See

discussions, stats, and author profiles for this publication at:

https://www.researchgate.net/publication/228561805

Application of reverse osmosis to remove

aniline from wastewater

Article in Desalination · September 2009

Impact Factor: 3.76 · DOI: 10.1016/j.desal.2009.02.038

CITATIONS READS

21 86

6 authors, including:

Asuncion Maria Hidalgo Maria D. Murcia

University of Murcia University of Murcia
46 PUBLICATIONS 435 CITATIONS 65 PUBLICATIONS 456 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Maria D. Murcia

Retrieved on: 12 May 2016
 Desalination 246 (2009) 314–320

Application of reverse osmosis to remove aniline

from wastewater

J.L. Gómeza, G. Leónb, A.M. Hidalgoa,*, M. Gómeza, M.D. Murciaa,

G. Griñána
Departamento de Ingeniería Química, Grupo de Análisis y Simulación de Procesos Químicos,
a

Bioquímicos y de Membrana, Universidad de Murcia, Campus de Espinardo, 30071 Murcia, Spain

Tel. +34-968-367355; Fax +34-968-364148; email: ahidalgo@um.es
bDepartamento de Ingeniería Química y Ambiental, Universidad Politécnica de Cartagena, Cartagena, Spain

Received 08 July 2008; revised 30 December 2008; accepted 09 February 2009

Abstract
The presence of organic toxic solutes in industrial wastewater is a common environmental problem. Aniline is
known to be a harmful and persistent pollutant and its presence in wastewater requires treatment before disposal.
The performance of reverse osmosis to remove aniline from aqueous solutions is studied in this paper. The study
has been carried out in a flat membrane test module using three thin layer composite membranes, two of polyamide,
HR98PP and SEPA-MS05, and one of polyether sulphone, DESAL-3B. Recycling of both concentrate and perme-
ate has been carried out in order to keep the feed concentration practically constant and so simulate a continuous
process in a quasi-stationary state. The influence of different operational variables (pressure, feed volumetric flow
rate, feed concentration and pH) on the performance of the aniline removal process is analyzed.
Keywords: Membrane processes; Aniline; Reverse osmosis; Wastewater

1. Introduction as a solvent in perfumes, varnish and resins

[1,2]. Aniline is released to the environment
Aniline is widely used as raw material in
directly in industrial wastewater and indirectly
many industrial processes including the manu-
through the degradation of some the above-
facture of dyes and pigments, herbicides and
mentioned organic compound (herbicides, pes-
pesticides, pharmaceuticals and explosives, and
ticides, dyes, etc.) [3,4].
Great care should be taken concerning the
*Corresponding author. contamination of groundwater because aniline is

Presented at the conference Engineering with Membranes 2008; Membrane Processes: Development, Monitoring and
Modelling – From the Nano to the Macro Scale – (EWM 2008), May 25–28, 2008, Vale do Lobo, Algarve, Portugal.

0011-9164/09/$– See front matter © 2009 Elsevier B.V. All rights reserved.
doi: 10.1016/j.desal.0000.00.000
 J.L. Gómez et al. / Desalination 246 (2009) 314–320 315

