Toru WATANABE 3

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of Cambodia
, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Faculty of Agriculture, Yamagata University, Japan
* in@itc.edu.kh

Abstract

This study aimed to observe the accumulation of heavy metals and trace elements in six fish species sampled from four
locations in Tonle Sap Lake. The amount of three kilograms of each fish sample was randomly collected from fishermen at
ChhnokTrou commune (Kampong Chhnang Province), Phat Sanday commune (Kampong Thom Province), Kbal Tor
commune (Battambang Province) and Kampong Plouk commune (Siemreap Province) between 2017 and 2018. The level of
Zn, Pb, Cr, Cu, Cd and As in the fish samples were below the Maximum Permissible Limit (MPL) set by FAO/WHO. The
contents of Mn and Ni in the six studied fish species ranged from 0.8 to 10.1 mg/kg in wet weight and 0.18 to 0.79 mg/kg,
respectively. The mean concentration of Mn and Ni were higher than the respective MPL. The highest concentration of Mn
was found in Anabas testudineus (10.1 mg/kg), and the highest concentration of Ni was found in A. testudineus (0.79
mg/kg). Among studied fish species, A. testudineus accumulated all of heavy metals and other trace elements most
effectively. A. testudineus is the predatory fish (carnivorous fish) and this may be resulted from the bioaccumulation.

Keywords: Fish species, accumulation, trace elements and heavy metals

I. Introduction water [3, 4]. In recent year, the increasing of pollution is

risen which caused by agricultural, domestic, industrials
Tonle Sap Lake, the largest freshwater lake in Southeast andtechnological applications. HMs sources in
Asia, is surrounded by five provinces: Kampong Thom, theenvironment are included agriculture waste,
Siem Reap, Battambang, Pusat and KampongChhnangand pharmaceutical, domestic effluents, industrial activities [5,
its surface area stretch to 2500 – 3000 km2 during dry 6]. Moreover, heavy metals penetrate into the water
season and 10 000 – 16 000 km2 during rainy season [1]. reservoirs via atmosphere, drainage of the waste, soil
Due to the immense size of Tonle Sap Lake, this lake is the erosion and human activities [7, 8]. Those polluted of
main supply of local fish consumption and exported heavy metals was contaminated to the live in the water,
fishproduct, which is the fifth largest freshwater fish especially fishes. The quality of water was contaminated by
capturing in the world after China, India and Bangladesh heavy metals and continued accumulate in organisms,
and Myanmar [2]. which are consumed by fish or penetrate into fish directly
Heavy metals (HMs) are the element, which have a high through skin and gill later [8].
atomic weight and a density at least 5 times greater than Due to the increasing of development of area surrounding

155
Tonle Sap Lake and poor wastewater management these is was digested 20 ml with hydrochloric acid concentrate
raising the concern of heavy metal presence in Tonle Sap (1mol/L) and heated at 150oC, and then with 20 ml of nitric
lake ecosystem. Moreover, there are more than 900 000 acid concentrate (1mol/L) and heated at 170 oC,. After
people live on floating village.Most of the people, who live finished digestion, the final volume was adjusted to 100ml
on the floating village, are lacking waste management, so with ultra-pure water for analysis (Ministry Environment of
they throw their daily wastes including plastic and also Japan, 2001) with inductively Coupled Plasma-Mass (ICP-
human waste directly into the lake, in order to make the MS) (Model Elan DRCII of PerkinElmer Japan Co., Ltd.,
water and soil pollution [9]. These activities also raise even Japan).
more concern about the heavy metal contamination in the
fish and aquatic live in the lake. Consequently, fish in
Tonle Sap Lakemight have high chance of accumulation of As analysis was adopted from Ministry Environment of
heavy metals by the absorption of contaminated water. Japan, 2001[10], 1g of sample was added with 10ml o
Therefore, the heavy metals effect on human health after HNO3 and heated at 100oC-150oC) covered by cap, and
consumption of the fishes, which contain the heavy metals. then 5ml of concentrated H2SO4 (98%) and 5ml of
The main objective of this study is to quantify several concentrated HNO3 (65%) were added to solution with and
heavy metals such as As, Fe, Cu, Hg, Pb, Zn, Cr, Mn, Co, heated at 200oC. The solution was filtered through filter
Ni, and Cd in the muscle of six fish species, were collected paper ADVANTEC 5B (4µm) into 100ml flask. The
from Chhnuk Tru village (Kampong Chhnang), Phat concentration of arsenic and mercury in the solution was
Sandai (Kompong Thom), Kampongs Phlork (Siem Reap) analyzed by using atomic absorption spectroscopy (AAS )
and KbalTorl (Battambang). machine Model AA7000 with hydride vapor generator
HVG-1 (Shimadzu, Kyoto, Japan).
II. Materials and Methods

2.1. Fish Sampling 2.3 Statistical analysis

Six fish species in Tonle Sap lake were randomly collected
The results were statistically analyzed using the SPSS 20.0
from fishermen at Chhnok Trou commune (Kampong
package (IBM Corp., Armonk, NY, USA). ANOVA
Chhnang Province), Phat Sanday commune Kampong
followed by Tukey’s test, correlation analysis, and
Thom, Kbal tor commune Battambang and Kampong plouk multivariate analysis was also employed. Results of
commune Siemreap Province (see in Figure 1 and Table 1) principle component analysis (PCA) were visualized using
between September 2017 to December 2018. 3 kg of each
RStudio.
sample contained variable numbers of fish of uniform size
were bought and coolimmediately; and transferred in
III. Results and Discussion
icebox to laboratory of Chemical and Food Engineering
Faculty, of Institute of Technology of Cambodia. Uniform
Table 1. Trace element (loid) content (in wet weight) of
fish size was divided into three categories big, medium and
fish species from Tonle Sap Lake (unit mg/kg)
small size and then weight and length of these fishes were
recorded. When arrived at laboratory, fish samples were Scientific name Zn Pb Cr Cu
frozen at about -20oC and stored until analyses.
Boesemaniamicrolepis 14.32 0.02 0.12 0.22
2.2. Sample preparation and Analytical methods Channamicropeltes 7.77 0.06 0.07 0.17
Hampaladispar 11.53 0.05 0.04 0.30
Samples of raw fish were cleaned with tap water before the
Anabas testudineus 20.27 0.06 0.16 0.44
edible part and non-edible part were separated by using the
knife, and were washed in distilled water for further Channa striata 10.86 0.10 0.08 0.22
process. Fish muscles were freeze dried at -50oC, pressure Notopterusnotopterus 6.68 0.07 0.08 0.21
0.04mbar for 24h in freeze dryer machine model ALPHA * a a b
MPL 100 1 2 20a
1-2 LD plus. After freeze-dry, the sample was broken into
* MPL: Maximum Permissible Limit
small pieces, and then passed through a nylon sieve (0.5 a Thailand. Ministry of Public Health. Notification of Ministry of Public
mm) and storage at -20oC for further analysis. Health, No. 98, B.E. 2529, Re: Prescribing Standards of Contaminated
Substances. R. Thai Gov. Gazette 1986, 103, 98. (In Thai)
The digestion of fish sample for trace elements and heavy b Food and Agricultural organization (FAO). Compilation of Legal Limits
for Hazardous Substances in Fish and Fishery Products; Fisheries
metals (except As) analysis was adapted from Ministry circular No. 764; Food and Agricultural Organization: Rome, Italy,
Environment of Japan, 2001 [10].2 g of freeze dried sample 1983.http://www.fao.org/3/q5114e/q5114e.pdf (accessed on 7 July 2020).

156
 Chhnang, Kampong Thom and Siemreap was presented in
Table 3 and 4. The results showed that the accumulation of
Table 2. Heavy metal(loid) content (in wet weight) of trace elements and heavy metal was varied among the
fish species from Tonle Sap Lake (unit mg/kg) studied location. However, the distribution of concentration
of different trace elements and heavy metal, especially Mn
Scientific name Cd As Mn Ni and Ni were highly accumulated in fish from
* KampongThom Province and Battambang province.
Boesemaniamicrolepis ND 0.006 4.07 0.36
Channamicropeltes ND 0.005 1.19 0.41 Table 3. Trace elements(loid) content in fish muscle in
Hampaladispar 0.004 0.003 0.85 0.18 difference location
Anabas testudineus 0.006 0.017 10.11 0.79
Location Zn Pb Cr Cu
Channa striata 0.006 0.002 1.10 0.22
Battambang 10.91 0.02 0.07 0.19
Notopterusnotopterus ND 0.009 0.80 0.21
MPL** 0.5a 1b 0.5b 0.05c Kampong Chnag 7.68 0.10 0.07 0.24
*ND: Not Detected; **MPL: Maximum Permissible Limit Kampong Thom 20.89 0.06 0.17 0.43
a Thailand. Ministry of Public Health. Notification of Ministry of Public Siemreap 8.54 0.07 0.06 0.19
Health, No. 98, B.E. 2529, Re: Prescribing Standards of Contaminated
Substances. R. Thai Gov. Gazette 1986, 103, 98. (In Thai)
b Food and Agricultural organization (FAO). Compilation of Legal Limits Table 4. Heavy metal (loid) content in fish muscle in
for Hazardous Substances in Fish and Fishery Products; Fisheries difference location
circular No. 764; Food and Agricultural Organization: Rome, Italy,
1983.http://www.fao.org/3/q5114e/q5114e.pdf (accessed on 7 July 2020).
c U.S. Department of Health and Human Services. ATSDR Toxicology Location Cd As Mn Ni
Profile for Nickel. 2005. Available online:
http://www.atsdr.cdc.gov/toxprofiles/tp15.pdf (accessed on 7 July 2020). Battambang ND* 0.005 1.87 0.33
Kampong Chnag 0.006 0.013 1.32 0.29
The mean trace element and heavy metal concentrations in
muscle of all six fish species was presented in Table 1 and Kampong Thom ND 0.010 7.17 0.52
2. The concentrations of different trace elements and heavy Siemreap 0.004 0.010 1.80 0.30
metal varied within species and location. The level of Zn, *ND: Not Detected
Pb, Cr, Cu, Cd and As were below the Maximum
Permissible Limit (MPL). However, the contents of Mn Pb level was higher in Trey Ptouk origin from Siemreap
and Ni ranged from 0.8 to 10.11 mg/Kg in wet weight and and Kampong Chnag province. Cr was detected
0.18 to 0.79 mg/Kg wet weight, respectively. The mean higherconcentration in Trey Bromaand Trey Kranh from
concentrations of Mn and Ni were higher than the KampongThom province. Cu was more accumulating in
Maximum Permissible Limit. The highest concentration of Trey Kranh and Trey Khman from Kampong Thom
Mn and was found in Anabas testudineus (10.11 mg/Kg province. Cad was mostly ND in the studied fish samples.
wet weight) follow by Boesemaniamicrolepis (4.07 mg/Kg The level of As was found higher concentration in Trey
wet weight), and the highest concentration of Ni was found Kranh from Kampong Chnang, Kampong Thom and
in Anabas testudineus (0.79 mg/Kg wet weight)follow by Siemreap, and in Trey Slat from Battambang. Mn was
Channamicropeltes (0.41 mg/Kg wet weight) and highly accumulated in Trey Kranh from Kampong Thom.
Boesemaniamicrolepis (0.36 mg/Kg wet weight). According to many studies in literatures, trace elements
and heavy metal bioaccumulation by fish can influence by
The distribution of concentration of different trace and living environment (geographical location,
elements and heavy metal, especially Mn and Ni were contamination sources in water), fish type (predator or non
highly accumulated in fish from Kampong Thom Province, predator), sex, age, size, reproductive cycle, swimming
follow by Battambang province. Trace element and patterns, feeding behavior (demersal fish, neritic or pelagic
heavy metals bioaccumulations were affected by the fish) [13,14,15,16].
aquatic ecosystem environmental conditions such as
industrial, municipal and agricultural wastes in the water
IV. Conclusion
[11,12].
Trace elements content varies depending on the zone,
environmental conditions, the contamination level of the
The accumulation of individual trace elements in fish (all
fishing area, and the characteristics of the fish (size,
species combined) collected Battambang, Kampong

157
endurance, and diet) constitute relevant factors, being some Kumaresan S, Balasubramanian T., 2012. Metals in
fish more prone to accumulating higher concentrations of fish along the southeast coast of India. Bull Environ
these metals in the muscle. Contam Toxicol; 88(4):582–8.
[12] Nicole Colin, Alberto Maceda-Veiga, Núria Flor-
Acknowledgement Arnau, Josep Mora, Pau Foruno, Cristiana Vieira, Narci
s Prat, Jaume Cambra and Adolfode Sostoa., 2016. Eco
We are thankful to the Science and Technology logical impact and recovery of a Mediterranean river af
Research Partnership for Sustainable Development ter receiving the effluent from a textile dyeing industry.
(SATREPS), the Japan Science and Technology Agency
Ecotoxicology and Environmental Safety, Volume
(JST)/Japan International Cooperation Agency (JICA) for
their financial support. 132, Pages 295-303.
[13] Zhao S, Feng C, Quan W, Chen X, Niu J, Shen Z.,
References 2012. Role of living environments in the
[1] FAO., 2004. Flooded Forest and Fishing Villages. pdf accumulation characteristics of heavy metals in fishes
(1st ed.; P. T. Evan, M. Marschke, & K. Paudyal, and crabs in the Yangtze River Estuary, China. Mar
Eds.). Siem reap: Asia forest network. Pollut Bull, 64:1163e71.
[2] Baran, E., & Gallego, G., 2015. Cambodia’s fisheries: [14] Mustafa C, Guluzar A., 2003. The relationships
a decade of changes and evolution. Catch and Culture, between heavy metal (Cd, Cr, Cu, Fe, Pb, Zn) levels
Vol. 21, pp. 28–31. and the size of six Mediterranean fish species.
[3] Tchounwou, P. B., Yedjou, C. G., Patlolla, A. K., & Environ Pollut, 121:129e36.
Sutton, D. J., 2012. Heavy Metals Toxicity and the [15] M Al-Busaidi, P Yesudhason, S Al-Mughairi, W A K
Environment. EXS, 101, 133–164. Al-Rahbi, K S Al-Harthy, N A Al-Mazrooei, S H Al-
[4] Zhenli He, Xiaoe E Yang, Peter J Stoffella., 2005. Habsi., 2011. Toxic metals in commercial marine
Trace Eelements in Agroecosystems and Impacts on fish in Oman with reference to national and
the Environment. Journal of Trace Elements in internationalstandards. Chemosphere, 85(1):67e73.
Medicine and Biology 19(2-3):125-40. [16] Khaled A., 2004. Seasonal determination of some
[5] Kh. M. El-MoselhyKh. M. El-Moselhy a, A.I. heavy metals in muscle tissues of Siganusrivulatus
Othman b, H. Abd El-Azem a, M.E.A. El-Metwally., and Sargussargus fish from El-Mex Bay and
2014. Bioaccumulation of heavy metals in some Eastern Harbor, Alexandria, Egypt. Egypt J Aquat
tissues of fish in the Red Sea, Egypt. egyptian journal Biol Fish, 8(1):65e81.
of basic and applied sciences, 97-105.
[6] Jan AT, Azam M, Siddiqui K, Ali A, Choi I, Haq QM.,
2015. Heavy Metals and Human Health: Mechanistic
Insight into Toxicity and Counter Defense System of
Antioxidants. Int J Mol Sci;16(12):29592-29630.
[7] P. Part, O. Svanberg and A. Kiessling, “The
Availability of Cadmium to Perfused Rainbow
Troutgills in Different Water Qualities. Water
Research, Vol. 19, No. 2, 1985, pp. 427-434.
[8] B. Staniskiene, P. Matusevicius1, R. Budreckiene ,
K.A. Skibniewska., 2006. Distribution of Heavy
Metals in Tissues of Freshwater Fish in Lithuania.
Polish J. of Environ. Stud. Vol. 15, No. 4, 585-591
[9] Johnstone Gareth, PuskurRanjitha, Declerck Fabrice,
Mam Kosal, Il Oeur (ADIC), MakSithirith,
PechBunna, SeakSophat, CHAN Sokheng,
HakSochanny, Lov Samnang, SuonSokheng,
ProumKimhor, Rest Sameth., 2013. Tonle Sap
scoping report. CGIAR Research Program on Aquatic
Agricultural Systems. Penang, Malaysia.Project
Report: AAS-2013-28.
[10] Ministry of Japan, T.E (2001). Methode of sediment
quality.
[11] Thiyagarajan D, Dhaneesh K V, Kumar TTA,

158
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Difference in Basic Water Quality Before and After Moving Floating Houses to Upland in
Chhnok Tru Village, Tonle Sap Lake, Cambodia

May Phue WAI1, Rina HEU3,4, *,Vibol CHEM1, Sochetra SEN2,

Kimheang THAI2, Khy Eam EANG3, Sokly SIEV2,4

1
Graduate School of Water and Environmental Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd., P.O.
Box 86, 12156 Phnom Penh, Cambodia
3
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia, Russian Federation
Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Water and Environmental Research Unit, Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
*Corresponding author: heu.rina@itc.edu.kh

Abstract

The ecosystem of Tonle Sap Lake (TSL) is widely nominated as a biosphere reserve by UNESCO in October 1997.
Approximately five million people live around TSL; the discharges of sewages and wastes from the households into the lake
also concern with the water quality of the lake. Government declared to move floating houses in Chhnok Tru to upland and
some of the floating houses in Chhnok Tru had been already moved according to Government’s plan. Thus, this paper aims
to investigate water quality parameters such as temperature, pH, dissolved oxygen (DO), oxidation-reduction potential (ORP),
conductivity, chlorophyll before and after moving floating houses Chhnok Tru. The water quality parameters of TSL were
measured in 19 sampling sites around Chhnok Tru by using EXO sensor. The result showed that, before moving floating
houses, the average temperature was 32.24 °C, pH was 6.96, conductivity was 166.9 µS/cm, salinity was 0.0673 psu (Practical
Salinity Unit), DO was 4.33 mg/l, ORP was 135.5 mV and total dissolved solid was 93.92 mg/l. After moving floating houses,
the average temperature was 30.9 °C, pH was 8.69, DO was 5.24 mg/l, conductivity was 203.67 µS/cm, chlorophyll was 4.39
and ORP was 81.5 mV. The average pH and DO after moving floating houses were higher than those before moving floating
houses because of the less discharging of wastes from floating houses, while the average ORP was lower than that before
moving floating houses because one sampling site that is close to the Chhnok Tru port had a negative ORP value showing
that only high contamination in this site. The output from this study is expected to provide the information about water quality
of TSL before and after moving floating houses for the support of water quality management to sustain the TSL ecosystem.
Keywords: Ecosystem, Floating houses, Tonle Sap Lake, Water quality

159
I. Introduction 2.2. Water quality measurement
Water quality analysis in both 2018 and 2020 was
Tonle Sap Lake (TSL) is the socio-economic
performed by using Multi- Exo Sondes (YSI Incorporated,
dependency of the Cambodian communities and an
USA). In situ measurement of water quality parameters such
important reservoir for Mekong system [1][2]. Approximately
as temperature, pH, dissolved oxygen (DO), oxidation-
1.7 million people in six provinces around the lake rely
reduction potential (ORP), conductivity (Cond), were
mainly on water for drinking and domestic uses. In this
measured. The Multi-EXO Sondes was set to 5 minutes and
regard, Chhnok Tru is one of the busiest communities, the
repeated at least 10 times for every measurement and it was
discharging of sewage from households causes water quality
washed with sampled water before running to prevent
degradation [3]. According to government plan, floating
contamination from one sample to another.
houses in Chhnok Tru village had been already removed. As
2.3 Inverse Distance Weighted (IDW) Method
a consequence, the comparison and correlation between
A simple interpolation method, IDW method has been
basic water quality parameters before and after moving
applied in environmental modelling. In this study, IDW was
floating houses still need to understand for knowing the
used to interpolate water quality parameters distribution in
effect of moving houses in Chhnok Tru village. Thus, this
2018 and 2020 by ArcMap (10.5).
study aims to evaluate water quality after moving floating
houses in Chhnok Tru village by comparing them with the 2.4. Statistical analysis
water quality before moving the floating houses in Chhnok
2.4.1 Student’s t-test
Tru village. Besides, the Student's t-test, Pearson’s
T-test for independent samples is used for comparison of
correlation test, and Principal component analysis (PCA)
mean values of results. Risk possibility level of 5% was
were used for statistical analysis. In addition, inverse
taken into account in defining the statistical significance of
distance weighted (IDW) interpolation was applied for the
obtained results, t value must be at least 1.96. In this study,
distribution of water quality in both 2018 and 2020.
a t-test was performed with MS excel 2019 to test the
II. Materials and Methods significance between before moving floating houses and
after moving floating houses for the water quality
2.1. Sampling sites parameters.
This study was conducted in the Chhnok Tru village, 2.4.2 Correlation analysis
TSL. Water samples were collected in 19 sites of the Chhnok Correlation is the way of assessing to find the strength of
Tru village on both 7th March 2018 and 5th March 2020. The the association. The Pearson product-moment correlation
floating houses distribution in 2018 was approximately 1881 coefficient is the most widely used in testing correlation
houses from one study of Chhnok Tru village [4]. Based on between two variables. The correlation coefficient is
government plan, the floating houses were supposed to expressed by r that is ranged from -1 to +1. In this study,
move to upland about 90% by 2020. The sample location Pearson’s correlation test was performed by R program
map is shown in Fig. 1. (3.6.1).
2.4.3 Principal component analysis (PCA)
PCA is used to identify the most important components
among the large variable numbers in the entire original data
set. The outputs of biplot were interpreted to be a better
understanding of the water quality in the sampling locations
for 2018 and 2020. In this study, PCA was performed using
Origin pro-2021.

III. Results and Discussion

From the graph of Fig 2, the results revealed that
temperature was slightly decreased in 2020. For pH, the
values in pH were increased in 2020 with DO was also
Fig. 1. Map of sampling locations
higher than in 2018. Conductivity (Cond) in 2020 was

160
higher than in 2018. Meanwhile, the oxidation-reduction
potential (ORP) was so much different in both 2018 and
2020.
The comparative graph of water quality in both 2018 and
2020 is shown in Fig 2.

Temp(°C) pH DO(mg/l)
Cond(µS/cm) ORP(mV)
400.0

300.0
187.6
200.0 166.9
135.5
100.0
32.2 30.5 20.8
7.0 4.9 8.6 5.2
0.0 Fig. 4. Water quality distribution map in 2020
2018 2020
YEAR
The independent t-test was performed and the results
Fig. 2. Water quality in 2018 and 2020 were shown in Table 1. According to results of t-test, as
shown in Table 1, values of temperature were statistically
According to water quality distribution maps shown in significant at significance level p<0.05, confirming that the
Fig 3 and Fig 4, minimum DO value in 2018 is 0.6 while in values of temperature were lower in 2020. Besides, pH
2020, the minimum DO value is 0.5. Maximum pH value in values were higher in 2020 with the significance level
2020 has become higher than in 2018. In 2018, conductivity p<0.05. As far as DO, Conductivity, and ORP were
distribution in both 2018 and 2020 is a similar distribution. concerned, they were not statistically significant by the
But for ORP values, in 2018, most sites show low ORP results of t-test.
values while in 2020, only some parts have low ORP values.
Table 1: t-test analysis for water quality parameters in
2018 and 2020

Parameters Year Mean T-test p-value

Temperature 2018 32.24
4.45 <0.001*
(o C) 2020 30.51
2018 6.96
pH -10.40
2020 8.59 <0.001*
DO 2018 4.90
-0.44 0.662
(mg/l) 2020 5.17
Conductivity 2018 166.91
-0.34 0.732
(µS/cm) 2020 187.62
ORP 2018 135.53
1.73 0.1016
(mV) 2020 20.82

Correlation analysis was performed for each specific

Fig. 3. Water quality distribution map in 2018 parameter between 2018 and 2020 as shown in Table 2, the
results showed that temperature values in both 2018 and
2020 were not correlated with each other (r=0.1, p>0.05).
but for pH values in both 2018 and 2020 were positively
correlated (r=0.5, p<0.05) while DO values in both 2018 and
2019 were positively correlated with r=0.66 and p<0.01.

161
Meanwhile, conductivity was strongly correlated in both JS14, JS15. This showed that these sites were mainly
2018 and 2020. But for ORP values in both 2018 and 2020, controlled by water quality parameters in 2018. From the
these values were not correlated with each other (r= -0.1, results of biplot, as shown in Fig 6, temp, Chl and Cond had
p>0.05). a positive score with PC1 on the sites of JS1, JS3, JS18 while
temp, DO, pH Chl had the positive score with PC2 on the
Table 2: Correlation analysis for water quality sites of JS3, JS5, JS4, JS6, JS7, JS8, JS15.
parameters in 2018 and 2020
IV. Conclusion
Parameters Year r p-value
From the results of this study, t-test revealed that
Temperature 2018
0.0905 0.721 temperature and pH were significantly different between
(o C) 2020
2018 and 2020, and the correlation of pH, DO Conductivity,
2018 and ORP between 2018 and 2020 were also found out. By
pH 0.5437 0.01972*
2020
biplot results, in 2018, JS3, JS4, JS5, JS6, JS7, JS9, JS12,
DO 2018
0.6638 0.00267* JS14, and JS15 were the main influenced sites by the water
(mg/l) 2020
quality parameters. JS1, JS3, JS4, JS5, JS6, JS7, JS8, JS15,
Conductivity 2018
(µS/cm) 2020
0.8805 <0.001* and JS18 were the main influenced sites by water quality
ORP 2018 parameters in 2020. Moreover, the distribution of each water
-0.1805 0.488 quality in both 2018 and 2020 can be seen clearly by IDW
(mV) 2020
method showing that some parts of study area of Chhnok Tru
village still need to consider for lake ecosystem management.

Acknowledgment
We are thankful to the Science and Technology Research
Partnership for Sustainable Development (SATREPS, Grant
No: JPMJSA1503), the Japan Science and Technology
Agency (JST)/Japan International Cooperation Agency
(JICA), and French Development Agency (AFD, Grant No:
CKH 1236 02P) for their financial support.
References
Fig.5. Biplot of water samples in 2018
[1] Campbell, I. C., Say, S., Beardall, J. (2009). Tonle Sap Lake, the
Heart of the Lower Mekong. The Mekong, 251–272.
https://doi.org/10.1016/B978-0-12-374026-7.00010-3
[2] Sarkkula, J., Kiirikki, M., Koponen, J., & Kummu, M. (2003).
Ecosystem processes of the Tonle Sap Lake. The 1st Workshop of
Ecotone Phase II, 1–14. https://doi.org/10.2307/171246
[3] Shivakoti, B. R., Bao, P. N. (2020). Environmental Changes in
Tonle Sap Lake and its Floodplain: Status and Policy
Recommendations. In Institute for Global Environmental
Strategies, Tokyo Institute of Technology, Institute of Technology
of Cambodia. https://www.iges.or.jp/en/pub/tonlesapsatreps/en
[4] Chem, V., Hieav, S., Kim, C.(2019). Assessment of Seasonal
Fig. 6. Biplot of water samples in 2020 Nutrient Loads from Floating Community of Chhnok Tru at Tonle
Sap Lake, Cambodia. A Senior Project Report Presented to the
Academic Member of the Faculty of Mathematics, Science, and
According to the results of the biplot, as shown in Fig 5, Engineering (Bachelor’s degree thesis). Paññāsāstra University of
in 2018, Cond, temp, DO, pH and ORP had a positive score Cambodia
with PC1 on the sites of JS3, JS4, JS5, JS6, JS7, JS9, JS12,

162
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

The effectiveness of different types of polyaluminum chloride (PAC) and aluminum

sulfate (alum) with Ca(OCl)2 dosing for treatment surface water of Tonle Sap River

Theara YANN 1, Kazuhiko MIYANAGA2 and Reasmey TAN 1,3,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
School of Life Science and Technology, Tokyo Institute of Technology, 4259 J 3-8
Nagatsuta-cho, Midori-ku, Yokohama, 226-8501, Japan
3
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: rtan@itc.edu.kh

Abstract

As a treatment of water, coagulation of minute particles and disinfection of microbes are the most commonly used. The
aim of this study is to optimize the appropriate condition of two different kinds (yellow and brown) of polyaluminum chloride
(PAC), and alum with Ca(OCl)2 for the treatment of surface water of Tonle Sap River. The optimum dosage of coagulants was
obtained based on residual turbidity and pH change. The optimum dosage of Ca(OCl) 2 was obtained based on efficiency of
removal bacterial cells and residual free chlorine. The series experiment was conducted to study the impact of dosing of
(yellow and brown) polyaluminum chloride and alum with Ca(OCl)2 on Tonle Sap River within high turbidity, The result of
experiment showed that the highest efficiency of turbidity removal within 97.1%, 95.6% and 95.6% of 60 mg/L of yellow
PAC, brown PAC and alum respectively, while pH was changed from 7.45 to 7.08, 6.88 and 6.77 respectively. As a result of
using those optimum coagulant with Ca(OCl)2, 100% of E.coli removal was obtained at 1.0ppm and 1.7ppm of free chlorine
dosing, while the residual of free chlorine was range from 0.21-0.35ppm. Therefore, the amount of those coagulants and free
chlorine are good for water treatment to get high quality by following the standard Ministry of Industry Mines and Energy,
and World Health Organization guideline.

Keywords: Alum, pH change, polyaluminum chloride (PAC), turbidity removal, residual of free chlorine.

I. Introduction fractions of natural organic matter (NOM) contained in

surface water are humic acid and fulvic acid, hydrophobic,
Nowadays, due to many factors such as population
hydrophilic, carbohydrate, carboxylic acid, an amino acid.
growth, domestic wastes, agriculture wastes, sewage, and
The characteristic and amount of NOM depend on the
urbanization, the water source was contaminated by
origins of the water. It comes from the degradation of plant
inorganic and organic matter. Livelihood surrounding Tonle
and animal, bacterial, algae [2, 3]. NOM is not only the
Sap River and lake confront with serious health risks due to
cause taste, odor, color and bacterial growth, but also result
use this water source, especially people who is living in
in the formation of disinfection by-products during
floating communities on Tonle Sap River and lake [1]. The
disinfection process [4]. Reducing the cloudiness and

163
particle substance as well as turbidity was needed before below 7.5 and is a more disinfectant, while hypochlorite ions
disinfection process [5]. are more effective at pH above 7.5 [16]. Aluminum
Turbidity is one of the important physical parameters of coagulant are aluminum sulfate (Alum), polyaluminum
environmental water which presenting the amount of chloride (PAC), poly aluminum silicate sulfate (PASS),
particle suspended and colloidal matters such as clay, silt, polyaluminum silicate sulfate chloride (PASC) are widely
finely divided organic and inorganic, and other used in water treatment, using these coagulant are associated
microorganisms [6, 7]. These particles can scatter the light with series problem due to over amount of residual
creating muddily and cloudy in water [8]. Colloidal particles aluminum cause by overdose of coagulant [17]. The
are normally carry negative charge on their surface which different color of polyaluminum chloride have different in
lead to the stabilization of the suspension [9]. Turbidity itself application and production technology. Yellow
does not represent a direct risk to public health, nevertheless, polyaluminum chloride (Y. PAC) was mainly used for
it indicates the possibilities to include hazardous chemicals sewage disposal and drinking water treatment. Brown
and/or pathogenic microorganisms. The high turbidity leads polyaluminum chloride (B. PAC) was used for waste water.
to impair the disinfection of water which acts as a barrier The characteristic of yellow and brown PAC waw showed
between disinfectant and microbial, and the cause of in Table 1. The main objective of this study is to optimize
forming the disinfection by-products which may have long- the treatment condition of yellow PAC, Brown PAC and
term health effects. Therefore, the turbidity was needed to alum with Ca(OCl)2 for removal of contaminants from the
be reduced before disinfection [7, 10]. As Ministry of water of Tonle Sap River.
Industry Mines and Energy reported that to prevent any
contamination of water, water system needs to go through II. Materials and Methods
four stages such as the storage reservoir, coagulation-
2.1. Sampling sites
flocculation, filtration and disinfection and based on
Raw water in this study was collected from Tole Sap
standard quality within turbidity below 5 nephelometric
River at Phnom Penh (with coordinate 11.5787626,
turbidity unit (NTU) [11].
104.9249167) on Sept. 9, 2020 at 6:30am.
Coagulation and flocculation are utilized to remove the
organic matter from the water, which are taken place in serial
2.2. Experimental set-up
steps that permit particle collision and growth into floc, then
2.2.1. Prepare of stock solution
followed by sedimentation. The serial steps are connected,
1% of coagulants stock were prepared by 1 g of
if coagulation is incomplete, the following steps are also
coagulant was suspended in 99 ml of distilled water and
incomplete and will be unsuccessful [12]. In coagulation
stirring in room temperature. Therefore, added 1ml of stock
process, colloidal or particles were destabilized by charge
solution to 1L of water is 10 mg/L. 0.1g of calcium
neutralization a coagulant or flocculant with a high positive
hypochlorite granule was suspended in 100 ml of distilled
charge is adsorbed on to the surfaces of negatively charged
water to produce Ca(OCl)2 stock solution and stirring at
colloids to form the small aggregate of particle, while in
room temperature. The concentration of free chlorine was
flocculation, aggregation particles are gathered into larger
obtained by an actual measurement using N,N-dimethyl-𝜌-
[13].
phenylenediamine (DPD) method.
Coagulation and disinfection are the most commonly
2.2.2. Coagulation and flocculation process
used for removing turbidity, color, algae and kill or
250 ml of raw water was filled in 4 bottles and were
inactivate infective microorganisms from surface water.
treated with three types of coagulant such as yellow PAC,
Chemical coagulant was used to form floc for removal of
brown PAC and alum with various dosing 20, 40, 60 and 80
those impurities [14]. Chlorine was used as a chemical
mg/l were performed stirring 3 mins with 300 rpm and
disinfectant due to its effectiveness, low cost, and easy to
settling time for 60 mins.
manage. Chlorine can be used in the form of calcium
2.2.3. Coagulation and disinfection process
hypochlorite, sodium chlorite, or chlorine gas. When
250 ml of raw water was filled in 4 bottles and were
chlorine undergo with water, it produced hypochlorous acid
treated with optimize coagulants with 300 rpm stirring speed
(HOCl) and hypochlorite (OCl- ) [15]. Hypochlorous acid
for 1 min before added Ca(OCl)2, after added Ca(OCl)2
act as a prime disinfection agent and its dominance is at pH
within 0.84, 1.0 and 1.7ppm of chlorine concentration were

164
continue stirring for 2 mins and settle for 60 mins. Turbidity (NTU) 83.53
Coliform (CFU/100ml) 1.33105
2.3. Analytical methods E. coli (CFU/100ml) 2.19105
10 ml of samples were collected at 3 cm below the pH 7.45
surface level. The turbidity was measured by using
HANNA-HI 98703 Portable Turbidimeter in three replicates. III. Results and Discussion
WM-32EP/pH meter was used to measure pH. The turbidity
removal efficiency was calculated by using Eq.1. 3.1. Effective of dosing coagulants on turbidity removal

(𝑇𝑢𝑟𝑖 − 𝑇𝑢𝑟𝑓 ) × 100 The effective of yellow PAC, brown PAC and Alum dose
%𝑇𝑢𝑟 𝑟𝑒𝑚𝑜𝑣𝑎𝑙 = (Eq.1) on turbidity removal were shown in Fig. 1. For figure was
𝑇𝑢𝑟𝑖
shown clearly that each dose of yellow PAC is more
where 𝑇𝑢𝑟𝑖 and 𝑇𝑢𝑟𝑓 are the initial and final turbidity, effective than each dose of Alum. The percentage turbidity
respectively. removal of yellow PAC was reached; 95.2%, 95.8%, 97.1%
and 97.5% at the dose of 20 mg/L, 40 mg/L, 60 mg/L and 80
Table 1. The characteristic of yellow and brown mg/L, respectively. Whereas, the removal of turbidity of
polyaluminum chloride alum was obtained; 89.9%, 94.6%, 95.6% and 95.5% at the
dose of 20 mg/L, 40 mg/L, 60 mg/L and 80 mg/L, in turn. In
Item Y.PAC B.PAC case, 20 mg/L of those coagulants were used for water
Al2O3 (%) ≥30 ≥30 treatment, brown PAC has efficiency lower than yellow PAC,
Basicity (%) 70-85 50-90 while the effective turbidity removal of brown PAC is more
pH Value(1% aqueous solution) 3.5-5.0 3.5-5.0 effective than yellow PAC with 96.8% at 40 mg/L and its
Water insoluble (%) ≤0.1 ≤1.5 efficiency was decrease where the dosage is more than 40
mg/L. Hence the highest effective of brown PAC is 40 mg/L.
Alternately, the comparation of all dose of coagulants
After each performed treatment, the sample were taken
was significant, the amount of yellow PAC ranges from 40-
for total coliform and E.coli count. 20 ml of sample was
80mg/L, brown PAC between 40-60mg/L and the amount of
filtered by using Whatman® sterile membrane filters
alum between 40-80mg/L can completely remove turbidity
without absorbent pads with 47 mm and pore size 0.45 μm
from 83.5 NTU to below 5 NTU. According to Zouboulis et
and was cultured on each Chromocult® Coliform agar plate
al., 2008 reported that polyaluminum chloride proved to be
which was then incubated at 36 ±2℃ for 21-24h. The
more effective coagulant than alum in water treatment plan
number of colony-forming unit (CFU) of 100ml was
[18]. In addition, Wang et al., 2011 illustrated that the
expressed in Eq.2. and the bacterial removal efficiency was
effectiveness turbidity removal of polyaluminum chloride
calculated by using Eq.3.
was slighly low that alum but it has highest effective on
(𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑜𝑙𝑜𝑛𝑦) × 100 cloudness water with algae compare with alum. Hence the
𝐶𝐹𝑈/100 = (Eq.2) ability of polyaluminum depend on the condictional of water
𝑠𝑎𝑚𝑝𝑙𝑒 𝑣𝑎𝑙𝑢𝑚𝑒 𝑓𝑖𝑙𝑡𝑒𝑟𝑒𝑑 𝑖𝑛 𝑚𝑙
[19].
(𝐵𝑎𝑐𝑖 − 𝐵𝑎𝑐𝑓 ) × 100
𝐵𝑎𝑐𝑡𝑒𝑟𝑖𝑎 𝑟𝑒𝑚𝑜𝑣𝑎𝑙 = (Eq.3) 3.2. Effective of dosing coagulants on pH change
𝐵𝑎𝑐𝑖
Fig. 2 illustrates the change of pH by using yellow
where 𝐵𝑎𝑐𝑖 and 𝐵𝑎𝑐𝑓 are the bacteria of raw sample
PAC, brown PAC and alum. As a result, the change of pH in
and treated sample, respectively. The characteristics of
using each dose of coagulant were decrease as the dosage of
source water were shown in Table 2.
coagulant was increased. When yellow PAC and brown PAC
were used, the pH of water were slightly change from 7.5 to
Table 2. Physicochemical characteristics of river water
7.2-7.0 and 7.2-6.8. while the high pH change (range from
Parameter Value 7.1-6.5) was occurred by using alum. Base on Table 1, the
basicity of yellow PAC is between 70-85% and brown PAC

165
is between 50-90%, while the basicity of alum is zero due to 1.7ppm, respectively. When the highest percentage of E.coli
it has no OH ions in its structure [20]. High concentration and Coliform removal of alum with free chlorine was
dosage of PAC to be used, the consumption of pH value is 100%;99.9% at 1.7ppm.
also changed. Accordingly, the optimal pH for removing
humic acid and clay is between 5-6 and 6.5-7.5 [21]. Also,
the treatment of using PAC, the optimum pH is 7.5 for its

Fig. 2. Effects of coagulant dosage on pH

After each performed of experiment, the sample were

Fig. 1. Effects of yellow PAC, Brown PAC and alum dose taken for analysis the residual free chlorine. As in Table 3
on turbidity shows the residual free chlorine after treated by using yellow
high effectiveness in removing turbid water [22]. Therefore, PAC, brown PAC and alum with Ca(OCl)2. The residual of
using yellow PAC has gradually changed in pH and it did free chlorine was obtained 0.12ppm, 0.005ppm, and
not need to adjust pH before or after treatment and it is 0.04ppm at using alone yellow PAC, brown PAC and alum,
suitable for the small scale of drinking water treatment. respectively. While yellow PAC was used with Ca(OCl)2 at
According to the guideline of drinking water-quality of 0.84ppm, 1ppm and 1.7ppm of free chlorine were increased
WHO [15] and the Ministry of Industry Mines and Energy to 0.17ppm, 0.22ppm and 0.29ppm, respectively. In case
[11] reported that pH range for drinking water is 6.5-8.5. using brown PAC with Ca(OCl)2, the residual free chlorine
Therefore, the amount of these coagulants can be used for was almost same as yellow PAC with Ca(OCl)2. When the
treatment water. residual free chlorine of alum with Ca(OCl)2 0.84ppm,
1ppm and 1.7ppm achieved 0.16ppm, 0.19ppm and 0.34ppm,
3.2. Effective of dosing Ca(OCl)2 on bacterial removal respectively.
and residual free chlorine
As Fig. 3 shows the effectiveness of bacteria removal of
Ca(OCl)2 with various coagulants. The percentage of
bacteria removals by optimum coagulant alone (control)
achieved 99.9% of E.coli removal and 97.9% of coliform
removal of 40mg/L of brown PAC, while the percentage of
E.coli removal achieved 99.4%, 99.6%, coliform removal
achieved 97.8%, and 98.2% with turns of 60mg/L of yellow
PAC and alum, respectively. Besides, the efficiency removal
of bacteria of Ca(OCl)2 with coagulant was obtained, the
percentage of E.coli and Coliform, 99.99%;99.84%,
100%;99.91%, 100%;99.97% of yellow PAC within the
concentration of free chlorine, 0.84ppm, 1ppm, and 1.7ppm,
respectively. Whereas the percentage of E.coli and Coliform Fig. 3. Effects of yellow PAC, brown PAC and alum
of brown PAC with free chlorine was 100%;99.9%, with Ca(OCl)2 on bacteria removal
100%;99.9%, 100%;99.9% at 0.84ppm, 1.0ppm, and

166
Table 3. Residual free chlorine of treated water by using References
yellow PAC, Brown PAC and alum with Ca(OCl)2 [1] J. Brown, S. Cairncross, and J. H. J. Ensink, “Water,
sanitation, hygiene and enteric infections in
Residual of free chlorine of treated water (ppm) children,” Arch. Dis. Child., vol. 98, no. 8, pp. 629–
Ca(OCl)2 (ppm Yellow Brown Alum 634, 2013, doi: 10.1136/archdischild-2011-301528.
of free chlorine) PAC PAC [2] A. Matilainen, E. T. Gjessing, T. Lahtinen, L. Hed,
Control 0.12 0.05 0.04 A. Bhatnagar, and M. Sillanpää, “An overview of
0.84 0.17 0.17 0.16 the methods used in the characterisation of natural
organic matter (NOM) in relation to drinking water
1.0 0.22 0.23 0.19
treatment,” Chemosphere, vol. 83, no. 11, pp. 1431–
1.7 0.29 0.26 0.34 1442, 2011, doi:
10.1016/j.chemosphere.2011.01.018.
Coagulant coupled with chlorination has efficiency for [3] F. U. Kac, M. Kobya, and E. Gengec, “Removal of
NOM chemical structure removal which was proposed the humic acid by fixed-bed electrocoagulation reactor:
formation of aggregates during the coagulation process Studies on modelling, adsorption kinetics and
HPSEC analyses,” J. Electroanal. Chem., vol. 804,
inhibits the reaction between chlorine and organics that have
no. October, pp. 199–211, 2017, doi:
been entrapped in the aggregates [23]. In contrast, the 10.1016/j.jelechem.2017.10.009.
effective disinfection level of residual concentration of [4] M. Yan, D. Wang, J. Yu, J. Ni, M. Edwards, and J.
available chlorine to disinfected is below or equal to 0.5 Qu, “Enhanced coagulation with polyaluminum
mg/L for at least 30 mins at pH lower than 8.0. The minimum chlorides: Role of pH/Alkalinity and speciation,”
residual concentration of available chlorine should be 0.2 Chemosphere, vol. 71, no. 9, pp. 1665–1673, 2008,
mg/L [15]. And according to Roy et al, 2016, the use of 2 doi: 10.1016/j.chemosphere.2008.01.019.
[5] J. McClean, “Disinfecting drinking water,” Water
mg/L of chlorine was obtained the residual free chlorine
Wastes Dig., vol. 55, no. 2, pp. 34–35, 2015.
below 0.5 ppm which is within the prescribed limit. [6] P. Hayden, “Water resources,” Fire Risk Manag., no.
Therefore, the concentration of free chlorine that suitable for MAY, pp. 39–42, 2009.
treatment water is 1.7ppm with those amount coagulants. [7] M. J. Allen and R. Copes, “Turbidity and microbial
risk in drinking water. Prepared by : Martin J . Allen ,
IV. Conclusion Ronald W . Brecher , Ray Copes , Steve E . Hrudey
( Chair ), and Pierre Payment Ministerial Technical
The effective of different dosage of coagulant and Advisory Committee Prepare,” no. May, 2014.
Ca(OCl)2 were studied experimentally. The efficiency of [8] W. Quality, “Turbidity: Description, Impact on
coagulant operation in coagulation and the efficiency of Water Quality, Sources, Measures -,” Water Qual.,
vol. 3, no. March, p. 2.4, 2008.
Ca(OCl)2 operation disinfection. The highest turbidity
[9] T. Tripathy and B. R. De, “Flocculation: A new way
removal was obtained at 40 mg/L of brown PAC within to treat the waste water,” J. Phys. Sci., vol. 10, pp.
96.8%, 60 mg/L of yellow PAC and alum within 97.1% and 93–127, 2006.
95.6%, respectively. In addition, the highest bacteria [10] World Health Organization, “Water Quality and
removal and appropriate residual free chlorine were Health- Review of Turbidity: Information for
achieved when using the optimum dose of coagulant couple regulators and water suppliers,”
Who/Fwc/Wsh/17.01, p. 10, 2017, [Online].
with Ca(OCl)2 at 1.7ppm of chlorine concentration.
Available:
https://www.who.int/water_sanitation_health/publi
Acknowledgement cations/turbidity-information-
200217.pdf%0Ahttp://www.who.int/water_sanitati
We are thankful to the Science and Technology Research
on_health/publications/turbidity-information-
Partnership for Sustainable Development (SATREPS), the 200217.pdf.
Japan Science and Technology Agency (JST)/Japan [11] MOIMAE, “Kingdom of Cambodia Drinking Water
International Cooperation Agency (JICA) for their financial Quality Standards,” Drink. Water Qual. Stand., pp.
support. 1–20, 2004.
[12] N. B. Prakash, V. Sockan, and P. Jayakaran, “Waste
Water Treatment by Coagulation and Flocculation,”
Certif. Int. J. Eng. Sci. Innov. Technol., vol. 9001, no.

