Document 0LknVb1ddg5oakZENdXqndGvR

SELECTED FLUOROCHEMICALS IN THE DECATUR, ALABAMA AREA Preparedfor: 3M St. Paul, Minnesota Prepared by: Entrix, Inc. John P. Giesy, Ph.D. John L. Newsted, Ph.D. East Lansing, Michigan Project No. 178401 June, 2001 1 1 EXECUTIVE SUMMARY A monitoring study was conducted to assess the current extent, status, and magnitude o f fluorochemical concentrations from a 3M Facility near the Tennessee River in the Decatur Alabama area. The objectives o f the study were to determine the concentrations o f selected fluorochemicals in biotic and abiotic phases within the Tennessee River in the vicinity o f the 3M Facility. Surface water, sediments, clams and fish were collected from locations upstream and downstream o f the 3M facility and analyzed for four fluorochemicals: perfluorooctanesulfonate (PFOS), perfluorooctanesulfonamide (FOSA), perfluorooctanoate (PFOA), and perfluorohexanesulfonate (PFHS). Surface waters and sediments were collected from five primary locations. These areas consisted of a location near the 3M point o f discharge (Outfall), a stream the receives the 3M Facility effluent (Bakers Creek), a downstream location (Fox Creek), a location upstream o f 3M Facility but below the City o f Decatur (DWTP), and a reference location upstream o f Guntersville Dam. Clams were collected from the Guntersville location and the Fox Creek location. Fish were collected from the 3M Outfall and from the Guntersville location. All samples were collected according to standard procedures and analyzed for fluorochemicals according to 3M Environmental Laboratory methods. Results from the monitoring study indicate that the while the effluent from the 3M Outfall did cause changes in some water quality parameters within Bakers Creek, it did not cause any statistically significant changes in water quality parameters at the farthest downstream location, Fox Creek. Average surface water PFOS concentrations ranged from 0.009 to 151 pg/L for Guntersville and Outfall, respectively. Average PFOS sediment concentrations ranged from 0.180 to 5930 pg/kg (wet weight) for Guntersville and the 3M Outfall, respectively. However, while fluorochemical concentrations in sediments and surface waters were greater in locations downstream of the 3M Outfall, the downstream sample locations within the main stem o f the Tennessee River (Fox Creek) were not statistically greater than those upstream (DWTP and Guntersville) o f the 3M Outfall. Only samples collected from the 3M Outfall and Bakers Creek, locations that are in the immediate vicinity the 3M Outfall, were significantly greater than those in the 2 reference location. From these results, it can be concluded that the effluent from the 3M Outfall does not significantly affect fluorochemical concentrations in sediments or surface waters in the main stem o f the Tennessee River. Fluorochemicals were detected in all fish collected from both the Guntersville reference area and 3M Outfall. The least fluorochemical concentrations were observed in fish collected from Guntersville while the greatest concentrations were observed in fish from the Outfall. For instance, at Guntersville, average whole body concentrations for PFOS, FOSA, PFOA, and PFHS were 59.1, 9.43, 11.7 and 7.50 pg/kg, (wet weight) respectively. For fish from the outfall, average whole body concentrations for PFOS, FOSA, PFOA and PFHS were 1332.0, 20.1, 106.4, and 86.9 pg/kg, (wet weight) respectively. While there were species specific differences in the accumulation o f the selected fluorocarbons into fish collected from either Guntersville or 3M Outfall locations, these differences were not statistically significant due to the small sample size and among-individual variability. Concentrations o f FOSA and PFOA in clams collected from the Outfall were greater than concentrations observed in clams collected from the Guntersville reference site. However, PFOS and PFHS tissue concentrations in Outfall clams were not greater than concentrations in clams collected from Guntersville. In conclusion, fish downstream o f the 3M Outfall contained greater concentrations o f fluorochemicals than did fish from the Guntersville area but the significance o f these differences is difficult to assess due to the small sample size and variability. 3 2 INTRODUCTION Entrix, Inc (ENTRIX) is providing a report on activities to assess the impact o f effluents containing fluorochemicals from the 3M Facility in Decatur, Alabama to the Tennessee River. Water, sediment, clams and fish were collected from upstream and downstream locations o f the 3M Facility and were used to evaluate the magnitude o f releases o f target compounds from the Facility to the Tennessee River. The objectives o f the research are to determine concentrations o f selected fluorochemicals in abiotic and biotic phases o f the Tennessee River system in the vicinity o f the 3M Facility at Decatur, Alabama. Fluorochemicals selected for evaluation in this study included: perfluorooctanesulfonate (PFOS), perfluorooctanesulfonamide (FOSA), perfluorooctanoate (PFOA) and perfluorohexane sulfonate (PFHS) (Appendix A). In addition, trifluoroacetic acid was also evaluated in water and sediment samples taken from the Tennessee River. 