Professional Documents
Culture Documents
Water Research
journal homepage: www.elsevier.com/locate/watres
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.
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
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
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
Synthetic wastewaterf
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 ipal WWTP with biological nutrient removal process (14 h HRT);
this source of AS reduced the NDMA FP of RNTD by 59%. Further
For the NDMA FP test, samples were buffered with 20-mM lower reduction in NDMA FP of RNTD was found after 24-h in-
phosphate (pH 7.8±0.2) and then chloraminated by adding a cubation with the textile (i.e., 41%) and lab-grown AS (i.e., 34%).
pre-determined volume of monochloramine stock solution (~20 0 0 Both AS 1 and AS 2 exhibited relatively high reductions in NDMA
mg Cl2 /L) to achieve the target dosage (i.e., 100 mg Cl2 /L). The FPs of the three amine-based pharmaceuticals (i.e., RNTD, MNCL
monochloramine stock solution was freshly prepared by adding a and SMTR), while the lab-grown AS (i.e., AS 4) showed the lowest
sodium hypochlorite solution (NaClO, ~40 0 0 mg Cl2 /L) to an am- reduction. Previous studies have shown that municipal AS has di-
monium chloride solution (NH4 Cl, ~10 0 0 mg N/L) drop by drop at verse microbial communities due to the complexity of organic con-
pH 9 with a Cl:N mass ratio of 4:1. The chloraminated samples taminants (including PPCPs) present in municipal wastewater in-
were then incubated in the dark at 23±2°C for 5 d. The residual fluents, while AS fed with simple nutrients (such as AS 4 fed with
chloramines were then quenched by adding an excess amount of a synthetic wastewater and AS 3 fed with textile wastewater) likely
sodium thiosulfate (Na2 S2 O3 ). The quenched samples were then has less diverse microbial communities (Zhao et al., 2006). The di-
extracted and analyzed for NDMA according to US EPA Method versity and richness of microbial species likely favors removal of
521 (US EPA, 2004). Details on the NDMA extraction and anal- xenobiotic compounds (Kraigher et al., 2008; Chonova et al., 2016;
ysis method are described elsewhere (Selbes et al., 2013; Beita- Kim et al., 2017).
Sandi et al., 2019). The minimum reporting level for NDMA is 2.0 Among the four types of AS tested, AS 3 collected from a tex-
ng/L. tile WWTP exhibited the largest seasonal variations in the reduc-
tion of some NDMA precursors (e.g., RNTD and SMTR; Table S6).
3. Results and discussion The decreases in NDMA FPs of RNTD and SMTR varied by 36%
and 23%, respectively, after 24-h incubation with AS 3 collected
3.1. Reduction in NDMA FPs of model precursors in three different seasons, while reductions in NDMA FPs of TMA
and MNCL varied by 9-12%. In contrast, <10% variations in NDMA
NDMA FP of RNTD was reduced by 34-87% after 24-h incu- FP reduction were observed after 24-h incubation with municipal
bation with four different types of AS (Fig. 1). NDMA FPs of WWTP AS (i.e., AS 1 or AS 2) collected during three seasons. In the
TMA and MNCL were reduced by 92-98% and 77-97%, respec- textile WWTP, seasonal changes in the manufacturing processes
tively, while NDMA FP of SMTR decreased to a lesser extent (i.e., may result in fluctuations in the wastewater influent components,
29-46% depending on the types of AS). These overall reductions and thus the AS 3 properties (i.e., color, MLSS, settleability). How-
were achieved via both biotic (i.e., biodegradation) and abiotic (i.e., ever, the operating conditions for the two municipal WWTPs (i.e.,
biosorption, volatilization) processes. TMA is known to be read- WW TP 1 and WW TP 2) were consistent during our sampling cam-
ily biodegradable, with >71% removals previously reported dur- paigns. Occasionally, WWTP 1 was influenced from rain events,
ing biological treatment (Mitch and Sedlak, 2004; Wang et al., producing increased influent flow rates. During one such event, air
2014). MNCL contains hydroxyl functional groups attached to ring supply to the aeration tank was shut off and the MLSS in AS 1
structures, which can enhance its biodegradability (Okey and dropped to ~10 0 0 mg/L. AS 1 collected during such rain events
Stensel, 1996). Unlike TMA or MNCL, RNTD contains a nitryl (- (~10 0 0 mg/L MLSS) showed 12%-34% lower levels of decreases in
NO2 ) functional group, and SMTR contains a sulfonamide (-NH- NDMA FPs for all tested model precursor compounds, compared
SO2 -) functional group, both of which may impede biodegradation to those measured under routine operating conditions with ~30 0 0
(Okey and Stensel, 1996). mg/L MLSS (Fig. S2).
