Water Research 186 (2020) 116315

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Impact of biological wastewater treatment on the reactivity of

N-Nitrosodimethylamine precursors
Xiaolu Zhang, Daekyun Kim, David L. Freedman, Tanju Karanfil∗
Department of Environmental Engineering and Earth Sciences, Clemson University, Anderson, SC 29625, USA

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

Article history: N-Nitrosodimethylamine (NDMA) is a probable human carcinogen which forms during chloramination of
Received 2 March 2020 wastewater-impacted drinking waters. Municipal wastewater effluents are considered as major sources of
Revised 7 July 2020
NDMA precursors affecting downstream water quality. To evaluate the deactivation mechanisms and effi-
Accepted 17 August 2020
ciencies of NDMA precursors during secondary treatment with the activated sludge (AS) process, NDMA
Available online 18 August 2020
formation potentials (FPs) of selected model precursor compounds and sewage components (i.e., black-
Keywords: waters and greywaters) were monitored in batch AS treatment tests. After 24-h incubation with four dif-
N-Nitrosodimethylamine ferent types of AS (i.e., domestic rural, domestic urban, textile and lab-grown AS), NDMA FP of trimethy-
Disinfection by-products lamine (TMA) and minocycline (MNCL) decreased by 77-100%, while there was only 29-46% reduction in
Biological treatment NDMA FP of sumatriptan (SMTR). The reduction in NDMA FP associated with ranitidine (RNTD) varied
Activated sludge between 34% and 87%. The decrease in NDMA FP of RNTD depended on the AS type, hydraulic retention
Trace organic contaminants
time (HRT) and solids retention time (SRT). The domestic AS (rural and urban) achieved higher decreases
in NDMA FPs of the tested model precursors than the textile AS or lab-grown AS. Increasing the HRT
or SRT enhanced NDMA FP decrease for RNTD. Among different processes tested (i.e., biodegradation,
biosorption and volatilization), biosorption was the major mechanism responsible for the NDMA FP de-
crease of RNTD, MNCL and SMTR, while biodegradation was the major NDMA FP reduction mechanism
for TMA. The reduction in NDMA FP of RNTD via biodegradation depended on the AS activity which may
vary with sampling seasons and SRT. NDMA FPs in all tested sewage components (i.e., blackwaters and
greywaters) decreased after 24-h AS treatment. Urine in blackwater was the predominant (i.e., >90%)
contributor to NDMA FP in domestic sewage and AS-treated effluents.
© 2020 Elsevier Ltd. All rights reserved.

1. Introduction in the United States Environmental Protection Agency (US EPA)

N-Nitrosodimethylamine (NDMA) is a carcinogenic disinfection
by-product which can form from the reactions between disinfec-
tants (e.g., chloramines, ozone) and nitrogen-containing organic
precursors (US EPA, 2001; Mitch et al., 2003; Chuang et al.,
2019). NDMA precursors can originate from wastewater effluent or-
ganic matters, algal organic matters, pharmaceuticals and personal
care products (PPCPs), and amine-based water treatment poly-
mers (e.g., polydiallyldimethylammonium chloride (polyDADMAC),
polyamines) (Krasner et al., 2013; Zeng et al., 2016; An et al., 2019;
Beita-Sandi et al., 2019). During chloramination, the dimethylamine
(DMA) group from NDMA precursors can be oxidized to NDMA
via nitrosation reaction, following a nucleophilic substitution on
the DMA group by chloramine (Liu et al., 2014; Zhang et al.,
2020). Due to its adverse health effects, NDMA has been included
∗
Corresponding author.
E-mail address: tkaranf@clemson.edu (T. Karanfil).
Contaminant Candidate List 4 for possible regulation in the fu-
ture (US EPA, 2016). Other countries and regions, such as Canada,
Australia and the European Union, have established guidance or
guidelines to address the health issues related to NDMA in drink-
ing water and wastewater for direct or indirect potable reuse
(UK DWI, 20 0 0; Health Canada, 2011;NHMRC and NRMMC, 2011).
Municipal wastewater discharges are considered as one of the
major sources of NDMA precursors that may impact downstream
drinking water quality (Krasner et al., 2013; Sgroi et al., 2018).
NDMA precursors from wastewater can be deactivated to varying
degrees during secondary biological treatment (e.g., the activated
sludge (AS) process) at wastewater treatment plants (WWTPs). Pre-
cursors such as dimethylamine (DMA) and trimethylamine (TMA)
were shown to be readily biodegradable during the AS pro-
cess, with >75% and >71% removal, respectively (Mitch and Sed-
lak, 2004; Sedlak et al., 2005; Wang et al., 2014). Other precursors
such as dimethylformamide and dimethylformamide were removed
to a lesser extent by AS (i.e., 54-68% and 57-72%, respectively)

https://doi.org/10.1016/j.watres.2020.116315
0043-1354/© 2020 Elsevier Ltd. All rights reserved.
2 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315