167
 2, pp. 2319–5967, 2014, [Online]. Available:
http://www.ijesit.com/Volume 3/Issue
2/IJESIT201402_61.pdf.
[13] T. Suopaj, Functionalized nanocelluloses in
wastewater treatment applications, no. April. 2015.
[14] DEFRA, “Water Treatment Processes - Drinking
Water Inspectorate,” Water Treat. Process, pp. 1–36,
2018.
[15] W. H. Organization, “Guidelines for Drinking-water
Quality 4th ed,” Who Libr. Cat. Data., pp. 1–564,
2011, doi: 10.1007/978-1-4020-4410-6_184.
[16] World Health Organization, “Chapter 1 - Summary
and Evaluation,” pp. 1–499, 2000.
[17] X. Huang et al., “Effect of dosing sequence and raw
water pH on coagulation performance and flocs
properties using dual-coagulation of polyaluminum
chloride and compound bioflocculant in low
temperature surface water treatment,” Chem. Eng. J.,
vol. 229, pp. 477–483, 2013, doi:
10.1016/j.cej.2013.06.029.
[18] A. Zouboulis, G. Traskas, and P. Samaras,
“Comparison of efficiency between poly-aluminium
chloride and aluminium sulphate coagulants during
full-scale experiments in a drinking water treatment
plant,” Sep. Sci. Technol., vol. 43, no. 6, pp. 1507–
1519, 2008, doi: 10.1080/01496390801940903.
[19] Y. Wang, S. Zhuo, N. Li, and Y. Yang, “Influences
of various aluminum coagulants on algae floc
structure, strength and flotation effect,” Procedia
Environ. Sci., vol. 8, no. November, pp. 75–80, 2011,
doi: 10.1016/j.proenv.2011.10.014.
[20] R. E. Pieroni, “Dosage Calculation,” Arch. Intern.
Med., vol. 140, no. 12, p. 1673, 1980, doi:
10.1001/archinte.1980.00330230119026.
[21] G. Annadurai, S. S. Sung, and D. J. Lee,
“Optimization of Floc Characteristics for Treatment
of Highly Turbid Water,” Sep. Sci. Technol., vol. 39,
no. 1, pp. 19–42, 2005, doi: 10.1081/ss-120027399.
[22] S. Ghafari, H. A. Aziz, and M. J. K. Bashir, “The use
of poly-aluminum chloride and alum for the
treatment of partially stabilized leachate: A
comparative study,” Desalination, vol. 257, no. 1–3,
pp. 110–116, 2010, doi:
10.1016/j.desal.2010.02.037.
[23] D. Ghernaout, “Coagulation and Chlorination of
NOM and Algae in Water Treatment: A Review,” Int.
J. Environ. Monit. Anal., vol. 2, no. 6, p. 23, 2014,
doi: 10.11648/j.ijema.s.2014020601.14.

168
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Occurrence, Transportation, Regulation and Treatment Methods of Heavy Metals in

Groundwater: A Review on Case of Well Water around Tonle Sap Lake

Laty Ma 1, Rina Heu 2, 4,*, Mardi Meas 1, Khy Eam Eang2, and Sokly Siev3, 4

1
Graduate School of Water and Environmental Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd, P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia, Russian Federation
Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd., P.O.
Box 86, 12156 Phnom Penh, Cambodia
4
Water and Environmental Research Unit, Research and Innovation Centre, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
*Corresponding author: heu.rina@itc.edu.kh

Abstract
Tonle Sap Lake (TSL), the largest natural freshwater lake in Southeast Asian area with rich endowment biodiversity
and various ecosystems, provides a huge potential on agriculture and aquaculture to Cambodians who live on the floating
villages and around the floodplain area. The lake and its ecosystems, however, are widely under threat due to
anthropogenic activities occurring inside and outside of its basin such as activities by agricultures, water infrastructure
development and land use change. Currently, one of the most concerns in this watershed is lack of clean water for
consumption. While the surface water is being polluted from various human activities (e.g., fishing, navigation, using
pesticide & herbicide for plant, and floating village), the groundwater was being utilized along the TSL for domestics,
small industrials, and agricultural purposes. Most of shallow groundwater in Cambodia has been contaminated by heavy
metals including arsenic (As), manganese (Mn), and iron (Fe). The exposure of the heavy metal substances will affect
the environment and human health when its concentration is over the limitation threshold of the Drinking Water Quality
Standard in Cambodia and WHO. This study aims to review the state knowledge on the well water, as well as the
essential research approach by focusing on several topics including the occurrences, transportation, monitoring and
treatment method of heavy metals. The findings of this review paper will propose and apply for the water treatment
technologies on heavy metal removal of contaminated groundwater along the TSL. More efforts will spread the
awareness of water quality for citizen surrounding the TSL area to understand the impact of utilizing well water with
the presence of heavy mental that can affect their health in the long-term.

Keywords: Heavy metals, Occurrences, Tonle Sap Lake, Treatment methods, Well water

I. Introduction its ecosystems are widely under threat due to anthropogenic

Tonle Sap Lake (TSL), the largest permanent freshwater activities occurring inside and outside its basin. The surface
body in Southeast Asia, is an important natural reservoir with water in TSL was concerned and being polluted from the
the total watershed extending over 43% ~ 86,786 km2 of the various activities such as fishing, navigation, using
national land. The lake surface area, which underpins its pesticide/ herbicide, and floating houses. Outside the basin,
uniqueness not only in hydrology, but also in water quality, citizen living along the TSL faced with lack of clean water
production, and biodiversity [1]. Unfortunately, the lake and for common use and for drinking.

169
 As a result, numerous dug/tube wells along in the ion exchange capacity of the geological materials and organic
floodplain zone have been installed and abstracted for matter content [3].
myriad purposes such as domestic, drinking, irrigation and Groundwater plays a dominant role in purposing of
small-scale industries (Fig. 1). According to Ministry of domestic, water supply, irrigation and small scale industries
Rural Development (MRD, 2014), groundwater is major for both rural and urban due to polluted surface water and
source for drinking water supply in Cambodia [5]; 53% of lacking of usable water in dry season[3]. Many dug/tube wells
Cambodian households drink water from groundwater have installed as the water resource for people live in the
sources in the dry season and ~270,000 tube-wells with hand floodplain area along the TSL. A study found the major
pump are functioning. The previous studies have shown that issues and challenges threatening of groundwater resource;
most of tube and drill well around the TLS area have groundwater quality problems with high arsenic (As) and
contaminated by heavy metals [6, 7]. The existence of heavy iron (Fe) contents in Mekong and Tonle Sap River basin [5].
metals with high concentration in water sources and drinking Based on [5,8,9,11], the chemical distribution assessment of
water is a most concerning problem to human health. well water in the floodplain areas along the TSL shows that
the present of arsenic (As), manganese (Mn), and iron (Fe)
with high contaminants. Some provinces along the Mekong
River Basin and TSL Basin have (As) problem in
groundwater. About 38% of tube-wells in those provinces
area contaminated with arsenic above 50ppb [5]. This
review has shown the significant chemical contamination
occurred in well water in the floodplain areas around the TSL
(Table 1) [8-11]. Previous data analyses from 2007 to 2020
have confirmed that the major concentration of heavy metals
in well water such as As, Mn, and Fe with high concentration
comparing to the standard of drinking water quality
regulated by Ministry of industrial and handicraft [13] and
standard for drinking water by World Health Organization [14].
Surface water naturally interacts with the groundwater
Figure1: Location of well water around the TSL, Cambodia [5] system noticeably shallow groundwater [7]. Therefore, the
In this paper, several major categories focusing on the seasonal fluctuation of surface water in TSL may contribute
occurrences and transportation of heavy metals in the to the change of groundwater composition in the aquifer
groundwater were reviewed; a case study of well water system. Well water quality in floodplain areas will affect by
along the TSL, the regulation for monitoring the water the geochemical evolution comparing to other well water
quality, and the treatment methods of heavy metals, which areas. During the rainy season, the elemental composition of
were applied as conventional water treatment at those areas. lake water appeared to greatly influence by the intrusion of
In addition, the review had summarized and gave the water from the Mekong River through the Tonle Sap River.
recommendation for research required. During the dry season, the type of lake water shifted,
suggesting that the lake water stored during the rainy season
2. Occurrence and Transportation of Heavy Metals replaced by inflow from other tributaries and groundwater
Heavy metal is general collective term, which applies to in its vicinity [8]. The past study showed that the seasonal
the group of metals and metalloids with atomic density changes in the elemental composition of the lake water were
greater than 5g.cm-3 or 5 times more than water [38], and largely controlled by surface water and groundwater
they are the natural earth’s crust components. A variety of circulation. However, potential risks of As and other heavy
reactions in soil environment, e.g., acid/base precipitation/ metal poisoning in the lake water system are growing, which
dissolution, oxidation/reduction, and sorption/iron exchange depend on the changes in the upper reaches of Mekong River
process can influence the speciation and mobility of metal and material inputs into the lake. Therefore, we need to
contaminants. The rate and extent of these reactions will monitor the status of the surface water and groundwater,
depend on factors such as pH, oxidation potential (Eh), simultaneously, in TSL and its tributary rivers.
complexation with other dissolved constituents, sorption and
2

170
 Table2: Typical of heavy metals, their effect on Human Health
Heavy Metals Permissible Effects on
Level (mg/l) Human Health
Arsenic(As) 0.01* Skin manifestations, visceral
cancers, vascular disease.
Cadmium(Cd) 0.003* Kidney damage, renal disorder,
human carcinogenic.
Chromium(Cr) 0.05*** Headache, diarrhea, nausea,
vomiting, carcinogenic.
Copper(Cu) 0.25** Liver damage, Wilson disease,
insomnia.
Iron(Fe) 0.3*** Liver disease, heart disease,
arthritis and osteoporosis.
Nickel(Ni) 0.2** Dermatitis, nausea, chronic
asthma, coughing, carcinogen.
Lead(Pb) 0.006** Liver damage the fatal brain,
kidney, circulatory system
Manganese(Mn) 0.1*** Weakness, muscle pain, and
dizziness.
Mercury(Hg) 0.001*** Rheumatoid arthritis, kidneys,
and nervous system
Nickel(Ni) 0.2** Dermatitis, nausea, coughing,
chronic asthma, carcinogen
Sodium(Na) 250*** High blood pressure, stroke and
heart, stomach cancer, kidney
stones osteoporosis,
Zinc(Zn) 0.8** Depression, neurological signs,
lethargy, and increased thirst
Note sources: (*) WHO, (**) USEPA, (***) MIH, Cambodia [12-16]

To protect human health, regulatory agencies have

recommended the maximum concentrations of few heavy
metals in drinking water (e.g., USEPA, 2015; WHO, 2011)
[34]. The royal government of Cambodia has enacted the
national drinking water quality standard to ensure safe
drinking water for everybody by adapting to WHO. Some
parameters are different in values from the WHO (Table 3).
Table 3: Drinking water quality standard (MIH, WHO) [13, 14]
Parameters Unit Cambodia WHO
3. Toxicity and Standard Regulation of Heavy Metals
pH - 6.5-8.5 6.5-8.5
in Drinking Water Turbidity NTU 5 ≤ 5
Studies reported various effects of heavy metals in TDS mg/l 800 -
EC µS/cm - 250
drinking water [39, 38]. The chemical contamination such as
Hardness mg/l 300
arsenic, fluoride, nitrate, and manganese presented in Calcium (Ca) mg/l 200 -
groundwater might pose human health and environmental Cadmium (Cd) mg/l 0.003 0.003
issues. Among the heavy metals, As, Cd, Pb, Cr, Cu, Hg, and Chloride (Cl) mg/l 250 250
Chromium (Cr) mg/l 1 0.05
Ni are the major concern (Table 2), mainly due to their Copper (Cu) mg/l 0.3 2
presence at relatively high concentrations in drinking water Iron (Fe) mg/l 0.3 -
and their effects on human health [2]. The most toxic effect Magnesium(Mg) mg/l 150 -
of these metals in their ionic species are the most stable Sodium (Na) mg/l 200 200
Manganese(Mn) mg/l 0.1 0.5
oxidation state (Cd, Pb, Hg, and As) which react with the Lead (Pb) mg/l 0.01 0.01
body’s bio-molecules to form extremely stable bio toxic Sulfate (SO4) mg/l 250 500
compounds [11]. Arsenic (As) mg/l 0.05 0.01

171
4. Treatment Methods of Heavy Metal risk to human body for long-term daily consuming. A
Removal of heavy metals is an important step toward safe standard for drinking water regulated to protect public
drinking water. About 71% of rural Cambodia households health.
 The conventional water treatment method is important for
reported always treat their drinking water prior consumption [20].
communities at TSL area, but there are no data shown the
Traditional methods were used for water treatment in effectiveness of heavy metals removal.
Cambodian such as boiling, filtration, sedimentation and Thus, this paper encourages the future research to
solar radiation, which is discussed in table 4. Our field develop the water treatment method to remove the
survey on 15 November 2020 at the Chhnok Tru Commune contaminant of heavy metal from well water with high
located along the TSL area reported that the local residents efficiency in order to protect people and improve human
use raw water for common utilities and 20-litre plastic jugs health around the TSL area.
for drinking. Some of them also used water treatment Table 5: Summary of treatment methods of heavy metals [23-33]
products including ceramic water filters, bio-sand filters and Treatment Methods Advantages Disadvantage
mineral pot filter. Ion exchange High transformation of Removes only limited metal ions.
(As, Fe, Mn, Pb) components. Operational cost is high.
Table 4: Comparison of traditional treatment methods [21] High removal efficiency, Not handle concentrated metal
solution as the matrix as fouled
Methods Advantages Disadvantages fast kinetic reaction, by organics and solids.
Less sludge production.
Filters To eliminate microbes Not affordable for Coagulation Cost effective, Dewatering Generation of sludge, Utilization
No particles in water poor households (As, Ni, Zn, F, Cd) qualities. of chemicals is high.
Water still contains Cost effective, high removal Jar test, sludge and hardness
Ease use for large family efficiency. production, and pH dependence.
No need to boil water some microbes
Membrane High removal efficiency of Very expensive, Membrane fouling,
Produce clean, clear water Filtrations heavy metals. Complex process.
(As, Fe, Mn, Lower space requirement High cost membrane fouling, scaling,
Boiling Most effective way to eli Time-consuming Reusable heavy metals, High-energy consumption.
Cu, Cd, Zn, Pb,) Sludge production.
minate microbes Requires fuel
Produces clean, safe water Slow Adsorption Easy operation, Desorption
Poses risk of fire (As, Fe, Mn) Less sludge production,
Easy method for community. Low cost adsorbents.
High removal efficiency
Tablets & Effective in eliminating Chemical substance Flexibility and simplicity
powders microbes maybe harmful. of design & insensitivity to
toxic pollution
Ease to use Efficient for the removal of Initial investment is high, need
Electrochemical
No need to boil water. Treatment important metal ions, low high electrical supply
(Cu, Cr, and Ni) chemical usage. High initial
High removal investment and electricity
There is no data or research to confirm that the classical Efficiency, reusable heavy requirement
metals, space saving.
product treatment could remove contaminants of heavy metals
in water, but [20] shows some effectiveness of traditional Acknowledgement
methods. The above treatment methods, which are being The authors would like to acknowledge financial support
practiced in the community and rural household, cannot from French Development Agency (AFD) with research
remove the heavy metals with high efficiency. The water grant number CKH 1236 02P and the Ministry of Education,
treatment methods for removal heavy metals with efficiency Youth and Sport through Higher Education Improvement
are discussed in Table 5. Adsorption is considered as a suitable Project (HEIP) Credit No 6221-KH at Institute of
method for removal As, Fe, Mn because it is easy to operate, Technology of Cambodia.
less sludge production and low cost. This is the best water References
treatment for rural. [1] Campbell, I. C., Poole, C., Giesen, W., & Valbo-
Jorgensen, J. (2006). Species diversity and ecology of
5. Conclusion and Future Research need Tonle Sap Great Lake, Cambodia. Aquatic Sciences,
68(3), 355-373.
To sum up, some significant reviews about the [2] Duruibe, J. O., Ogwuegbu, M. O. C., & Egwurugwu, J.
occurrences, transportation, toxicity, regulation, and N. (2007). Heavy metal pollution and human biotoxic
treatment method for removal heavy metals in groundwater effects. International Journal of physical sciences, 2(5),
around the TSL were found as summarized: 112-118.
 The contaminations of heavy metals in well water along [3] Hashim, M. A., Mukhopadhyay, S., Sahu, J. N., &
the TSL area were detected includes As, Fe, Mn with high Sengupta, B. (2011). Remediation technologies for
concentration comparing to the standard. heavy metal contaminated groundwater. Journal of
 The most toxic heavy metals that is harm to human health environmental management, 92(10), 2355-2388.
is As. Another toxic heavy metal is Fe, which can also
4

172
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Lake Water Temperature Characteristics and Long-Term Variations in Water

Temperature in the Tonle Sap Lake

Yoichi Fujihara 1,*, Keisuke Hoshikawa 2, Hideto Fujii 3, Takashi Nakamura 4, Sokly Siev 5, and Sambo Lun 5

1
Ishikawa Prefectural University, 1-308, Suematsu, Nonoichi, Ishikawa 921-8836, Japan
2
Toyama Prefectural University, 5180, Kurokawa, Imizu, Toyama 939-0398, Japan
3
Yamagata University, 1-23, Wakaba, Tsuruoka, Yamagata 997-8555, Japan
4
Tokyo Institute of Technology, 4259-G5-3, Nagatsuta, Midori, Yokohama 226-8501, Japan
5
Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: yfuji@ishikawa-pu.ac.jp

Abstract

Water temperature is an important parameter for lake management because it has a major impact on material circulation
and the local ecosystems. In addition, various regions have witnessed increased lake water temperatures owing to the increased
air temperatures caused by the recent climate changes. However, there is little information on the water-temperature
characteristics and long-term water-temperature trends for Tonle Sap Lake, Cambodia, as there are no regular, long-term
water-temperature observations, and little is known about the changing lake water environment. Therefore, we investigated
the characteristics of and trends in surface water temperatures of the Tonle Sap Lake from 2000 to 2019 using MOD11A1, a
MODIS product to obtain daily land surface temperature data using Google Earth Engine. Moreover, the relationship between
water temperature fluctuations and hydrological–meteorological conditions (e.g., air temperature, precipitation, and lake
water level) were analyzed. The maximum and minimum water temperatures were recorded as 30.7 ℃ and 24.6 ℃ in May
and January, respectively, indicating that these trends were one month behind those of the air temperatures. One of the spatial
characteristics of this lake is that the northern part is cooler than the southern part. With respect to the water temperature
trends, the maximum value increased at the rate of 0.20 ℃ in 10 years, minimum value decreased at the rate of 0.78 ℃ in 10
years, and difference between the maximum and minimum temperatures is increasing. There was a significant correlation
between the maximum water temperature and lake water level. The lake water level has declined recently, with the maximum
water level for 2000–2009 being 9.3 m (maximum: 10.4 m, minimum: 9.3 m) and that for 2010–2019 being 8.2 m (maximum:
10.5 m, minimum: 6.0 m). This indicates that the declining water levels are causing higher maximum water temperatures with
increasing air temperatures. Conversely, the correlation between the minimum water temperature and lake water level was
not strong. Factors contributing to the lower minimum water temperatures were not identified and must be studied in more
detail in the future.

Keywords: Surface water temperature, MODIS, Climate change, Lake water level, Google Earth Engine

I. Introduction management because it has a major impact on material

circulation and ecosystems. In addition, various regions
Water temperature is an important parameter for lake

173
have witnessed a rise in lake water temperature due to the (maximum 26.5 °C, minimum 25.7 °C), while for the 2010–
rise in air temperature caused by recent climate change. 2019 period it was 25.9 °C (maximum 26.7 °C, minimum
However, there is little information on long-term water 25.2 °C). We found that the annual fluctuation of water
temperature trends for Tonle Sap Lake, as there are no temperature has been increasing since 2009. Moreover, the
regular, long-term water temperature observations. surface temperature showed a decreasing trend of 0.78 °C
Although Oyagi et al. (2017) reported intensive observations /10 years, and there was no trend for the air temperature of
of water temperature and no thermocline in the lake, little is Phnom Penh city. Considering that the temperature rise due
known about the changes in the lake's water environment. to global warming has shown an upward trend of 0.34 °C /10
years worldwide (O’Reilly et al., 2015), the decrease in
II. Materials and Methods temperature of Tonle Sap Lake is an interesting phenomenon.
We investigated the trend in surface water temperature
of Tonle Sap Lake from 2000 to 2019 using MOD11A1,
which is a MODIS data product, and there are 365/366
images per year. The R-square between the observed water
temperature and MODIS-derived temperature was 30
approximately 0.54 in Tonle Sap Lake. We used the Google

Temperature ( ℃)
95 percentile
Earth Engine, a cloud-based platform for global-scale Median
geospatial analysis that uses Google's computational 5 percentile
capabilities for various high-impact societal applications,
including water management, climate monitoring, and 25
environmental protection. It is an integrated platform
designed to empower traditional remote sensing scientists
and a much wider audience lacking the technical capacity
required to use conventional supercomputers or large-scale
cloud computing (Gorelick et al., 2017).
We investigated the lake water temperature fluctuations 20
2000 2010 2020
for open water areas throughout the year. The data tended to Year
be influenced by clouds. The median image for the year was Fig. 1 Trends of water temperature
obtained for each pixel from 365/366 land surface (maximum, minimum, and mean)
temperature values to remove this influence. In addition, the
median surface temperature within the lake boundary was
obtained to calculate the annual and lake average water Max
temperatures. The trend of this median surface temperature Average
12
Min
was analyzed for data spanning 20 years. We also analyzed
the yearly median air temperature of ERA5. In addition, as
Water Level (m)

the maximum water level of the lake has been shown to

affect water quality (e.g., Hoshikawa et al., 2019), its 8
relationship with water temperature was also analyzed.

III. Results and Discussion

As a result, the northwestern part of the lake was found 4
to have lower water temperatures than the southeastern part,
with a difference of 1.5 °C. The annual water temperature
data for the 2000–2019 period and the annual maximum
water level variation of the lake are shown in Figure 1 and 0
2000 2010 2020
Figure 2. We determined that the annual mean water Year
temperature for the 2000–2009 period was 26.0 °C Fig. 2 Trends of water level

174
 2000–2009 2010–2019

Max water temp. (℃)

r=0.57 r=0.14 r=0.59
32 32 32

30 30 30

28 28 28
27 28 29 30 1000 1500 2000 6 8 10
Annual air temp ( ℃) Annual precipitation (mm) Max. water level (m)
Fig.3 Factors of maximum surface water temperature

2000–2009 2010–2019
Min water temp. (℃)

23 23 23

22 22 22

21 21 21
r=0.10 r=0.03 r=0.14
27 28 29 30 1000 1500 2000 6 8 10
Annual air temp ( ℃) Annual precipitation (mm) Max. water level (m)
Fig.4 Factors of minimum surface water temperature

For the 2000–2009 period, there was a positive be studied in more detail in the future.
correlation between annual water temperature and annual The construction of a group of large dams is currently
maximum water level. However, after 2009, it changed to a underway in the up-stream area, and it has been suggested
negative correlation, and the variability between annual that this will change the hydrological environment in the
water temperature and annual maximum water level Mekong River (for example, Hecht et. al., 2019). While the
increased. It appears that both annual water temperature and relationship between water temperature fluctuation and dam
annual maximum water level have fluctuated more since construction cannot be accurately established in our study,
2009. There was a significant correlation between the such an investigation is necessary in the future. Due to the
maximum water temperature and lake water level (Figure 3). lack of hydrological data on Tonle Sap Lake, there has been
The lake water level has declined recently, with the no regular water quality monitoring, including water
maximum water level for 2000–2009 being 9.3 m temperature, but water temperatures have definitely begun
(maximum: 10.4 m, minimum: 9.3 m) and that for 2010– to change since 2009.
2019 being 8.2 m (maximum: 10.5 m, minimum: 6.0 m).
This indicates that the declining water levels are causing IV. Conclusion
higher maximum water temperatures with increasing air We investigated the characteristics of and trends in
temperatures. Conversely, the correlation between the surface water temperatures of the Tonle Sap Lake from 2000
minimum water temperature and lake water level was not to 2019 using MOD11A1, a MODIS product to obtain daily
strong (Figure 4). Factors contributing to the lower land surface temperature data using Google Earth Engine.
minimum water temperatures were not identified and must Google Earth Engine is an effective tool for satellite image

175
analysis. The maximum and minimum water temperatures variable warming of lake surface waters around the
are recorded in May and January, respectively, indicating globe, Geophysical Research Letters, 42,
that these are one month behind those of the air temperatures. 2015GL066235, doi:10.1002/2015GL066235
One of the spatial characteristics is that the northern part is [6] Oyagi H, Endoh S, Ishikawa T, Okumura Y,
cooler than the southern part. Maximum water temperature Tsukawaki S (2017) Seasonal changes in water
increased at the rate of 0.20℃/10-year, minimum water quality as affected by water level fluctuations in Lake
temperature decreased at the rate of 0.78℃/10-year. Tonle Sap, Cambodia, Geographical Review of Japan
There was a significant correlation between the Series B, 90(2): 53-65,
maximum water temperature and lake water level. The lake doi:10.4157/geogrevjapanb.90.53
water level has declined recently. This indicates that the
declining water levels are causing higher maximum water
temperatures. The correlation between the minimum water
temperature and lake water level was not strong. Factors
contributing to the lower minimum water temperatures were
not identified and must be studied in more detail in the future.

Acknowledgement
We are thankful to the Science and Technology Research
Partnership for Sustainable Development (SATREPS), the
Japan Science and Technology Agency (JST)/Japan
International Cooperation Agency (JICA) for their financial
support.

References
[1] Gorelick N, Hancher M, Dixon M, Ilyushchenko S,
Thau D, Moore R (2017) Google Earth Engine:
Planetary-scale geospatial analysis for everyone,
Remote Sensing of Environment, 202: 18-27,
doi:10.1016/j.rse.2017.06.031
[2] Hecht JS, Lacombe G, Arias ME, Dang TD, Pimanh
T (2019) Hydropower dams of the Me-kong River
basin: A review of their hydrological impacts, Journal
of Hydrology, 568: 285-300,
doi:10.1016/j.jhydrol.2018.10.045
[3] Hortle KG (2007) Consumption and the yield of fish
and other aquatic animals from the Low-er Mekong
Basin. MRC Technical Paper No. 16, Mekong River
Commission, Vientiane, Lao PDR
[4] Hoshikawa K, Fujihara Y, Siev S, Arai S, Nakamura
T, Fujii H, Sok T, Yoshimura C (2019)
Characterization of total suspended solid dynamics in
a large shallow lake using long-term daily satellite
images, Hydrological Processes, 33(21): 2745-2758,
doi:10.1002/hyp.13525
[5] O’Reilly CM., Sharma S, Gray DK, Hampton SE,
Read JS, Rowley RJ et al. (2015) Rapid and highly

176
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Meta-analysis of Photocatalytic Degradation of Pharmaceuticals and Personal Care Products in

Water

Cui LI 1, and Chihiro Yoshimura 1*

1
School of Environment and Society, Tokyo Institute of Technology, 2-12-1-M1-4
Meguro-ku, Tokyo, 152-8552, Japan
* yoshimura.c.aa@m.titech.ac.jp

Abstract

Pharmaceuticals and personal care products (PPCPs) are a unique group of emerging environmental
contaminants which are widely found in water environment in the world. An increasing number of studies has
confirmed the presence of various PPCPs in different environmental compartments, which raises concerns
about the potential adverse effects to humans and wildlife. To overcome this issue of PPCPs, heterogeneous
photocatalysis, for example using TiO2, is an emerging and viable treatment technology. Its combination with
carbonaceous materials (e.g., activated carbon, carbon nanotubes and graphene nanosheets) has been found to
significantly improve the performance. In this study, meta-analysis with random effects model was applied to
literature data from more than fifty series of experiments. As a result, the pooled mean removal efficiency of
PPCPs by carbonaceous/TiO2 materials was 89.5% (95 % CI = 83.4 – 94.4%, k= 64). Subgroup analysis
expounded the important and influential variables (e.g., different carbon-based TiO2 photo-catalysts, process
parameters and PPCPs characteristics) in the degradation processes. Furthermore, artificial neural network
models were applied to modeling the performance (removal efficiency) of these carbonaceous/TiO2
photocatalytic experiments.

Keywords: artificial neural network modeling, carbonaceous/TiO2 composites, heterogeneous photocatalysis, meta-
analysis, modified kinetic modeling

I. Introduction compartments, including lakes, waste water effluents and

Pharmaceuticals and personal care products (PPCPs) are even drinking tap waters. Heterogeneous photocatalysis was
emerging environmental contaminants and have attracted considered as a promising alternative for degrading various
the world’s attention since the late 1990s. PPCPs are toxic organic pollutants in water, which is drawing
essential in maintaining high human health and well-being, increasingly attention due to its ability to mineralize the
has been reported that more than 50,000 PPCPs are pollutants into water and carbon dioxide[2]. Compared with
produced[1]. An increasing number of studies has confirmed pure TiO2, the modification of TiO2 with carbonaceous
the presence of various PPCPs in different environmental

177
nanomaterials is an innovative approach for the retrieved articles. The quality of a study was evaluated by
enhancement of photocatalytic activity. the criteria of clarity, for instance, the associated
The study of carbonaceous/TiO2 has now focused on methodologies. The availability of the data was analyzed by
synthesis and optimization of the photocatalysts and whether the study implied the mean and standard deviation
treatment of single PPCP compounds[3].There are rare of data. After these assessments, papers will be compiled as
models which combined element reaction and operation the primary dataset, including the first author, year of
conditions together for prediction of photocatalytic publication, the type of PPCPs, composition and dosage of
performance. Similar photocatalytic processes showed a photocatalysts, initial concentration of experiment, source
variation in degradation performance for the same PPCP and intensity of light irradiation, reaction time, temperature,
compound . For instance, the removal efficiency of removal efficiency (RE) and apparent rate constant (Kapp).
Ibuprofen by AC/TiO2 photocatalytic system, varied from The other properties recorded from pubmed.com, and they
78.2% to 99%[4].This fragmented information makes it were: molecular weight (MW), topological polar surface
difficult to compare the removal of a certain pollutant and to area (TPSA), complexity, pKa and octanol/water
investigate the influential parameters, operation settings and partitioning coefficient (logKow). The total data points
the technique application in actual industry. In addition, included in each analysis are shown in Table 1.
present study only focused on one or several PPCPs in one
study. While the understanding of removal performance of Table 1. Data points included in each analysis
multiple PPCPs under different reaction condition is needed
to support the real application step. Category Data points (k)
Thus, it is helpful to optimize bias and variation between
Total 64
studies and utilize results from multiple PPCPs and reaction
conditions. In addition, effort could be taken to build a Meta-analysis 64
model based on the existing experimental results, which Artificial neural network modeling 49
could be applicable for future prediction of
photodegradation performance of PPCPs. The specific
objectives of this study are:
2.2. Description of statistical approaches
(1) To statistically understand the removal efficiency of
(a) Meta-analysis
PPCPs by carbonaceous/TiO2 photocatalytic treatment;
Meta-analysis is a statistical procedure for combining
(2) To identify the important variables in causing the
results from multiple studies which originated from medical
difference in removal performance, getting insight into the
research. A meta- analytical approach can access the drivers
degradation process;
of the variability relating to the inherent properties of the
(3) To build a model based on the data collected from
chemicals and operations parameter of a treatment. process.
literature, addressing both the element reaction and
In the field of environmental research, for the effect of
operation conditions.
interest is removal efficiency, correspondingly proportion
meta-analysis is applied[5]. Unless otherwise specified, R
II. Materials and Methods program (version 3.6.3) was used for statistical analysis.
2.1. Data source, searching strategy and data processing Meta-analysis was conducted using ‘metafor’ and ‘meta’
A systematic search of the peer-reviewed literature packages. Random effects model was assumed to
was carried out in the Web of Science database for articles accommodate the anticipated heterogeneity.
that evaluated the PPCPs photodegradation processes in
aqueous solution. The literature search was limited to peer- (b) Artificial neural network modeling
reviewed publications written in English between 2010 until Artificial neural network modeling (ANN) is used to
June, 2020. recognize the relationship between variables and
The inclusion criteria were applied to each publication and degradation performance[6]. The input variables were set as
they included scope, study quality, and data availability. molecular related variables (molecular weight of
Scope criteria is the evaluation of titles and abstracts of the compounds, logKow, pKa, complexity and TPSA) and

178
process related variables (concentration, dosage of irradiation time from 1 to 4 hours. The pooled average RE
photocatalyst, TiO2 content of photocatalyst, light source for PPCPs degradation by the carbonaceous/TiO2
and intensity). And the output variables were set as removal photocatalysis was 89.5% (95 % CI = 83.4 – 94.4%, k= 64).
efficiency. MATLAB 2019a (ANN toolbox) is used for the The results of subgroup parameters analysis were shown in
development of ANN model. A supervised learning Table 3.
technique called Levenberg-Marquardt was used for the The highest removal efficiency was obtained from
training of the network. The total number of hidden neurons CNT/TiO2 subgroup with an average percentage of 94.6%
was decided by the mean square error (MSE) and the (95% CI = 89.8 – 98.0%, k= 29), followed by AC/TiO2
performance of the ANN model was evaluated by (79.6%, k= 11) and Graphene/TiO2 (78.9%, k= 8).
correlation coefficient (R). The settings of ANN model are Regarding light source, UVC has the highest pooled average
shown in Table 2. removal efficiency (93.2%,95%CI=86.4–97.9%, k=28).
Sensitivity analysis was conducted by Garson’s Most PPCPs were persistent under visible photocatalytic
method[6].Each weight decides which proportion of the process (RE < 10%), while the graphene/TiO2 has appeared
incoming signal will be transmitted into the neuron’s body . to improve the overall degradation.
It is based on partitioning of connection weights and
expressed as the follow equation: Table 3. Results of meta-analysis

where !! is the relative importance of the ""# input variable

on output variable; #! and #h are the number of input and
hidden neurons; $s are connection weights; the superscripts
%, h, and ' refer to input, hidden and output layers,
respectively; and subscripts ), *, and + refer to input,
hidden and output neurons, respectively.

Table 2. Parameters of ANN model

The performance of established ANN model was

depicted in Fig.1. The X-axis “Target” indicates the
III. Results and Discussion observed value which the model targets to reach, and the Y-
axis “Output” means the output of the predicted value from
In total, 64 experiments of 32 PPCPs treatment were
the model. Two lines in the graph were used to show the
included in the analysis, among which more data points were
success of the prediction. The solid line is the line that best
available in the CNT/TiO2 photocatalytic system (k = 29)
fits on the data of the scatter plot which is obtained by
compared with Graphene/TiO2 treatment (k = 8). The
regression analysis. In a perfect fit, when the predicted value
reported removal efficiency ranged from 54% to 100% with
equals to the observed value, the solid line would lay on the

179
dotted line. For the overall model performance, the carbonaceous/TiO2 system published in literatures from
correlation coefficient (R) is 0.94. 2010 to 2020, applying both statistical and modelling
method to characterize and model the process. The results
from meta-analysis found out the pooled average removal
efficiency at 60-minutes’ reaction time of PPCPs by
carbonaceous/TiO2 photocatalysis is 89.5%, and CNT/TiO2
composites have the best removal performance of 94.6%.
Artificial neural network with good prediction performance
was built based on both molecular related features and
process related parameters. A better ANN model may be
realized by combing quantum structural features of
molecular and process parameters, for prediction of multiple
PPCPs under different reaction conditions[7]. However,
model over-fitting is a problem needed to be discussed in the
future studies.

