Ecological Engineering 82 (2015) 272–275

Contents lists available at ScienceDirect

Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng

Short communication

Treatment of a low-strength bilge water of Caspian Sea ships by HUASB

technique
Seyyed Mohammad Emadian a , Morteza Hosseini b , Mostafa Rahimnejad c,d, * ,
Mohammad Hassan Shahavi b , Behnam Khoshandam a
a
Department of Chemical Engineering, Semnan University, Semnan, Iran
b
Department of Chemical Engineering, Babol University of Technology, Babol, Iran
c
Biofuel & Renewable Energy Research Center, Faculty of Chemical Engineering, Babol Noshirvani University of Technology, Babol, Iran
d
Advanced Membrane & Biotechnology Research Center, Faculty of Chemical Engineering, Babol Noshirvani University of Technology, Babol, Iran

A R T I C L E I N F O A B S T R A C T

Article history: Oily bilge water is one of the major pollutants that threats marine environment due to its direct discharge
Received 9 July 2014 from ships into sea. The aim of this paper is to investigate the possibility of dilute bilge water treatment
Received in revised form 16 January 2015 by using hybrid up-flow anaerobic sludge blanket (HUASB) bio-reactor. The reactor operated at two
Accepted 5 April 2015
hydraulic retention times (HRTs) of 10 h and 8 h. The organic loading rate (OLR) was gradually increased
Available online 16 May 2015
from 0.12 g to 0.6 g chemical oxygen demand (COD)/l day. After the immobilization of sludge on the
surface of the support materials and 10 days of batch feeding of the reactor with the waste water as
Keywords:
acclimation period (with COD removal of 59%), the continuous operation of the reactor started. At the end
Anaerobic treatment
HUASB reactor
of the experiment, with the HRT of 8 h and OLR of 0.6 g COD/l day, the COD removal efficiency reached the
COD amount of 75%. Furthermore, the bio-reactor showed a good performance in removing oil from the waste
pH stream which was significantly lower than the standard value which has been laid down for the discharge
Oil content of the bilge water from ships by the International Maritime Organization (IMO). The obtained data
demonstrated that the HUASB reactor is an appropriate system for the treatment of a low-strength bilge
water.
ã 2015 Elsevier B.V. All rights reserved.

1. Introduction the microfiltration of bilge water as a pretreatment stage is

desirable because used oils and particulates can block the feed
Bilge water is a corrosive mixture of seawater containing a channels of UF spiral. However, some reports indicated the
variety of constituents including cleaning agents, solvents, fuel, disadvantages which associated with the application of membrane
lubricating oils and hydraulic oils. The International Maritime in treatment of bilge water. These disadvantages contain their
Organization (IMO) regulates the discharge limit for ships which is relatively high cost of production due to the expensive raw
15 mg/l (MARPOL, 1973). Oil water separators (OWS) are usually materials, fouling which has a number of negative effects such as
employed to treat bilge water. OWS processes are gravity the reduction in membrane flux, additional capital and mainte-
separators based on the density variation between oil and water nance cost because of membrane replacement and regeneration.
phases. Cleaning agents in bilge water can create an emulsion of oil Karakulski et al. (1998) used a laboratory-scale ultrafiltration pilot
in water. When the emulsification takes place, buoyancy difference plant with tubular membranes for the treatment of bilge water.
of oil and water is too small to be treated properly via the existing However, the use of additional photo catalytic oxidation stage was
OWS technology. necessary to eliminate the residual oil. Rincon and La Motta, (2014)
In recent years, several researches are conducted on bilge water investigated the treatment of synthetic ship bilge water in an
treatment methods including ultrafiltration (UF), electrocoagula- electrocoagulation system. They concluded that the electro-
tion and UF/photocatalytic oxidation. Peng et al. (2005) stated that coagulation process was an effective method in destabilizing of
oil in water emulsions and removing the heavy metals. However,
the authors applied additional flotation method to improve the
* Corresponding author. Tel.: +98 1113234204.
treatment performance. Sun et al. (2010) investigated on the
E-mail addresses: rahimnejad@nit.ac.ir, rahimnejad_mostafa@yahoo.com performance of biofilm-membrane bio-reactor (MBR) for treating
(M. Rahimnejad). bilge water. Although the efficiency of the reactor in removal of

http://dx.doi.org/10.1016/j.ecoleng.2015.04.055
0925-8574/ ã 2015 Elsevier B.V. All rights reserved.
 S.M. Emadian et al. / Ecological Engineering 82 (2015) 272–275 273

