Environmental Toxicology and Chemistry, Vol. 33, No. 8, pp. 1720–1725, 2014 # 2014 SETAC Printed in the USA
Environmental Challenges in China OCCURRENCE AND DISTRIBUTION OF ORGANOPHOSPHATE FLAME RETARDANTS/PLASTICIZERS IN WASTEWATER TREATMENT PLANT SLUDGES FROM THE PEARL RIVER DELTA, CHINA XIANGYING ZENG,y LIXIONG HE,yz SHUXIA CAO,x SHENGTAO MA,yz ZHIQIANG YU,*y HONGYAN GUI,k GUOYING SHENG,y and JIAMO FUy yState Key Laboratory of Organic Geochemistry, Guangdong Key Laboratory of Environment and Resources, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, China zUniversity of Chinese Academy of Sciences, Beijing, China xMaterial Science Laboratory, Technologies Development (Dongguan) Co., Dongguan, China kGuangzhou Development District Construction and Environment Administration Bureau, Guangzhou, China (Submitted 3 December 2013; Returned for Revision 30 December 2013; Accepted 31 March 2014) Abstract: Organophosphate esters (OPs) are widely used as flame retardants or plasticizers and are ubiquitously distributed in the environment. In the present study, the occurrence and distribution of 7 widely used OPs were analyzed in sludge samples collected from 19 municipal wastewater treatment plants in the Pearl River Delta, South China. All analytes were detected in these samples, and the total concentration of OPs ranged from 96.7 mg/kg to 1312.9 mg/kg dry weight, with a mean value of 420.1 mg/kg dry weight. In most sludge samples OPs exhibited a similar distribution pattern, for example, tris(2-butoxyethyl) phosphate (TBEP) and triphenyl phosphate (TPhP) were identified as the dominant compounds. However, the results also indicated significantly higher levels of OPs in specific sludges, such as tri-n-butyl phosphate (804.9 mg/kg), TBEP (783.7 mg/kg), TPhP (656.7 mg/kg), and tritolyl phosphate (265.0 mg/kg), which implied different discharge sources in the studied areas. Environ Toxicol Chem 2014;33:1720–1725. # 2014 SETAC Keywords: Organophosphate esters
Flame retardants
Plasticizers
Sludge
their obvious toxicities and widespread usage, TCEP, TCPP, and TDCPP were included in the second (1995) and fourth (2000) European Union priority lists for risk assessment [5]. Organophosphate esters are easily released or leached into the ambient environment from OP-containing products during their life cycle [1,4,7,10–12]. Accordingly, many studies have indicated that OPs are ubiquitous in the environment, being found in air and dust [7,11,13–15]; wastewater and sludge [15– 17]; surface water, groundwater, and drinking water [15,17–23]; sediment [15,24,25]; soil [26]; and in biota samples and human urine [27–29]. However, most environmental studies have focused on the occurrence and distribution of these compounds in air and dust, and limited information is available pertaining to sludge, despite it being an important pathway for organic pollutants to enter soil [3,30]. Furthermore, certain chemicals in sludge might serve as crucial indicators of industrial location and usage in local regions; therefore, sludge is the optimal matrix for assessing the presence of certain contaminants in local regions. The Pearl River Delta, South China, has undergone rapid economic development and urbanization since the 1980 s. This region is a well-known electrical/electronic manufacturing center, as well as an extensive plastic and textile manufacturing zone. Previous studies have shown significantly high concentrations of flame retardants, including polybrominated diphenyl ethers (PBDEs) and several novel brominated flame retardants (decabromodiphenyl ethane, tetrabromobisphenol A bis[2, 3dibromopropyl ether]) in the sediment in this area, indicating that these harmful chemicals might pose a high ecological risk to local biota [31,32]. As mentioned, OPs are widely used flame retardants in diverse products, and we can safely speculate that OPs are ubiquitous in this region because of their extensive application in the electrical/electronic, plastic, and textile manufacturing
INTRODUCTION
Organophosphate esters (OPs) are widely used as flame retardants or plasticizers in industrial and household products. Generally, chlorinated OPs such as tris(2-chloroethyl) phosphate (TCEP), tris(2-chloroiso-propyl) phosphate (TCPP), and tris(1, 3-dichloro-2-propyl) phosphate (TDCPP) are used as flame retardants in flexible or rigid polyurethane foam, rubber, electronic equipment, textile coatings, and furniture. Conversely, nonchlorinated OPs such as tri-n-butyl phosphate (TBP) and triphenyl phosphate (TPhP) are used as plasticizers, extreme pressure additives, or antiwear agents in hydraulic fluids, lubricants, and transmission and motor oils. Triphenyl phosphate is also used as a flame retardant and plasticizer in polyvinylchloride (PVC), cables and water pipes. Tris(2butoxyethyl) phosphate (TBEP) is commonly added to rubber and PVC [1–5], and tritolyl phosphate (TTP) is used in hydraulic fluids, PVC, cellulose, plastic, and transmission fluid [6]. Detailed information pertaining to the properties, production, and application of these compounds is available elsewhere [6]. Organophosphate esters are listed as high-productionvolume chemicals because of the rate at which they are consumed worldwide [7]. Extensive studies have demonstrated their adverse effects on animals and humans, which include neurotoxic [8], carcinogenic, and mutagenic or hormone disturbance activities [9]. Detailed information regarding their adverse effects on organisms and human health, as well as their risk assessment, have been reviewed recently [6]. Because of All Supplemental Data may be found in the online version of this article. * Address correspondence to [email protected]. Published online 14 April 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2604 1720
Organophosphate esters in sludge from the Pearl River Delta
industries. However, few data are available regarding the occurrence and distribution of organophosphate flame retardants in the Pearl River Delta. The levels and distribution of contaminants in sludge could reflect roughly the pollution status in the local region. Thus, in the present study, 19 representative municipal wastewater treatment plants (WWTPs) in 6 cities located in the Pearl River Delta were selected as target plants on the basis of wastewater origin, treatment technology, and service area, as well as inhabitants within their catchment area. Dewatered sludge from these target WWTPs was collected and investigated for the occurrence and distribution of 7 commonly used OPs, and preliminary information regarding the contamination status of organophosphate flame retardants/ plasticizers in the region was obtained. MATERIALS AND METHODS