Acknowledgement
We are thankful to the Science and Technology Research
Partnership for Sustainable Development (SATREPS), the
Japan Science and Technology Agency (JST)/Japan
Fig. 1 Plot of the performance of the build ANN model. International Cooperation Agency (JICA) for their financial
(Target: the observed value; Output: the predicted value by support.
ANN model; R: correlation coefficient)
References
By Garson’s sensitivity analysis method, we obtained the [1] Daughton, C. G. & Ternes, T. A. Pharmaceuticals and personal
importance of each input variable (Table 4). There is no care products in the environment: Agents of subtle change?
particularly noticeable difference between the importance of Environ. Health Perspect. 107, 907–938 (1999).
each variable, but process related variables have more [2] Awfa, D., Ateia, M., Fujii, M., Johnson, M. S. & Yoshimura, C.
relative importance to the model output (RE) than molecular Photodegradation of pharmaceuticals and personal care products in
related variables, counting to 61.2% of the total importance. water treatment using carbonaceous-TiO2 composites: A critical
review of recent literature. Water Res. 142, 26–45 (2018).
[3] Gu, Y. et al. Adsorption and photocatalytic removal of
Table 4. Relative importance of input variables on RE
Ibuprofen by activated carbon impregnated with TiO2 by UV–Vis
monitoring. Chemosphere 217, 724–731 (2019).
[4] Xu, X. & Li, X. QSAR for photodegradation activity of
polycyclic aromatic hydrocarbons in aqueous systems. J. Ocean
Univ. China 13, 66–72 (2014).
[5] Mikolajewicz, N. & Komarova, S. V. Meta-analytic
methodology for basic research: A practical guide. Front. Physiol.
10, (2019).
[6] López, M. E., Rene, E. R., Boger, Z., Veiga, M. C. & Kennes,
C. Modelling the removal of volatile pollutants under transient
conditions in a two-stage bioreactor using artificial neural networks.
J. Hazard. Mater. 324, 100–109 (2017).
IV. Conclusion [7] Huang, X. et al. Mechanistic QSAR models for interpreting
degradation rates of sulfonamides in UV-photocatalysis systems.
Chemosphere 138, 183–189 (2015).
This study collected photocatalytic process parameters
and performance of PPCPs degradation in

180
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Application of Multivariate Techniques in the Evaluation of Spatial Surface Water Quality in an

Urbanized Floodplain Area in Cambodia

Vinhteang Kaing, Ilan Ich, Lieng Thoun, Chantharath Yos, Chan Rathborey, Chan Rathboren, Layheang Song,
Marith Mong, Chantha Oeurng, Ty Sok*

Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia

Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
*sokty@itc.edu.kh

Abstract: Lakes and rivers are important multi-usage components, particularly lakes are not only important sources of
surface water and livelihood for both rural and urban communities but they are also functioning as water retention receiving
stormwater and wastewater from the urban area. Located in the northern part of Phnom Penh capital city of Cambodia, Tamouk
Lake receives discharges from combined sewer systems by which a mixture of industrial wastewater, urban surface runoff,
domestic wastewater, and sewer deposits from the drainage system of Phnom Penh. The water quality of Tamouk Lake has
become more concern due to the pollutant load in wastewater discharge. This study aims to assess the spatial variation of
water quality parameters and to determine the main contamination sources in the Tamouk Lake by using multivariate
techniques. This study focuses on seven target water parameters (Total suspended solid (TSS), pH, Electrical conductivity
(EC), Dissolved oxygen (DO), Chemical Oxygen Demand (COD), Phosphate (PO 4), and Nitrate (NO3)) from 11 sampling
points. The analysis was employed individually of lake conditions; normal condition and receiving storm; for understanding
the effect of storm water to the water quality of the lake. The principle component analysis (PCA) was used to identify the
sources of contaminant that effected the lake water quality and cause degradation. In lake normal condition, the first three
principle componence explained 78.5% of the total variance in the water quality data set, while in receiving storm water
condition, the first two principle componence explained for 83.15%. Overall, the study showed that most of factors determine
the water quality of Tamouk lake is associated with the anthropogenic pollution sources from urban wastewater and biogenic
sources in the lake. The study suggested that the management of wastewater discharge from Phnom Penh City is required to
lower the accumulation of pollutants and minimize lake water quality degradation.
Keywords: floodplain urbanized area, multivariate techniques, principal component analysis, urban wastewater

I. Introduction phosphorus and nitrogen, hydrocarbons, heavy metals,

Lakes and rivers are important multi-usage components, endocrine disruptors, and organic matter [3]. It is becoming
particularly lakes are not only important sources of surface the pressure to waterbodies (lake and reservoirs located near
water and livelihood for both rural and urban communities the urban area) as they function as the storage in biological
but they are also functioning as water retention receiving treatment to collect the wastewater from the drainage system
stormwater and wastewater from the urban area. In urban [4]. Thus, the deterioration of lake water quality is
area, the water quality is such a key factor which is hard to unavoidable. The declining water quality in freshwater lakes
identify as its characteristic vary based on numerous factors is an increasing problem that threatens the ecosystem
such as street sweeping, stormwater, wastewater from services to the riparian communities, especially in
domestic and small to medium industrial activity [1, 2]. developing countries [5].
These pollutants were pathogenic microorganisms, Located in Cambodia, Tamouk Lake is in the northern

181
part of Capital Phnom Penh and has been selected as one site
of wastewater treatment by JICA since 2015 [6] and to be
constructed in years due to the fact that this lake is the largest
lake in the northern part of Phnom Penh and has the low
level of the population density in the area. Additionally, the
developing status in the area is still immature [6]. It
interesting to note that Tamouk Lake receiving discharges
from combined sewer systems, by which a mixture of
industrial wastewater, urban surface runoff, domestic
wastewater and sewer deposits are discharged into receiving
waters. Since the lake functions as the storage to receive
storm water and wastewater from the drainage system in the
northern part of Phnom Penh, the water quality in Tamouk
Lake has become more concerning due to the pollutant load Figure 1. Location map and sampling points of Tamouk
in wastewater discharge. Moreover, the study on water lake
quality of Tamouk Lake is normally looked over. 2.2. Water Quality Variables Selection
As mentioned about the advantage of urban lake, the The field monitoring was conducted two times: the pre-
water quality concerning and lacking of study, the study on storm event called ‘normal condition’ and post-storm event
water quality in the lake and identification major pollution call ‘receiving storm water’ at 11 sampling points (Figure1
source of lake is crucial required. Therefore, this study is to &Table 1). The field survey was conducted during the pre-
assess the spatial variation of water quality parameters (Total storm event on 28 May 2019. After that, the second field
suspended solid (TSS), pH, Electrical conductivity (EC), survey was conducted during the post-storm event on 7 June
Dissolved oxygen (DO), Chemical Oxygen Demand (COD), 2019. The 11 sampling points from the depth of 10 cm in
Phosphate (PO4), and Nitrate (NO3)) and to determination Tamouk Lake were established for both field surveys.
the main contamination sources in the Tamouk Lake, a
Table 1. Water quality variable and site location consider
floodplain Urbanized Area in Cambodia, by using
in Tamouk Lake
multivariate technique.
Date Variables Site location
II. Materials and Methods EC, DO, pH, COD, Inlet, outlet, and
28/05/2019
PO4, NO3, TSS middle of the lake
2.1. Sampling sites
EC, DO, pH, COD, Inlet, outlet, and
The study was conducted in Tamouk Lake (or Kob Srov 07/06/2019
PO4, NO3, TSS middle of the lake
Lake) situated in Prek Pnov district of Phnom Penh,
Cambodia. Tamouk Lake is the largest natural lake and the 2.4. Principal component analysis (PCA)
major reservoir of wastewater treatment in the northern area Principal Components Analysis (PCA) was used to
of Phnom Penh city. This lake is surrounded by 2 inlets, 2 provide information on the most meaningful parameters
outlets, and three water gates as shown in the map (Figure which describe the whole data set interpretation and data
1). As reported by [6], this lake covers 3270 ha and the depth reduction, and summarize the statistical correlation among
water is 3.0-4.5 m in the dry season and accumulate 2-3 m constituent in the water with minimum loss of original
more in the rainy season. Moreover, the Tamok sewage information [7]. PCA data are mainly reflected by the
system is a separate system that covers the treatment area of variance of data variables. The greater the variance, the more
6,019.2 ha with the population of around 481,000 information is included as measured by the cumulative
(population projection to 2035). The total length of trunk variance contribution ratio. PCA were performed on
sewer is 66.1 km (diameter from 200 mm to 1,650 mm), and standardized data with z-scale of mean zero and standard
the pumping stations were installed at nine locations, seven deviation of one in order to have an equal weight to every
of which were pumping stations were manhole type and two variable, this is for minimizing the influence of variable
were installed at the inlet of Tamouk Lake [6]. whose variance is large and adjusting for the disparity in the
variable sizes, in the units of measurement and renders the

182
data dimensionless [8-10]. conservation planning. It is affordable for us to take water
quality conservation measures to the condition with good
III. Results and Discussion
water quality while take positive counter measures to
3.1. Water quality characteristic condition with poor water quality.
From spatial location, pH decreased (7.7 to 9.6) at the
3.2. Identification of potential sources of monitored
inlet of the lake and slightly increase at the downstream as
variables
the middle of lake and outlet (9.2 to 9.5, 9.6 to 9.7). However,
The PCA was performed to understand the relationship
the DO level in the water is opposite with pH value due to
between the water quality variables of all sampling points
the increasing the anoxic in the surface runoff from the urban
and to identify their characteristics. Factor analysis of the
area at the receiving storm water. TSS is dramatically
present data set of Tamouk lake further reduced the
increased at the post-storm event due to the erosion and
contribution of less significant variables obtained from PCA.
sediment that washed from the road surface by the runoff.
The two lake conditions; normal condition and receiving
Additionally, EC is a bite increase in the lake at the post-
storm water; were individually examined to see the effect of
rainfall event due to the additional ion and salt that picked
storm water to the lake’s water quality. The PCs with the
from the road surface during the storm event. Although, EC
eigenvalue greater than 1 were extracted entire PCs of data
of sample 2 and 3 are very high at the pre-storm event due
set to be the retained PCs and the components with
to the level of ion and salinity in wastewater discharge.
eigenvalues lower than 1 were removed due to their low
Furthermore, the different level of PO4 and NO3 in the lake
significance [13]. The PC loadings of the water quality
due to the correlation of NO3 with PO4 [11, 12]. Then, COD
variables were categorized as ‘strong’, ‘moderate’, and
at the inlet is drop-down, but it rises a bite near the outlet at
‘weak’ corresponding to the absolute loading values of
the post-storm event. This comes from the positive
>0.75, 0.75–0.50 and 0.50–0.30, respectively [10].
correlation between COD and pH [12].
In lake normal condition, the first three PCs explain 78.5%
Table 2. Range of surface water quality parameters in
of the total variance in the water quality data set (Table 3).
Tamouk Lake
The first factor (PC1) has moderate negative loading on EC
Parameters Min Mean Max SD
pH 7.90 8.93 9.63 0.62 and COD and weak negative loading with PO4. This factor
DO 9.14 11.01 11.69 0.68 can be attributed to biogenic and anthropogenic from urban
Normal condition

EC 224.00 387.27 763.00 194.93 wastewater pollution sources. The second factor (PC2)
TSS 1.00 28.64 97.50 27.43 showed moderate negative loading on TSS and weak
COD 10.20 38.95 112.00 38.84 positive loading on pH. This factor could be associated with
PO4 2.90 6.26 14.50 4.01 the physical and chemical properties of the water, and to
natural weathering of the basin. These two originated
NO3 2.80 5.13 9.20 1.81
primarily from run-off with high load of solids from point
Receiving storm water

pH 7.75 8.72 9.65 0.81

sources of pollution like domestic areas, agricultural fields
DO 4.13 7.77 11.35 1.98
[14]. The third factor (PC3) showed strong positive loading
EC 231.00 377.91 567.00 120.61
on DO and weak positive loading on NO3. The PC3 should
TSS 10.00 59.01 152.00 49.11
attribute the chemical and biological process in the lake and
COD 9.01 22.32 49.00 15.83
can be associated the low level of dissolved organic matter
PO4 1.80 4.79 8.00 1.84
NO3 2.00 6.45 15.30 4.12 that lead to consume less amount of oxygen.
During receiving storm water condition, the first two PC
Description and characteristic values of 7 key indicators with eigenvalue>1 explained for 83.15% of the total
of surface water quality parameters in Tamouk Lake are variance (Table 3). The PC’1 has weak positive loading on
shown in Table 2. EC, COD, and TSS have larger standard pH and DO, and weak negative loading with EC and COD.
deviations, the Tamouk Lake water quality shows obvious This PC represents the physiochemical source of variability.
differences among the above 4 indicators both normal The PC’2 has strong positive loading with TSS and NO 3.
condition and receiving storm water. Thus, different These two elements originated primarily from run-off with
condition should be treated individually in the next stage of high load of solids and wastes from point sources of
water pollution control and water environmental pollution like domestic areas, agricultural fields [14]. For

183
Tamouk lake, the main point source should be from References
wastewater with storm discharge from the Phnom Penh City. 1. Samsudin MS, Juahir H, Zain SM, Adnan NH: Surface river
The results from PCA suggested that most of factor water quality interpretation using environmetric techniques:
Case study at Perlis River Basin, Malaysia. International
determine the water quality of Tamouk lake is associated
Journal of Environmental Protection 2011, 1(5):1-8.
with the anthropogenic pollution sources from urban 2. Vaze J, Chiew FH: Experimental study of pollutant
wastewater and its biogenic sources in the lake. accumulation on an urban road surface. Urban Water 2002,
Table 3. Factor loadings of principle components of 4(4):379-389.
3. Akpor O, Otohinoyi D, Olaolu D, Aderiye B: Pollutants in
individual water quality variables in normal condition and wastewater effluents: impacts and remediation processes.
receiving storm water of the Tamouk Lake International Journal of Environmental Research and Earth
Principle components Science 2014, 3(3):050-059.
Receiving storm 4. Verma M, Negandhi D: Valuing ecosystem services of
Variable Normal condition wetlands—a tool for effective policy formulation and poverty
water
PC1 PC2 PC3 PC’1 PC’2 alleviation. Hydrological sciences journal 2011, 56(8):1622-
1639.
pH 0.313 0.482 0.164 0.465 - 5. Ndungu J, Augustijn DC, Hulscher SJ, Fulanda B, Kitaka N,
DO - -0.310 0.732 0.425 0.229 Mathooko JM: A multivariate analysis of water quality in
Lake Naivasha, Kenya. Marine and freshwater research 2015,
EC -0.595 0.003 - -0.471 - 66(2):177-186.
TSS - -0.637 -0.332 -0.218 0.648 6. JICA: The Study on Drainage and Sewerage Improvement
Project in Phnom Penh Metropolitan Area. Final Report 2016,
COD -0.536 0.237 -0.261 -0.487 -0.146
II.
PO4 -0.445 0.359 0.311 -0.255 -0.160 7. Helena B, Pardo R, Vega M, Barrado E, Fernandez JM,
Fernandez L: Temporal evolution of groundwater
NO3 -0.229 -0.284 0.401 -0.177 0.687 composition in an alluvial aquifer (Pisuerga River, Spain) by
Eigen- principal component analysis. Water research 2000,
2.413 1.894 1.188 3.921 1.900
value 34(3):807-816.
PV(%) 34.47 27.05 16.97 56.01 27.14 8. Simeonov V, Stratis J, Samara C, Zachariadis G, Voutsa D,
Anthemidis A, Sofoniou M, Kouimtzis T: Assessment of the
CPV(%) 34.47 61.52 78.49 56.01 83.15 surface water quality in Northern Greece. Water research
Note: The variables with bold values of loading are variables 2003, 37(17):4119-4124.
which most explain or represent the principal components. 9. Liu W, Li X, Shen Z, Wang D, Wai O, Li Y: Multivariate
statistical study of heavy metal enrichment in sediments of the
IV. Conclusion Pearl River Estuary. Environmental pollution 2003,
The study assessed the spatial variation of water quality 121(3):377-388.
and identified the sources of contamination base on water 10. Liu C-W, Lin K-H, Kuo Y-M: Application of factor analysis
in the assessment of groundwater quality in a blackfoot
quality variables (TSS, pH, EC, DO, COD, PO4, and NO3) disease area in Taiwan. Science of the Total Environment 2003,
in the Tamouk Lake, a floodplain Urbanized Area in 313(1-3):77-89.
Cambodia. Multivariate technique included Pearson’s 11. Alsaqqar AS, Khudair BH, Hasan AA: Application of water
quality index and water suitability for drinking of the
correlation, PCA, and CA were applied to understand spatial
Euphrates river within Al-Anbar province, Iraq. Journal of
variation of lake water quality. The PCA was used to identify Engineering 2013, 19(12):1619-1633.
the sources or factor of contaminant that effected the 12. Tripathi B, Pandey R, Raghuvanshi D, Singh H, Pandey V,
Tamouk Lake water quality and cause degradation. Result Shukla D: Studies on the physico-chemical parameters and
correlation coefficient of the river Ganga at Holy Place
from PCA suggested that the lake water quality was control Shringverpur, Allahabad. IOSR Journal of Environmental
by the anthropogenic pollution sources (urban waste water), Science, Toxicology and Food Technology 2014, 8(10):29-36.
and biogenic sources in the lake. The study suggests that the 13. Kim J-O, Mueller CW: Factor analysis: Statistical methods
and practical issues: sage; 1978.
management of wastewater discharge from Phnom Penh
14. Gazzaz NM, Yusoff MK, Ramli MF, Aris AZ, Juahir H:
City is required to lower the accumulation of pollutants and Characterization of spatial patterns in river water quality using
minimize environmental degradation. chemometric pattern recognition techniques. Marine
Pollution Bulletin 2012, 64(4):688-698.
Acknowledgement: We are thankful to the Agence
Française de Développement (AFD) No. CKH 1236 02P for
financial support of the study.

184
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Effect of Water and Land Based Villages on Phosphorus Dynamics in a Lake-Floodplain

System, Tonle Sap Lake

Vouchlay Theng1*, Hashimoto Kana1, Sovannara Uk1, Sophanna Ly1, Tanaka Tomohiro2, Yoshioka Hidekazu3,
Yoshimura Chihiro1
1
Department of Civil and Environmental Engineering, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku,
Tokyo, 152-8552, Japan
2
Graduate School of Global Environmental Studies, Kyoto University, Kyotodaigaku-Katsura, Nishikyo-ku, Kyoto, 615-
8540, Japan
3
Graduate School of Natural Science and Technology, Shimane University, 1060 Nishikawatsu-cho, Matsue-shi, Shimane,
690-8504, Japan
*Corresponding author: vouchlaytheng@gmail.com

Abstract
The objective of this study is to develop a phosphorus (P) dynamic model by integrating the load from water and land
based villages at Tonle Sap Lake’s floodplain to investigate spatiotemporal changes of total P (TP) dynamics in the lake. A
two-dimensional local inertial model describing hydrostatic shallow surface water dynamics was applied to simulate
horizontal hydrodynamics of the lake. The processes such as advection-dispersion and internal and external loadings of TP
were considered in the developed model. The model application result in 1999-2003 showed that at least 240 villages
permanently stayed in the water from September 1999 to December 2003. The total input of TP from the tributary rivers was
estimated to be 137 ton/month in the dry season and 692 ton/month in the rainy season. On average, the estimated TP loads
from the villages were approximately 21 ton/month in the dry season and 47 ton/month in the rainy season, equivalent to 15%
and 7% of the total TP load from the tributaries, respectively. More than 83% of the lake area was in mesotrophic state and
some locations reached eutrophic state. The locations around the villages reached eutrophic state at the beginning of flooding
periods due to the release of TP mass from land to water. Increasing TP load from the villages will increase the eutrophication
problem in the rainy season which should be considered for water quality management.

Keywords: Floodplain, phosphorus dynamics, trophic state, water and land based villages, water quality model

I. Introduction effect of these villages to P dynamics in TSL should be

investigated to understand the anthropogenic impact on the
Phosphorous (P) is a limiting nutrient in most of the
ecosystem and solve such water environmental problems.
freshwater lakes [1]. Excessive phosphorus could change
Many process-based P models were developed to
tropic states of the lakes to be eutrophic, which is considered
estimate total P (TP) dynamics for freshwater lakes.
as a major stress for lake ecosystems. Tonle Sap Lake (TSL)
However, those models have not been applied to a tropical
is one of such examples being excessive nutrients and likely
lake- floodplain system (i.e., TSL), which has a seasonal
P-limited [2]. More than 1.7 million people living in TSL’s
reversal flow and villages on and around the lakes. The
ecosystem depend directly on the lake’s resources and
objective of this study is, therefore, to develop a process-
directly discard wastes into the lake [3]. Raw domestic
based dynamic model by integrating the load from the
wastewater generally contains high concentration of P,
villages to analyze TP dynamics in TSL.
ranging between 0.2–2.0 g/day/capita [4, 5]. Therefore, the

185
II. Methodology
2.1. Description of study area
TSL is a tropical lake-floodplain system in Cambodia.
During the dry season, the lake’s surface area is 2500 km2
and expands up to 15 000 km2 in the rainy season, extending
the lake over the vast floodplain [6]. The villages on and
around TSL were categorized into three types: land based,
land-water based, and floating (water based) [7]. The TP
load from these villages were included in this study.

2.2. Mathematical model

The developed model is governed by horizontal
advection-dispersion and processes of sedimentation, Fig. 1. The framework of the P dynamic model
internal loading, and input of TP from tributaries,
atmosphere, and the villages on the lake in two dimensions. Geomorphology-Based Hydrological Model and 1D
The dynamic mass balance of TP can be expressed as hydraulic model Mike 11. Hydrodynamic data was obtained
from the two-dimensional local inertial model, 2D-LIE [10].
𝜕(ℎ𝐶) 𝜕 𝜕𝐶 The observed TP concentration was obtained from the
= (𝐷𝑥 ℎ − 𝑣𝑥 ℎ𝐶) Mekong River Commission’s database and Water Utilisation
𝜕𝑡 𝜕𝑥 𝜕𝑥
(Eq.1)
𝜕 𝜕𝐶 Programme. The villages and population data in 2008 were
+ (𝐷𝑦 ℎ − 𝑣𝑦 ℎ𝐶) − 𝑣𝑠𝑒𝑑 𝐶 + 𝑘𝑖 + 𝑘𝑎
𝜕𝑦 𝜕𝑦 obtained from National Institute of Statistic [11]. Then, the
population of villages were converted based on the annual
where, C is the TP concentration; 𝑡 is the time; vx and vy are population growth rate of 1.54%/year. Most of floating
the water velocity in x- and y-directions, respectively; h is villages moved to nearby flooded forest. Therefore, we
the water depth; vsed is the settling velocity; ka is assumed that each village area stayed within 1 km2. The TP
atmospheric loading rate; ki is the internal loading rate; Dx load from the villages was assumed to be 0.9 g/day/capita.
=0.6hxvx* and Dy = 0.6hyvy* are the dispersion coefficients All TP mass from the villages was assumed to accumulate
in x- and y-directions, respectively [8]; hx and hy are the on land whenever dried and be immediately released to the
respective water depths in x- and y-directions. The friction lake water whenever inundated. The spatial resolution was
velocities, vx* and vy* are calculated by the Manning’s set to be 500 m.
equation. The internal loading rate is the total loading rate
from resuspension and diffusional flux of sediment pore III. Results and Discussions
water phosphorus to overlying water column.
3.1. Numerical accuracy and model calibration
As shown in the conceptual framework (Fig. 1), the
The evaluation of the numerical accuracy of advection-
major components of the model are numerical discretization,
dispersion was performed against 2D benchmark tests of the
computational domain, and input data. A finite volume
rotating Gaussian pulse where an exact solution is available.
method is employed and the mass conservation is locally
The mass conservation rates of the tests were higher than
preserved by using the total variation diminishing (TVD)
99%. The preliminary test with the case of TSL also
scheme in spatial discretization and 1st-order Euler-forward
confirmed reasonable mass conservation and an assumed
method in temporal discretization. The monotonic upstream-
bell shape distribution of the concentration was not
centered scheme for conservation laws approach is applied
fluctuated after 1-month transport. The accumulated relative
into the TVD scheme as it is one of the most successful high-
error was ranged from 0% to 1% in mass from 1999 to 2003.
resolution schemes for hyperbolic conservation laws, which
The model was manually calibrated for the period from
has second-order accuracy in space [9].
September 1999 to December 2001 at Kampong Loung
Daily water flow discharges from the boundary tributary
(KGL1) and validated for the period from January 2002 to
rivers in TSL were obtained from hydrological models,
December 2003 at 2 locations of Kampong Loung (KGL1

186
and KGL2), Peam Bang (PBA), and 3 locations of Phnom a
Krom (PNK1, PNK2, and PNK3) (Fig. 2). The root mean
square error of TP in the calibration was 17 µg/L (13 µg/L
in the validation) while the observed TP concentration
ranged from 5 µg/L to 69 µg/L. The calibrated settling
velocity was 0.4 m/day, and the summation of internal
loading and atmospheric loading was 8.0 mg/m2/day.

3.3. Phosphorus dynamics in Tonle Sap Lake

The total input of TP from the tributary rivers was
estimated to be 137 ton/month in the dry season and 692
ton/month in the rainy season. On average, only 14
ton/month of TP flowed out from TSL during the dry season
and 28 ton/month during the rainy season. During the study
periods at least 240 villages stayed permanently on the water.
The average numbers of villages on the water and inundated b
were 405 and 608 in the dry and the rainy seasons,
respectively. The estimated TP loads from the villages into
the lake were approximately 21 ton/month in the dry season
and 47 ton/month in the rainy season, equivalent to 15% and
7% of the total TP load from the tributaries, respectively.
Most of the time, the TP settling was higher than internal and
atmospheric loading. The total of net settling and
atmospheric flux was 350 ton/month in the dry season and
509 ton/month in the rainy season.
On average, the TP mass stored in the lake was 492 ton
and 828 ton in the dry and the rainy seasons, respectively.
The water surface was observed to be mostly mesotrophic
(11 µg/L ≤ TP ≤ 30 µg/L), and its surface coverage was 86%
in the dry season and 83% in the rainy season. The
concentration of TP in the lake in the dry season was almost Fig. 2. Distribution of simulated TP concentration in TSL
higher than that in the rainy season (Figs. 2. a & b). The during: a) the dry season in March 15, b) the rainy
eutrophic area expanded from 4% in November to 9% in season in October 15, 2001.
May and contracted during the rainy season. This shift was downstream part (e.g., Fig. 2. a). The eutrophic area was 7%
possibly caused by the low net settling flux in the low-water of the lake area in March 2001. In July, TP influx via TSR
period and the high concentration of external input from the quickly flowed into the middle of the lake causing
tributaries in such a period. Furthermore, approximately 49% eutrophication along that way. In the late rainy season, the
of the wet cells were dried out during the dry season which lake water was diluted by rainwater and lowered the
contained TP mass about 30 ton or 6% of total TP mass and concentration of most tributaries input. The eutrophic area
was released back to the lake due to the annual flood pulse. in October 2001 was 3% of the lake area due to high TP
Around the villages, water became eutrophic (TP > 30 µg/L) concentration from few tributaries and villages (Fig. 2. b).
at the beginning of flooding periods due to the release at
accumulated TP mass on the land to water. Then, the lake 3.4. Impact assessment of water and land based villages
was recovered by sedimentation process. on eutrophication
During the dry season, the model simulated Population in Cambodia increases rapidly. Sewages are
eutrophication near the outlets of the tributaries and at the usually discharged into the water sources without any
bank line along KGL1, PBA, and the outlet of the TSL treatment due to the lack of wastewater treatment facility [3].

187
 10
No village References
8 2020 [1] Schindler, D.W., Hecky, R.E., Findlay, D.L., Stainton,
2040 M.P., Parker, B.R., Paterson, M.J., Beaty, K.G., Lyng,
Eutrophic area increase (%)

6 2060
M., Kasian, S.E.M., 2008. Eutrophication of lakes
4 cannot be controlled by reducing nitrogen input:
Results of a 37-year whole-ecosystem experiment.
2
Proceedings of the National Academy of Sciences of
0 the United States of America 105, 11254–11258.
[2] Burnett, W.C., Wattayakorn, G., Supcharoen, R.,
-2
Supcharoen, R., Sioudom, K., Kum, V., Chanyotha,
-4 S., Kritsananuwat, R., 2017. Groundwater discharge
and phosphorus dynamics in a flood-pulse system:
-6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Tonle Sap Lake, Cambodia. Journal of Hydrology
549, 79–91.
Fig. 3. Effect of water and land based villages on [3] Ung, P., Peng, C., Yuk, S., Tan, R., Ann, V., Miyanaga,
eutrophication area compared to the baseline of TSL in K., Tanji, Y., 2019. Dynamics of bacterial community
2000 in Tonle Sap Lake, a large tropical flood-pulse system
Thus, the amount of TP load from the villages will also in Southeast Asia. Science of the Total Environment
664, 414–423.
increase. By assuming the population growth rate keeps
[4] Okada, M., Sudo, R., 1986. Per Capita Loadings of
constant, 1.54%/year, the impact of the villages in 2020, Domestic Wastewater. Research Report from the
2040, 2060; and no village loading on eutrophic area is National Institute for Environment Studies 95, 7-20.
found by comparing to the baseline of TSL in 2000 using [5] Mesdaghinia, A., Nasseri, S., Mahvi, A.H., Tashauoei,
model calibration result (Fig. 3). The results show that the H.R., Hadi, M., 2015. The estimation of per capita
villages are highly effect on eutrophication in rainy season loadings of domestic wastewater in Tehran. Journal
due to the inundation at the lake flood-plain area. The of Environmental Health Science & Engineering 13,
1–9.
highest increases of eutrophic area are 2%, 5%, and 10% in
[6] Keskinen, K., 2006. The Lake with Floating Villages :
October in 2020, 2040, and 2060, respectively. However, if Socio- economic Analysis of the Tonle Sap Lake.
there is no TP load from the villages (e.g., people move to Water Resources Development 22, 463-480.
the land or the wastewater treatment facility is developed), [7] Binaya, R.S., Pham, N.B., 2020. Environmental
the eutrophic area will be reduced from 1% to 5%. Changes in Tonle Sap Lake and Its Floodplain: Status
and Policy Recommendations. Institute for Global
IV. Conclusion Environmental Strategies, Tokyo Institute of
Technology and Institute of Technology of Cambodia.
The developed P dynamic model is useful for describing [8] Huang, L., Fang, H., He, G., Jiang, H., Wang, C.,
mass balance of TP processes, the spatiotemporal changes of 2016. Effects of internal loading on phosphorus
P, and the effect of the villages on eutrophication in TSL. distribution in the Taihu Lake driven by wind waves
and lake currents. Environmental Pollution 219, 760–
However, the biological process of phytoplankton uptake
773.
and release and speciation of P (e.g., dissolved and [9] Bai, F., Yang, Z., Zhou, W., 2017. Study of total
particulate P) were not explicitly included. The uncertainty variation diminishing (TVD) slope limiters in dam-
of data should be considered in further work. break flow simulation. Water Science and
Engineering 11, 68–74.
Acknowledgement [10] Tanaka, T., Yoshioka, H., Siev, S., Fujii, H., Fujihara,
Y., Hoshikawa, K., Ly, S., Yoshimura, C., 2018. An
We are thankful to the Science and Technology Research integrated hydrological-hydraulic model for
Partnership for Sustainable Development (SATREPS), the simulating surface water flows of a shallow lake
Japan Science and Technology Agency /Japan International surrounded by large floodplains. Water (Switzerland)
10, 1–23.
Cooperation Agency for their financial support, and the [11] National Insitute of Statistics (NIS), 2010. Census
Mekong River Commission and Water Utilisation Map Layer and Databases 2008. National Institute of
Programme of Finland Component Project for the provision Statistic, Phnom Penh, Cambodia.
of the observed data of phosphorus in Tonle Sap Lake.

188
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Groundwater Quality Assessment in the Floodplain Area around the Tonle Sap Lake,
Cambodia

Bunhuot Ruos1, Ratana Kheang1, Sreyleang Ya2, Vuthy Chork2, Kong Chhuon2, Ratha Doung2, Ratino Sith2,
Sokly Siev3, Boreborey Ty3, Khy Eam Eang2*

1
Master of UWE, Graduate School, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86,
Phnom Penh, Cambodia
2
Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia, Russian Federation B
lvd., P.O. Box 86, Phnom Penh, Cambodia
3
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd.,
P.O. Box 86, Phnom Penh, Cambodia

*Corresponding author: khyeam_eang@yahoo.com

Abstract
This study attempted to evaluate the groundwater quality in the floodplain area around the Tonle Sap lake. Thirty-six and
forty well water samples were collected from the study area in the wet and dry season, respectively and analyzed for major
physicochemical properties. As a result, temperature, pH, TDS (total dissolved solids), and EC (electrical conductivity) varied
considerably between both seasons and among the sampling sites. The major concentrations of cations and anions were in the
respective order of Ca > Na+K > Mg and HCO3 > SO4 > Cl > NO3. Three major hydro-chemical facies (Ca-HCO3, Ca-Cl, and
Na-HCO3) were identified using a Piper trilinear diagram. The value of pH, K, Mn, and Fe, nearly 50% of water samples were
unsuitable for drinking purposes within both seasons, except Fe in the dry season, probably due to high precipitation of iron-
hydroxide. The total hardness (TH) indicated that the well water more than 50% were defined as soft water for both seasons.
Residual sodium carbonate (RSC) revealed that 40% samples of dry season were unsuitable for irrigation purposes. However,
Wilcox and USSL (US salinity laboratory) classified that the groundwater was suitable for irrigation. The overall quality of
groundwater in both seasons could be considered as moderately higher than drinking standard, but adequate for irrigation.

Keywords: Drinking water, Floodplain area, Groundwater quality assessment, Irrigation purpose

I. Introduction flood season. This unique phenomenon could change the

chemical distribution in groundwater over time. It is very
Cambodia groundwater regarded as a widely available
beneficial to be known all these mentioned problems.
supplemental source to support 53% of Cambodian
Therefore, this study aims to carry out the evaluation of
households in the dry season [1] . Until now, groundwater
groundwater quality in floodplain area of Tonle Sap lake
quality concern is still being doubtful for Cambodian people
which focus on the current status of water quality and the
particularly, the part of floodplain area around the Tonle Sap
suitability in drinking and irrigation purpose.
lake. The groundwater from this area is being used without
knowing whether it satisfied with drinking and irrigation II. Materials and Methods
standards. Some researches were conducted in this area, but
2.1. Sampling sites
it was rarely related to groundwater quality. Additionally, the
water depth of the lake varied seasonally with an average The present study was conducted in floodplain area
depth less than 2 m in dry season and increased to 10 m in around the Tonle Sap lake as shown in Fig. 1. Thirty-six and

189
forty samples from selected locations were collected in wet III. Results and Discussion
season (August, 2019) and dry season (December,2019),
3.1. Current status of water quality
respectively.
The pH ranged from 4.12 to 7.65 in rainy season and 3.1
to 7.86 in dry season. The downing pH would be result in
buffering calcium minerals but high in dissolved carbon
dioxide (H2CO3) [9] . EC has reached 3.6 mS/cm in dry
season. Higher EC when its depth downs to 30m and pH
greater than 8. Temperature in both seasons ranges from
28oC to 32oC excepting the dry season in Battambang (35oC).
Higher temperatures can reduce oxygen and dissolve more
minerals to higher electrical conductivity. The pH, EC, and
Temperature are provided in Table 2.
The platform hydro-chemical facies use to describe
characteristic chemical composition, Table 1. It is observed
that water samples in both seasons are influenced by cation:
Ca > Na+K > Mg and anion: HCO3 > SO4, > Cl but, HCO3
Fig. 1. Study area and sampling site increased to 75%. The majority of the water samples fall in
calcium-bicarbonate type (Ca-HCO3) and calcium-chloride
2.2. Experimental set-up type (Ca-Cl) field, in rainy season. In rainy, it is apparent
that the total hydrochemistry is dominated by bicarbonate
The physicochemical parameters included pH,
(HCO3). In the dry season, the majority of waters defined as
temperature, TDS, EC were analyzed in-situ by using Multi-
Ca-HCO3 type and following by sodium-bicarbonate type
EXO Sondes. Seven of major ions: Ca2+, Mg2+, K+, Na+, Cl-,
(Na-HCO3).
NO3-, and SO42- were analyzed by ion chromatography (IC)
while, six heavy metals such as Mn, Fe, Cu, Pb, Cd, and Cr Table 1. Result from piper diagram
were analyzed via atomic adsorption spectrophotometry
% Sample
(AAS). IC and AAS analyses were conducted at SATREPs Ions
Rainy (34 Samples) Dry (40 samples)
laboratory and chemical laboratory of Institute of
Ca2+ 33% 34%
Technology of Cambodia (ITC).
Mg2+ 23% 20.5%
2.3. Analytical methods Na++ K+ 32% 30.5%
No dominant 12% 15%
To study on current status of water quality, three
HCO3- 56% 75%
parameters such as pH, EC, and temperature were carried
SO42- 20% 10%
out by using Box and Whisker plot while ion components Cl- 15% 10%
were plotted in piper diagram [2] . No dominant 9% 5%
Instantly, to assess to drinking water this study has used Ca-HCO3 Ca-HCO3
the current data to compare with Cambodia Drinking Water Water Type Ca-Cl Na-HCO3
Standard (DWS) by Ministry of Industry, Mining and
3.3. Suitability in drinking and irrigation purpose
Energies and World Health Organization [3] [4] . The total
hardness (TH) was classified for domestic purpose based on The results in Table 2 showed that EC, TDS, pH, K, Fe,
Environmental Protection Agency (1976) [5] . In terms of and Mn are slightly out of the standard range for both
irrigation purpose, sodium adsorption ratio (SAR), residue seasons except for Fe in the dry season. The pH is not
sodium carbonate (RSC), and sodium percentage (%Na) desirable for drinking when smaller than 6.5. Fe within 50%
were calculated based on US salinity laboratory (USSL), is detected more than the permissible limit (PL) of 0.3 mg/l
Food Agriculture Organization (FAO), and Wilcox (1948), in the rainy period, while Mn also exceeded the PL more
respectively [6] -[8] . than 40% in both seasons. K+ has reached more than a
thousand mg/l which surpassed the permissible limit of 12

190
mg/l. The highest K would be provoked by the leaching from suitable for irrigation and agriculture. RSC greater than
fertilizer use in agriculture. Some ions in well water such 5meq/l are considered harmful to the growth of plants. In
NO3, SO4, HCO3 also found to overtake the PL with a few Fig. 4, it is revealed that most samples low in sodium and
sites but HCO3 is higher in the dry season. TH showed that high in SAR. Most sample is located in C1S1, C2S1, and
3% in rainy and none in dry samples, which are considered C3S1 which is suitable for all crops. The classification
as very hard (Table 4), exceeding 300 mg/l and may lead to Wilcox 1948 is applied for the study as given in Fig. 5.
the incidence of parental mortality and some types of cancer. Samples ranged from excellent to poor category. The graph
showed that 4 samples in the rainy and 1 sample in the dry
The RSC stretches from -2.8 to 12.6 in rainy season and
are unsuitable while 6 samples from both seasons are still
-2.5 to 21.5 in dry season (Table 3). Based on RSC, water
doubtful to unsuitable condition. However, samples are most
can be classified as safe, marginally suitable, and unsuitable.
likely in good condition based on zone 1 and zone 2.
It is observed that RSC was 40% in dry season and was not
Table 2. Ranges of physicochemical parameters and their comparison with drinking water standards of MIME (2004) and
WHO (2017)
Range Permissible limit (PL) Percentage of sample exceeded PL
Parameter
Rainy Dry MME, WHO Rainy (n=34) Dry (n=40)
pH 4.12-7.65 3.1-7.86 6.5-8.5 41% 40%
EC 36-3697 37-3032 1500 21% 15%
TDS 21-2209 23-1826 800 21% 18%
+
K 0-1247 0-1108 12 35% 40%
HCO3- 0-827 0-1396 600 3% 20%
-
NO3 0-146 0-94 50 6% 5%
SO42- 0-503 0-296 250 6% 5%
Mn 0-2.23 0-2.7 0.1 41% 45%
Fe 0-23.92 0-0.31 0.3 50% 2.5%
*Ca, Mg, Na, Cl are all under permissible limit (PL), while Cu, Pb, Cr, Cd are under detected limit
Table 3. Classification of well water for agricultural purposes

Sample Range % of Sample

Parameter Range Classification
Rainy Dry Rainy (n=34) Dry (n=40)
<1.25 Good (G) 67.5% 60%
RSC
-2.8-12.6 -2.5-21.5 1.25-2.5 Medium (M) 15% 0%
(meq/l)
>2.5 Bad (B) 17.5% 40%

Table 4. Classification of groundwater in term of hardness

Sample Range % of Sample

Parameter Range Classification
Rainy Dry Rainy (n=34) Dry (n=40)
<75 Soft 53% 55%
0.8-307 0-286 75-150 Moderated hard 18% 25%
Hardness (mg/l)
150-300 Hard 26% 20%
>300 Very hard 3% 0%

191
 S4 30
Acknowledgement
28 EC SAR
C1: Low S1: Low
We are thankful to the Science and Technology Research
26
24
C2: Medium S2: Medium Partnership for Sustainable Development (SATREPS), the
S3

C3: High S3: High

22 Japan Science and Technology Agency (JST)/Japan
C4: Very High S4: Very High
Sodium Hazard SAR (meq/l)

20
18 Rainy Dry
International Cooperation Agency (JICA) for their financial
16 support.
S2

14
12
10
References
8
6
[1] Sok, S. (2011). Groundwater Research in Cambodia.
4 https://archive.iges.or.jp/en/natural-
S1

2 resource/groundwater/PDF/activity20110602/S1-
0 3_Mr.Sok-Sophally_GW_Cambodia.pdf. Accessed
0 500 1000 1500 2000 2500 3000 3500 4000
15 December 2020.
C1 C2 C3 C4 [2] Piper, A. M. (1944). A graphic procedure in the
Salinity Hazard EC (µS/cm) geochemical interpretation of water‐analyses. Eos,
Transactions American Geophysical Union, 25(6),
Fig. 4. Diagram rating of well water samples in relation 914–928. https://doi.org/10.1029/TR025i006p00914
to salinity hazard and sodium hazard [3] MIME. (2004). Kingdom of Cambodia Drinking
Water Quality Standards. In Drinking Water Quality
Standards.
100 [4] WHO. (2017). Guideline for Drinking-water Quality
90 Rainy Dry (4th Editio). Word Health Organization.
80
[5] EPA, U. S. E. P. A. (1976). Quality Criteria for Water.
Percentage Na (%)

70 1 Excellent
(3) 2 Good Office of Water Planning and Standards.
60
(1) 3 Unsuitable [6] Ayers, R. S., & Westcot, D. W. (1976). Water Quality
50 (5) 4 Doubtful for Agriculture. FAO Irrigation and Drainage Paper
40 (4) 5 High risk
30
29 Rev.1. In FAO Irrigation and Drainage Paper: Vol.
20
(2) No.29.
10 [7] USSL. (1954). Diagnosis and Improvement of Saline
0 and Alkaline Soils. Soil Science Society of America
0 1000 2000 3000 4000 Journal, 18(3), 348.
EC (µS/cm) https://doi.org/10.2136/sssaj1954.036159950018000
30032x
Fig. 5. Rating of well water samples on the basis of [8] Wilcox, L. V. (1948). Classification and use of
electrical conductivity and percent sodium irrigation water. Agricultural Circular No.
969,Washington, DC: USDA., 969, 1–19.
IV. Conclusion https://doi.org/USDA
[9] Appelo, C. A. . and D. P. (2005). Geochemistry,
Higher values of EC, TDS, Mn, and Fe were observed in Groundwater and Pollution, 2nd Edition. In A.A.
rainy season while higher HCO3- was recorded in dry season. BALKEMA PUBLISHERS.
The groundwater quality of the study area had a principal
issue of low pH. The water should be treated before
consumption such concern parameters of HCO3, Fe, Mn, and
K while the salt water should be checked before irrigated on
soil or plants. For a better interpretation and clarification on
geochemistry controlling the well water quality, Saturation
Index (SI) may be needed and geological formation of the
well profile are one of the important information to be
suggested for the future study.