(Wang et al., 2014). Reported removal efficiencies for ranitidine molar basis (Le Roux et al., 2011; Selbes et al., 2013; Shen and
(RNTD) during the AS process varied from <0% to 95% at different Andrews, 2013b). RNTD, MNCL, and SMTR are the main ac-
WWTPs (Castiglioni et al., 2006; Gros et al., 2006; Sedlak and Ka- tive ingredients of the most widely sold and prescribed amine-
vanaugh, 2006; Radjenovic et al., 2007; Guerra et al., 2014). Overall based pharmaceuticals in the US in recent years (Shen and An-
NDMA precursor levels, which were measured by NDMA formation drews, 2013b; Zeng and Mitch, 2016). TMA is a tertiary amine com-
potentials (FPs), were reduced by 10-98% during the AS process monly present (i.e., 0-2.7 mmol/mol creatinine) in human urine
(Mitch and Sedlak, 2004; Sedlak et al., 2005; Wang et al., 2014). and feces (Svensson et al., 1994; Mitch and Sedlak, 2004; Lee et al.,
However, NDMA FPs were also found to increase after the AS treat- 2010). RNTD, MNCL, and SMTR were purchased from TCI (Duncan,
ment under certain circumstances (Sedlak et al., 2005; Yu et al., US) as solids, which were then dissolved in methanol (0.4 mM)
2012). and further diluted in deionized and distilled water (DDW) to 20
Little is currently known about the conditions that impact re- μM (RNTD) or 200 μM (MNCL and SMTR). An aqueous solution of
moval efficiencies for NDMA precursors during the AS process. TMA (4% mass concentration) was purchased from Sigma-Aldrich
Many known NDMA precursors such as PPCPs are xenobiotics, and (St. Louis, US) which was further diluted in DDW to 200 μM. The
they are often biologically recalcitrant (Radjenovic et al., 2007; chemical structures and key physiochemical properties of selected
Sgroi et al., 2018). The characteristics of an AS microbial commu- model precursor compounds are shown in Table S1.
nity (e.g., bacterial speciation and population) and operating con-
ditions (e.g., hydraulic retention time (HRT), solids retention time
2.2. Sewage components
(SRT)) can be important factors affecting biodegradation of per-
sistent NDMA precursors, as reported for other trace-level syn-
Urine and feces of blackwaters, and greywaters including laun-
thetic organic contaminants in wastewaters (Ahmed et al., 2007;
dry, shower, bathroom washbasin, and kitchen greywaters were se-
Zhang et al., 2009; Gros et al., 2010; Jin et al., 2010; Li and
lected as representative components of domestic sewage. Raw hu-
Zhang, 2010; Sarma and Joshi, 2015; Liu et al., 2017).
man urine and feces samples were collected from a volunteer be-
NDMA precursors can be removed via biotic (i.e., biodegra-
fore (U and F, pre-RNTD) and after (UR and FR, post-RNTD) taking a
dation) and/or abiotic processes (i.e., biosorption, volatilization)
Zantac 150 tablet which contains 150 mg of RNTD as an active in-
during AS treatment (Byrns, 2001). At near neutral pH, tertiary
gredient. Laundry greywater samples were collected from washing
amine NDMA precursors are positively charged (Shen and An-
machine discharges after a medium-size load of white and colored
drews, 2013a), and their biosorption onto negatively charged AS
clothes were washed using only detergent (LD) and detergent plus
solid particles could be important. RNTD and tetracycline tend to
fabric softener (LF), respectively. More details about collecting and
adsorb onto AS particles rather than being biodegraded (Li and
processing the urine, feces and laundry greywater samples used in
Zhang, 2010; Jelic et al., 2011). Precursors like DMA and TMA,
this study are described elsewhere (Zeng and Mitch, 2015).
however, are more readily biodegradable (Mitch and Sedlak, 2004;
Shower greywater samples were collected from a bathtub af-
Sedlak et al., 2005; Wang et al., 2014). Significant spatial and tem-
ter a volunteer took a hot shower not using any personal care
poral variations in wastewater influents and WWTP operational
products (S), using shampoo only (SS), or using body wash only
parameters leading to varying AS properties (Yuan and Black-
(SB). A sample of bathroom washbasin greywater (W) was col-
all, 2002; Zhao et al., 2006; Grady, 2011) may also impact the roles
lected by grabbing a mixture of handwashing, tooth-brushing, and
of biotic and abiotic processes in the removal of NDMA precursors.
face-cleaning wastewaters after the volunteer conducted a routine
Wastewater influents contain human urine and feces, laundry,
morning wash, with each step of activities using a single brand of
shower, washbasin and kitchen greywaters as major constituents
personal care product. Kitchen greywater samples were collected
(Friedler et al., 2013; Zeng and Mitch, 2015). Trace-level pharma-
from a stoppered kitchen sink after a manual washing of clean
ceuticals and their metabolites are also present through human ex-
dishes using a single brand of detergent (KD), and after food mate-
cretion and drug disposal (Mitch and Sedlak, 2004). Among these
rials (i.e., raw and cooked vegetables, grains, meats and seafood,
sewage components, laundry greywater was found to be the most
each ~50 g/L) were soaked (for 30 min) and boiled (for 2 min)
significant contributor to NDMA precursor pools when NDMA for-
in tap water, respectively. The soaking tap water and boiled soup
mation was measured under the uniform chloramination condition
were then mixed in a 1:4 volumetric ratio (KF) to mimic an ordi-
(i.e., 2.5 mg Cl2 /L and 3-d reaction time) (Zeng and Mitch, 2015).
nary diet consisting of raw and cooked food. All collected blackwa-
The occurrence of NDMA precursors (measured by FPs) from differ-
ters and greywaters were filtered (0.45-μm, Whatman, GE Health-
ent sewage components and their removal during the AS process
care Life Sciences, US) before being stored or used. Urine and feces
remain largely unexplored.
blackwaters, and KF greywater were stored at -20°C, while other
The major objectives of this study were to (i) investigate the re-
greywaters were stored at 4°C until use. The selected water quality
duction of NDMA FPs from selected model NDMA precursors and
parameters of filtered blackwater and greywater samples (Table S2)
different sewage components during batch AS treatment tests, (ii)
were measured based on Standard Methods (APHA et al., 2005).
examine the removal mechanisms (i.e., biodegradation, biosorp-
The dissolved organic carbon (DOC) in all tested samples ranged
tion, and volatilization) for NDMA precursors, (iii) evaluate the ef-
from 2 to 78 mg/L, while the ammonia nitrogen (NH3 -N) concen-
fects of AS types (i.e., domestic rural, domestic urban, textile and
trations were <1.6 mg/L, except that laundry greywaters showed
lab-grown AS) and selected treatment conditions (i.e., HRT, SRT,
20-41 mg/L NH3 -N. The specific ultraviolet absorbance at 254 nm
seasonal variations in AS activity) on the reduction of NDMA FP,
(SUVA254 ) was generally low (i.e., 0.3-1.5 L/mg-m) in all samples.
and (iv) assess the major sources of NDMA precursors present in
All experimental protocols testing human urine and feces samples
wastewater effluents.
were approved by Clemson University Institutional Biosafety Com-
mittee (IBC).
2. Materials and methods

2.1. Model NDMA precursors 2.3. Batch AS treatment tests (biotic and abiotic)

Three amine-based pharmaceuticals (i.e., RNTD, minocycline A modified version of EPA test method OPPTS 835.3280 -314B
(MNCL) and sumatriptan (SMTR)) and TMA were selected as model (US EPA, 2008) was employed to assess decreases in NDMA FP.
NDMA precursors, with ~2-90% of their NDMA FP yields on a This method involves batch incubation of the target compounds or
 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315 3

sewage components with AS. Four treatment scenarios were eval-

uated: (i) overall decreases in NDMA FP, via biotic (biodegradation)

Industrial impactc
plus abiotic (biosorption and volatilization) processes, (ii) biosorp-
tion only, (iii) volatilization only, and (iv) biodegradation only. The
biodegradation process was separately assessed with a modified
respirometric method (OECD, 2003). NDMA FPs were monitored

None
100%
25%
before and after batch tests to evaluate the removal (or deactiva-

1%
tion) in NDMA precursors.
Four types of AS were collected from aeration basins to treat

Nitrogen and phosphorus removal

NDMA precursors during the batch tests: (i) a rural municipal
WWTP (AS 1), (ii) an urban municipal WWTP (AS 2), (iii) a tex-
tile WWTP (AS 3), and (iv) a sequencing batch reactor (SBR, AS 4)
operated in the laboratory using a synthetic wastewater contain-

Partial nitrification
None (N is added)
Nutrients removal
ing glucose and yeast extract as sources of biochemical oxygen de-
mand. AS 4 was acclimated to the synthetic wastewater for at least

Nitrification
three SRTs (i.e., 24 d for an 8-d SRT) in the lab-scale SBR before be-
ing collected. Key information for the WWTPs and lab-scale SBR is
summarized in Table 1. The composition of the synthetic wastew-
ater and detailed operating conditions for the SBRs are provided in
Table S3 and Table S4.

Anaerobic-anoxic-oxic
Samples of AS 1, AS 2 and AS 3 mixed liquor were transported

Extended aeration

Extended aeration
Treatment process
to the lab within 1 h of collection, and then aerated for no more
than 12 h at 23±2 °C for preconditioning. Prior to the batch tests,
AS liquor was centrifuged (20 0 0 × g, 5 min), the supernatant was
decanted, and the solids were resuspended in a mineral solution

Oxic
(Table S5, pH = 7.4 buffered with 4 mM phosphate; OECD, 2003) to
a concentration of ~60 0 0 mg/L. The same procedure was repeated

SRTb (d)
three times to remove background NDMA precursors from the AS
liquor. The washed AS biomass was then resuspended in the min-

20
12
26
8g
eral solution to a concentration of ~30 0 0 mg/L. AS liquor (50 0 mL,
washed and resuspended) was transferred to a 1-L incubation bot-

HRTa (h)
tle, dosed with a predetermined amount of precursor solution (i.e.,

22-24
13.5
with a target NDMA FP of ~10 0 0 ng/L for each compound) under

5
6
well mixing conditions, and then aerated for 6 or 24 h at 25±2
°C. The air flow rate (i.e., 150 L/m3 •min) was set as close as possi-
Treatment capacity

ble to the rates used in practice (i.e., 20-90 L/m3 •min) in aeration
basins at WWTPs (Gray, 2004). At the end of incubation (i.e., after
6 or 24 h), 125-mL samples were harvested and filtered (0.45-μm).
d

1.7 mgd
2.0 mgd
70 mgd

12 L/d

Nominal SRT without considering the loss of suspended solids via discharging effluents.
The filtrate was then used in the NDMA FP test.
To monitor the reduction in NDMA FP via biosorption alone,
sodium azide (NaN3 , 5 g/g mixed liquor suspended solids (MLSS))
Domestic wastewater and industrial discharge

was added to the washed and resuspended AS liquor under gen- Contains glucose and yeast extract as sources of biological oxygen demand.
tle mixing conditions. After 24-h contact time, NaN3 was removed
by washing the AS biomass four times in the mineral solution (i.e.,
Predominantly domestic wastewater

60 0 0 mg/L). The washed AS biomass was then resuspended in the

mineral solution with the MLSS adjusted to ~30 0 0 mg/L. To con-
The volume fraction of industrial discharge in WWTP influent.

firm the effectiveness of NaN3 deactivation, oxygen (O2 ) uptake

was monitored using a respirometer (Oxymax ER-10, Columbus In-
Key information of selected wastewater treatment facilities.