Chemicals
Organophosphate ester standards were purchased from Sigma-Aldrich. The purity of each compound was as follows: TBP, 99%; TPhP, 99%; TCEP, 97%; TBEP, 94%; TTP, 90%; TDCPP, 97%; and TCPP, 99.5%. Surrogate standard tri-n-butyld27 phosphate (d27-TBP, 98%) was purchased from C/D/N Isotopes. All solvents were of chromatographic grade. Acetonitrile and methanol were acquired from Merck, and ethyl acetate was obtained from CNW Technologies GmbH. Oasis hydrophilic-lipophilic-balanced (HLB) extraction cartridges (200 mg, 6 mL) were purchased from Waters Corporation, and the internal standard hexamethylbenzene (HMB; 99.5%), was acquired from Ehrenstofer-Schäfer Bgm-Schlosser. Sample collection
Sludge samples were collected from 19 municipal WWTPs located in 6 cities in the Pearl River Delta. Of the 19 WWTPs, 9 primarily treated domestic sewage, 2 treated industrial wastewater, and the remainder treated a mixture of domestic sewage and industrial wastewater with different ratios. Detailed descriptions of these WWTPs have been published previously [33] and also can be found in the Supplemental Data, Table S1. The sludge samples were freeze-dried, ground, and homogenized by sieving through a stainless steel 80 sieves (178 mm), after which they were stored in an amber glass bottle until extraction. Extraction and cleanup
The analytes were extracted using ultrasonic assistance as described previously, with slight modification [25,34]. Briefly, approximately 0.5 g dry sludge was loaded into the bottom of a 50-mL Teflon centrifuge tube (Nalge Nunc International), spiked with 5 mL d27-TBP at 10 mg/L as a surrogate, and extracted by ultrasonic assistance with a 20-mL mixture of acetonitrile/water (25:75, v/v) for 30 min at 20 8C. The raw extract was then centrifuged at 6163 g and 20 8C for 15 min (Anke TGL-10B; Shanghai Anting Scientific Instrument Factory), after which the supernatant was decanted into a flask. Next, the residual sludge was extracted again, following the same procedure. Thereafter, the extract was diluted with ultrapure water to 500 mL and then subjected to solid-phase extraction for concentration and cleanup. An Oasis HLB cartridge (200 mg, 6 mL) was conditioned with 4 mL ethyl acetate, followed by 4 mL methanol and 4 mL ultrapure water. After the extract was loaded, the cartridge was dried using highpurity N2, and the analytes were finally eluted with 8 mL ethyl
Environ Toxicol Chem 33, 2014
1721
acetate. The eluate was then concentrated to almost dryness under gentle N2, after which it was redissolved in 100 mL n-hexane. Five microliters of internal standard HMB at concentration 20 mg/L was added before instrumental analysis. Instrumental analysis
Determination and identification of OPs were performed using a Shimadzu 2010 gas chromatograph equipped with a mass spectrometer using a TG-5 ms column (30 m 0.25 mm inner diameter 0.25 mm film; Thermo Technology). The oven temperature was programmed as follows: 70 8C for 2 min, followed by an increase of 10 8C/min to 160 8C, then 5 8C/min to 295 8C, where it was held for 15 min. Helium was used as the carrier gas at 37 cm/sec, and the injection was in splitless mode at 290 8C. The mass spectrometer was operated in selected ion mode with EI-mode, and the source temperature and transfer line were set at 220 8C and 300 8C, respectively. Quantification of analytes was carried out based on a 6-point calibration curve, using standards at concentrations ranging from 50 mg/L to 5000 mg/L. Good correlation coefficients (r2) between 0.9975 and 0.9997 were obtained. Detailed information pertaining to the quantitative ion and qualitative ion for each chemical have been reported elsewhere [34]. The limit of detection (LOD) for individual compounds was calculated according to the suggestion proposed by the Wisconsin (USA) Department of Natural Resources [35]. The limit of quantification (LOQ) was the minimum amount that could be quantitated accurately. For the chemicals present in blanks, such as TCEP and TCPP, the LOQ was defined as twice as the maximum value detected in blanks, whereas LOQ was defined as twice the LOD for those not detected (TBP, TBEP, and TDCPP) or those found below the LOD (TPhP and TTP) in blanks. In the present study, the LOD was 1.6 mg/kg for TBP and TCPP; 0.8 mg/kg for TBEP; 2.0 mg/kg for TCEP, TPhP, and TTP; and 3.0 mg/kg for TDCPP. Their LOQs also are listed in Table 1. Quality control and quality assurance
Procedural blanks (n ¼ 4), spiked blanks (standards spiked into solvent, n ¼ 4), spiked matrix (standards spiked into preextracted sludge, n ¼ 4), and replicate samples (n ¼ 4) were analyzed with real samples in every batch. In each spiked sample, a 100-ng mixture of 7 OPs was added. All samples, including quality assurance/quality control samples, were spiked with d27-TBP as a surrogate. Tri-n-butyl phosphate, TBEP, and TDCPP were not found in blanks; TPhP and TTP were found in every method blank at the LOD; and TCEP and TCPP were detected at 3.4 0.9 mg/kg and 2.0 0.1mg/kg in blanks (Supplemental Data, Figure S1). With the exception of TTP, acceptable recoveries were achieved with ranges of 85 4.4% to 108 4.4%, and 78 2.3% to 117 7.0% being observed for spiked blanks and spiked matrix, respectively (Table 1). Considerably lower recoveries for TTP were achieved in spiked sludge (33 12.0%) than in spiked blanks (85 4.4%) and spiked sediments (66 8%) [34]. Furthermore, the recoveries of OPs were also evaluated by addition of a 100-ng mixture of standards containing 7 OPs into 3 types of sludge samples, including those from plant LD (domestic sewage), plant SGSC (industrial wastewater), and plant DTS (mixed wastewater), and the results are listed in Table 1. Table 1 shows that acceptable recoveries have been achieved in the range of 83 4.1% and 102 11.8%, with the exception of TTP (43 9.1%). The concentrations of OPs (Table 2) measured in sludge samples were background-subtracted but not recovery corrected.
