N-Nitrosodimethylamine Formation upon Ozonation and

semicarbazides) or sulfamide (e.g., N,N-dimethylsulfamide (DMS)) functional group, ... (2) As an alternative, a source control or elimination of N...
19 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/est

N‑Nitrosodimethylamine Formation upon Ozonation and Identification of Precursors Source in a Municipal Wastewater Treatment Plant Massimiliano Sgroi,†,‡ Paolo Roccaro,†,‡ Gregg L. Oelker,§ and Shane A. Snyder*,‡,∥ †

Department of Civil & Environmental Engineering, University of Catania, Viale A. Doria 6, 95125, Catania, Italy Department of Chemical & Environmental Engineering, University of Arizona, 1133 E James E Rogers Way, Tucson, Arizona 85721, United States § United Water, Edward C. Little Water Reclamation Facility, 1935 Hughes Way, El Segundo, California 90245, United States ∥ NUS Environmental Research Institute, National University of Singapore, 5A Engineering Drive 1, Singapore 117411, Singapore ‡

S Supporting Information *

ABSTRACT: Ozone doses normalized to the dissolved organic carbon concentration were applied to the primary influent, primary effluent, and secondary effluent of a wastewater treatment plant producing water destined for potable reuse. Results showed the most N-Nitrosodimethylamine (NDMA) production from primary effluent, and the recycle streams entering the primary clarifiers were identified as the main source of NDMA precursors. The degradation of aminomethylated polyacrylamide (Mannich) polymer used for sludge treatment was a significant cause of precursor occurrence. A strong correlation between NDMA formation and ammonia concentration was found suggesting an important role of ammonia oxidation on NDMA production. During ozonation tests in DI water using dimethylamine (DMA) as model precursor, the NDMA yield significantly increased in the presence of ammonia and bromide due to the formation of hydroxylamine and brominated nitrogenous oxidants. In addition, NDMA formation during ozonation of dimethylformamide (DMF), the other model precursor used in this study, occurred only in the presence of ammonia, and it was attributable to the oxidation of DMF by hydroxyl radicals. Filtered wastewater samples (0.7 μm) produced more NDMA than unfiltered samples, suggesting that ozone reacted with dissolved precursors and supporting the hypothesis of polymer degradation. Particularly, the total suspended solids content similarly affected NDMA formation and the UV absorbance decrease during ozonation due to the different ozone demand created in filtered and unfiltered samples.

■

INTRODUCTION The N-nitrosamines are a group of emerging disinfection byproducts (DBPs) that are considered as probable human carcinogens, with nanogram per liter water concentrations associated with a 10−6 lifetime cancer risk.1 As a result, California’s Department of Public Health set 10 ng/L notification levels for three nitrosamines (N-nitrosodimethylamine, N-nitrosodiethylamine, and N-nitrosodipropylamine),2 and California’s Office of Environmental Health Hazard Assessment established a 3 ng/L public health goal for Nnitrosodimethylamine (NDMA).3 Of all the nitrosamines, NDMA has been most commonly detected in drinking water and wastewater4 and USEPA placed it, along with other four nitrosamines, on the drinking water contaminant candidate list 3 (CCL3).5 Several studies have shown that NDMA can be generated in water and wastewater treatment systems by chlorine-based © 2014 American Chemical Society

disinfection processes and often dimethylamine (DMA) served as the model NDMA precursor.6,7 More recently, researchers have reported that alternative disinfectants, including chlorine dioxide (ClO2) and ozone (O3), are also able to produce nitrosamines.8,9 Ozone utilization is gaining more popularity in wastewater treatment, particularly for water reuse applications, because of its strong ability to degrade trace organic contaminants10 and to provide microbial disinfection.11 In addition, ozonation can also be used to reduce organic fouling on reverse osmosis (RO) membranes due to the transformation of organic matter.12 Received: Revised: Accepted: Published: 10308

March 8, 2014 July 14, 2014 July 16, 2014 July 16, 2014 dx.doi.org/10.1021/es5011658 | Environ. Sci. Technol. 2014, 48, 10308−10315

Environmental Science & Technology

Article

Sampling. Different campaigns of collection were conducted at a southwestern United States wastewater treatment plant (WWTP) operating without nitrogen removal and with anaerobic sludge digestion in thermophilic conditions. A detailed description of the WWTP is reported in the SI (Text S2 and Figure S1). Samples of the WWTP primary influent, primary effluent, and secondary effluent were collected as 24 h composite samples. Composite samples (24-h) of the centrates from waste activated sludge (WAS) thickening centrifuges and dewatering centrifuges were also collected. Samples were shipped overnight to the laboratory and kept refrigerated at 4 °C until the ozonation tests. Typical water quality parameters for the investigated water matrixes are provided in Table 1.