Synthetic wastewaterf

struments, US) throughout the 6-h (occasionally 10-h) biosorption

Textile wastewater

test. A negligible O2 uptake was presumed to be an adequate met-

ric for a lack of microbial activity (Fig. S1).
Influent type

Reduction in NDMA FP via biodegradation alone was examined

by employing a low concentration of MLSS (~200 mg/L) and pre-
biosorbed biomass (i.e., 30-min pre-contact of AS biomass with
NDMA precursors; more details shown in Supplement Information)
to minimize interferences from the biosorption process. The incu-
Lab-scale SBRe

Sequencing batch reactor.

bation time for the biodegradation tests was extended up to 10 d

Hydraulic retention time.

Million gallons per day.

to obtain measurable reduction in NDMA FPs. Reduction in NDMA

Solids retention time.
WWTP 1
WWTP 2
WWTP 3
Facility

FP via volatilization was evaluated by dosing NDMA precursors

into the mineral solution and aerating for 6 h in the absence of
any AS biomass. NDMA FPs were measured before and after 6-h
AS sample

incubation.
Table 1

1
2
3
4
AS
AS
AS
AS

d
b

g
e
a

f
4 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315
2.4. NDMA FP test
For the NDMA FP test, samples were buffered with 20-mM
phosphate (pH 7.8±0.2) and then chloraminated by adding a
pre-determined volume of monochloramine stock solution (~20 0 0
mg Cl2 /L) to achieve the target dosage (i.e., 100 mg Cl2 /L). The
monochloramine stock solution was freshly prepared by adding a
sodium hypochlorite solution (NaClO, ~40 0 0 mg Cl2 /L) to an am-
monium chloride solution (NH4 Cl, ~10 0 0 mg N/L) drop by drop at
pH 9 with a Cl:N mass ratio of 4:1. The chloraminated samples
were then incubated in the dark at 23±2°C for 5 d. The residual
chloramines were then quenched by adding an excess amount of
sodium thiosulfate (Na2 S2 O3 ). The quenched samples were then
extracted and analyzed for NDMA according to US EPA Method
521 (US EPA, 2004). Details on the NDMA extraction and anal-
ysis method are described elsewhere (Selbes et al., 2013; Beita-
Sandi et al., 2019). The minimum reporting level for NDMA is 2.0
ng/L.
3. Results and discussion
3.1. Reduction in NDMA FPs of model precursors
NDMA FP of RNTD was reduced by 34-87% after 24-h incu-
bation with four different types of AS (Fig. 1). NDMA FPs of
TMA and MNCL were reduced by 92-98% and 77-97%, respec-
tively, while NDMA FP of SMTR decreased to a lesser extent (i.e.,
29-46% depending on the types of AS). These overall reductions
were achieved via both biotic (i.e., biodegradation) and abiotic (i.e.,
biosorption, volatilization) processes. TMA is known to be read-
ily biodegradable, with >71% removals previously reported dur-
ing biological treatment (Mitch and Sedlak, 2004; Wang et al.,
2014). MNCL contains hydroxyl functional groups attached to ring
structures, which can enhance its biodegradability (Okey and
Stensel, 1996). Unlike TMA or MNCL, RNTD contains a nitryl (-
NO2 ) functional group, and SMTR contains a sulfonamide (-NH-
SO2 -) functional group, both of which may impede biodegradation
(Okey and Stensel, 1996).
In addition to the structures of NDMA precursors, the sources
of AS biomass also affected the overall removal of NDMA FPs. AS
1 was collected from a rural municipal WWTP with an extended
aeration process (i.e., 22-24 h HRT) and a high level of nitrifica-
tion (i.e., undetectable ammonia in the secondary effluent). With
this source of AS, there was an 87% decrease in NDMA FP of RNTD
after 24-h incubation. AS 2 was collected from an urban munic-
100 AS 1
Reduction in NDMA FP
AS 2
80 AS 3
AS 4
60
(%)
40
20
0 4. For nominal SRTs of 4, 8, and 12 d, NDMA FP reductions were
RNTD TMA MNCL SMTR
Fig. 1. Reductions in NDMA FPs of model precursor compounds during batch AS
treatment tests. Bar graph represents the average reduction in NDMA FP after 24-h
treatment with each AS collected during three different seasons (i.e., winter, spring
and summer). Error bar represents the standard deviation for three batch tests with
different AS. AS 4 was collected only in spring, and the error bar represents stan-
dard deviation for duplicate tests. RNTD: ranitidine, TMA: trimethylamine, MNCL:
minocycline, SMTR: sumatriptan. Initial NDMA FP ~10 0 0 ng/L from each compound.
ipal WWTP with biological nutrient removal process (14 h HRT);
this source of AS reduced the NDMA FP of RNTD by 59%. Further
lower reduction in NDMA FP of RNTD was found after 24-h in-
cubation with the textile (i.e., 41%) and lab-grown AS (i.e., 34%).
Both AS 1 and AS 2 exhibited relatively high reductions in NDMA
FPs of the three amine-based pharmaceuticals (i.e., RNTD, MNCL
and SMTR), while the lab-grown AS (i.e., AS 4) showed the lowest
reduction. Previous studies have shown that municipal AS has di-
verse microbial communities due to the complexity of organic con-
taminants (including PPCPs) present in municipal wastewater in-
fluents, while AS fed with simple nutrients (such as AS 4 fed with
a synthetic wastewater and AS 3 fed with textile wastewater) likely
has less diverse microbial communities (Zhao et al., 2006). The di-
versity and richness of microbial species likely favors removal of
xenobiotic compounds (Kraigher et al., 2008; Chonova et al., 2016;
Kim et al., 2017).
Among the four types of AS tested, AS 3 collected from a tex-
tile WWTP exhibited the largest seasonal variations in the reduc-
tion of some NDMA precursors (e.g., RNTD and SMTR; Table S6).
The decreases in NDMA FPs of RNTD and SMTR varied by 36%
and 23%, respectively, after 24-h incubation with AS 3 collected
in three different seasons, while reductions in NDMA FPs of TMA
and MNCL varied by 9-12%. In contrast, <10% variations in NDMA
FP reduction were observed after 24-h incubation with municipal
WWTP AS (i.e., AS 1 or AS 2) collected during three seasons. In the
textile WWTP, seasonal changes in the manufacturing processes
may result in fluctuations in the wastewater influent components,
and thus the AS 3 properties (i.e., color, MLSS, settleability). How-
ever, the operating conditions for the two municipal WWTPs (i.e.,
WW TP 1 and WW TP 2) were consistent during our sampling cam-
paigns. Occasionally, WWTP 1 was influenced from rain events,
producing increased influent flow rates. During one such event, air
supply to the aeration tank was shut off and the MLSS in AS 1
dropped to ~10 0 0 mg/L. AS 1 collected during such rain events
(~10 0 0 mg/L MLSS) showed 12%-34% lower levels of decreases in
NDMA FPs for all tested model precursor compounds, compared
to those measured under routine operating conditions with ~30 0 0
mg/L MLSS (Fig. S2).
The effects of selected operational parameters (i.e., HRT and
SRT) on deactivations of model precursor compounds were also ex-
amined. When the duration of the batch tests (i.e., simulating HRT)
was increased from 6 to 24 h, the reduction in NDMA FP of RNTD
increased from 81% to 88% for AS 1, from 34% to 54% for AS 2, or
from 32% to 62% for AS 3 (Fig. 2). These observations are consis-
tent with previous studies showing a significant (p<0.05) positive
correlation between removal of RNTD and the HRTs of the AS pro-
cess at different WWTPs (Fig. S3). In contrast, an increase in HRT
did not affect the extent of NDMA FP reduction for TMA, MNCL,
or SMTR. The reduction in NDMA FPs for TMA and MNCL were
nearly 100% after 6 h of incubation, and further incubation thus
has no impact. Increasing HRT did not impact the extent of NDMA
FP reduction for SMTR, which is relatively bio-refractory. On the
other hand, RNTD is considered to be moderately biodegradable
(e.g., with average removal efficiencies between 40% and 70% af-
ter 24-h AS treatment at WWTPs; Gros et al., 2010). Consequently,
removal of its NDMA FP was more affected by HRT.
The effect of SRT was examined for RNTD using lab-grown AS
26%, 30%, and 38%, respectively. Increases in SRT are expected to
increase diversity in the AS microbial community by retaining mi-
croorganisms with slower growth rates (e.g., nitrifiers) (Jelic et al.,
2011). This may enhance deactivation of compounds with a mod-
erate level of biodegradability such as RNTD (Jelic et al., 2011). A
longer SRT may also favor biosorption of polar compounds onto AS
particles (Petrie et al., 2014) such as RNTD.
 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315 5