1722
Environ Toxicol Chem 33, 2014
X. Zeng et al.
Table 1. Limits of detection, limits of quantification, and recoveries of seven organophosphate flame retardants/plasticizers Recoveries from spiked samples (%) Compound TBP TBEP TCEP TCPP TDCPP TPhP TTP
LOD (mg/kg)
LOQ (mg/kg)
Blanks
Pretreated sludge
Untreated sludge
1.6 0.8 2.0 1.6 3.0 2.1 1.9
3.2 1.6 8.6 4.2 6.0 4.2 3.8
102 5.0 97 3.2 106 1.1 102 0.3 108 4.4 101 3.0 85 4.4
104 7.7 117 7.0 106 4.3 93 2.1 95 0.1 78 2.3 33 12.0
99 7.0 93 5.5 93 12.8 101 14.4 102 11.8 83 4.1 43 9.1
LOD ¼ limit of detection; LOQ ¼ limit of quantification; TBP ¼ tri-n-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; TPhP ¼ triphenyl phosphate; TTP ¼ tritolyl phosphate.
Table 2. Concentrations and distributions of organophosphate flame retardant/plasticizers in sludge in the Pearl River Deltaa Treatment plant code names
TBP
TBEP
TCEP
TCPP
TPhP
TTP
TDCPP
SOPs
DTS SGSC FSD NHPZ NH XT FSZA KFQ DG KFQD ZHGB SZBH JMWC FSSG ZHGD JM SZLF LD SZNS
20.6 7.7 17.4 11.7 13.9 804.9 7.1 35.7 42.5 117.9 44.0 11.8 14.4 9.7 35.4 14.7 20.5 17.2 165.1
146.7 25.8 28.8 36.0 169.2 25.1 62.6 117.4 93.4 250.5 110.1 783.7 51.7 26.2 269.1 63.6 102.3 119.5 366.5
14.1 7.1 17.1 10.0 8.6 6.9 15.2 14.2 14.4 17.0 11.4 12.1 13.7 8.9 9.8 8.1 11.4 9.5 15.3
15.9 7.6 24.1 7.9 13.1 6.3 16.1 44.2 17.1 54.4 11.0 25.5 35.0 8.1 12.0 8.8 14.5 25.1 43.0
30.2 26.1 321.0 16.3 22.7 LOQ 30.3 165.2 49.9 134.5 39.1 59.5 46.7 62.9 183.8 37.1 58.4 27.5 656.7
14.9 10.6 27.7 5.2 LOQ 9.9 13.6 66.9 8.9 265.0 9.0 8.4 6.5 13.1 69.7 12.0 115.5 11.8 24.8
23.2 11.8 14.6 13.8 15.2 15.0 19.3 16.8 64.0 38.2 16.3 27.0 16.9 12.6 24.9 18.9 35.7 19.6 41.5
265.6 96.7 450.6 100.9 242.7 868.1 164.2 460.4 290.2 877.5 240.9 928.0 184.9 141.5 604.7 163.2 358.3 230.2 1312.9
a Data reported are mg/kg. TBP ¼ tri-n-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TPhP ¼ triphenyl phosphate; TTP ¼ tritolyl phosphate; TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; SOPs ¼ total concentration of organophosphate esters.
RESULTS AND DISCUSSION
Concentration of OPs in sludge samples
The concentration and distribution of 7 OPs in 19 sludge samples are listed in Table 2. The OPs were detected P in all sludge samples, and the total concentration of the 7 OPs ( OPs) ranged from 96.7 mg/kg to 1312.9 mg/kg, showing obvious variation among the WWTPs. Sludge containing industrial input showed comparatively higher OP concentrations than domestic sludge. Specifically, WWTPs treating mixed wastewater had OPs ranging from 265.6 mg/kg to 1312.9 mg/kg. The highest concentration of OPs (1312.9 mg/kg) was found in plant SZNS, followed by plant SZBH (928.0 mg/kg). For the 9 WWTPs that mainly treated domestic sewage, comparable levels of OPs (141.5–290.2 mg/kg) were observed, implying that their major source was domestic sewage. The OPs are commonly added in diverse household applications, including furniture, wallpaper, textiles, cotton, baby products, and electronics. Accordingly, these household applications are important emission sources [6,11,16,36,37]. Until recently, limited data were available regarding the occurrence and distribution of OPs in sludge (Table 3). In 2005, Marklund et al. [3] reported the levels and distribution of 8 OPs (TBP, TCPP, TBEP, TPhP, TCEP, TDCPP, ethylhexyl diphenyl
phosphate, and tri-iso-butyl phosphate) in 11 sludge samples collected from Swedish WWTPs. Specifically, they found that the total concentrations of the 8 OPs varied between 620 mg/kg and 6900 mg/kg. Bester [30] reported considerably higher levels of TCPP (1000–20 000 mg/kg) in 20 sludges from German WWTPs than those detected in the present study (6.3–54.4 mg/kg). Chen and Bester [16] found several OPs in 1 sludge sample collected in Germany. Furthermore, they reported concentrations of TCEP, TDCPP, TBP, and TPhP similar to those observed in the present study; however, TCPP levels (18 400 mg/kg) were significantly higher than those observed in the present study. Profiles of OPs
The compositional profiles of the 7 OPs in these 19 sludge samples are shown in Figure 1. As indicated in Figure 1, the distribution pattern varied with the type of wastewater and industry. In the 8 sludges containing mixed domestic and industrial wastewater, TBEP was dominant in 3 WWTPs (DTS, SZBH, and ZHGD), TPhP was dominant in 3 WWTPs (FSD, KFQ, and SZNS), and TTP was the main component in plant KFQD and plant SZLF, implying that OPs have widespread commercial application. Tri-n-butyl phosphate was the dominant analyte in 1 industrial sludge (plant XT), accounting for
Organophosphate esters in sludge from the Pearl River Delta
Environ Toxicol Chem 33, 2014
1723
Table 3. Organophosphate flame retardants/plasticizers detected in the reported sludge studies (mg/kg)a
Sludge (n ¼ 5) Sludge (n ¼ 11) Sludge Sludge (n ¼ 19)
Location
TBP
TBEP
TCEP
TCPP
TDCPP
TPhP
TTP
References
Germany Sweden Germany China
— 39–850 90 7.1–804.9
— LOD–1900 — 25.1–783.7
— 6.6–110 70 6.9–17.1
1000–20 000 61–1900 18 400 6.3–54.4
— 3.0–260 90 11.8–64.0