COD and oil was promising, the membrane was seriously fouled the sample surface. The sample surface was also collected in a
without a recycling side-stream to the bio-reactor. closed bottle.
In this study, anaerobic treatment has been chosen because it is
an efficient, simple and economical method. It is also environ- 2.4. Analytical methods
mentally-friendly way. The hybrid up-flow anaerobic sludge
blanket (HUASB) reactor configuration has combined both Several monitoring parameters were evaluated during the
suspended and fixed growth of bacteria in one reactor. It has also entire operation, including COD and oil concentrations, as well as
used the advantages of both up-flow anaerobic sludge blanket pH and temperature. For COD analysis, HACH’s Method 8000, a
(UASB) and up-flow anaerobic fixed film (UAFF) reactors. This kind combination of reactor digestion method and colorimetric method,
of reactor is efficient in the treatment of dilute to high strength was used (DR/890, 2009). This method is equivalent to standard
wastewaters at low to high organic loading rates. Bilge water is method 5220D: closed reflux, colorimetric method (APHA, 2008).
classified in the low-strength group of wastewater. Although Analysis of oil was determined according to USEPA 1664 and
anaerobic process is used for the treatment of medium and high n-hexane gravimetric methods. Temperature and pH were
strength wastewaters, it has already been applied successfully for a measured by using a pH/temperature probe (HANNA, PH212,
number of waste streams including low-strength wastewaters. Germany) with automatic temperature compensation. The method
Also, the efficiency of HUASB reactor (on the basis of COD and oil used in pH measurement was generally in compliance with
removal) in treatment of low-strength bilge water under different Standard Method 4500B (APHA, 2008).
low organic loading rates has been studied.
2.5. Start-up and operation scheme
2. Material and methods
Start-up period is usually a time consuming period. In order to
2.1. Experimental apparatus decrease the time, the immobilization of biomass on the support
material was done; therefore, the mentioned mixture of sludge
The fabricated Plexiglass reactor column with an internal was used by means of a technique described by Zaiat et al. (1994).
diameter of 4.4 cm and a liquid height of 194 cm was used. The The support material together with the sludge was stored in 1.5 l
column included three sections: bottom, middle and top sections. closed bottle and homogenized for the period of a week so as to
The bottom part of the column, with a volume of 1823 ml operated secure steadier immobilization of bio-particles in the supporting
as a UASB reactor whereas the middle part of the column with a material. After this stage, the packing material filled in its place in
volume of 855 ml was used as a fixed film reactor. The top part of the HUASB reactor. The reactor was inoculated with 500 ml of the
the bio-reactor with a volume of 273 ml was an unpacked column same sludge mixture. In order to acclimatize the sludge with bilge
prior to the effluent overflow. The fixed film section of the column water, the daily batch feed reactor with the bilge water (50 mg/l)
was randomly packed with 270 billowy pieces of PVC rings which was selected for 10 days. After each feed, the liquid content of the
had a diameter of 15 mm. The height of each ring was equal to reactor was continuously circulated for 1 day (until the next feed).
13 mm (150 m2/m3 specific surface areas for each one). The media The acclimation period permitted oxygen level to decrease to
in the reactor were stabled by using a plastic mesh. The wastewater prevent inhibition of anaerobic bacteria as well as the bacteria
as a substrate was continuously fed to the base of the reactor which population to adjust with the feed wastewater. The TSS concen-
was located under the bed of active sludge and through a T-inlet tration of the sludge after the 10-day batch-fed period was 16.5 g/l.
connected to a peristaltic pump. An outlet was provided at the top A COD removal efficiency of about 59% was achieved at the end of
of the reactor that was connected to a 1 l funnel shaped settling this acclimation period.
compartment served as a sedimentation part where the final The start-up was carried out by using stepped organic loading
effluent was collected from the top of this tank. The reactor to produce the most rapid biomass development. The HRT of 10 h
operated at ambient temperature (15–25  C). was kept constant throughout the start-up duration and the OLR
increased from 0.12 g to 0.24 g COD/l day. The reactor was allowed
to reach steady state condition before each OLR change. When
2.2. Wastewater characteristics
effluent COD reached a relatively constant value, the steady state
condition was achieved and then influent OLR can be raised. The
The bilge water was collected from the ships which had
experimental procedure has been illustrated in Fig. 1.
anchored at Amirabad port, Behshahr, Mazandaran, Iran. The
HUASB reactor was fed with bilge water pre-settled for 10 min. The
characteristics of pre-settled bilge water were as follows: pH: 8–9, 250 start-up-
COD: 20–200 mg/l, total solid (TS): (800–2400) mg/l, TSS: later opration
0.9
220–1760 mg/l, total nitrogen: 836 mg/l and total phosphorus
Influent COD (mg/l)