192
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Analytical Studies on Water Quality Index of Prek Te River,

A Mekong River Tributary in Cambodia

Kong Chhuon 1,*, Alpy Math 1 , Hua Lin 2, Yu Guo 2

1
Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
College of Environmental Science and Engineering, Guilin University of Technology, Guilin 541004, China
* Corresponding author: chhuon.k@itc.edu.kh

Abstract

Prek Te river is a tributary of Mekong river in Cambodia. It is of particular importance to start studying on this river water
quality because of the concern on the impact due to the agricultural development, land cover change and water uses inside its
catchment. Water quality is needed to evaluate the physical, chemical, biological state. The Water Quality Index (WQI)
reduces the number of parameters to a simple expression in order to facilitate the interpretation of the data. This study is to
analysis Prek Te river water quality in 2018 and 2020 using WQI method to express the average quality of water at a time,
based on analytical values of physico-chemical parameters. The 13 water samples were collected to analyze physico-chemical
and heavy metal. In dry season of 2018, the WQI was 128 which made it fell into the moderate water quality, but the value
decreased to only 88 in wet season of 2020 which corresponded to the good quality. The high value of WQI is influenced
mainly from the high concentration of TDS and COD. TDS and COD were found to be higher with 153mg/L and 42mg/L
respectively, at the tributary of Prek Te. Cr, Fe, Mn, and Arsenic were found to be below the standard WHO. The pesticide is
also analyzed in this study at five sites but it turned out only detectable of some molecules. It can be concluded that, water in
Prek Te has good quality in wet season, while moderate quality in dry season. This study showed the current situation of water
quality in Prek Te which will be served as the baseline for future impact evaluation. It is recommended to do regularly
monitoring for identifying any changes and further study is needed.
Keywords: Prek Te, WQI, physico-chemical, heavy metal, pesticide

I. Introduction domestic and agricultural purposes. The water quality from

the rivers has considerable importance for the reason that
A tributary of the Mekong River in Cambodia, Prek Te
these water resources are generally used for multiple matters.
River is draining water via valleys and falls from the
Typically, water quality is determined by comparing the
mountainous area in Mundul Kiri province and crossing
physical and chemical characteristics of a water sample with
alluvial plains before flowing into the Mekong River in
water quality guidelines or standards. One of the simplest
Kratie province with a total area of 11.094 km2 with 81.42%
methods to assess water quality conditions is by using a
of the population are farmers. The crops such as rice, corn,
water quality index. It is a tool that provides meaningful
and bean are along the Mekong river of the southwest
summaries of water quality data that are useful to both
segment and agro-industry plantations such as rubber,
technical and non-technical individuals interested in water
pepper, and cashew nuts are at the northeast plateau region.
quality results. This study is to analysis Prek Te river water
Prek Te River is the main water source for farmers in both
quality in 2018 and 2020 using WQI method to express the

193
average quality of water at a time, based on analytical values water for drinking purposes. The weight was assigned
of physico-chemical parameters. between 1 and 5 based on their relative significance in the
II. Materials and Methods water quality. The overall Water Quality Index was
calculated by aggregating the quality rating (Q) with unit
2.1. Sampling sites
weight (Wi):
A tributary of Mekong River in Cambodia, Prek Te ∑ 𝑞𝑖 𝑤𝑖
watershed (Fig 1) is situated in Kratie province which lays Overall WQI = (Eq.2)
∑ 𝑤𝑖
from the eastern plateau in Mondulkiri province at an
altitude of around 900 m to the Mekong River flood plain
Principle Component Analysis: The PCA is a
south of Kratie town at an elevation less than 14 m. Its basin
descending dimension algorithm, in which the complex raw
area is 4,372 km2 which equals to 2.41% of Cambodia total
data are replaced by several unrelated variables (Wu et al.
land area.
2014). Only the components with the high eigenvalues (> 1)
were extracted (Wang et al. 2017). All the results are
conducted by using the analysis unit in software Origin
2018 .

III. Results and Discussion

Concentrations of Water Quality Parameter
pH is a crucial indicator that can be used for assessing
water quality and degree of contamination in water bodies.
The variation of pH values in 2018 showed not different
Fig. 1. Map of Prek Te river catchment of the study
area values between site 1, 2, and 4. However, the pH value in
which higher average values of 8.10 are obtained at site 5
2.2. Experimental set-up during the wet season, whereas the lower average values of
On-site Analysis: Temperature, pH, turbidity, EC, DO, 7.22 are obtained at site 1. In 2020, the pH values recorded
and TDS were carried out at the site of sample collection by in the Prek Te river from 6.95 to 7.21.
using EXO2 Multiparameter Sonde YSI. The polyethylene EC is another useful water quality indicator observed by
bottles were rinsed with sampled water at least three times this study. Except for site 4, which had values of the high
before being filled and stored. conductivity of 573.35 µS/cm and 270.56 µS/cm during in
Laboratory Analysis: The heavy metal (Cr, Mn, Fe and 2018 and 2020, respectively, all EC levels for site 1, 2, 3,
cyanide), Alkalinity, COD, As, TS and pesticides will be and 5 were recorded to fall outside the recommended range
analyzed in the laboratory at the Institute of Technology of of the WHO standard.
Cambodia. The quality of water will be classified following DO is one of the most important indicators of water
the Cambodia water pollution standards and the WHO quality. To maintain acceptable or good water quality. The
standards. highest DO value in the Prek Te River was observed at site
2.3. Analytical methods 3 (7.78 mg/L) at the end of the wet season in 2018, while the
Water Quality Index (WQI): The WQI was calculated lowest was observed at site 4 (4.17 mg/L). In 2020, DO data
using the Weighted Arithmetic Index method. The quality along the river that on average concentrations tended to be
rating scale for each parameter qi was calculated by using higher in the upper stream of the river, the highest DO
this expression: concentration was recorded in site 4 at 7.62 mg/L.
Turbidity measures the relative clarity of the water by the
Ci
q1 = ( ) × 100 (Eq.1) presence of organic and mineral suspended particles and
Si color producing substances. The mean turbidity readings of
the samples at all sampling sites were in the range of 10.51
Each of the 10 parameters was assigned a weight (wi) to 78.02 NTU. The results indicate that the turbidity of all
according to its relative importance in the overall quality of the samples studied was higher the maximum standard limit

194
of 5 NTU. sites have only detectable some molecules.
TDS is usually estimated by EC and there is a strong Water Quality Index (WQI)
relationship between EC and TDS. The values found from The WQI in 2018 and 2020 has been calculated which
the Prek Te water samples are all within the maximum limit has been applied also for 5 sampling sites, along the Prek Te
of 800 mg/L. The TDS ranged from 35 mg/L at site 5 to 153 River. For computing WQI three steps are followed. In the
mg/L at site 4 during the end of dry season. These higher first step, each of the 11 parameters (pH, EC, turbidity, TDS,
TDS values could be due to the natural weathering of certain TS, Alkalinity, DO, COD, Fe, Cr, and Mn) has been assigned
sedimentary rocks or a certain anthropogenic source, e.g., a weight (wi) according to its relative importance in the
irrigation discharge, domestic effluents, and sewage effluent. quality of water for drinking purposes. In 2018, except for
TS, the TS concentrations in analyzed samples ranged site 4, located on the upstream, the water quality status is
from 54 mg/L to 186 mg/L during the end of dry season in poor quality (128.18), which are influenced by the nutrients,
2020. The TS values of all the water samples were far below as a result of the agricultural practices, municipal and
the prescribed limit (500 mg/L) as per WHO standards. industrial, manure from farms. In Fig. 2 the five study sites
Alkalinity refers to the capability of water to resist located in the Prek Te River, four study sites were rated as
changes in pH. Carbonate and bicarbonate alkali substances good water (79.58-80.90).
represent the major forms of alkalinity in natural waters.
Alkalinity values for the water samples in the study ranged 300
Unsuitable for drinking (>300)
Water Quality Value

from 48.8 mg/L to 146.84 mg/L during the end of dry season.
According to the WHO standards for the river, the total 250
Very poor quality (200-300)

Water Quality Value

alkalinity values of all water samples below the permissible 200

limit of 200 mg/L.

Poor water quality (100-200)
COD concentrations of water samples were fluctuating 150

between minimum 24.6 mg/L at site 4 and the maximum 100

COD concentration was recorded at 42 mg/L at site 3 in the Good water quality (50-100)

middle stream, this is higher than the recommended WHO 50 Excellent quality (<50)

guidelines. In 2020, COD showed that 5 water quality Site 1 Site 2 Site 3 Site 4 Site 5
Sampling Location
monitoring sites reported COD values water samples
exacted the permissible limit of 5 mg/L. Fig. 2. Water quality classification based on WQI in 2018
Heavy Metals and Pesticides
Fe, the values of all water samples below the permissible An analysis of the 2020 water quality data, using the
limit of 0.3 mg/L, while the Chromium values below the WQI showed that water quality of the Prek Te River is still
permissible limit of 0.05 mg/L. of good quality, with all sites rated as either “good”. Of the
Cyanide (CN-): In 2020, samples were analyzed for total five sites located in the Prek Te Rive, poor water quality
cyanides at five sites. Results were all below the detection (136.19) is noted at site 4 a tributary of Prek Te River.
limit. Unfortunately, the detection limit used in this study is Unsuitable for drinking (>300) Water Quality Value

300
higher than those values.
As, the arsenic concentration in the five sampled water 250
Very poor water quality (200-300)

located in Prek Te River were below the permissible limit

Water Quality Value

(0.05 mg/L). The same concentration of As (0.005 mg/L) 200

Poo water quality (100-200)

was observed at site 2, site 3, and site 5, where, the 150

concentration of site 1 and site 4 not detected. The variation
of concentration of heavy metal from locations to locations 100
Good water quality (50-100)

may be correlated with the flow of the rivers and the location
50
of industries. Excellent quality (<50)

Pesticides: Pesticides are chemical compounds used to Site 1 Site 2 Site 3 Site 4 Site 5
Sampling Location
control or eradicate pests. At the end of wet seasons, the
analysis of 2018 showed that all 5 water quality monitoring Fig. 3. Water quality classification based on WQI in 2020

195
Principal Component Analysis quality. There were only a small number of measurements
PC1 and PC2 was selected to plot in biplot in order to exceeding the standard. The value of water quality
indicate the influence parameters link to the sample sites parameters of heavy metals and Arsenic from all sample
(Fig 4). The result of biplot in 2018 shown that site 3, 4 and collection in the end of dry season were also measure and
5 was influenced by PC1 under pH and DO while site 1 and found to be well below the standard of WHO, while
2 were controlled by PC2, temperature and pH. Cyanides are not detected at 5 sampling sites. Therefore, the
results of pesticides have only detectable some molecules.
Furthermore, at site 4 a tributary of Prek Te River has
strongly influenced by TS, temperature, TDS, pH, and DO.
The overall WQI showed that, by including all physico-
chemical parameters, all water samples were classified as
good quality at the end of wet season while it has poor
quality at end of dry season.
Based from the results of this study, the water quality of
Prek Te Rivers is generally still be considered as in good
quality for domestic use.

Acknowledgement
This work is supported by Guangxi science and technolo
gy program (Guike AD17195023；2018AD16013-04).
Fig. 4. Bivariate plot of the scores of PCs 1 and 2 in 2018 References
[1] Chhuon, K., Herrera, E., Nadaoka, K,. 2016.
In 2020, result shown that there are 61.79% of total
Application of Integrated Hydrologic and River
variance were captured by PC1 which strong influenced by
Basin Management Modeling for the Optimal
temperature, TDS, TS, pH, and DO at site 4 and 5 in the
Development of a Multi-Purpose Reservoir Project.
upstream while 24.43% was controlled by PC2, under
Water Resources Management, 30(9), 3143–3157.
controlled by Fe at site 3 in the middle stream. The result of
https://doi.org/10.1007/s11269-016-1336-4
biplot shown that site 1 and 2 was influenced by PC3 under
[2] Wang J, Liu G, Liu H, Lam, PKS,. 2017. Multivariate
Alkalinity and COD.
statistical evaluation of dissolved trace elements and
a water quality assessment in the middle reaches of
Huaihe River, Anhui, China. Sci Total Environ
583:421–431.
https://doi.org/10.1016/j.scitotenv.2017.01.088
[3] Wu, J., Li, P., Qian, H., Duan, Z., Zhang, X., 2014.
Using correlation and multivariate statistical analysis
to identify hydrogeochemical processes affecting the
major ion chemistry of waters: case study in Laoheba
phosphorite mine in Sichuan China. Arab J Geosci
7(10):3973–3982. https://doi.org/10.1007/s12517-
013-1057-4

Fig. 5. Bivariate plot of the scores of PCs 1 and 2 in 2020

IV. Conclusion
The water quality of Prek Te Rivers is still of good

196
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Pesticide Distribution in the Hydrological Compartment in Koh Thum

Nalin Hak1, Sombath Keo1, Chhovin Long1, Melvin Frick1, Sylvain Massuel2, Chanvorleak Phat3, Chhuon
Kong4, Sambo Lun4, Sokly Siev3, Ratha Doung4, Jean-Philippe Venot2, Khy Eam Eang4*

1
Master of UWE, Graduate School, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, Ph
nom Penh, Cambodia
2
UMR G-EAU, IRD, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, Phnom Penh, Cambodia
3
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Bo
x 86, Phnom Penh, Cambodia
4
Faculty of Hydrology and Water Resources Engineering, Institute of Technology of Cambodia, Russian Federation B
lvd., P.O. Box 86, Phnom Penh, Cambodia
*Corresponding author: khyeam_eang@yahoo.com
Abstract
This study aimed to make an investigation of the environmental state with regards to pesticides during the dry season after
10 years of intensive use in Koh Thum. Water from irrigation canal (Prek) and crop field and groundwater from mango farm
were sampled and pesticides were detected by semi-quantitative analysis using gas chromatography coupled to mass-
spectrometry (GC-MS). Firstly, 77 pesticides were inventoried through interviews of farmers and resellers in Kandal province.
Next, among the MS database of 451 molecules, 167 pesticides were detected including 21 which were part of the inventoried
pesticides from interviews while other 146 detected pesticides were off-list. The major types of biocides were insecticides
(33%), fungicides (25%) and herbicides (21%) and the major chemical families were carbamates (14%), organophosphates
(13%), triazoles (12%), organochlorines (11%), and pyrethroids (11%). Most of pesticides detected were moderately
hazardous (40%). Otherwise, 119 pesticides detected were not allowed in European Union while 10 were banned in Cambodia.
Chloroneb was detected in every water sample, with particularly high concentrations in groundwater (3.4010 ± 0.3644 μg/L)
and in water in the middle of Prek Touch (3.8314 ± 0.0826 μg/L). The Bassac river was the least contaminated (54 molecules),
the rice field water appeared to be the source of pesticides with 99 molecules, while 84 molecules were detected in the middle
of Prek Touch and 71 molecules were detected at Prek Chann drainage gate. Furthermore, 26 pesticides were common to
every compartment, suggesting water exchanges occurring between all compartments at the same time, probably during the
wet season. Almost twice as many pesticides were detected in the Preks than in the Bassac river and the Prek Touch West.
The pesticides would accumulate by remaining locally in the water near the treated locations while the flush effect would
remain limited. The unrehabilitated prek - the most hydrologically isolated - counted the greatest number of pesticides.

Keywords: Pesticides, hazardous pesticides, gas chromatography coupled to mass-spectrometry (GC-MS)

I. Introduction ecosystem. The classification of pesticide is dependent on
the intended target they control including herbicides,
Pesticides are natural or chemical, organic or inorganic,
insecticides, fungicides, rodenticides and many others [2].
substances or mixtures intended to prevent, destroy
Furthermore, the four main groups of toxicity pesticides are
populations of insects, weeds, rodents, fungi or other
including organochlorine, organophosphorus, carbamates,
harmful pests [1]. Pesticides consist of different products
and nitrogen based pesticides; they are persistent in the
with different functions. The extensive use of pesticides has
environment. However, these banned pesticides are still
impacted seriously on environmental, human, animal, and

197
used in the developing countries [3]. The classification SATREPS lab in the Institute of Technology of Cambodia.
toxicity hazard of each pesticide based on the acute oral and
dermal toxicity followed the WHO: Ia (extremely
hazardous), Ib (highly hazardous), II (moderately hazardous)
and III (slightly hazardous) (see Table 1) [4].
The use of pesticides in Cambodia started in the early 60s.
Between 2002 and 2012, the import of pesticides increased
by seventeen times, while most of pesticides in Cambodia
were imported from neigbour countries such as Vietnam,
Thailand and China since no factories were reported in
Cambodia. Based on the different survey studied, Cambodia
is the first range among other 13 countries with the highest
pesticide residue on vegetables, especially leaf of vegetables
from Kandal province [5]. On the other hand, many
Cambodian farmers have experienced health problem by
pesticide poisoning. Those chemicals have been extensively
misused in term of time, strength and way of use because of
insufficient understanding and the lack of relevant
instruction manual written in the native Khmer language [6].
Fig. 1. Prek system in Koh Thum district, Kandal province
Three main challenges to pesticide risk reduction were
identified such as the rapid spreading of pesticide trade 2.2. Clean-up procedure
associated with a weak regulatory, the strong satisfaction of
pesticides associated with insufficient awareness of risks
and a lack of knowledge in the use of pesticides, and the non-
regular monitoring of pesticide risks [7].
Table 1. Classification of hazards followed by WHO (2006)
𝑳𝑫𝟓𝟎 for the rat (mg/Kg body weight)
Class Oral Dermal
Solids Liquids Solids Liquids
Ia Extremely Hazardous
Ib Highly hazardous
<5
5-50
< 20
20-200
< 10
10-100
< 40
40-400
Fig. 2. SPE: Sample Elution & PLS3 Sorbent Washing
II Moderately hazardous 50-500 200-2000 100-1000 400-4000
III Slightly hazardous Over 500 Over 200 Over 1000 Over 4000 The samples were proceeded to purity by solid-phase
extraction (SPE) method following the method of Jinya,
II. Materials and Methods 2013 [8]. Sodium phosphate buffer solution (pH=7, 1mol/L)
was added to the water samples and left for rest for 1h. The
2.1. Sampling sites
sorbents PLS3 and AC in their respective cartridges were
The samples were collected on March 20th, 2020 in the activated by conditioning with 5 mL of dichloromethane, 5
Prek system (Prek Chan and Prek Touch) and Bassac river mL of acetone and two-time 5 mL of distilled water. The
which located in Koh Thum district, Kandal province (see samples were next passed through the cartridges at the flow
Fig.1). The Prek system is a canal perpendicular to the rate of about 15 mL/mn (see Fig.2). The cartridges were
Bassac river which was used for irrigating the crop located dried using nitrogen gas stream for 30 mn. After that, the
on the high bank of the alluvial plain called “Chamcar”. Five dried cartridges were disassembled and washed: 2 mL of
surfaces water and one groundwater sample were collected acetone followed by 5 mL of dichloromethane for the PLS3
in triplicate, which were as representative as possible in the cartridge, while both of the cartridge PLS3 and AC were
Prek system. 1000 mL of each sample were collected and washed with 5 mL of acetone. The mixed eluted solvent was
directly filtrated at the field site through 2μm multilayered collected and concentrated to approximately 1 mL, using
glass microfiber in order to remove debris and suspended nitrogen gas stream. Then, 10 mL of hexane were added to
materials, and then transported in iced box (around 4˚C) to the previous concentrated solution of 1 mL, to be then

198
dehydrated by elution through sodium sulfate. The solution predicted the structural formula of compounds, and
obtained was concentrated to 1 mL using nitrogen gas stream, compared them using the MS database.
transferred into vial, and finally stored at -20 ˚C.
2.5. Database software for simultaneous analysis
2.3. Calibration for Analysis
GC-MS (GC-MSTQ8040, Shimadzu, Japan) is equipped
A 1ppm mixture of standard pesticides containing about with automated identification and quantification system
950 molecules similar to those to be detected was added to with database. The database containing the mass spectra,
the samples. In addition, an external calibration was retention time and calibration curves about 1000 substances
performed with 1ppm, 0.5ppm, 0.125ppm, 0.1ppm, included 451 pesticides compound, 194 compounds of CH,
0.05ppm, and 0.025ppm solution containing 24 target 150 compounds of CHO, 113 compound of CHN (O), 14
pesticides; to which was also added 10ppm mixture of compounds of PPCPs, 12 compounds of CHS (NO), and 8
standard pesticides containing the other similar pesticides. A compounds of CHP (NOS), which permitted simultaneous
quality analysis could be performed on the 24 target identification and quantification of about 1000 plus
pesticides (see Table 1) while other pesticides could be substances without the use of chemical standards.
detected in a semi-quantitative way. Maximizing the performance of this database required high-
sensitivity instrument together with feature-rich quantitative
Table 1. Target pesticides for an external calibration of
soft.
quantitative analysis
No. Name No. Name 2.6. Statistical analysis
1 Methamidophos 13 Terbacil
All essays were carried out three times and the results
2 O,p’-DDT 14 Methyl-parathion were expressed as mean (±) standard deviation. Single-
3 Metalaxyl 15 Parathion
factor analysis of variance (ANOVA) were performed to test
4 Isoxathion 16 Pyroquilon
the differences between mean concentrations of pesticides,
5 Hexachlorobenzene 17 Anilofos
for completely randomized design (CRD) using Excel Data
6 Heptachlor 18 Azaconazole
analysis software. A p-value < 0.05 was considered
7 Aldrin 19 Isazofos statistically significant for every sample location.
8 Dieldrin 20 Mefenoxam
9 Endrin 21 BHC/HCH III. Results and Discussion
10 Chlordane (α, β, γ, δ) 3.1. General profile of pesticides detected in Koh Thum
11 Chloroneb 22 Lindane
In the area of Koh Thum, 167 pesticides were detected,
12 Atrazine 23 Malathion
the majority of which were insecticides (32%), then
24 Triadimefon
fungicides (25%) and herbicides (21%). Acaricides
2.4. Detection procedure represented 14%, nematicides were 3% and a further
minority included bactericides, molluscicides and plant
The detection procedures were inspired from Jinya
growth regulators. Carbamates (14%), organophosphate
(2013), and analyzed by using GC-MS device, model
(13%), triazoles (12%), organochlorines (11%), and
TQ8040 series. The column was DB-5ms, whose length was
pyrethroids (11%) were the most present chemical family of
30m, thickness 0.25μm and diameter 0.25mm. Injection was
the pesticides detected. Furthermore, pesticides moderately
splitless with volume 1μL. The temperature of the oven
hazardous were detected as predominant (40%), slightly
containing the column was maintained for 2mn at 40˚C, and
hazardous (16%) and unlikely to present acute hazard (16%);
then reached 310˚C for 5mn at a speed of 8˚C/mn. The flow
highly hazardous pesticides (8%) and extremely hazardous
rate of the carrier gas, ultra-pure helium, was 50mL/mn, and
pesticides 1%.
the GC column flow rate was 1.23 mL/mn. The MS ion
source temperature was 200˚C, and the MS interface 3.2 Pesticide detection and its regulation
temperature was 300˚C. The detection threshold for every Various pesticide molecules were detected in Koh Thum
compound was 0.0001 μg/L. The identification of pesticides district through semi-quantitative analysis, including 119
was performed by the data treatment system and the molecules either no approved or not registered in the
computer, which calculated the monoisotopic mass, European Union, while 10 were banned in Cambodia since

199
 Table 2. Target pesticides detected at the sample location
Concentrations detected at the sample locations EU water
drinking
Pesticide Bassac D.Gate Mango F Middle PT PT West Rice F standard
(μg/L) PC(μg/L) (μg/L) (μg/L) (μg/L) (μg/L) (μg/L)
Methamidophos - - - 0.0019±0.0032 - - 0
O,P’-DDT - - 0.0029±0.0051 - - - 0
Metalaxyl - - 0.2319±0.2452 - - - 0.1
Isoxathion - - - - - 0.0055±0.0096 0
Dieldrin - 0.0278±0.0254 - - - 0.0035±0.0061 0
Endrin - 0.0114±0.0197 - - - - 0
Chloroneb 0.9449±0.8682 1.1099±0.1544 3.4010±0.3644 3.8314±0.0826 0.4282±0.0491 0.5231±0.0061 0
Pyroquilon - - - 0.0018±0.0031 - - 0
Azaconazole - - 0.0123±0.0213 - - 0.0011±0.0019 0
Mefenoxam - - 0.3121±0.3293 - - - 0.1
Triadimefon 0.0283±0.0184 0.0176±0.0155 0.0158±0.0137 0.0197±0.0134 0.0331±0.0119 0.0225±0.0295 0
*Note: (-) is not detected, (0) not allowable to present in drinking water; D: Drainage; F: Farm; PT: Prek Touch

2012 [5]; namely, o,p’-DDT, isozathion, dieldrin and endrin. these pesticides 5 highly hazardous, 42 moderately
Moreover, 11 molecules were detected among the 24 hazardous, 20 slightly hazardous and 14 molecules were not
targeted pesticides through a quantitative analysis. Among listed in WHO classification. Mango farm and Prek Touch
those 11 molecules, only 2 were authorized in European West were showed the similar number of pesticides (60 and
Union, namely metalaxyl and mefenofam, whose water 57 pesticides). In Mango farm, no highly hazardous
drinking standards were 0.1 μg/L (see Table 2). pesticides were detected while 1 extremely hazardous, 33
moderately hazardous, and 7 slightly hazardous. Prek Touch
3.3. Pesticide distribution, its persistence and toxicity
West the number of pesticides were detected including 4
Among the 167 pesticides detected in the hydrological highly hazardous, 26 moderately hazardous, 8 slightly
compartments, Bassac river was observed to be hazardous, and 8 pesticides of unknow toxicity. In Middle
contaminated with the least number of pesticides than the of Prek Touch, there were 84 number of pesticides were
Preks, the Prek Touch West, and the rice field. Most of detected including 1 extremely hazardous, 7 highly
pesticides detected in Bassac river were moderately hazardous, 41 moderately hazardous and 13 slightly
hazardous (22 molecules) followed by slightly hazardous hazardous and 6 molecules unknown toxicity. In addition,
(11 molecules), and 11 molecules of unknown toxicity. Rice 71 pesticides were detected in the drainage gate of the Prek
field is the highest contaminated (99 pesticides), among of Chan including 4 highly hazardous, 30 moderately

Fig 3. Distribution of toxicity with regard to persistence of pesticides detected in the Koh Thum district

200
hazardous, 9 slightly hazardous and 10 unknown toxicities which was banned in Cambodia was detected in the
(see Fig. 3). groundwater, the rice field water, the water of the Prek Touch
West and the water in middle of the Prek. Moreover, 26
3.4 Hydrological flows and pesticide movement
pesticides were common to every hydrological compartment.
Firstly, in dry season, potential exchanges may happen A separation between Prek and Bassac river and the Prek
between the Bassac river and the rehabilitated Prek Chan Touch West on the other hand seemed to emerge in the dry
since they shared 37 pesticides in common (52.1% of the season. The surface water flow from the Bassac river
pesticides detected in Prek chan and 68.5% of the pesticides connected to the Prek Touch West seemed to maintain a
detected in the Bassac river. At this stage, it was impossible lower number of pesticides while these numbers tended to
to determine if the source of pesticide came from the prek, build up in the Preks. Those conclusions should be
the river, or both of them. The presence of pesticide in Prek confronted to analysis in raining season to confirm the
Chan could either come from the neighboring treated plots pesticides transport during the flood period.
by excess irrigation return flow and by leakage drainage, or
come from exchange with groundwater. Secondly, we have Acknowledgement
known that the water flow was very limited in Prek Chan. We are thankful to the COSTEA project for the financial
Prek Touch was isolated from the Bassac and the Prek Touch support under the management of Institut de Recherche pour
West. We could determine that the pesticide found in the le Développement (IRD).
preks were mostly tied to the presence of pesticides in the
References
surrounding treated plots in dry season. The detection of
pesticide in non-rehabilitated Prek Touch (84 pesticides) [1] Mahmood, I., Imadi, S. R., Shazadi, K., Gul, A.,
more than in the rehabilitated Prek Chan (71 pesticides) Hakeem, K. R. (2016). Effects of pesticides on
could be explained either by the spatial variability that environment. In Plant, Soil and Microbes. Springer,
Cham, 253-269.
would imply poor mixing of prek water or by limited
[2] Uqab, B., Mudasir, S., Nazir, R. (2016). Review on
exchanges with the Bassac river and Prek Touch West for bioremediation of pesticides. J Bioremed Biodeg, 7,
Prek Chan. Compare to the prek, the Bassac river and the 343.
Prek Touch West clearly showed lower and similar numbers [3] Saravi, S. S. S., Shokrzadeh, M. (2011). Role of
of pesticides detected (54 and 57 pesticides). The Prek pesticides in human life in the modern age: a review.
Touch west, running from north toward the south collected Pesticides in the modern world-risks and benefits, 3-12.
water from the Bassac river through other preks upstream [4] WHO-UNEP (2006). Sound management of pesticides
and diagnosis and treatment of pesticide poisoning: a
and flow directly to the boeung. Compare to the
resource tool. Geneva: Inter-organization Programme
groundwater 60 pesticide were detected and 41 pesticides for the Sound Management of Chemicals.
were found in Prek Touch (48.8% of the pesticides detected [5] FFTC Agricultural Policy Platform (FFTC-AP) (2015).
in Prek Touch and 68.3% of pesticides detected in the Current Use of Pesticides in the Agricultural Products
groundwater). This would suggest possible exchanges even of Cambodia. https://ap.fftc.org.tw/article/986.
in dry season with groundwater where pesticides could be Accessed 25 December, 2020.
[6] Matsukawa, M., Ito, K., Kawakita, K., Tanaka, T.
stored, but it was difficult to draw any conclusion with a
(2016). Current status of pesticide use among rice
single groundwater sample. farmers in Cambodia. Applied Entomology and
IV. Conclusion Zoology, 51(4), 571-579.
[7] Schreinemachers, P., Afari-Sefa, V., Heng, C. H., Dung,
The result obtains from this study indicated that the three P. T. M., Praneetvatakul, S., Srinivasan, R. (2015). Safe
major types of biocides were insecticides, fungicides and and sustainable crop protection in Southeast Asia:
herbicides, and most of pesticides were moderately status, challenges and policy options. Environmental
Science & Policy, 54, 357-366.
hazardous. The Bassac river was the least contaminated and
[8] Jinya, D. (2012). Development of solid-phase
the least hazardous water in term of potency of toxicity. extraction method for simultaneous analysis of semi-
However, the rice field appeared to be the source of pesticide volatile organic compounds using a GC-MS database
application, which accumulated mostly in waters of the system. SHIMADZU Technical Report.
rehabilitated and non-rehabilitated Preks. The water in https://www.shimadzu.com.cn/solution/green/pdf/jjfa/
middle Prek Touch was the most hazardous, while captafol s19.pdf. Accessed 25 January, 2021.

201
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Assessment of Pesticide Residues in Surface Water and Fish from Chhnok Tru, Kampong
Chhnang

Chanvorleak Phat 1,2,*, Kearakvattey Kun 1, Voleak Pheap 1, Sereyvath Yoeun 1, Eden G. Mariquit 3, Winarto
Kuriniawan 3 and Hirofumi Hinode 3

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
School of Environment and Society, Tokyo Institute of Technology, Japan
* Corresponding author: phatchanvorleak@itc.edu.kh

Abstract

Surface water from 5 different sites and 4 types of fish from the Cyprinidae family were collected from Chhnok Tru,
floating community of Tonle Sap Lake. Those samples were subjected to pesticide residues analysis using gas chromatography
mass spectrometry. Qualitative results showed that water samples were detected with fungicides fluquinconazole and
fenpropimorph. Insecticides such as lenacil and mevinphos were also found in some studied sites. For fish samples, two active
compounds of fungicides were detected in muscle part of fish from Cyprinidae family. Quantitative results could be drown
only from water samples as targeted compounds were not found in fish samples. Three fungicides including azaconazole,
chloroneb, and pyroquilon were detected with the concentrations of 16.46 µg/L in CT3, 0.28 µg/L in CT5, and 0.51 µg/L in
CT2, respectively. Four active compounds of insecticides were found in CT3 and CT5, which CT3 were detected with
chlordane and CT5 were contaminated with three compounds such as HCHs, isazofos, and parathion. The presences of these
pesticides in water and fish sample pose a high risk to human as well as to the lake environment.

Keywords: Pesticides, Chnhnok Tru, Tonle Sap Lake, Qualitative analysis, Quantitative analysis

I. Introduction Agriculture is the economic foundation of Cambodia,

where more than 70% of the population engage in
The Tonle Sap Lake (TSL), or the Great Lake of
agricultural activities [3, 4]. However, agriculture sectors
Cambodia, is located in the central plains of Cambodia. The
face some constrain such as crop losses due to pest
lake is known for its rich biodiversity and outstanding water
infestication making the application of pesticides crucial for
regime, with vast seasonal variations in water levels and
plant protection and increase crop yield. A number of global
volumes [1]. Fishing and agricultural activities in the TSL
monitoring studies have shown the ability of pesticides to
basin benefited from abundant freshwater, nutrients and rich
contaminate surface and ground water by runoff,
soils contributed by seasonal flood pulse and rainfall;
groundwater leaching and spray drift [4–6]. Each year, over
together, these ecosystem services have preserved the
3 million cases of pesticide poisoning with an estimated
region's livelihoods for centuries [2].
220,000 deaths occur globally especially in developing

202
countries. Water samples were then filtered using 0.45 μm fiber glass
Scientific evidence is required to raise public awareness filter (Whatman GMF 150) to remove suspended solids.
of the adverse effects of pesticides on all stakeholders, Suspended solids retained by the filters were eluted with 5
including farmers and policy makers. However, there is mL of acetone and 5 mL of dichloromethane to dissolve
limited information on pesticide contamination in Cambodia. potentially adsorbed compounds. Filtered samples were
Therefore, this study aims to identify the presence of subjected to extraction process by using solid-phase
commonly used pesticides and evaluate the levels of some extraction method (SPE) as described by Jinya (2013).
banned pesticides in surface water and fish from Chhnok Tru,
Kampong Chhnang, Cambodia. Chhnok Tru is one of TSL's 2.3. Fish sample preparation
floating communities. With its unique characteristics, the Fish muscles were then grind and freeze-dried. Five
majority of the village is submerged during the rainy season, grams of freezed dry fish muscle in powder form was
while broad cultivated land is accessible during the dry homogenized with 50 mL of distilled water and 100 mL of
season. This contributes to the abundance of rice and other acetone/hexane mixture (50/50, v/v) was added to the
crop cultivation in this region, making it important to study suspension and stirred for 15 min. After that, this mixture
on pesticide contamination levels from agriculture to water was filtered with Whatman filter paper (pore size of 2 µm).
and aquasystem. Then, ten grams of sodium chloride and 1 mL of phosphate
buffer solution (1 mol/L, pH 7.0) were added into the
II. Materials and Methods filtrated solution and stirred again for 10 min. After that, the
decanter was used to separate the layer of organic solvent
2.1. Sample collection
from fish fat part. Then, anhydrous sodium sulfate was used
Surface water were collected in March 2020 from 5 site
to remove any remaining water and impurities from the
at Chhnok Tru community (Fig.1). Water samples were
solution. Lastly, the solution was applied to SPE following
stored in cleaned 1 L plastic bottles and transported to the
the method of Jinya (2013).
laboratory at Institute of Technology of Cambodia for
analysis.
2.4. GC-MS analysis
The pesticide compounds were analyzed using gas
chromatograph and mass spectrometer (GC-MS), model
TQ8040 series (Shimadzu, Japan). The column used was
DB-5ms with the length of 30 m. One microliter of sample
was injected to GC-MS by auto injector using spitless mode.
After that, the column oven temperature was programmed
from 40 to 310°C by holding for 2 minutes at 40°C,
increased the temperature to 310°C with a rate of 8°C/min,
and then held for 5minutes. The carrier gas was ultra-pure
Fig.1. Water sampling points at Chhnok Tru helium at the total flow of 50 mL/min while the column flow
was 1.23 mL/min. The ion source temperature was 200°C
Freshwater fish samples of the Cyprinidae family (Trey and the temperature of the interface was 300°C. All
Chhkork (Cyclocheilichthys enoplos), Trey Chhpin pesticides were both qualitatively and quantitatively
(Hypsibarbus suvattii spp), Trey Chrakaing (Puntioplites analyzed by retention time and specific ions, and quantified
falcifer), and Trey Khmann (Hampala macrolepidota)) were by the external standard method.
collected from fishermen at the various sites in Chhnok Tru,
Kompng Chhnang Province. III. Results and Discussion

2.2. Water sample preparation 3.1. Qualitative analysis of pesticides

Sample preparation has been adopted from Jinya (2013) Qualitative analysis was carried out by scanning method
in which 1 L of surface water sample has been adjusted with to identify the presences of pesticides in tested samples. The
1 mL of phosphate buffer solution (1 mol/L, pH 7.0) [7]. results showed that water samples were detected with
fungicides fluquinconazole and fenpropimorph in CT3 and

203
CT2, respectively (Table1). Fungicides have been greatly Fish species
Fungicide
used for the control of pathogen fungi in crops; however, it Biphenyl Fluquinconazole
Trey Chhkork (Cyclocheilic - +
can exhibit undesirable effects on non-target plant-beneficial
hthys enoplos)
microorganisms [8]. Besides, fenpropimorph has a very Trey Chhpin (Hypsibarbus + -
negative impact on the plant root development which lead to suvattii spp)
growth deduction in some cereal seedlings [8]. Insecticides Trey Chrakaing (Puntioplit + -
es falcifer)
such as lenacil and mevinphos were also found in some Trey Khmann (Hampala m - +
studied sites (CT1 and CT2). Lenacil is an odorless acrolepidota)
herbicide employed for selective weed control. The
detection of this compound in water sample might be from 3.2. Quantitative analysis of pesticides
agricultural application and its solubility and stability in Among 23 target pesticides (fungicides include
water [9]. azaconazole, chloroneb, hexachlorobenzen, mefenoxam,
metalaxyl, pyroquilon, and triadimefon; herbicides include
Table 1. Presences of pesticides in surface water from anilofos, atrazine, and terbacil; insecticides include Aldrin,
Chhnok Tru chlordane, dieldrin, endrin, HCHs, heptachlor, isazofos,
isoxathion, malathion, methamidophos, methyl parathion,
o,p’-DDT, and parathion), seven active compounds were
Category

Active
CT1 CT2 CT3 CT4 CT5 detected in water sample from Chhnok Tru. CT5, which is
compounds
surrounded by rice and crop field was the most contaminated
Lenacil - + - - -
sites with four pesticide compounds (Table 3). Some banned
Insecticide

pesticides such as chlordane, HCHs, and parathion were

Mevinphos + - - - - detected in this study with the concentrations of 9.64 µg/L
in CT3, 0.50 µg/L in CT5, and 0.36 µg/L in CT5,
Fenpropimorph - + - - - respectively. The presences of these compounds might be
Fungicide

due to their highly persistence in environment or their illegal

Fluquinconazole - - + - - application. For instance, chlordane may remain in soils for
20 years and travel long-distances to higher latitudes
through repeated deposition and evaporation [11]. The levels
For fish samples, two active compounds (biphenyl and of pesticides detected in this study were highly over the
fluquinconazole) of fungicides were detected in muscle part WHO single pesticide limit of 0.1 µg/L for drinking water
of fish from Cyprinidae family (Table 2). Biphenyls are [12]. As Chhnok Tru is a floating community, most of
important structural analog, which are used considerably in populations there use lake water as the main source for
synthesis of various compounds. Earlier, biphenyl drinking, cooking, and household uses. This elaborate the
derivatives were widely used as pesticides in the form of primary route of exposure of those pesticides to human
polychlorinated biphenyls (PCBs). Biphenyl is a chemical which can cause severe health issues as chlordane initially
fungicide used in agricultural application to kill or inhibit accumulates in the kidney and liver, and it is then
the growth of fungi. Biphenyl is insoluble in water and vapor redistributed to adipose tissue, where it can be stored for
are heavier than air cause a persistent in environment, could several years due to its high lipophilicity [11].
absorbed through the skin, the mucous membrane and
pulmonary system of organism. Fish can accumulate of Table 3. Levels of target pesticides in surface water from
biphenyl by direct absorption through the gills, exposure to Chhnok Tru
contaminated sediments, and consumption of insects and
smaller fish [10]. Fish were contaminated with biphenyl and Compounds CT1 CT2 CT3 CT4 CT5

fluquinconazole in their muscle due to the accumulation of

pollutant from their surrounded environment include live in Azaconazole - - 16.46±0.03 - -

contaminant water or consume the contaminant food. Chloroneb - - - - 0.28±0.01

Pyroquilon - 0.51±0.01 - - -

Table 2. Presences of pesticides in fish from Chhnok Tru

204
 Chlordane - - 9.64±0.15 - - Pesticide use and self-reported symptoms of acute
HCHs - - - - 0.50±0.03 pesticide poisoning among aquatic farmers in
Isazofos - - - - 0.61±0.01
phnom penh, cambodia. J Toxicol 2011:.
https://doi.org/10.1155/2011/639814
Parathion - - - - 0.36±0.00
5. Kapsi M, Tsoutsi C, Paschalidou A, Albanis T (2019)
Environmental monitoring and risk assessment of
The residues of pesticide in aquatic system can be from pesticide residues in surface waters of the Louros
a historic or recently used around the study area. For our River (N.W. Greece). Sci Total Environ 650:2188–
study, all of 23 target compounds were not detected in all of 2198.
fish samples include muscle, gill and inner part of the four https://doi.org/10.1016/j.scitotenv.2018.09.185
6. Papadakis EN, Vryzas Z, Kotopoulou A, et al (2015)
fish species in Cyprinidae family. Base on the previous
A pesticide monitoring survey in rivers and lakes of
studies, most of organochlorine compounds were detected in northern Greece and its human and ecotoxicological
fish, where the agriculture area using large amount of risk assessment. Ecotoxicol Environ Saf 116:1–9.
pesticides around the lake or river. Most of the compounds https://doi.org/10.1016/j.ecoenv.2015.02.033
were detected in a low or high level depend on the 7. Jinya D (2013) Report Development of Solid-Phase
environment they lived [13]. Extraction Method for Simultaneous Analysis of
Semi-Volatile Organic Compounds Using a GC-MS
Database System. Shimadzu Tech Rep 1–8
IV. Conclusion 8. Campagnac E, Fontaine J, Sahraoui AL-H, et al
(2008) Differential effects of fenpropimorph and
In this study, surface water samples from 5 sites and four
fenhexamid, two sterol biosynthesis inhibitor
fish species were collected from Chhnok Tru floating fungicides, on arbuscular mycorrhizal development
community and evaluated the levels pesticide contamination. and sterol metabolism in carrot roots.
The results showed that 80% of water samples and 100% of Phytochemistry 69:2912–2919.
fish samples were contaminated with at least one of pesticide https://doi.org/10.1016/j.phytochem.2008.09.009
compound. Among 23 target compounds, azaconazole, 9. Worden AN, Noel PRB, Mawdesley-Thomas LE, et
chloroneb, pyroquilon, chlordane, HCHs, isazofos, and al (1974) Feeding studies on lenacil in the rat and
dog. Toxicol Appl Pharmacol 27:215–224.
parathion were detected in surface water from Chhnok Tru
https://doi.org/10.1016/0041-008X(74)90192-6
area. Routine monitoring studies should be conducted to 10. Kampire E, Rubidge G, Adams JB (2015)
raise public awareness of pesticide contamination and to Distribution of polychlorinated biphenyl residues in
prevent the worsening of pesticide pollution issues. sediments and blue mussels (Mytilus
galloprovincialis) from Port Elizabeth Harbour,
Acknowledgement South Africa. Mar Pollut Bull 91:173–179.
https://doi.org/10.1016/j.marpolbul.2014.12.008
This research was financially supported by SATREPS - 11. Singh K, Nong A, Feeley M, Chan HM (2019)
JST/JICA: grant-number JPMJSA1503 - Establishment of Development of Biomonitoring Equivalents for
Environmental Conservation Platform of Tonle Sap Lake. chlordane and toxaphene with application to the
general Canadian population. Regul Toxicol
Pharmacol 106:262–269.
References https://doi.org/10.1016/j.yrtph.2019.05.015
12. WHO (1993) WHO-Guidelines for Drinking Water
1. Keskinen M (2006) The Lake with Floating Villages: Quality, Chemical Aspects.
Socio-economic Analysis of the Tonle Sap Lake. Int https://www.who.int/water_sanitation_health/dwq/
J Water Resour Dev 22:463–480. GDW8rev1and2.pdf
https://doi.org/10.1080/07900620500482568 13. Kafilzadeh F (2015) Assessment of Organochlorine
2. Lin Z, Qi J (2017) Hydro-dam – A nature-based Pesticide Residues in Water, Sediments and Fish
solution or an ecological problem: The fate of the from Lake Tashk, Iran. Achiev Life Sci 9:107–111.
Tonlé Sap Lake. Environ Res 158:24–32. https://doi.org/10.1016/j.als.2015.12.003
https://doi.org/10.1016/j.envres.2017.05.016
3. FAO (2014) Country fact sheet on food and
agriculture policy trends. Fapda 6
4. Jensen HK, Konradsen F, Jørs E, et al (2011)

205
Session 4: Food Technology and Microbiology
The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Optimization of Young Mango Fermentation and Effect of Different Preservation

Methods on its Shelf-life

Kimleang NGOUN 1, Guechleang CHHUN 1, Reasmey TAN 1,2,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: rtan@itc.edu.kh

Abstract

Fermented young mango is one of the most famous mango based products in Cambodia and it is
particularly popular among teenagers. However, even if this were to be the case, we observe that the quality of this
product is considerably low and has a short shelf-life. The objectives of this study were to produce a fermented
young mango that is well-received by consumers and study on the effect of two food preservation methods on its
shelf life. The optimization of young mango fermentation in this study focused on the mango’s preparation (mango
fermented as whole fruit and peeled fruit), blanching temperature and time (65 oC for 15 min and 70 oC for 10
min), fermentation conditions (ferment at room temperature and 4 oC), salt concentrations (2% and 5%) and sugar
concentrations (20 oBrix, 23 oBrix, and 25 oBrix). The sensory evaluation was conducted in each step except for
blanching temperature and time and the duration of soaking in cold water by using 9 points hedonic scale. For the
study of shelf life, the vacuum packaging was utilized and two food preservatives (sodium benzoate and sodium
metabisulfite) were used individually on the fermented young mango that was fermented after one day at the
concentration of 0.1%. During the three weeks’ observation on its shelf life, it was revealed that the sample treated
with sodium benzoate stored at 4 oC was the best condition as in the final week of observation, it attained the
highest pH (2.41±0.04), lowest acidity (1.35±0.00), lowest total soluble solids (13.55±0.07), lowest salt content
(0.59±0.00) and yeast and mold was not detected in the sample.