— 52–320 400 LOQ–656.7
— — — LOQ–265.0
Bester [30] Marklund et al. [3] Chen and Bester [16] The present study
— indicates that data were not reported. TBP ¼ tri-n-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; TPhP ¼ triphenyl phosphate; TTP ¼ tritolyl phosphate; LOD ¼ limit of detection; LOQ ¼ limit of quantification. a
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%
Industrial wastewater
Domestic wastewater
Combined wastewater
Figure 1. Congener profiles of organophosphate flame retardants/plasticizers in the sludges in the Pearl River Delta. TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; TTP ¼ tritolyl phosphate; TPhP ¼ triphenyl phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TBP ¼ trin-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate.
93% of the total OPs and suggesting obvious industrial input in plant XT. Several factors might affect the concentration and distribution pattern of OPs in sludge, such as their emission source in the catchment area, the wastewater treatment technology adopted in specific WWTPs, and distinctive physicochemical properties for each compound. For example, high concentrations of OPs (1312.9 mg/kg) were found in plant SZNS, with TPhP as the main component, in which mixed wastewater combined with 60% domestic sewage and 40% industrial wastewater was treated using the modifiedP University of Cape Town method. In addition, in the same city, OPs in plant SZBH were detected at 928.0 mg/kg, with TBEP as the main component, in which 91.5% of domestic sewage combined with limited industrial wastewater was P treated using an oxidation ditch. Moreover, 877.5 mg/kg OPs was found in plant KFQD, a small-scale WWTP located in a development zone in Guangzhou City in which mixed wastewater was treated using a sequencing batch reactor. There is no clear explanation for the variations in levels and distribution of OPs in these samples nor a way to identify key influencing factors. We also found that the distribution was substantially different from that in previously published studies. Specifically, TCPP and TBEP were identified as the most abundant OPs in 11 Swedish sludges [3], and TCPP was recognized as the main constituent of German sludge [16], and both TCPP and TEHP were dominant over the other OPs in Spanish sludge [15]. The obviously different distribution may be attributed to their different usages in different countries.
Potential sources
Generally, compositional profiles in sludge would reflect their specific origins in local regions, depending on their usage mode and dosage of different products. For example, as shown in Figure 1, TBP was detected in all samples at levels ranging from 7.1 mg/kg to 44.0 mg/kg and 11.8 mg/kg to 165.1 mg/kg for sludge from domestic wastewater and combined wastewater, respectively. However, the highest concentration (804.9 mg/kg) was detected in plant XT, which treated almost 100% industrial wastewater with a daily capacity of 100 000 tons. Tri-n-butyl phosphate is widely added to plastic products as a plasticizer [3,5] and often as an additive in the textile and clothing dyeing industry [6]; therefore, it is reasonable to assume that a high quantity of TBP was released into wastewater from these businesses, resulting in a higher level of TBP in this WWTP. We also detected significantly higher concentrations of TBEP, TPhP, and TTP in the sludge from the plants SZBH, SZNS and KFDQ, respectively, suggesting that specific emission sources existed in nearby areas [6]. All sludge samples contained relatively lower concentrations of TDCPP, TCPP, and TCEP, suggesting that chlorinated OPs are ubiquitous and have no specific source in the Pearl River Delta. Furthermore, chlorinated OPs have much higher water solubility. For example, the water solubility of TCEP and TCPP are 7000 mg/L and 1200 mg/L, respectively [34]; therefore, they would be distributed primarily in the water phase [38], resulting in their comparatively low presence in the sludge. In addition, chlorinated OPs have been shown to be very resistant to
1724
Environ Toxicol Chem 33, 2014
X. Zeng et al.
Figure 2. Concentrations of organic pollutants reported in the sludges. OPs ¼ organophosphate esters; PBDEs ¼ polybrominated diphenyl ethers; PAHs ¼ polycyclic aromatic hydrocarbons; SMs ¼ synthetic musks.
wastewater treatment technology as well as traditional drinking water treatment technology [22,39,40]. As a result, these chemicals may remain in the effluent and be discharged into the receiving water or remain in drinking water and reach the human body. Consequently, further investigations should be conducted to determine the occurrence and levels of OPs in wastewater and effluent in WWTPs, as well as the receiving water in this region. Several tributaries of the Pearl River serve as drinking water sources; accordingly, much attention should be paid to their occurrence in source water and in drinking water, and to the risk to human health posed by exposure through drinking water.
