Desalination 246 (2009) 389–396

Performances of an activated sludge process for the treatment

of fish processing saline wastewater

Fathi Aloui, Sonia Khoufi, Slim Loukil, Sami Sayadi*

Laboratoire des Bioprocédés, Pôle d’Excellence Régional AUF (PER-LBP),
Centre de Biotechnologie de Sfax, BP: « K », 3038, Sfax, Tunisie
Tel. / Fax 00 216 74874 452; email: [email protected]
Received 05 September 2007; revised 04 March 2008; accepted 11 March 2008

Abstract
In recent years, there has been a rapid proliferation of fish industries across the world. The conditions in the
fish industries are often squalid with the continual generation of liquid wastes. The generated wastewaters are rich
in salts (NaCl) and nitrogen ammonia. In this work an industrial wastewater containing 1 to 6% salt (NaCl) was
treated in an activated sludge unit operating in a continuous mode. Effects of important process parameters such
as organic loading and O2 rates on COD, N-NH4+ and BOD5 removal rates were investigated. Increasing the aeration
rate resulted in the improvement of the performance of the system. However, the pollution abatement rates
decreased with increasing the COD loading rate and salt content. Inhibition process was found to be significant
for salt concentrations higher than 4% NaCl. Experimental results demonstrate that the acclimatized consortium
can be efficient for the treatment of saline wastewaters, even in the presence of high salt concentrations up to 4%
and an organic loading rate reaching 855 mg COD/l.d. These experimental results may be used in the design of
activated sludge units treating saline wastewaters.
Keywords: Activated sludge; Chemical oxygen demand; Saline wastewater, Acclimatized consortium

1. Introduction ture of chemicals such as pesticides, pharmaceu-

ticals, and herbicides and during oil and gas
Several industries are dealing with high salt
recovery processes [1]. High salt concentration is
concentration in their wastewaters. Further, the
also found in the landfill leachates [2].
policy of more economic use of water and water
In recent years, there has been a rapid growth
reuse will result in an increase of salt content of
of commercial fish markets and industries across
the ultimately produced wastewater. Hypersaline
the world. The liquid wastes generated from fish
wastewaters are generated during the manufac-
cleaning and processing activities are saline in
nature. Moreover, fish processing wastewaters
*Corresponding author. are rich in nitrogen, which when discharged into
Presented at the: MEDA WATER International Conference on Sustainable Water Management, Tunis, March 21–24 2007.

0011-9164/09/$– See front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.desal.2008.03.062
390 F. Aloui et al. / Desalination 246 (2009) 389–396