Keywords: Sodium benzoate, Sodium metabisulfite, Lactic acid bacteria, Titratable acidity

I. Introduction vulnerability to low temperature, and perishable nature of

the fresh fruit limit the transport of the product from the
Mangoes are fruits with high valuable nutritional location to farther areas. This issue can be illuminated by
properties but like other fruits, they are perishable and have utilizing green fresh fruits to turn them into pickle or chutney
a short shelf-life. Not to mention, they are seasonal, or sundried acidifying condiment (Amchur is a fruity spice
therefore, they are not available all year long in the shops powder made from dried unripe green mangoes and is used
which were the reason processes are developed to extend as a citrus seasoning), whereas ripe mangoes are utilized for
their shelf-life. After collecting, mango ripe rapidly since it the preserve, jam, sauces and so on [1]. Fermented mangoes
is classified as a climacteric fruit. Problems with the disease, are prepared with high sugar content and a low salt

206
concentration to help the lactic acid fermentation. Since the method, 0.1 ml of the sample was dropped onto the solid
sugar is the main ingredient and raw mangoes were utilized, media then using the spreader to gently and evenly distribute
the prepared mangoes taste sweet and sour and it also the sample. Finally, bring it to incubate 5 days at 25-30 ºC.
provides energy [2]. Nevertheless, due to it being a low-salt Yeasts will grow creamy to white colonies and mold will
fermentation, it has a lot of risks in contaminating by other grow as filamentous colonies, black spores.
spoilage bacteria which end up with a short shelf-life. That
is why the current study was conducted to find a food 𝑵=
∑𝑪
(E.q. 2.2)
preservation method to prolong its shelf-life. 𝑽[(𝟏×𝒏𝟏 )+(𝟎.𝟏×𝒏𝟐 )]×𝒅

II. Materials and Methods 2.4. Sensory analysis

2.1. Sampling sites The sensory evaluation of the two fermented young
A type of mangoes as known as Keo Romeat in mangoes that participated by 12 panelists using 9 points
Cambodian. The selected mangoes were small in size not Hedonic scale.
fully ripe nor mature with bright green color skin indicating 2.5. Statistical analysis
that they were still young. Mangoes were stored in ambient The data are expressed as mean ± standard deviation
temperature overnight before bringing to the work area to (Χ ± 𝑆𝐷). Statistical test one-way ANOVA was applied to
process the fermented young mango. find a significant difference between the value of various
2.2. Analytical methods parameters recorded for the effect of different preservation
2.2.1.pH methods on the fermented young mango’s shelf life, p-value
STARTER300 was used to measure the pH value of ≤ 0.05 was considered statistically significant.
fermented young mango. The pH meter was calibrated at 25
ºC using buffer solutions of acidic and basic values of 4 and III. Results and Discussion
7, respectively. The sample of each condition was put into a 3.1.Sensory Evaluation
small beaker around 20 ml then immerse the electrode of pH
meter enough to submerge. Each sample was done in
triplicate.
Appearance
2.2.2. Total soluble solids (TSS) 10
Brix value was measured by a refractometer (ATAGO, Overall
Odor
Japan) to determine the total soluble solids in the liquid. acceptance 5
Only a few drops of the sample were required to drop on the
prism surface. Each sample was done in triplicate. 0
2.2.3. Acidity Sweetness Taste
The acidity of fermented young mango was determined
through an analytical method adopted by AOAC
(Association of Official Analytical Chemists). Bring 10 ml Sourness Texture
of sample to titrate with NaOH (0.1N) using phenolphthalein
as an indicator for 2-3 drops. Keep the titration going until
the solution turned a light pink color. Each sample was done Whole fruit Peeled fruit
in triplicate.

%𝑻𝑨 =
𝑽𝒔 ×𝑵×𝟗𝟎×𝟏𝟎𝟎
(E.q. 2.1) Fig. 1. Sensory evaluation of mango’s preparation
𝑽×𝟏𝟎𝟎𝟎

2.2.4. Salt . There was a significant difference (p-value ≤ 0.05) in

the appearance of the two fermented mango of which the
The salt content was determined by using a salt meter, fermented mango with a whole fruit without peeling and
ATAGO ES-421, Japan. Firstly, click start to power on then cutting received 6.08±1.16 due to its drastic change in skin
drop a few drops of distilled water onto the titanium color which was visible to the eyes and the coagulated sap
electrode and click ‘zero’. Each sample was done in that stuck on the surface of the skin resulting in a negative
triplicate. response from the panelist. Meanwhile, the fermented
mango with peeled skin and cut into thin strips obtained a
2.3. Yeast and Mold higher score which was 6.92±1.16 receiving some positive
comments from the consumer due to its hygienic and good
Potato dextrose agar (PDA), a selective media for the looking appearance.
growth of yeast and media was used. Using the spread plate

207
3.2. Effect of sodium benzoate, sodium metabisulfite and stored at 4 °C which varied from 13.7±0.14 to 12.6±0.14.
vacuum packaging on pH During the fermentation, LAB converted sugars into lactic
acid which caused a decline in TSS [3] and at the same time,
2.7 there were also three simultaneous mass transfer phenomena
occurred; water flows from the product to the solution, a
solute transfer from the solution to the product and a minor
pH

transfer of product’s own solutes (sugars, organic acids,

2.2 minerals, and vitamins) to the concentrated solution [2].
0 1 2 3
Storage time (Week) 3.4.Effect of sodium benzoate, sodium metabisulfite and
RT 4˚C vacuum packaging on acidity

Fig. 2. pH of vacuum packaging samples 2.6

TA (%)
stored in both conditions 1.6

Regardless of any conditions, pH value of the samples 0.6

decreased for 3 weeks of which after one-week fermentation 0 1 2 3
the pH dropped quickly in comparison to two-weeks and Storage time (Week)
three-weeks fermentation. For, fermented young mango RT 4˚C
with the addition of E211 and E223 stored at room Fig. 4. TA of vacuum packaging samples stored in both
temperature, it dropped significantly from 2.95 ± 0.03 to
2.56±0.02 and 2.77±0.02 to 2.54±0.03 respectively after conditions
one-week fermentation with no significant difference to be The acidity for vacuum packaging stored at room
found in week 1 but there was a significant difference temperature was lower than the one stored at 4 °C possibly
between Con, 2.45±0.02 and the two preservatives in week due to the Ascorbic acid degradation with the exposure to
one. However, it had to be noted that the pH for every sample the sunlight because it was well-known that Normally, the
stored at 4ºC was higher than that of room temperature. The fermentation started with Leuconostoc mesenteroides as
reason for the decrease in pH can be attributed to the they produced lactic acid until it reached 0.25 to 0.3%. As
increase in acidity which due to lactic acid fermentation in the acidity increased Leuconostoc species started to die off,
which LAB converted carbohydrates into lactic acid. the Lactobacilli species (plantarum and cucumeris) took
Another reason that contributed to the strong acidity was due over as they continued until the acidity was 1.5 to 2.0%. The
to the Ascorbic acid content contained in the young mango. last bacteria in the chain was the Lactobacilli pentoaceticus
that continued until the acidity was 2-2.5% [3]. The highest
3.3.Effect of sodium benzoate, sodium metabisulfite and acidity in this study was 1.64±0.03% of fermented mango
vacuum packaging on total soluble solids (TSS) with vacuum packaging stored at 4 °C while others were
around 1.5% showing that the fermentation hasn’t reached
22 the end yet.
TSS (˚Brix)

12 3.5.Effect of sodium benzoate, sodium metabisulfite and

vacuum packaging on salt content
2
0 1 2 3 0.9
Salt (%)

Storage time (Week) 0.7

RT 4˚C 0.5
0.3
Fig. 3. TSS of vacuum packaging samples 0 1 2 3
stored in both conditions Storage time (Week)
RT 4˚C
For vacuum packaging, fermented young mango for
both conditions had its TSS fluctuated throughout the Fig. 5. Salt content of vacuum packaging samples stored in
observation with the one stored at room temperature greatly both conditions
changed from 13.97±0.35 to 5.65±0.70 unlike when it was

208
 Salt decreased over time for every sample regardless of IV. Conclusion
with or without the addition of preservatives. For both
conditions and no significant difference was found for each In the young mango’s fermentation, it was found that
sample in different storage conditions. Salt allowed water fermented mango tasted better if it was to be peeled and cut
and needed sugars to be pulled from the product [4]. For into thin strips and the ideal blanching temperature and time
vacuum packaging, fermented young mango stored at room was at 65 °C for 10 min. After the mild-heating process,
temperature is no longer safe to be consumed after 3 days- bringing it to ferment for one day. According to the sensory
storage as it exhibited an unpleasant appearance and odor evaluation, it was discovered that young mango fermented
this spoilage was caused by excess lactic acid fermentation at room temperature worked better than the fermentation
and other spoilage bacteria present due to the fermented process at 4°C as it had a more desirable odor, texture, and
young mango’s high sugar content and low salt taste. Besides, young mango fermented with a salt
concentration which is a favorable condition for them to concentration of 2% and sugar content of 23 °Brix received
grow [4]. a positive response from the panelists. Regarding the study
of shelf life, in consideration of the slower fermentation
3.6. Effect of sodium benzoate, sodium metabisulfite and process, fermented young mango added with sodium
vacuum packaging on yeast and mold benzoate stored at 4 °C was the best condition as it had
highest pH and lowest acidity among the studied conditions.
15 For further study, it is recommended to do a more detailed
Log CFU/ml

10 preliminary study on the blanching temperature and time,

5 salt concentration and study the effect of the combination
between the two preservatives (sodium benzoate and sodium
0 metabisulfite), the addition of preservatives should be done
0 1 2 3 while processing the fermented young mango and it is also
Storage time (Week) highly suggested to choose another preservative that has an
antimicrobial effect against LAB in the future study.
RT 4 ˚C
References
Fig. 6. Yeast and mold present in vacuum packaging
samples stored in both conditions [1] Saroj, Kumar, A., Singh, A., Sharma, M., 2010.
Standardization of recipe and method for mango
Due to the high acidity of the product, it inhibited the chutney. Haryana Journal of Horticultural Sciences.
growth of spoilage bacteria, therefore, only LAB, yeast, and 39, 247–249.
mold were able to grow in an acidic environment. Also, this [2] Khan, S., Ali, W., Kamal, T., Usman Khan, M., Gul,
was a low-salt concentration which promoted the growth of S., 2016. Preservation of mango slices in sucrose
more types of fermenting organisms, especially in natural solution with various concentrations. Journal of Food:
fermentation, leading to faster acid production and a more Microbiology, Safety & Hygiene. 1, 1–7.
acidic product. However, the lower salt concentrations can [3] Food and Agriculture Organization., 2012. Pickles.
allow for mold and other contaminant growth resulting in a
General information. 4, 15–16.
softer texture or off-flavored final product [4]. As it is shown
in Figure 6. at room temperature, it was found that yeast was [4] Eifert, J.A., Safety, F., Agent, E., Extension, V.C.,
present in the Con sample with 4.49 log CFU/ml in week 0 Boyer, R.R., Science, F., Tech, V., Williams, R.C.,
then rose dramatically to 10.72 log CFU/ml in week 2 and 2015. Vegetable fermentation. Fermentation. 3, 10–11.
finally showed a slight decline in week 3, 9.53 log CFU/ml [5] Shahmohammadi, M., Javadi, M., Nassiri-Asl, M.,
possibly due to the increase in acidity which hindered the 2016. An overview on the effects of sodium benzoate
growth of yeast and the decrease in sugar and salt content as a preservative in food products. Biotechnology and
which were considered to be their nutrients. Meanwhile, on Health Sciences. 3, 44–49.
the other hand, the effect of preservatives on the shelf-life
could be seen with the naked eye. Antimicrobial activity of
E211 is deeply related with the pH as most of its activity
occurred at low pH and it was also deemed as an effective
deterrent force against yeast and mold that were able to grow
in an acidic environment [5].

209
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

The effect of blanching on curcumin content and chemical composition of essential oils of
dried Turmeric (Curcuma longa)

Chhanthyda Nong1, Sokneang In1,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* in@itc.edu.kh
Abstract

Turmeric is normally dried and grind into powder prior to storage and usage, and the conditions of blanching is one of
important factors that can affect on the quality of dried turmeric. The objective of this study was to observe the impact of
blanching on total phenolic content (TPC), antioxidant activity, curcuminoids content and essential oil composition of dried
turmeric. Fresh turmeric rhizome was processed with blanching in boiling water for 0, 15, 30 and 45 minutes before oven
dried at 55°C for 16 hours. The curcuminoids content and essential oil composition was determined by HPLC and GC-MS,
respectively. The highest content of TPC (107.62 ± 6.46 mg GAE/g dried turmeric) and the total of antioxidants activity (9.58
± 0.11 mg VCE/g dried turmeric) were found in the sample blanching for 30 minutes. The curcumin content of all blanching
condition was ranged from 1.78 ± 0.12% to 2.20 ± 0.07% for bisdemethoxycurcumin, 1.00 ± 0.08% to 1.24 ± 0.05% for
demethoxycurcumin and 2.24 ± 0.06% to 2.62 ± 0.14% for curcumin. The essential oil was ranged from 4.44 ± 0.55% to 4.96
± 0.30% and the non-blanching sample and blanching sample for 30 minutes (37 and 31 compounds of essential oils,
respectively) could detect more compounds than the other blanching conditions. Among detected compounds in essential oil,
there were three main compounds, Ar-turmerone (33.37 ± 0.08%), Turmerone (32.27 ± 0.11%) and Curlone (20.66 ± 0.15%)
in the essential oils of turmeric. Blanching turmeric before drying did affected on the quality of turmeric, especially curcumin
content and essential oil yield and essential oil composition.

Keywords: blanching, phenolic content, curcuminoids, essential oils

I. Introduction 80% of the world’s supply of commercial turmeric. Between

this, the rhizomes have been harvested then dried before
As one of the important spices, turmeric known as
distributing to the world [2]. Regarding to this, Cambodia is
Curcuma longa and belong to Zingiberaceae family is an
one among the Asian countries that cultivated and used
indigenous plant of India. This spice has been used as a
turmeric since ancient time for any cooking recipe such as
seasoning, coloring agent, preservative, flavouring as well
curry, amok and other Khmer foods to give the aroma and
as folk medicine to treat many types of disease since ancient
beautiful yellow color to the dishes.
time and also noted to be used as cosmetic application. The
Based on the potential application of turmeric due to its
special color from turmeric is yellow and it came from the
bioactive compounds, especially curcumin compound and
presence of the compound which is named “Curcuminoids”
the huge amount of the turmeric plants in Cambodia, hence
[7]. In like manner, about 400,000 tone of turmeric were
turmeric rhizomes became a potential spice for apply in
produced by India in fresh weight per year and it was about
different important sectors. Between this, the drying and

210
blanching process have both positive and negative effect on 2.3.3. Curcumin analysis by HPLC
the quality and the content of those bioactive compounds,
HPLC was used for quantification the curcuminoids
therefore the studying of the effect of blanching on quality
from turmeric and the method was modified from Hirun et
of turmeric with different blanching condition of dried
al.[4]. A Shimadzu LC-20AT HPLC system connected to a
turmeric have been investigated.
Shimadzu SPD-M20A UV-VIS detector and apHera C18 –
Polymer (25cm x 4.6mm,5µm) column was use in this
II. Materials and Methods
experiment. (60:40 of acetonitrile and acidified water with
2.1. Sample preparation acetic acid 1%) were used as mobile phase with 0.7ml/min
and 25min, 5µL of curcuminoids were detected at 425nm.
Fresh turmeric rhizome was obtained from a farm at
Kampong Cham province, Cambodia. After blanching in 2.3.4. Essential oil and composition analysis
boiling water for 15, 30 and 45 minutes and then transf-
erred to oven, the samples were packed with aluminum Essential Oils were extracted from dried turmeric
paper and stored in freezer at -20°C for further analysis. rhizomes by hydro-distillation in Clevenger’s type apparatus
and the results were expressed as % of Essential Oils yield
2.2. Materials (AOAC 2012).
The method was modified from Tran et al.[9] for GCMS
Folin-Ciocalteu (Sigma, USA), gallic acid (Himedia.
analysis. In chromatographic system, the polar phase
India) and sodium bicarbonate (Merck, Germany) are
capillary column, TG-5MS (acid-deactivated polyethylene
phenolic’s reagents. Ascorbic acid(Merck,Germany) and
glycol) with the size of 30 m length × 0.25 mm internal
DPPH ( sigma, USA ) were used for Total antioxidant
diameter × 0.25µm film thickness (Thermo Fisher Scientific,
activity analysis. Curcumin, bisdemethoxycurcumin and
USA) were performed on a GC-2010 Plus chromatograph
demethoxycurcumin (sigma, USA, purity ≥ 98%) were
(Shimadzu, Japan) with AOC-20i autosampler (Shimadzu,
use as standard of curcuminoids analysis. Acetonitrile
Japan). GC-MS were operated under the following
(Merck, Germany), acetic acid (Sigma,USA) and water
conditions : Carried gas He; flow rate 1.0ml/min; split 1:50,
grade (Merck, USA) were used for HPLC analysis of
injection volume 1.0 µl; injection temperature 250°C; oven
curcuminoids. Methanol (Merck, Germany) was used as
temperature progress included an initial hold at 50°C for 2
solvent for extraction and dilution in this experiment.
min, a rise to 80°C at 2°C/min, a rise to 150°C at 5°C/min,
a rise to 200°C at 10°C/min and a rise to 300°C at 20°C/min
2.3. Analytical methods
for 5 min. Total running time is 46 min.
2.3.1. Total phenolic content
III. Results and Discussion
The determination of Total phenolic compounds was
done by followed Wijaya, [10] with slightly modification 3.1. Total phenolic content of the extract
by using Folin-ciocalteu assay. The mixture was measured
Table.1 were presented the total phenolic of dried
with UV-Vis spectrophotometer at 765 nm. Result was
turmeric. The highest value of total phenolic content was
expressed as mg gallic acid equivalent/g of dried turme-
recorded in dried turmeric at 55°C with blanching 30
ric (mgGAE/gDT).
minutes. Moreover, the results shown a different number but
not significantly different between the results of TPC for
2.3.2. Antioxidants activity of the extracts
these 4 conditions in this study (p>0.05). Bamidele et al.[1]
The radical scavenging activity of the extracted turmeric stated that there might be some increasing of total phenolic
was determined by using the DPPH free radical activity (2,2- content while increasing the blanching time. The increasing
diphenyl-1picrylhydrazyl) according to Choi et al., [3]. The in the total phenolic content were attributed to the reduction
mixture of reagents and sample was then measured with the of enzyme-mediated polyphenol degradation (complete
absorbance 517nm by using the spectrophotometer. The inactivation of native polyphenol oxidase) and might due to
result was as expressed mg VCE/g DT. the release of bound phenolic acids from the breakdown of
cellular constituents of cell wall.

211
3.2. Antioxidant activity of the extract revealed the significantly different among non-blanching
and blanching condition of bisdemethoxycurcumin and
Antioxidant activities of dried turmeric in this study have
demethoxycurcumin in dried turmeric. From the results of
shown a different number (Table 1.) but there are not
this current study, it could be indicated that blanching has
significantly different (p>0.05) among the studied condition.
affected on the amount of bisdemethoxycurcumin and
Not far from total phenolic content, 30 minutes was a good
demethoxycurcumin as it decreased while the blanching
condition among 3 blanching condition which could remain
time increased to 45 minutes. Non-blanching condition
higher of antioxidants activity. This meant that blanching
could remain the high value of curcuminoids and blanching
process could prevent the major loss of antioxidant activity,
15 to 30 minutes could preserve the curcuminoids with
but it should be performed with the minimal heat treatment.
higher amount if compared to blanching 45 minutes.
The other publication has said that, it is possible to believe
that total phenolic group are highly responsible for
3.4. Essential oils of dried turmeric rhizomes and its
antioxidants activity of plants, that is why both of them have
composition
strongly correlation. [6].

Table 1. Total phenolic content and antioxidant activity The highest value of essential oil from turmeric was
of the extract from dried turmeric found in dried turmeric at 55°C with blanching 30 minutes
(4.96 ± 0.30 %) while the lowest one was found in blanching
TPC Antioxidant 45 minutes condition (4.04 ± 0.05 %). The result of these 4
Sample (mg GAE/g DT) activity conditions showed in Table 3. was not significantly different
(mgVCE/g DT) (p value > 0.05).
NB 105.47±5.37a 9.60±0.18a
a
B15 93.42±4.53 9.44±0.08a Table 3. Percentage of essential oils yield
a
B30 107.62±6.46 9.58±0.11a
B45 96.28±5.35a 9.37±0.18a Condition % yield of EOs
Values are presented as mean (n=3). Means with different NB 4.82 ± 0.25a
superscripts alphabets in the same column are significantly B15 4.84 ± 0.53a
different (p<0.05). B30 4.96 ± 0.30a
B45 4.44 ± 0.55a
3.3. Curcuminoids content by HPLC Mean with different small letters (a,b) in the same column
Table2. Curcuminoids content in dried turmeric are significantly different.. Number of replicates = 2
rhizomes
A studying has done on the seed and plants, and it has
Sample Curcumin Bisdemethoxy Demethoxy- been claimed that blanching the oil seeds could be a
(%) -curcumin curcumin promising technique to improve the yield and quality of the
(%) (%) oil. It is an inexpensive technique that disintegrates the
NB 2.62 ± 0.14a 2.20 ± 0.07b 1.24 ± 0.05b seed’s cell walls and increases the extractability of the
B15 2.24 ± 0.06a 1.89 ± 0.13a 1.03 ± 0.01a intracellular material and the extraction method could be the
B30 2.33 ± 0.20 a
1.90 ± 0.03 a
1.05 ± 0.07a most effect factor of essential yield [5]. Link with this
B45 2.25 ± 0.24 a
1.78 ± 0.12 a
1.00 ± 0.08a result, blanching 30 minutes has been noticed as the
Mean with different small letters (a-b) in the same optimum condition which could maintain the essential oil
column are significantly different at p<0.05. Number of with higher amount if compared to other studied conditions.
replicated = 3. Around 46 compounds were detected in turmeric oil for this
study among 4 conditions, and the similarity of the
The result was showed that the curcumin, bisdemethoxy- compounds were compared with compound library NIST11s
curcumin and demethoxycurcumin was the highest in non- which percentage was higher than 90%. Table 4 only shows
blanching condition and dried at 55°C while the lowest one the main compounds in dried turmeric oil with 4 condition
was found in blanching 45 minutes (Table. 2). The result of studying.

212
 Table 4. Percentage of main compounds of essential oils References
yield [1] Bamidele, O., Fasogbon, M., Adebowale, O., Ade
yanju, A., 2017. Effect of blanching time on total
Phenolic content. Current Journal of Applied Scie
nce and Technology. 24, 1–8.
[2] Borah, A., Hazarika, K., Khayer, S.M., 2015. Dr
ying kinetics of whole and sliced turmeric rhizom
e in a solar conduction dryer. Information Process
ing in Agriculture. 2, 85–92.
[3] Choi, Y., Ban, I., Lee, H., Baik, M.Y., Kim, W.,
2019. Puffing as a novel process to enhance the a
ntioxidant and anti-inflammatory properties of Cu
rcuma longa L. (turmeric). Antioxidants MDPI. 8,
RI (Cal): Retention Index, identify based on the calculation 2–4.
of RI value which relative to C7-C31n-alkanes on TG-5MS [4] Hirun, S., Utamaang, N., Roach, P.D., 2014. Tur
Column, % are given by the mean of two determinations. (-) meric (Curcuma longa L.) drying: an optimization
None detected and not found. approach using microwave-vacuum drying. Journ
The differences number of compositions may due to al of Food Science and Technology. 51, 2127 – 2
some factors such as method of blanching and parameters 133.
that are characteristic of the product subjected to blanching. [5] Kaseka, U.L.O., 2006. Physicochemical attributes.
For blanching, in some case of formation new compounds
Bioactive Compounds. 6, 234–289.
after blanching or cooking process has been recorded
probably because of oxidation reactions, hydrolysis of [6] Oboh, G., 2005. Effect of blanching on the antiox
glycosylated forms, or the release of compounds by the idant properties of some tropical green leafy vege
rupture of cell walls [8]. tables. LWT - Food Science and Technology. 38,
513–517.
IV. Conclusion [7] Osorio-Tobón, J.F., Carvalho, P.I.N., Barbero, G.F.
,Nogueira, G.C., Rostagno, M.A., Meireles, M.A.
The study on determination of bioactive compounds in D.A., 2016. Fast analysis of curcuminoids from t
dried turmeric rhizomes from Kampong Cham province urmeric (Curcuma longa L.) by high-performance
with different conditions of blanching (NB, B15, B30 and liquid chromatography using a fused-core column.
B45) prior to oven drying at 55°C has been investigated. Food Chemistry. 200, 167 – 174.
Blanching turmeric before drying did affected on the quality [8] Ratseewo, J., Tangkhawanit, E., Meeso, N., Kaew
of turmeric, specially curcumin content and essential oil seejan, N., Siriamornpun, S., 2016. Changes in an
composition. Among 3 blanching conditions, dried turmeric tioxidant properties and volatile compounds of kaf
with blanching 30 minutes is the optimum condition which fir lime leaf as affected by cooking process. Inter
could preserve yield of crude extracts, total phenolic content, national Food Research Journal. 23(1), 188–196.
antioxidants activity, yield of essential oils and its [9] Tran, T.H., Ha, L.K., Nguyen, D.C., Dao, T.P., N
compositions and curcuminoids content close to non- han, L.T.H., Nguyen, D.H., Nguyen, T.D., Vo, D.
blanching condition. V.N., Tran, Q.T., Bach, L.G., 2019. The study on
extraction process in essential oils of black peppe
Acknowledgement r (Piper nigrum L.) seeds harvested in Gia Lai Pr
Thank to my advisor Dr. IN Sokneang for all the ovince, Vietnam. Processes. 7(2), 56.
supporting and thankful to the world bank (HEIP Project) for [10] Wijaya, C., 2018. Optimization of Extraction time
their financial support. and temperature for Java tea. Food and Nutrition
Open Access. 2, 1-8.

213
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Production of White Pepper from Ripe Pepper Berries (Piper nigrum L.)

Seanghai Hoeun 1, Sreyleak Meas 1, Elen Morm 1 and Sokneang In 1,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* in@itc.edu.kh

Abstract

White pepper (Piper nigrum L.) is commonly produced from ripe pepper berries by removing the outer skin before drying.
This spice is preferred using by consumers due to its light color, milder flavour and pungency. White pepper exists in the
market as the whole and ground form. It is commonly used in food products such as sauces, soups, salad dressing and
mayonnaise. The main purpose of this study is to determine the suitable conditions for white pepper production from ripe
pepper berries. The study was conducted in different blanching times (0, 1, 3, and 5 min), soaking in water for 24 hours and
drying at 65ºC for 16 hours. Drying kinetic, color measurement and bioactive compounds including total phenolic compound,
total antioxidant capacity, essential oils and an alkaloid (piperine) content were evaluated. According to the results, soaking
at 24 hours within non blanching and drying at 65°C for 12 hours, was considered as a suitable condition for white pepper
production However, total phenolic content and total antioxidant capacity decreased in overall for white pepper after
processing due to the skin removed while the piperine content and total essential oil remained stable.

Keywords: Blanching, color, Piper nigrum L., piperine and white pepper

I. Introduction According to Aziz et al. [3], blanching time should be from

3 to 10 min to restrain the loss of color, aroma, and essential
In general, white pepper is produced from fully ripe
oil components of pepper. However, blanching for less than
berries by removing their pericarp before drying. This spice
1 minute is insufficient to remove the outer skin. Even
could provide economical, nutritional, and medicinal
though, the process of white pepper has been studied and a
benefits [1]. Also, it is used in food industries and
lot of study has been done in other countries. Therefore,
pharmaceutical industries for medicinal purposes. In
white pepper processing techniques should be developed,
European countries, white pepper is commonly used in light-
improved and optimized to obtain better white pepper
colored sauces, soups, salad dressing, mayonnaise, balsamic
quality and provide information to the local farmer.
tomatoes and mostly used as a flavoring agent [2]. As for
processing technique, it is quite different from area to area II. Materials and Methods
and from country to country. For instant in Malaysia, the
2.1. Sample preparation
pepper berries were soaked in water one or two weeks to
remove the skin before drying, while in Cambodia, the The ripe pepper berries (Khmer cultivate) were collected
berries were boiled in water for a certain period, from Rosemoric farm in Ratanakiri province in April 2020
subsequently soaked in water to remove the skin prior drying. and transported by storing in the icebox. The pepper berries

214
were washed with distilled water thrice to remove the
The effect of pretreatment and soaking on drying kinetic
foreign matters. Then, the ripe berries were pretreated in
of white pepper produced from ripe pepper berries in
boiling water for 1, 3, and 5 min and soaked for 24 hours.
different pretreatment conditions (0, 1, 3, and 5 min) after
The soaking condition was also applied to non-boiling
drying at 65 ºC for 16 hours was shown in Fig. 1.
peppers. Next, they were rubbed manually to remove
D65-NB, D65-1B, D65-3B, and D65-5B represented
pericarp and rested. Eventually, the white pepper was
pepper berries after drying at 65ºC for 16 hours at different
weighted prior drying.
pretreatment conditions including without pretreatment,
2.2. Drying kinetic pretreatment for 1, 3 and 5 min, respectively.
All the samples were oven-dried and compared to un-
40
pretreated pepper berries. The amount of 100 g of green D65-NB1S
D65-1B1S
pepper berries were oven-dried at the optimum condition of

Moisture content (%)

D65-3B1S
30 D65-5B1S
65ºC for 16 hours. The samples were taken out and weighted
for every one hour.
20

2.3. Crude extraction

10
An amount of 0.5 g of ground pepper was extracted with
10 ml of methanol then placed in the ultrasonic tank 120 W 0
(VWR International bvba/sprt, Malaysia) for 20 min. After 0 2 4 6 8 10 12 14 16 18

Time (h)
extraction, the solvent was centrifuged (Rotofix 32A,
Germany) at 2000 rpm for 15 min. The clear supernatant was Fig. 1. Drying kinetic of white pepper
poured out by putting in the evaporated flask and the The drying kinetic indicated that drying from the early
extraction was repeated 3 times with 10 ml of methanol stage to 8 hours, each moisture content went down similarly.
added. The rotary evaporator (IKA RV-10 Control, IKA, After drying for 8 hours, pretreated sample for 5 min
Germany) was performed to evaporate the solvent until the declined sharply and followed by pretreated sample for 1
sample was dried. The extract was diluted with 10 ml of and 3 min, respectively. Meanwhile, the unpretreated sample
methanol and stored at -20ºC for further analysis. moderately decreased. The results showed that the longer the
pretreatment in boiling water improved the drying kinetic of
2.4. Initial moisture content, color, and bioactive
white pepper. All moisture content of all conditions
compounds analysis
remained stable after 12 hours drying. The pretreated sample
The initial moisture content of peppercorns was for 1, 3 and 5 min could reduce the moisture content to lower
determined by followed an analytical method AOAC than 10-13 % for 8 hours, while the unpretreated sample was
(Association of Official Analytical Chemists) with a slight about 15%. In general, the moisture content of white pepper
modification [4]. The fresh pepper was oven-dried at 105ºC is around 12 to 13% [9]. Additionally, the yield of white
for 24 hours. Color of white pepper after peeling and drying pepper obtained from ripe pepper ranged from 35 to 37%.
process was measured by using Chroma meter (CR-400, The obtained results showed higher yield than the study Aziz
Konica Minolta). The total phenolic content (TPC) of the et al. [3] that found the mature berries were able to yield
pepper was determined by the Folin-Ciocalteu’s reagent test about 25-28 % of white pepper.
[5, 6]. The antioxidant activity was determined by using
3.2. Color
DPPH (2,2-diphenyl-1-picrylhydrazy) radical scavenging
capacity assay with the slight modifications [7]. Piperine The result of the chromatic coordinates L*, a*, b*value of
content was detected by HPLC (High Performance Liquid peeled pepper and pepper after drying at 65 ºC for 16 hours
Chromatography, Shimadzu, LC 2010A, Japan) coupled were indicated in Fig. 2.
with UV detector [8].

III. Results and Discussion

3.1. Drying kinetic of white pepper

215
 RP represented fresh ripe pepper berries. PNB, P1B, P3B, Table 2. The amount of TPC and TAC
and P5B represented the peeled pepper berries in different
pretreatment conditions including without pretreatment, Sample TPC TAC
pretreatment for 1, 3 and 5 min, respectively. (mg GAE/g DW) (mg Vit.CE/g DW)
RP 6.75 ± 0.45d 3.71 ± 0.26e
50 Lightness(L*)
a
Red/Green(a*) Yellow/Blue(b*)
a PNB 2.60 ± 0.27a,b,c 1.08 ± 0.15a
a
a a
a

a a P1B 3.35 ± 0.32c 1.64 ± 0.06b,c,d

40 a
P3B 3.12 ± 0.54c 2.05 ± 0.40d
± 0.39b,c ± 0.18d
Color value

30
P5B 2.95 2.06
b
D65-NB 1.96 ± 0.04a 1.17 ± 0.09a,b,c
b b
20
a
a,b a,b
a,b a,b D65-1B 1.90 ± 0.06a 1.23 ± 0.05a,b,c
D65-3B 2.14 ± 0.07a,b 1.17 ± 0.11a,b
10
a a a a a a a
D65-5B 1.91 ± 0.15a 1.70 ± 0.06c,d
a
The lowercase letter represents a significant different for each condition (p-
0
B B B B B B B B value < 0.05)
PN P1 P3 P5 5-N 5-1 5-3 5-5
D6 D6 D6 D6 The TPC of fresh ripe pepper berries was found the
Sample highest value compared to peeled pepper and dried white
Fig. 2. The color value (L*, a*, b*) of white pepper
pepper. The obtained results were similar to the study of
The lightness L* value was found 42.12 ± 2.44, 42.27 ± Aziz et al. [3] that found the TPC of white pepper contained
0.12, 42.13 ± 0.68, and 44.37 ± 1.52 on peeled pepper berries around 3-6 mg GAE/g dried pepper. However, the TPC of
PNB, P1B, P3B, and P5B, respectively. For dried pepper dried white pepper was not significantly different, but lower
berries, lightness L* value was 39.56 ± 2.68, 37.60 ± 1.35, than peeled and fresh pepper. The loss of TPC was probably
39.42 ± 2.21, and 38.77 ± 8.47 of D65-NB, D65-1B, D65- due to the fact that the phenolic compounds are mostly
3B, and D65-5B, accordingly. The lightness L* value of all located in the mesocarp of pepper berries [11]. Therefore,
the conditions of peeled pepper and dried pepper were the TPC lost while the skin of pepper berries was removed
similar. The redness/greenness a* value of peeled pepper and to produce white pepper.
dried pepper conditions were not significantly different. The TAC of ripe pepper berries (RP) was greater value
Furthermore, the yellowness b* value of peeled pepper than peeled and dried pepper berries samples. Both TAC
condition were increased slightly. For dried pepper berries, results of PNB and D65-NB were found similarly and lower
the yellowness b* value of D65-NB and D65-1B showed than the one of peeled and dried pepper berries condition.
lower than the one of D65-3B and D65-5B. It seems that For the dried pepper conditions, the TAC of D65-5B showed
drying condition has no influence on the color of white higher compared to D65-NB, D65-1B, and D65-3B. It was
pepper. According to Mey [10] found that drying below demonstrated that the pretreatment could be increased the
70ºC did not affect the color of pepper because it could not amount of antioxidant capacity except drying slightly effects
deactivate the enzymes responsible for browning reaction of on antioxidant capacity in white pepper after processing.
color. Comparison between TPC and TAC, there showed that
3.3. Total phenolic compound and total antioxidant TPC and TAC had a positive strong correlation (0.929) in
capacity white pepper. From the previous study, Eyenga et al. [12]
stated that all of the antioxidant activities were positively
The results of TPC and total antioxidant capacity (TAC) and strongly (> 90%) correlated with TPC. It makes sense
were described in Table 2 and expressed as milligram of that antioxidant activity was declined if the TPC was
gallic acid equivalent (GAE) in gram of dried pepper and decreased.
milligram of vitamin C equivalent (Vit.CE) in gram of dried
pepper. 3.4. Essential oil and piperine
As shown in Table 3, the amount of essential oil and
piperine in pepper were expressed as percentage (% w/w).

