Sei sulla pagina 1di 12

Journal of Hazardous Materials 323 (2017) 329–340

Contents lists available at ScienceDirect

Journal of Hazardous Materials

journal homepage:

Fate and mass balance of triclosan and its degradation products:

Comparison of three different types of wastewater treatments and
aerobic/anaerobic sludge digestion
Fatemeh Tohidi a , Zongwei Cai a,b,∗
State Key Laboratory of Environment and Biological Analysis, Department of Chemistry, Hong Kong Baptist University, Kowloon Tong, Hong Kong, China
School of Environmental Science and Engineering, Guangdong University of Technology, Guangzhou, China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

• Fate of triclosan was investigated

in different wastewater and sludge
• TCS was considerably biodegraded
in aerobic/anaerobic sludge digestion
• TCS transformation products were
observed based on the treatment
• Kd values were estimated for TCS,
2,4-DCP and MTCS in primary and
secondary sludge.
• TCS temporal variations as well as
few relationships were investigated.

a r t i c l e i n f o a b s t r a c t

Article history: Triclosan (TCS) as an antimicrobial agent has been ubiquitously found in wastewater and sewage
Received 28 January 2016 sludge. TCS may undergo transformation/degradation during wastewater treatment. Some of the resulted
Received in revised form 12 April 2016 products such as 2,4-dichlorophenol (2,4-DCP), 2,8-dichlorodibenzoparadioxin (2,8-DCDD) and methyl
Accepted 13 April 2016
triclosan (MTCS) are presumed toxic/persistent compounds. In this study, fate of TCS and the probability of
Available online 27 April 2016
formation of important degradation products were investigated in three susceptible wastewater/sludge
treatment practices. 24.1% and 27.2% of the loading TCS was adsorbed to the generated sludge, whereas
up to 60% of the loading TCS was biotransformed. Up to 9.9% and 13.0% of TCS loss was attributed to
Sewage treatment plant
the formation of 2,4-DCP and 2,8-DCDD in chlorination and UV disinfection, respectively. Anaerobic and
Mass balance aerobic sludge digestion processes eliminated up to 23.0% and 56.0% of TCS, respectively. About 7.4% of
2,4-Dicholorophenol TCS in aerobic digestion was transformed to methyl triclosan (MTCS). Significant temporal variation of
2,8-Dichlorodibenzoparadioxin TCS was observed in primary sedimentations, except for chemically enhanced primary treatment that
Distribution coefficient was suggested to be governed by chemical-forced sedimentation. Distribution coefficient (Kd ) of TCS
Biodegradation was directly correlated to the total organic carbon of the sludge (TOC). Moreover, strong correlation was
observed between elimination efficiency in primary sedimentation and loading concentration.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction

The production and use of triclosan (TCS) (Table 1S, Supplemen-

∗ Corresponding author at: State Key Laboratory of Environment and Biological
tary information) as a broad-spectrum anti-microbial agent in a
Analysis, Department of Chemistry, Kowloon Tong T1215, Hong Kong China.
E-mail address: (Z. Cai).
wide range of health care products has extensively become perva-
0304-3894/© 2016 Elsevier B.V. All rights reserved.
330 F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340

