Environ. Sci. Technol. 2008, 42, 6566–6572
Occurrence and Sources of Perfluorinated Surfactants in Rivers in Japan M I C H I O M U R A K A M I , †,‡ E I J I I M A M U R A , † HIROYUKI SHINOHARA,† KENTARO KIRI,† YUKI MURAMATSU,† ARATA HARADA,§ A N D H I D E S H I G E T A K A D A * ,† Laboratory of Organic Geochemistry (LOG), Institute of Symbiotic Science and Technology, Tokyo University of Agriculture and Technology, Fuchu, Tokyo 183-8509, Japan, “Wisdom of Water” (Suntory), Corporate Sponsored Research Program, Organization for Interdisciplinary Research Projects, The University of Tokyo, 7-3-1 Hongo, Bunkyo, Tokyo 113-8656, Japan, and Water Environment Research Group, Public Works Research Institute, 1-6, Minamihara, Tsukuba, Ibaraki, 305-8516, Japan
Received February 5, 2008. Revised manuscript received May 12, 2008. Accepted May 12, 2008.
We analyzed perfluorinated surfactants (PFSs) in 20 river samples and 5 wastewater secondary effluent samples in Japan to reveal their occurrence and sources. Nine PFS species were determined: perfluorooctanesulfonate (PFOS), perfluorooctane sulfonamide (FOSA), perfluoroheptanoate (PFHpA), perfluorooctanoate (PFOA), perfluorononanoate (PFNA), perfluorodecanoate (PFDA), perfluoroundecanoate (PFUA), perfluorododecanoate (PFDDA), and perfluorotridecanoate (PFTDA). PFSs were detected in all rivers, revealing nationwide contamination of rivers. In particular, 11 out of 20 river samples exceeded New Jersey guidance for PFOA in drinking water (40 ng/L). PFOS, PFHpA, PFOA, and PFNA were major species in Japan. Concentrations of PFOS, PFHpA, and PFNA in rivers were strongly correlated with population density, suggesting that the chemicals were derived from urban activities. PFOA showed a significant but weak correlation. We used crotamiton, a marker of sewage effluent, for further source analysis. Concentrations of PFOS, PFHpA, and PFNA were strongly correlated with those of crotamiton, and plots of secondary effluents fell near the regression lines of rivers, indicating that the PFOS, PFHpA, and PFNA in rivers were derived from sewage effluent. On the other hand, PFOA was found at remarkably high levels (54-192 ng/L) in seven river samples containing low levels of crotamiton, suggesting that it was derived from nonsewage point sources, as well as sewage effluent. The total fluxes of sewage-derived PFOS, PFHpA, PFOA, and PFNA from Japan were estimated to be 3.6, 2.6, 5.6, and 2.6 t/year, respectively. This is the first report to identify PFOA in several rivers, derived from nonsewage point sources, by using a marker of sewage effluent.
Introduction Perfluorinated surfactants (PFSs) such as perfluorooctanesulfonate (PFOS; C8F17SO3-) and perfluorooctanoate (PFOA; C7F15COO-) are widely used in industry as surfactants, fire retardants, carpet cleaners, waxes, and paper and textile coatings. PFSs are water soluble and bioaccumulative (1–4). Because the C-F bond is very strong, PFSs are thermally and chemically stable and thus environmentally persistent. PFSs were detected in surface waters (5–14), bays (2, 6, 7, 15), sediments (4, 16), arctic ice (17), aquatic organisms (1–4, 10), polar bears (1), and humans (2, 18). Global distributions of PFSs in oceans were also reported by Yamashita et al. (19). PFSs have adverse effects on animals and humans and are chronic toxic and carcinogenic compounds (20, 21). Recently, the New Jersey, the Department of Environmental Protection developed a preliminary health-based drinking water guidance for PFOA of 40 ng/L. Because of their features, concern about PFSs is increasing. PFS pollution in water has been reported in the U.S.A. (5, 8–10, 12), Canada (8, 22), Europe (13, 14), Japan (2, 6, 7, 23), Hong Kong (15), China (15), Korea (15), and Sri Lanka (11). Concentrative sampling approaches have been used to assess the occurrence and behavior of PFOS and PFOA in the Tennessee River, U.S.A. (5). Saito et al. investigated PFOS and PFOA in surface waters in Japan to map their geographical distribution (6, 7). However, other PFS homologs, including longer-chain perfluorocarboxylates, in Japanese rivers have been little studied. Because bioconcentration factors increase with increasing length of the perfluoroalkyl chain (24), analyses of longer-chain PFSs are warranted. The identification of sources of PFSs in water environments is key to effective strategies for controlling them. Because atmospheric reactions of fluorotelomer alcohols produce PFSs (17, 25), both atmospheric transport and direct discharge such as in wastewater (26, 27) are potential sources of PFSs. Past studies showed that PFS concentrations increased downstream of urban areas, indicating urban activities as sources of PFSs (6, 15). Simcik and Dorweiler used the ratios of PFSs to infer their sources in Lake Michigan, U.S.A., and found that the primary source was most likely wastewater treatment effluent (9). Kim and Kannan showed that surface runoff water contributed to contamination by PFOA in urban lakes (12). However, sources of PFSs in many rivers are not still unraveled. In this study, we investigated the occurrence of 9 PFS species in 18 rivers (20 samples) and 5 wastewater secondary effluents in Japan: PFOS, perfluorooctane sulfonamide (FOSA; C8F17SO2NH2), perfluoroheptanoate (PFHpA; C6F13COO-), PFOA, perfluorononanoate (PFNA; C8F17COO-), perfluorodecanoate (PFDA; C9F19COO-), perfluoroundecanoate (PFUA; C10F21COO-), perfluorododecanoate (PFDDA; C11F23COO-), and perfluorotridecanoate (PFTDA; C12F25COO-). Because surface water is the major source of drinking water in Japan, this information is important in evaluating the potential risks posed by PFSs in drinking water. We also analyzed the sources of PFSs in rivers and total fluxes of sewage-derived PFSs from Japan by using crotamiton, a molecular marker of sewage effluents (28, 29).