Table 3. The percentage of essential oil and piperine in

white pepper

216
Sample Essential oil (%) Piperine (%) IV. Conclusion
RP 2.22 ± 0.08a 3.58 ± 0.24a White pepper produced from ripe pepper berries with
PNB 2.21 ± 0.01a 5.37 ± 0.61b soaking at 24 hours without pretreatment and drying at 65°C
P1B 3.02 ± 0.37a 5.68 ± 0.18b,c
for 16 hours, was considered as a suitable condition for
P3B 3.11 ± 0.19a 6.08 ± 0.16b,c
P5B 2.86 ± 0.02a 5.39 ± 0.25b white pepper production because it obtained higher
D65-NB 2.44 ± 0.38a 6.29 ± 0.15c production yield, better color, high amount of bioactive
D65-1B 2.63 ± 0.16a 4.33 ± 0.18a compounds. The total phenolic content and antioxidant
D65-3B 2.26 ± 0.41a 3.62 ± 0.35a activity decreased after processing while the piperine
D65-5B 2.57 ± 0.19a 3.86 ± 0.26a content and total essential oil remained stable.

According to Table 3, the essential oil yield of all ACKNOWLEDGMENTS

samples were not significantly different. The study of Weil This research has been partially supported by funds from
et al. [13] mentioned that blanching caused the structure of the ARES-CCD.
the peppercorn may protect the volatile compounds from
transfer during processing. However, pretreatment in boiling REFERENCES
water and soaking for one day had no effects on the yield of [1] Olalere, O.A., Abdurahman, N.H., Alara, O.R.,
essential oil from white pepper. Moreover, the essential oil Habeeb, O.A., 2017. Parametric optimization of
cells were mainly in the inner portion of the skin [11]. microwave reflux extraction of spice oleoresin from
According to Jayatunga and Amarasinghe [14], drying white pepper (Piper nigrum). Journal of Analytical
pepper at 65ºC obtained the highest yield of essential oil. Science and Technology. 8, 1–7.
Therefore, there was no impact of pretreatment, soaking and [2] Azam Ali., 2007. Pepper processing. Technology
drying on the essential oil content of white pepper produced challenging poverty. 44, 1–8. Om, C., Daily, F.,
from ripe pepper berries. Based on FAO (Food and Vlieghe, E., MaLaughlin, J.C., Mclaws, M.L., 2017.
Agriculture Organization) [9] the essential oil contained 1.5 % [3] Aziz, N.S., Sofian Seng, N.S., Razali, N.S.M., Lim,
was classified in grade I and II and 1.0 % of essential oil was S.J., Mustapha, W.A.W., 2019. A Review on
classified in grade III. Thus, the percentage of essential oil Conventional and Biotechnological Approaches in
results was about 2-3 % which classified in grade I white White Pepper Production. Journal of Science of Food
pepper. and Agriculture. 99, 2665–2676.
In term of piperine content, ripe pepper berries was [4] AOAC, 2000. Official methods of Analysis. 4th ed.
found the piperine content (3.58 ± 0.24 %) and there was not Association of Official Analytical Chemist,
significantly different for piperine value in peeled pepper Washington DC, USA.
berries. For peeled pepper berries, the piperine was found [5] Chan, E.W.C., Lim, Y.Y., Wong, S.K., Lim, K.K., Tan,
greater than dried pepper berries. However, the D65-NB S.P., Lianto, F.S., Yong, M.Y., 2009. Effects of
contained the highest piperine among of those from dried different drying methods on the antioxidant
pepper condition. Based on these obtained results, there was properties of leaves and tea of ginger species. Food
no impact of pretreatment and soaking on piperine content, Chemistry. 113, 166–172.
while the drying that had a very slight influence. It could be [6] Rabeta, M.S., Vithayia, M., 2013. Effect of different
explained that the piperine molecule located in the inner core drying methods on the antioxidant properties of Vitex
of pepper grain. As a result, this molecule did not much negundo Linn. tea. International Food Research
effect when the pepper skin removed during processing [11]. Journal. 20, 3171–3176.
The study of Weil et al. [13] showed that piperine contents [7] Bopitiya, D., Madhujith, T., 2014. Antioxidant
were not affected by the wet process as blanching. The potential of rice bran oil prepared from red and white
piperine of those dried pepper berries was around 3-6 %. 4.0% rice. Tropical Agricultural Research. 26, 1–11.
of piperine and it was classified in grade I, 3.5 % classified [8] Sharma, N., Joshi, H.M., Malik, A., Mishra, M.,
in grade II and 3.0% classified in grade III [9]. Singh, B.P., Tripathi, S., Upadhyay, V., 2013.
Development and Validation of Rapid RP- HPLC
Method for Estimation of Piperine in Piper nigrum L.

217
 International Journal of Herbal Medicine IJHM. 1,
6–9.
[9] FAO, 2017. Standard for Black, White and Green
Peppers Cxs 326-2017. Codex Alimentarius.
[10] Mey, P., 2017. Effects of blanching temperature on
quality of black pepper (Piper nigrum L.).
International Journal of Environmental and Rural
Development. 8, 1–6
[11] Attokaran, M., 2011. Natural Food Flavors and
Colorants. 1st ed. Blackwell, Wiley.
[12] Eyenga, M., Youovop, J., Ngondi, J., Sindic, M.,
2020. Temperature dependent studies on nutritional,
total polyphenols, flavonoids content and antioxidant
activities of Aframomum citratum (C.Pereira)
K.Schum and Tetrapleura tetraptera (Schum. &
Thonn.) Taub. fruits. Biotechnol. Agron. Soc. Environ.
2020. 24, 207–220.
[13] Weil, M., Shum Cheong Sing, A., Méot, J.M.,
Boulanger, R., Bohuon, P., 2017. Impact of blanching,
sweating and drying operations on pungency, aroma
and color of Piper borbonense. Food Chemistry. 219,
274–281.
[14] Jayatunga, G.K., Amarasinghe, B.M.W.P.K., 2019.
Drying kinetics, quality and moisture diffusivity of
spouted bed dried Sri Lankan black pepper. Journal
of Food Engineering. 263, 38–45.

218
The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Development of Fermented Small Cucumbers with Different Tastes Using Isolated Lactic
Acid Bacteria

Marinich NET 1, Daly DIM1 and Reasmey TAN 1,2,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: rtan@itc.edu.kh

Abstract

Young cucumber is a popular vegetable in Cambodia and the preservation of this vegetable by fermentation is widely
practiced. However, the methodology of the fermentation process is not specified yet and the quantity of lactic acid bacteria
was not determined clearly and the sensory evaluation was also not remained the same. Regarding this, pure isolated LAB is
the best way to provide the best quality, prolong shelf life and the taste is remained stable. The objective of this study was
the development of fermented cucumber by using pure isolated LAB and shelf life determination by the addition of food
preservatives including sodium metabisulphite and sodium benzoate. To achieve this purpose, the brine solution of
fermented cucumbers was purchased in local markets in Phnom Penh and yogurt was bought from supermarket. Then,
culture and isolated in three times to purified it, and 10 pure isolated LAB were collected to fermented. For physicochemical
characteristics and sensory evaluation were analyzed after fermentation for two days. Acidity value was in range 0.22% to
0.32%, pH was 3.82 to 4.14, salt concentration was 1.62% to 1.75%, and TSS was 2.5 to 3-degree brix of LAB S1, S2 and
S3. They were collected to continue fermentation with the addition of sugar 4% and without sugar addition to choosing
fermentation by blanching at 75℃ in 5min. According to sensory evaluation and physicochemical analysis again, only one
pure isolated lactic acid bacteria was conducted on shelf life base on physicochemical value including total acidity 0.22%,
pH 4.14 and reducing sugar value 0.85% and sensory evaluation. To sum up, sodium metabisulphite is the best one to
preserve shelf life and provide good quality depends on physicochemical analysis and microbial count.

Keywords: Young cucumber, sodium metabisulphite, sodium benzoate, Total plate count, Yeasts, Molds

I. Introduction fermentation is the cheap and easy process to conserve the

Food is one of our basic needs that all humans have vegetables, which can be afforded by most of the
been recognized in the world. Be a part of the food, Cambodian people [1]. Similarly, the principal of the
vegetables play an important role in maintaining proper fermentation process is to preservation, shelf life-extending,
health and provide our body a lot of nutritional [7]. other effects such as the improvement of safety, nutritional
Cambodia is a developing country where has less industrial value, and organoleptic quality of the food. Therefore, the
activities, it is hard to transform or to preserve vegetables best method to store those products healthily and
by any modern methods besides fermentation because economically is lactic acid fermentation. Fermentation
process needs less energy simple for handling and storage surface and spin for 30s to distribute the 0.1ml of dilution
without cooling and easy method for handling of raw evenly over the entire agar surface. The plates were
material before further processing. Based on the benefits of incubated in aerobic at 37℃ for a period of 48h. After a
vegetable, fermented products become the common foods 48h incubation period, the colonies were counted in the
which were utilized in the local [8]. The object of this range 30-300 CFU/ml.
study was the development of fermented small cucumbers 2.3. Sensory evaluation
with different tastes using isolated LAB by selecting the A sensory evaluation test using descriptive test was
pure isolated LAB from brine solution of fermented evaluated by 16 panelists and affective tests was evaluated
cucumber and yogurt in cucumber fermentation and to by 30 panelists participate using 9 points Hedonic scale.
observe the shelf life in cucumber fermentation with 2.4. Statistical data analysis
addition food preservative base on physicochemical All data were subjected to determine the significant
analysis and number of microbial count. difference among mean on those samples by one way
II. Materials and Methods ANOVA analysis of variance in SPSS version 16.0. The
predetermined acceptable level of probability was 5% for
2.1. Sampling sites
all comparisons (p≤0.05).
Raw materials were used in cucumber fermentation
III. Results and Discussion
including fresh cucumber, salt, and sugar were purchased
from Kilo Lekh buon market. 3.1. The influence of food preservative on total acidity
2.2. Analytical methods The total acidity value store in room temperature and
2.2.1. Total acidity 4ͦC were shown in Figure 1. In week 0, sodium
The total acid produced during fermentation is metabisulphite was the highest quantity of total acidity
determined by the method acid-base titration. At each came behind by control and sodium benzoate respectively
sampling time, 10ml of brine is placed in Erlenmeyer and 3 0.14%, 0.07%, and 0.06%. In contrast, sodium
drops of phenolphthalein are added. Then the titration with metabisulphite was the lowest amount of total acidity in
NaOH 0.1N is going until the light pink color persisted [6]. week 1 and week 3, whereas, Sodium benzoate and control
2.2.2. Determination of pH contained a similar value around 0.40% and 0.39%
The pH of the fermented liquors was determined at the respectively.
specified interval by a pH meter (Model HI2020, HANNA
instruments Woonsocket, Romania) [5].
2.2.3. Salt concentration
The salt concentration is determined by using a salt
meter (Model ES-421, ATAGO Co., Ltd, Japan). The value
of salt concentration is indicated in percent [5].
2.2.4. Total soluble solid
Brix value was measured by a refractometer to
determine the total soluble solids in the liquid [5]. Only a
few drops of the sample were required to drop on the prism Fig 1. Total acidity store in room temperature and 4oC
surface.
For storage condition 4ͦC, sodium metabisulphite is the
2.2.5. Determination of moisture content
best food preservative because it is the lowest quantity of
To determine moisture, samples were weighed 5g and
total acidity, and it is small amount of microorganism
dried for 3 hours in an oven at 105±2°C and it is rested in a
contain in this brine solution.
desiccator for 15mn then the dried sample is weighed [2].
3.2. The influence of food preservative on pH Value
2.2.6. Determination of yeast mold count
Figure 2 illustrates pH value store in room temperature
Appropriately 15 to 20ml of agar was poured in the
and at 4℃. In week 0, control, sodium benzoate and
sterile Petri dishes and allows them to solidify. The marked
sodium metabisulphite had pH value 3.88, 4.62, and 3.91
sterile Petri dishes were inoculated with 100µl of
respectively. Furthermore, these 3 samples were also
appropriate tenfold serial dilution of the samples by using
significantly different with p-value (p ≤ 0.05) in
brine. Place the bent portion of the glass rod on the agar
respectively 3.27, 3.41, and 3.61 of control, sodium The total soluble solid value of brine solution in room
benzoate and sodium metabisulphite. temperature condition was proved in Figure 4. It noticed
According to the result at 4℃, pH value of these two sodium metabisulphite and control had the same total
preservatives and control samples were declined soluble solid value in week 0 in value 1.7 degrees Brix and
significantly from week 0 to week 1 despite they dropped sodium benzoate was the biggest value in week 0 2 degrees
slightly from week 1 to week 3. As a consequence, sodium Brix. As well as to this, sodium benzoate and sodium
metabisulphite was the preservative that was the best one to metabisulphite were in the same value from week 1 to
preserve cucumber fermentation and prolong longer shelf week 3 in 3 degrees Brix while control was also stable in 2-
life in storage conditions in 4℃. degree Brix from week 1 to week 3. In assumption, sodium
benzoate and sodium metabisulphite were considered as
adding into cucumber fermentation to prolong shelf life at
room temperature. However, for condition 4℃, it was
pointed out that sodium benzoate and sodium
metabisulphite were in the same value each week, and they
were rose quickly from week 0 to week 1 but they were
remained stabled from week 1 to week 3.

Fig 2. pH value store in room temperature and 4oC

3.3. Food preservative influence on salt concentration

The salt concentration value in brine solution by
addition of food preservative and store in room temperature
and refrigerator at 4℃ show in Figure 3. Depending on this
figure, it was noted that salt concentration was declined
dramatically from week 0 to week 3 both in room
Fig 4. Total soluble solid store in room temperature and 4oC
temperature and 4℃ storage condition. It is agreed with
this study that salt extracts liquid from the vegetable which
serves as a substrate for the growth of lactic acid bacteria 3.5. Food preservative influence on moisture content
[4]. In addition, control was the lowest amount of salt Figure 5 was observed about the moisture content value
concentration each week for 4℃ but RT was only in week1 of brine solution in cucumber in room temperature and 4℃
and week 3. However, sodium benzoate was the highest storage condition. In obviously, sodium metabisulphite was
quantity in week 1 and week 3 both in RT and 4℃ too. the lowest amount of moisture content from week 1 to
Especially, sodium metabisulphite was the food week 3 both in RT and 4℃ conditions in respectively 95.03%
preservative that contained a large quantity of salt to 97.04% and 94.78 to 95.23%. Besides, control had the
concentration 1.44% and 1.41% in week 1 and week 3 largest amount around 99.78% and came after by sodium
respectively in RT and 1.37%, 1.31% in week1, and week 3 benzoate in week 3 for RT condition in respectively. Apart
in 4℃. from this, sodium benzoate was the highest amount in
wee1 and week 3 by approximately 95.08% and 97.05%.

Fig 3. Salt concentration store in room temperature and 4oC

3.4. Food preservative influence on total soluble solid Fig 5. Moisture content store in room temperature and 4oC
3.6. Result of yeast and mold count IV. Conclusion
Figure 6 shown yeast mold count of brine solution with
According to the result, LAB S1, S2, and S3 were
the addition of 2 different preservatives. It displayed that
chosen for studying the shelf life. To prolong the shelf life
sodium benzoate and sodium metabisulphite were declined
of cucumber fermentation, sodium benzoate and sodium
considerably from week 0 to week3 both in RT and 4℃. On
metabisulphite 0.1% were added into the brine solution
the other hand, control increased massively from week 0 to
compare to the control sample. Following this experiment,
week 3 not only in RT but also at 4℃. Control was the
it was found that sodium metabisulphite is the food
highest amount of yeast mold count both in RT and 4℃
preservative that can prolong longer shelf life than sodium
from week 1 to week 3 in respectively 4.39 log CFU/ml to
benzoate depending on data of physicochemical parameters
4.67 log CFU/ml and for 4℃, they were 4.18 log CFU/ml
including total acidity 0.22%, pH 4.14 and reducing sugar
to 4.52 log CFU/ml. In contrast, sodium metabisulphite
value 0.85% and the number of microorganisms count.
was the lowest quantity of yeast mold count from week 0 to
week 3 in respectively 2.54 log CFU/ml, 1.69 log CFU/ml,
and 1.39 log CFU/ml for 4℃. Along with this, they were References
2.81log CFU/ml, 2.35log CFU/ml, and 2.35 log CFU/ml in [1] Chrun, R., Hosotani, Y., Kawasaki, S., Inatsu, Y.,
respectively in each week in RT condition. 2017. Microbioligical hazard contamination in
fermented vegetables sold in local markets in
Cambodia. Biocontrol Science. 22, 181–185.
[2] Defrise, M., Grangeat, P., 2010. Analytical methods.
Tomography. 79 , 21–62.
[3] Egbe, J., Lennox, J., Rao, P., Anitha, Umoafia, G.,
2017. Lactic acid bacteria profile of fermenting
cucumber in 7% brine solution. Journal of Advances
in Microbiology. 3, 1–8.
[4] Khanna, S., 2019. Effects of salt soncentration on
Fig 6. Yeast mold count
the physicochemical properties and microbial safety
3.7. Result of sensory evaluation of spontaneously fermented cabbage. Food Science
Figure 7 was represented of sensory evaluation in 10 and Human Nutrition. 48 , 67–74.
different isolated lactic acid bacteria of cucumber [5] Nations, U., 2001. Food and agriculture
fermentation on the second day. According to running one organization of the United nations. Vegetable
way ANOVA, all attributes were significantly different product. 26, 98–104.
with p-value (p≤ 0.05) in these 10 LAB except the color. [6] Niro, G.E.A., 2006. A 19 a - Titratable acidity.
Following running one-way ANOVA, it was assumed that Interlaboratory testing of methods for assay. 39 , 1–
pure isolated lactic acid bacteria S1, S2, and S3 were 2.
picked up to confirm fermentation again by adding sugar [7] Pe, I.M., Ferna, A.G., Gallego, J.B., Yoon, S.S.,
4%. Johanningsmeier, S.D., 2013. Fermented and
acidified vegetables. 29, 59–65.
[8] Zhai, Y., 2017. Fermentation cover brine
reformulation for cucumber processing with low salt
to reduce bloater defect. Cucumber Fermentation.
55, 78-81.

Fig 7. Sensory of cucumber fermentation in second day

 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Development of Fermented Young Melon using Isolated Lactic Acid Bacteria

Tetoutdam KONG1, Sreyleang HEAN1 and Reasmey TAN1,2*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: rtan@itc.edu.kh

Abstract

The young melon is a popular crop for Cambodian people. Some of them eat it as a vegetable while others as a pickle.
Inoculation with a pure culture with lactic acid bacteria (LAB) is known a large-scale method for vegetable fermentation.
Therefore, the purpose of this study was to produce high quality and delicious young melon fermented using pure varieties of
different acidic bacteria. Initially, lactic acid bacteria were isolated from four samples of young melon sold on different
markets in Phnom Penh. The isolated LAB were applied for young melon fermentation for 2 days. The analysis
physicochemical parameters, sensory tests for 2 times and microbials were also conducted. Sodium benzoate and sodium
metabisulfite were used as preservatives in the products and their effects were investigated for 2 weeks. Based on the results
obtained, fresh young melon used in this study had a pH of 6.30, a moisture content of 93.74%, a total acidity of 0.25%,
reducing sugar of 1.97% and an ash content of 0.53%. Moreover, four strains of LAB were found; S1, S2, S3 and S4. After
the first sensory test, fermented young melon products with S2, S3 and S4 were selected to be tested again to select the best
one that give good quality in term of crunchiness and flavor. Consequently, S4 which was isolated from a sample of 7 Makara
market; was considered as the best LAB since the fermented young melon with this LAB got the highest score of flavor and
crunchiness attributes. For long-term preservation of fermented young melon, both sodium benzoate and sodium metabisulfite
could prevent the microbial growth but sodium benzoate is more beneficial than sodium metabisulfite.

Keywords: Young melon, Fermentation, Lactic acid bacteria

I. Introduction traits that can add variety to the diet and is a means of
extending the availability of vegetables beyond their season
Young melon is a kind of hybrid melon that belongs to
[1]. The traditional process usually gives lower organoleptic
the family of Cucurbitaceae and in scientific name called
quality and the fermented vegetables cannot be stored for a
Cucumis melo Linnaeus. These young fruits are occasionally
long period. To guarantee the consistent quality of
harvested and eaten as a vegetable either raw or cooked and
fermentation of vegetables for large-scale production,
they are available year-round. Young melon is a kind of
inoculation with a pure culture of lactic acid bacteria (LAB)
vegetables in Cambodia. The fermentation of vegetables is
should be done. LAB are defined as beneficial bacteria of
an important method for vegetable preservation, especially
which the main metabolites from carbohydrate metabolism
for rural communities, as it gives products with organoleptic
is lactic acid. For the food application, LAB is used in the

223
industry of fermented vegetables where they contribute to III. Results and Discussion
the preservation of organoleptic quality. Therefore, the aim
3.1. Physicochemical characteristics of young melon
of this study is to make fermented young melon products
The physicochemical characteristics of young melon is
with different tastes using isolated LAB. Moreover, the
shown in Table 3.1.
quality of the products would be also investigated.
Table 3.1. Physicochemical characteristics of young melon
Parameters Value
II. Materials and Methods
pH 6.30 ± 0.02
2.1. Materials Moisture content (%) 93.74 ± 0.74
The ingredients for fermentation were fresh young Total acidity (%) 0.25 ± 0.52
melon, salt without iodine and sugar bought from the local Reducing sugar (%) 1.97 ± 0.10
markets. Young melons selected were fresh with lower than Ash (%) 0.53 ± 0.18
27mm diameter and no injuries; and prepared at ambient
temperature in the laboratory before fermentation. 3.2. LAB isolated from fermented young melon
Furthermore, 4 commercially fermented young melon were Four different strains were isolated from commercially
also purchased for LAB isolation. Sodium benzoate and fermented young melon; S1, S2, S3 and S4. S1 and S2 were
sodium metabisulfite were used as preservatives. There was from Orussey market and S3 and S4 were from 7 makara
still other modern equipment which was further described. market; Phnom Penh; Cambodia.
2.2. LAB isolation 3.3. Physicochemical characteristics while fermentation
The samples were first serially diluted with brine 5% 3.3.1. Changes of pH
from 10-1 to 10-7; and then spread on MRS agar and According to Figure 3.3.1 (a). The initial pH in the brine
incubated at 37°C for 24 h. The colonies (LAB) were chosen solution was 6.29 and then the pH dropped significantly to
to isolate on MRS agar and incubate at 37°C for 24h, then less than 4 and approximately 3.5 at day 1 and day 2;
picked up, and stood overnight in 1ml of broth. Next, 700 ml respectively.
of sterilized glycerol was mixed up with broth; and the (a) (b)
mixture was stored in the fridge of -81°C. Finally, LAB were 7 7
harvested. 6 6
5 5
2.3. Fermentation process of young melon 4 4
pH

Young melon was first washed with tap water at least pH

3 3
2 2
three times. The bloom in the bottom of the melon should be 1 1
removed. Young melon was then cut in half and first 0 0
blanched at a temperature at 75 °C for 5 minutes. Then, it S1 S2 S3 S4 S1 S2 S3 S4
Strains of LAB Strains of LAB
was fermented in brine 4% with ratio 1:1 including with Day 0 Day 1 Day 2 Fresh Day 2
adding 4% of sugar. Obviously 10% of each isolated LAB
Figure 3.3.1. Changes of pH in (a) brine solution and (b)
was added to the mixture with different containers. The
fermented young melon
fermentation process was done for 2 days. Finally, 0.1% of
Based on Figure 3.3.1 (b), the initial pH of fresh young
both preservatives were added exclusively and the final
melon was 6.30 but it ranged from 3.54 to 3.83 at day 2.
products were stored with two temperature condition; at
When LAB was inoculated at the beginning of fermentation,
room temperature and 4°C for 2 weeks for further analysis.
it caused a decrease in pH in short period due to the
2.4. Sampling
production of lactic acid from fermentable sugar.
Brine solution, fresh and processed young melon would
3.3.2. Changes of salt concentration
be used to conduct some experiment such as
Depending on Figure 3.3.2 (a), the salt concentration
physicochemical and microbial analysis and sensory
was 4% at day 0 while those ranged between 1.11 to 1.70%
evaluation.
at day 1.

224
 (a) (b) 6.05. For saltiness attribute, S4 had the highest score of 4.35
5 2 while S2 got the highest score; 5.64 of sourness. In this case,
Salt concentration (%)

Salt concentration (%)

4 the gap of sourness between these four strains of LAB were
1.5
3 not much different, so the results would get variable from
1
2 the sensory evaluation. Therefore, to reduce this kind of
1 0.5 problem, there was a preference from the panelists who
0 0 likely to eat pickle or fermented products and the best score
S1 S2 S3 S4 S1 S2 S3 S4 of acceptance of sourness was found around 4.5 to 5.
Strains of LAB Strains of LAB
Day 0 Day 1 Day 2 Fresh Day 2 7
6

Sensory score
Figure 3.3.2. Changes of salt concentration in (a) brine 5
solution (b) fermented young melon 4
During fermentation, some amount of salt in the brine 3
solution absorbed into young melon through osmosis. And 2
1
then the salt was slightly decreased in day 2. The higher 0
NaCl concentration in salt solution resulted a rapid salt S1 S2 S3 S4
Strains of LAB
penetration into young melon while the rate of changes in
Color Flavor Firmness Saltiness Sourness
pH and the total acidity were reduced [2]. According to the
Figure 3.4.1. Sensory evaluation of fermented young
result in Figure 3.3.2 (b), different types of LAB had
melon
influence on the final concentration of fermented young
The 3 selected LAB after the first sensory test, gave
melon. The lowest salt concentration was found in S2 while
similar results to all parameters except crunchiness and
the highest one was found in S4 at day 2.
overall acceptance (Figure 3.4.2). The crunchiness score
3.3.3. Changes of total soluble solids
ranged from 6.48, 7.15 and 7.18 for S3, S2 and S4;
The changes of total soluble solids in fermented young
respectively. For overall acceptance, S4 got highest score of
melon during fermentation are shown in Figure 3.3.3.
7.33 and was followed by S2 and S3. The flavor and
Therefore, fresh young melon had a sugar content of 3 ⁰Brix,
crunchiness could make a big difference for the 33 panelists.
and then at day 2 fermentation, it fell to 2.4-2.7 ⁰Brix.
According to metabolism of lactic acid bacteria [3], they S2 S3 S4
Crunchiness
convert sugar to lactic acid. 7.5
Overall
Color
3.5 acceptance
6.5
3
2.5 5.5
TSS (⁰Brix)

2 Saltiness Odor

1.5 4.5
1
0.5 Sourness Aroma
0
Fresh S1 S2 S3 S4
Strains of LAB Sweetness Flavor

Figure 3.3.3. Changes of TSS in fermented young melon at

Figure 3.4.2. Sensory evaluation of 3 selected fermented
day 2
young melon
So, that means the losses of sugar content in fresh young
3.5. Effect of preservatives on fermented young melon
melon are because sugar is converted to lactic acid. So, S2
during storage
and S4 are the resented the high amount of LAB.
Adding preservatives was done for the best fermented
3.4. Sensory evaluation
young melon chosen, the fermented young melon with S4 of
According to Figure 3.4.1. All the strains of LAB had a
lactic acid bacteria after the sensory tests.
similar result of color and sourness. S4 was the most popular
3.5.1. Physicochemical characteristics
one according to panelists due to its highest score of 6.17 for
Based on Table 3.5.1, the pH decreased slightly for all
flavor. Regarding the firmness, S2 got the highest score of

225
conditions at day 7 and day 14. 3.5.3 (b). Then, none of them was present at day 7 and day
Table 3.5.1. Physicochemical characteristics of fermented 14 at both temperature conditions for the preservative-
young melon using food preservatives containing samples. For the control samples investigated at
Sample Day pH Salt (%) both temperature conditions, it was observed that yeasts and
Control 0 3.36 1.43 molds grew normally at day 7 and day 14.
7 3.24 1.50 (a) (b)
14 3.15 1.49 8 8
7 7
With S. Meta 0 3.63 1.47

log (CFU/ml)
log (CFU/ml)
6 6
7 3.37 1.54 5 5
4 4
14 3.15 1.52 3 3
With S. Ben 0 3.78 1.41 2 2
1 1
7 3.19 1.50 0 0
0 7 14 0 7 14
14 2.62 1.50 Storage time (day) Storage time (day)
In this case, it could be because of replacement of brine Control S. Meta S. Ben Control S. Meta S. Ben
solution. In the low salt concentration, the growth of lactic Figure 3.5.3. Effect of preservatives on yeasts and molds at
acid bacteria occurred [4]. Thus, a change of brine including (a) room temperature and (b) 4℃
with using low salt concentration enhanced lactic acid
bacteria growth. So, that was why the pH decreased. IV. Conclusion
The salt concentrations were approximately 1.45% at
day 0 for all samples. But at day 7, they increased slightly to The fermented young melon using S4 isolated from a
about 1.50%. At day 14, the salt concentration of the sample commercially fermented young melon sold at 7 makara
with sodium metabisulfite was nearly stable but it decreased market was considered as the best one due to its highest
a little for that with sodium benzoate. This might due to the score of flavor and crunchiness after the sensory tests.
osmosis process during storage. It may be that added brine Sodium benzoate and sodium metabisulfite both could
was not equivalent with the salt content in the fermented prevent microbial growth during storage time and preserve
young melon. the products for a long time, but sodium benzoate had a more
3.5.2. Total plate count beneficial health effect than sodium metabisulfite. However,
At day 0, it was detected in all samples based on Figure the shelf-life study should be done by doing more research
3.5.2 (a) and (b). But there was no total plate count for about its characteristics and methodology in the future.
samples with preservatives at day 7 and day 14 at both
temperatures. References
(a) (b) [1] Yamani, M.I., Hammouh, F., 1999. Production of
10 10 fermented cucumbers and turnips with reduced levels
9 9
log (CFU/ml)

log (CFU/ml)

8 8 of sodium chloride.
7 7
6 6 [2] K. Jong-Goon; Hee-Sook, Choi; Sang-Song, Kim;
5 5
4 4 Woo-Jung, K., 1989. Changes in Physicochemical and
3 3
2 2 Sensory Qualities of Korean Pickled Cucumbers during
1 1
0 0 Fermentation.pdf., 838–844.
0 7 14 0 7 14
Storage time (day) Storage time (day) [3] Liu, S., 2003. Practical implications of lactate and
Control S. Meta S. Ben Control S. Meta S. Ben pyruvate metabolism by lactic acid bacteria in food and
Figure 3.5.2. Effect of preservatives on total plate count at beverage fermentations. 83, 115–131.
(a) room temperature and (b) 4℃ [4] Mcmurtrie, E.K., Johanningsmeier, S.D., Jr, F.B., Price,
3.5.3. Yeast and mold R.E., 2019. Effect of Brine Acidification on
Based on Figure 3.5.3 (a), they were both detected in all Fermentation Microbiota, Chemistry, and Texture
samples at room temperature. However, yeasts and molds Quality of Cucumbers Fermented in Calcium or
were not detected in all samples at 4℃ based on Figure Sodium Chloride Brines.

226
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Determination of Antibiotic Resistance of Enterococcus spp. Isolated from Drinking

Water Collected from Stoung District
Chanchakriya SAM 1, Sovannmony NGET 1, 2, Soukim HENG 1,2, Sokneang IN 1, 2, Msateru NISHIYAMA 3,
Toru WATANABE 3 and Hasika MITH 1, 2,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Faculty of Agriculture, Yamagata University, Japan,
1-23 Wakaba-machi, Tsuruoka, Yamagata 997-8555, Japan

* hasika@itc.edu.kh

Abstract

The rapid evolution of antibiotic resistant bacteria is occurring worldwide, threatening the efficacy of
antibiotics. Antibiotic resistance becomes the main problem for public health in developing countries including
Cambodia that suffers from poverty and poor access to safe water in rural areas, especially at the areas around the Tonle
Sap Lake. Therefore, the purpose of the current study was to detect the presence of Enterococcus spp. in drinking water
collected from Stoung District in Kampong Thom province, and to determine their antibiotic susceptibility. The samples
of drinking water were collected from three communes, Trea (land based), Msa Krang (water -land based) and Peam
Bang (water based) in Stoung District. The detection and isolation of Enterococcus spp. were carried out, followed by
antibiotic susceptibility test. Out of 145 samples, 83 samples were found contaminated with Enterococcus spp. The
results showed that, among the 257 isolated strains of Enterococcus spp., 23% were resistant to tetracycline (TC), 20%
to erythromycin (EM), 11% to vancomycin (VCM), 7.39% to doxycycline (DOX), 4.3% to ciprofloxacin (CIP), 2.7%
to levofloxacin (LVX), and 1.2% ampicillin (AMP). In addition, 38% of isolated strains were resistant to at least one
antibiotic agent as resulting in high-risk contamination of Enterococcus spp. from different water samples. The results
also revealed that 11 isolated strains of enterococci were multidrug-resistant. Remarkably, two strains of enterococci
isolated from Msa Krang commune showed high multiple antibiotic resistance to LVX -AMP-DOX-CIP-VA-TC-EM
corresponding to five antibiotic categories. Nevertheless, the presence of antibiotic resistant enterococci in drinking
water at Kampong Thom province could potentially be a serious harm for public health, particularly the communities
in those area.

Keywords: drinking water, Enterococcus spp., antibiotic resistance

I. Introduction Contaminated water and inadequate sanitation are involved

in the pathogenesis transmission, such as cholera, dysentery,
The quality of drinking water is the main crucial factor
hepatitis A, typhoid and polio. Water and sanitation facilities
that could cause infection and diseases [1]. The
that are deficient, insufficient, or poorly handled expose
microbiological quality in drinking water was interested by
individuals to preventable health danger [3]. The bacteria
worldwide because of implied public health impacts [2].

227
can be the reasons of infection in human being when it enters °C (Standard Methods 9230C). After incubation, the c
blood, urine, or wound [4]. Antimicrobial resistance is a olonies were identified as enterococci with light and d
measure of an antimicrobial agent’s decreased ability to kill ark red colonies. The colonies assumed to be enteroco
cci were picked and transferred into Enterococcosel br
or inhibit the growth of a microorganism. This is determined
oth and incubated for 24 h at 37 °C, the broth chang
practically by testing a bacterial isolate in an in vitro system ed into black when the enterococci are present and th
against various antimicrobials. There are many pathogenic en a loop of inoculated broth was streaked on Todd
bacteria that might contaminate in drinking water. But the Hewitt agar (TH, Becton, Le Pont de Claix, France) a
purpose of the current study was to focus on identification nd incubated for 24 h at 37 °C. Then enterococci stai
of antibiotic resistance of Enterococcus spp. potentially ns were picked and transferred into TH broth with pre
present in drinking water. There are several types of sence 15 percent of glycerol. Finally, all isolated strai
ns of enterococci were stored at -80 °C.
antibiotics used in study namely ampicillin, ciprofloxacin,
levofloxacin, tetracycline, doxycycline, vancomycin and 2.3. Antibiotic susceptibility test
erythromycin. The antibiotic susceptibility testing of enterococci was
conducted by using the disk diffusion method recommended
II. Materials and Methods by CLSI (2017). The enterococci strains were grown on
2.1. Sampling sites Todd Hewitt agar to culture the bacteria for 16 h – 24 h from
The current study was carried out at three communes, -80 °C. Then, the colonies were picked into 10 mL of saline
Trea (land-based), Msa Krang (water land-based), and Peam water (0.85%) and compare with 0.5 MacFarland standard
Bang (water-base area) located in Stoung district, in which contains the concentration of standardized 108
Kampong Thom province. The samples were collected in CFU/mL [6]. After that, a sterile cotton swap was dipped
three different villages per commune (Fig. 1) and kept into bacteria suspension and transfer into Mueller Hinton
cooled and sent to laboratory for microbiological analysis. agar (MH, Himedia, Mumbai, India) by streaking the over
The water sample were analyzed within 24h to 48h as they the surface of MH agar. The streaking method of bacteria
keep in 1℃ to 4℃ in the dark [5]. suspension three times at a different angle of 60 ° on MHA
and then, let the plate dry 3 - 5 minutes. The concentration
of antibiotics contains in disks were chosen according to
recommend from CLSI (2017). The diameter of each zone
was measured in millimeter by using digital Caliper and the
interpretation of measurement results as sensitive,
intermediate resistance and resistance was made by CLSI
(2017) that demonstrated in Table 1.

Table 1. CLSI diameter of inhibition zone

Diameter of inhibition zone (mm)

Antibiotics
Susceptible Intermediat Resistance
(S) e (I) (R)

Ampicillin ≥17 - ≤16

Fig. 1. Sampling sites areas in Kampong Thom
Ciprofloxacin ≥21 16-20 ≤15
2.2. Detection and isolation of Enterococcus spp.
Doxycycline ≥16 13-15 ≤12
Enterococci were isolated using membrane filtration
method, considered as “gold standard” for quality asse Erythromycin ≥23 14-22 ≤13
ssment in drinking water by culturing on mEnterococc
Levofloxacin ≥17 14-16 ≤13
us agar (Bacton, Le Pont de Claix, France). A volume
of 100 mL of drinking water sample were filtered th Tetracycline ≥19 15-18 ≤14
rough a membrane filter (0.45 μm pore size, 47 mm
diameter, sterile membrane filters without absorbent pa Vancomycin ≥17 15-16 ≤14
ds, Whatman), then incubated for 24 h – 48 h at 37

228
 III. Results and Discussion antibiotic classes. Those antibiotic classes are β-lactam,
fluoroquinolone, tetracycline, macrolides, and vancomycin.
3.1. Detection and isolation of Enterococcus spp.
257 isolated colonies from 83 samples were conducted disk
The presence of Enterococcus spp. was observed from
diffusion test. As the Fig. 3 show the variation of sensitivity
different commune Trea, Msa Krang, and Peam Bang. In
degree to the test antibiotic in Stoung district. Most strains
comparison, drinking water sample in Msa krang and Trea
are sensitive to ampicillin 96% (n=247), doxycycline 77%,
are mostly contaminated compare to Peam bang commune.
ciprofloxacin 77%, tetracycline 73%, levofloxacin 71%,
The concentration of Enterococcus spp. in Msa Krang have
vancomyci 33% and erythromycin 13% (n=34). There are
a wide range of 1 to 867 CFU/100mL while Peam Bang have
still have some strains are resistant to those antibiotics
concentration range of 0.5 to 776 CFU/100mL. In the
mostly resistant to tetracycline 24% (n=62) and
previous study of [7], the concentration of enterococci
erythromycin 19.84% (n=51).
increase range from 32 to 82× 104 CFU/100mL. The most
contaminated site was from water-land based and most of
the sample detected were filtered water sample. Therefore, 100

Percentage of antibiotic-resistance
the contamination of enterococci might be because people in 90
those area are lack of knowledge of changing filter regularly 80
3.2. Detection and isolation of Enterococcus spp. 70
Out of the 145 samples tested, which were collected from 60
three different communes in Stoung district in Kampong 50

(%)
Thom province, 83 samples 57% were found to be 40
contaminated with enterococci in drinking water sample. In 30
these 83 samples 57%, 40% (n=33) were collected from Msa 20
Krang commune which is the water-land based communities, 10
36% (n=30) were from land base area in Trea commune, and 0
LVX AMX DOX CIP VA TC EM
24% (n=20) were collected from Peam Bang commune;
water based or floating communities. In this study, 257 R I S
strains of Enterococcus spp. were isolated from 83 samples
Fig. 3. Percentages of antibiotic-resistance Enterococcus
with seven different types of samples such as rain water,
spp. in Stoung District
bottle water (20L), filter, boiling water, water supply, well
water, and lake water as shown in Fig.2.
Enterococci, which are found in many aquatic
compartments, have an advantage in terms of persistence
55
50 rain water and multiplication because of their tolerance to various
Percentage of samples types

45 bottle water (20L) environmental factors, such as alkaline pH, increased

40 filter
35 boiling water
temperature, and sodium chloride concentration. Because
30 long-term survival is possible in the aquatic environment,
(%)

water supply
25
20
well water antibiotic resistant enterococci present a higher risk of
Lake water
15 spreading antibiotic resistance genes than other bacteria [8].
10
5
According to [6], more than 45% of the enterococci studied
0 were resistant to ciprofloxacin, erythromycin and
Trea Msa krang Peam Bang
Fig. 2. Percentage of samples types collected from St tetracycline, antibiotics are often used by humans and in
oung district. animals as therapeutic agents

3.3. Antibiotic resistance of Enterococcus spp. in 3.4 Multi-antibiotic resistance of Enterococcus spp.
drinking water In antibiotic susceptibility test, 51% of colonies is
The isolated Enterococcus spp. from drinking water susceptible and intermediate resistant to the antimicrobial
samples in Stoung district were conducted antibiotic agent. Whereas, 38% were resistance to the at least one
susceptibility test with seven antibiotics agent in five antimicrobial agent, while 11% (n=11) were multidrug-

229
resistant strains. Only two strains (2.0%) were resistance to
all type of antibiotic resistance LVX-AMP-DOX-CIP-VA- Acknowledgement
TC-EM in different five antimicrobial categories as
We are thankful to the Science and Technology Research
demonstrated in Fig 4.
Partnership for Sustainable Development (SATREPS), the
Japan Science and Technology Agency (JST)/Japan
11% International Cooperation Agency (JICA) for their financial
support.
38% 51%
References
[1] Emad Abada, Zarraq Al-Fifi, Abdul Jabbar Al-
Rajab, M.M. and M.S., 2019. Molecular identi fi
cation of biological contaminants in different
drinking water resources of the Jazan region , Saudi
Arabia. , 622–632.
[2] Abdelrahman, A.A., 2011. Bacteriological quality
S&I R MDR
of drinking water in Nyala , South Darfur , Sudan. ,
37–43.
Fig. 4. Percentage of antibiotic resistant of Enterococcus [3] WHO., 2019. Drinking-water. World Health
spp. in drinking water indicated as S&I: susceptible and Organization., 1.
intermediate resistant, R: resistant, MDR: multidrug- [4] CLSI., 2012. Methods for Dilution Antimicrobial
resistance. Susceptibility Tests for Bacteria That Grow
Aerobically.
[5] Water, department of., 2009. Surface water
Poor hygiene and infection control, lack of sanitation,
sampling methods and analysis — technical
weak or unenforced medicines regulation, and sub-standard appendices Standard operating procedures for
quality drugs have been suggested as some of the main water sampling-.
drivers of AMR in SE Asia [9]. In Cambodia, the behavior [6] Macedo, A.S., Freitas, A.R., Abreu, C., Machado,
of people using antibiotics in rural areas including trained E., Peixe, L., Sousa, J.C., Novais, C., 2011.
and untrained pharmacists, drugstores and nurses are International Journal of Food Microbiology
provided drugs to patients without go through any diagnose. Characterization of antibiotic resistant enterococci
isolated from untreated waters for human
Moreover, the villagers in the communities mostly
consumption in Portugal. International Journal of
purchasing antibiotic by self-prescription. Food Microbiology. 145, 315–319.
[7] Lata, P., Ram, S., Agrawal, M., Shanker, R., 2009.
IV. Conclusion Enterococci in river Ganga surface waters :
In conclusion, more than 50 percent of drinking water, Propensity of species distribution , dissemination
mostly filtered water, were contaminated with pathogenic of antimicrobial-resistance and virulence-markers
among species along landscape. 10, 1–10.
enterococci and could be a health risk for the communities
[8] Nishiyama, M., A.I.& Y.S., 2015. Journal of
living in those area. in order to reduce the multiple Environmental Science and Health , Part A : Toxic
antibiotic resistance, the authorities should educate people / Hazardous Substances and Environmental
to strengthen the hygienic practice including cleaning and Identification of Enterococcus faecium and
boiling water consumption, especially cleaning regularly the Enterococcus faecalis as vanC-type Vancomycin-
water containers with clean water to reduce recontamination Resistant Enterococci ( VRE ) from sewage and
and filter change. Moreover, proper use of antibiotics should river water in th. , 37–41.
[9] Reed, T.A.N., Krang, S., Miliya, T., et al., 2019.
be considered based on doctor’s prescription. Moreover,
International Journal of Infectious Diseases
wastewater from health care should be treated to ensure Antimicrobial resistance in Cambodia : a review.
absence of pathogenic bacteria before discharge to 85, 98–107.
environment since microorganisms will continue to adapt to
environment by developing resistance to new drugs and to
cause serious infection.