In addition to the structures of NDMA precursors, the sources The effects of selected operational parameters (i.e., HRT and
of AS biomass also affected the overall removal of NDMA FPs. AS SRT) on deactivations of model precursor compounds were also ex-
1 was collected from a rural municipal WWTP with an extended amined. When the duration of the batch tests (i.e., simulating HRT)
aeration process (i.e., 22-24 h HRT) and a high level of nitrifica- was increased from 6 to 24 h, the reduction in NDMA FP of RNTD
tion (i.e., undetectable ammonia in the secondary effluent). With increased from 81% to 88% for AS 1, from 34% to 54% for AS 2, or
this source of AS, there was an 87% decrease in NDMA FP of RNTD from 32% to 62% for AS 3 (Fig. 2). These observations are consis-
after 24-h incubation. AS 2 was collected from an urban munic- 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,
100 AS 1
or SMTR. The reduction in NDMA FPs for TMA and MNCL were
Reduction in NDMA FP
AS 2
nearly 100% after 6 h of incubation, and further incubation thus
80 AS 3
AS 4 has no impact. Increasing HRT did not impact the extent of NDMA
FP reduction for SMTR, which is relatively bio-refractory. On the
60
(%)
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 all reduction during 6-h AS treatment (Fig. 3). Similarly, biodegra-
dation accounted for only <1% and 2-11% of NDMA FP reduction
The contribution of biodegradation to overall NDMA FP reduc- for MNCL and SMTR, respectively, when treated with the three
tion was defined as the difference between the overall reduction WWTP AS (i.e., AS 1, AS 2 and AS 3). In contrast, the NDMA FP of
and the reduction via biosorption plus volatilization. Reductions TMA was mainly decreased via biodegradation (i.e., 74-98%) during
due to biosorption (based on inhibition of respiration with NaN3 ) treatment with AS 2, AS 3 and AS 4, although nearly 100% of its
and volatilization (in the absence of AS) were measured separately. NDMA FP decrease was observed via biosorption with AS 1. The
The reduction in NDMA FP of RNTD occurred mainly via biosorp- lower biosorption of TMA as compared to the other model precur-
tion, while biodegradation contributed less than 25% to the over- sors is probably due to its relatively higher water solubility (i.e.,
100 a 100
Reduction in NDMA FP
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 b
Reduction in NDMA FP
Reduction in NDMA FP
100
t=6h
80 80
t = 10 h
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. 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 (Savci, 2013; Vasiliadou et al., 2013; Martinez-Hernandez et al.,
all tested precursor compounds, volatilization during 6-h incuba- 2016). To examine the biosorption kinetics, NDMA FPs of model
tion in the absence of any AS biomass had a negligible impact on precursor compounds were monitored up to 10 h during the
NDMA FPs (Fig. S4). These results indicate that biosorption was an biosorption tests with NaN3 -deactivated AS liquor (~30 0 0 mg/L
important removal mechanism for the tested NDMA precursors, es- MLSS). The two municipal AS (AS 1 and AS 2), and a new sample
pecially for the high-molecular-weight compounds, while the effect of AS 3 (designated AS 3’ in Fig. 4) were employed for biosorption
of volatilization was negligible. kinetics tests. AS 3’ was collected from the textile WWTP after a
Reduction in NDMA FP via biosorption also depended on the new manufacturing process was launched, which resulted in sub-
sources of AS. AS 1 exhibited higher uptakes for RNTD and TMA, stantially altered AS properties (e.g., lighter color, increased MLSS,
while AS 4 exhibited the lowest level of biosorption of TMA, MNCL and enhanced settleability). AS 3’ exhibited 2-56% more reduction
and SMTR. The physiochemical properties of AS, such as its set- of NDMA FP via biosorption than AS 3 after 6-h incubation (Fig.