Typical secondary amines (such as DMA), some dyes, and related compounds have been shown to be NDMA precursors upon ozonation, but their molar conversion yields resulted in very low (i.e., 0.02%) NDMA formation.9,13,14 On the contrary, a limited subset of precursors containing hydrazine (e.g., unsymmetrical dimethylhydrazine and semicarbazides) or sulfamide (e.g., N,N-dimethylsulfamide (DMS)) functional group, have generated NDMA yields >50%.15,16 It is noteworthy to observe that the presence of inorganic water constituents, such as bromide and hydroxylamine (NH2OH), has been shown to significantly increase the NDMA yields of DMA and DMS, highlighting the importance of these substances on NDMA formation during ozonation.17−19 In addition, Padhye et al.17 observed NDMA formation from ozonation of several water treatment polymers. Particularly, poly(diallyldimethylammonium chloride) (polyDADMAC) yielded the highest amount of NDMA (i.e., yield of 0.003%) among the studied polymers, including polyamines and cationic polyacrylamides, and the NDMA formation was related to the polyDADMAC degradation and DMA release. Overall, NDMA formation has always been shown to be related to polymer degradation.17,20,21 NDMA presence and formation are generally low in pristine water bodies,22 while wastewater often contains significant concentrations of NDMA and other nitrosamines23,24 as well as considerable concentrations of NDMA precursors25 and inorganic constituents, including ammonia and bromide. Particularly, ozonation of wastewater can produce elevated amounts of NDMA15 that may be difficult to lower to the required concentration for water reuse applications.2 As an alternative, a source control or elimination of NDMA organic and inorganic precursors could be a worthier option. To date, few studies have investigated the source of NDMA precursors within a WWTP, particularly during ozonation. Examples using chloramination are Sedlack et al.,23 which looked at the potential importance of industrial precursors in wastewater treatment plants, and Padhye et al.,26 which reported the occurrence of NDMA precursors along the sludge treatment train of municipal wastewater treatment plants. Furthermore, Mitch and Sedlack,27 always upon chloramination formation potential test, detected the presence of particleassociated NDMA precursors in secondary wastewater effluent of some treatment plants using Mannich polymer for sludge thickening and dewatering operations. The main objectives of the current study were (i) to identify the source of NDMA precursors within a wastewater treatment plant upon ozonation, and (ii) to investigate the effect of inorganic water constituents (i.e., ammonia and bromide) on NDMA formation during ozonation and the subsequent possible formation mechanisms. Results of this study can contribute to the development of new strategies to reduce NDMA occurrence in drinking water impacted by wastewater.

Table 1. Typical Water Quality Parameters water parameter

influent

primary effluent

secondary effluent

conductivity (μS/cm) alkalinity (mg/L as CaCO3a) sulfate (mg/L) chloride (mg/L) bromide (μg/L) nitrate (mg/L as Nb) nitrite (mg/L as N) ammonia (mg/L as N) TKNc (mg/L) pH SUVAd (L/mg m) TOCe (mg/L) DOCf (mg/L) UV254g (1/cm) TSSh (mg/L)

1290 314 111 154 254 0.25 0.43 38.8 57 7.6 1.94 n.a.j 33.0 0.639 n.a.

1150 354 106 194 349 0.23 0.46 50.5 69 7.7 1.90 66.7 (2.6) 19.4 0.368 2105 (259.9)

1348 (104.7)h 298 (24.8) 110 (5.6) 199 (16.4) 370 4.0 (2.1) 1.7 (0.7) 40.6 (4.7) 50 (0.5) 7.7 (0.4) 1.52 (0.1) 14.7 (1.6) 14.3 (1.9) 0.210 (0.02) 12 (2.7)

a

CaCO3, calcium carbonate. bN, nitrogen. cTKN, total Kjeldahl nitrogen. dSUVA, specific UV absorbance. eTOC, total organic carbon. f DOC, dissolved organic carbon. DOC was measured after 0.7 um filtration. gUV254, UV absorbance at 254 nm. hTSS, total suspended solids. i(±one standard deviation of the measurement derived from the standard curve). jn.a., not available. Reported TOC and TSS values of WWTP primary effluent are related to October 2012 monthly average.