100 a t=6h 100 b

Reduction in NDMA FP

Reduction in NDMA FP
80 t = 24 h 80

60 60

(%)

(%)
40 40

20 20

0 0
AS 1 AS 2 AS 3 AS 1 AS 2 AS 3

100 c 100 d
Reduction in NDMA FP

Reduction in NDMA FP
80 80

60 60
(%)

(%)
40 40

20 20
0 0
AS 1 AS 2 AS 3 AS 1 AS 2 AS 3
Fig. 2. Effects of incubation time (i.e., simulating HRT) on the reductions in NDMA FPs of (a) RNTD, (b) TMA, (c) MNCL, and (d) SMTR. Bar graph represents the average
NDMA FP reduction from two batch tests with AS collected in spring and summer, respectively. Error bars represent the standard deviation from the two batch tests.

3.2. Biosorption of model NDMA precursors
The contribution of biodegradation to overall NDMA FP reduc-
tion was defined as the difference between the overall reduction
and the reduction via biosorption plus volatilization. Reductions
due to biosorption (based on inhibition of respiration with NaN3 )
and volatilization (in the absence of AS) were measured separately.
The reduction in NDMA FP of RNTD occurred mainly via biosorp-
tion, while biodegradation contributed less than 25% to the over-
all reduction during 6-h AS treatment (Fig. 3). Similarly, biodegra-
dation accounted for only <1% and 2-11% of NDMA FP reduction
for MNCL and SMTR, respectively, when treated with the three
WWTP AS (i.e., AS 1, AS 2 and AS 3). In contrast, the NDMA FP of
TMA was mainly decreased via biodegradation (i.e., 74-98%) during
treatment with AS 2, AS 3 and AS 4, although nearly 100% of its
NDMA FP decrease was observed via biosorption with AS 1. The
lower biosorption of TMA as compared to the other model precur-
sors is probably due to its relatively higher water solubility (i.e.,

100 a 100
Reduction in NDMA FP

Reduction via biodegradation b

Reduction in NDMA FP

Reduction via biosorption

80 80

60
(%)

60
(%)

40 40

20 20

0 0
AS 1 AS 2 AS 3 AS 4 AS 1 AS 2 AS 3 AS 4

100 c 100
Reduction in NDMA FP

Reduction in NDMA FP

d
80 80

60 60
(%)
(%)