sive since its first introduction in 1972. Singer et al. [1] estimated ing aerobic/anaerobic sludge digestion. The information obtained
that TCS is globally produced over 1500 tons per year by which based on the specific treatment processes is very valuable for set-
Europe participates in 23.3% of the total production. As a result ting the correct strategies toward the improvement of treatment
of widespread use of TCS, the chemical has been found in nearly efficiency which in turn causes the reduction in introduction of
every compartment of the environment such as surface water [2], organic compounds into the environment. For this purpose, analy-
wastewater [3], sediment [4], sludge [5], human plasma and milk sis of raw and treated wastewater and sludge samples were carried
[6], urine [7], fish [8] and etc. TCS has low acute and chronic toxicity out and the mass balance of TCS in three STPs were performed
on estuarine organisms [9,10]. However, there are some big con- accordingly based on the experimental data. TCS fate was probed
cerns about this compound. Firstly, TCS has been reported to have in real-based treatment plant during wastewater and sludge aer-
adverse effect on the aquatic livings [11,12] as well as the micro- obic/anaerobic treatment condition. In addition, the occurrence of
bial and algae community [13–15]. Secondly, endocrine disruption probable degradation products, 2,8-DCDD, 2,4-DCP and MTCS and
effect of this compound has been documented vastly [12,16]. More the transformation rate were investigated. Finally, the adsorption
importantly, TCS can be the precursor of more toxic compounds. behavior of TCS and some of the occurred degradation products was
Yet, due to insufficient information about the safety and potential studied.
hazard of TCS, the chemical is still under review by the U.S. Food
and Drug Administration (FDA) and has been banned as biocidal
2. Materials and methods
product in Europe.
Basically, TCS can be transferred to wastewater through indus-
2.1. STPs description
trial or domestic sewage flow, ranging from ng/l to ␮g/l in sewage
treatment plants (STPs). Subjected to treatment, a considerable
Three different STPs, STP (I), STP (II) and STP (III) were selected
amount of TCS will adsorb to the sewage sludge [17] leaving a
based on different treatment practices. STP (I) is a primary treat-
fraction to be discharged to the environment [18]. During aerobic
ment plant serving a population of 23,000 people. The plant is
treatment, TCS can be partially degraded through microbial activ-
designed to treat about 8600 cubic meters per day domestic
ity [19]. Thus, elimination of TCS may be mainly physical sorption
wastewater. The treatment train is comprised of pretreatment
onto the sludge and biodegradation, and to less extent other trans-
(grit screening) and primary treatment. After passing the pri-
formations [20–22]. Even though the portion of TCS transformation
mary sedimentation tank with a hydraulic residence time of 2–3 h,
is fairly low, it is still considered as the major concern because
the wastewater is directly discharged to the environment. Sludge
the products are known or presumed carcinogens/toxicants. It
treatment process consists of aerobic digestion and the treated
has been well documented that TCS can transform to chlorophe-
sludge is dewatered to be disposed to landfill. STP (II) runs chem-
nols, dioxins, methylated TCS, chlorinated TCSs and etc. in certain
ically enhanced primary treatment (CEPT) that is supposed to
circumstances [3,4,8]. For instance 2,4-dichlorophenol (2,4-DCP)
improve treatment process by enhanced flocculation. The plant
stemmed from TCS molecule breakdown, is merged in Group 2B,
is capable of handling daily 1.4 million cubic meters of sewage
possibly carcinogenic to humans (World Health Organization 1999)
on average for serving a population of 3.5 million people. After
and is considered as the priority toxic pollutant by US Environ-
contacting with chemicals and settling in primary sedimenta-
mental Protection Agency (USEPA). 2,8-dichlorodibenzoparadioxin
tion for about 2–3 h, the wastewater is conducted to disinfection
(2,8-DCDD) originated from TCS transformation, is a member of
chamber for treating with sodium hypochlorite and being dis-
dioxin family with a recently reported relative potency factors of
charged accordingly. Produced sludge is only centrifuged and
0.0001 compared to 2,3,7,8-tetrachlorodibenzoparadioxin (2,3,7,8-
dewatered for landfill disposal. STP (III) as a secondary treat-
TCDD) [23]. 2,8-DCDD can transform to higher chlorinated dioxins
ment employs activated sludge treatment (biological treatment)
with higher toxicity under certain conditions such as chlorination
with a daily capacity of 200,000 cubic meters per day for a pop-
[4]. Lastly, microbial degradation of TCS under aerobic condition
ulation of 600,000 people. The influent passes grit removal and
results in methylation of hydroxyl group to form methyl triclosan
primary sedimentation to flow into biological tank where bio-
(MTCS). Although formation of MTCS mitigates the possibility of
logically degradable compounds are prone to degradation by the
dioxin formation by shielding the OH group, it ends up with a
microorganisms. The hydraulic residence time of aerobic biologi-
more persistent compound than the parent TCS which tends to
cal treatment is between 8 and 10 h. The effluent from the biological
bioaccumulate in the sludge or living organisms after being dis-
tank settles in secondary sedimentation and finally is exposed to
charged [24].
UV radiation in UV chamber for disinfection purpose. The sludge
Although TCS occurrence in wastewater has been reported
originated from primary sedimentation and secondary sedimen-
vastly, little attention has been paid to its significance at different
tation are blended and fed into an anaerobic digester where the
STPs sites related to treatment practice. In addition, the behav-
sludge is anaerobically digested at 35–37 ◦ C for an average period
ior of compound in the sludge treatment has been barely noticed
of 19 days. After treatment, the sludge is centrifuged and dewa-
probably due to the complexity of sludge matrix that makes the
tered for final disposal. The description about STPs and wastewater
analysis quite troublesome. TCS and its degradation products are
quality is summarized in Tables 2S–4S, Supplementary informa-
prone to partition to the sludge due to their hydrophobic charac-
teristic. If not treated appropriately, polluted sludge may increase
public health risk severely due to accumulation of organic pollu-
tants and ultimate transfer to the environment. Thus, finding out 2.2. Sample collection and preparation
the behavior and fate of TCS along a STP without reliable knowl-
edge about the occurrence in wastewater, partitioned fraction Five campaigns were conducted during one year to collect com-
to the sludge, degradation and generated products is impossible. posite samples from wastewater and sludge. Campaigns 1, 4 and
More important is that the products might exhibit higher tox- 5 were related to December to March (cold dry periods) while
icity and persistency than TCS. Therefore, current research was campaigns 2 and 3 were performed in May to September (warm
aimed to determine the fate of TCS during different wastewater rainy periods). Basically, the samples were taken from inlet, after
and sludge treatment practices that were susceptible to affect TCS each treatment unit and the outlet of wastewater and sludge. Aver-
and investigate how each treatment may contribute to the fate of age climate temperature ranged from 8 to 16 ◦ C and 28–36 ◦ C for
TCS. Besides, attempt was done to focus on TCS degradation dur- cold and warm periods, respectively. Sampling points are presented
F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340 331

Fig. 1. (a) Maass balance analysis of TCS in wastewater in STP (I). (b) Mbass balance analysis of TCS in sludge in STP (I), PS: primary sludge.

in Fig. 1S (Supplementary information). 24-h composite wastew- secondary sludge samples were collected in amber glass jars and
ater samples were taken from a flow-proportional autosampler. appropriate amount of methanol was in situ added. Dewatered
Wastewater samples were collected in two 4-L amber bottles. sludge sample was collected in aluminum foil. The samples were
The bottles were prewashed with distilled water, methanol, ethyl then stored in an ice-filling cooler to be delivered to the analytical
acetate, methanol, distilled water. In situ pH measurement was laboratory. Upon arrival, wastewater samples were filtered through
performed and phosphoric acid was added for stabilization. The pre-ashed glass-fiber filters (GF/F, Whatman, UK) and stored in the
samples were then threaded with caps and Teflon and stored in amber bottles at 4 ◦ C for immediate extraction. Filters were oven
to be delivered to the analytical laboratory. Sludge samples were dried at 70 ◦ C until constant weight and stored at −20 ◦ C for anal-
collected from a blended 24-h composite sample. Primary and ysis of particulates. Primary and secondary sludge samples were
332 F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340

Fig. 2. (a) Maass balance analysis of TCS in wastewater in STP (II). (b) Mbass balance analysis of TCS in sludge in STP (II).

centrifuged at 6000 rpm for 8 min. The water was separated and storfer GmbH. Dichloromethane (DCM) and ethyl acetate were
the solid sludge as well as dewatered sludge sample was further provided by ACS Chemical Inc. Nonane and n-hexane were pur-
lyophilized in the freeze-dryer. chased from Anaquachemicals Supply. Methanol and acetonitrile
were obtained from TEDIA Company, Inc. Anhydrous sodium
sulfate, sodium hydroxide, formic acid, N-(tert-Butyldimethylsilyl)-
2.3. Reagents and chemicals N-methyl-trifluoroacetamide (MTBSTFA) and Alumina activated,
acidic Brockmann grade1 (mesh ∼ 150) were from Sigma-Aldrich
TCS and 2,8-DCDD were purchased from Wellington Labora- Corporation. Silica (0.063-0.2 mm) was purchased from Merck Lim-
tories (Ontario, Canada). TCS-d3 was provided by Accustandard ited.
Inc., USA. 2,4-DCP and MTCS were purchased from Dr Ehren-
F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340 333

Fig. 3. (a) Maass balance analysis of TCS in wastewater in STP (III). (b) Mbass balance analysis of TCS in sludge in STP (III), PS: primary sludge, SS: secondary sludge.