tleability (e.g., sludge volume index), the amounts and compo- S5). In general, increasing the contact time from 2 to 10 h during
sition of extracellular polymeric substances (EPS), and the pres- the biosorption tests had small impacts on the NDMA FP removals
ence of floc-forming microbes may affect the surface charges of (i.e., <18% differences) for RNTD, MNCL and SMTR despite the dif-
AS (Sponza, 2002; Wilen et al., 2003, 2008) and thus the adsorp- ferent types of AS (i.e., AS 1, AS 2 and AS 3’). The only exception
tion of polar compounds onto AS particles (Steiner et al., 1976). was TMA, for which NDMA FP reductions were 33-41% higher after
In this study, the inconsistent biosorption efficiencies among the 6 or 10-h incubation compared to 2-h incubation with AS 2 and
four AS were likely because of their distinct properties or char- AS 3’. The increased reduction of NDMA FP may have been due
acteristics. The influent compositions and operating conditions at to a recovery in biodegradation activity, rather than biosorption.
WWTPs (or lab-scale SBR for AS 4) can both impact AS properties As the incubation duration increased from 2 to 10 h, the oxygen
(Sponza, 2002; Ahn et al., 2013; Zhang et al., 2015). uptake rates by NaN3 -deactivated AS slightly increased (Fig. S1),
The solid-water distribution coefficients (Kd ) of precursor com- suggesting the gradually recovering microbial activities during the
pounds biosorbed onto different AS samples were calculated based biosorption tests. Overall, these results indicate that reductions in
on NDMA FPs measured in both aqueous and solid phases (Table NDMA FP for the model precursors via biosorption were completed
S7). The log Kd values of the tested model precursors biosorbed in less time (i.e., 2 h) than the lower end of HRTs at most WWTPs
onto different sources of AS were at least twofold higher than log (Gray, 2004).
Kow . This may suggest that adsorption of precursor compounds
onto AS particles was governed by factors other than hydrophobic 3.3. Biodegradation of model NDMA precursors
interactions, such as electrostatic interactions (Golet et al., 2003).
Under the near neutral pH condition of the biosorption experi- Decreases in NDMA FP via biodegradation were investigated for
ments (pH = 7.4±0.2), DMA functional groups (i.e., -N(CH3 )2 ) of RNTD (100 nM) and SMTR (200 nM) using AS 1 (collected dur-
precursor compounds are positively charged and tend to bind to ing the spring and summer) and AS 4 (at 8 and 24-d SRTs) with a
negatively charged carboxyl and/or hydroxyl groups on AS surfaces MLSS concentration of ~150-200 mg/L. After 10 d of incubation, the
(Steiner et al., 1976; Shen and Andrews, 2013b). decrease in NDMA FP for RNTD was highest for AS 1 collected in
Biosorption of xenobiotic compounds is typically considered to summer, followed by AS 1 collected in spring, AS 4 at an 8 d SRT,
be a rapid process, e.g., completed within hours or even minutes 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 7
12000
4000
2000
0
U UR F FR LD LF S SS SB W KD KF
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) human urine from domestic sewage may thus significantly reduce
was reduced via biosorption, while >64% of NDMA FP from feces NDMA precursors present in wastewater influents and effluents.