40 40

20 20

0 0
AS 1 AS 2 AS 3 AS 4 AS 1 AS 2 AS 3 AS 4
Fig. 3. Contributions of biosorption and biodegradation to deactivations of (a) RNTD, (b) TMA, (c) MNCL, and (d) SMTR during batch AS treatment tests. HRT = 6 h. Bar
graph hereafter represents the average NDMA FP reduction from duplicate batch AS treatment tests, error bars hereafter represent the standard deviations for duplicate tests.
6 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315
100 a t=2h
Reduction in NDMA FP
t=6h
80
t = 10 h
60
(%)
40
20
0 0
AS 1 AS 2 AS 3'
100 c 100
Reduction in NDMA FP
80
60
(%)
40
20
0 0
AS 1 AS 2 AS 3'
Fig. 4. Reductions in NDMA FPs of (a) RNTD, (b) TMA, (c) MNCL, and (d) SMTR during 2-h, 6-h and 10-h biosorption tests. AS 3’ was collected from the textile WWTP after
a new manufacturing process was launched and AS characteristics were substantially altered.
410 g/L, compared to 0.025-52 g/L of the other compounds). For
all tested precursor compounds, volatilization during 6-h incuba-
tion in the absence of any AS biomass had a negligible impact on
NDMA FPs (Fig. S4). These results indicate that biosorption was an
important removal mechanism for the tested NDMA precursors, es-
pecially for the high-molecular-weight compounds, while the effect
of volatilization was negligible.
Reduction in NDMA FP via biosorption also depended on the
sources of AS. AS 1 exhibited higher uptakes for RNTD and TMA,
while AS 4 exhibited the lowest level of biosorption of TMA, MNCL
and SMTR. The physiochemical properties of AS, such as its set-
tleability (e.g., sludge volume index), the amounts and compo-
sition of extracellular polymeric substances (EPS), and the pres-
ence of floc-forming microbes may affect the surface charges of
AS (Sponza, 2002; Wilen et al., 2003, 2008) and thus the adsorp-
tion of polar compounds onto AS particles (Steiner et al., 1976).
In this study, the inconsistent biosorption efficiencies among the
four AS were likely because of their distinct properties or char-
acteristics. The influent compositions and operating conditions at
WWTPs (or lab-scale SBR for AS 4) can both impact AS properties
(Sponza, 2002; Ahn et al., 2013; Zhang et al., 2015).
The solid-water distribution coefficients (Kd ) of precursor com-
pounds biosorbed onto different AS samples were calculated based
on NDMA FPs measured in both aqueous and solid phases (Table
S7). The log Kd values of the tested model precursors biosorbed
onto different sources of AS were at least twofold higher than log
Kow . This may suggest that adsorption of precursor compounds
onto AS particles was governed by factors other than hydrophobic
interactions, such as electrostatic interactions (Golet et al., 2003).
Under the near neutral pH condition of the biosorption experi-
ments (pH = 7.4±0.2), DMA functional groups (i.e., -N(CH3 )2 ) of
precursor compounds are positively charged and tend to bind to
negatively charged carboxyl and/or hydroxyl groups on AS surfaces
(Steiner et al., 1976; Shen and Andrews, 2013b).
Biosorption of xenobiotic compounds is typically considered to
be a rapid process, e.g., completed within hours or even minutes
b
Reduction in NDMA FP
100
80
(%)
60
40
20
AS 1 AS 2 AS 3'
d
Reduction in NDMA FP
80
60
(%)
40
20
AS 1 AS 2 AS 3'
(Savci, 2013; Vasiliadou et al., 2013; Martinez-Hernandez et al.,
2016). To examine the biosorption kinetics, NDMA FPs of model
precursor compounds were monitored up to 10 h during the
biosorption tests with NaN3 -deactivated AS liquor (~30 0 0 mg/L
MLSS). The two municipal AS (AS 1 and AS 2), and a new sample
of AS 3 (designated AS 3’ in Fig. 4) were employed for biosorption
kinetics tests. AS 3’ was collected from the textile WWTP after a
new manufacturing process was launched, which resulted in sub-
stantially altered AS properties (e.g., lighter color, increased MLSS,
and enhanced settleability). AS 3’ exhibited 2-56% more reduction
of NDMA FP via biosorption than AS 3 after 6-h incubation (Fig.
S5). In general, increasing the contact time from 2 to 10 h during
the biosorption tests had small impacts on the NDMA FP removals
(i.e., <18% differences) for RNTD, MNCL and SMTR despite the dif-
ferent types of AS (i.e., AS 1, AS 2 and AS 3’). The only exception
was TMA, for which NDMA FP reductions were 33-41% higher after
6 or 10-h incubation compared to 2-h incubation with AS 2 and
AS 3’. The increased reduction of NDMA FP may have been due
to a recovery in biodegradation activity, rather than biosorption.
As the incubation duration increased from 2 to 10 h, the oxygen
uptake rates by NaN3 -deactivated AS slightly increased (Fig. S1),
suggesting the gradually recovering microbial activities during the
biosorption tests. Overall, these results indicate that reductions in
NDMA FP for the model precursors via biosorption were completed
in less time (i.e., 2 h) than the lower end of HRTs at most WWTPs
(Gray, 2004).
3.3. Biodegradation of model NDMA precursors
Decreases in NDMA FP via biodegradation were investigated for
RNTD (100 nM) and SMTR (200 nM) using AS 1 (collected dur-
ing the spring and summer) and AS 4 (at 8 and 24-d SRTs) with a
MLSS concentration of ~150-200 mg/L. After 10 d of incubation, the
decrease in NDMA FP for RNTD was highest for AS 1 collected in
summer, followed by AS 1 collected in spring, AS 4 at an 8 d SRT,
and AS 4 at a 24 d SRT (Fig. 5). These results suggest that NDMA FP
 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315
1.2 AS 1 (spring)
[NDMA FP]/[NDMA FP]0 AS 1 (summer)
1.0 AS 4 (8-d SRT)
AS 4 (24-d SRT)
0.8
0.6
0.4
0.2
0 2 4 6 8 10
Incubation time (d)
Fig. 5. . Biodegradation curves of NDMA FP from RNTD during 10-d incubation with
AS 1 and AS 4. Initial RNTD concentration = 100 nM, MLSS ≈ 200 mg/L.
reduction for RNTD via biodegradation depended on seasonal vari-
ations in AS and SRT. AS 1 had higher removals of organic carbon
and ammonia in summer than in spring (Table S8). On the other
hand, the MLSS in AS 4 with a shorter SRT may have had more
active biomass than that with a longer SRT, and thus exhibited a
better reduction in NDMA FP for RNTD via biodegradation. As SRT
decreases, AS biomass may contain more abundant EPS which has
been shown to favor the initial uptake and subsequent biodegrada-
tion of compounds (Al-Halbouni, et al., 2008; Xu et al., 2013). Ex-
tending the incubation time of the assay to 10 d enhanced NDMA
FP reduction for RNTD to 52-71% and 34-44% with AS 1 and AS 4,
respectively, while only to 7-29% for SMTR.
The decrease in NDMA FP of RNTD via biodegradation followed
pseudo-first-order kinetics (R2 >0.86, Fig. S6) during the 10-d incu-
bation with AS 4 (8 and 24-d SRT) and AS 1 collected in spring.
With the ratios of MLSS to RNTD concentration (i.e., MLSS/[RNTD])
increasing from 0.5 to 1.5, 3.0 and 6.0 mg/μg, the pseudo-first-
order biodegradation rate constants increased from 0.067 to 0.146,
0.301 and 0.398 d−1 , respectively, during incubation with AS 4 at
an 8 d SRT. For AS 4 at a 24 d SRT, the rate constants increased
from 0.043 to 0.059, 0.230 and 0.355 d−1 , respectively (Fig S6).
The slower biodegradation rate for a lower MLSS/[RNTD] was likely
because of the toxicity of RNTD to microorganisms in AS. A high
concentration of RNTD (e.g., 2-5 mg/L) has been found to inhibit
microbial activities of AS and thus reduce its biodegradation rates
(Carucci et al., 2006).
3.4. Deactivation of NDMA precursors from sewage components
Among the tested sewage components, urine blackwaters ex-
hibited the highest NDMA FP, i.e., 11338±1958 ng/L in urine prior
to ingestion of RNTD (U), 7077±1414 ng/L in urine following inges-
tion of RNTD (UR). This result occurred despite a 250-fold dilution
of the urine in dechlorinated tap water to mimic toilet flushing
and maintain the NDMA FP in a proper range (i.e., not exceeding
