Process Biochemistry 48 (2013) 941944

Contents lists available at SciVerse ScienceDirect

Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio

Short communication

Dairy wastewater polluting load and treatment performances of an industrial

three-cascade-reactor plant
Carlo Tocchi a,b , Ermanno Federici c , Silvia Scargetta c , Alessandro DAnnibale a , Maurizio Petruccioli a,
a
b
c

Department for Innovation in Biological, Agro-Food and Forest Systems (DIBAF), University of Tuscia, Via San Camillo De Lellis s.n.c., 01100 Viterbo, Italy
Manzi Aurelio s.r.l., Via Cassia Km 94.100, 01027 Monteascone, VT, Italy
Department of Cellular and Environmental Biology, University of Perugia, Via del Giochetto, 06122 Perugia, Italy

a r t i c l e

i n f o

Article history:
Received 5 February 2013
Received in revised form 20 March 2013
Accepted 14 April 2013
Available online 22 April 2013
Keywords:
Dairy wastewater
Industrial three-cascade-reactor plant
Treatment performance
Inuent load impact
Non-ionic and anionic surfactants

a b s t r a c t
An industrial three-cascade-reactor plant treating 45 m3 d1 of dairy wastewater (DW) was monitored
for approx. one year to investigate the effect of variable daily inuent loads. It removed more than 85%
COD, NH4 -N and non-ionic and anionic surfactants from DW within the loads 724, 0.42.3, 0.40.7
and 0.10.5 kg d1 , respectively; NH4 -N removal, in particular, was almost quantitative. Although the
degradation of the above parameters below the lower load thresholds declined to 78.7, 87.5, 50.2 and
64.7%, respectively, their residual concentrations met efuent discharge standards. The biomass settling
properties, assessed as sludge volume index (SVI), were satisfactory (generally lower than 150 ml g1 )
regardless of the organic load of the inuent. The depletion of the pollutant load took mainly place in
the rst reactor albeit a signicant contribution to the removal of the slowly degradable organic matter
fraction was given by the two subsequent reactors.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
The dairy industry is one of the largest sources of agro-industrial
wastewaters worldwide. In a typical manufacturing plant, the efuent is primarily generated by cleaning and washing operations and
its volume generally ranges between 2- and 3-fold the volume of
processed milk depending on the plant typology [1,2].
The dairy wastewater (DW) is characterized by high biological
(BOD) and chemical oxygen demand (COD) and generally contains fats, nutrients, lactose, as well as detergents and sanitizing
agents [1,3]. Serious environmental problems can, thus, arise if
the DW is discharged either untreated or partially treated. To
meet stringent efuent discharge criteria, aerobic biological treatments relying on conventional activated sludge plants are generally
employed [3].
The intrinsic chemical characteristics of DW, including a high
C:N ratio, phosphorus limitation and the presence of highly soluble and degradable organic substances, often referred to as rapidly
biodegradable COD (rb-COD), might result in poor sludge settling (i.e., bulking phenomena) which, in turn, negatively affects
the efciency of these plants [4]. To overcome these problems,
the use of reactors in series, often referred to as cascade reactors, has been proposed since, at the same total working volume,

Corresponding author. Tel.: +39 0761 357332; fax: +39 0761 357242.
E-mail address: petrucci@unitus.it (M. Petruccioli).
1359-5113/$ see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.procbio.2013.04.009