2.4. Sample extraction, cleanup and analysis order to a) improve the peak shape and reduce the analytical error
due to peak tailing, and b) enhance the retention time in chro-
The analytical method applied for determination of the target matogram, derivatization of 2,4-DCP was performed on Fraction
compounds in wastewater and sludge samples were the same as 3 using MTBSTFA. For solid samples, extraction was carried out by
the method developed by the authors [25] with some modifica- accelerated solvent extractor. The extract was cleaned up by gel
tion. Briefly, liquid–liquid extraction was used for extraction of permeation chromatography (Bio-Beads S-X beads from Bio-Rad,
liquid samples. After concentration, the extract was loaded onto eluted with 1:1 DCM/hexane) followed by multilayer silica col-
an acidified silica column and eluted with n-hexane/DCM (6:4), umn chromatography. Then, activated copper was used to remove
DCM and DCM/Acetone (98:2) to obtain Fraction 1 (2,8- DCDD and sulfur compounds. For qualitative and quantitative analysis, a gas
MTCS), Fraction 2 (TCS), and Fraction 3 (2,4-DCP), respectively. In chromatograph/mass spectrometer Agilent 6890 with 5975B MSD
334 F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340

Table 1
Summary data of the concentration (n = 5) of target compounds at various units of STPs.


STP (I) (II) (III) (I) (II) (III) (I) (II) (III) (I) (II) (III)
Influent (ng/l) 547 (±22) 422 (±15) 389 (±19) NDb 68.4 (±5.7) ND ND ND ND ND ND ND
Primary sedimentation (ng/l) 421 (±22) 301 (±10) 327 (±14) ND 53.4 (±5.6) ND ND ND ND ND ND ND
Secondary sedimentation (ng/l) NAa NA 162 (±10) NA NA ND NA NA ND NA NA 33.3 (±3.4)
Effluent (ng/l) 421 (±22) 278 (±8) 144 (±9) ND 64.5 (±5.5) ND ND ND 20.9 (±4.1) ND ND 31.3 (±3.0)
Primary sludge (ng/g) 2247 (±185) 447 (±21) 196 (±14) ND 56.7 (±5.2) ND ND ND ND ND ND ND
Secondary sludge (ng/g) NA NA 174 (±12) NA NA ND NA NA ND NA NA 50.6 (±7.3)
Dewatered sludge (ng/g) 2299 (±228) 393 (±21) 310 (±18) ND 111 (±8) ND ND ND ND 262 (±14) ND 36.2 (±4.9)
NA: Not applicable.
ND: Not detected.

was applied. A HP-5MS GC column (30 m × 250 ␮m internal diam- Table 2

Kd values obtained for TCS, MTCS and 2,4-DCP in primary and secondary sludge in
eter × 0.25 ␮m film thickness, (5%-Phenyl)-methylsiloxane) was
three STPs.
used for the separation. GC/MS condition was as follows: Temper-
ature initially increased from 75 ◦ C (for 1 min) to 230 ◦ C at 10 ◦ C, Kd (l/kg) PSa Kd (l/kg)SSb
then to 280 ◦ C at 20 ◦ C/min and was held for 15 min. Injection vol- TCS STP (I) 5362 NAc
ume was 1 ␮l in the splitless mode and 5 min solvent delay was set STP (II) 1470 NA
to prevent filament damage. The injector and the transfer line tem- STP (III) 631 637
perature were 280 ◦ C and 300 ◦ C, respectively. The carrier gas was
2,4-DCP STP (II) 1096 NA
helium at a flow rate of 1.2 ml/min. Mass spectrometer was oper-
PS: primary sludge.
ated in electron ionization mode at 70 eV and calibrated against b
SS: secondary sludge.
perfluorotributylamine (PFTBA) on a monthly basis. Selected ion c
NA: Not applicable.
monitoring was applied for quantitative analysis.
Method validation was performed on spiked reagent blank
compound in the solid phase (ng/kg d.w.) and the dissolved phase,
and pre-determined spiked sample. Isotope dilution mass spec-
respectively (ng/l).
trometry (IDMS) method was employed to ensure the precise
quantitation of TCS. For 250 ml sample, the method detection Cs
Kd = (2)
limit was 10.0, 6.0, 4.0 and 5.0 ng/l for TCS, 2,8-DCDD, 2,4-DCP Ca
and MTCS, respectively. Moreover, for 0.5 g sludge sample, the
method detection limit was 10.0, 1.0, and 5.0 ng/g for TCS, 2,4- 2.5.3. Mass balance calculation
DCP and MTCS, respectively. For 2,8-DCDD, the method detection In order to investigate how various treatments affect TCS and the
limit was 3 ng/g for 10 g sludge sample. Mean recoveries were probability of formation of any treatment related products, mass
obtained as 91(±4)–106(±7)%, 82(±3)–87(±4)%, 86(±6)–87(±8)%, balance of the target compounds was performed throughout the
88(±4)–105(±3)% in wastewater sample and, 88(±5)–96(±8)%, entire process of wastewater and sludge treatment. For this pur-
84(±2)–87(±3)%, 84(±7)–89(±4)% and 88(±3)–97(±5)% in sludge pose, experimental approach was carried out; the experimental
sample for TCS, 2,4-DCP, 2,8-DCDD and MTCS, respectively. More data obtained from the analysis of wastewater and sludge sample
detail about the recovery, precision and accuracy of the developed were applied for the mass balance equations shown below (Eqs.
method is given in Table 5S (Supplementary information). Daily (3)–(10)).
RRF obtained from 3 calibration standards was calculated and com- For primary treatment:
pared with RRF from the calibration curve. For each sequence of min (g/d) = Cin (g/m3 ) × Q in (m3 /d) (3)
samples, a method blank and for each wastewater sample a spiked
sample were analyzed with the same protocol described for sam-
3 3
mout (g/d) = Cout (g/m ) × Q out (m /d) + Cps (g/kg) × Wps (kg/d) (4)
ples to check for any contamination. More detail about quality For secondary treatment:
assurance/quality control and method validation is given in Sup-
plementary information. min (g/d) = Cin (g/m3 ) × Q in (m3 /d) (5)
3 3
mout (g/d) = Cout (g/m ) × Q out (m /d) + Css (g/kg) × Wss (kg/d) (6)
where min and mout , mass fluxes entering and exiting of the unit;
2.5. Calculations Cin and Cout , concentration of influent and effluent of the unit; Cps
and Css , concentrations in primary and secondary sludge; Qin , and
2.5.1. Elimination Qout , flow rates of influent and effluent of each unit; Wps and Wss
The elimination efficiency of TCS in each specific unit was cal- weight of primary and secondary sludge, respectively.
culated based on the difference between total mass flux entering For aerobic sludge treatment
(minf ) and leaving (meff ) in each unit divided by minf of each unit
(Eq. (1)). min (g/d) = Cps (g/kg) × Wps (kg/d) (7)

minf − meff mout (g/d) = Cds (g/kg) × Wds (kg/d) × 0.3
Eliminationefficiency (%) = × 100 (1)
+ Csup (g/m3 ) × Q sup (m3 /d) (8)