treatment systems, few have focused on transformation and biodegradation reactions throughout the treatment system, especially for OPs and synthetic musks. The transformation reaction will change the fate of organic contaminants and their toxicity to health and the environment. For example, decabromodiphenyl ether (BDE-209) has low water solubility (
Environmental Challenges in China OCCURRENCE AND DISTRIBUTION OF ORGANOPHOSPHATE FLAME RETARDANTS/PLASTICIZERS IN WASTEWATER TREATMENT PLANT SLUDGES FROM THE PEARL RIVER DELTA, CHINA XIANGYING ZENG,y LIXIONG HE,yz SHUXIA CAO,x SHENGTAO MA,yz ZHIQIANG YU,*y HONGYAN GUI,k GUOYING SHENG,y and JIAMO FUy yState Key Laboratory of Organic Geochemistry, Guangdong Key Laboratory of Environment and Resources, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, China zUniversity of Chinese Academy of Sciences, Beijing, China xMaterial Science Laboratory, Technologies Development (Dongguan) Co., Dongguan, China kGuangzhou Development District Construction and Environment Administration Bureau, Guangzhou, China (Submitted 3 December 2013; Returned for Revision 30 December 2013; Accepted 31 March 2014) Abstract: Organophosphate esters (OPs) are widely used as flame retardants or plasticizers and are ubiquitously distributed in the environment. In the present study, the occurrence and distribution of 7 widely used OPs were analyzed in sludge samples collected from 19 municipal wastewater treatment plants in the Pearl River Delta, South China. All analytes were detected in these samples, and the total concentration of OPs ranged from 96.7 mg/kg to 1312.9 mg/kg dry weight, with a mean value of 420.1 mg/kg dry weight. In most sludge samples OPs exhibited a similar distribution pattern, for example, tris(2-butoxyethyl) phosphate (TBEP) and triphenyl phosphate (TPhP) were identified as the dominant compounds. However, the results also indicated significantly higher levels of OPs in specific sludges, such as tri-n-butyl phosphate (804.9 mg/kg), TBEP (783.7 mg/kg), TPhP (656.7 mg/kg), and tritolyl phosphate (265.0 mg/kg), which implied different discharge sources in the studied areas. Environ Toxicol Chem 2014;33:1720–1725. # 2014 SETAC Keywords: Organophosphate esters
Flame retardants
Plasticizers
Sludge
their obvious toxicities and widespread usage, TCEP, TCPP, and TDCPP were included in the second (1995) and fourth (2000) European Union priority lists for risk assessment [5]. Organophosphate esters are easily released or leached into the ambient environment from OP-containing products during their life cycle [1,4,7,10–12]. Accordingly, many studies have indicated that OPs are ubiquitous in the environment, being found in air and dust [7,11,13–15]; wastewater and sludge [15– 17]; surface water, groundwater, and drinking water [15,17–23]; sediment [15,24,25]; soil [26]; and in biota samples and human urine [27–29]. However, most environmental studies have focused on the occurrence and distribution of these compounds in air and dust, and limited information is available pertaining to sludge, despite it being an important pathway for organic pollutants to enter soil [3,30]. Furthermore, certain chemicals in sludge might serve as crucial indicators of industrial location and usage in local regions; therefore, sludge is the optimal matrix for assessing the presence of certain contaminants in local regions. The Pearl River Delta, South China, has undergone rapid economic development and urbanization since the 1980 s. This region is a well-known electrical/electronic manufacturing center, as well as an extensive plastic and textile manufacturing zone. Previous studies have shown significantly high concentrations of flame retardants, including polybrominated diphenyl ethers (PBDEs) and several novel brominated flame retardants (decabromodiphenyl ethane, tetrabromobisphenol A bis[2, 3dibromopropyl ether]) in the sediment in this area, indicating that these harmful chemicals might pose a high ecological risk to local biota [31,32]. As mentioned, OPs are widely used flame retardants in diverse products, and we can safely speculate that OPs are ubiquitous in this region because of their extensive application in the electrical/electronic, plastic, and textile manufacturing
INTRODUCTION
Organophosphate esters (OPs) are widely used as flame retardants or plasticizers in industrial and household products. Generally, chlorinated OPs such as tris(2-chloroethyl) phosphate (TCEP), tris(2-chloroiso-propyl) phosphate (TCPP), and tris(1, 3-dichloro-2-propyl) phosphate (TDCPP) are used as flame retardants in flexible or rigid polyurethane foam, rubber, electronic equipment, textile coatings, and furniture. Conversely, nonchlorinated OPs such as tri-n-butyl phosphate (TBP) and triphenyl phosphate (TPhP) are used as plasticizers, extreme pressure additives, or antiwear agents in hydraulic fluids, lubricants, and transmission and motor oils. Triphenyl phosphate is also used as a flame retardant and plasticizer in polyvinylchloride (PVC), cables and water pipes. Tris(2butoxyethyl) phosphate (TBEP) is commonly added to rubber and PVC [1–5], and tritolyl phosphate (TTP) is used in hydraulic fluids, PVC, cellulose, plastic, and transmission fluid [6]. Detailed information pertaining to the properties, production, and application of these compounds is available elsewhere [6]. Organophosphate esters are listed as high-productionvolume chemicals because of the rate at which they are consumed worldwide [7]. Extensive studies have demonstrated their adverse effects on animals and humans, which include neurotoxic [8], carcinogenic, and mutagenic or hormone disturbance activities [9]. Detailed information regarding their adverse effects on organisms and human health, as well as their risk assessment, have been reviewed recently [6]. Because of All Supplemental Data may be found in the online version of this article. * Address correspondence to [email protected]. Published online 14 April 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2604 1720
Organophosphate esters in sludge from the Pearl River Delta