3 FIELD SAMPLING 3.1 Site Description Four primary locations were selected for the field investigation (Figure 1, Appendix B). These areas consisted o f a location near the 3M point o f discharge (Outfall), a downstream location, a location upstream o f 3M but below the City o f Decatur, and a reference location upstream o f Guntersville Dam. Specifically, samples were collected adjacent to the 3M Facility Outfall (includes Outfall and Bakers Creek) that corresponds River miles 301-302. Samples were also collected downstream o f the 3M Facility near the mouth o f Fox Creek (River miles 296-297). Sediment and water were collected upstream o f the 3M Facility adjacent to the Decatur Wastewater Treatment Plant (approximately river mile 303.5). Finally, fish, clams, sediment, and water were collected from a reference site above the Guntersville Dam (approximately river mile 370). To better evaluate the impact o f the effluent from the 3M Outfall, the Bakers Creek location was divided into two sub-locations, Bakers Creek-1 included Stations 1, 2 and 3 that were collected near the outfall and within Bakers Creek. Stations 4, 5 and 6 and were collected near the interface o f the mouth o f Bakers Creek and the Tennessee River and are identified as Bakers Creek-2. 4 2 Bakers Creek Figure 1. Tennessee River and sampling locations evaluated during study. 5 Fox Creek sampling locations were also divided into two sub-locations. Fox Creek-1 included Stations 7, 11, 12 and 26/27 that were collected from the mouth o f Fox Creek. Fox Creek-2 included Stations 8, 9, and 10 that were collected on the northern side o f the Tennessee River across from the mouth o f Fox Creek. Locations and sample codes for samples collected from the Tennessee River are given (Table 1). 3.2 Field Sampling Logistics Sampling at selected sites on the Tennessee River was initiated on June 19, 2000 by ENTRIX. Collection o f water and sediment samples continued until June 21, 2000. During the collection o f sediments, clams were also collected. Fish collection was initiated on June 21 and continued through June 22, 2000. Water, sediment, clams and fish were collected as outlined in Entrix sampling protocols (Appendix C). 3.3 SAMPLE COLLECTION 3.3.1 Water Collection Water samples were collected using ENTRIX sampling protocols from a total o f 26 locations (Table 1, Appendix B). At each location, water quality parameters were measured using YSI Model 63 and 95 meters. Six samples were collected from the areas surrounding the 3M Outfall, the City o f Decatur wastewater treatment facility and the upstream reference location. Seven water samples were collected from the area downstream o f the 3M Facility near Fox Creek. In addition to these 25 samples o f surface water, water was collected from the 3M Facility effluent stream. All sampling locations were documented using IRIMLBE PRO-XRS Global Positioning Satellite technology and these locations were recorded on topographic maps. Samples were properly labeled, stored on ice at 4C, and delivered to the 3M Environmental Laboratory, St. Paul, MN. 3.3.2 Sediment Collection Sediment samples were collected from the same location, as were the water samples (Table 1, Appendix B). A total o f 25 grab samples were collected from the Tennessee River. Six samples were collected from areas surrounding the 3M Outfall, the City o f Decatur's wastewater treatment facility, and the upstream reference location using either a Petite PONAR and/or Eckman dredge. Seven samples were collected from the 6 downstream area in the vicinity o f Fox Creek. AH sampling locations were documented using TRIMLBE PRO-XRS Global Positioning Satellite technology and these locations were recorded on topographic maps. Sediments were collected by use o f a PONAR dredge following standard protocols. Sediments were placed into labeled 250 ml widemouth LDPE containers, stored on ice at 4C, and shipped to the 3M Environmental Laboratory, St. Paul, MN. Table 1. Location and identification o f sediment and water samples collected from the Tennessee River in Alabama Water Sample Sediment Sample Location Sublocation/Code SW-01 SED-01 Bakers Creek BCR1 SW-02 SED-02 Bakers Creek BCR1 SW-03 SW-04 SW-05 SED-03 SED-04 SED-05 Bakers Creek Bakers Creek Bakers Creek BCR1 BCR2 BCR2 SW-06 SED-06 Bakers Creek BCR2 SW-07 SW-08 SED-07 SED-08 Fox Creek Fox Creek FCR1 