(pre- and post-RNTD) was reduced via biodegradation. For the 2-
d, 7-d and 10-d biodegradation tests with fresh AS 1, reductions Declaration of Competing Interest
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- The authors declare that they have no known competing finan-
spectively). These results suggest that NDMA precursors in urine cial interests or personal relationships that could have appeared to
may have a moderate biodegradability like RNTD. Because removal influence the work reported in this paper.
of NDMA precursors with moderate biodegradability is dependent
on treatment conditions (i.e., HRT, SRT, MLSS) and AS activity, re- Acknowledgments
ductions in NDMA FPs from wastewaters are thus varying among
WWTPs. This work was financially supported by the Water Research
Foundation (Project #4591). Authors would like to thank Dr.
4. Conclusions Christophe Darnault for allowing us to use biosafety cabinet, Part-
iban Raja for assisting in operating the lab-scale SBRs and Anne
In this study, decreases in NDMA FPs from model NDMA pre- Cumming for maintaining instruments. We especially appreciate
cursors and unidentified precursors present in different compo- Drs. Teng Zeng and Dr. William Mitch for their collaboration in col-
nents of sewage were investigated using a batch AS treatment as- lecting and shipping blackwaters and laundry greywaters to Clem-
say. The effects of treatment conditions (i.e., HRT, SRT, MLSS, AS son University. We also want to thank the three WWTPs in SC for
sources) and deactivation mechanisms (i.e., biosorption, biodegra- supporting our AS sampling campaigns.
dation, volatilization) were also carefully evaluated. AS treatments
decreased NDMA FPs to different degrees depending on the func- Supplementary materials
tional groups of the precursor compounds, sources of precursors,
AS types, and operating conditions (i.e., HRT, SRT, MLSS) of the Supplementary material associated with this article can be
AS process. To better remove NDMA precursors, strategies are found, in the online version, at doi:10.1016/j.watres.2020.116315.
needed to control the discharge of precursor compounds with bio-
refractory functional groups. Increasing the HRT and SRT may en- References
hance the removal of moderately biodegradable NDMA precursors
(such as RNTD, and those from urine), although the necessary in- 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-
crease in aeration basin capacity make this a costly solution. actor. J. Membrane Sci. 287 (2), 211–218.
Biosorption was the most important mechanism responsible Ahn, K.H., Ko, K.B., Sung, B.W., 2013. Settling and dewatering characteristics of
for NDMA FP reduction, especially for high-molecular-weight com- mixed microorganisms according to changes in the SRT. Desalination and Water
Treatment 51 (31-33), 6201–6209.
pounds (e.g., amine-based pharmaceuticals). Removal of NDMA Al-Halbouni, D., Traber, J., Lyko, S., Wintgens, T., Melin, T., Tacke, D., Janot, A.,
precursors via biosorption may depend on AS properties which Dott, W., Hollender, J., 2008. Correlation of EPS content in activated sludge at
vary with influent compositions and operating conditions at different sludge retention times with membrane fouling phenomena. Water Res.
42 (6–7), 1475–1488.
WWTPs. Biodegradation tests demonstrated that AS activity (which
An, D., Chen, Y., Gu, Bin, Westerhoff, P., Hanigan, D., Herckes, P., Fischer, N., Dono-
depended on AS sources, sampling seasons and SRT) may impact van, S., Croue, J.P., Atkinson, A., 2019. Lower molecular weight fractions of Poly-
removal rates for NDMA precursors. Strategies to enhance AS ac- DADMAC coagulants disproportionately contribute to N-nitrosodimethylamine
formation during water treatment. Water Res. 150, 466–472.
tivity are needed to enhance inactivation of precursors. Targeted
APHA, AWWA, WEF., 2005. In: Washington, DC. (Ed.), Standard Methods for the Ex-
addition of MLSS with microbes that biodegrade NDMA precursors amination of Water and Wastewater, 21st edn..
is one such approach. Ayesh, R., Mitchell, S.C., Zhang, A., Smith, R.L., 1993. The fish odor syndrome: Bio-
Urine blackwater, among the tested blackwaters and greywa- 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
ters, was the most important contributor to NDMA precursors in and nitrosamine precursors from scrap tires. Environ. Sci. Tech. Lett. 6 (4),
domestic sewage and AS-treated secondary effluents. Separating 251–256.