2.5.2. Distribution coefficient (Kd ) For anaerobic sludge treatment

The solid–water distribution coefficient (Kd ) was estimated min (g/d) = Cps (g/kg) × Wps (kg/d) + Css (g/kg) × Wss (kg/d) (9)
based on Eq. (2) where Cs and Ca are the concentrations of target
F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340 335

Table 3 [26]. TCS elimination in primary treatment ranged between 18.6%

Seasonal variation of elimination of TCS and degradation products.
to 25.5% with lowest value in warm period and highest value in
PSa STP (I) PS STP (II) PS STP (III) SSb STP (II) cold period. None of the aimed degradation products were detected
TCS +*
− **
+ + in wastewater. The applied practice for sludge treatment and sta-
DCP NAc − NA NA bilization is aerobic digestion. Mass balance analysis showed TCS
MTCSd NA NA NA + mass loss of 43.7% during aerobic digestion. The mass losses were
PS: primary sludge. comparable for five campaigns (34–56%). TCS loss in this pro-
SS: secondary sludge. cess could be directly attributed to the degradation during aerobic
NA: Not applicable. sludge digestion. To the best of our knowledge, only few researchers
Data are all elimination values except for MTCS that are mass flux values.
have reported aerobic degradation of TCS in the sludge aerobic
+: statistically significant difference at P < 0.05.
−: not statistically significant difference at P < 0.05. treatment in a full scale plant [18]. MTCS as the biodegradation
product of TCS was detected in the dewatered sludge while it was
not detected in the primary sludge. This observation might be an
evidence for TCS biodegradation in the aerobic digestion process.
mout (g/d) = Cds (g/kg) × Wds (kg/d) × 0.3 6.8%–15.1% of TCS loss in the sludge treatment was attributed to
MTCS formation and the rest might be due to other degradation
+ Csup (g/m3 ) × Q sup (m3 /d) (10) processes. Other target compounds were not detected either in
where Cds and Csup , concentration in dewatered sludge and super- primary sludge or in dewatered sludge.
natant originated from sludge digestion; Qsup , flow of supernatant; STP (II) showed higher elimination of TCS (24.2%–30.3%) com-
Wds , weight of dewatered sludge produced per day. Since the dewa- pared to primary treatment. Statistical test on elimination data
tered sludge is not completely dried and still contains 70% water, (paired sample t-test, p < 0.05) also confirmed significant differ-
the weight of dewatered sludge should be corrected by considering ence between primary treatment in STP (I) and STP (II), which
a factor of 0.3 of the total weight to obtain the dry weight. implies that TCS elimination might improve by applying floccu-
The sludge amount before and after treatment and the sludge lants to the wastewater. 2,4-DCP was detected along the treatment
amount reduction can be calculated according to the Eqs. (11) and train with a small increase in the samples collected after chlori-
(12): nation. Meanwhile, TCS concentration showed a small drop (up to
9.9%) in the same samples. This shows that chlorination can cause
Wps = Q ps (m3 /d) × TS% (11) TCS transformation to other compounds including 2,4-DCP [27].
Wds = Wds × 0.3 (12) Other target compounds were not observed in the wastewater sam-
ples of STP (II). TCS elimination for the entire process ranged from
where, Wps and Wds are the actual weight of primary and dewa- 28.8%–37.2%. Sludge treatment in STP (II) consists of only centrifu-
tered sludge, respectively. TS is the total solid of the primary sludge. gation and dewatering for the purpose of sludge volume reduction
and easy handling. Expectedly, no significant change in TCS mass
2.5.4. Statistical analyses was observed before and after dewatering process. Moreover, no
Statistical data analysis was performed in Excel, SPSS 16 and Ori- target degradation product was detected in primary and dewatered
gin 8 Pro. Temporal variation analysis was performed using a paired sludge.
two tailed student t-test with ␣ = 0.05 as threshold for significance. Among studied plants, STP (III) with biological treatment
Evaluation of all the elimination rates were based on two tailed stu- showed the highest efficiency for elimination of TCS. Primary sed-
dent t-test of two samples with equal variance. Evaluation of STPs imentation in STP (III) eliminated 13.1%–17.5% of TCS that was
comparisons were based on ANOVA one way analysis. Correlation comparable to primary treatment in STP (I). TCS elimination in
analyses were carried out using a linear model with non-forced biological treatment unit was 95%–98% that could be due to the
interception in Origin 8 pro. All the data presented for concentra- contribution of two probable processes: biodegradation and sorp-
tion, elimination efficiency and mass balance are the average of 5 tion of TCS. The total elimination of TCS throughout the train was
sampling campaigns unless stated. found to be 59.6%–76.0%. TCS mass balance between secondary
sludge and secondary sedimentation revealed that 3.1% of load-
3. Results and discussion ing TCS was adsorbed to the sludge and the rest (96.9%) was lost
in this unit. This confirms that the biodegradation of TCS as a
Table 1 shows a summary of the concentrations of TCS and the well-known phenomenon is the dominant process in the biolog-
degradation products detected from three STPs. TCS concentration ical treatment [19]. TCS mass flux adsorbed to secondary sludge
showed a decreasing trend from influent to effluent in wastewa- was lower in warm period than in cold period, however no sig-
ter. All the concentration changes were significantly different in nificant difference was observed between the two data set. MTCS
each operation unit (p < 0.05), indicating TCS elimination during was observed only in the secondary sedimentation that implies
treatment. Detection frequency for the target compounds during the formation of this compound during biological treatment. MTCS
5 campaigns were 100%, 56%, 31% and 7% for TCS, MTCS, 2,4-DCP occurrence in wastewater was almost constant from secondary
and 2,8-DCDD, respectively. The occurrence depended on the plant sedimentation to effluent, accounting for 25.5% and 15.9% of the
treatment practice, which could be an implication that treatment loading TCS mass loss in secondary sedimentation in the warm
technology might affect TCS transformation to other compounds. and cold period, respectively. 2,8-DCDD as another transforma-
tion product was detected in the effluent that had been exposed
3.1. Fate of TCS to UV irradiation for disinfection purpose. It was observed that
for three campaigns in which 2,8-DCDD was detected, TCS loss
TCS elimination during primary sedimentation in STP (I) could was 13.0%, however for the two campaigns in which 2,8-DCDD
be mainly related to TCS adsorption to the sludge as other factors was not detected, TCS loss was only 4.9%. This indicates that TCS
such as biodegradation or stripping are less likely to take place transformation might take place during UV disinfection in wastew-
due to shorter hydraulic residence time in primary sedimentation ater treatment and result in the formation of 2,8-DCDD or other
tank compared to biological treatment and also negligible volatil- transformation compounds. STP (III) utilizes anaerobic digestion
ity of TCS (Henry’s Law Constant 1.5 × 10−7 atm m3 /mol at 25 ◦ C) treatment for generated sludge. The digester is fed with a mix-
336 F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340