(TP): 211 mg/l (TN and TP were measured in COD = 50 mg/l). The pH 200
OLR (g COD/l day)

0.75
of the feed was adjusted to 6.8–7.2 by adding diluted HCl. The only
HRT (h)

supplementary nutrient, MgNO3, was added to yield a COD (N ratio 150 0.6
of 250:5) as a nitrogen supply.
0.45
100
2.3. Inoculum (seed sludge) 0.3
50
The reactor was seeded with a mixture of activated sludge from 0.15
the aerobic wastewater treatment of the Mazandaran pulp, paper
0 0
industry and a non-granular sludge obtained from a UASB reactor 1 21 41 61 81 101 121 141
operating with cheese whey wastewater from the Gela food Operation time (day)
industry of Amol, Mazandaran, Iran. The TSS of the mixture was
Influent COD HRT OLR
13 g/l. The non-granular sludge was methanogenically active as the
biogas bubbles which were apparently observed from stripping of Fig. 1. Start-up and operation scheme for UASFF reactor.
274 S.M. Emadian et al. / Ecological Engineering 82 (2015) 272–275

During the experiment, COD reduction and pH were monitored later operation 120
start-up
daily. Also, oil reduction was checked 2 times throughout the 220

Influent and Effluent COD (mg/l)

experiment, firstly after the end of the start-up period. The second 200

COD removal efficiency (%)

100
check was also after the completion of the whole experiment. 180
160 80
140
3. Results and discussion
120 60
100
3.1. Bio-reactor performance
80 40
60
3.1.1. pH 40 20
Changes in acidity (pH) of the effluent from the HUASB reactor 20
during the operation has been shown in Fig. 2. As it is shown in 0 0
Fig. 2, the pH was comparatively stable (varying from 8.04 to 8.78) 1 21 41 61 81 101 121 141
during the operation, which was suitable for efficient methano- Operating time (day)
genesis. This indicated that the system had sufficient alkalinity to
neutralize organic acids coming from the hydrolysis and fermen- Influent COD Effluent COD COD Removal
tation stage. One can see in Fig. 2 that there are numbers of Fig. 3. Bioreactor performance during operation.
decrease in pH through the operation attributed to the accumula-
tion of the produced volatile fatty acid (VFA) due to enhancement
of the influent OLR. Accumulation of VFA in the reactor did not sour decrease is attributed to the more VFA production due to the
the reactor. The similar result was reported by Van Haandel and introduction of new OLR to the reactor. Similar observation was
Lettinga (1994) which was about the treatment of domestic reported by other authors. The system recovered shortly and
wastewater. There was a sudden decrease in pH from 8.52 at day of adapted to the new condition with time. Additionally, it is clearly
96 to 8.1 at day of 111 because the effect of the nutrient was tested understood that the initial immobilization of microorganisms on
at this stage of the operation. For testing this effect, from day of the surface of the support materials had a key role in shortening
96 to day of 101, the addition of the nutrient was ceased and after the start-up procedure.
that the new nutrient, NH4Cl, was added to the reactor till day of As it was mentioned before, the effect of the nutrient on the
111. In this period of the operation, decrease in pH connected to the performance of the reactor was tested during the days of 96–111.
lower activity of the methanogenic bacteria which is responsible According to Fig. 3, the COD removal efficiency decreased from 77%
for consuming of the VFA. However, the reactor recovered itself to 42% during the days of 96–101. In this period, adding MgNO3 to
because of reintroducing of MgNO3 as the nutrient to the reactor the reactor was ceased. After that, by addition of new nutrient
and pH increased again which is indicative of the increment in (NH4Cl) to the reactor, the COD removal efficiency showed a little
methanogenic bacteria activity. increase and reached an amount of 50% at the day of 111. Therefore,
MgNO3 was reintroduced to the reactor from day of 111. By
3.2. COD removal efficiency increasing the COD influent and reintroducing MgNO3 as the
nutrient to the reactor, the COD removal efficiency raised again and
The bio-reactor performance during the operation is shown in it reached an amount of the 75% at the end of the study which was
Fig. 3. The reactor was fed with an influent COD of 50 mg/l and much higher than the result was obtained by Sun et al. (2010). They
100 mg/l during the start-up stage. As it is shown in Fig. 3, the COD reported that the COD removal efficiency of moving biofilm bio-
removal efficiency increased from 40% at the beginning of the reactor (an aerobic bio-reactor) in treating shipboard wastewater
start-up to 75% at the end of the operation implying that the sludge (including synthetic bilge water) was about 59% in HRT of 8 h. For
was acclimated appropriately to the bilge water. Each increment in improving the performance of the bio-reactor, they applied a
OLR during the operation led to a reduction in COD removal membrane as a post treatment stage with a recycle-stream to the
efficiency. A comparison between Figs. 2 and 3 shows a similar bio-reactor (Sun et al., 2010).
trend between effluent pH and COD removal efficiency which The TSS concentration of the sludge in the reactor increased
concurs with results obtained by Zhang et al. (2008). The sudden from 16.5 g/l at the beginning of the start-up to 67 g/l at the end of
the study so that the sludge acted as a filter for removing the
suspended solids from the wastewater. Subsequently, the UASB
reactor had a noticeable effect on removing the TSS content of the
9.5 0.8 wastewater.
start-up later operation
9.3 0.7
OLR (g COD/l.day)
Effluent pH