industries. However, few data are available regarding the occurrence and distribution of organophosphate flame retardants in the Pearl River Delta. The levels and distribution of contaminants in sludge could reflect roughly the pollution status in the local region. Thus, in the present study, 19 representative municipal wastewater treatment plants (WWTPs) in 6 cities located in the Pearl River Delta were selected as target plants on the basis of wastewater origin, treatment technology, and service area, as well as inhabitants within their catchment area. Dewatered sludge from these target WWTPs was collected and investigated for the occurrence and distribution of 7 commonly used OPs, and preliminary information regarding the contamination status of organophosphate flame retardants/ plasticizers in the region was obtained. MATERIALS AND METHODS
Chemicals
Organophosphate ester standards were purchased from Sigma-Aldrich. The purity of each compound was as follows: TBP, 99%; TPhP, 99%; TCEP, 97%; TBEP, 94%; TTP, 90%; TDCPP, 97%; and TCPP, 99.5%. Surrogate standard tri-n-butyld27 phosphate (d27-TBP, 98%) was purchased from C/D/N Isotopes. All solvents were of chromatographic grade. Acetonitrile and methanol were acquired from Merck, and ethyl acetate was obtained from CNW Technologies GmbH. Oasis hydrophilic-lipophilic-balanced (HLB) extraction cartridges (200 mg, 6 mL) were purchased from Waters Corporation, and the internal standard hexamethylbenzene (HMB; 99.5%), was acquired from Ehrenstofer-Schäfer Bgm-Schlosser. Sample collection
Sludge samples were collected from 19 municipal WWTPs located in 6 cities in the Pearl River Delta. Of the 19 WWTPs, 9 primarily treated domestic sewage, 2 treated industrial wastewater, and the remainder treated a mixture of domestic sewage and industrial wastewater with different ratios. Detailed descriptions of these WWTPs have been published previously [33] and also can be found in the Supplemental Data, Table S1. The sludge samples were freeze-dried, ground, and homogenized by sieving through a stainless steel 80 sieves (178 mm), after which they were stored in an amber glass bottle until extraction. Extraction and cleanup
The analytes were extracted using ultrasonic assistance as described previously, with slight modification [25,34]. Briefly, approximately 0.5 g dry sludge was loaded into the bottom of a 50-mL Teflon centrifuge tube (Nalge Nunc International), spiked with 5 mL d27-TBP at 10 mg/L as a surrogate, and extracted by ultrasonic assistance with a 20-mL mixture of acetonitrile/water (25:75, v/v) for 30 min at 20 8C. The raw extract was then centrifuged at 6163 g and 20 8C for 15 min (Anke TGL-10B; Shanghai Anting Scientific Instrument Factory), after which the supernatant was decanted into a flask. Next, the residual sludge was extracted again, following the same procedure. Thereafter, the extract was diluted with ultrapure water to 500 mL and then subjected to solid-phase extraction for concentration and cleanup. An Oasis HLB cartridge (200 mg, 6 mL) was conditioned with 4 mL ethyl acetate, followed by 4 mL methanol and 4 mL ultrapure water. After the extract was loaded, the cartridge was dried using highpurity N2, and the analytes were finally eluted with 8 mL ethyl
Environ Toxicol Chem 33, 2014
1721
acetate. The eluate was then concentrated to almost dryness under gentle N2, after which it was redissolved in 100 mL n-hexane. Five microliters of internal standard HMB at concentration 20 mg/L was added before instrumental analysis. Instrumental analysis
Determination and identification of OPs were performed using a Shimadzu 2010 gas chromatograph equipped with a mass spectrometer using a TG-5 ms column (30 m 0.25 mm inner diameter 0.25 mm film; Thermo Technology). The oven temperature was programmed as follows: 70 8C for 2 min, followed by an increase of 10 8C/min to 160 8C, then 5 8C/min to 295 8C, where it was held for 15 min. Helium was used as the carrier gas at 37 cm/sec, and the injection was in splitless mode at 290 8C. The mass spectrometer was operated in selected ion mode with EI-mode, and the source temperature and transfer line were set at 220 8C and 300 8C, respectively. Quantification of analytes was carried out based on a 6-point calibration curve, using standards at concentrations ranging from 50 mg/L to 5000 mg/L. Good correlation coefficients (r2) between 0.9975 and 0.9997 were obtained. Detailed information pertaining to the quantitative ion and qualitative ion for each chemical have been reported elsewhere [34]. The limit of detection (LOD) for individual compounds was calculated according to the suggestion proposed by the Wisconsin (USA) Department of Natural Resources [35]. The limit of quantification (LOQ) was the minimum amount that could be quantitated accurately. For the chemicals present in blanks, such as TCEP and TCPP, the LOQ was defined as twice as the maximum value detected in blanks, whereas LOQ was defined as twice the LOD for those not detected (TBP, TBEP, and TDCPP) or those found below the LOD (TPhP and TTP) in blanks. In the present study, the LOD was 1.6 mg/kg for TBP and TCPP; 0.8 mg/kg for TBEP; 2.0 mg/kg for TCEP, TPhP, and TTP; and 3.0 mg/kg for TDCPP. Their LOQs also are listed in Table 1. Quality control and quality assurance
Procedural blanks (n ¼ 4), spiked blanks (standards spiked into solvent, n ¼ 4), spiked matrix (standards spiked into preextracted sludge, n ¼ 4), and replicate samples (n ¼ 4) were analyzed with real samples in every batch. In each spiked sample, a 100-ng mixture of 7 OPs was added. All samples, including quality assurance/quality control samples, were spiked with d27-TBP as a surrogate. Tri-n-butyl phosphate, TBEP, and TDCPP were not found in blanks; TPhP and TTP were found in every method blank at the LOD; and TCEP and TCPP were detected at 3.4 0.9 mg/kg and 2.0 0.1mg/kg in blanks (Supplemental Data, Figure S1). With the exception of TTP, acceptable recoveries were achieved with ranges of 85 4.4% to 108 4.4%, and 78 2.3% to 117 7.0% being observed for spiked blanks and spiked matrix, respectively (Table 1). Considerably lower recoveries for TTP were achieved in spiked sludge (33 12.0%) than in spiked blanks (85 4.4%) and spiked sediments (66 8%) [34]. Furthermore, the recoveries of OPs were also evaluated by addition of a 100-ng mixture of standards containing 7 OPs into 3 types of sludge samples, including those from plant LD (domestic sewage), plant SGSC (industrial wastewater), and plant DTS (mixed wastewater), and the results are listed in Table 1. Table 1 shows that acceptable recoveries have been achieved in the range of 83 4.1% and 102 11.8%, with the exception of TTP (43 9.1%). The concentrations of OPs (Table 2) measured in sludge samples were background-subtracted but not recovery corrected.