they may enable greater treatment efciency than conventional

activated sludge plants [5]. To date, however, little information
is available on the treatment performances of this plant typology at the industrial level and, above all, it is limited to either
municipal wastewaters [5,6] or paper mill efuents [7]. An industrial three-cascade-reactor plant with a treatment capacity of
45 m3 DW d1 was operated under varying aeration conditions,
obtained by means of on/off cycles of the blower, and highest treatment performances were obtained with a 45/15 min on/off aeration
regime providing 45.4 kg O2 d1 [8]; in the same study, however,
the plant performances were not investigated as a function of the
pollutant load of the feeding wastewater and thus the process
robustness was not assessed. An additional nding of that previous
study was that the removal of organic load was highly and significantly correlated with both richness and ShannonWeaver index
of protozoa rather than with those of bacteria [8].
Objective of this study was to assess the effect of markedly different loads, under the aforementioned optimal aeration regime,
on DW treatment performances in a three-cascade-reactor plant.
Efuent quality was inferred from removals of conventional
organic and inorganic pollutants (COD, BOD5 , total nitrogen,
NH4 -N and PO4 -P). An additional aim was to assess the impact
of the treatment process on anionic (Methylene Blue Active
Substances, MBAS) and non-ionic (Bismuth Active Substances,
BiAS) surfactants, the fate of which, despite their common presence in DW, is often neglected in treatment studies of this
efuent.

C. Tocchi et al. / Process Biochemistry 48 (2013) 941944

The treatment plant of Buona Tavola Sini factory was designed and manufactured in glass ber tanks by the Manzi Aurelio Srl (Monteascone, VT, Italy) and
consisted of a primary sedimentation section (200 m3 ), also acting as an equalization
basin, three aerated cascade reactors (R1, R2 and R3) (18 m3 capacity each) and a
secondary sedimentation section (18 m3 ) [8]. A digestion section (18 m3 ), connected
with three drying beds, collected the excess sludge (approx. 3 m3 d1 ); the recirculation ratio was kept constant at 150%, while the average sludge age was 11.7 days.
Aeration regime was maintained for the whole experimentation at 45/15 min (on/off
cycle of the blower) [8]. During the study, the plant worked at a hydraulic load of
about 45 m3 d1 with a hydraulic retention time of roughly 10 h for each aerated
reactor. Samples of wastewater inuent, mixed liquor of the three aerated reactors,
liquid efuent and recirculation sludge were taken every week in duplicate and subjected to physico-chemical and biological analyses. Due to the large volume of the
equalization basin, the ensuing variations of inuent loads were attenuated. At each
inuent load, measurements were done under operational steady-state conditions,
which were assumed to be reached after approx. 10 reactor volumes of treated DW.
2.3. Physico-chemical analyses
Dissolved oxygen (DO), temperature, redox potential (ORP) and pH were measured directly on-site [8]. The samples were analyzed for TOC, COD, BOD5 , NH4 -N,
NO2 -N, NO3 -N, Ntot (Kjeldahl), PO4 -P, BiAS, MBAS, suspended solids, mixed liquor
suspended solids (MLSS), sludge volume index (SVI) and chloride ions (Cl ) using
procedures accordingly to the Standard Methods [9].
In order to quantify the whole treatment performance of the cascade reactors,
removal efciencies (RE%) were calculated for the chemical parameters (COD, NH4 N, PO4 -P, BiAS and MBAS) using the following equation:
RE (%) =

Cin Ceff
Cin

100

where Cin and Ceff are the concentrations in the inuent (IN) and in the efuent
(after settling) out of reactor R3, respectively. Data from the two sets of samples
were averaged.
2.4. Statistical analysis
For each sampling time, duplicate samples were taken and two replicates were
done for each parameter in each sample (4 total replicates). Data were analyzed by
one-way ANOVA and multiple pair-wise comparisons were performed by the Tukey
test (P 0.05) using the Sigmastat 3.5 software (Jandel Corp, San Rafael, CA).