230
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Investigation on Antibiotic Resistance of Escherichia coli Isolated from Drinking Water

Collected in Stoung District

Boren BUN 1, Sovannmony NGET 1, 2, Soukim HENG 1,2, Sokneang IN 1, 2, Nishiyama MASETERU 3, Toru
WATANABE 3 and Hasika MITH 1, 2,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3 Faculty of Agriculture, Yamagata University, Japan
* Corresponding author: hasika@itc.edu.kh

Abstract

Escherichia coli has been known as key indicator of fecal contamination for drinking water. Even most of E. coli were not
pathogenic but some of them could be a threat in causing serious illness and outbreaks in many countries. Misuse of antibiotics
is currently becoming a major concern in relation to increase of antibiotic resistance of E. coli and prolongation of
hospitalization. Therefore, the current study was carried out to determine antibiotic resistance of E. coli isolated from drinking
water collected from three communes, Trea, Msa Krang and Peambang, in Stoung District, Kampong Thom province. The
drinking water samples were subjected for detection and isolation of E. coli, followed by antibiotic susceptibility test with
seven antibiotics. Among the 145 collected samples, there were 48 samples from Trea, 50 samples from Msa Krang and 47
samples from Peambang. The results showed that the presence of E. coli was detected in 54 out of 145 samples with 94
isolated strains. In terms of antibiotic susceptibility, E. coli were found to be resistant to the tested antibiotics as following:
67% to ampicillin, 52.31% to ticarcillin, 49.98% to tetracycline, 13.45% to ciprofloxacin, 10.46% to fosfomycin, 3.19% to
gentamicin, and 1.06% to ceftazidime. In addition, 44.68% were identified as multi-drug resistant (MDR) with MAR index
value > 0.2 and remarkably 2.12% were resistant to six antibiotics corresponding to six classes of antibiotics. These findings
could be an indicator showing a concern in terms of health risk for the communities living in those areas.

Keywords: antibiotic resistance, Escherichia coli, drinking water

I. Introduction more than five million people suffer from diseases

associated with contaminated drinking water, unclean
For infectious diseases, antibiotics are the main
domestic conditions, and inappropriate disposal of excreta
therapeutic strategy for bacterial infection [1]. In Cambodia,
[4]. More fundamentally, resistance to antibiotics is
with general frost, diarrhea and even general unwellness,
associated with overuse and abuse of antibiotics. As well as
antibiotics are found to be used [2]. Escherichia coli is an
the lack of production of new medicines from the
opportunistic pathogenic bacterium found in water [3], and
pharmaceutical industries [5]. It is impossible to quantify the

231
real value of bacterial resistance to antibiotics since bacterial 2.3. Antibiotics susceptibility test
antibiotic resistance is the natural process that happens after Susceptibility test was done by using the disc diffusion
the bacteria are exposed to it [6]. Consequently, there is method and following the categories of Clinical laboratory
uncertainty about post-antibiotics about the issue of standard institute (CLSI) for determining the susceptibility
antibiotics in the future and the broader effect on the health or resistance. Muller Hinton agar was used as the media for
system in emerging and developed countries, and even basic the test. Only the fresh culture can be used for susceptibility
infections would be unmanageable and potentially fatal [7]. test, using the LB agar to subculture the colonies. Prepare
the 0.5 McFarland concentration of colonies in 10 ml saline
II. Materials and Methods solution. Placing antibiotics discs with on the media surface,
make sure they are well touched the media. The position of
2.1. Sampling sites
disk on the inoculated plate, the center-center distance is 24
145 samples were taken from three communes in Stoung
mm and from edge to edge is 10 to 15 mm, then incubated
District, Kampong Thom Province, Cambodia (Figure 3.1).
in the optimum temperature of 37 ℃ for 16-18 h. In
Trea (land-based) 48 samples, Msa Krang (water-land-based)
interpretation of result, measured the diameter of zone
50 samples, and Peambang (water-based) 47 samples were
inhibition and report the result as Resistant (R), Intermediate
such communities. Sterilized bottles of 500 ml of drinking
(I) and Susceptible (S) by comparing the reference in CLSI,
water (rain water, purified water, bottled water, lake water,
as shown in Table 1.
well water and water supply) per household were dispensed
and then sent to the microbial laboratory of the Institute of
Table 1. Diameter of inhibition zone [8].
Technology of Cambodia.
Diameter of inhibition (mm)
Antibiotics
Susceptible Intermediate Resistance
(S) (I) (R)
AMP ≥17 14-16 ≤13
CAZ ≥21 18-20 ≤17
CIP ≥21 16-20 ≤15
GEN ≥15 13-14 ≤12
FOS ≥16 13-15 ≤12
TE ≥15 12-14 ≤11
Fig. 1. Map of sampling locations TIC ≥20 15-19 ≤14
2.2. Detection and isolation of Escherichia coli
Using the membrane filtration for detection of III. Results and Discussion
Escherichia coli from the samples in 24 h after collection.
The Chromocult coliform agar was used as the selective 3.1. Detection of Escherichia coli
media, and incubation was at 37 ℃ for 18-24 h. The dark- Among 145 samples in three communes, there are 54
blue or violet formed colonies were considered as samples were found positive of Escherichia coli, with 17
Escherichia coli and would then be picked by loop to samples in Trea commune, 20 samples in Msa Krang
another Chromocult coliform agar plate for isolation. This commune and 17 samples from Peambang. The total of
process would be continued until all the colonies form on the isolated strains from these three locations are 94 samples
plates are purple or only dark-blue colonies appeared. Then which then be tested for antibiotics susceptibility.
the colonies were ready for stock culture storage. Using the
LB broth for the media to grow the isolated strains, prepared
in the Eppendorf, picking the colonies strains to the
Eppendorf and incubated at 37 ℃ for 24 h. After incubation,
adding the glycerol to make the glycerol concentration is 15-
20%, vortex then keeping in the -80 ℃ freezer.

232
 and TIC with 28.13%. While, only 12.5% are resistant to CIP
Number of each parameter 60 and Following by GEN and CAZ with 6.25% of total strains.
50
50
48 47 In Cambodia, the level of understanding about antibiotics is
40
40 still low. They lack of education about drug consumption
32
30 and the potential impacts [2]. Using drug freely, discharging
22
20 17
20
17 to the environment without treatment, bacteria can evolve
their gene to be resistant to antibiotics [11]. While, waste
10
water treatment plant is currently not specifically tasked
0
Trea Msa Krang Peam Bang with the removal of antibiotics resistance gene yet [13]. With
Results by commune presence of antibiotics in the environment, it may create
Total samples Detected Samples Detected Strains selective pressure which antibiotic resistance, in
consequence [14].
Fig.2. Results of detected samples.
100%
3.2. Antibiotics resistance of Escherichia coli

Percentage of antibiotic resistant strains

80%
3.2.1. Antibiotics susceptibility test in Trea
With 22 strains for conducting on study of antibiotic 60%
susceptibility of E. coli, 59.09% of isolated strains were 40%
resistant to TE, 54.55% were resistant to AMP, 50% to TIC, 20%
9.09% to CIP and 4.55% to FOS. Predictably, AMP, TE [7] 0%
AMP GEN CIP CAZ FOS TE TIC
and TIC [9] resistance is common in E. coli. AMP and TE is
%R 37.50 6.25% 12.50 6.25% 0.00% 34.38 28.13
used as common treatment among Cambodians [2]. There
%I 15.63 31.25 3.13% 3.13% 0.00% 0.00% 0.00%
are up to over 200 βlactamase genes that responds for %S 46.88 62.50 84.38 90.63 100.00 65.63 71.88
antibiotics resistance to that used to be detected [10], and Antibiotics
also TE, it is used for growth promoter in animal husbandry
%R %I %S
[11], animal feed contains antibiotics, is to stimulate growth
becomes a global practice [12], antibiotic is widely used in
agriculture [10]. As it is available freely in the region [2], Fig.4. Percentage of R, I, S in each antimicrobial agent in
this might be the cause reason that make bacteria becomes Msa Krang commune.
resistant to drugs. 3.2.3. Antibiotics susceptibility test in Peambang
100% AMP and TIC in the study of E. coli strains from
Percentage of antibiotic resistant strains

80% Peambang commune, are the noteworthy ones which 80%

and 70% among 40 strains are resistant to them, respectively.
60%
Moreover, TE is still in high level, with 55% of all strains
40%
are resistant to this drug. While, as shown in Figure 4.3,
20% following by with 15%, CIP with 12.5% and GEN with just
0% 5%. Related factors associated with location as aquatic
AMP GEN CIP CAZ FOS TE TIC
%R 54.55 0.00% 9.09% 0.00% 4.55% 59.09 50.00 environment, usage of antimicrobials in humans or
%I 4.55% 13.64 4.55% 4.55% 4.55% 0.00% 0.00% terrestrial food animal, sewage contamination or run-off
%S 40.91 86.36 86.36 95.45 90.91 40.91 50.00 from agricultural areas with grazing animals to aquaculture
Antibiotics operations [15] are the great cause for antibiotics resistant
%R %I %S among microbiota. There is also a suggestion that fish
consumption that contain resistant bacteria can contribute to
resistant bacteria [16]. Also, in Peambang, people tend to
Fig. 3. Percentage of R, I, S in each antimicrobial agent in
have fish farming under or nearby their floating houses, the
Trea commune.
drinking water source and the fish farm is almost the same
3.2.2. Antibiotics susceptibility test in Msa Krang
one. Some of them use the feed from the market. This may
There are 32 strains of E. coli were conducted with
create another factor to bacterial resistance. The animal feed
antibiotics susceptibility study. The resistance result of E.
contains antibiotics as growth promoter and disease
coli strains contributed to AMP with 37.5%, TE with 34.38%
prevention [14], and will then push antibiotics resistance

233
among the microbiota as antibiotics exposure to bacteria [2], [5] Mobarki, N., Almerabi, B., Hattan, A., 2019. Antibiotic
and antimicrobial resistance is accelerated when Resistance Crisis. International Journal of Medicine in
overexposure of bacteria to antibiotics [17]. Developing Countries. 40, 561–564.
[6] Prestinaci, F., Pezzotti, P., Pantosti, A., 2015.
100% Antimicrobial resistance: A global multifaceted
Percentage of antibiotic resistant strains

phenomenon. Pathogens and Global Health. 109, 309–

80%
318.
60% [7] Li, X., Watanabe, N., Xiao, C., Harter, T., McCowan, B.,
Liu, Y., Atwill, E.R., 2014. Antibiotic-resistant E. Coli
40% in surface water and groundwater in dairy operations in
20% Northern California. Environmental Monitoring and
Assessment. 186, 1253–1260.
0% [8] Dolinsky, A.L., 2017. M100 Performance Standards for
AMP GEN CIP CAZ FOS TE TIC
Antimicrobial Susceptibility Testing.
%R 80.00% 5.00% 12.50% 0.00% 15.00% 55.00% 70.00%
[9] Li, X., Watanabe, N., Xiao, C., Harter, T., McCowan, B.,
%I 5.00% 15.00% 10.00% 2.50% 0.00% 5.00% 7.50%
Liu, Y., Atwill, E.R., 2014. Antibiotic-resistant E. Coli
%S 15.00% 80.00% 77.50% 97.50% 85.00% 40.00% 22.50%
in surface water and groundwater in dairy operations in
Antibiotics Northern California. Environmental Monitoring and
Assessment. 186, 1253–1260.
%R %I %S
[10] Alm, E.W., Zimbler, D., Callahan, E., Plomaritis, E.,
2014. Patterns and persistence of antibiotic resistance in
Fig. 5. Percentage of R, I, S in each antimicrobial agent in faecal indicator bacteria from freshwater recreational
Peambang commune. beaches. Journal of Applied Microbiology. 117, 273–
285.
IV. Conclusion [11] Karami, N., Nowrouzian, F., Adlerberth, I., Wold,
A.E., 2006. Tetracycline resistance in escherichia coli
In conclusion, among such sampling area, Trea, Msa and persistence in the infantile colonic microbiota.
Krang and Peambang, there is a high contamination of Antimicrobial Agents and Chemotherapy. 50, 156–161.
drinking water with E. coli with 94 strains of E. coli isolated [12] Chattopadhyay, M.K., 2014. Use of antibiotics as
from 54 out of 145 collected drinking water samples, E. coli feed additives: A burning question. Frontiers in
is mostly contaminated in the rain water source as the Microbiology. 5, 1–3.
container was exposed to the surrounded environment, lack [13] Kraemer, S.A., Ramachandran, A., Perron, G.G.,
2019. Antibiotic pollution in the environment: From
of hygiene and there is no proper treatment. The antibiotic
microbial ecology to public policy. Microorganisms. 7,
susceptibility study revealed that the isolated E. coli were 1–24.
found to be more resistant to AMP, TIC and TET as in order. [14] Manyi-Loh, C., Mamphweli, S., Meyer, E., Okoh, A.,
There were 44.68% of total strains found to be MAR. 2018. Antibiotic use in agriculture and its consequential
Particularly, some strains (2.12%) could be resistant to six resistance in environmental sources: Potential public
groups of antibiotics, such as AMP-GEN-CAZ-CIP-TIC-TE. health implications.
[15] Alday, V., Guichard, B., Smith, P., Uhland, C., 2006.
References Joint FAO / WHO / OIE Expert Consultation on
[1] Rayner, C., Munckhof, W.J., 2005. Antibiotics currently Antimicrobial Use in Aquaculture and Antimicrobial
used in the treatment of infections caused by Resistance Towards a risk analysis of antimicrobial use
Staphylococcus aureus. Internal medicine journal. 35, in aquaculture. , 29 pp.
1–16. [16] Petersen, A., Andersen, J.S., Kaewmak, T., Somsiri,
[2] Om, Chhorvoin, Daily, F., Vlieghe, E., McLaughlin, T., Dalsgaard, A., 2002. Impact of integrated fish
J.C., McLaws, M.L., 2017. Pervasive antibiotic misuse farming on antimicrobial resistance in a pond
in the Cambodian community: Antibiotic-seeking environment. Applied and Environmental Microbiology.
behaviour with unrestricted access. Antimicrobial 68, 6036–6042.
Resistance and Infection Control. 6, 1–8. [17] Smith, G.L., 2018. Infectious Diseases and
[3] Gorchev, H.G., Ozolins, G., 2004. Guidelines for Nanomedicine III. Springer Singapore.
Drinking-water Quality, 3rd Edition. Who. 1, 564.
[4] Peter., 2002. Dirty Water: Estimated Deaths from
Water-Related Diseases 2000-2020 Pacific. Dirty Water:
Estimated Deaths from Water-Related Diseases 2000-
2020 Pacific., 1–12.

234
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Antibiotic Resistance along Kopsrov Lake Impacted By Domestic Wastewater

Sopheavattey MONIROTH 1, Sornpisey KHUT 2, Rithineth TORTH 2, Monychot Tepy CHANTO2,3 and
Chanthol PENG 2,4,*
1
Faculty of Hydrology and Water Resource Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
4
Water and Environment Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
*peng@itc.edu.kh

Abstract
The emerging of Antibiotic Resistant Bacteria (ARB) in wastewater and freshwater systems is a severe health problem
and water quality concern. Thus, the monitoring and assessment of ARB in the water system are essential for public health
protection. This study aims to assess the water quality in term of physicochemical, total bacteria and ARB concentration at
Kopsrov lake (KL) that is impacted by domestic wastewater. Three different sites at KL, including upstream, middle, and
downstream, were chosen. The physicochemical properties such as pH, temperature, turbidity, dissolved oxygen (DO),
conductivity, and oxidation-reduction potential (ORP) were analyzed. Moreover, culture-dependent methods were used to
determine the concentration of total bacteria and ARB (Escherichia coli and Coliform) that are resistant to ampicillin (AMP),
kanamycin (KAN) and ciprofloxacin (CIP). The result showed that the physicochemical properties (pH and DO) were
followed the national effluent standard for pollution sources while Coliform concentration were higher than the standard. The
concentration of total bacteria at the upstream site of KL showed the highest concentration (4.04E+05 CFU/ml) which
contained coliform (7.90E+04 CFU/ml) that was higher than national standard (<1.0E+01 MPN/ml). While at the downstream,
the concentration of total bacteria was reduced to 3.23E+04 CFU/ml where coliform concentration (2.60E+03 CFU/ml), but
remained higher than standard. Notably, the middle lake water showed the lowest concentration (3.91E+03 CFU/ml) while
Coliform (1.33E+03 CFU/ml) was still higher than compared standard. Similarly, the antibiotic-resistant E. coli and Coliform
at upstream showed the highest, followed by the downstream site. The antibiotic resistant E. coli was not detected in the
middle lake water sample. E. coli and Coliform had the highest resistant to AMP, followed by CIP and minor resist to KAN
at each sampling site. This finding implies that, (KL) is constituted of some significant ARB which are considered as a
potential threat to humans and/or animal health since they may lead to more cases of difficult-to-treat infections. Moreover,
part of the ARB released from wastewater will not only be able to cause disease in humans or animals, the risk of worsening
the environmental problem which results in contributing to the emergence of resistant pathogenic bacteria.
Keywords: Antibiotic Resistance Bacteria, E. coli, Coliform, Wastewater

I. Introduction and photogenic microorganisms from sewage and

wastewater, technological combinations of physicochemical
Wastewater Treatment Plants (WWTPs) have a crucial
and biological treatments are needed to achieve for allowing
role in the protection of the environment, in particular – the
the return of good quality water to the environment [1].
natural water bodies. In order to remove chemical pollutants
Antibiotics, either are cytotoxic or cytostatic to the

235
microorganisms, allow the body’s natural defenses, the (Tem), turbidity, dissolved oxygen (DO), conductivity
immune system, to eliminate them by inhibiting the (Cond), and oxidation-reduction potential (ORP) were
synthesis of a bacterial cell, synthesis of proteins, analyzed by using YSI EXO2 multi-parameter instrument.
deoxyribonucleic acid (DNA), ribonucleic acid (RNA) [2]. 2.3. Quantification of Total Bacteria and ARB
In fact, antibiotics are widely used in many applications such Culture-dependent methods were used to determine of total
as for the treatment of infectious diseases to protection of bacteria and ARB (Escherichia coli and Coliform) that are
human health and added into feed to promote animal growth resistant to ampicillin (AMP), kanamycin (KAN) and
which is partially leaded to emerge of antibiotic-resistant ciprofloxacin (CIP).
bacteria (ARB) [3]. Antibiotics has become an emerging Lysogeny Broth agar was used to culture Total Bacteria
micropollutant and their presence in the water lead bacteria (TPC) which colonies appeared in white colour with Pour
to develop resistant gene which is often transferred from Plate Method. For Escherichia coli (E. coli) and Coliform
bacterial cells to other cells, occasionally from commensal were done by Pour Plate technique with Chromocult®
bacteria to pathogenic ones, promote ARB and antibiotic Coliform Agar by Merck Millipore. Bacteria was
resistance genes (ARGs) in the environment. Thus, distinguished by colonies color which can be known as dark
environment can promote the development and blue or violet is E. coli while pink represented Coliform.
dissemination of ARGs, which have been identified as a Three antibiotics were added into culture media. Stock
global public health crisis [4]. solution was prepared by dissolving antibiotics powder into
Currently, wastewater released from households contains sterilized pure water and diluting it into culture media to
234 tons of feces, 2,335 m3 of urine and 8,154 m3 of grey achieve the concentration of 50 µg/mL for AMP and KAN,
water per day. There are 14 pumping stations in Phnom and 5 µg/mL for CIP. Antibiotic-resistant bacteria
Penh’s 732-kilometer drainage system and 44,807 enumerated by using spread plate technique. All sampled
converted-holds in Phnom Penh for wastewater plates were done triplicate and control plate contained no
management. Kopsrov water treatment facility on the lake antibiotic. The concentration of bacteria was calculated by
became the 14th pumping station, with the construction of a using the following equation
water treatment facility that covers about 500-hectares, the ∑𝐶
lake is now being used to discharge sewage and flood water 𝑁= (Eq.1)
𝑛×𝑑×𝑣
from Phnom Penh [5]. Since Kopsrov Lake (KL) has Where 𝑁 is the concentration of bacteria (CFU/ml), Σ𝐶 is
received many sources of wastewater which may contained the sum of counted colonies, 𝑛 is the amount of plates, 𝑑 is
high concentration of pollutant and various bacteria. Thus, the dilution factor, and 𝑉 is the inoculated volume.
the monitoring and assessment of bacteria, especially ARB
are essential for public health protection. This study III. Results and Discussion
investigates the current status of water quality of KL that is
impacted by the flow of untreated wastewater. The physicochemical characteristics at each site, are
shown in Table 1. It can be seen that Tem, DO, turbidity,
II. Materials and Methods Cond and ORP of UP site were 22.72oC, 7.36, 2.92 mg/l,
102.00 NTU, 312.33 µs/cm, and 209.50 mV respectively. At
2.1. Sampling sites MS site, the Tem, pH, DO, turbidity, Cond and ORP were
Samples were collected on 07th October, 2020, during 22.13 oC, 7.89, 5.93 mg/l, 18.27 NTU, 278.93 µs/cm, and
rainy season from three different sites as surface water and 357.17 mV while at DS were 20.98oC, 7.74, 6.61 mg/l, 22.03
wastewater discharging point in KL including upstream NTU, 270.73 µs/cm, and 423.53 mV respectively. Moreover,
(US), middle stream (MS), and downstream (DS). 1000 mL pH and DO value at each site were ranged within national
of each sample was collected and stored in sterilized standard (pH: 6.5-8.5) and DO (7.5-2.0), according to sub-
polyethylene bottle, and placed in ice box while transporting degree on water pollution control [6].
to ITC laboratory for analysis.

2.2. Physicochemical analysis

Physicochemical properties such as pH, temperature

236
Table 1. Physicochemical characteristics of lake water Concentration of Total Plate Count (TPC), Antibiotic
Resistance of Coliform and antibiotic resistance of E. coli
Parameter UP MS DS SD which expressed in CFU/ml were shown in Fig. 1, Fig. 2 and
Tem (oC) 22.72 22.13 20.98 - Fig. 3. respectively. In Fig. 1, concentration TPC at UP
pH 7.36 7.89 7.74 6.5-8.5 (4.04E+05 CFU/ml) was higher than those at DS (3.23E+04
DO (mg/l) 2.92 5.93 6.61 7.5-2.0 CFU/ml) and MS (3.91E+03 CFU/ml). Out of TPC in US,
Tur (NTU) 102.00 18.27 22.03 - MS and DS contained Coliform (7.90E+04, 1.33E+03 and
Cond (µs/cm) 312.33 278.93 270.73 - 2.60E+03 CFU/ml) which equaled to 19.55%, 34.02% and
ORP (mV) 209.50 357.17 423.53 -
8.05% while E. coli (3.82E+04, 0.00E+00 and 4.30E+02
1.00E+06 CFU/ml) that equaled 9.46%, 0.00% and 5.72% respectively.
Coliform concentration at all three sites were higher than
Concentration (CFU/ml)

1.00E+05
given standard on water pollution control (<1.0E+01
1.00E+04 MPN/ml) which meant that those sample were not complied
1.00E+03 with national standard [6]. Coliform, primary indicator of
fecal contamination that found at three sites could be
1.00E+02
brought into aquatic environment through treated or
1.00E+01 untreated wastewater release. In Fig. 2 suggested that,
1.00E+00 the highest antimicrobial resistance of Coliform that
US MS DS compared to control, at DS tolerant to AMP (70.77%),
Sampling locations
followed by CIP (29.12%) and KAN (13.73%). At UP,
antimicrobial resistance of Coliform for AMP (13.92%),
Fig. 1. Concentration of Total Plate Count followed by KAN (1.72%) and CIP (1.55%) while at MS for
AMP (69.92%), followed by CIP (0.25%) and KAN (0.00%).
1.00E+05 In Fig. 3, antimicrobial resistance of comparing to control,
Concentration (CFU/ml)

1.00E+04 the highest resistance of E. coli is at DS for AMP (71.40%),

1.00E+03 followed by CIP (57.44%) and KAN (13.19%). At UP,
antimicrobial resistance of E. coli for CIP (4.08%), followed
1.00E+02
by AMP (1.48%) and KAN (0.50%) while at MS E. coli was
1.00E+01
not detected.
1.00E+00 The frequency of antibiotic resistance among Coliform
Control AMP KAN CIP
and E. coli bacteria varied from sample to sample, and this
Antiibiotics
is probably dependent upon the sources of contamination of
US MS DS the water. The highest concentration of TPC is located at UP
which is correlate to Coliform and E. coli. For UP samples
Fig. 2. Antibiotic Resistance in Coliform were taken at the surface of discharging flow from various
sources to KL where water flow from sewage system that
1.00E+05 may contain high pollution. Additionally, DO value at UP is
Concentration (CFU/ml)

1.00E+04 quite low as there are many bacteria or aquatic animals in

1.00E+03 that area. They may be overpopulated which resulting
1.00E+02
consuming great amount of DO At the same time, that site
presented the lowest value of ORP (209.50 mV) which
1.00E+01
indicated the high amount of presented reducing agent and
1.00E+00 pollution than the other two sites as water pollution levels
Control AMP KAN CIP
tended to increase with low ORP readings [7].
Antibiotic
Concentration of Coliform in control, plate without
US MS (Not Detected) DS antibiotic at each site are higher than antibiotic plates which
is respected to the theory of antibiotic’s action whether
Fig. 3. Antibiotic Resistance in E. coli

237
killing or decreasing growth of bacteria. Even though inhibitory concentration (MIC) which is varied by
Coliform was not fully removed, its concentration at DS is depending on antibiotic group [10]. Even if low
still highly resistant to all three drugs as shown in Fig. 2 concentrations of CIP, a group of fluoroquinolones could
comparing to control. The percentage of antibiotic resistance promote resistance for both detected E. coli and Coliform.
at DS, MS, or UP to AMP is higher than other two drugs. This might be from overuse of one of the fluoroquinolone
For MS site is clearly indicated that Coliform is highly leads to the development of resistance to the whole group of
tolerant to AMP (69.92%) while resistance level is less for quinolone antibiotics. Fluoroquinolones which are mostly
CIP (0.25%) and KAN (0.00%), so Coliform growth were added to the animal feed and used to treat humans caused
inhibited by KAN and CIP at MS site. Coliform is an selected E. coli strains resistant to these drugs [9]. However,
indicator microorganism which subdivided into total and E. coli also known as less resistance to KAN which located
fecal coliforms (E. coli, Klebsiella, and Enterobacter) while in aminoglycosides group that used to treat serious bacterial
total coliform includes both soil intermediates and fecal infections. These group are poorly absorbed by orally, so
forms, fecal coliforms confines to those from fecal origin, they are usually injected into a vein [10].
used as standard microbial indicators of water quality. IV. Conclusion
Because some pathogenic microorganism also known as
Results of our research have shown that TPC is a
Coliform group, so its highly resistant to specific drugs
contained Coliform and E. coli. Also frequency of TPC at
appeared as a serious concern of multi-drug resistant of
UP was the highest. E. coli showed major resistance to AMP
pathogenic one [9]. E. coli at MS site was not detected
and CIP, and minor resistance to KAN. Emerging of these
while the highest concentration is at UP which can be
resistance while a growing of common infection such as
expressed in higher contaminated from human activities.
waterborne and foodborne disease are becoming harder, and
This result paralleled to Coliform which the most
sometime impossible, to treat as antibiotic become less
contaminated site known as UP. Since MS sample is absent
effective.
of E. coli, so this site could not use to express antibiotic
resistance while the other two have shown. Each studied site
shown the percentage of resistance where at DS known as Acknowledgement
highly resistant one. Detected E. coli (5.72%) from total This research was supported by the Science and
bacteria indicated the highest tolerance to AMP (71.40%), Technology Research Partnership for Sustainable
followed by CIP (57.44%) and KAN (13.19%). This result Development (SATREPS), the Japan Science and
indicated that E. coli resisted to AMP. E. coli were not killed Technology Agency (JST)/Japan International Cooperation
and it continued to grow even the presence of AMP. E. coli Agency (JICA) with the grant-number JPMJSA1503, and
too had developed ability and defeated the drug which were AFD/EU.
usually applied to kill them [8]. Moreover, from resistance References
percentage, it is indicated that E. coli was not resistant to
[1] Manaia, C.M., Rocha, J., Scaccia, N., Marano, R.,
KAN as it showed small resistant percentage, so KAN is the Radu, E., Biancullo, F., Cerqueira, F., Fortunato, G.,
most effective one among studied drugs. E. coli is the most
Iakovides, I.C., Zammit, I., Kampouris, I., Vaz-
common cause of uncomplicated and community acquired
Moreira, I., Nunes, O.C., 2018. Antibiotic resistance
urinary tract infection [8]. According to Fig.2. and Fig.3. in wastewater treatment plants: Tackling the black
both E. coli and Coliform are highly resistant to AMP and
box. Environ. Int. 115, 312–324
minor resistant to CIP and KAN. AMP which is considered [2] Zaman, S.B., Hussain, M.A., Nye, R., Mehta, V.,
as a resistance drug to either Coliform or E. coli could not
Mamun, K.T., Hossain, N., n.d. A Review on
use to treat infected human. The reason for this is common
Antibiotic Resistance: Alarm Bells are Ringing.
practice of prescribing ampicillin in the tract infections, and Cureus 9.
the fact that ampicillin was the most commonly prescribed
[3] Qadir, M., Wichelns, D., Raschid-Sally, L.,
antibiotics in developing countries. World Health McCornick, P.G., Drechsel, P., Bahri, A., Minhas,
Organization donated large doses of these antibiotics in the
P.S., 2010. The challenges of wastewater irrigation
treatment of many infectious diseases [9]. Concentration of
in developing countries. Agric. Water Manag. 97,
KAN (50µg/L) and CIP (5µg/L) indicated their minimal 561–568.

238
[4] Liu, X., Guo, X., Liu, Y., Lu, S., Xi, B., Zhang, J.,
Wang, Z., Bi, B., 2019. A review on removing
antibiotics and antibiotic resistance genes from
wastewater by constructed wetlands: Performance
and microbial response. Environ. Pollut. 254,
112996.
[5] Pumping station pollution plagues community,
2018. . Khmer Times. URL
https://www.khmertimeskh.com/528735/pumping-
station-pollution-plagues-community/ (accessed
12.15.20).
[6] Royal Government Concil of Minister, n.d. Sub-
Degree-27-on-Water-Pollution-Control_990406.pdf. [7]
Marín Galvín, R., Rodríguez Mellado, J.M., Ruiz
Montoya, M., Jiménez Gamero, C., 2001. OXIDATION-
REDUCTION POTENTIAL (ORP) IN PREPARED
AND INDUSTRIALLY TREATED WATERS. Bol. Soc.
Chil. Quím. 46, 387–397.
[8] Cooke, M.D., 1976. Antibiotic resistance in coliform
and faecal coliform bacteria from natural waters and
effluents. N. Z. J. Mar. Freshw. Res. 10, 391–397.
[9] Vranic, S.M., Uzunovic, A., 2016.
ANTIMICROBIAL RESISTANCE OF
ESCHERICHIA COLI STRAINS ISOLATED
FROM URINE AT OUTPATIENT POPULATION:
A SINGLE LABORATORY EXPERIENCE. Mater.
Socio-Medica 28, 121–124.
[10] Brain J., W., n.d. Aminoglycosides - Infections.MSD
Man. Consum. Version. URL
https://www.msdmanuals.com/home/infections/anti
biotics/aminoglycosides (accessed 12.8.20).

239
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Study on Antibiotic Resistance of Pseudomonas aeruginosa Isolated from Drinking Water

Collected from Three communes in Kampong Thom Province

Seab SAT 1, Sovannmony NGET 1, 2, Soukim HENG 1,2, Sokneang IN 1, 2, Masateru NISHIYAMA 3, Toru
WATANABE 3 and Hasika MITH 1, 2,*

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
Faculty of Agriculture, Yamagata University, Japan
1-23 Wakaba-machi, Tsu Yamagata 997-8555, Japan
* hasika@itc.edu.kh

Abstract

The opportunistic pathogenic Pseudomonas aeruginosa has been categorized as the critical prioritized pathogen for
research and development of new antibiotics according to the World Health Organization. However, the awareness of P.
aeruginosa’s antibiotic resistance is still not widely and specifically studied in developing countries. The current study
was conducted to determine antibiotic resistance of P. aeruginosa isolated from drinking water collected in Kampong
Thom province. The enumeration of P. aeruginosa was carried out using filtration method, followed by isolation and
antibiotic susceptibility test. The results showed that out of 145 samples, 35 drinking water samples were detected with
presence of P. aeruginosa. Furthermore, a total number of 123 strains were isolated prior to antibiotic susceptibility
test. As a result, only nine isolated strains were found to be resistant to ticarcillin among the tested antibiotics. Even
though no single strain showed its resistance to meropenem and fosfomycin, but 2 and 17 strains of P. aeruginosa were
observed to be intermediate to the two antibiotics, respectively. However, colistin, gentamicin, and ciprofloxacin
showed a good effectiveness against these isolated P. aeruginosa. Based on this study, the prevalence of antibiotic-
resistance of P. aeruginosa is not high in the drinking water collected from the three communes in Stoung District.

Keywords: Antibiotic resistance, drinking water, Pseudomonas aeruginosa

I. Introduction population in developing country is still suffering from

waterborne diseases [3]. As a result, the study of antibiotic
The antibiotic resistance is considered a potential global
resistance in developing country on P. aeruginosa was
threat for food safety and development, since it sharply rose
conducted in order to assess this problem in drinking water.
the mortality of population [1]. According to the World
To deeply investigate this issue, the study area was located
Health Organization, P. aeruginosa is classified as
in Kampong Thom with three different sites including land-
“Critical” priority pathogen for research and development of
based, water- and land-based, and water-based.
new antibiotics. Moreover, the potable water is defined as
consumable water and safety for lifetime [2], meanwhile the
II. Materials and Methods

240
2.1. Sampling sites gentamycin (10 g/disk), meropenem (10 g/disk),
A total number of 145 drinking water samples were ciprofloxacin (5 g/disk), colistin (10 g/disk), Fosfomycin
col-lected, located in Kampong Thom province along To (200 g/disk) and ticarcillin (75 g/disk). The antibiotic
nle Sap Lake. The sampling was carried out in three co containing disks were placed on the Mueller Hinton agar
mmunes such as Trea, Msa Krang and Peam Bang with (MH, Himedia, Mumbai, India). The diameters of inhibitory
three villages from each commune, and the map of sam zone (mm) were measured and compared to the reference to
pling sites was shown in Fig. 1 The samples were imm determine whether the isolated strains were susceptible,
ediately sent to the laboratory to analyze within 24 h [4 intermediate or resistant, as shown in the Table 1.
]. Table 1. CLSI diameter of inhibition zone

Diameter of inhibition zone (mm)

Antibiotics
Intermediate Resistance
Susceptible (S)
(I) (R)

Ticarcillin ≥24 16-23 ≤15

Meropenem ≥19 16-18 ≤15

Colistin ≥11 - ≤10

Gentamicin ≥15 13-14 ≤12

Fosfomycin ≥16 13-15 ≤12

Fig. 1. Sampling sites areas in Kampong Thom. Ciprofloxacin ≥21 16-20 ≤15
2.2. Detection of P. aeruginosa by filtration method
The collected drinking water samples were enumerated III. Results and Discussion
for P. aeruginosa using filtration method. A volume of 100 3.1. Detection and isolation of P. aeruginosa
ml of samples was filtrated using 0.45 µm mixed cellulose In overall 145 samples were taken from nine villages in
ester filters (Merck Millipore, Germany). The filters were three different communes in Kampong Thom province to
placed onto the prepared Pseudomonas CFC/CN agar detect the presence of P. aeruginosa. As the result, the
(Merck KGaA, Darmstadt, Germany) with CN supplement presence of P. aeruginosa was found in 35 samples among
and incubated at 37 ˚C for 24 h, the blue/green colonies were those 14, 11, 10 samples were from Trea, Msa Krang, and
detected and considered as P. aeruginosa. Peam Bang respectively. They were originated from various
2.3. Isolation of P. aeruginosa from water samples resources including rain water, bottle water (20L), lake
water, filter, and boiling in 23%, 11%, 14%, 43%, 9%
The grown colonies were selected continuously for
respectively.
isolation to ensure the purity of each strain out of diverse
bacterial population. The colonies were looped and streaked 60.00
detected of P. aeruginosa (%)

50.00 50.00
Percentage of sample types

on another agar plate (Pseudomonas agar base CN 50.00

40.00
36.36
supplement) for few times in order to obtain the same 40.00
28.57 27.27
30.00
morphology and color indicator of colonies. The collected 18.18
20.00 14.29
strains were then stored in Luria-Bertani broth (LB, Difco, 7.14 9.09 9.09 10.00
10.00
Franklin Lakes, United State) with 15% glycerol in 0.00 0.00 0.00
0.00
cryogenic tubes. Trea Msa Krang Peam Bang

2.4. Antibiotic susceptibility test Commune

In the current study,), the disk diffusion method was used Rain water Bottle water (20L) Lake water Filter Boiling
to determine the susceptibility of the isolated P. aeruginosa
following the Clinical and Laboratory Standard Institute Fig. 2. Different sample types detected of P. aeruginosa
(CLSI). Six antibiotics from different categories were from each commune in Kampong Thom province.

241
 As shown in Fig. 2, the presence of P. aeruginosa was that have to be tested with disk diffusion method and zone
found in filtered, boiling drinking water which could be an measurement to determine its resistance.
explanation of water being stored inappropriately that could 93.55 100 100 100 100 100
100
cause contamination from the container or lack of 90
effectiveness of the filtration method. 80

Percentage (%)
70
3.2. Antibiotic susceptibility of isolated P. 60
50
aeruginosa 40
30
3.2.1. Trea Commune 20
6.45
The sampling sites in Trea commune are located in t 10 0 0 0 0 0 0 0 0 0 0 0
0
hree different villages such as SA, TT, and HB that are R I S R I S R I S R I S R I S R I S
full of villagers. There are 48 isolated strains in Trea co TIC MEM CL GEN FOS CIP
mmune that have to be applied with disc diffusion meth
Antibiotics
od, and in order to determine its resistance, zone measur
ements were compared to the reference in CLSI (2017) Fig. 4. Antibiotic resistance of P. aeruginosa in Msa Krang
[5]. Commune.