known to be a toxic and persistent pollutant that The flow or permeation rate, J, is defined as
is harmful not only to aquatic life but also to the volume flowing through the membrane per
humans [5,6]. Indeed, aniline is toxic through unit area and time.
ingestion, inhalation and contact with the skin. The solution–diffusion model [19] assumes
The short-term effects of aniline in humans are that both the solute and the solvent dissolve in the
mainly connected with the lung, and include non-porous homogeneous surface layers of the
upper respiratory tract irritation and congestion. membranes and each diffusing across it in an
Repeated exposure may have effects on the liver, uncoupled manner due to its chemical potential
kidneys, blood (methaemoglobinaemia, resulting gradient, which is result of concentration and
in cyanosis) and spleen. It goes without saying, pressure differences across the membrane. The
then, that industrial wastewater containing signif- effect of concentration polarization and fouling
icant levels of aniline should be treated to avoid are not considered in this study because model
pollution. dilute feed solutions and high feed velocities
Several processes to remove aniline from were used to minimize deviations from ideal
wastewater have been described, including mass transfer.
biodegradation [7,8], adsorption [9,10], oxidation The solvent flux depends on the hydraulic
[11,12] and different membrane processes such pressure applied across the membrane, ΔP, minus
as pervaporation [13], liquid membranes [14,15], the difference in the osmotic pressures of the
nanofiltration [16] and reverse osmosis [17]. solutions on the feed and permeate side of the
In this paper, aniline removal from aqueous membrane, Δπ
solutions by reverse osmosis using different
Jw = Aw (ΔP – Δπ) (2)
membranes and different operational variables
(pressure, feed volumetric flow rate, feed concen- where Aw is the water permeability constant, which
tration and pH) is studied. depends on the structure of the membrane, ΔP is
the membrane pressure gradient and π is the
osmotic pressure. The solute flux depends on the
2. Theory solute concentration gradient across the membrane
The performance of a given membrane Js = Bs (Cs – Cp) (3)
process is determined by two parameters, the
where Bs is the solute permeability constant,
selectivity and the flow through it [18]. For dilute
which is a function of the solute composition and
aqueous mixtures consisting of water and a
the membrane structure, with the following value:
solute, the selectivity of a membrane towards the
mixture is usually expressed in terms of the solute
Ds K s
rejection coefficient. This parameter, R, is a Bs = (4)
measure of the ability of the membrane to sepa- l
rate the solute from the feed solution, and is
where Ds being the solute diffusion coefficient, Ks
defined, as a percentage, by the equation
is the solute distribution coefficient and l is the
membrane thickness. Expressing permeate con-
Cf − Cp ⎛ Cp ⎞ centration as Cp = Js/Jw [18] and combining Eqs.
R = 100 × = 100 × ⎜1 − ⎟ (1)
Cf ⎝ Cf ⎠ (2)–(4), the rejection coefficient can be written as:

Aw ( ΔP − Δπ )
where Cf and Cp are the solute concentration in R= (5)
the feed and in the permeate, respectively. Aw ( ΔP − Δπ ) + Bs
316 J.L. Gómez et al. / Desalination 246 (2009) 314–320

with a high selectivity towards salts, which can

be used in a relatively wide range of tempera-
FI tures, pressures and pH values. The characteris-
A tics of the membranes are described in Table 1.
Typical experimental conditions were operat-
PI
ing pressure of 40 × 105 N/m2, feed aniline con-
B centration of 0.1 kg/m3, feed volumetric flow
TI rate of 2.78 × 10–5 m3/s, pH=7 and temperature
of 25ºC.
C
To study the influence of the different opera-
tional variables on the performance of the aniline
Fig. 1. Flow diagram of reverse osmosis test unit flat
removal process, the following experimental
membrane module: (A) feed tank, (B) membrane module,
(C) high pressure pump.
series were carried out: operational pressure vari-
ation (30 × 105, 35 × 105, 40 × 105, and 45 × 105
N/m2), feed volumetric flow rate variation (2.78
3. Experimental equipment and procedure
× 10–5, 4.17 × 10–5 and 5.56 × 10–5 m3/s) feed ani-
Experimental tests were performed in an line concentration variation (0.02, 0.05, 0.1 and
INDEVEN flat membrane test module, which 0.2 kg/m3) and pH variation (6, 7, 8, 9 and 10).
consists of a unit that provides data on the behav- The aniline concentrations in feed and perme-
iour of the membranes in cross-flow conditions ate solutions were determined spectrophotomet-
with a reduced surface area, low feed and short rically at 280 nm, after dilution with 1 M NaOH,
times. Aniline aqueous solutions were treated in using an UV spectrophotometer Shimazdu UV-
the test module, recycling both concentrate and 160A.
permeate in order to keep the feed concentration
practically constant and to simulate a continuous
4. Results and discussion
process in a quasi-stationary state (Fig.1).
Three membranes were used, HR98PP from The influence of the different operational vari-
Dow/Filmtec, SEPA-MS05 from Osmonics and ables on aniline rejection is shown in Fig. 2. The
DESAL-3B from Desalination Systems. Those rejection percentage slightly increases with pres-
membranes are thin layer composite membranes, sure for all three tested membranes, the highest