3. Results and discussion

Dairy wastewater (DW) was treated in an industrial plant
equipped with three cascade reactors and the treatment process
was followed for approximately one year. To best show efcacy
and exibility of this plant in treating DW, typically characterized
by large variability in both composition and load, the experimental
results have been grouped in classes based on specic load ranges.
3.1. Removal of organic load
COD removal was higher in the rst reactor (R1) than in the
other ones (R2 and R3), regardless of the TOC load range: in fact,
the differences between the COD load of the inuent wastewater
(IN) and the residual one after treatment in R1 were always signicant (P 0.01) (Fig. 1). Throughout the whole experimentation, the
residual pollutant content of the nal efuent was almost always
low with average residual COD and BOD5 of 80.3 and 24.4 mg l1 ,
respectively. Best COD removal performances (between 92.7 and

100

b
a

cdB

70

bB

40

60
50
40

aB

30
abA
abA
aA

abA
abA
abA

aA
aA
aA

abcA
abA
abA

bcA
bcA
bcA

abB

30

10

IN
R1
R2
R3

bcB

50

20

90
80

cB

60

aB
aB
aB

2.2. Wastewater treatment plant, operative conditions and sampling procedures

70

ab

20
10

Total removal efficiency (%)

The DW came from the cheese factory Buona Tavola Sini (Monterosi, Viterbo,
Italy) which processes 1520 m3 of milk daily and produces about 45 m3 d1 of
wastewaters; whey was disposed of separately. The DW characteristics expressed
in terms of mean values and standard deviations during the one-year monitoring were (mg l1 ): TOC, 320.4 145.2; COD, 898.1 399.7; BOD5 , 514.4 187.3;
Ntot , 26.3 16.8; NH4 -N, 24.1 14.9; NO2 -N, 0.1 0.9; NO3 -N, 2.3 3.0; PO4 -P,
12.5 13.5; MBAS, 4.1 3.3; BiAS, 1.5 1.8; Chloride (Cl ), 3131.0 2396.1; suspended solids, 335.9 219.0; pH, 6.4 0.7. Sampling was made on a weekly basis.

Influent COD load (kg COD d-1);

COD residual load (kg COD d-1)

2.1. Dairy wastewater (DW)

dB

80

2. Materials and methods

cA
cA
cA

942

0
(24-22) (21-19) (18-16) (15-13) (12-10)

(9-7)

(6-4)

TOC (kg C d-1)

Fig. 1. Inuent (IN) and COD residual loads in the cascade reactors (R1, R2 and R3)
at different ranges of TOC loads and related total efciencies, calculated between IN
and out of R3. Data are the means of 4 replicates and error bars indicate standard
deviations. Pairwise comparisons were performed by the Tukey test (P 0.05): for
each parameter, same lowercase or uppercase letters denote absence of statistically
signicant differences within the same group (i.e., IN, R1, R2, R3 and Total removal
efciency) at different TOC load ranges or between different groups at the same TOC
load range, respectively.

94.2%) were recorded when the inuent TOC was in the range
1018 kg d1 . Conversely, TOC loads lower than 10 kg d1 led
to reduced performances (< 84%) probably due to reduction of
the food/microorganisms ratio (F/M, kg BOD5 kg MLSS1 d1 ), that
were lower than 0.46, 0.09 and 0.09 d1 in R1, R2 and R3,
respectively, and the possible occurrence of concomitant predation phenomena [10,11]. The distribution of the activated sludge
biomass (MLSS) in the three reactors (Fig. 2A) was greatly inuenced by the daily inuent organic load consistently decreasing
with its decrease. On the other hand, no differences in sludge
biomass were observed among the three reactors, likely consequence of the free passage of the sludge from the rst to the
following reactors which were connected in series. The biomass
settling properties, assessed as sludge volume index (SVI), were
satisfactory regardless of the organic load of the inuent: SVI values were generally lower than 150 ml g1 (Fig. 2B) value that was
suggested to be the boundary between bulking and non-bulking
conditions [4].
The
highest
substrate
removal
rate
(SSRR,
kg BOD5 kg MLSS1 d1 ) was observed in R1 since the large
majority of the readily degradable organic matter fraction was
depleted therein (Fig. 2C) in agreement with other studies [5,12].
In particular, the average SSRR steeply decreased from 0.0217 in
R1 to 0.0014 and 0.0006 d1 in R2 and R3, respectively. Nevertheless, R2 and R3 were of fundamental importance ensuring the
metabolization of the recalcitrant and/or the slowly degradable
organic fraction, mainly made up of fats, proteins and detergents
[13,14].
The highest metabolic activity was observed in R1 as indicated
by respirometric data (Fig. 2D). Indeed, both dissolved oxygen (DO)
and redox potential (ORP) were always lower in this reactor than
in R2 and R3, consistently with the fast degradation rate of the
organic matter shown in Fig. 2C [15,16]. Particularly, loads higher
than 20 kg BOD5 d1 in R1 led to marked declines in both ORP and
DO reaching levels as low as 130 mV and 0.2 mg l1 , respectively
(Fig. 2C), thus determining a typical anoxic environment [8,17].
After the depletion of the readily degradable substrate in R1, DO
and ORP increased in R2 and R3. This was probably due to two
concomitant factors, namely, a lower oxygen requirement in processes acting on the residual organic matter fraction [13,15] and the
occurrence of starvation conditions in R2 and R3 as indicated by the
respective F/M mean values (0.18 and 0.16 d1 , respectively) [18].
Moreover, it is worth noting that the starvation conditions which