1722
Environ Toxicol Chem 33, 2014
X. Zeng et al.
Table 1. Limits of detection, limits of quantification, and recoveries of seven organophosphate flame retardants/plasticizers Recoveries from spiked samples (%) Compound TBP TBEP TCEP TCPP TDCPP TPhP TTP
LOD (mg/kg)
LOQ (mg/kg)
Blanks
Pretreated sludge
Untreated sludge
1.6 0.8 2.0 1.6 3.0 2.1 1.9
3.2 1.6 8.6 4.2 6.0 4.2 3.8
102 5.0 97 3.2 106 1.1 102 0.3 108 4.4 101 3.0 85 4.4
104 7.7 117 7.0 106 4.3 93 2.1 95 0.1 78 2.3 33 12.0
99 7.0 93 5.5 93 12.8 101 14.4 102 11.8 83 4.1 43 9.1
LOD ¼ limit of detection; LOQ ¼ limit of quantification; TBP ¼ tri-n-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; TPhP ¼ triphenyl phosphate; TTP ¼ tritolyl phosphate.
Table 2. Concentrations and distributions of organophosphate flame retardant/plasticizers in sludge in the Pearl River Deltaa Treatment plant code names
TBP
TBEP
TCEP
TCPP
TPhP
TTP
TDCPP
SOPs
DTS SGSC FSD NHPZ NH XT FSZA KFQ DG KFQD ZHGB SZBH JMWC FSSG ZHGD JM SZLF LD SZNS
20.6 7.7 17.4 11.7 13.9 804.9 7.1 35.7 42.5 117.9 44.0 11.8 14.4 9.7 35.4 14.7 20.5 17.2 165.1
146.7 25.8 28.8 36.0 169.2 25.1 62.6 117.4 93.4 250.5 110.1 783.7 51.7 26.2 269.1 63.6 102.3 119.5 366.5
14.1 7.1 17.1 10.0 8.6 6.9 15.2 14.2 14.4 17.0 11.4 12.1 13.7 8.9 9.8 8.1 11.4 9.5 15.3
15.9 7.6 24.1 7.9 13.1 6.3 16.1 44.2 17.1 54.4 11.0 25.5 35.0 8.1 12.0 8.8 14.5 25.1 43.0
30.2 26.1 321.0 16.3 22.7 LOQ 30.3 165.2 49.9 134.5 39.1 59.5 46.7 62.9 183.8 37.1 58.4 27.5 656.7
14.9 10.6 27.7 5.2 LOQ 9.9 13.6 66.9 8.9 265.0 9.0 8.4 6.5 13.1 69.7 12.0 115.5 11.8 24.8
23.2 11.8 14.6 13.8 15.2 15.0 19.3 16.8 64.0 38.2 16.3 27.0 16.9 12.6 24.9 18.9 35.7 19.6 41.5
265.6 96.7 450.6 100.9 242.7 868.1 164.2 460.4 290.2 877.5 240.9 928.0 184.9 141.5 604.7 163.2 358.3 230.2 1312.9
a Data reported are mg/kg. TBP ¼ tri-n-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TPhP ¼ triphenyl phosphate; TTP ¼ tritolyl phosphate; TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; SOPs ¼ total concentration of organophosphate esters.
RESULTS AND DISCUSSION
Concentration of OPs in sludge samples
The concentration and distribution of 7 OPs in 19 sludge samples are listed in Table 2. The OPs were detected P in all sludge samples, and the total concentration of the 7 OPs ( OPs) ranged from 96.7 mg/kg to 1312.9 mg/kg, showing obvious variation among the WWTPs. Sludge containing industrial input showed comparatively higher OP concentrations than domestic sludge. Specifically, WWTPs treating mixed wastewater had OPs ranging from 265.6 mg/kg to 1312.9 mg/kg. The highest concentration of OPs (1312.9 mg/kg) was found in plant SZNS, followed by plant SZBH (928.0 mg/kg). For the 9 WWTPs that mainly treated domestic sewage, comparable levels of OPs (141.5–290.2 mg/kg) were observed, implying that their major source was domestic sewage. The OPs are commonly added in diverse household applications, including furniture, wallpaper, textiles, cotton, baby products, and electronics. Accordingly, these household applications are important emission sources [6,11,16,36,37]. Until recently, limited data were available regarding the occurrence and distribution of OPs in sludge (Table 3). In 2005, Marklund et al. [3] reported the levels and distribution of 8 OPs (TBP, TCPP, TBEP, TPhP, TCEP, TDCPP, ethylhexyl diphenyl
phosphate, and tri-iso-butyl phosphate) in 11 sludge samples collected from Swedish WWTPs. Specifically, they found that the total concentrations of the 8 OPs varied between 620 mg/kg and 6900 mg/kg. Bester [30] reported considerably higher levels of TCPP (1000–20 000 mg/kg) in 20 sludges from German WWTPs than those detected in the present study (6.3–54.4 mg/kg). Chen and Bester [16] found several OPs in 1 sludge sample collected in Germany. Furthermore, they reported concentrations of TCEP, TDCPP, TBP, and TPhP similar to those observed in the present study; however, TCPP levels (18 400 mg/kg) were significantly higher than those observed in the present study. Profiles of OPs
The compositional profiles of the 7 OPs in these 19 sludge samples are shown in Figure 1. As indicated in Figure 1, the distribution pattern varied with the type of wastewater and industry. In the 8 sludges containing mixed domestic and industrial wastewater, TBEP was dominant in 3 WWTPs (DTS, SZBH, and ZHGD), TPhP was dominant in 3 WWTPs (FSD, KFQ, and SZNS), and TTP was the main component in plant KFQD and plant SZLF, implying that OPs have widespread commercial application. Tri-n-butyl phosphate was the dominant analyte in 1 industrial sludge (plant XT), accounting for
Organophosphate esters in sludge from the Pearl River Delta
Environ Toxicol Chem 33, 2014
1723
Table 3. Organophosphate flame retardants/plasticizers detected in the reported sludge studies (mg/kg)a
Sludge (n ¼ 5) Sludge (n ¼ 11) Sludge Sludge (n ¼ 19)
Location
TBP
TBEP
TCEP
TCPP
TDCPP
TPhP
TTP
References
Germany Sweden Germany China
— 39–850 90 7.1–804.9
— LOD–1900 — 25.1–783.7
— 6.6–110 70 6.9–17.1
1000–20 000 61–1900 18 400 6.3–54.4
— 3.0–260 90 11.8–64.0
— 52–320 400 LOQ–656.7
— — — LOQ–265.0
Bester [30] Marklund et al. [3] Chen and Bester [16] The present study