the environment can lead to eutrophication and NaCl [9]. Kincannon and Gaudy [10] found that
depletion of receiving water oxygen resources by relatively more oxygen was used by cells grown
oxidation of ammonia to nitrate [3]. in the presence of high salt concentrations and
Saline effluents are conventionally treated they explained that NaCl may exert a selective
through physicochemical means, as biological pressure which fosters the predominance of
treatment is strongly inhibited by salts. However, species which have an inherently higher ratio of
the costs of physicochemical treatments being par- respiration to synthesis.
ticularly high, alternative systems for the treatment Recent studies using respirometric tests
of organic matter are increasingly the focus of showed an inhibiting effect of high salinity on
research. Most of such systems involve anaerobic respiration rates of activated sludge [11]. The val-
or aerobic biological treatment. Even though bio- ues of respiration inhibition were well correlated
logical treatment of carbonaceous, nitrogenous and to the loss in carbon removal efficiency.
phosphorous pollution has proved to be feasible at In spite of the detrimental effect of salt on
high salt concentrations, the performance obtained microbial activity, moderate acclimation of acti-
depends on a proper adaptation of the biomass or vated sludge to high salinity is possible.
the use of halophilic microorganisms. The aim of this work is the application of an
High salt content in wastewater is known to acclimatized consortium to treat saline waste-
significantly reduce the treatment efficiency of waters at different salt concentrations in an acti-
conventional activated sludge, anaerobic, nitrifi- vated sludge process.
cation and denitrification processes [4].
Conventional microbiological treatment
processes do not function at high salt concentra- 2. Materials and Methods
tions, and therefore the use of moderately
2.1. Analytical methods
halophilic bacteria should be considered [5]. Bio-
logical treatment of ammonia-rich wastewater Dry weight and moisture content were deter-
has been investigated using the activated sludge mined by weighing samples before and after dry-
process [6]. These studies showed that the acti- ing overnight at 105°C. Biomass was analysed, by
vated sludge process could be optimised for nitri- loss on ignition at 600°C for 2 h. pH was meas-
fication of high ammonia concentrations (up to ured using a pH-meter (Metrohm). The Total Kjel-
1000 mg NH4+) at long solid retention times in a dahl nitrogen content (TKN) and the ammoniacal
two-stage treatment scheme. nitrogen (N-NH4+) were analysed as Kjeldahl-N.
Biological treatment systems offer the best COD was estimated as described by Knechtel [12]
alternative to treat high strength wastewaters after a total digestion with H2SO4 and potassium
containing carbonaceous organic and nitrogen dichromate at 148°C for 2 hours. The interference
matter [7]. However, high percentages of salt of chlorides can be compensated for, at least up to
are known to compromise the correct operation 3000 mg/L, by matching the Cl− concentration of
of conventional aerobic wastewater treatment the blanks to that of the samples. Therefore, the
processes above chloride concentrations of 5–8 interference of chloride was compensated for, by
g/L.d [8]. preparing blank solutions with similar chloride
In addition to the effect of high salinity on concentration to the samples. Biological oxygen
organics removal, its impact on respiration rates demand (BOD5) was determined by the manomet-
is a well-known phenomenon. Early studies with ric method with a respirometer [BSB-Controller
Bacillus cereus indicated that the endogenous res- Model 620 T (WTW)]. Microtoxicity was carried
piration rate of the culture decreased above 1% out with Vibrio fischeri (luminescent bacteria
 F. Aloui et al. / Desalination 246 (2009) 389–396 391

LCK 480, LUMIStox Test Kit, Dr. Lange) using management strategy or treatment process. Two
LUMIStox 300 measuring instrument according samples of wastewater were taken and used for
to ISO 11438 method (1998) [13]. The inhibition this study. Table 1 illustrates the average values
of the bioluminescence (LI) of Vibrio fischeri of the various parameters determined on the
determination was determined as described by sample of wastewater used to study the aeration
Dhouib et al [14]. effect on removal efficiencies. The BOD5/COD
ratio equal to 0.47 indicates that the organic
2.2. Experimental setup and conditions matter in the sample of wastewater can be
The activated sludge process consists of an biodegradable. The nitrogen is present essen-
aerated reactor with a working volume of 6 L, and tially in ammonia form and exceeds the Tunisian
a static decanter. Air was supplied using a dif- standards of rejection in public canalisations
fuser. The wastewater was supplied to the reactor (100 mg/L). However, a pH around 6.85 is
by means of a peristaltic pump. The decanter was favourable for biological treatment.
used to separate the activated sludge and the Acclimation implies the exposure of non-
treated wastewater. While a part of the sludge is salt-adapted microorganisms to increasing salt
removed as so called excess or surplus sludge, the concentrations in order to achieve satisfactory
rest is sent back in the reactor in order to increase effluent treatment performance at a given salt
the sludge concentration. The clarified effluent concentration. At the start-up of the acclimatiz-
was then collected and analysed. ing phase, the pH values increased, exceeding
8.5. For each step of COD increase, the pH val-
2.3. Consortium origin ues presented some fluctuations before their sta-
bilisation around 8 indicating that the consortia
An activated sludge, from Sfax wastewaters
was adapted to the progressive COD increase.
treatment plant, was acclimatised to saline waste-
The pH of the wastewater was not adjusted dur-
waters containing 2.5% salt concentration. To
ing treatment.
adapt the consortia to the effluent, introduced
COD, N-NH4+ and BOD5 removal efficiencies
COD values were increased progressively. The
for an acclimatized sludge at different salt concen-
batch was firstly fed with low COD starting-up
trations, organic loading rate (OLR) and aeration
from 0.4 g COD and reaching 3.2 g COD per litre
ratio were studied. Sludge retention time (SRT)
daily, this lasted for 4 months. The sludge reten-
was kept constant at 10 days. Table 1 shows the
tion time (SRT) of the system was kept constant
positive effect of the aeration on microbial growth.
about 10 days.
In fact, at 1 v.v.m, higher biomass concentration
2.4. Wastewater sampling was attained. Also, an increasing aeration ratio
improves the COD and BOD5 removals.
The effluent used in this work is an industrial Results concerning the aeration ratio effect on
wastewater generated from SOCEPA fish indus- COD removal are presented in Fig.1. For each
try located in Sfax (Tunisia). Different quantities aeration ratio, a control test was realised in the
of NaCl were added to this effluent (containing same conditions without biomass. These test
originally 10 g/L of NaCl) in order to study the experiences, show a slight difference on removal
salinity effects on the treatment efficiency. of COD and BOD5, thus proving the insignificant
effect of volatilization on these parameters. How-
3. Results and discussion
ever, N-NH4+ removals are proportional to aera-
Wastewater characterization is a critical fac- tion ratio. In fact, volatilisation can be a factor
tor in establishing a corresponding effective causing N-NH4+ removal.
392 F. Aloui et al. / Desalination 246 (2009) 389–396