dB

DO-R1
DO-R2
DO-R3

cA

bB

dA

dB

(40-34)

6
4

bA

aB
aA

-100

dA
cAB
bB

bB
aC

aC

aC

2
0

(33-27)

(26-21)

bcB

cB

aA
aA
aA

aA

abA

bA

cAB
bA
bcB

0.00

(20-13)

(12-6)

Organic Load (kgBOD 5 d -1)

Fig. 2. (AD) Mixed liquor suspended solid (MLSS, A), sludge volume index (SVI,
B), specic substrate removal rate (SSRR, C) and redox potential (ORP) and dissolved oxygen (DO) (D) in the three reactors (R1, R2 and R3) at different ranges of
organic loads (kg BOD5 d1 ). Data are the means of 4 replicates and error bars indicate standard deviations. Pairwise comparisons were performed by the Tukey test
(P 0.05): for each parameter, same lowercase or uppercase letters denote absence
of statistically signicant differences within the same group (i.e., R1, R2 and R3) at
different organic load ranges or between different groups at the same organic load
range, respectively.

occurred in these two reactors could have a positive role in favoring the aggregation of the activated sludge ocs [18] as apparently
indicated by the decrease of average SVI values that passed from
108.5 in R1 to 81.3 and 78.2 ml g1 in R2 and R3, respectively.
3.2. Removal of the nitrogen substrate and phosphate
The continuous plant of the present study was not equipped
with a traditional separate anaerobic sector for denitrication but
operated alternating on/off aeration cycles in the three reactors
where nitrication and denitrication could occur both by alternate
anoxic and aerobic conditions and by the co-existence within the
activated sludge oc of heterotrophic, nitrifying and denitrifying
populations [8,19,20].
The cascade-reactor plant showed high performance in the
removal of the ammonium ion (NH4 + ) with efciencies that
reached, in most cases, 100% (Fig. 3A). Interestingly, and despite

dB

bAB
cB

0.4

bA

aA

0.1

aA
aA
aA

0.2

aA
bA

aA

0.3

abA
abA
bA

aA
aA
aA

aB

abB

bB
abA
aA
aA

abA
aA
aA

bA
abA
bA

bA
abA
abA

0.03

110
100
90
80
70
60
50
40
30
20
10
0

bcA

0.06

0.0
(2.3-2)

100

bcA

0.09

aA
aA
aB

DO (mg -1
l)

ORP (mV)

cC

200

-200

aA
aA
abA

aA
dB

aA
cA

aA

aA
bA

cB

0.01

300

0.0
0.12

bA

aC

0.02

0.00

0.5

IN
R1
R2
R3

-1

bcB

bcB

bB

0.03

cC

bA
bA

SSRR (kgBOD5 kgMLSS-1 d-1)