Experimental Section * Corresponding author phone: + 81-423-67-5825; fax: + 81-42360-8264; e-mail:
[email protected]. † Tokyo University of Agriculture and Technology. ‡ The University of Tokyo. § Public Works Research Institute. 6566
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 17, 2008
Sample Collection. Out of 109 rivers throughout Japan that are designated as “the first-grade”, we selected two from each of nine regions administered by regional development bureaus to cover the range of basin areas and population densities in Japan (Figure S1a, b, Supporting Information). 10.1021/es800353f CCC: $40.75
2008 American Chemical Society
Published on Web 08/05/2008
TABLE 1. Basin Characteristics of Rivers
river R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. R.
Tokorogawa Sarugawa Yoneshirogawa Narusegawa Nakagawa Tamagawa Tamagawa Tamagawa Arakawa Kurobegawa Shounaigawa Abegawa Yuragawa Yamatogawa Asahikawa Ashidagawa Niyodogawa Monobegawa Oonogawa Kimotsukigawa a
station
district
sampling date
Chuushi-bashi Osachinai-bashi Futatsui Ono Shimokunii Haijima-bashi (st.1) Sekido-bashi (st.2) Denenchofuzeki (st.3) Asahi-bashi Shimokurobe-bashi Biwajima-bashi Abegawa-bashi Habi-bashi Oriono-bashi Otoidezeki Kominomi-bashi Nakajima Fukabuchi Shirataki-bashi Matase-bashi
Hokkaido Hokkaido Tohoku Tohoku Kanto Kanto Kanto Kanto Hokuriku Hokuriku Chubu Chubu Kinki Kinki Chugoku Chugoku Shikoku Shikoku Kyushu Kyushu
Nov. 8, 2005 Dec. 8, 2005 Nov. 8, 2005 Nov. 16, 2005 Dec. 6, 2005 Dec. 7, 2005 Dec. 7, 2005 Dec. 7, 2005 Nov. 30, 2005 Dec. 7, 2005 Nov. 16, 2005 Dec. 8, 2005 Dec. 14, 2005 Dec. 8, 2005 Nov. 15, 2005 Nov. 28, 2005 Dec. 6, 2005 Dec. 6, 2005 Dec. 1, 2005 Dec. 12, 2005
population/ population population flow fresh natural basin area in basin density volume percentage flow (×1000 (km2) area (×1000) (km-2) (m3/s) (%)a s/m3) 1930 1350 4100 1130 3270
142 13 280 190 912
74 10 68 168 279
16.1 22.9 94.8 14.5 56.1
93 99 99 90 91
9.5 0.6 3.0 15 18
1240 1150 682 1010 567 1880 1070 1810 860 1560 508 1465 485
4250 40 71 2500 170 300 2150 335 269 105 40 207 116
3427 35 104 2475 300 160 2009 185 313 67 79 141 239
13.4 47.2 59.6 8.6 5.6 29.7 16.6 15.2 2.1 19.9 2.9 22.7 18.2
36 100 100 69 90 97 38 97 95 97 100 83 86
881 0.8 1.2 421 34 10 341 23 135 5.4 14 11 7.4
Percentage of intact river flow.