Table 1
Saline wastewater composition before and after treatment in a continuous activated sludge process (18th day) at 0.5 and
1 volume of air/ volume of reactor/ minute (v.v.m) (salt concentration = 2.5%; HRT = 1 day)

COD BOD5 TN N-NH4+ Biomass

Saline wastewater pH (mg/L) (mg/L) (mg/L) (mg/L) (g/L)
Before treatment 6.85 3400 ± 300 1600 ±200 335 ± 40 298 ± 30
After treatment
Aeration ratio = 0.5 v.v.m 8.50 740 ± 60 400 ±50 140 ± 20 111 ±10 3.8 ± 0.4
Aeration ratio = 1 v.v.m 8.60 490 ± 40 50 ± 8 129 ±20 92 ± 10 5.6 ± 0.4

It was observed that the increase in the aera- tions in BOD removal. Ludzack and Noran [8]
tion ratio resulted in the improvement of the COD investigated the effects of salt concentrations up
removal. Moreover, BOD5 and N-NH4+ concen- to 20 g/L on the activated sludge process. Salt
trations in the treated effluent were strongly concentrations above 1% resulted in low BOD5
reduced at 1 v.v.m (Table 1). removal efficiencies [8].
Fig.1 shows that COD values in the treated A response to rapid changes for the salinity is
wastewater were clearly reduced at 1 v.v.m. the release of cellular material, resulting in an
The effects of salt content on the abatement increase of soluble COD [15]. Although the adap-
rates of COD, N-NH4+ and BOD5 after 24 days of tation of activated sludge has already proved to be
treatment in continuous mode were investigated. possible, a major bottleneck is that the proper per-
In this case an industrial wastewater having a formance of such salt-adapted systems is usually
concentration equal to 1% salinity was used. Dif- limited to less than 5% salt [16]. Consequently the
ferent quantities of NaCl were added to this use of specialised organisms (i.e. halophiles) usu-
wastewater in order to attain 2%, 4% and 6% ally remains the best way to enhance the biologi-
salinity concentrations. Fig. 2 shows that COD cal treatment of saline wastewater. This is in line
and BOD5 of the treated wastewater increase with with Uygur and Kargi [17], They observed a drop
increasing salt content. Severe changes in salinity of COD removal efficiency from 90% to 32%,
combined with high organic and hydraulic load- when salinity increased from 0% to 6% in a SBR
ing rates were reported to cause temporary reduc- used to purify a synthetic saline effluent.

1200
0.5 v.v.m 1 v.vm
1000
COD(mg /l)

800

600

400

200

0
0 5 10 15 20
Days

Fig. 1. Effect of the aeration rate on COD of the activated sludge outlet during the treatment of a saline wastewater in
a continuous activated sludge process (HRT = 1 day, salt concentration = 2.5%, CODi = 3400 ± 300 mg/L). (■): 0.5 v.v.m;
(▲): 1 v.v.m.
 F. Aloui et al. / Desalination 246 (2009) 389–396 393

1000 500
a 1% 2%
b
4% 6%
800 400 1% 2%
4% 6%

BOD5 (mg/l)
COD (mg/l)

600 300

400 200

200 100

0 0
0 5 10 15 20 25 0 5 10 15 20 25
Days Days

Fig. 2. Effect of salt concentration on COD (a), and BOD5 (b) of the activated sludge outlet during the treatment of a
saline wastewater in activated sludge process (aeration ratio = 1 v.v.m, CODi = 5300± 600 mg/L, HRT= 3 days). □: 1%;
◊: 2%; ▲: 4%; X: 6%.