50

1.0

bA

aA

aA
aA

bB
aA
bA

aA
bA

abA
aA

100

aA
aA
abA

abB

150

aA

SVI (ml g-1)

0.0
200

1.5

cA

0.5

bA
aA

1.0

cB

1.5

aB
aAB
aA

2.0

abA
bA
bA

bA
bA
bA

2.5

2.0

cA
cA

R1
R2
R3

cA
cA
cA

3.0

943

Total removal efficiency (%)

MLSS (g -1
l)

3.5

cA
dA
cA

4.0

Influent NH4-N load (kg d-1);

-1
Residual
NO
-N
load
(kg
d
)
Residual NH4-N load (kg d-1)
2
Residual NO3-N load (kg d )

C. Tocchi et al. / Process Biochemistry 48 (2013) 941944

(1.9-1.6) (1.5-1.2) (1.1-0.8) (0.7-0.4) (0.4-0.1)

Nitrogen load (kg Ntot d-1)

Fig. 3. (AC) Inuent (IN) and NH4 -N residual loads in the three reactors (R1, R2
and R3) at different ranges of total nitrogen loads and the related total efciencies,
calculated between IN and out of R3 (A); NO2 -N (B) and NO3 -N (C) residual loads
in the reactors (R1, R2 and R3) at the same total nitrogen loads. Data are the means
of 4 replicates and error bars indicate standard deviations. Pairwise comparisons
were performed by the Tukey test (P 0.05): for each parameter, same lowercase or
uppercase letters denote absence of statistically signicant differences within the
same group (i.e., IN, R1, R2, R3 and Total removal efciency) at different total nitrogen
ranges or between different groups at the same total nitrogen range, respectively.

the fact that the DO and ORP conditions were not always favorable to nitrication, the largest part of the NH4 + was removed in
R1. Differently, Metcalf and Eddy [5] had found high nitrication
levels only from the second reactor in a similar plant typology
working, however, on municipal efuents that are, generally, characterized by signicantly lower C/N ratios than DW [3]. In the
present study, specic NH4 -N removal rates, in terms of kg of
NH4 -N oxidized/removed daily per kg of MLSS, were higher in R1
(0.070.01 d1 ) than in R2 and R3 (0.010.0002 and 0.0020 d1 ,
respectively). The nitrication process took place in all three reactors albeit the amounts of nitrites and nitrates depended on the
inuent total nitrogen load (Fig. 3B and C). The sum of nitrites plus
nitrates did not ever match the amount of removed nitrogen (Fig. 3B
and C) and this probably depended on both NH4 -N assimilation by
the growing biomass and denitrication that mainly occurred in
R1 (Fig. 3B) [21]. Noteworthy, as already observed for the organic
substrate, low nitrogen loads (0.10.4 kg Ntot d1 ) resulted in a signicantly lower nitrogen removal efciency (87.5 3.3%).
The overall PO4 -P, removal percentage was generally above 90%.
Phosphorous release phenomena, however, occasionally occurred
in all three reactors, even under aerobic conditions, leading to a
higher phosphate concentration in the efuent than in the inuent
(data not shown). It must be noted, however, that the concentration
of phosphates in the efuent never exceeded 10 mg l1 , even in the
presence of occasional PO4 -P release phenomena.

944

C. Tocchi et al. / Process Biochemistry 48 (2013) 941944

0.40

cC

0.35

IN
R1
R2
R3

0.20

80

bB

70
60

aA

40

dB

10
100
90

bcB

80
a

bA

0.25

cdB

70
60
50
40

(0.3-0.2)

(0.2-0.1)

aA
aA
aA

aA
aA
aA

aA
aA
aA

aA
aA

0.10
0.05

30

aA

abA

bA

bB

bA

MBAS (kg d-1)

0.35

0.15

20
0

c
bc

0.20

aA
aA
aA
aA

aA
aA
aA
aA

aA

aA
aA
aA

aA
aA
aA

aA

0.05

Total removal efficiency (%)

30

0.10

0.30

Acknowledgements

50

0.15

0.00
0.40

performances for COD, BOD5 , NH4 -N, MBAS and BiAS, thus showing good exibility and reliability. Also when the polluting load was
very low and the corresponding removal efciencies decreased, the
residual pollutant content of the efuent was always largely below
the legal limits.