Ozonation. The ozonation system used to ozonate the collected samples is described in the SI (Text S3 and Figure S2). Unless otherwise specified, all the samples were filtered (0.7 μm) before ozonation. Experiments in deionized water (DI water) were buffered by 2.5 mM phosphate buffer, and sodium hydroxide was added to reach the desired pH. Residual dissolved ozone concentration was measured by indigo method.28 The tested samples were ozonated using the same or similar ozone to dissolved organic carbon (DOC) ratio. Indeed, in this study, almost all the ozonation tests were carried out with an ozone to DOC ratio of 0.8 to 1.0 mg/mg. An ozone dose normalized to the DOC concentration has often been used as an operating parameter to compare waters with varying DOC concentration.29,30 Ozone doses included in the range 0.3−1.0 mg/mg as O3/DOC ratios are generally usual for wastewater ozonation.30 Furthermore, in all the wastewater ozonation tests in this work, the dissolved ozone residual measured immediately after ozone application was zero due to the high instantaneous ozone demand (IOD) common for wastewater,31 and the utilized reactor configuration and experimental setup. Since the synthetic water spiked had little ozone demand, especially as compared to the wastewater, the

■

EXPERIMENTAL SECTION Materials. All purchased solvents, standards, and reagents were of high purity. The details concerning these materials are reported in the Supporting Information (SI Text S1). Anionic polyacrylamide polymer emulsion (Clarifloc LA-2691) and cationic Mannich polymer emulsion (Clarifloc WE-1073) were given by the facility’s lab staff of the investigated wastewater treatment plant. The active ingredient percentage was 3%, as purchased in both polymer emulsions. 10309

dx.doi.org/10.1021/es5011658 | Environ. Sci. Technol. 2014, 48, 10308−10315

Environmental Science & Technology

Article

Figure 1. NDMA concentration in ozonated and non-ozonated WWTP primary influent (Influent), primary effluent (Prim. Effl.), and secondary effluent (Sec. Effl.). Ozonated samples were analyzed in triplicate, whereas non-ozonated samples were analyzed in duplicate. Error bars represent one standard deviation of the measurement derived from the standard curve.

Table 2. NDMA Formation from Anionic Polyacrylamide and Mannich Polymer water matrix

pH

polymer

polymer (mg/L as a.i.a)

O3 (mg/L)

NDMA (ng/L)

NDMA (ng/mg)b

wastewater effluent wastewater effluent DI water DI water

7.7

polyacrylamide

3

11

NDc

7.7

Mannich

3

11

126

42

8.0 8.0

polyacrylamide Mannich

10 10

11 11

ND 1044

104

a

a.i., active ingredient. bProduced amount of NDMA divided by the polymer active ingredient concentration to yield units in ng NDMA per mg of polymer active ingredient. cND, not detectable.

Table 3. Mannich Polymer Dosage and NDMA Formation in Centrates Entering the Primary Clarifiers

a

recycle stream

flow rate (m3/s)

contribution to the total flow rate at the WWTP (%)

dewatering centrate WASTa centrate

0.120 0.238

1.3 2.6

NDMA (ng/L)

pH

DOC (mg/L)

Ammonia (mg/L as N)

Mannich polymer dosage (mg/L as a.i.b)

O3/DOC 0.0 mg/mg

O3/DOC 0.16 mg/mg

O3/DOC 1.0 mg/mg

7.9 6.9

269 42

1250 61

110−130c 9−11c

568 50

6270

17739 2110

WAST, waste activated sludge thickening. ba.i., active ingredient. crange of typical dosages at the WWTP.

NDMA precursors enter the WWTP from the sewage streams; however, most production of NDMA was observed in the primary effluent (Figure 1). This suggests that either the precursors entering with the raw sewage are transformed within the plant or that chemicals added during the primary process and/or the sludge treatments are contributing to the increased NDMA formation observed. With regard to NDMA concentration without ozone, the highest value was also found in the WWTP primary effluent. Duplicate measurements of NDMA yielded concentrations of 35 ng/L (±0.5%), 58 ng/L (±0.03%), and 30 ng/L (±1.3%) in primary influent, primary effluent, and secondary effluent, respectively. Biological treatment resulted in a 47% reduction of NDMA. Since the primary effluent had a higher NDMA concentration than the raw sewage, it is likely that centrate streams returned to the primary clarifier are contributing to the elevated NDMA observed. In the investigated WWTP, an anionic polyacrylamide polymer is used in primary treatment to aid coagulation, and a cationic Mannich polymer is employed for digested sludge dewatering and for WAS thickening. To verify the polymer