X. Zhang, D. Kim and D.L. Freedman et al. / Water Research 186 (2020) 116315 9
Byrns, G., 2001. The fate of xenobiotic organic compounds in wastewater treatment lines Paper 6 National Water Quality Management Strategy. Commonwealth of
plants. Water Res. 35 (10), 2523–2533. Australia, Canberra, AUS.
Carucci, A., Cappai, G., Piredda, M., 2006. Biodegradability and toxicity of pharma- Okey, R.W., Stensel, H.D.A., 1996. QSAR-based biodegradability model -A QSBR. Wa-
ceuticals in biological wastewater treatment plant. J. Environ. Sci. Health Part A ter Res. 30 (9), 2206–2214.
41 (9), 1831–1842. Organization for Economic Co-operation and Development (OECD), 2003. Guide-
Castiglioni, S., Zuccato, E., Crisci, E., Chiabrando, C., Fanelli, R., Bagnati, R., 2006. lines for Testing Chemicals. Paris. http://www.oecd.org/chemicalsafety/testing/
Identification and measurement of illicit drugs and their metabolites in urban 17131516.pdf (accessed on May 10th, 2014).
wastewater by liquid chromatography-tandem mass spectrometry. Anal. Chem. Petrie, B., McAdam, E.J., Hassard, F., Stephenson, T., Lester, J.N., Cartmell, E., 2014. Di-
78 (24), 8421–8429. agnostic investigation of steroid estrogen removal by activated sludge at varying
Chonova, T., Keck, F., Labanowski, J., Montuelle, B., Rimet, F., Bouchez, A., 2016. Sep- solids retention time. Chemosphere 113, 101–108.
arate treatment of hospital and urban wastewaters: A real scale comparison Radjenovic, J., Petrovic, M., Barcelo, D., 2007. Analysis of pharmaceuticals in wastew-
of effluents and their effect on microbial communities. Sci. Total Environ. 542, ater and removal using a membrane bioreactor. Anal. Bioanal. Chem. 387 (4),
965–975. 1365–1377.
Chuang, Y., Shabani, F., Munoz, J., Aflaki, R., Hammond, S.D., Mitch, W.A., 2019. Com- Sarma, B., Joshi, S.R., 2015. Environmental toxicity and the role of Pseudomonads in
paring industrial and domestic discharges as sources of N-nitrosamines and biodegradation of xenobiotics. In: Choudhury, H. (Ed.), Biology, Biotechnology
their chloramine or ozone-reactive precursors. Environ. Sci. 5 (4), 726–736. and Sustainable Development. Research India Publications, Delhi, India.
Craig, S.C., Craig, S.A., Ganio, M.S., Maresh, C.M., Greg, H., Costa, K., Zeisel, S.H., 2010. Savci, S., 2013. Biosorption of ranitidine onto live activated sludge. Asian J. Chem.
The betaine content of sweat from adolescent females. J. Int. Soc. Sports Nutr. 7 25 (6), 3175–3178.
(1), 3. doi:10.1186/1550- 2783- 7- 3. Sedlak, D.L., Deeb, R.A., Hawley, E.L., Mitch, W.A., Durbin, T.D., Mowbray, S., Carr, S.,
Friedler, E., Butler, D., Alfiya, Y., 2013. Wastewater Composition. In: Larsen, T.A., Ud- 2005. Sources and fate of nitrosodimethylamine and its precursors in municipal
ert, K.M., Lienert, J. (Eds.), Source Separation and Decentralization for Wastew- wastewater treatment plants. Water Environ. Res. 77 (1), 32–39.
ater Management. IWA Publishing, London, UK. Sedlak, D.L., Kavanaugh, M., 2006. Removal and destruction of NDMA and
Golet, E.M., Xifra, I., Siegrist, H., Alder, A.C., Giger, W., 2003. Environmental exposure NDMA precursors during wastewater treatment. WaterReuse Foundation report,
assessment of fluoroquinolone antibacterial agents from sewage to soil. Environ. Alexandria, VA Report No. ISBN: 0-9747586-8-X.