It was further observed from Fig. 2a that at salt high salt concentrations were significantly allevi-
concentrations below 40 g NaCl/l, the bacterial ated using salt-tolerant organism (Halobacter)
system displayed improved COD removal effi- supplemented activated sludge culture.
ciencies higher than 97%. The BOD5 was also NH4+-N values in the influent and the effluent
clearly reduced at salt concentrations below 4% show that the concentration of ammoniacal nitro-
(Fig. 2b). This may be an indication that once the gen in the saline wastewater can be strongly
inhibitory effect of salt was reduced, the bacterial reduced essentially at salt concentrations under
system regained its efficiency. At 6% salt concen- 40 g/L (Fig. 3a). High effluent ammonia level
tration, the COD and BOD5 of the treated effluent was obtained at 60 g/L salt content. This is in line
were comparatively high than to salt concentra- with Panswad and Anan [21] findings who
tions below 4%. In fact, the presence of high salt reported that nitrification is the most sensitive
concentrations in wastewaters induces salt stress process to saline concentrations (55% inhibition
to the microbial species, resulting in the inhibi- at 30 g NaCl /l).
tion of many enzymes, decreased cell activity and High saline concentrations have negative
eventually leads to plasmolysis [18]. effects on organic matter and nitrogen removals
Saline wastewater biological treatment sys- as reported by Intrasungkha et al [22]. The activ-
tems usually result in low BOD removal because ity of autotrophic organisms responsible for nitri-
of the adverse effects of salt on the microbial flora. fication was determined in several batch tests and
High salt concentrations cause plasmolysis or loss was found to be about 1/2 or 1/3 of the degree of
of activity of cells. The application of salt-tolerant activity observed in fresh water [23]. Also, Hunik
bacteria for biological treatment of wastewater has et al (1993) [24] used batch assays to study the
been found to improve treatment efficiency under effect of different salts on ammonia and nitrite
such conditions. For example, the use of oxidation and found that ammonia oxidation is
halophilic microorganisms (e.g. Halobacter halo- sensitive to saline effect.
bium) in an activated sludge process resulted in A total detoxification of the effluent was
better treatment performances at salt contents achieved after treatment at salt concentrations
above 2% [19]. Fed-batch treatment of saline under 2% (Fig. 3b). At 6% of salinity, the lumi-
wastewater has been investigated by Kargi and nescent inhibition (LI) of the tested bacteria is
Dinçer [20] in an aerated tank. Adverse effects of about 14%.
394 F. Aloui et al. / Desalination 246 (2009) 389–396

b 100 Initial toxicity

a
500 90 Residual toxicity
Initial N-NH4+ Residual N-NH4+
80
400
70
N-NH4+ (mg/l)

60

LI (%)
300
50
40
200
30
100 20
10
0 0
1% 2% 4% 6% 1% 2% 4% 6%
Salinity concentration Salinity concentration

Fig. 3. Effect of salt concentration on N-NH4+ (a) and luminescence inhibition (LI), (b) of the activated sludge outlet
during the treatment of a saline wastewater in an activated sludge process (aeration rate = 1 v.v.m, CODi = 5300 ± 600
mg/L, HRT = 3 days). □: Initial N-NH4+ and toxicity; ▄: Residual N-NH4+ and toxicity.