— indicates that data were not reported. TBP ¼ tri-n-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; TPhP ¼ triphenyl phosphate; TTP ¼ tritolyl phosphate; LOD ¼ limit of detection; LOQ ¼ limit of quantification. a
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%
Industrial wastewater
Domestic wastewater
Combined wastewater
Figure 1. Congener profiles of organophosphate flame retardants/plasticizers in the sludges in the Pearl River Delta. TDCPP ¼ tris(1, 3-dichloro-2-propyl) phosphate; TTP ¼ tritolyl phosphate; TPhP ¼ triphenyl phosphate; TCPP ¼ tris(2-chloroiso-propyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TBP ¼ trin-butyl phosphate; TBEP ¼ tris(2-butoxyethyl) phosphate.
93% of the total OPs and suggesting obvious industrial input in plant XT. Several factors might affect the concentration and distribution pattern of OPs in sludge, such as their emission source in the catchment area, the wastewater treatment technology adopted in specific WWTPs, and distinctive physicochemical properties for each compound. For example, high concentrations of OPs (1312.9 mg/kg) were found in plant SZNS, with TPhP as the main component, in which mixed wastewater combined with 60% domestic sewage and 40% industrial wastewater was treated using the modifiedP University of Cape Town method. In addition, in the same city, OPs in plant SZBH were detected at 928.0 mg/kg, with TBEP as the main component, in which 91.5% of domestic sewage combined with limited industrial wastewater was P treated using an oxidation ditch. Moreover, 877.5 mg/kg OPs was found in plant KFQD, a small-scale WWTP located in a development zone in Guangzhou City in which mixed wastewater was treated using a sequencing batch reactor. There is no clear explanation for the variations in levels and distribution of OPs in these samples nor a way to identify key influencing factors. We also found that the distribution was substantially different from that in previously published studies. Specifically, TCPP and TBEP were identified as the most abundant OPs in 11 Swedish sludges [3], and TCPP was recognized as the main constituent of German sludge [16], and both TCPP and TEHP were dominant over the other OPs in Spanish sludge [15]. The obviously different distribution may be attributed to their different usages in different countries.
Potential sources
Generally, compositional profiles in sludge would reflect their specific origins in local regions, depending on their usage mode and dosage of different products. For example, as shown in Figure 1, TBP was detected in all samples at levels ranging from 7.1 mg/kg to 44.0 mg/kg and 11.8 mg/kg to 165.1 mg/kg for sludge from domestic wastewater and combined wastewater, respectively. However, the highest concentration (804.9 mg/kg) was detected in plant XT, which treated almost 100% industrial wastewater with a daily capacity of 100 000 tons. Tri-n-butyl phosphate is widely added to plastic products as a plasticizer [3,5] and often as an additive in the textile and clothing dyeing industry [6]; therefore, it is reasonable to assume that a high quantity of TBP was released into wastewater from these businesses, resulting in a higher level of TBP in this WWTP. We also detected significantly higher concentrations of TBEP, TPhP, and TTP in the sludge from the plants SZBH, SZNS and KFDQ, respectively, suggesting that specific emission sources existed in nearby areas [6]. All sludge samples contained relatively lower concentrations of TDCPP, TCPP, and TCEP, suggesting that chlorinated OPs are ubiquitous and have no specific source in the Pearl River Delta. Furthermore, chlorinated OPs have much higher water solubility. For example, the water solubility of TCEP and TCPP are 7000 mg/L and 1200 mg/L, respectively [34]; therefore, they would be distributed primarily in the water phase [38], resulting in their comparatively low presence in the sludge. In addition, chlorinated OPs have been shown to be very resistant to
1724
Environ Toxicol Chem 33, 2014
X. Zeng et al.
Figure 2. Concentrations of organic pollutants reported in the sludges. OPs ¼ organophosphate esters; PBDEs ¼ polybrominated diphenyl ethers; PAHs ¼ polycyclic aromatic hydrocarbons; SMs ¼ synthetic musks.
wastewater treatment technology as well as traditional drinking water treatment technology [22,39,40]. As a result, these chemicals may remain in the effluent and be discharged into the receiving water or remain in drinking water and reach the human body. Consequently, further investigations should be conducted to determine the occurrence and levels of OPs in wastewater and effluent in WWTPs, as well as the receiving water in this region. Several tributaries of the Pearl River serve as drinking water sources; accordingly, much attention should be paid to their occurrence in source water and in drinking water, and to the risk to human health posed by exposure through drinking water.
treatment systems, few have focused on transformation and biodegradation reactions throughout the treatment system, especially for OPs and synthetic musks. The transformation reaction will change the fate of organic contaminants and their toxicity to health and the environment. For example, decabromodiphenyl ether (BDE-209) has low water solubility (