Sci. Tech. 37 (15), 3243–3249. Selbes, M., Kim, D., Ates, N., Karanfil, T., 2013. The roles of tertiary amine structure,
Grady, C.P.L., Daigger, G.T., Love, N.G., Filipe, C.D.M., 2011. Biological Wastewater background organic matter and chloramine species on NDMA formation. Water
Treatment, 3rd edn. CRC Press, FL. Res. 47 (2), 945–953.
Gray, N.F., 2004. Biology of Wastewater Treatment, 2nd edn. Imperial College Press, Sgroi, M., Vagliasindi, F.F.A., Snyder, S.A., Roccaro, P., 2018. N-Nitrosodimethylamine
London, UK. (NDMA) and its precursors in water and wastewater: A review on formation
Gros, M., Petrovic, M., Barcelo, D., 2006. Development of a multi-residue analyti- and removal. Chemosphere 191, 685–703.
cal methodology based on liquid chromatography-tandem mass spectrometry Shen, R., Andrews, S.A., 2013a. Formation of NDMA from ranitidine and sumatrip-
(LC-MS/MS) for screening and trace level determination of pharmaceuticals in tan: The role of pH. Water Res. 47 (2), 802–810.
surface and wastewaters. Talanta 70 (4), 678–690. Shen, R., Andrews, S.A., 2013b. NDMA formation from amine-based pharmaceuticals
Gros, M., Petrovic, M., Ginebreda, A., Barcelo, D., 2010. Removal of pharmaceuticals -Impact from prechlorination and water matrix. Water Res. 47 (7), 2446–2457.
during wastewater treatment and environmental risk assessment using hazard Sponza, D.T., 2002. Extracellular polymer substances and physicochemical properties
indexes. Environ. Int. 36 (1), 15–26. of flocs in steady and unsteady-state activated sludge systems. Process Biochem.
Guerra, P., Kim, M., Shah, A., Alaee, M., Smyth, S.A., 2014. Occurrence and fate of an- 37 (9), 983–998.
tibiotic, analgesic/anti-inflammatory, and antifungal compounds in five wastew- Steiner, A.E., McLaren, D.A., Forster, C.F., 1976. The nature of activated sludge flocs.
ater treatment processes. Sci. Total Environ. 473-474, 235–243. Water Res. 10 (1), 25–30.
Health Canada, 2011. Guidelines for Canadian Drinking Water Quality: Guideline Subramaniam, S., Fletcher, C., 2018. Trimethylamine N-oxide: Breathe new life. Br. J.
Technical Document – N-Nitrosodimethylamine. Water, Air and Climate Change Pharmacol. 175, 1344–1353.
Bureau. Healthy Environments and Consumer Safety Branch, Ottawa, Canada. Svensson, B.G, Akesson, B., Nilsson, A., Paulsson, K., 1994. Urinary excretion of
Jelic, A., Gros, M., Ginebreda, A., Cespedes-Sanchez, R., Ventura, F., Petrovic, M., methylamines in men with varying intake of fish from the Baltic Sea. J. Toxi-
Barcelo, D., 2011. Occurrence, partition and removal of pharmaceuticals in col. Environ. Health 41, 411–420.
sewage water and sludge during wastewater treatment. Water Res. 45 (3), United Kingdom Drinking Water Inspectorate (UK DWI), 20 0 0. Guidance on the
1165–1176. Water Supply (Water Quality) Regulations Specific to N-Nitrosodimethylamine
Jin, T., Zhang, T., Yan, Q., 2010. Characterization and quantification of ammonia-ox- (NDMA) Concentrations in Drinking Water. Drinking Water Inspectorate, Lon-
idizing archaea (AOA) and bacteria (AOB) in a nitrogen-removing reactor using don, UK.