The effects of increasing OLR on COD, It was observed that OLR values higher than
BOD5, N-NH4+ and toxicity removals after 18 1140 mg/L.d affected the COD (Fig. 4a), BOD5
days of treatment in continuous mode were inves- (Fig. 4b) and N-NH4+ (Fig. 5a) removals. COD,
tigated. For this, an industrial wastewater with BOD5 and N-NH4+ in the treated effluent were
2.5% of salinity was used. Different OLR were clearly reduced at OLR of 570 and 855 mg/L.d.
adjusted with the same sample of wastewater by High organic and hydraulic loadings were
the decrease of the Hydraulic Retention Time. reported to cause temporary reductions in BOD5
COD, N-NH4+ and BOD5 effluent concentrations removal efficiency [8].
increased with increasing OLR, resulting in high A total detoxification of the effluent was
effluent nutrient levels, essentially at OLR of achieved after treatment at OLR under 855 mg/L.d
1710 mg/L.d (Fig. 4). (Fig. 5b). At an OLR of 1710 mg/L.d, the LI of the

a 600 b
1500 570 mg/L.d 855 mg/L.d
1140 mg/L.d 1710 mg/L.d 500
1200
400
BOD5 (mg/l)
COD (mg/l)

900
300
600 200
570 mg/L.d 855 mg/L.d
300 100 1140 mg/L.d 1710 mg/L.d

0 0
0 3 6 9 12 15 18 0 3 6 9 12 15 18
Days Days

Fig. 4. Effect of OLR on COD (a), and BOD5 (b) of the activated sludge outlet during the treatment of a saline wastewater
in activated sludge process (aeration ratio = 1 v.v.m, salt concentration = 2.5%, CODi = 3400 ± 300 mg/L). ●: OLR =
570 mg/L.d; □: OLR = 855 mg/L.d; ▲: OLR = 1140 mg/ l.d; X: OLR = 1710 mg/L.d.
 F. Aloui et al. / Desalination 246 (2009) 389–396 395

500 100 b
a
Initial N-NH4+ Residual N-NH4+
Initial Itoxicity Residual toxicity
400
75
N-NH4+ (mg/l)

LI (%)
300
50
200

25
100

0 0
570 855 1140 1710 570 855 1140 1710
OLR (mg/l.d) OLR (mg/l.d)

Fig. 5. Effect of the OLR on N-NH4+ (a) and luminescence inhibition (LI) (b), of the activated sludge outlet during the
treatment of a saline wastewater in activated sludge process (aeration ratio = 1 v.v.m, salt concentration = 2.5%, CODi
= 3400 ± 300 mg/L). □: Initial N-NH4+ and toxicity; ▄: Residual N-NH4+ and toxicity.

tested bacteria is about 33%. Considering that the achieved with salt concentrations below 2% and
toxicity was measured using the marine lumines- OLR under 855 mg /l.d.
cent bacteria Vibrio fischeri, which tolerates the
used salt concentrations. It can be concluded that
Acknowledgment
the LI can be due to the residual ammonia nitrogen
in the treated effluent (180 mg/L in the treated This research was supported by PRF “eau”
effluent at an OLR of 1710 mg /l.d). project (2004–2007), financed by the MESRST.

4. Conclusions References
An activated sludge process was used, in order [1] M. Henze, P. Harremoës, J. Jansen and E. Arvin,
to treat saline wastewaters at different salt con- Wastewater treatment — Biological and Chemical
Processes. Springer-Verlag, Berlin, Heidelberg,
centrations and OLR. Experimental results
New York (1995).
demonstrate that the acclimatized culture is effi- [2] M. Ellouze, F. Aloui and S. Sayadi, Detoxification
cient for the treatment of saline wastewaters, even of Tunisian landfill leachates by selected fungi, J.
in the presence of salt concentrations reaching Hazard. Mater, 150 (2008) 642–648.
4% and OLR under 855 mg /l.d. [3] J.L. Campos, A. Mosquera-Corral, M. Sanchez, R.
At a salt concentration of 6%, high effluent Mendez and J.M. Lema, Nitrification in saline waste-
water with high ammonia concentration in an acti-
ammonia, COD and BOD5 levels were reached. vated sludge unit, Water Res., 36 (2002), 2555–2560.
Results showed that an OLR higher than 1140 mg [4] F. Kargi and A. Uygur, Biological treatment of
/l.d affected the COD, BOD5 and N-NH4+ saline wastewater in an aerated percolator unit uti-
removals. lizing halophilic bacteria, Environ. Technol., 17
During this activated sludge process, toxicity (1996) 320–325.
[5] A. Oren, P. Gurevich, M. Azachi, and Y. Henis,
was considerably reduced. The percentage of LI
Microbial degradation of pollutants at high salt con-
of V. fischeri dropped down under 15% after centrations, Biodegradation, 3 (1992) 387–398.
treatment at 1 to 6% salt concentrations. A total [6] LT. Thiem and E.A. Alkhatib, In situ adaptation of
detoxification of the saline wastewater was activated sludge by shock loading to enhance treat-
396 F. Aloui et al. / Desalination 246 (2009) 389–396