100
90

0.25
bB

BiAS (kg d-1)

0.30

0.00

20
10
0

(0.7-0.6)

(0.5-0.4)

(0.1-0)

TSL (kg d-1)

Fig. 4. (AB) Inuent (IN) and surfactant residual loads in the three reactors (R1, R2
and R3) at different ranges of total surfactant loads (TSL) and related total removal
efciencies, calculated between IN and out of R3. (A) MBAS (Methylene Blue Active
Substances); (B) BiAS (Bismuth Active Substances). Data are the means of 4 replicates
and error bars indicate standard deviations. Pairwise comparisons were performed
by the Tukey test (P 0.05): for each parameter, same lowercase or uppercase letters
denote absence of statistically signicant differences within the same group (i.e., IN,
R1, R2, R3 and Total removal efciency) at different TSL ranges or between different
groups at the same TSL range, respectively.

3.3. Removal of the total surfactants

The majority of surfactants have been shown to be good
substrates for the activated sludge [22]. Some of them (e.g., linear alkylbenzene sulfonates, LAS, and nonylphenol ethoxylates,
NPnEO), however, are only partially biodegradable due to the accumulation of dead-end degradation products which, in turn, can
cause environmental problems [22,23].
The cascade-reactor plant showed great efciency in the
removal of non-ionic (BIAS) as well as anionic (MBAS) surfactants
with efciencies higher than 92% when the inuent surfactant load
was 0.40.5 kg d1 . The removal efciency, however, tended to
decrease reaching minimum values (50.2 4.3 and 64.7 4.4% for
BiAS and MBAS, respectively) when the load was less than 0.1 kg d1
(Fig. 4A and B).
The large majority of surfactants were removed in R1, where
adsorption phenomena on activated sludge biomass could mainly
occur [23,24]. A further contribution of R2 and R3 reactors was
invariably observed and found to increase with increasing loads.
MBAS were removed more efciently than BiAS, regardless
of the surfactant load (average removal efciency, 81.9 3.6 vs.
75.7 4.2%, P = 0.043) thus conrming some ndings obtained with
municipal wastewaters [25].
Noteworthy, the residual concentrations of both MBAS and BiAS
out of reactor R3 were lower than 1.0 mg l1 in the majority of cases
with the only exception of the former the residual concentration of
which was 1.9 mg l1 at the highest surfactant load.
In conclusion, the three-cascade-reactor industrial plant monitored in this study efciently treated DW with high removal