9.1
0.6
8.9
8.7 0.5
8.5 0.4
8.3 0.3
8.1
0.2
7.9
7.7 0.1
7.5 0
1 21 41 61 81 101 121 141
Operating time (day)
pH OLR

Fig. 2. Change of pH during operation. Fig. 4. Oil removal at two point of operation.
 S.M. Emadian et al. / Ecological Engineering 82 (2015) 272–275 275

3.3. Oil content References

The reduction of oil content for the wastewater at the end of APHA, 2008. Standard Methods for the examination of Water and Wastewater.
American Public Health Association/American Water Work Association/Water
start-up and operation of the reactor has been shown in Fig. 4. It Environmental Federation, Washington, DC.
can be seen in Fig. 4 that either at the end of the start-up or the DR/890, 2009. colorimeter, Procedures Manual, Method 8000. in: Hach Company L.,
operation, the oil effluent concentration was below 15 mg/l which CO, ed.
Karakulski, K., Morawski, W.A., Grzechulska, J., 1998. Purification of bilge water by
is a standard level for discharging the wastewater from ships hybrid ultrafiltration and photocatalytic processes. Sep. Purif. Technol. 14,
(MARPOL, 1973). Sun et al. (2010) obtained that the residual oil 163–173.
concentration of about 30 mg/l from the effluent of the MBBR MARPOL, 1973. International Convention for the Prevention of Pollution from Ships,
1973, as modified by the protocol of 1978 relating thereto (MARPOL 73/78). in:
reactor in HRT of 8 h which was double times higher than the (IMO) I.M.O., ed.
standard level (15 mg/l). Peng, H., Tremblay, A.Y., Veinot, D.E., 2005. The use of backflushed coalescing
microfilteration as pretreatment for the ultrafilteration of bilge water.
Desalination 181, 109–120.
4. Conclusions
Rincon, G.J., La Motta, E.J., 2014. Simultaneous removal of oil and grease, and heavy
metals from artificial bilge water using electrocoagulation/flotation. J. Environ.
In this study, anaerobic treatment of dilute bilge water was Manage. 144, 42–50.
performed by using HUASB reactor at ambient temperature. After a Sun, C., Leiknes, T., Weitzenbock, J., Thorstensen, B., 2010. Development of an
integrated shipboard wastewater treatment system using biofilm-MBR. Sep.
good resulted immobilization of sludge in the support materials and Purif. Technol. 75, 22–31.
start-up period, the COD removal efficiency and oil residual Van Haandel, A.C., Lettinga, G., 1994. Anaerobic Sewage Treatment – A Practical
concentration were promising at the end of the operation. The Guide for Regions with a Hot Climate. John Wiley and Sons, England, pp. 226.
Zaiat, M., Cabral, A.K.A., Foresti, E., 1994. Horizontal-flow anaerobic immobilized
immobilization of the biomass in the support materials had an sludge reactor for wastewater treatment: conception and performance
important role in reducing the influent COD. According to the evaluation. Revista Brasileira de Engenharia 11, 33–42.
obtained results, it can be concluded that the HUASB reactor is a Zhang, Y., Yan, L., Chi, L., Long, X., Mei, Z., Zhang, Z., 2008. Startup and operation of
anaerobic EGSB reactor treating palm oil mill effluent. J. Environ. Sci. 20,
promising option for the treatment of the low-strength bilge water, 658–663.
produced from the ships in Caspian Sea at the ambient temperatures.