Fig. 4. Correlation between the primary sedimentation of TCS and its occurrence in primary sludge for (a)STP (I) and (b)STP (III).

ture of primary and secondary sludge. Mass balance suggests an ence in degradation values may stem from the difference in the
overall add-up of TCS mass for dewatered sludge (from primary treatment methods applied. Since the mass flux of MTCS had neg-
and secondary sludge) if no digestion occurs in anaerobic process. ligible change before and after treatment, it is assumed that MTCS
Some reports also mentioned that TCS does not undergo anaero- was originally produced during wastewater biological treatment,
bic digestion [18,28]. However, significant TCS loss was observed partitioned to secondary sludge, and did not undergo any signifi-
between the untreated and dewatered sludge which may only cant degradation during anaerobic digestion. So far, this is the first
be attributed to TCS degradation during anaerobic digestion. The report on MTCS degradability under anaerobic sludge treatment.
observation is in agreement with the report published by Samaras It is suggested that methyl group attached to oxygen on the phe-
et al. [29] who found TCS elimination from 60%–80% in anaerobic nol group on TCS structure might retain further degradation thus
digestion depending on the digestion method applied. In this study, leaving a persistent compound against biological degradation. Tak-
TCS anaerobic degradation ranged from 18.6%–23.0%. The differ- ing the amount of MTCS in wastewater into account, it reveals that
F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340 337

Fig. 5. Correlation between distribution of TCS to the particulates in raw wastewater and elimination in primary treatment.

average 7.4% of TCS loss in biological treatment was transformed loss (20.6%) was observed in anaerobic digestion of sludge and the
to MTCS with higher (9.5%) and lower amount (6.6%) in warm and remaining mass in dewatered sludge was disposed of to landfill
cold period, respectively. (Fig. 3b). Formed MTCS in this process was directly discharged to
Figs. 1–3 represent a summary of mass balance analysis of TCS the receiving environment without being affected by sludge anaer-
in wastewater and sludge in three studied STPs. For STP (I), 77.6% of obic digestion. The average mass balances for each unit in three
the entering TCS mass flux to the plant was directly discharged to STPs have been illustrated in Figs. 2S–4S.
the receiving environment (Fig. 1a). The mass that was settled along
the sludge underwent aerobic digestion treatment of which 52.2% 3.2. Correlation between Kd and TOC
was aerobically degraded in the digester. However, 5.4% of TCS in
the sludge was observed to transform to MTCS and 47.8% remained Distribution coefficient (Kd ) is a very important parameter
in the dewatered sludge to be disposed of to landfill (Fig. 1b). in estimating how potent the pollutant is to be adsorbed by
Fig. 2a shows the mass flux partition of TCS in STP (II). Enhanced soil/sediment/sludge. Using default Kd can result in significant
sedimentation process in STP (II) appeared to improve TCS elimina- errors especially when the impact of contaminants migration or
tion (25.5%) compared to STP (I). Furthermore, chlorination caused site remediation evaluation is intended. Therefore, to study the
5.3% TCS loss during disinfection process. Nevertheless, still 68.7% potential sorption of organic pollutants, applying case-defined Kd
of the loading TCS was discharged to the environment through is essential. Distribution behavior of TCS and degradation products
wastewater. All the TCS mass adsorbed to the sludge was appeared was aimed based on the experimental data (Eq. (2)). There are only
to directly be discharged to the environment without significant few researches addressing the partition behavior of TCS of which,
change in mass flux (Fig. 2b). Moreover, no degradation product mostly are focused on the adsorption of TCS onto the sediment and
was detected in the dewatered sludge. Therefore, it is implied soil or in bench scale study [29–33]. While there is limited knowl-
that centrifugation and dewatering had negligible effect on TCS edge on the adsorption of TCS onto sludge in full scale plant [34,35],
removal/transformation. no such data is so far available for MTCS and 2,4-DCP. Table 2 shows
For STP (III), 14.2% and 3.1% of the initial TCS mass flux was Kd values for TCS in primary and secondary sludge for the three
adsorbed to primary and secondary sludge, respectively (Fig. 3a). plants. Mean Kd values were in the order of STP (I) >STP (II) >STP
Mass balance between the TCS mass lost in the biological treat- (III). The obtained values were incomparable which might be due
ment, the mass adsorbed to the sludge and the mass in the effluent the difference in operational and physical condition of each treat-
revealed that 51.6% of the initial TCS mass flux was biodegraded ment plant that may result in difference in sludge characteristics.
in biological treatment of which 19.8% was transformed to MTCS This indicates that applying default Kd that is vastly intended due
(data not shown in the figure). MTCS formation was higher in warm to difficulty in analysis of complicated sludge samples, may lead to
period (25.5%) than cold period (15.9%). In addition, average 10.3% huge bias in analytical determination. Therefore, in order to obtain
of TCS mass was lost in UV irradiation disinfection unit of which accurate information from data analysis, all the involved media
90.8% was transformed to 2,8-DCDD. TCS loss was found to be including wastewater and sludge must be analyzed.
higher in the campaigns in which 2,8-DCDD was observed, indicat- Estimation of Kd for MTCS and 2,4-DCP was aimed as well
ing that this compound was either not formed or formed as small (Table 2). MTCS showed a higher Kd value (1580 ± 248 l/kg) com-
amount as the method detection limit in these two campaigns. This pared to that for TCS in secondary sludge, that can be attributed to
might be due to higher flow rate of wastewater in warm period (due higher lipophilic characteristic of MTCS (values are from secondary
to rainfall) that causes lower probability of TCS exposure to UV irra- sludge samples as MTCS was not observed in primary sludge). For
diation in the required transformation time. Significant TCS mass 2,4-DCP, Kd value was found to be less than (1096 ± 15 l/kg) that for
338 F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340