ment of high ammonia content in petrochemical [16] A.R. Dinçer and F. Kargi, Performance of rotating
wastewater, J. Water Pollut. Control Fed., 60 (1988) biological disc system treating saline wastewater,
1245–1252. Process Biochem., 36 (8–9) (2001) 901–906.
[7] F. Kargi and A.R. Dincer, Salt inhibition effects in [17] A. Uygur and F. Kargi, Salt inhibition on biological
biological treatment of saline wastewater in RBC, J. nutrient removal from saline wastewater in a
Environ. Eng., 125 (1999) 966–971. sequencing batch reactor, Enzyme Microb. Tech-
[8] F.J. Ludzack and D.K. Noran, Tolerance of high nol., 34 (3–4) (2004) 313–318.
salinities by conventional wastewater treatment [18] A. Uygur, Specific nutrient removal rates in saline
processes. J. Water Pollut. Control Fed., 37 (1965) wastewater treatment using sequencing batch reac-
1404–1416. tor, Proc. Biochem., 41 (2006) 61–66.
[9] M. Ingram, The influence of sodium chloride and [19] F. Kargi and A.R. Dinçer, Use of halophilic bacteria
temperature on the endogenous respiration of Bacil- in biological treatment of saline wastewater by fed-
lus cereus, J. Gen. Physiol., 23 (1940) 773–778. batch operation, Water Environ. Res., 72 (2) (2000)
[10] D.F. Kincannon and A.F. Gaudy, Some effects of 170–174.
high salt concentrations on activated sludge, J. Water [20] F. Kargi and A.R. Dinçer, Effect of salt concentra-
Pollut. Control Fed., 38 (7) (1966) 1148–1159. tion on biological treatment of saline wastewater by
[11] M. Pernetti and L. Di Palma, Experimental evalua- fed-batch operation, Enzyme Microb. Technol., 19
tion of inhibition effects of saline wastewater on (7) (1996) 529–537.
activated sludge. Environ. Technol., 26 (6) (2005) [21] T. Panswad and C. Anan (1999), Specific oxygen,
695–703. ammonia and nitrate uptake rates of a biological
[12] R.J. Knechtel, A more economical method for the nutrient removal process treating elevated salinity
determination of chemical oxygen demand, Water wastewater, Bioresource Technol., 70 (1999)
Pollut. Control, (1978), 25–29. 237–43.
[13] ISO 11348-2 (1998), Water quality — determination [22] N. Intrasungkha, J. Keller and L. Blackall, Biolog-
of the inhibitory effect of water samples on the light ical nutrient removal efficiency in treatment of
emission of Vibrio fischeri (Luminescent bacteria saline wastewater, Water Technol., 39(6) (1999)
test), Part 2, Method using liquid-dried bacteria. 183–90.
[14] A. Dhouib, M. Ellouze, F. Aloui, and S. Sayadi, [23] S.L. Tang and T.H. Lee, Treatment of mixed (fresh
Effect of bioaugmentation of activated sludge with and salt) wastewater, 28th WEDC Conference, Sus-
white-rot fungi on olive mill wastewater detoxifi- tainable environmental and water services, Kolkata,
cation, Lett. Appl. Microbiol., 42 (2006) 405–411. India (2002).
[15] D.F. Kincannon and A.F. Gaudy, Response of bio- [24] J.H. Hunik, H.J.G. Meijer and J. Tramper, Kinetics
logical waste treatment systems to changes in salt of Nitrobacter agilis at extreme substrate, product
concentrations, Biotechnol. Bioeng., 10 (1968) and salt concentrations, Appl. Microbiol. Biotech-
483–496. nol., 40 (1993).