This research was partially funded by Manzi Aurelio s.r.l. (Monteascone, Italy). The Authors thank the Buona Tavola Sini dairy
(Monterosi, Italy) staff for sampling assistance.
References
[1] Britz TJ, van Schalkwyk C, Hung YT. Treatment of dairy processing wastewaters.
In: Yapijakis C, Hung YT, Lo HH, Wang LK, editors. Waste Treatment in the Food
Processing Industry. Boca Raton, FL: CRC Press; 2006. p. 128.
[2] Munavalli GR, Saler PS. Treatment of dairy wastewater by water hyacinth.
Water Sci Technol 2009;59:71322.
[3] Kushwaha JP, Srivastava VC, Mall ID. An overview of various technologies for
the treatment of dairy wastewaters. Crit Rev Food Sci Nutr 2011;51:44252.
[4] Donkin MJ. Bulking in aerobic biological systems treating dairy processing
wastewaters. Int J Dairy Technol 1997;50:6772.
[5] Metcalf, Eddy. Wastewater Engineering: Treatment and Reuse. NY, USA:
McGraw-Hill; 2003.
[6] Dockhorn T, Dichtl N, Kayser R. Comparative investigations on COD-removal
in sequencing batch reactors and continuous ow plants. Water Sci Technol
2001;43:4552.
[7] Mobius CH. Biological treatment of papermill wastewater in an activated sludge
cascade reactor. Water Sci Technol 1989;21:18258.
[8] Tocchi C, Federici E, Fidati L, Manzi R, Vinciguerra V, Petruccioli M. Aerobic
treatment of dairy wastewater in an industrial three-reactor plant: effect of
aeration regime on performances and on protozoan and bacterial communities.
Water Res 2012;46:333444.
[9] APHA. Standard Methods for the Examination of Water and Wastewater. 21st
ed. Baltimore: American Public Health Association, AWWA, WEF; 2005.
[10] Comas J, Rodrguez-Roda I, Sanchez-Marre M, Cortes U, Freixo A, Arraez J, Poch
M. A knowledge-based approach to the deocculation problem: integrating
on-line, off-line, and heuristic information. Water Res 2003;37:237787.
[11] Coello Oviedo MD, Lpez-Ramirez JA, Sales Mrquez D, Quiroga Alonso JM.
Evolution of an activated sludge system under starvation conditions. Chem
Eng J 2003;94:13946.
[12] Farizoglu B, Uzuner S. The investigation of dairy industry wastewater treatment in a biological high performance membrane system. Biochem Eng J
2011;57:4654.
[13] Miksch K. Optimization of the cascade system of activated sludge by means
of the metabolic activity measurements of microorganisms. Water Sci Technol
1991;24:6571.
[14] Hwang S, Hansen CL. Characterization of and bioproduction of shortchain organic acids from mixed dairy-processing wastewater. Trans ASAE
1998;41:795802.
[15] DImporzano G, Adani F. The contribution of water soluble and water insoluble
organic fraction to oxygen uptake rate during high rate composting. Biodegradation 2007;18:10313.
[16] Ndegwa MP, Wang L, Vaddella KV. Potential strategies for process control and
monitoring of stabilization of dairy wastewaters in batch aerobic treatment
systems. Process Biochem 2007;42:12728.
[17] Dubber D, Gray NF. The effect of anoxia and anaerobia on ciliate community in
biological nutrient removal systems using laboratory-scale sequencing batch
reactors (SBRs). Water Res 2011;45:221326.
[18] Bossier P, Verstraete W. Triggers for microbial aggregation in activated sludge.
Appl Microbiol Biotechnol 1996;45:16.
[19] Dytczak MA, Londry KL, Oleszkiewicz JA. Activated sludge operational regime
has signicant impact on the type of nitrifying community and its nitrication
rates. Water Res 2008;42:23208.
[20] Yuan X, Gao D. Effect of dissolved oxygen on nitrogen removal and process
control in aerobic granular sludge reactor. J Hazard Mater 2010;178:10415.
[21] Akpor OB, Momba MNB, Okonkwo JO, Coetzee MA. Nutrient removal from activated sludge mixed liquor by wastewater protozoa in a laboratory scale batch
reactor. Int J Environ Sci Technol 2008;5:46370.
[22] Ying GG. Fate, behavior and effects of surfactants and their degradation products in the environment. Environ Int 2006;32:41731.
[23] Gori R, Camilli L, Petrovic M, Gonzales S, Barcelo D, Lubello C, Malpei F. Fate of
surfactants in membrane bioreactors and conventional activated sludge plants.
Environ Sci Technol 2010;44:82239.
[24] Temmink H, Klapwijk B. Fate of linear alkylbenzene sulfonate (LAS) in activated
sludge plants. Water Res 2004;38:90312.
[25] Othman MZ, Ding L, Jiao Y. Effect of anionic and non-ionic surfactants on
activated sludge oxygen uptake rate and nitrication. Int J Civ Environ Eng
2009;58:1199205.