TCS in primary sludge of the same plant that is in good agreement mation to this hypothesis is significant MTCS formation variation
with the theory as 2,4-DCP is a weak acid with a single aromatic in cold and warm periods (paired sample student t-test, p < 0.05).
ring, thus it presents less lipophilic characteristic than TCS. There- MTCS showed higher mass flux in warm periods (10.73 g/d d.w.)
fore, less partitioning to the sludge and lower Kd value is expected. and lower mass flux in cold periods (5.28 g/d d.w.) that might indi-
As the key factor for the adsorption of organic compounds is cate temperature-influenced biological activity of microorganisms
the organic matter content of the sludge, total organic carbon for biological degradation.
(TOC) of the freeze-dried sludge samples was determined using 3.4. Relationship between distribution of TCS to the particulates
Shimadzu TOC-analyzer model SSM-5000A. Mean TOC for STP (I), in raw wastewater and elimination in primary treatment
STP (II) and STP (III) were obtained as 430 ± 37, 280 ± 22 and TCS was found to be associated with the particulates of raw
200 ± 18 mg/g, respectively. Paired two tailed student t-test anal- wastewater in the range of 17%–34%. The distribution pattern var-
ysis confirmed significant difference between TOC values of plants ied dramatically from cold to warm period (higher in cold period
(n = 29, p < 0.05). Meanwhile, correlation test applied to Kd and TOC and lower in warm period). Fig. 5 shows a trend comparison
data set resulted in a Pearson correlation coefficient of R = 0.92 between TCS fraction partitioned to the particulates and the elim-
(n = 88, p < 0.05). This result indicates that organic matter of the ination in primary treatment. The parallel trend suggests that a
sludge can highly affect the partition of hydrophobic compounds relationship might be attributable between TCS fraction adsorbed
into the solid phase. to the particulates in raw wastewater and elimination efficiency in
primary treatment (R(28) = 0.83, 0.66 and 0.87 for STPs (I), (II) and
3.3. Temporal variation of TCS elimination (III), respectively). This implies that sorption to sludge is the main
elimination mechanism for primary treatment and TCS is partially
TCS elimination exhibited meaningful variation during the cam- removed by adsorption and settlement along the sludge forma-
paigns in a way that higher concentration in influent and sludge tion.
was observed in cold period. Since concentration variation was 3.5. Comparison between mass flux in water phase of primary
important for mass balance investigation, temporal variation of TCS sludge and effluent of the treatment unit
elimination and contribution of treatment practice to this varia- Once in wastewater stream, TCS tends to partition between the
tion was studied. There are only few researches reporting seasonal particulates and water phase. The fraction in each phase depends
variation of other organic pollutants such as PAH elimination dur- on the physical properties of TCS (Kd ). In order to obtain TCS mass
ing primary treatment [36,37]. Applying one-way ANOVA on the flux, determination of the fractions in both phases is necessary.
time-classified data showed that STP (I) and (III) had significant sea- In an attempt to determine TCS fraction in the water and solid
sonal variation in primary sedimentation (p < 0.05). Both plants had mass of the primary sludge, it was hypothesized that the water
higher elimination in cold period and lower elimination in warm phase of primary sludge and the effluent of the primary sedimen-
period (24.7% and 17.0% vs. 18.7% and 13.0% for STP (I) and STP (II), tation unit are of similar characteristic. Thus, the fraction of TCS
respectively). As the TCS influent concentration was significantly in these two matrices should be comparable. Applying two-tailed
higher in cold period, a relationship between influent concentration student t-test on experimental data revealed no meaningful dif-
and elimination in different climate conditions might be implied. ference between TCS fraction in the water phase of the primary
This may be explained by better availability of TCS for association sludge and that in the effluent of primary sedimentation unit of
to sludge in cold period. Nevertheless, the result for STP (II) was not three STPs (p < 0.05). Therefore, in order to facilitate the sample
consistent with the other two STPs and no significant difference was analysis procedure, determination of TCS fraction is suggested to be
observed between the elimination efficiencies in cold and warm performed in either one matrix and still enjoy the negligible biased
periods (p < 0.05). This could be attributed to the fact that sedimen- data.
tation of sludge in STP (II) is more a factor of coagulants applied
to the wastewater and is governed by the chemical-forced precip-
itation. Therefore, sedimentation is not affected by the temporal 4. Conclusion
variation. Since 2,4-DCP was detected throughout the STP (II) train
with the same influent variation trend, ANOVA test was applied TCS fate in wastewater and sludge treatment was investigated
(p < 0.05) to the relevant data set as well. The result confirmed the in three sewage treatment plants with different treatment prac-
hypothesis and no significant difference in 2,4-DCP elimination was tices. Adsorption to sludge and biodegradation were considered as
observed between cold and warm periods. The results of statistical two main processes for TCS elimination. However, transformation
tests are shown in Table 3. to toxic products (2,4-DCP and 2,8-DCDD) in chlorination and UV
In order to confirm the above results, complementary test on disinfection, and microbial formation of persistent product (MTCS)
correlation between elimination in primary sedimentation and in biological treatment raised a big concern.
concentration in primary sludge was performed for STP (I) and (III) Mass balance analysis in wastewater showed that around
using Origin 8 pro. The scatter plots in Fig. 4a and b are based on 13.1–30.3% of TCS was adsorbed to sludge while 59.6–76% of
the elimination efficiency in primary sedimentation and the con- TCS underwent biodegradation in secondary treatment. TCS also
centration of TCS in primary sludge in cold and warm periods. The showed considerable degradation during aerobic (34–56%) and
result showed correlation coefficients of R(28) = 0.80 and 0.76 for anaerobic (18.6–23.0%) sludge digestion process. Chlorination and
STP (I) and (III) (p < 0.05), respectively that indicates elevated con- UV irradiation could cause 9.9% and 13.0% TCS transformation to
centration of TCS in primary sludge could affect the sedimentation 2,4-DCP and 2,8-DCDD, respectively.
process. The partition behavior of target compounds was correlated to
Biological treatment process is not supposed to be much affected the TOC values of the dried sludge from three plants and it was
by temporal variation as operational conditions can be easily found that the sludge with higher TOC, exhibited higher partition
altered to control the stability of the biological system. However, of compound to the sludge.
paired sample student t-test on time classified elimination data Elevated concentration of TCS in the influent in cold period
showed that TCS elimination efficiencies in secondary treatment resulted in higher elimination efficiencies in primary sedimenta-
were significantly different (34.8% vs. 57% in cold and warm peri- tion and higher concentration in solid phase. Moreover, comparable
ods, respectively; p < 0.05). This observation might be explained by TCS mass fluxes in water phase of the sludge and the effluent of the
variation in biological activity of microorganism [38,39]. A confir- primary sedimentation unit implied that determination of TCS can
F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340 339