Figure S2 shows sampling locations. We collected samples from the most downstream freshwater estuary of each river during dry weather in November or December 2005. In the Tamagawa River, samples were collected at three stations: st.1 (Haijimabashi; 46 km from Tokyo Bay), st.2 (Sekidobashi; 35 km from Tokyo Bay), and st.3 (Denenchofuzeki; 13 km from Tokyo Bay). In total, 20 river water samples were collected. We also investigated pharmaceutical and personal care products in a parallel study (29). Information on rivers, including basin areas and populations, is shown in Table 1. To confirm the representativeness of water quality, we compared concentrations of total nitrogen (T-N) and total phosphorus (T-P) between this study and the annual average based on monthly monitoring (30). T-N and T-P were measured by flow injection analysis (Branluebbe Traacs 2000) using the Japanese Industrial Standard (JIS) K0102 (31). Figure S3 shows that T-N and T-P concentrations in this study agreed well with their annual averages. Figure S1c, d also shows that distributions of T-N and T-P concentrations in the rivers in this study were similar to those in the 109 first-grade rivers (30). Therefore, the conventional water quality in the selected rivers could be regarded as being representative. We also collected five wastewater secondary effluents in Japan during dry weather in 2006 and 2007: four 24-h composite samples and one grab sample. Detailed information on individual treatment plants is summarized in Table S1. Sample IDs followed ref (28, 29). River and secondary effluent samples were filtered through prebaked glass fiber filters (GF/F, pore size ) 0.7 µm; Whatman) normally within 1-2 days and stored at 5 °C before analyses. Most samples were extracted within one month of collection. Chemicals. The potassium salt of PFOS (98%) was purchased from Fluka (Buchs, Switzerland). FOSA (97%) was purchased from Matrix Scientific (Columbia, U.S.A.). PFHpA (95%), PFNA (95%), and PFDA (97%) were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). PFOA (95%) and PFDDA (96%) were purchased from Alfa Aesar (Ward Hill, USA). PFUA (95%), PFTDA (97%), and crotamiton (97%) were purchased from Aldrich (Milwaukee, U.S.A.). 13C -labeled PFOA (97.6%) was purchased from Hayashi Pure 2 Chemical Industries, Ltd. (Osaka, Japan). HPLC-grade methanol, distilled water, and acetonitrile and special-grade hydrochloric acid and ammonium acetate were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan) and used for PFS measurements. Special-grade n-hexane,
dichloromethane, acetone, methanol, and isooctane, also purchased from Wako, were distilled in glass before crotamiton measurements. Analyses. PFSs in the filtrate (dissolved phase) were concentrated by a factor of 500-1000 by solid-phase extraction. 13C-labeled PFOA (25-100 ng) was spiked into 500-1000 mL aliquots of samples, and the samples were passed through Sep-Pak Plus tC18 cartridges (Waters) preconditioned with 20 mL methanol and 10 mL distilled water. A flow rate of less than 10 mL/min was maintained. To remove interfering compounds from secondary effluent samples, we washed the cartridges with 7 mL of 30% (v/v) methanol in distilled water and then with 7 mL of 55% (v/v) methanol in distilled water that was acidified with 4 M HCl to pH 2.0-2.5. The target compounds were eluted with 20 mL methanol. The eluate was concentrated to 0.5-1 mL and further diluted if necessary. PFSs were analyzed by liquid chromatographytandem mass spectrometry (LC-MS/MS; Agilent 1100 and TSQ Quantum) in electrospray negative ionization mode. The extract (10 µL) was injected into a Zorbax Rx-C8 column (4.6 mm × 150 mm, 5 µm, Agilent). A mobile phase of 0.5 mM ammonium acetate in distilled water and 0.5 mM ammonium acetate in acetonitrile was used. At a flow rate of 0.3 mL/min, the mobile phase gradient was ramped from 20% to 100% 0.5 mM ammonium acetate in acetonitrile in 5 min, maintained at 100% for 11 min, and then ramped down again to 20%. The capillary temperature was maintained at 350 °C. Ions were monitored in selected reaction monitoring mode. Details of parent and product ions and collision energies are listed in Table S2. Concentrations of all target analytes were quantified from calibration curves drawn using external standards. The coefficient of determination (r2) for each calibration (5 point calibration) was normally >0.99. Only PFHpA, PFOA, PFNA, and PFDA were label-recovery-corrected (see Quality Accuracy and Quality Control section). Crotamiton was determined as described earlier (28, 29). Briefly, filtrate samples (500-1000 mL) were passed through Sep-Pak Plus tC18 cartridges that had been washed with hexane, dichloromethane, methanol, and distilled water. The compounds retained on the cartridges were eluted with 20 mL of methanol. The eluate was purified by 5% H2Odeactivated silica gel column chromatography, in which crotamiton was eluted in the 30% (v/v) acetone-in-dichloromethane fraction. The eluate was dried by evaporation, VOL. 42, NO. 17, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6567
TABLE 2. Reproducibility and Recovery Rates of PFSs wastewater secondary effluent (n ) 4)
PFOS FOSA PFHpA PFOA PFNA PFDA PFUA PFDDA PFTDA
concentrations (ng/L)b
RSDc
57 ( 2