ozone residual was quenched by sodium thiosulfate immediately after ozonation in order to better represent the absence of ozone residual in the wastewater evaluated. Analytical Methods. The NDMA extraction was performed using a modified version of EPA Method 521. Analysis of NDMA was performed on an Agilent 7000 Triple Quadrupole GC/MS-MS, and is described in detail in the SI (Text S4). For TOC analysis, samples were acidified to pH 7, when free NH3 is present).35,36 However, if high levels of ammonia concentration are present in water, then the ozone-ammonia reaction rate kinetics can be significant. In these conditions, ammonia may be a considerable source of NH2OH, increasing the NDMA yields during secondary amines ozonation. In agreement, good correlation has been found between NDMA formation and ammonia concentration in primary influent, primary effluent, final effluent, and recycle streams ozonated according to a O3/ DOC ratio of 0.9−1.0 mg/mg (R2 > 0.9, SI Figure S4). In order to better investigate the role of ammonia in NDMA formation, DMA and dimethylformamide (DMF) were ozonated in DI water with and without ammonia addition. These two compounds were chosen because DMA is a degradation product of Mannich polymer20 and it has often been used as the model NDMA precursor,6,7 whereas DMF was qualitatively detected in the investigated wastewater, it belongs to a different class of compounds than DMA (i.e., amides), and it has a different reactivity with ozone.37 Because the objective of the experiments was to investigate NDMA formation mechanisms, solutions of model compounds were prepared at high concentrations. Table 4 reports the different conditions of DMA ozonation and the related NDMA formation. Ammonia concentration in DI water and pH were similar to the conditions measured in the investigated wastewater (Table 1). Results confirmed that the presence of ammonia enhances NDMA formation during ozonation, supporting the hypothesis of increased NH2OH production due to ammonia oxidation. In addition, NDMA formation increased with pH. At higher pH a larger amount of unprotonated DMA present in water is able to react with NH2OH to form UDMH.14 Furthermore, at higher pH, ammonia oxidation is faster38 and a larger amount of free NH3 can be oxidized to NH2OH by ozone. It was reported that, at different pHs, ammonia oxidation results in different concentrations and speciations of end products (i.e., different yield of nitrite and nitrate)39 and probably results in different yield of NH2OH. Experiments were also performed with different ammonia concentrations. Surprisingly, the highest NDMA formation was found at the lowest ammonia concentration (20 mg/L as N) and the highest ammonia concentration (100 mg/L as N) lowered the NDMA production. The reason for the obtained results is not clear. Padhye et al.17 observed a reduced NDMA formation by ozonation when nitrites were spiked in DMA containing solutions. Likely, elevated nitrites production during the oxidation of high amounts of ammonia may increase the ozone consumption and hinder the NDMA formation via

hydroxylamine pathway. Alternatively, high ammonia concentration may favor the production of hydrazine instead of UDMH during hydroxylamine reactions resulting in a reduced NDMA formation. In any case, an optimum ratio O3/NH3 (likely, pH dependent) can be supposed for enhancing NDMA formation in ammonia and DMA containing water. However, more research is needed to elucidate this phenomenon. NDMA production was also investigated in the presence of bromide ion (Br−), which was reported to increase NDMA formation during chloramination and ozonation of several compounds.17,19,40 When 4 mg/L of Br− was added, the NDMA formation was increased almost 8-fold, and when bromide ion and ammonia were present concurrently, the NDMA concentration was approximately 15-fold larger (Table 4). Bromide ion improves ammonia oxidation by ozone leading to the production of brominated nitrogenous species, such as bromamines.41 It seems that when ammonia and bromide are present in water, ozonation of DMA can lead to the formation of NDMA in a manner similar to its proposed formation when chlorine is added to ammonia and bromide containing water.7,40,42 Therefore, the NDMA formation pathways reported in Figure 2 can be proposed during ozonation of DMA. Ozonation of

Figure 2. Proposed pathways for NDMA formation during ozonation of DMA in the presence of ammonia and bromide. 10312

dx.doi.org/10.1021/es5011658 | Environ. Sci. Technol. 2014, 48, 10308−10315

Environmental Science & Technology

Article

Table 5. NDMA Formation during Ozonation of DMF and Ammonia Containing Water

a

water matrix

pH

DMF (μg/L)

ammonia (mg/L as N)

TBA (mM)

H2O2 (mg/L)

ozone (mg/L)

NDMA (ng/L)

DI water DI water DI water DI water wastewater effluent wastewater effluent wastewater effluent

8.0 8.0 8.0 8.0 7.7 7.7 7.7

2500 2500 2500 2500 2500 2500 0

0 50 50 50 41 41 41

0 0 0 10 0 0 0

0 0 5 0 0 5 0

11 11 11 11 11 11 11

NDa 59 101 ND 285 308 282

ND = not detectable.

Figure 3. NDMA formation upon ozonation of unfiltered and 0.7 μm filtered samples of the WWTP effluent. Ozonated samples were analyzed in experimental duplicates. Error bars represent one standard deviation of the measurement derived from the standard curve.