be performed in either one matrix. Although sorption to sludge and [11] M. Ricart, H. Guasch, M. Alberch, D. Barceló, C. Bonnineau, A. Geiszinger, M.I.
biodegradation are the major TCS removal processes, the treated Farré, J. Ferrer, F. Ricciardi, A.M. Romaní, S. Morin, L. Proia, L. Sala, D. Sureda, S.
Sabater, Triclosan persistence through wastewater treatment plants and its
wastewater and sludge are still as the main reservoir of TCS and its potential toxic effects on river biofilms, Aquat. Toxicol. 4 (2010
degradation products. This outlines the importance of wastewater ) 346–353.
and sludge management strategies. [12] L. Zorrilla, E. Gibson, S. Jeffay, K. Crofton, W. Setzer, R. Cooper, T. Stoker, The
effects of triclosan on puberty and thyroid hormones in male Wistar rats,
Toxicol. Sci. 107 (2009) 56–64.
Novelty statement [13] M.L. Ciusa, L. Furi, D. Knight, F. Decorosi, M. Fondi, C. Raggi, J.R. Coelho, L.
Aragones, L. Moce, P. Visa, A.T. Freitas, L. Baldassarri, R. Fani, C. Viti, G. Orefici,
J.L. Martinez, B. Consortium, A novel resistance mechanism to triclosan that
Although TCS occurrence has been reported vastly, its behavior suggests horizontal gene transfer and demonstrates a potential selective
during wastewater/sludge treatment is still mostly unknown such pressure for redduced biocide suseptibility in clinical strains of Staphylococcus
aureus, J. Antimicrob. Agents 40 (2012) 210–220.
as how each treatment may affect TCS elimination or transforma-
[14] B. Drury, J. Scott, E.J. Rosi-Marshall, J.J. Kelly, Triclosan exposure increases
tion in the full scale study. Quantitative data on the occurrence triclosan resistance and influences taxonomic composition of benthic
and fate of TCS in sewage sludge treatment are either scarce or not bacterial communities, Environ. Sci. Technol. 47 (2013)
investigated so far. To top, we showed that wastewater treatment 8923–8930.
[15] D.R. Orvos, D.J. Versteeg, J. Inauen, M. Capdevielle, A. Rothenstein, V.
may act as a point source of TCS transformation products which are Cunningham, Aquatic toxicity of triclosan, Environ. Toxicol. Chem. 21 (2002)
toxic/persistent. The findings of this study can be interesting for the 1338–1349.
readers of Journal of Hazardous Materials that has long history in [16] S.A. Raut, R.A. Angus, Triclosan has endocrine-disrupting effects in male
western mosquitofish, Gambusia affinis, Environ. Toxicol. Chem. 29 (2010)
organic pollutants in wastewater. 1287–1291.
[17] A.S. Stasinakis, G. Gatidoua, D. Mamaisc, N.S. Thomaidis, T.D. Lekkas,
Occurrence and fate of endocrine disrupters in Greek sewage treatment
plants, Water Res. 42 (2008) 1796–1804.
[18] D.C. McAvoy, B. Schatowitz, M. Jacob, A. Hauk, W.S. Eckhoff, Measurement of
This study was supported by the National Natural Science Foun- triclosan in wastewater treatment systems, Environ. Toxicol. Chem. 21 (2002)
dation of China (21175025).
[19] T.W. Federle, S.K. Kaiser, B.K. Nuck, Fate and effects of triclosan in activated
sludge, Environ. Toxicol. Chem. 21 (2002) 1330–1337.
Conflict of interest [20] J.M. Buth, M. Grandbois, P.J. Vikesland, K. McNeill, Aquatic photochemistry of
chlorinated triclosan derivatives: potential source of
polychlorodibenzo-p-dioxins, Environ. Toxicol. Chem. 28 (2009)
The authors declare that they have no conflict of interest. 2555–2563.
[21] D.E. Latch, J.L. Packer, W.A. Arnold, K. McNeill, Photochemical conversion of
triclosan to 2,8-dichlorodibenzop-dioxin in aqueous solution, Photochem.
Acknowledgement Photobiol. A: Chem. 158 (2003) 63–66.
[22] N. Lozano, C.P. Rice, R. Mark, A. Torrents, Fate of triclocarban, triclosan and
methyltriclosan during wastewater and biosolids treatment processes, Water
The authors would like to acknowledge the financial supports
Res. 47 (2013) 4519–4527.
from National Natural Science Foundation of China (21175025). [23] F. Samara, B. Gullett, R. Harrison, A. Chu, G. Clark, Determination of relative
assay response factors for toxic chlorinated and brominated dioxins/furans
using an enzyme immunoassay (EIA) and a chemically-activated luciferase
Appendix A. Supplementary data gene expression cell bioassay (CALUX), Environ. Int. 35 (2009)
Supplementary data associated with this article can be found, in [24] T. Leikera, S. Abneya, S. Goodbred, M. Rosenc, Identification of methyl
triclosan and halogenated analogues in male common carp (Cyprinus carpio)
the online version, at from Las Vegas Bay and semipermeable membrane devices from Las Vegas
034. Wash Nevada, Sci. Total Enviorn. 407 (2009) 2102–2114.
[25] F. Tohidi, Z. Cai, GC/MS analysis of triclosan and its degradation by-products
in wastewater and sludge samples from different treatments, Environ. Sci.
References Pollut. Res. Int. 22 (2015) 387–400.
[26] M.J. O’Neil, The Merck Index: An Encyclopedia of Chemicals, Drugs, and