T-RFLP and qPCR. Appl. Microbiol. Biotech. 87 (3), 1167–1176. US EPA, 2001. Integrated Risk Information System. United States Environmental Pro-
Kemper, J.M., Walse, S.S., Mitch, W.A., 2010. Quaternary amines as nitrosamine pre- tection Agency. Office of Research and Development (ORD). National Center for
cursors: A role for consumer products? Environ. Sci. Tech. 44, 1224–1231. Environmental Assessment http://www.epa.gov/iris/subst/0045.htm.
Kim, S., Rossmassler, K., Broeckling, C.D., Galloway, S., Prenni, J., De Long, S.K., 2017. US EPA., 2004. Method 521: Determination of N-nitrosamines in Drinking Water by
Impact of inoculum sources on biotransformation of pharmaceuticals and per- Solid Phase Extraction and Capillary Column Gas Chromatography with Large
sonal care products. Water Res. 125, 227–236. Volume Injection and Chemical Tandem Mass Spectrometry (MS/MS). https://
Kraigher, B., Kosjek, T., Heath, E., Kompare, B., Mandic-Mulec, I., 2008. Influence of cfpub.epa.gov/si/si_public_record_report.cfm?Lab=NERL&dirEntryId=103912 (ac-
pharmaceutical residues on the structure of activated sludge bacterial commu- cessed on August 19th, 2013).
nities in wastewater treatment bioreactors. Water Res. 42 (17), 4578–4588. US EPA, 2008. OPPTS 835.3280: Simulation Tests to Assess the Primary and Ul-
Krasner, S.W., Mitch, W.A., McCurry, D.L., Hanigan, D., Westerhoff, P., 2013. Forma- timate Biodegradability of Chemicals Discharged to Wastewater. https://www.
tion, precursors, control, and occurrence of N-nitrosamines in drinking water: A regulations.gov/document?D=EPA- HQ- OPPT- 2009- 0152- 0036 (accessed on May
review. Water Res. 47 (13), 4433–4450. 10th, 2014).
Le Roux, J., Gallard, H., Croue, J.P., 2011. Chloramination of nitrogenous contaminants US EPA, 2016. Contaminant Candidate List 4 (CCL4). U.S. Environmental Protection
(pharmaceuticals and pesticides): NDMA and halogenated DBPs formation. Wa- Agency https://www.epa.gov/ccl/chemical-contaminants-ccl-4.
ter Res. 45, 3164–3174. Vasiliadou, I.A., Molina, R., Martinez, F., Melero, J.A., 2013. Biological removal of
Lee, S.K., Kim, D., Jin, C., Yoo, H.H., 2010. Determination of urinary trimethylamine pharmaceutical and personal care products by a mixed microbial culture: Sorp-
and trimethylamine N-oxide by liquid chromatography-tandem mass spectrom- tion, desorption and biodegradation. Biochem. Eng. J. 81, 108–119.
etry using mixed-mode stationary phases. Bull. Korean Chem. Soc. 31, 483–486. Wang, L., Li, Y., Shang, X., Shen, J., 2014. Occurrence and removal of
Li, B., Zhang, T., 2010. Biodegradation and adsorption of antibiotics in the activated N-nitrosodimethylamine and its precursors in wastewater treatment plants in
sludge process. Environ. Sci. Tech. 44 (9), 3468–3473. and around Shanghai. Front. Environ. Sci. Eng. 8 (4), 519–530.
Liu, C., Olivares, C.I., Pinto, A.J., Lauderdale, C.V., Brown, J., Selbes, M., Karanfil, T., Wilen, B.M., Jin, B., Lant, P., 2003. The influence of key chemical constituents
2017. The control of disinfection byproducts and their precursors in biologically in activated sludge on surface and flocculating properties. Water Res. 37 (9),
active filtration processes. Water Res. 124, 630–653. 2127–2139.