observed when hydroxyl radicals were quenched by addition of tert-butyl alcohol (TBA), a known hydroxyl radical scavenger. However, when H2O2 was added, the NDMA formation increased. The small amount of NDMA (13 ng/L) observed in Oya et al.13 in ultrapure water without ammonia addition is, likely, due to the much higher DMF concentration and ozone dose used in that study. As previously theorized, ammonia oxidation increases the NH2OH production improving the NDMA yield from the formed DMA, and thus from DMF. Other unknown reaction mechanisms may also take place. Spiked DMF in wastewater did not significantly increase the NDMA formation, even when H2O2 was added. It is likely that all the produced •OH radicals were quenched by effluent organic matter (EfOM) and they were not able to react with the spiked DMF in wastewater. For both model precursors (DMA and DMF), the ammonia presence increased the NDMA yield during ozonation, confirming its hypothesized importance on NDMA formation. Particle-Associated NDMA Precursors and Total Suspended Solids Effect. In this study, primary influent, primary effluent, and secondary effluent of the investigated WWTP were ozonated after 0.7 μm filtration to explore the NDMA formation within the treatment plant. A prior study reported that the use of DMA-based polyacrylamide (i.e., Mannich) sludge thickening polymers can be a significant source of particle-associated NDMA precursors and a higher NDMA-FP can be present in unfiltered samples compared to 0.7 μm filtered samples in tests accomplished with high monochloramine dose and extended reaction time.27 In order to compare the NDMA formation upon ozonation of filtered and unfiltered samples, unfiltered primary influent, primary effluent, and secondary effluent of the studied WWTP were

ammonia and secondary amines forms hydroxylamine; brominated nitrogenous oxidants are formed in the presence of bromide. Similar to proposed pathways during chloramination reactions,7,40 DMA could react with NH2OH and brominecontaining oxidant species (i.e., bromamines, etc.) to form UDMH or a hypothetical brominated unsymmetrical dimethylhydrazine (UDMH-Br), that subsequently would be oxidized by ozone to NDMA. The higher amount of NDMA formed in the presence of bromide can be explained by the improved oxidation of ammonia and amines and the subsequent larger formation of brominated oxidants than NH2OH when bromide is absent. It is likely that brominated nitrogenous oxidants also have higher reactivity than NH2OH leading to increased amount of UDMH or UDMH-Br. DMA was also spiked in the investigated wastewater effluent in order to verify if inorganic water constituents present in this water are able to improve the NDMA yield from DMA. As expected, the net NDMA production in wastewater by spiked DMA was around one and a half times greater than in DI water confirming the important role of inorganic constituents (i.e., ammonia and bromide) even in wastewater matrix. DMF is an organic compound with dimethylamine functional group. It has a low reactivity with ozone (KDMF O3 = 2.4 × 10−1) but it is highly reactive with •OH radicals (KDMF •OH = 1.7 × 109).37 Therefore, DMF may be partially transformed by its reaction with •OH radicals during ozonation. In previous studies, during ozonation of DMF, Lee et al.37 observed low yields of DMA (less than 5%) and Oya et al.13 found very low yields of NDMA (3.5 × 10−4%). In this study, as shown in Table 5, no NDMA was detected unless ammonia was present in water and the NDMA formation is attributable to the oxidation of DMF by hydroxyl radicals. In fact, no NDMA was 10313

dx.doi.org/10.1021/es5011658 | Environ. Sci. Technol. 2014, 48, 10308−10315

Environmental Science & Technology

Article

biological processes to degrade complex molecules and release potential NDMA precursors. More studies into the occurrence and control of inorganic substances, such as nitrite, nitrate, ammonia, hydroxylamine, bromide, and brominated oxidants are also needed.

ozonated according to the ratio O3/DOC = 0.9 mg/mg and the NDMA formation (SI Figure S5) and UV absorbance spectra (SI Figure S6) were compared to those of the filtered samples. The ozonated filtered sample consistently resulted in higher NDMA concentration as compared to the unfiltered ozonated sample. The biggest difference in NDMA formation between unfiltered and filtered samples was observed for the WWTP influent (281 ng/L as compared to 130 ng/L). A difference of 95 ng/L was observed for the primary effluent (895 ng/L versus 800 ng/L), whereas the lowest difference was observed for the final effluent (413 ng/L as compared to 387 ng/L). The differences in UV absorbance spectra for ozonated unfiltered and filtered samples reflect the differences in NDMA formation. In particular, the UV spectra of unfiltered and filtered samples from secondary effluent seems exactly overlapped. Likely, filtration lowers the ozone demand created by total suspended solids (TSS), allowing more ozone to react with NDMA precursors. This effect is most dramatic in influent samples, which would have the highest levels of TSS. Similarly, TSS hinder the degradation of dissolved chromophores in wastewater organic matter resulting in less decreased UV absorbance. In addition, an NDMA FP test was accomplished with unfiltered wastewater effluent samples applying different ozone doses until reaching a plateau in NDMA formation. The ozonation test was repeated with a 0.7 μm filtered sample using the ozone dose corresponding to the highest NDMA formation observed in unfiltered samples (Figure 3). The NDMA formation was again slightly higher in the 0.7 μm filtered samples, confirming the absence of ozone particle-associated NDMA precursors related to the use of DMA-based polyacrylamide polymers. Hence, NDMA formation at the investigated ozonation conditions mainly seems to be related to the reaction between ozone and dissolved precursors supporting the hypothesis of polymer degradation. Environmental Significance and Implications. Results of this study offer important guidance for strategies to minimize the potential of NDMA formation in wastewater and may help to better address future research on NDMA source control. Ozonation of wastewater characterized by elevated ammonia, bromide, and NDMA precursors occurrence can induce high NDMA formation. The absence of ammonia reduces hydroxylamine or brominated nitrogenous oxidants formation during ozonation and NDMA yields of precursors such as ammines and amides will then be significantly lower. Therefore, a complete biological nitrification may represent an effective treatment method for reducing NDMA formation potential. Furthermore, biological treatment for nitrogen removal of supernatants would significantly reduce the ammonia load in the main stream of a wastewater treatment plant43 and may help to degrade NDMA precursors released during the previous sludge treatments. The use of DMA-based polyacrylamide polymers (i.e., Mannich polymer) in sludge thickening and dewatering should be avoided in wastewater treatment plants due to their easily degradable structure that can release elevated amount of dissolved NDMA precursors in the centrates sent back to the head of the treatment plant becoming an important NDMA precursor source in wastewater. This is the first study to identify the main cause of NDMA formation by ozonation within a full scale wastewater treatment plant, including the sludge treatment train, indicating that further research is needed to better evaluate the role of