[1] H. Singer, S. Muller, C. Tixier, L. Pillonel, Triclosan: occurrence and fate of a Biological, 15 ed., RCS, 2013.
widely used biocide in the aquatic environment: field measurements in [27] E.M. Fiss, K.L. Rule, P.J. Vikesland, Formation of chloroform and other
wastewater treatment, plants, surface waters, and lake sediments, Environ. chlorinated byproducts by chlorination of triclosan-containing antibacterial
Sci. Technol. 36 (2002) 4998–5004. products, Environ. Sci. Technol. 41 (2007) 2387–2394.
[2] T. Okumura, Y. Nishikawa, Gas chromatography-mass spectrometry [28] B.F. Pycke, A. Crabbe, W. Verstraete, N. Leys, Characterization of
determination of triclosans in water, sediment and fish samples via triclosan-resistant mutants reveals multiple antimicrobial resistance
methylation with diazomethane, Anal. Chim. Acta 325 (1996) 175–184. mechanisms in Rhodospirillum rubrum S1H, Appl. Environ. Microbiol. 76
[3] K. Bester, Fate of triclosan and triclosan-methyl in sewage treatment plants (2010) 3116–3123.
and surface waters, Arch. Environ. Contam. Toxicol. 49 (2005) 9–17. [29] V.G. Samaras, A.S. Stasinakis, N.S. Thomaidis, D. Mamais, T.D. Lekkas, Fate of
[4] J.M. Buth, P.O. Steen, C.H. Sueper, P.J. Vikesland, W.A. Arnold, Dioxin selected emerging micropollutants during mesophilic, thermophilic and
photoproducts of triclosan and its chlorinated derivatives in sediment cores, temperature co-phased anaerobic digestion of sewage sludge, Biores.
Environ. Sci. Technol. 44 (2010) 4545–4551. Technol. 162 (2014) 365–372.
[5] X. Chen, J.L. Nielsen, K. Furgal, Y. Liu, I.B. Lolas, Biodegradation of triclosan and [30] J.C. Duran-alvarez, B. Prado-Pano, B. Jiménez-Cisneros, Sorption and
formation of methyl-triclosan in activated sludge under aerobic conditions, desorption of carbamazepine, naproxen and triclosan in a soil irrigated with
Chemosphere 84 (2011) 452–456. raw wastewater: estimation of the sorption parameters by considering the
[6] M. Allmyr, G. Sandborgh-Englund, Triclosan in plasma and milk from Swedish initial mass of the compounds in the soil, Chemosphere 88 (2012) 84–90.
nursing mothers and their exposure via personal care products, J. Sci. Total [31] A. Karnjanapiboonwong, A.N. Morse, J.D. Maul, T.A. Anderson, Sorption of
Environ. 372 (2006) 87–93. estrogens, triclosan, and caffeine in a sandy loam and a silt loam soil, J. Soils
[7] A.M. Calafat, X.Y. Ye, L.Y. Wong, J.A. Reidy, L.L. Needham, Urinary Sediments 10 (2010) 1300–1307.
concentrations of triclosan in the U.S. population: 2003–2004, Environ. [32] N. Lozano, C.P. Rice, R. Mark, A. Torrents, Fate of triclosan and methyltriclosan
Health Perspect. 116 (2008) 303–307. in soil from biosolids application, Environ. Pollut. 160 (2012) 103–108.
[8] M.C. Balmer, T. Pioger, C. Droz, K. Romanin, P.A. Bergqvist, M.D. Muller, H.R. [33] G.G. Ying, R.S. Kookana, Triclosan in wastewaters and biosolids from
Buser, Occurrence of methyl triclosan, a transformation product of the Australian wastewater treatment plants, J. Environ. Int. 33 (2007) 199–205.
bactericide triclosan, in fish from various lakes in Switzerland, Environ. Sci. [34] N. Nakada, M. Yasojima, Y. Okayasu, Y. Suzuki, Mass balance analysis of
Technol 38 (2004) 390–395. triclosan diethyltoluamide, crotamiton and carbamazepine in sewage
[9] M.E. Delorenzo, J.M. Keller, C.D. Arthur, M.C. Finnegan, H.E. Harper, V.L. treatment plants, Water Sci. Technol. 61 (2010) 1739–1747.
Winder, D.L. Zdankiewicz, Toxicity of the antimicrobial compound triclosan [35] A.S. Stasinakis, N.S. Thomaidis, A.G. Asimakopoulos, V.G. Samaras, A. Ajibola,
and formation of the metabolite methyl-triclosan in estuarine systems, D. Mamais, T.D. Lekkas, Contribution of primary and secondary treatment on
Environ. Toxicol. 23 (2008) 224–232. the removal of benzothiazoles benzotriazoles, endocrine disruptors,
[10] A. Dann, A. Hontela, Triclosan: environmental exposure, toxicity and pharmaceuticals and perfluorinated compounds in a sewage treatment plant,
mechanisms of action, J. Appl. Toxicol. 31 (2011) 285–311. Sci. Total Environ. 463–464 (2013) 1067–1075.
340 F. Tohidi, Z. Cai / Journal of Hazardous Materials 323 (2017) 329–340

[36] E. Manoli, C. Samara, The removal of polycyclic aromatic hydrocarbons in the [38] E.C. Collins, The effect of temperature control on biological wastewater
wastewater treatment process: experimental calculations and model treatment processes, Water Res. 12 (1978) 547–554.
predictions, Environ. Pollut. 151 (2008) 477–485. [39] K.L. Wang, N.K. Shammas, Advanced biological treatment processes, in:
[37] H. Melcer, P. Steel, W. Bedford, Removal of polycyclic aromatic hydrocarbons Handbook of Environmrntal Engineering, 2009 ed., Humana Press, New York,
and heterocyclic nitrogen compounds in municipal treatment plants, Water 2009.
Environ. Res. 67 (1995) 926–934.