Liu, Y., Selbes, M., Zeng, C., Zhong, R., Karanfil, T., 2014. Formation mechanism of Wilen, B.M., Lumley, D., Mattsson, A., Mino, T., 2008. Relationship between floc
NDMA from ranitidine, trimethylamine, and other tertiary amines during chlo- composition and flocculation and settling properties studied at a full scale acti-
ramination: A computational study. Environ. Sci. Tech. 48 (15), 8653–8663. vated sludge plant. Water Res. 42 (16), 4404–4418.
Martinez-Hernandez, V., Meffe, R., Lopez, S.H., Bustamante, I., 2016. The role of sorp- Xu, J., Sheng, G., Ma, Y., Wang, L., Yu, H., 2013. Roles of extracellular polymeric
tion and biodegradation in the removal of acetaminophen, carbamazepine, caf- substances (EPS) in the migration and removal of sulfamethazine in activated
feine, naproxen and sulfamethoxazole during soil contact: A kinetic study. Sci. sludge system. Water Res. 47 (14), 5298–5306.
Total Environ. 559, 232–241. Yu, K., Li, B., Zhang, T., 2012. Direct rapid analysis of multiple PPCPs in munic-
Mitch, W.A., Gerecke, A.C., Sedlak, D.L., 2003. A N-nitrosodimethylamine (NDMA) ipal wastewater using ultrahigh performance liquid chromatography–tandem
precursor analysis for chlorination of water and wastewater. Water Res. 37 (15), mass spectrometry without SPE pre-concentration. Analytica Chimica Acta 738,
3733–3741. 59–68.
Mitch, W.A., Sedlak, D.L., 2004. Characterization and fate of N-nitrosodimethylamine Yuan, Z.G., Blackall, L.L., 2002. Sludge population optimization: A new dimension
precursors in municipal wastewater treatment plants. Environ. Sci. Tech. 38 (5), for the control of biological wastewater treatment systems. Water Res. 36 (2),
1445–1454. 482–490.
National Health and Medical Research Council (NHMRC), 2011. National Resource
Management Ministerial Council (NRMMC). Australian Drinking Water Guide-
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 Zhang, S., Zhou, Y., Liu, Y.D., Zhong, R., 2020. Reinvestigation of NDMA formation
domestic sewage by greywaters and blackwaters. Environ. Sci. Tech. 49 (22), mechanisms from tertiary amines during chloramination: a DFT study. Environ.
13158–13167. Sci. doi:10.1039/D0EW0 0 098A.
Zeng, T., Glover, C.M., Marti, E.J., Woods-Chabane, G.C., Karanfil, T., Mitch, W.A., Dick- Zhang, W., Yang, P., Xiao, P., Xu, S., Liu, Y., Liu, F., Wang, D., 2015. Dynamic variation
enson, R.V., 2016. Relative importance of different water categories as sources of in physicochemical properties of activated sludge floc from different WWTPs
N-nitrosamine precursors. Environ. Sci. Tech. 50 (24), 13239–13248. and its influence on sludge dewaterability and settleability. Colloids and Sur-
Zeng, T., Mitch, W.A., 2016. Oral intake of ranitidine increases urinary excretion of faces A Physicochem. Eng. Aspects 467, 124–134.
N-nitrosodimethylamine. Carcinogenesis 37 (6), 1–10. Zhao, Y., Wang, A., Ren, N., Zhao, Y., 2006. Microbial community structure in differ-
Zhang, B., Sun, B., Ji, M., Liu, H., 2009. Population dynamic succession and quantifi- ent wastewater treatment processes characterized by single-strand conforma-
cation of ammonia-oxidizing bacteria in a membrane bioreactor treating mu- tion polymorphism (SSCP) technique. Front. Environ. Sci. Eng. 2, 116–121.
nicipal wastewater. J. Hazard. Mater. 165 (1–3), 796–803.