■

ASSOCIATED CONTENT

S Supporting Information *

Materials; wastewater treatment plant; ozonation; analytical methods; scheme of the investigated WWTP (Figure S1); labscale apparatus used for semi-batch ozonation (Figure S2); synthesis of Mannich polymer (Figure S3); correlation between NDMA formation by ozonation and ammonia concentration in primary influent, primary effluent, secondary effluent, and WAS thickening recycle stream at the studied WWTP (Figure S4); NDMA formation upon ozonation of 0.7 μm filtered and unfiltered samples (Figure S5); and UV absorbance spectra of primary influent, primary effluent, and secondary effluent at the investigated WWTP (Figure S6). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +1 520 6212573; fax: +1 520 6216048; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was primarily funded by United Water Inc. to support collaborative research on disinfection byproducts at the University of Arizona. Funding for M. Sgroi and P. Roccaro was partially supported by the Italian Ministry of Instruction, University, and Research (MIUR), through the Research Projects of National Interest “Reuse of Wastewater in Agriculture: Emerging Pollutants and Operational Problems” (PRIN 2009Grant 20092MES7A_002). Views expressed in this paper do not necessarily reflect those of the funding agencies. P.R. acknowledges the U.S.−Italy Fulbright Commission for supporting his research in the U.S. through the “Fulbright Scholar Program Advanced Research and University Lecturing Awards in the United States”. The authors acknowledge Dr. Armando Durazo from the University of Arizona for his assistance in the use of laboratory equipment. The authors are also grateful to Joe Weitzel from Agilent Technologies (Santa Clara, California) and Jens Scheideler of Wedeco (Herdford, Germany) for their equipment and technical support used in this study.

■

REFERENCES

(1) United States Environmental Protection Agency. IRIS Database (1993). http://www.epa.gov/ncea/iris/. (2) California’s Department of Public Health. NDMA and other nitrosaminesDrinking water issues (2009). http://www.cdph.ca. gov/certlic/drinkingwater/Pages/NDMA.aspx. (3) California’s Office of Environmental Health Hazard Assessment. Public Health Goal for Chemicals in Drinking Water: N-Nitrosodimethylamine (2006). http://www.oehha.ca.gov/water/phg/pdf/ 122206NDMAphg.pdf. (4) USEPA. Unregulated Contaminant Monitoring Rule 2 (UCMR 2), 2012. http://water.epa.gov/lawsregs/rulesregs/sdwa/ucmr/data. cfm#ucmr1997. 10314