Table 1
Main characteristics of the membranes used in the experimental test module

Membrane
Manufacturer Dow/Filmtec Osmonics Inc. Desalination Systems Inc.
Product denomination HR98PP SEPA MS05 DESAL-3B
Type Thin film composite Thin film composite Thin film composite
Composition Polyamide Polyamide Polyether-sulphone
Effective membrane surface area (m2) 0.003 0.003 0.003
Maximum pressure (N/m2) 60 × 105 70 × 105 45 × 105
Maximum temperature (ºC) 60 80 50
NaCl rejection >97.5 >98 >98.5
pH range 2–11 3–11 4–11
Chlorine tolerance Low Low Low
 J.L. Gómez et al. / Desalination 246 (2009) 314–320 317

100 100

90 90

80 80
R (%)

R (%)
70 70
HR98PP MS05 DESAL-3B HR98PP MS05 DESAL-3B
60 60
(a) (b)
50 50
25 30 35 40 45 50 0 0.05 0.1 0.15 0.2
Pressure × 10−5 (N/m2) Cf (kg/m3)

100 100

90 90

80 80
R (%)

R (%)

70 70
HR98PP MS05 DESAL-3B HR98PP MS05 DESAL-3B
60 60
(c) (d)
50 50
5 6 7 8 9 10 11 0 2 4 6 8
pH Volumetric feed flow rate (m3/s)

Fig. 2. Influence of different operating conditions in rejection percentages: (a) pressure, (b) feed aniline concentration,
(c) pH, (d) volumetric feed flow rate.

rejections being obtained with HR98PP mem- followed by a slight decrease at pH values higher
brane (91.8%) and the poorest with MS05 mem- that 7, is observed for the HR98PP and for MS05
brane (79.0%) (Fig. 2a). These results agree with membranes, while no significant variations is
Eq. (5), where ΔP is the only variable, assuming observed for the DESAL-3B membranes.
that the constants Aw and Bs are independent of Rejection changes with pH are presumably
pressure. So, an increase in ΔP leads to an related to the presence of ionizable groups in
increase in R. In the same way, an increase in feed the membrane structure and to the net charge of
aniline concentration produces slight increments the aniline molecule as a result of its dissocia-
in aniline rejection in the three tested membranes tion equilibrium [20]. Polyamide membranes
(Fig. 2b). When the feed concentration increases, have free carboxylic acids in their structure,
the permeation concentration increases, but as the which become negatively charged at pH values
increase of permeate concentration is lower than in the order of 5. This means that in the exper-
the increase in feed concentration, rejection imental pH range the membrane surface has
increases according to Eq. (1). negative charge. On the other hand, the aniline
Variations in rejection at different pH values are pKa is 4.6 and so, at pH values higher than 4.6,
not very important (Fig. 2c) in the experimental the anilinium proportion will decrease because
range of pH used in this work (the surroundings of of the formation of neutral aniline.
the typical values of aqueous aniline solutions pH). The initial slight increase in rejection between
A slight increase of rejection between pH 6 and 7, pH 6 and 7 could be related with the retention of
318 J.L. Gómez et al. / Desalination 246 (2009) 314–320

5 5

4 4
J × 10−5 (m/s)

J × 10−5 (m/s)
3 3

2 2

1 HR98PP MS05 DESAL-3B

1
(a) HR98PP MS05 DESAL-3B (b)
0 0
25 30 35 40 45 50 0 0.05 0.1 0.15 0.2
Pressure × 10−5 (N/m2) Cf (kg/m3)

5 5

4 4
J × 10−5 (m/s)