tens of thousands ng/L) for GC/MS/MS analysis (Table S2). Kitchen
greywater containing food ingredients (KF) also exhibited relatively
high NDMA FP (i.e., 11001±3696 ng/L) despite a 100-fold dilution
in dechlorinated tap water. NDMA FP in other sewage components
were generally less than 30 0 0 ng/L. Although RNTD is considered
to be a high yielding NDMA precursor, NDMA FPs in urine or fe-
ces after ingestion of RNTD were comparable or even lower than
those in pre-RNTD urine or feces. This suggests that RNTD and its
metabolites may not be significant sources of NDMA precursors
from human urine or feces. The main contributors to NDMA FP
from urine are still unknown. NDMA FPs of known precursors that
are found in urine, such as DMA, TMA, trimethylamine N-oxide, are
relatively low (Mitch and Sedlak, 2004).
7
NDMA FP in laundry greywater containing detergent only (LD,
470±97 ng/L NDMA FP) was similar to that containing both de-
tergent and fabric softener (LF, 501±96 ng/L NDMA FP), suggest-
ing that fabric softener may not be an important source of NDMA
precursors. Shower greywater with the use of shampoo only (SS,
449±0 ng/L NDMA FP) exhibited slightly higher NDMA FP than
that not containing any personal care products (S, 378±8 ng/L
NDMA FP), or that with the use of body wash only (SB; 353±5
ng/L NDMA FP). While shampoo might be a source of NDMA
precursors (e.g., surfactant ingredients; Zeng and Mitch, 2015),
the majority of NDMA precursors present in shower greywa-
ters may originate from sources other than personal care prod-
ucts, such as trimethylamine N-oxide, choline and betaine ex-
creted in human sweat (Ayesh et al., 1993; Craig et al., 2010;
Subramaniam and Fletcher, 2018) which have been shown to yield
NDMA FP (Mitch and Sedlak, 2004; Kemper et al., 2010). The DOC-
normalized NDMA FP (i.e., NDMA FP/DOC) was calculated to eval-
uate the reactivities of sewage components in forming NDMA FP.
Urine showed the highest DOC-normalized NDMA FP (2344-7314
ng/mg), followed by feces (133-505 ng/mg) and kitchen greywa-
ters (168-487 ng/mg), while the DOC-normalized NDMA FPs in
the other sewage components were <35 ng/mg. This suggests that
NDMA precursors from the collected biological samples (i.e., hu-
man urine and feces, food ingredients) may have higher reactivities
in forming NDMA FP than anthropogenic constituents (e.g., laun-
dry detergent, personal care products such as shampoo and body
wash).
Treating sewage components with WWTP AS (i.e., AS 1-3) gen-
erally decreased NDMA FPs (Fig. 6 and Fig. S7). NDMA FPs in LD,
SS and KD greywaters which contain detergents or shampoo, in-
creased after 6-h incubation with AS 1 and then decreased af-
ter 24-h incubation. Biodegradation of surfactants, foam stabiliz-
ers, emulsifiers, and some dyes used for manufacturing detergents
or shampoos may increase NDMA FP due to formation of interme-
diates containing DMA groups (Fig. S8). A longer HRT (i.e., 24 h)
favored the removal of such intermediates. After 24-h incubation
with the three WWTP AS, most sewage components showed rela-
tively low NDMA FPs (i.e., <370 ng/L) except for the urine black-
water (pre- and post-RNTD) which still exhibited relatively high
NDMA FPs (i.e., 1157-6621 ng/L).
The relative importance of each sewage component contributing
to the overall NDMA FP in domestic sewage and secondary effluent
was evaluated using the following equation:
CiVi
Ri ( % ) =  (2)
CiVi
where Ri is the relative importance (%) of an individual sewage
component contributing to NDMA FP, Ci is NDMA FP measured for
an individual sewage component (ng/L), and Vi is the volume frac-
tion (%) of each individual component in domestic sewage or sec-
ondary effluent (Table S9).
According to the calculated contributions to NDMA FP, urine
contributed 92% of the overall NDMA FP in domestic sewage, fol-
lowed by kitchen greywater (~7%) and feces (~1%). After 6 or 24-h
incubation with different sources of AS, urine was still the domi-
nant contributor (90-98%) to NDMA FP, followed by kitchen grey-
water (<5%) and laundry greywater (<4%). Known ingredients in
secondary effluents such as DMA, TMA, amino acids, peptides and
amino sugars have been found not to be the major sources of
NDMA precursors (Mitch and Sedlak, 2004; Wang et al., 2014). Ad-
ditional studies are needed to investigate the constituents in hu-
man urine responsible for its dominant contribution to NDMA FPs
in wastewater influents and effluents.
The mechanisms responsible for reducing NDMA FPs from
urine and feces were evaluated using NaN3 -treated AS 2 (i.e., via
biosorption) and fresh AS 1 collected in spring (i.e., via biodegra-
8 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315
12000
NDMA FPs in sewage components
10000
8000
(ng/L)
6000
4000
2000
0
U UR
Fig. 6. NDMA FPs in sewage components during batch treatment tests with AS 1. U and UR: urine samples collected before (U) and after (UR) taking RNTD, respectively,
both diluted 250 folds in dechlorinated tap water; F and FR: feces samples collected before (F) and after (FR) taking rntd, respectively, both diluted 150 folds; LD and LF:
laundry greywaters containing detergent only (LD) and detergent plus fabric softener (LF); S, SS and SB: shower greywaters not containing any personal care products (S),
containing shampoo only (SS), or containing body wash only (SB); W: bathroom washbasin greywater; KD and KF: kitchen greywaters containing dishwashing detergent
(KD), or containing food leachates diluted 100 folds in dechlorinated tap water (KF). Control tests: NDMA FPs prior to batch AS treatment tests.
dation). Nearly 100% of NDMA FP in urine (pre- and post-RNTD)
was reduced via biosorption, while >64% of NDMA FP from feces
(pre- and post-RNTD) was reduced via biodegradation. For the 2-
d, 7-d and 10-d biodegradation tests with fresh AS 1, reductions
in NDMA FP for urine were 24%, 38% and 49%, respectively, which
were comparable to those for RNTD (i.e., 24%, 46% and 52%, re-
spectively). These results suggest that NDMA precursors in urine
may have a moderate biodegradability like RNTD. Because removal
of NDMA precursors with moderate biodegradability is dependent
on treatment conditions (i.e., HRT, SRT, MLSS) and AS activity, re-
ductions in NDMA FPs from wastewaters are thus varying among
WWTPs.
4. Conclusions
In this study, decreases in NDMA FPs from model NDMA pre-
cursors and unidentified precursors present in different compo-
nents of sewage were investigated using a batch AS treatment as-
say. The effects of treatment conditions (i.e., HRT, SRT, MLSS, AS
sources) and deactivation mechanisms (i.e., biosorption, biodegra-
dation, volatilization) were also carefully evaluated. AS treatments
decreased NDMA FPs to different degrees depending on the func-
tional groups of the precursor compounds, sources of precursors,
AS types, and operating conditions (i.e., HRT, SRT, MLSS) of the
AS process. To better remove NDMA precursors, strategies are
needed to control the discharge of precursor compounds with bio-
refractory functional groups. Increasing the HRT and SRT may en-
hance the removal of moderately biodegradable NDMA precursors
(such as RNTD, and those from urine), although the necessary in-
crease in aeration basin capacity make this a costly solution.
Biosorption was the most important mechanism responsible
for NDMA FP reduction, especially for high-molecular-weight com-
pounds (e.g., amine-based pharmaceuticals). Removal of NDMA
precursors via biosorption may depend on AS properties which
vary with influent compositions and operating conditions at
WWTPs. Biodegradation tests demonstrated that AS activity (which
depended on AS sources, sampling seasons and SRT) may impact
removal rates for NDMA precursors. Strategies to enhance AS ac-
tivity are needed to enhance inactivation of precursors. Targeted