dx.doi.org/10.1021/es5011658 | Environ. Sci. Technol. 2014, 48, 10308−10315

Environmental Science & Technology

Article

(5) USEPA. Contaminant Candidate List 3 (CCL3), 2009. http:// water.epa.gov/scitech/drinkingwater/dws/ccl/ccl3.cfm. (6) Schreiber, I. M.; Mitch, W. A. Influence of the order of reagent addition on NDMA formation during chloramination. Environ. Sci. Technol. 2005, 39 (10), 3811e3818. (7) Schreiber, I. M.; Mitch, W. A. Nitrosamine formation pathway revisited: The importance of chloramine speciation and dissolved oxygen. Environ. Sci. Technol. 2006, 40 (19), 6007−6014. (8) Najm, I.; Trussel, R. R. NDMA formation in water and wastewater. J. Am. Water Works Assoc. 2001, 93, 92−99. (9) Andrzejewski, P.; Kasprzyk-Hordern, B.; Nawrocki, J. Nnitrosodimethylamine /NDMA) formation during ozonation of dimethylamine-containing waters. Water Res. 2007, 42, 863−870. (10) Snyder, S. A.; Wert, E. C.; Rexing, D. J.; Zegers, R. E.; Drury, D. D. Ozone oxidation of endocrine disruptor and pharmaceuticals in surface water and wastwater. Ozone: Sci. Eng. 2006, 28 (6), 445−460. (11) Rakness, K. L.; Corsaro, K. M.; Hale, G.; Blank, B. D. Wastewater disinfection with ozoneProcess control and operating results. Ozone: Sci. Eng. 1993, 15 (6), 497−514. (12) Standford, B. D.; Pisarenko, A. N.; Holbrook, D. R.; Snyder, S. A. Preozonation effects on the reduction of reverse osmosis membrane fouling in water reuse. Ozone: Sci. Eng. 2011, 33, 379−388. (13) Oya, M.; Kosaka, K.; Asami, M.; Kunikane, S. Formation of Nnitrosodimethylamine (NDMA) by ozonation of dyes and related compounds. Chemosphere 2008, 73, 1724−1730. (14) Yang, L.; Zhonglin, C.; Shen, J.; Xu, Z.; Liang, H.; Tian, J.; Ben, Y.; Zhai, X.; Shi, W.; Li, G. Reinvestigation of the nitrosamineformation mechanism during ozonation. Environ. Sci. Technol. 2009, 43, 5481−5487. (15) Kosaka, K.; Asami, M.; Konno, Y.; Oya, M.; Kunikane, S. Identification of antiyellowing agents as precursors of N-nitrosodimethylamine production on ozonation from sewage treatment plant influent. Environ. Sci. Technol. 2009, 43, 5236−5241. (16) Schmidt, C. K.; Brauch, H. J. N,N-Dimethylsufamide as precursor for N-nitrosodimethylamine (NDMA) formation upon ozonation and its fate during drinking water treatment. Environ. Sci. Technol. 2008, 42, 6340−6346. (17) Padhye, L.; Luzinova, Y.; Cho, M.; Mizaikoff, B.; Kim, J.; Huag, C. PolyDADMAC and dimethylamine as precursors of N-nitrosodimethylamine during ozonation: reaction kinetics and mechanism. Environ. Sci. Technol. 2011, 45, 4353−4359.22. (18) Andrzejewski, P.; Fijolek, L.; Nawrocki, J. An influence of hypothetical products of dimethylamine ozonation on N-nitrosodimethylamine formation. J. Hazardous Mater. 2012, 229−230, 340−345. (19) Von Gunten, U.; Salhi, E.; Schmidt, C. K.; Arnold, W. A. Kinetics and mechanisms of N-nitrosodimethylamine formation upon ozonation of N,N-dimethylsulfamidecontaining waters: bromide catalysis. Environ. Sci. Technol. 2010, 44 (15), 5762−5768. (20) Park, S.-H.; Piyachaturawat, P.; Taylor, A. E.; Huang, C. H. Potential N-nitrosodimethylamine (NDMA) formation from aminebased water treatment polymers in the reactions with chlorine-based oxidants and nitrosifying agents. Water Sci. Techno.: Water Supply 2009, 9 (3), 279−288. (21) Park, S.-H.; Wei, S.; Mizaikoff, B.; Taylor, A. E.; Favero, C.; Huang, C. H. Degradation of amine-based water treatment polymers during chloramination as N-nitrosodimethylamine (NDMA) precursors. Environ. Sci. Technol. 2011, 45, 8290−8297. (22) Gerecke, A. C.; Sedlak, D. L. Precursors of N-nitrosodimethylamine in natural waters. Environ. Sci. Technol. 2003, 37 (7), 1331− 1336. (23) Sedlak, D. L.; Deeb, R. A.; Hawley, E. L.; Mitch, W. A.; Durbin, T. D.; Mowbray, S.; Carr, S. Source and fate of nitrosodimethylamine and its precursors in municipal wastwater treatment plants. Water Environ. Res. 2005, 77 (1), 32−39. (24) Krauss, M.; Longrée, P.; Dorusch, F.; Ort, C.; Hollender, J. Occurrence and removal of N-nitrosamines in wastewater treatment plants. Water Res. 2009, 43 (17), 32−39.