100 91.67 100 100 100 100 As shown in Fig. 4, the antibiotic resistance of P.
90 77.08 aeruginosa is quite similar to the previous result that two P.
80
Percentage (%)

70 aeruginosa strains were resistant and 29 strains were

60 susceptible to TIC, whereas MEM, CL, GEN, FOS, and CIP
50
40 were still highly effective. In this scenario, TIC was
30 22.92
20 8.33
considered a concern for clinical treatment because in
10 0 0 0 0 0 0 0 0 0 0 another investigate with 40 isolates of P. aeruginosa were
0
R I S R I S R I S R I S R I S R I S resulted in 85% of TIC resistance [8].
TIC MEM CL GEN FOS CIP 3.2.3. Peam Bang Commune
Three villages in Peam Bang commune were chosen as
Antibiotics the sampling sites including PV, DS, and PB that are
Fig. 3. Antibiotic resistance of P. aeruginosa in Trea alongside the Tonle Sap Lake. Total 44 isolated strains were
commune. “R” means “Resistant”, “I” means also tested with disk diffusion method.
“Intermediate”, and “S” means “Susceptible”.
As shown in Fig. 3, four strains of P. aeruginosa were
resistant and 44 strains were susceptible to TIC, meanwhile
11 strains were also intermediate to FOS. However, MEM,
CL, GEN, and CIP were highly effective against this
bacterium. Another study of [6] also showed similar result
as above one, antibiotic resistance was often resistant to TIC,
FOS whereas there were no resistant strains to CIP, GEN,
and MEM. According to this observation, the occurrence of
antibiotic resistance could be an evidence of environmental
antibiotic resistance genes because of extensive use of Fig. 5. Antibiotic resistance of P. aeruginosa in Peam B
antibiotic in agricultural purpose and therapy [7] that lead to ang commune.
selective pressure in environmental bacteria and was found As shown in Fig. 5, TIC is still a less effective drug
in a populated commune. against P. aeruginosa with 41 intermediate-strains, three
3.2.2. Msa Krang Commune resistant-strains. Moreover, MEM notably consists two
Three villages of sampling sites in Msa Krang commune intermediate-strains which is an interesting result compare
were done in PK, BT, and KD. There are 31 isolated strains to previous two communes, and FOS also contains six
intermediate-strains. Meanwhile CL, GEN and CIP continue

242
to play a strong role to inhibit this pathogen. This result Aztreonam and Ticarcillin in Pseudomonas aeruginosa
could be evidence of a possible presence of antibiotic isolated from soil of different crops in Brazil. Science
resistance genes (ARGs) contamination from Tonle Sap of the Total Environment. 473–474, 155–158.
Lake along with Peam Bang commune, and the occurrence [9] Ash, R.J., Mauck, B., Morgan, M., 2002. Antibiotic
and transfering of ARGs could be caused by pollution from resistance of gram-negative bacteria in rivers, United
human activities [9]. States. Emerging Infectious Diseases. 8, 713–716.
IV. Conclusion
In conclusion, a high concentration of P.
aeruginosa was found in drinking water that originated
from home-collected boiling and rain water. The
prevalence of antibiotic-resistant P. aeruginosa is not
high in drinking water for those sampling sites while
the presence of multi-drug resistance is not detected for
any isolated strain.
Acknowledgement
We are thankful to the Science and Technology
Research Partnership for Sustainable Development
(SATREPS), the Japan Science and Technology Agency
(JST)/Japan International Cooperation Agency (JICA) for
their financial support and Institute of Technology of
Cambodia for their laboratory cooperation.
References
[1] Jasovský, D., Littmann, J., Zorzet, A., Cars, O., 2016.
Antimicrobial resistance—a threat to the world’s
sustainable development. Upsala Journal of Medical
Sciences. 121, 159–164.
[2] Alam, M.J.B., Islam, M.R., Muyen, Z., Mamun, M.,
Islam, S., 2007. Water quality parameters along rivers.
International Journal of Environmental Science and
Technology. 4, 159–167.
[3] Brander, N., 2003. Drinking water in schools. Nursing
times. 99, 50–51.
[4] WHO., 1997. Guidelines for Drinking-water Quality.
World Health. 1, 104–8.
[5] CLSI., 2017. M100 Performance Standards for
Antimicrobial Susceptibility Testing.
[6] Vaz-Moreira, I., Nunes, O.C., Manaia, C.M., 2012.
Diversity and antibiotic resistance in Pseudomonas spp.
from drinking water. Science of the Total Environment.
426, 366–374.
[7] Martinez, J.L., 2009. Environmental pollution by
antibiotics and by antibiotic resistance determinants.
Environmental Pollution. 157, 2893–2902.
[8] Pitondo-Silva, A., Martins, V.V., Fernandes, A.F.T.,
Stehling, E.G., 2014. High level of resistance to

243
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

_________________________________________________________________________

Detection of Antibiotic-Resistant Bacteria in Water Environment of Tonle Sap Area and

Wastewater

Sovathana PHUONG1, Monychot Tepy CHANTO1,3, Chanthol PENG1,3, Kazuhiko MIYANAGA2 and
Reasmey TAN 1, 3,*
1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
School of Life Science and Technology, Tokyo Institute of Technology, 4259 J3-8
Nagatsuta-cho, Midori-ku, Yokohama, 226-8501, Japan
3
Food Technology and Nutrition Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
* Corresponding author: rtan@itc.edu.kh

Abstract

Use, misuse, and overuse of antibiotic have led to the emergence of antibiotic-resistant bacteria. However, the
prevalence of such bacteria in water environment of Cambodia is still less studied. Thus, this study was done in order to
detect antibiotic-resistant bacteria in water environment of Tonle Sap area (Tonle Sap River and Tonle Sap Lake) and
wastewater. Enumeration and isolation of antibiotic-resistant Klebsiella pneumoniae, Escherichia coli, and Staphylococcus
aureus were done by spread plate method and streaking on selective media supplemented with antibiotics. The isolated
strains were then used in antibiotic susceptibility experiment by disc diffusion method in order to confirm its resistance. The
result shows that highest concentration of antibiotic-resistant bacteria was found in wastewater, and a few sites of Tonle Sap
area where influence of anthropogenic pollution may cause promotion of gene transfer and acquisition of resistance genes.
From this study, it could be said that now antibiotic-resistant bacteria is highly prevalent in all sample sites which are in
close proximity with human population of water environment of Tonle Sap area.

Keywords: antibiotic-resistant bacteria, disc diffusion method, Tonle Sap area, wastewater

I. Introduction
Pollution of antibiotics from usage in medicine,
livestock, and aquaculture into water environment in
combination with wastewater discharge containing II. Methodology
antibiotics residues in environment lead to occurrence of
antibiotic resistance in environmental bacteria [12, 15].
2.1. Sampling sites and enumeration of antibiotic-
However, little is known about antibiotic-resistant bacteria resistant bacteria
that represent environmental isolates. As a result, this study Seven sampling sites which were chosen in this study
is conducted in order to detect antibiotic-resistant bacteria are located in Phnom Penh city (one from Tonle Sap River
in the water environment and wastewater in order to and one from Cheung Ek wastewater), Kampong Chhnang
contribute to lack of data prevalence of drug resistance in province (four from around Chnok Trou floating village)
environment. Recently, with the finding that there is a high
and Pursat province (one from Kampong Luong floating
concentration of Proteobacteria and Firmicutes in the
water environment of Tonle Sap area especially near the village). Antibiotic-resistant bacteria was enumerated by
floating villages [16], K. pneumoniae, E. coli, and S. using viable plate count method with spread plate
aureus are selected to study its antibiotic resistance. technique on DHL Agar (for K. pneumoniae), Chromocult
Agar (for E. coli), and Mannitol Salt Agar (for S. aureus).

244
The selective media were supplemented with antibiotics
such as ampicillin, ciprofloxacin, kanamycin, meropenem,
and vancomycin at breakpoint concentration.

2.2. Disc diffusion method for confirmation of

resistance
Four colonies with different size and morphology were
isolated from different media with different antibiotics by
streak plate method. Then, the isolated single colony was
used to prepare the overnight culture using LB broth.
Culture from LB broth was used for disc diffusion method,
in which six antibiotics discs (disc content: ampicillin 25
μg, kanamycin 30 μg, vancomycin 30 μg, ciprofloxacin 5
μg, meropenem 10 μg , and tetracycline 30 μg) were
applied. The diameter of the zone of inhibition was
measured and interpreted by using CLSI standard.
(b)
III. Results and Discussions

3.1. Concentration of antibiotic-resistant bacteria

Viable plate count of K. pneumonia, E. coli, and S.
aureus shows that high concentration of these bacteria
resistant to antibiotic used were found abundantly in
wastewater site since wastewater provides good condition
for proliferation of bacteria and transferring of resistance
genes [2]. For Tonle Sap area, the highest concentration of
antibiotic-resistant K. pneumonia and E. coli were found in
Tonle Sap river site, followed by Port site, Lake Water site, (c)
and Kampong Luong site which could indicate fecal
contamination in these area as well as influence of Fig. 1. (a). Concentration of antibiotic-resistant K.
anthropogenic pollution [3, 11]. However, high pneumoniae, (b). Concentration of antibiotic-resistant E.
concentration of antibiotic-resistant S. aureus are also coli, (c). Concentration of antibiotic-resistant S. aureus
found in Port site, UP site, and KL site which could be sue (AMP: ampicillin, KAN: kanamycin, VAN: vancomycin,
to shedding from users of water for recreational purpose CIP: ciprofloxacin, MEM: meropenem, WW:
Wastewater, TSR: Tonle Sap river, DS: Downstream, UP:
Upstream, LW: Lake Water, KL: Kampong Luong)

in addition to human and animal waste and waste disposal

[4, 5, 10]. Overall, the prevalence of antibiotic-resistant
bacteria in Phnom Penh sample sites was more than that in
sample sites of Kampong Chhnang and Pursat as the
pollution activity in the capital is much higher than in
province [8].

(a)

245
3.2. Resistance of isolated strains to antibiotics
Measuring of the diameter of the zone of inhibition for
each antibiotics yields the following result. From 73% to
100% of isolated strains of antibiotic-resistant K.
pneumoniae are resistant to ampicillin and vancomycin
which are intrinsic resistant of this bacteria [13] while high
resistance to tetracycline could be due to possession of tetB
or tetD gene that encode efflux pump [9]. For E. coli, high
resistance to vancomycin is also due to intrinsic resistant to
this antibiotic. However, high percentage of ampicillin-
resistant strain is expected as ampicillin is a drug that was
introduced in 1961 and has been used for a long time [14].
In addition, high resistance to tetracycline could be due to
the use of this antibiotic in aquaculture as prophylactics
which then lead to selection of resistant bacteria [7]. For

(c)

Fig. 2. (a). Resistance of K. pneumoniae to antibiotics, (b).

Resistance of E. coli to antibiotics, (c) Resistance of S.
aureus to antibiotics (AMP: ampicillin, KAN: kanamycin,
VAN: vancomycin, CIP: ciprofloxacin, MEM: meropenem,
WW: Wastewater,
S. aureus TSR:level
isolates, high TonleofSap river, DS:
resistance Downstream,
to ampicillin and
UP:
kanamycin were found for all sites that could Luong)
Upstream, LW: Lake Water, KL: Kampong be due to
production of staphylococcal β-lactamase [6] and
dissemination of aacA genes in the water environment [1].

IV. Conclusion
In conclusion, high concentration of these antibiotic-
resistant bacteria was found in Wastewater site and in
sampling sites of water environment of Tonle Sap area
(a) such as Tonle Sap river site, Port site, Lake Water site, and
Kampong Luong site where their locations are in close
proximity to human population.

Acknowledgement
We are thankful to the Science and Technology
Research Partnership for Sustainable Development
(SATREPS), the Japan Science and Technology Agency
(JST)/Japan International Cooperation Agency (JICA) for
their financial support.

(b)
Reference
[1] Baquero, F., Martínez, J. L., and Cantón, R., 2008.
Antibiotics and antibiotic resistance in water
environments. Current Opinion in Biotechnology, 19,
260–265.

246
[2] Bouki, C., Venieri, D., and Diamadopoulos, E., 2013. riverine antibiotic resistance genes. Environmental
Detection and fate of antibiotic resistant bacteria in Science and Technology, 46, 11541–11549.
wastewater treatment plants: A review. Ecotoxicology [12] Rozman, U., Duh, D., Cimerman, M., and Turk, S. Š.,
and Environmental Safety, 91, 1–9. 2020. Hospital wastewater effluent: hot spot for
[3] Czekalski, N., Sigdel, R., Birtel, J., Matthews, B., and antibiotic resistant bacteria. Journal of Water,
Bürgmann, H., 2015. Does human activity impact the Sanitation and Hygiene for Development, 1–8.
natural antibiotic resistance background? Abundance [13] Sarkar, P., Yarlagadda, V., Ghosh, C., and Haldar, J.,
of antibiotic resistance genes in 21 Swiss lakes. 2017. A review on cell wall synthesis inhibitors with
Environment International, 81, 45–55. an emphasis on glycopeptide antibiotics.
[4] El-Shenawy, M., 2005. Staphylococcus aureus and MedChemComm, 8, 516–533.
fecal indicators in Egyptian coastal waters of Aqaba [14] Simon, H. J. and Miller, R. C., 1966. Ampicillin in
Gulf, Suez Gulf, and Red Sea. Egypt Journal of the Treatment of Chronic Typhoid Carriers. New
Aquatic Research, 31, 113–124 England Journal of Medicine, 274, 807–815.
[5] Hatcher, S. M., Myers, K. W., Heaney, C. D., Larsen, [15] Suzuki, S. and Hoa, P. T. P., 2012. Distribution of
J., Hall, D., Miller, M. B., and Stewart, J. R., 2016. quinolones, sulfonamides, tetracyclines in aquatic
Occurrence of methicillin-resistant Staphylococcus environment and antibiotic resistance in Indochina.
aureus in surface waters near industrial hog operation Frontiers in Microbiology, 3, 1–8.
spray fields. Science of the Total Environment, 565, [16] Ung, P., Peng, C., Yuk, S., Tan, R., Ann, V.,
1028–1036. Miyanaga, K., and Tanji, Y., 2019. Dynamics of
[6] Kesharwani, A. K. and Mishra, J., 2019. Detection of bacterial community in Tonle Sap Lake, a large
β-lactamase and antibiotic susceptibility of Clinical tropical flood-pulse system in Southeast Asia. Science
Isolates of Staphylococcus aureus. Biocatalysis and of the Total Environment, 664, 414–423.
Agricultural Biotechnology, 17, 720–725.
[7] Li, D., Yu, T., Zhang, Y., Yang, M., Li, Z., Liu, M.,
and Qi, R., 2010. Antibiotic resistance characteristics
of environmental bacteria from an oxytetracycline
production wastewater treatment plant and the
receiving river. Applied and Environmental
Microbiology, 76, 3444–3451.
[8] MOE, 2009. Cambodia Environment Outlook. 1st ed.
Ministry of Environment, Phnom Penh.
[9] Nawaz, M., Khan, S. A., Tran, Q., Sung, K., Khan, A.
A., Adamu, I., and Steele, R. S., 2012. Isolation and
characterization of multidrug-resistant Klebsiella spp.
isolated from shrimp imported from Thailand.
International Journal of Food Microbiology, 155,
179–184.
[10] Plano, L. R., Garza, A. C., Shibata, T., Elmir, S. M.,
Kish, J., Sinigalliano, C. D., Gidley, M. L., Miller, G.,
Withum, K., Fleming, L. E., and Solo-Gabriele, H.
M., 2011. Shedding of Staphylococcus aureus and
methicillin-resistant Staphylococcus aureus from
adult and pediatric bathers in marine waters. BMC
Microbiology, 11, 1-10.
[11] Pruden, A., Arabi, M., and Storteboom, H. N., 2012.
Correlation between upstream human activities and

247
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
th
The 5 International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Isolation of Bacteriophages from Water Environment and Their Infectivities on

Multidrug Resistant Bacteria

Chakrya Theap1, Chanthol Peng*2), Monychot Tepy Chantho1, Kazuhiko Miyanaga3

1
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia,
Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
2
Water and Environment Research Unit, Research and Innovation Center, Institute of Technology of
Cambodia, Russian Federation Blvd., P.O. Box 86, 12156 Phnom Penh, Cambodia
3
School of Life Science and Technology, Tokyo Institute of Technology, 4259 J2-15 Nagatsuta-cho, Midori-ku, Yokohama, 226-850
1, Japan
* peng@itc.edu.kh

Abstract

The increasing of antibiotics resistance bacteria becomes a current critical problem which has to be solved as soon as
possible. Bacteriophage is found as an alternative treatment to antibiotic to treat the bacterial infection with its high specificity
to their host. The seven bacteriophages were screened from wastewater and used to treat with the six resistance strains of
Escherichia coli such as E. coli 09-02E, E. coli E302, E. coli papRG-06-5, E. coli Mt1B1, E. coli STEC711 and E. coli
ECCRA-119 which were isolated from river water and wastewater in the previous study. The host range of 7 phages were
determined by performing spot test with 6 primary hosts which mentioned above, and one phage was eventually selected for
further study due to its high infectivity. Phage ΦW1El had the widest host range, producing clear plaques on two hosts, E.
coli 09-02E and E. coli E302. The measurement of optical density at the wavelength of 660 nm (OD660) of ΦW1El treated
with both strains was confirmed the infectivity of phage and indicated that phage activity against both strains of bacterial was
obtained within 3 h to 9 h. However, the phage began to lyse at 3 h but then the phage-resistant bacteria appeared after 9h.

Keywords: Escherichia coli; multidrug-resistant bacteria; phage-antibiotic combination; phage therapy; the infectivity of
phage

I. Introduction including bacteriophage attach to the susceptible host

bacterium then infects the bacterial and insert its DNA into
The inappropriate use of antibiotics occurred from
bacterial chromosome, assembled the phage protein and
agricultural and hospitalized were considered as the biggest
then the process of phage lysis is started [2]. By the reason
major of increasing multi-drug resistant bacterial (MDR) [3,
of some antibiotics became less effective, this study aims to
4]. Bacteriophage (phage) was considered as an alternative
isolate phage from aquatic environment and detect their
way to antibiotics [2, 5, 6]. Phages are viruses that can infect
infectivity on multidrug resistance bacteria through their
and kill bacteria without any negative effect on human or
host range and also the phage infection in culturing by
animal cell. The phage pathway involves in several steps

248
observation of optical density at the wavelength of 660nm. The spot test method was mentioned in [7]. The positive spot
test will show up as total pulverization of the whole drop
II. Materials and Methods area, whereas a negative spot test will result within the
bacterial lawn growing normally in the region of the spot.
2.1. Sampling sites, E. coli strains, isolation of
The lytic ability of phages was considered by the plaques
bacteriophages
clarity which was observed as clear, turbid, and faint.

Three different samples were collected from Tonle Sap

2.4. Optical density of bacterial
River, Trabek Chanel and Cheoung Ek Lake. Pure strains E.
coli k12 and E. coli 0157 and the multidrug strains which
The effect of phage was observed by using the optical
were isolated from surface water and wastewater were used
density at the wavelength of 660nm (OD660) measurement.
for phages isolation. Initially, host cell of the bacterial was
Initially, 60μl of bacterial overnight culture with the
cultured in 2ml Luria-Bertani (LB) overnight. After getting
concentration of 107 CFU/ml was added into glass tube
host cell overnight culture, phage solution had to be diluted
containing 6ml of LB broth and shaking at 120rpm, 37℃
by Sodium-magnesium buffer (SM buffer) (5.8g of NaCl,
in Bio-shaker. After an hour incubating of bacterial culture,
2.0g of MgSO4, 50ml of 1M Tris-HCl (pH7.5), 0.1g of
phages concentration was 107 PFU/ml were added at the
gelatin and 1L of distilled water). LB soft agar (10g of
MOI=1 to each glass tubes and the OD was measured every
tryptone, 5g of yeast extract, 5g of NaCl, 5g of agar granule
one hour. This process was conducting until 24h.
and 1L of distilled water with 5ml of MgSO4 1M, 0.5% (w/v)
and 5ml of CaCl2 1M, 05% (w/v)) in the glass tubes were
melt by boiling water and cooling down to 45 to 46℃ in
III. Results and Discussion
heating block. Then, the plaque assay method was conducted. 3.1. Isolation of bacteriophages from aquatic
All of the LB plates were incubated at 37℃ upside down environment and host range
overnight. Thus, after overnight incubation, the plaques
appeared and individual plaque was picked up by truncated
Phages were screened from three different sources of
plastic tip then transfer the isolation phage with its host to
water such as from surface water and wastewater which
the 2ml fresh LB medium and then shaking for several hours.
resulted in the isolation of lytic phages against E. coli K12
After incubation, the solution was centrifuged at 10000×g,
and E. coli 0157. Phages isolated from three different water
10mn, 4℃ and use the supernatant as an isolated phage
samples were tested using plaque method. In an effort to
solution.
determine the most hopeful applicant for lysis of the widest
range of E. coli strains, each of the 7 phage types was spotted
2.2. Phage purification
onto the bacterial lawns. As the result we found only one
phage ΦW1El showed high infectivity against 50% of all
To obtain the pure phage, different dilution of
representative strain of E. coli (Figure 1). There are only
previous phage solution was used with different host
two multidrug resistant strains of E. coli, 09-02E, E302 that
bacterium by repeated plaque assay until getting the uniform
were able to get infected by ΦW1El. None of the phages
plaques and picking a single plaque. After that, 5ml of SM
were able to produce clear plaques Mt1B1 resistance cell of
buffer was added directly to the plate surface followed by
E. coli, although some phages did produce turbid plaques.
plate lysate and the PEG [#6000]-NaCl precipitation method
For multi-drug strain ECCRA-119, only one ΦW2P2 was
(10%(w/v) of PEG [#6000], 4% (w/v) of NaCl) to collect
able to produce turbid plaques. One phage, classified W1EL,
phage. produced the clearest plaques on two resistance cells such as
E. coli 09-02E, E. coli E302, and another two pure strains E.
2.3. Spot test coli K12 and E. coli 0157. To be the most promising
candidate, host range and the clarity of plaques which is
A spot test method is a quick way to check whether a considered as lytic ability are required. Among of seven
phage sample can infect a bacterium by dropping a small phages, there is only one phage, ՓW1EL was found to be
spot of phage titer onto a plate inoculated with the bacterium. the most auspicious candidate. ՓW1EL produced clear

249
plaques on two resistance strains whereas another six phages its host and then the bacterial start steadily recover after 3h.
only produce turbid plaques are judged as less infection. From 3h until 9h, the OD660 remained almost steady. After
9h, it was observed that the OD660 slightly increase as the
number of phage resistant bacteria increased in both strains.
However, the four other strains did not show any differences
even with or without phage added which means that phage
W1EL was not able to infect those bacteria hosts. The
increasing of bacterial density might be caused by lysis from
without. The lysis from without described an early bacterial
lysis induced by high multiplicity virion adsorption and that
occurs without phage production simply say, lysis does not
depend on phage infection but instead is influenced directly
by extracellularly contributed agents [1]. A large amount of
phage particles is adsorbed to the host bacterium, the
decreasing turbidity of bacterial is considered as phage lysis.
Lysis from without is different from normal lysis which
occurs at the end of the phage latent period with release of
new phage [8].

E. coli 09-02E
2.2
2.0
1.8
1.6
1.4
OD (660 nm)

1.2

Figure 1. Spot tests for the 7 phages with the widest 1.0

host ranges and representative E. coli lawns in 0.5% agar LB 0.8

0.6
medium. The host bacterial strain is indicated at the bottom
0.4
of each plate. A key for the phage spots is shown in the
0.2
bottom right panel. 0.0
0 5 10 15 20 25

3.2. Optical density of bacterial Time (h)

Control W1K12 W1K21 W1K22
W1K23 W1K32 W1El W2P2
By the reason of phage W1EL which was the only
one phage that show the widest host range and much
activities against two primary hosts that already mentioned
from above, it was considered to choose for the further study.
In order to identify the effects of ΦW1El on E. coli, each
phage was cultured in the LB liquid medium. The reduction
in the OD660 was considered as a consequence of the lytic
activity of a phage on its host. A following decrease in the
OD660 was considered to be the reduction in the nutrients
available as the OD660 of the bacterium-only control also
declined. The obtained results illustrate that a phage ΦW1El
infect strongly against on two primary hosts, E. coli 09-02E
and E. coli E302 (Figure 2). The decreasing of turbidity in
bacterial was notified during 3h as the phage began to lyse

250
 E. coli E302 Enfermedades Infecciosas y Microbiologia Clinica,
2.2
31, 3–11
2.0
[4] Khachatourians, G. G., 1998. Agricultural use of
1.8 antibiotics and the evolution and transfer of antibiotic-
1.6 resistant bacteria. Cmaj, 159, 1129–1136
1.4 [5] Lin, D. M., Koskella, B., and Lin, H. C., 2017. Phage
OD (660 nm)

1.2 therapy: An alternative to antibiotics in the age of

1.0 multi-drug resistance. World Journal of
0.8 Gastrointestinal Pharmacology and Therapeutics, 8,
0.6
162.
0.4
[6] Nilsson, A. S., 2019. Pharmacological limitations of
phage therapy. Upsala Journal of Medical Sciences,
0.2
124, 218–227.
0.0
0 5 10 15 20 25 [7] Synnott, A. J., Kuang, Y., Kurimoto, M., Yamamichi,
Time (h)
K., Iwano, H., and Tanji, Y., 2009. Isolation from
Control W1K12 W1K21 W1K22 sewage influent and characterization of novel
W1K23 W1K32 W1El W2P2 Staphylococcus aureus bacteriophages with wide host
Figure 2. The culture of bacteriophage and E. coli resistance ranges and potent lytic capabilities. Applied and
strains in LB medium. The bacterial strain was indicated in Environmental Microbiology, 75, 4483–4490
each graph. [8] Visconti, N., 1953. Resistance to lysis from without in
bacteria infected with T2 bacteriophage. Journal of
bacteriology, 66, 247–253
IV. Conclusion
In conclusion, only ΦW1EL showed infectivity on
multidrug resistant strains of E. coli. However, the
infectivity was obtained within 3 h to 9 h. After that, the
phage resistant appeared, yet lower than the bacterial control.
Therefore, the phage infection with the same concentration
of bacterial is not sufficient to use in this case.

Acknowledgement
This research was supported by the Science and
Technology Research Partnership for Sustainable
Development (SATREPS), the Japan Science and
Techonolgy Agency (JST)/Japan Internation Cooperation
Agency (JICA) with the grant-number JPMJSA1503.

References

[1] Abedon, S. T., 2011. Lysis from without.

Bacteriophage, 1, 46–49
[2] Azam, A. H. and Tanji, Y., 2019. Bacteriophage-host
arm race: an update on the mechanism of phage
resistance in bacteria and revenge of the phage with
the perspective for phage therapy. Applied
Microbiology and Biotechnology, 103, 2121–2131
[3] Cantón, R., Horcajada, J. P., Oliver, A., Garbajosa, P.
R., and Vila, J., 2013. Inappropriate use of antibiotics
in hospitals: The complex relationship between
antibiotic use and antimicrobial resistance.

251
 The 13th AUN/SEED-Net Regional Conference on Chemical Engineering 2020 (RCChE-2020)
Jointly held with
The 5th International Symposium on Conservation and Management of Tropical Lakes
“Insights and Challenges toward Achieving SDGs”

Price Evaluation and Quality Control of Different Soy Sauces Sold in the Markets

Luka LY 1, Monychot Tepy CHANTO2,3, Chanthol PENG2,3, Reasmey TAN 2,3*

1
Graduate School, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, Phnom Penh, Cambodia.
2
Food Technology and Nutrition Research Unit, Institute of Technology of Cambodia, Russian Federation Blvd.,
P.O. Box 86, Phnom Penh, Cambodia.
3
Faculty of Chemical and Food Engineering, Institute of Technology of Cambodia, Russian Federation Blvd.,
P.O. Box 86, Phnom Penh, Cambodia.
* rtan@itc.edu.kh
Abstract

This study is aimed to evaluate the price and quality control of different soy sauces sold in the markets. Four local
markets and five supermarkets were selected to survey the price and collecting samples. To analyze the quality of soy sauce,
pH, salt content, soluble salt - free solid, reducing sugar, total nitrogen, amino acid nitrogen, halophilic yeast and total
bacteria were selected in determining soy sauce quality. As a result, Cambodian soy sauce, Vietnamese, and Thai soy sauce
were abundant in the local markets, and Chinese soy sauce, Malaysian soy sauce, Taiwanese and Indonesian soy sauces
were found mostly in supermarkets. Within 42 soy sauce brands found, the price was between 400 riel/100mL and 12000
riel/100mL. Within the physiochemical analysis of the 42 soy sauce samples, their pH was between 3.79 to 4.62, Salt was
between 4.7 g/100 mL to 23.40 g/100mL, Reducing sugar was between 0.21 g/100 mL and 9.57 g/100 mL, soluble salt-free
solid content was between 1.21 g/100mL and 38.65 g/100 mL, Total nitrogen was between 0.08 g/100 mL and 2.2 g/100 mL
and Amino acid nitrogen was between 0.01 g/100mL and 1.2 g/100mL. For microbiological quality, there were 3 soy sauces
that had the presence of halophilic yeasts and Total plate count was detected in 50% of soy sauces but they were lower than
standard limits. By comparing the results in the physicochemical analysis and microbiological quality, there were 11
samples that were within the specifications of Cambodia’s national soy sauce standards. For further study, 3-MCPD and the
volatile compounds of 42 soy sauces will be analysed.

Keywords: Soy sauce, brand, Cambodia, physicochemical parameters, microbiological analysis

I. Introduction
Soy sauce has been a liquid condiment originated from that, scientists found that soy sauce also contains bioactive
China since approximately 2200 years ago. It is produced compounds with anticarcinogenic, antimicrobial,
by mixing the steamed soybeans with roasted and grided antioxidative, and antiplatelet activities [4] and also
wheat, and Koji mold spores and incubate for 3 to 5 days antiallergenic activity [5]. There are many different soy
followed by fermentation in brine for about 3 to 6 months sauces sold in Cambodia; however, the study of their
[1]. People in Asian countries such as China, Japan, Korea, quality is unknown. The price and quality are essential for
Thailand, the Philippines, Indonesia, and more than 90% of attracting the consumers. The main objective of this study
Cambodian people consumed soy sauce [2]. Soy sauce is was to know the price and quality of soy sauces currently
normally used in cooking, such as in soups, frying, roasting, sold in Cambodia's markets by providing awareness to the
marinating, and also for dipping. It is widely used in the consumers and providing information to the manufacturers
purpose of food, for improving the taste, flavour and aroma, to be more careful with the quality of soy sauce production.
as it is the main source of natural umami taste [3]. Not only

252
II. Materials and Methods
Halophile yeasts were cultured by using Potato
2.1. Survey of soy sauce price
Dextrose Salt agar. Media was made by weighting 24 g of
The survey was conducted in 2 weeks from 26 February potato Dextrose Broth powder and 15 g of agar powder
to 03 March 2020. Market survey was divided into 2 together into a 1000 mL bottle with 100g of sodium
sections in the local markets and supermarkets around and chloride, then stir and Autoclave 15 minutes at 121 ℃
in the middle of Phnom Penh city. First section was local after autoclaved keep it until 45 ℃ in the water bath and
markets such as Psa Thom Thmey, Samaki, Tu Tum Pung, add 10 mL of 10% tartaric acid solution, then media was
and Ou Reusey. In local markets, 4 sellers were selected in poured (about 25 to 30 mL) to the plate and keep it until
Psa Thom Thmey, 3 sellers were selected in Psa Samaki, 4 cool, 0.1 mL of soy sauce sample was pipetted to the hard
sellers were selected in Psa Tu Tum Pung, and 6 sellers agar media, then a spreader was used to spread the sample
were selected in Psa Ou Reusey. The distance between each over the agar until dried [6]. Plate Count Agar was used to
seller was 3 to 4 shops. And the second section was in grow and count the total number of aerobic microbial
supermarkets such as: Lucky Olympic, Lucky Steung contaminants in samples. 23.5 g of PCA was added and
Mean Chey, Aeon I and II, and Macro supermarket. Data 1000 mL of distilled water, and then stirred until becoming
were collected by recorded the volume of soy sauce bottles, a homogenous solution, sterilized in an autoclave set at
price and country of production origin. 121°C for 15 minutes. Sample was tested 3 times by taking
1 ml of soy sauce diluted with buffer sodium chloride, after
2.2. Physicochemical analysis of soy sauces that pouring 15-20ml of PCA at 45°C ±1°C into the Petri
and put it into the incubator at 36°C ± 1°C for 2 days.
Soy sauce samples were bought from the markets to
analyze their physicochemical characteristics. The salt
III. Results and Discussion
concentration was analysed using salt-meter (Atago model
ES-421), pH measurement was analysed by using pH There are 42 different soy sauce brands that were
meter (OHAUS, Starter300), soluble salt free solid content collected from local markets and supermarkets. Price of
and amino nitrogen was analysed as described in GB/T each soy sauce brand sold in local markets and
18186-2000 [14], and total nitrogen was analysed as supermarkets is shown in Fig. 1.. Within a price from 400
described in AOAC. (2000) [15] The soy sauce samples riel to 4000 riels per 100 mL in local markets, Cambodian
were analysed in triplicate and data was represented in the soy sauces were the same price or cheaper than Thai and
form of mean ± standard deviation (SD). Analysis of Vietnamese soy sauces, except the C7 that has high price of
variance (ANOVA) with Tukey's test was performed to 2,000 riels and, there were 38 brands of soy sauces sold in
evaluate significant difference in the physicochemical the supermarkets. Soy sauces imported from Japan, Hong
parameters from different samples, and a significant Kong, Taiwan, Korea and France can be found in the
difference was defined as p < 0.05, and one-way ANOVA supermarkets with the high price varying from 550 to
was conducted using SPSS software. 12,000 Riels per 100 mL among soy sauces sold in markets,
the price of J2 soy sauce was the highest.
2.2. Microbiological analysis of soy sauces
12000
Price in the market (100 mL)

10000 Price in supermarket

Price in local market

8000

6000

4000

2000

0
C1
C2
C3
C4
C5
C6
C7
C8
C9
Th1
Th2
Th3
Th4
Th5
Th6
Th7
Th8
Th9
Th10
Th11
Th12
Th13
Ch1
Ch2
Ch3
Ch4
Ch5
Ch6
Ch7
V1
V2
V3
V4
S1
S2
J1
J2
M
H
K
T
F

Soy sauce
Fig. 1. Price of soy sauce sold in super markets compared with local market, The alphabet codes (C, Th, Ch, V, S, J, M, H,

253
 K, T, and F) are represented: (Cambodia, Thailand, China, Vietnam, Singapore, Japan, Malaysia, Hong Kong, Korea,
Taiwan and French), respectively.

According to the institute of standard of Cambodia, soy ± 0.02 and 4.62 ± 0.01, soy sauce C4, C5, C6, C8, TH3,
sauce’s pH should be between 4.2 and 4.8 [8]. As shown in Ch6, V1, V4 were lower than the standard.
Fig 2. The pH of the soy sauce samples was between 3.79
5.2
5.0
4.8
4.6
4.4
pH

4.2
4.0
3.8
3.6

T
Th1
Th2
Th3
Th4
Th5
Th6
Th7
Th8
Th9
C1
C2
C3
C4
C5
C6
C7
C8
C9

Th10
Th11
Th12
Th13
Ch1
Ch2
Ch3
Ch4
Ch5
Ch6
Ch7
V1
V2
V3
V4
S1
S2
J1
J2
M
H
K

F
Soy sauce samples

Fig.2. pH of soy sauce samples. The alphabet codes (C, Th, Ch, V, S, J, M, H, K, T, and F) are represented: (Cambodia,
Thailand, Vietnam, Singapore, Japan, Malaysia, Hong Kong, Korea, Taiwan and French), respectively.

Salt is an important ingredient used in soy sauce Cambodia, soy sauce was divided into 2 parts, such as dark
fermentation, a good fermented soy sauce should have 18% soy sauce and light soy sauce. Light soy sauce had a
to 20% of salt [6]. Out of 42 soy sauce samples, soy sauce soluble salt free solid content about (15 to 18 g/100 mL)
code M has a salt content of only 4.7 g/100 mL, with an
and dark soy sauce has soluble salt free solid content about
amount so low, this product needs to add a food
preservative to ensure safety and a longer shelf life. (22 to 44 g/100 mL). Total nitrogen is another parameter
While Ch6 has salt content 23.40 g/100mL this amount is indicating the quality in soy sauce [6]. According to the
very high over the Cambodian soy sauce standard [8], national institute of standards in Cambodia, soy sauce must
consumption of too much sodium is not good for the health, have at least total nitrogen 0.8%. There are 22 soy sauce
especially for someone who suffers from heart disease, samples having the total nitrogen greater than 0.8 g/100mL.
hypertension and kidney disease [9]. Based on the national Total nitrogen and total amino nitrogen was used as an
institute of standards of Cambodia, the maximum amount indicator to show the effect of fermentation too, by the way
of salt allowed in soy sauce is 20%. Amount of 42 soy in sample C8 (0.08 ±0.02 g/100mL) total nitrogen so low it
sauce samples, there are only 7 soy sauce simples had high could be from poor fermentation, short period of time or
amount of salt over 20%. Reducing sugar is very important the digestion of soy beans was not done well, moreover this
for making a good taste and quality of soy sauce, and had soy sauce could be make from excessive dilution [10].
about 2 to 5% in a good quality of soy sauce [6]. Within 42 Amino nitrogen is one of the testing parameters showing us
soy sauce samples the lowest reducing sugar was found in the quality and effective of soy sauce fermentation.
“C8” has the total amount 0.21 ± 0.01% while the highest Amount of 42 soy sauce samples, there are 23 soy sauces
of reducing sugars was found in sample “S2” has the total having the amino nitrogen bigger than 0.26 g/ 100 mL,
amount 9.57 ± 0.06% by the way there were only 12 Th13 sample had amino nitrogen 1.18 ± 0.17 (g/100 mL)
samples found in the range of 2 to 5. The lowest amount of greater than other samples, by the way “C8” samples had
soluble salt free solid content was found in “C8” sample amino nitrogen only 0.01 ± 0.00 (g/100 mL) lower than the
(1.21±0.09 g/100mL) while the highest was found in “M” other samples. According to institute of standard of
sample (38.65 ± 0.14 g/100ml) the main reason that made Cambodia the amino nitrogen in soy sauce must be equal
sample “M” has the soluble salt free solid greater than or bigger than 0.26 g/100 mL [7].
other samples it could be from adding sugar [11] and it also Within the experiments there were 3 samples found having
described in soy sauce ingredient on soy sauce “M” about halophilic yeast inside. TH5 has halophilic yeast more than
the adding of sugar in the ingredient. There are 18 samples other samples followed by C6 and C2.as shown in Table 1.
having an amount of soluble salt free solid content bigger The presence of halophilic yeasts in the samples should be
than 15. According to the Institute of standards of from the lack of hygienic practice in the production,

254
pasteurize or sterilize not well. The halophile yeasts could [2] Theary, C., Panagides, D., Laillou, A., Vonthanak, S.,
deteriorate soy sauce quality; it could be given off an Kanarath, C., Chhorvann, C., Sambath, P., Sowath, S.
offensive odour and forming the thin pellicle during long & Moench-Pfanner, R. (2013). Fish sauce, soy sauce,
term storage [12]. and vegetable oil fortification in Cambodia: where do
we stand to date? Food Nutr. Bull. 34.
Table 1. Halophilic yeast detected in the soy sauce samples. [3] Marcus, J.B. (2019). Flavor Enhancement Ingredients,
Samples Halophilic yeast colony (CFU/mL)
in: Aging, Nutrition and Taste. Elsevier, pp. 173–206.
[4] Murooka, Y. & Yamshita, M. (2008). Traditional
C2 1.1± 1.00×102 healthful fermented products of Japan. J. Ind.
C6 2.2 ± 2.64×101 Microbiol. Biotechnol. 35: 791 –798
Th5 2.35 ± 8.02×102 [5] Kobayashi, M. (2005). Immunological functions of
soy sauce: Hypoallergenicity and antiallergic activity
of soy sauce. J.
Based on the Nation Institute of Standard of soy sauce in
[6] Food and Agriculture Organization, 1984. Manuals of
Cambodia, Halophilic yeasts much be absence in the soy
food quality control. 1st ed. Rome, Italy.
sauce. According to the nation food safety soy sauce
[7] Luh, B. S., 1995. Industrial production of soy sauce.
standard in China, the total number of bacterial colony
Journal of Industrial Microbiology, 14, 467–471.
should be lower than 5000 CFU/mL [13]. The highest
[8] CS 066:2011. (2012)., Institute of standards of
bacteria grown on “S1” sample about 1.6×103 CFU/mL, all
Cambodia., Soy sauce standard.
of samples had total bacteria lower than 5×103 CFU/mL or
[9] He, F.J., MacGregor, G.A., 2010. Reducing
not detected in the samples.
Population Salt Intake Worldwide: From Evidence to
IV. Conclusion Implementation. Prog. Cardiovasc. Dis. 52, 363–382.
[10] Röling, W.F.M., Van Verseveld, H.W., 1996.
From the market survey on 42 soy sauce samples, the
Characterization of Tetragenococcus halophila
price was between 400 riel/100mL to 12000 riel/100mL,
populations in Indonesian soy mash (kecap)
For the physiochemical tests, and microbiological tests, as
fermentation. Appl. Environ. Microbiol. 62, 1203–
the results, 11 samples has found to have high quality as
1207.
shown in Institute of Standard of Cambodia such as one of
[11] Garnida, Y., Taufik, Y., 2014. Fermentation in Salt
Cambodia soy sauce (C7), 2 of Thai’s soy sauce (Th6,
Solution to Produce Jack Beans (Canavalia
Th12), one of China’s soy sauce (Ch3), a Singapore’s soy
ensiformis L) SAUCE, International Congress on
sauce (S2), two of Japan’s soy sauce (J1, J2), a Hong
Challenges of Biotechnological Research in Food
Kong’s soy sauce (H), a Malaysia’s soy sauce (M), a
and Health. Slamet Riyadi University.
Taiwan’s soy sauce (T) and a French’s soy sauce (F).
[12] Lee, Y. T.-S., Chu, Y.-H., Shin, B.-K., and and Ju-
For further research, volatile compounds, 3-MCPD,
Hyun, 1975. Study on Preservation of Soy Sauce. , 7,
development of high-quality soy sauce with the benefits for
200–207
the human health and price competition should be
[13] GB/T GB 2717-2018. (2000)., Nation standard of the
conducted.
people's republic of china., Soy sauce. Chinese
Standard Publishing House.
Acknowledgements [14] GB/T 18186-2000. (2000)., Nation standard of the
people's republic of china., Fermented soy sauce.
We would like to acknowledge the Ministry of
Education, Youth and Sport for financial support through Chinese Standard Publishing House.
Higher Education Improvement Project (HEIP) Research [15] AOAC. (2000). Official Methods of Analysis 17th
Grant. Ed., AOAC IN-TERNATIONAL, Gaithersburg, MD,
Method 991.20.
References
[1] Leboffe, M.J. & Pierce, B.E. (2006). Microbiology:
Laboratory Theory and Application, Essentials -
Michael J. Leboffe, Burton E. Pierce - Google Bøker.
Morton Publishing Company.

255