addition of MLSS with microbes that biodegrade NDMA precursors
is one such approach.
Urine blackwater, among the tested blackwaters and greywa-
ters, was the most important contributor to NDMA precursors in
domestic sewage and AS-treated secondary effluents. Separating
Control
t=6h
t = 24 h
F FR LD LF S SS SB W KD KF
human urine from domestic sewage may thus significantly reduce
NDMA precursors present in wastewater influents and effluents.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
Acknowledgments
This work was financially supported by the Water Research
Foundation (Project #4591). Authors would like to thank Dr.
Christophe Darnault for allowing us to use biosafety cabinet, Part-
iban Raja for assisting in operating the lab-scale SBRs and Anne
Cumming for maintaining instruments. We especially appreciate
Drs. Teng Zeng and Dr. William Mitch for their collaboration in col-
lecting and shipping blackwaters and laundry greywaters to Clem-
son University. We also want to thank the three WWTPs in SC for
supporting our AS sampling campaigns.
Supplementary materials
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.watres.2020.116315.
References
Ahmed, Z., Cho, J., Lim, B., Song, K., Ahn, K., 2007. Effects of sludge retention time
on membrane fouling and microbial community structure in a membrane biore-
actor. J. Membrane Sci. 287 (2), 211–218.
Ahn, K.H., Ko, K.B., Sung, B.W., 2013. Settling and dewatering characteristics of
mixed microorganisms according to changes in the SRT. Desalination and Water
Treatment 51 (31-33), 6201–6209.
Al-Halbouni, D., Traber, J., Lyko, S., Wintgens, T., Melin, T., Tacke, D., Janot, A.,
Dott, W., Hollender, J., 2008. Correlation of EPS content in activated sludge at
different sludge retention times with membrane fouling phenomena. Water Res.
42 (6–7), 1475–1488.
An, D., Chen, Y., Gu, Bin, Westerhoff, P., Hanigan, D., Herckes, P., Fischer, N., Dono-
van, S., Croue, J.P., Atkinson, A., 2019. Lower molecular weight fractions of Poly-
DADMAC coagulants disproportionately contribute to N-nitrosodimethylamine
formation during water treatment. Water Res. 150, 466–472.
APHA, AWWA, WEF., 2005. In: Washington, DC. (Ed.), Standard Methods for the Ex-
amination of Water and Wastewater, 21st edn..
Ayesh, R., Mitchell, S.C., Zhang, A., Smith, R.L., 1993. The fish odor syndrome: Bio-
chemical, familial, and clinical aspects. Br. Med. J. 307 (6905), 655–657.
Beita-Sandi, W., Selbes, M., Ersan, M.S., Karanfil, T., 2019. Release of nitrosamines
and nitrosamine precursors from scrap tires. Environ. Sci. Tech. Lett. 6 (4),
251–256.
 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315
Byrns, G., 2001. The fate of xenobiotic organic compounds in wastewater treatment
plants. Water Res. 35 (10), 2523–2533.
Carucci, A., Cappai, G., Piredda, M., 2006. Biodegradability and toxicity of pharma-
ceuticals in biological wastewater treatment plant. J. Environ. Sci. Health Part A
41 (9), 1831–1842.
Castiglioni, S., Zuccato, E., Crisci, E., Chiabrando, C., Fanelli, R., Bagnati, R., 2006.
Identification and measurement of illicit drugs and their metabolites in urban
wastewater by liquid chromatography-tandem mass spectrometry. Anal. Chem.
78 (24), 8421–8429.
Chonova, T., Keck, F., Labanowski, J., Montuelle, B., Rimet, F., Bouchez, A., 2016. Sep-
arate treatment of hospital and urban wastewaters: A real scale comparison
of effluents and their effect on microbial communities. Sci. Total Environ. 542,
965–975.
Chuang, Y., Shabani, F., Munoz, J., Aflaki, R., Hammond, S.D., Mitch, W.A., 2019. Com-
paring industrial and domestic discharges as sources of N-nitrosamines and
their chloramine or ozone-reactive precursors. Environ. Sci. 5 (4), 726–736.
Craig, S.C., Craig, S.A., Ganio, M.S., Maresh, C.M., Greg, H., Costa, K., Zeisel, S.H., 2010.
The betaine content of sweat from adolescent females. J. Int. Soc. Sports Nutr. 7
(1), 3. doi:10.1186/1550- 2783- 7- 3.
Friedler, E., Butler, D., Alfiya, Y., 2013. Wastewater Composition. In: Larsen, T.A., Ud-
ert, K.M., Lienert, J. (Eds.), Source Separation and Decentralization for Wastew-
ater Management. IWA Publishing, London, UK.
Golet, E.M., Xifra, I., Siegrist, H., Alder, A.C., Giger, W., 2003. Environmental exposure
assessment of fluoroquinolone antibacterial agents from sewage to soil. Environ.
Sci. Tech. 37 (15), 3243–3249.
Grady, C.P.L., Daigger, G.T., Love, N.G., Filipe, C.D.M., 2011. Biological Wastewater
Treatment, 3rd edn. CRC Press, FL.
Gray, N.F., 2004. Biology of Wastewater Treatment, 2nd edn. Imperial College Press,
London, UK.
Gros, M., Petrovic, M., Barcelo, D., 2006. Development of a multi-residue analyti-
cal methodology based on liquid chromatography-tandem mass spectrometry
(LC-MS/MS) for screening and trace level determination of pharmaceuticals in
surface and wastewaters. Talanta 70 (4), 678–690.
Gros, M., Petrovic, M., Ginebreda, A., Barcelo, D., 2010. Removal of pharmaceuticals
during wastewater treatment and environmental risk assessment using hazard
indexes. Environ. Int. 36 (1), 15–26.
Guerra, P., Kim, M., Shah, A., Alaee, M., Smyth, S.A., 2014. Occurrence and fate of an-
tibiotic, analgesic/anti-inflammatory, and antifungal compounds in five wastew-
ater treatment processes. Sci. Total Environ. 473-474, 235–243.
Health Canada, 2011. Guidelines for Canadian Drinking Water Quality: Guideline
Technical Document – N-Nitrosodimethylamine. Water, Air and Climate Change
Bureau. Healthy Environments and Consumer Safety Branch, Ottawa, Canada.
Jelic, A., Gros, M., Ginebreda, A., Cespedes-Sanchez, R., Ventura, F., Petrovic, M.,
Barcelo, D., 2011. Occurrence, partition and removal of pharmaceuticals in
sewage water and sludge during wastewater treatment. Water Res. 45 (3),
1165–1176.
Jin, T., Zhang, T., Yan, Q., 2010. Characterization and quantification of ammonia-ox-
idizing archaea (AOA) and bacteria (AOB) in a nitrogen-removing reactor using
T-RFLP and qPCR. Appl. Microbiol. Biotech. 87 (3), 1167–1176.
Kemper, J.M., Walse, S.S., Mitch, W.A., 2010. Quaternary amines as nitrosamine pre-
cursors: A role for consumer products? Environ. Sci. Tech. 44, 1224–1231.
Kim, S., Rossmassler, K., Broeckling, C.D., Galloway, S., Prenni, J., De Long, S.K., 2017.
Impact of inoculum sources on biotransformation of pharmaceuticals and per-
sonal care products. Water Res. 125, 227–236.
Kraigher, B., Kosjek, T., Heath, E., Kompare, B., Mandic-Mulec, I., 2008. Influence of
pharmaceutical residues on the structure of activated sludge bacterial commu-
nities in wastewater treatment bioreactors. Water Res. 42 (17), 4578–4588.
Krasner, S.W., Mitch, W.A., McCurry, D.L., Hanigan, D., Westerhoff, P., 2013. Forma-
tion, precursors, control, and occurrence of N-nitrosamines in drinking water: A
review. Water Res. 47 (13), 4433–4450.
Le Roux, J., Gallard, H., Croue, J.P., 2011. Chloramination of nitrogenous contaminants
(pharmaceuticals and pesticides): NDMA and halogenated DBPs formation. Wa-
ter Res. 45, 3164–3174.
Lee, S.K., Kim, D., Jin, C., Yoo, H.H., 2010. Determination of urinary trimethylamine
and trimethylamine N-oxide by liquid chromatography-tandem mass spectrom-
etry using mixed-mode stationary phases. Bull. Korean Chem. Soc. 31, 483–486.
Li, B., Zhang, T., 2010. Biodegradation and adsorption of antibiotics in the activated
sludge process. Environ. Sci. Tech. 44 (9), 3468–3473.
Liu, C., Olivares, C.I., Pinto, A.J., Lauderdale, C.V., Brown, J., Selbes, M., Karanfil, T.,
2017. The control of disinfection byproducts and their precursors in biologically
active filtration processes. Water Res. 124, 630–653.
Liu, Y., Selbes, M., Zeng, C., Zhong, R., Karanfil, T., 2014. Formation mechanism of
NDMA from ranitidine, trimethylamine, and other tertiary amines during chlo-
ramination: A computational study. Environ. Sci. Tech. 48 (15), 8653–8663.