(25) Krasner, S. W.; Westerhoff, P.; Chen, B.; Rittmann, B. E.; Nam, S.-N.; Amy, G. Impact of wastewater treatment processes on organic carbon, organic nitrogen, and DBP precursors in effluent organic matter. Environ. Sci. Technol. 2009, 43 (8), 2911−2918. (26) Padhye, L.; Tezel, U.; Mitch, W. A.; Pavlostathis, S. G.; Huang, C.-H. Occurrence and fate of nitrosamine and their precursors in municipal sludge and anaerobic digestion system. Environ. Sci. Technol. 2009, 43, 3087−3093. (27) Mitch, W. A.; Sedlak, D. L. Characterization and fate of Nnitrosodimethylamine precursors in municipal wastewater treatment plants. Environ. Sci. Technol. 2004, 38 (5), 1445−1454. (28) Rakness, K. L.; Wert, E. C.; Elovitz, M.; Mahoney, S. Operatorfriendly technique and quality control considerations for indigo colorimetric measurement of ozone residual. Ozone: Sci. Eng. 2010, 32, 33−42. (29) Lee, Y.; Gerrity, D.; Lee, M.; Bogeat, A. E.; Salhi, E.; Gamage, S.; Trenholm, R. A.; Wert, E. C.; Snyder, S. A.; von Gunten, U. Prediction of micropollutant elimination during ozonation of municipal wastewater effluents: Use of kinetic and water specific information. Environ. Sci. Technol. 2013, 47, 5872−5881. (30) Wert, E. C.; Rosario-Ortiz, F. L.; Snyder, S. A. Effect of ozone exposure on the oxidation of trace organic contaminants in wastewater. Water Res. 2009, 43, 1005−1014. (31) Wert, E. C.; Rosario-Ortiz, F. L.; Drury, D. D.; Snyder, S. A. Formation of oxidation byproducts from ozonation of wastewater. Water Res. 2007, 41, 1481−1490. (32) Keefer, L. K.; Roller, P. P. N-Nitrosation by nitrite ion in neutral and basic medium. Science 1973, 181 (4106), 1245−1247. (33) Park, S.-H.; Wei, S.; Huang, C. H.; Mizaikoff, B.; Aral, M. M. A study of the effect of polymers on potential N-nitrosodimethylamine (NDMA) formation at water and wastewater treatment plants. Research project No.: MESL-03-07; June 2007, p 116; Georgia Institute of Technology: https://smartech.gatech.edu/bitstream/handle/1853/ 28703/MESL_SNF_Report_MESL-03-07S.pdf;jsessionid= C475CF542022377AF2AEADB709F370B5.smart2?sequence=1. (34) Tezel, U.; Padhye, L.; Huang, C.-H; Pavlostathis, S. G. Biotransformation of nitrosamines and precursor secondary amines under methanogenic conditions. Environ. Sci. Technol. 2011, 45, 8290− 8297. (35) Lee, Y.; Von Gunten, U. Oxidative transformation of micropollutants during municipal wastewater treatment: comparison of kinetic aspects of selective (chlorine, chlorine dioxide, ferrateIV, and ozone) and non-selective oxidants (hydroxyl radical). Water Res. 2010, 44, 555−566. (36) Hoigné, J.; Bader, H. Ozonation of water: Kinetics of oxidation of ammonia by ozone and hydroxyl radicals. Environ. Sci. Technol. 1978, 12 (1), 79−84. (37) Lee, C.; Schmidt, C.; Yoon, J.; Von Gunten, U. Oxidation of Nnitrosodimethylamine (NDMA) precursors with ozone and chlorine dioxide: Kinetics and effect on NDMA formation potential. Environ. Sci. Technol. 2007, 41, 2056−2063. (38) Singer, P. C.; Zilli, W. B. Ozonation of ammonia in wastewater. Water Res. 1975, 9, 127−134. (39) Yang, M.; Uesugi, K.; Myoga, H. Ammonia removal in bubble column by ozonation in the presence of bromide. Water Res. 1999, 33 (8), 1911−1999. (40) Le Roux, J.; Gallard, H.; Croue, J. P. Formation of NDMA and halogenated DBPs by chloramination of tertiary amines: The influence of bromide ion. Environ. Sci. Technol. 2012, 46, 1581−1589. (41) Haag, W. R.; Hoigné, J.; Bader, H. Improved ammonia oxidation by ozone in the presence of bromide ion during water treatment. Water Res. 1984, 18 (9), 1125−1128. (42) Luh, J.; Marinas, B. J. Bromide ion effect on N-nitrosodimethylamine formation by monochloramine. Environ. Sci. Technol. 2012, 46, 5085−5092. (43) Fux, C.; Boehler, M.; Huber, P.; Brunner, I.; Siegrist, H. Biological treatment of ammonium-rich wastewater by partial nitritation and subsequent anaerobic ammonium oxidation (anammox) in a pilot plant. J. Biotechnol. 2002, 99, 295−306. 10315

dx.doi.org/10.1021/es5011658 | Environ. Sci. Technol. 2014, 48, 10308−10315