J × 10−5 (m/s)
3 3

2 2

1 HR98PP MS05 DESAL-3B 1 HR98PP MS05 DESAL-3B

(c) (d)
0 0
5 6 7 8 9 10 11 0 2 4 6 8
pH Volumetric feed flow rate (m3/s)

Fig. 3. Influence of different operating conditions on permeation rate: (a) pressure, (b) feed aniline concentration, (c)
feed pH, (d) volumetric feed flow rate.

the remaining anilinium cations by the negative Permeation rate increases with operation pres-
carboxylate groups of the membrane. At pH val- sure, this increase being higher with HR98PP and
ues higher than 7, rejection decreases because the MS05 membranes than with DESAL-3B mem-
proportion of anilinium cations decreases signif- brane (Fig. 3a). According to Eq. (1) Jw increases
icantly at a higher pH, and neutral aniline is not with operation pressure, but Js is not affected and
so retained by the negative charge of the mem- is only determined by the concentration differ-
brane. At a pH higher that 8, no variations in ence across the membrane. So, a permeation rate
rejection are observed with pH. increase is only due to water flux increase. The
Since the DESAL-3B membrane does not lower permeation rate increase for DESAL-3B
possess these ionizable groups, no significant membrane would be related to its lower water
variations in rejection with pH are observed. permeability.
The increase of volumetric feed flow rate No significant influence of aniline feed concen-
increases the rejection in the case of the HR98PP tration on permeation rate is observed (Fig. 3b).
and DESAL-3B membranes and decreases the As mentioned above, when feed concentration
rejection when MS05 is used. increases, the permeate concentration increases,
The influence of the different operational vari- but the increase of permeate concentration is lower
ables on permeation rate is shown in Fig. 3. that the increase of feed concentration. So, Jw
Polyamide membranes (HR98PP and MS05) should decrease, as a consequence of the increase
show higher permeation rates than polyether sul- in Δπ, and Js should increase as a consequence
phone membrane (DESAL-3B) in the whole of the ΔC (Cf – Cp) increase. No influence of
range of conditions studied. pH on the permeation rate is observed (Fig. 3c).
 J.L. Gómez et al. / Desalination 246 (2009) 314–320 319