Martinez-Hernandez, V., Meffe, R., Lopez, S.H., Bustamante, I., 2016. The role of sorp-
tion and biodegradation in the removal of acetaminophen, carbamazepine, caf-
feine, naproxen and sulfamethoxazole during soil contact: A kinetic study. Sci.
Total Environ. 559, 232–241.
Mitch, W.A., Gerecke, A.C., Sedlak, D.L., 2003. A N-nitrosodimethylamine (NDMA)
precursor analysis for chlorination of water and wastewater. Water Res. 37 (15),
3733–3741.
Mitch, W.A., Sedlak, D.L., 2004. Characterization and fate of N-nitrosodimethylamine
precursors in municipal wastewater treatment plants. Environ. Sci. Tech. 38 (5),
1445–1454.
National Health and Medical Research Council (NHMRC), 2011. National Resource
Management Ministerial Council (NRMMC). Australian Drinking Water Guide-
9
lines Paper 6 National Water Quality Management Strategy. Commonwealth of
Australia, Canberra, AUS.
Okey, R.W., Stensel, H.D.A., 1996. QSAR-based biodegradability model -A QSBR. Wa-
ter Res. 30 (9), 2206–2214.
Organization for Economic Co-operation and Development (OECD), 2003. Guide-
lines for Testing Chemicals. Paris. http://www.oecd.org/chemicalsafety/testing/
17131516.pdf (accessed on May 10th, 2014).
Petrie, B., McAdam, E.J., Hassard, F., Stephenson, T., Lester, J.N., Cartmell, E., 2014. Di-
agnostic investigation of steroid estrogen removal by activated sludge at varying
solids retention time. Chemosphere 113, 101–108.
Radjenovic, J., Petrovic, M., Barcelo, D., 2007. Analysis of pharmaceuticals in wastew-
ater and removal using a membrane bioreactor. Anal. Bioanal. Chem. 387 (4),
1365–1377.
Sarma, B., Joshi, S.R., 2015. Environmental toxicity and the role of Pseudomonads in
biodegradation of xenobiotics. In: Choudhury, H. (Ed.), Biology, Biotechnology
and Sustainable Development. Research India Publications, Delhi, India.
Savci, S., 2013. Biosorption of ranitidine onto live activated sludge. Asian J. Chem.
25 (6), 3175–3178.
Sedlak, D.L., Deeb, R.A., Hawley, E.L., Mitch, W.A., Durbin, T.D., Mowbray, S., Carr, S.,
2005. Sources and fate of nitrosodimethylamine and its precursors in municipal
wastewater treatment plants. Water Environ. Res. 77 (1), 32–39.
Sedlak, D.L., Kavanaugh, M., 2006. Removal and destruction of NDMA and
NDMA precursors during wastewater treatment. WaterReuse Foundation report,
Alexandria, VA Report No. ISBN: 0-9747586-8-X.
Selbes, M., Kim, D., Ates, N., Karanfil, T., 2013. The roles of tertiary amine structure,
background organic matter and chloramine species on NDMA formation. Water
Res. 47 (2), 945–953.
Sgroi, M., Vagliasindi, F.F.A., Snyder, S.A., Roccaro, P., 2018. N-Nitrosodimethylamine
(NDMA) and its precursors in water and wastewater: A review on formation
and removal. Chemosphere 191, 685–703.
Shen, R., Andrews, S.A., 2013a. Formation of NDMA from ranitidine and sumatrip-
tan: The role of pH. Water Res. 47 (2), 802–810.
Shen, R., Andrews, S.A., 2013b. NDMA formation from amine-based pharmaceuticals
-Impact from prechlorination and water matrix. Water Res. 47 (7), 2446–2457.
Sponza, D.T., 2002. Extracellular polymer substances and physicochemical properties
of flocs in steady and unsteady-state activated sludge systems. Process Biochem.
37 (9), 983–998.
Steiner, A.E., McLaren, D.A., Forster, C.F., 1976. The nature of activated sludge flocs.
Water Res. 10 (1), 25–30.
Subramaniam, S., Fletcher, C., 2018. Trimethylamine N-oxide: Breathe new life. Br. J.
Pharmacol. 175, 1344–1353.
Svensson, B.G, Akesson, B., Nilsson, A., Paulsson, K., 1994. Urinary excretion of
methylamines in men with varying intake of fish from the Baltic Sea. J. Toxi-
col. Environ. Health 41, 411–420.
United Kingdom Drinking Water Inspectorate (UK DWI), 20 0 0. Guidance on the
Water Supply (Water Quality) Regulations Specific to N-Nitrosodimethylamine
(NDMA) Concentrations in Drinking Water. Drinking Water Inspectorate, Lon-
don, UK.
US EPA, 2001. Integrated Risk Information System. United States Environmental Pro-
tection Agency. Office of Research and Development (ORD). National Center for
Environmental Assessment http://www.epa.gov/iris/subst/0045.htm.
US EPA., 2004. Method 521: Determination of N-nitrosamines in Drinking Water by
Solid Phase Extraction and Capillary Column Gas Chromatography with Large
Volume Injection and Chemical Tandem Mass Spectrometry (MS/MS). https://
cfpub.epa.gov/si/si_public_record_report.cfm?Lab=NERL&dirEntryId=103912 (ac-
cessed on August 19th, 2013).
US EPA, 2008. OPPTS 835.3280: Simulation Tests to Assess the Primary and Ul-
timate Biodegradability of Chemicals Discharged to Wastewater. https://www.
regulations.gov/document?D=EPA- HQ- OPPT- 2009- 0152- 0036 (accessed on May
10th, 2014).
US EPA, 2016. Contaminant Candidate List 4 (CCL4). U.S. Environmental Protection
Agency https://www.epa.gov/ccl/chemical-contaminants-ccl-4.
Vasiliadou, I.A., Molina, R., Martinez, F., Melero, J.A., 2013. Biological removal of
pharmaceutical and personal care products by a mixed microbial culture: Sorp-
tion, desorption and biodegradation. Biochem. Eng. J. 81, 108–119.
Wang, L., Li, Y., Shang, X., Shen, J., 2014. Occurrence and removal of
N-nitrosodimethylamine and its precursors in wastewater treatment plants in
and around Shanghai. Front. Environ. Sci. Eng. 8 (4), 519–530.
Wilen, B.M., Jin, B., Lant, P., 2003. The influence of key chemical constituents
in activated sludge on surface and flocculating properties. Water Res. 37 (9),
2127–2139.
Wilen, B.M., Lumley, D., Mattsson, A., Mino, T., 2008. Relationship between floc
composition and flocculation and settling properties studied at a full scale acti-
vated sludge plant. Water Res. 42 (16), 4404–4418.
Xu, J., Sheng, G., Ma, Y., Wang, L., Yu, H., 2013. Roles of extracellular polymeric
substances (EPS) in the migration and removal of sulfamethazine in activated
sludge system. Water Res. 47 (14), 5298–5306.
Yu, K., Li, B., Zhang, T., 2012. Direct rapid analysis of multiple PPCPs in munic-
ipal wastewater using ultrahigh performance liquid chromatography–tandem
mass spectrometry without SPE pre-concentration. Analytica Chimica Acta 738,
59–68.
Yuan, Z.G., Blackall, L.L., 2002. Sludge population optimization: A new dimension
for the control of biological wastewater treatment systems. Water Res. 36 (2),
482–490.
10 X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315
Zeng, T., Mitch, W.A., 2015. Contribution of N-nitrosamines and their precursors to
domestic sewage by greywaters and blackwaters. Environ. Sci. Tech. 49 (22),
13158–13167.
Zeng, T., Glover, C.M., Marti, E.J., Woods-Chabane, G.C., Karanfil, T., Mitch, W.A., Dick-
enson, R.V., 2016. Relative importance of different water categories as sources of
N-nitrosamine precursors. Environ. Sci. Tech. 50 (24), 13239–13248.
Zeng, T., Mitch, W.A., 2016. Oral intake of ranitidine increases urinary excretion of
N-nitrosodimethylamine. Carcinogenesis 37 (6), 1–10.
Zhang, B., Sun, B., Ji, M., Liu, H., 2009. Population dynamic succession and quantifi-
cation of ammonia-oxidizing bacteria in a membrane bioreactor treating mu-
nicipal wastewater. J. Hazard. Mater. 165 (1–3), 796–803.
Zhang, S., Zhou, Y., Liu, Y.D., Zhong, R., 2020. Reinvestigation of NDMA formation
mechanisms from tertiary amines during chloramination: a DFT study. Environ.
Sci. doi:10.1039/D0EW0 0 098A.
Zhang, W., Yang, P., Xiao, P., Xu, S., Liu, Y., Liu, F., Wang, D., 2015. Dynamic variation
in physicochemical properties of activated sludge floc from different WWTPs
and its influence on sludge dewaterability and settleability. Colloids and Sur-
faces A Physicochem. Eng. Aspects 467, 124–134.
Zhao, Y., Wang, A., Ren, N., Zhao, Y., 2006. Microbial community structure in differ-
ent wastewater treatment processes characterized by single-strand conforma-
tion polymorphism (SSCP) technique. Front. Environ. Sci. Eng. 2, 116–121.