This agrees with other results described in the [2] M.E. Essington, Adsorption of aniline and tolu-
bibliography [21]. idines on montmorillonite, Soil Sci., 158(3) (1994)
181–188.
Finally, the permeation rate is not affected by
[3] R.D. Voyksner, R. Straub, J.T. Keever, H.S.
the volumetric feed flow rate in the case of MS05 Freeman and W.N. Hsu, Determination of aromatic
and DESAL-3B membranes, but decreases with amines originating from azo dyes by chemical
the HR98PP membrane. reduction combined with liquid
chromatography/mass spectrometry, Environ. Sci.
Technol., 27 (8) (1993) 1665–1672.
5. Conclusions [4] S. Laha and R.G. Luthy, Oxidation of aniline and
other primary aromatic amines by manganese
The performance of reverse osmosis to dioxide, Environ. Sci. Technol., 24 (3) (1990)
remove aniline from aqueous solutions is studied 363–373.
in this paper. Three thin layer composite mem- [5] U.S. Environmental Protection Agency, Health and
branes, two of polyamide, HR98PP and SEPA- Environmental Effects Profile for Aniline, Envi-
MS05, and one of polyether sulphone, ronmental Criteria and Assessment Office, Office
of Health and Environmental Assessment, Office
DESAL-3B, has been used. The influence of
of Research and Development, Cincinnati, OH,
operational variables such as pressure, feed vol- 1985.
umetric flow rate, feed concentration and pH on [6] U.S. Department of Health and Human Services,
the rejection and permeate flow rate has been Hazardous Substances Data Bank (HSDB, online
analyzed. The highest rejections are obtained database), National Toxicology Information Pro-
with HR98PP membrane (91.8%) and the lowest gram, National Library of Medicine, Bethesda,
MD, 1993.
rejections with MS05 membrane (79.0%). Ani- [7] S.H. Gheewala and A.P. Annachhatre, Biodegrada-
line rejection slightly increases with pressure and tion of aniline, Water Sci. Technol., 36 (1997)
feed aniline concentration for the three tested 53–63.
membranes. The observed changes in aniline [8] F.J. O'neill, K.C.A. Bromley-Challenor, R.J.
rejection with pH are related to the charge of ion- Greenwood and J.S. Knapp, Bacterial growth on
aniline: Implications for the bio-treatment of indus-
izables groups in the membrane structure and to
trial wastewater, Water Res., 34 (2000) 4397–4409.
the net charge of aniline molecule as a result of its [9] J. Niu and B.E. Conway, Adsorptive and electroad-
dissociation equilibrium. Permeation rate sorptive removal of aniline and bypyridyls from
increases with operation pressure, but no signifi- wastewaters, J. Electroanal. Chem., 536 (2002)
cant variations with feed aniline concentration 83–92.
and pH are observed. No discernable trend of [10] X. Gu, J. Zhou, A. Zhang, P. Wang, M. Xiao and
G. Liu, Feasibility study of the treatment of aniline
feed volumetric flow rate on performance is hypersaline wastewater with a combined adsorp-
obtained for all three membranes tested. tion/bioregeneration system, Desalination, 227
(2008) 139–149.
[11] G. Deiber, J.N. Foussard and H. Debellefontaine,
Acknowledgement Removal of nitrogenous compounds by catalytic
M.D. Murcia and M. Gómez are beneficiary wet air oxidation, Kinetic study, Environ. Pollut.,
96 (1997) 311–319.
of a pre- and postdoctoral scholarship, respec-
[12] L. Sánchez, J. Peral and X. Domenech, Aniline
tively, from Fundación Séneca of Murcia. degradation by combined photocatalysis and ozona-
tion, Appl. Catal., B 19 (1998) 59–65.
[13] C.C. Pereira, A.C. Habert, R. Nobrega and C.P.
References
Borges, New insights in the removal of diluted
[1] W. Gerhatz (Ed), Ullman's Encyclopaedia of Indus- volatile organic compounds from dilute aqueous
trial Chemistry, vol. 2, 5th ed., VCH, Weinheim, solution by pervaporation process, J. Membr. Sci.,
1985. 138 (1998) 227–235.
320 J.L. Gómez et al. / Desalination 246 (2009) 314–320

[14] S. Datta, P.K. Bhattacharya and N. Verma, Removal apparatus, Russ. J. Appl. Chem., 78(7) (2005)
of aniline from aqueous solutions in a mixed flow 1096–1100.
reactor using emulsion liquid membrane, J. Membr. [18] M. Mulder (Ed.), Basic Principles of Membrane
Sci., 226 (2003) 185–201. Technology, Kluwer Academic Publishers, Dor-
[15] J. Swai, N. Ito, T. Minami and K. Kikuchi, Separa- drecht, 1992.
tion of low volatile organic compounds, phenol and [19] J.G. Wijmams and R.W. Baker, The solution–diffu-
aniline derivatives, from aqueous solutions using sion model: a review, J. Membr. Sci., 107 (1995)
silicone rubber membrane, J. Membr. Sci., 252 1–21.
(2005) 1–7. [20] A. Kulkarni, D. Mukherjee and W.N. Gill, Flux
[16] C. Causserand, P. Aimar, J.P. Cravedi and enhancement by hydrophilization of thin film com-
E. Singlande, Dichloroaniline retention by posite reverse osmosis membrane, J. Membr. Sci.,
nanofiltration membranes, Water Res., 39 (2005) 114 (1996) 39–50.
1594–1600. [21] J.W. Lee, T.O. Kwon and I.S. Moon, Performance
[17] V.L. Golovashin, S.I. Lazarev and M. Mamantov, of polyamide reverse osmosis membranes for
Kinetic characteristics of reverse-osmosis separa- steel wastewater reuse, Desalination, 189 (2006)
tion of an aqueous solution of aniline in a flat-frame 309–322.