FCR2 SW-09 SED-09 Fox Creek FCR2 SW-10 SED-10 Fox Creek FCR2 SW-11 SED-11 Fox Creek FCR1 SW-12 SED-12 Fox Creek FCR1 SW-13 SED-13 Decatur WWTP WWTP SW-14 SED-14 Decatur WWTP WWTP SW-15 SED-15 Decatur WWTP WWTP SW-16 SED-16 Decatur WWTP WWTP SW-17 SW-18 SED-17 SED-18 Decatur WWTP Decatur WWTP WWTP WWTP SW-19 SW-20 SW-21 SED-19 SED-20 SED-21 Guntersville Guntersville Guntersville GTVL GTVL GTVL SW-22 SED-22 Guntersville GTVL SW-23 SW-24 SED-23 SED-24 Guntersville Guntersville GTVL GTVL SW-25 SW-26 SW-27 SED-25 SED-27 3M Outfall Fox Creek Fox Creek OUTF FCR1 FCR1 3.3.3 Clam Collection Samples o f the Asian clam (Corbicula fluminea) were targeted for collection from a location downstream o f the 3M Facility and from a location upstream o f Guntersville 7 Dam. Historically, various state and federal agencies have used clams in monitoring programs for organic and inorganic pollutants in aquatic systems, as a result, they were also included in this study. Furthermore, due to their ubiquitous distribution in the Tennessee River and their sessile behavior, they provide site-specific information concerning the potential accumulation o f fluorochemicals into aquatic biota. Due to the habitat requirements of C. fluminea, clam samples were not co-located with water or sediment samples. Clams were collected using a Petit PONAR dredge. However, due to difficulties in locating clam beds as well as interference o f substratum with the PONAR dredge, the collection o f clams was limited to two composite samples from locations in the vicinity o f sediment samples. One sample (approximately 25 clams/sample) was taken from a location downstream o f the 3M Facility (SED-08, -09, -10) and one upstream sample was collected from the Guntersville location (SED-20, - 21, - 22, -23, 24). Clams from each location were placed into labeled plastic bags and shipped on ice to the MSU-ATL (Michigan State University-Aquatic Toxicology Laboratory), East Lansing, MI. 3.3.4 Fish Collection Fish were collected from two locations (Table 2). One location was in the vicinity o f the 3M Outfall and the other was located within the reference area upstream o f Guntersville Dam. The fish samples were collected using both hook and line as well as gill net. Gill nets (passive gear) were placed in the lower energy environments where concentrations o f fish were generally greater. Due to the techniques employed for collection o f fish, all fish caught in each o f the sampling locations were retained for analysis. Collected fish were removed from the nets, measured (total length), weighed, and immediately wrapped in aluminum foil and stored in plastic bags. Samples were labeled placed on ice, and shipped to the MSU-ATL (East Lansing, MI) for processing. 4 CHEMICAL ANALYSIS Water and sediment samples were collected from the Tennessee River and analyzed by the Centre Analytical Laboratories, State College, PA (CLA, Report No. 023-0140; Appendix D). The analytical methods used for the water samples were validated by Centre. 8 The methods were modified for the analysis o f sediment samples but were not fully validated for this matrix. Fish livers and clam samples were extracted and processed at MSU-ATL. LC/MS/MS characterization o f fluorochemicals was conducted at 3M Environmental Laboratory. Summaries o f each protocol are given: Table 2. Sample identification and location o f individual fish collected from the Tennessee River. Sample ID Location Species LM1 Guntersville Largemouth bass Micropterus salmoides LM2 3M Outfall Largemouth bass Micropterus salmoides SHAD1 3M Outfall Skipjack Herring Alosa chiysochloris SHAD2 SHAD3 3M Outfall 3M Outfall Skipjack Herring Skipjack Herring Alosa chrysochloris Alosa chrysochloris SHAD4 3M Outfall Skipjack Herring Alosa chrysochloris SHAD5 SHAD6 SHAD7 3M Outfall 3M Outfall 3M Outfall Skipjack Herring Skipjack Herring Skipjack Herring Alosa chrysochloris Alosa chrysochloris Alosa chrysochloris SHAD8 WP1 3M Outfall 3M Outfall Skipjack Herring White Perch Alosa chrysochloris Morone americana WP2 3M Outfall White Perch Morone americana CAT1 CAT2 CAT3 CAT4 Guntersville Guntersville 3M Outfall 3M Outfall Catfish Catfish Catfish Catfish Ictalurus punctatus Ictalurus punctatus Ictalurus punctatus Ictalurus punctatus GAR1 Guntersville Gar Lepisosteus sp. GAR2 GAR3 GAR4 GAR5 GAR6 SB1 Guntersville Guntersville Guntersville 3M Outfall 3M Outfall Guntersville Gar Gar Gar Gar Gar Striped Bass Lepisosteus sp. Lepisosteus sp. Lepisosteus sp. Lepisosteus sp. Lepisosteus sp. Morone saxatilis SB2 Guntersville Striped Bass Morone saxatilis SB3 Guntersville Striped Bass Morone saxatilis SB4 Guntersville Striped Bass Morone saxatilis SB5 Guntersville Striped Bass Morone saxatilis SB6 Guntersville Striped Bass Morone saxatilis SB7 Guntersville Striped Bass Morone saxatilis SB8 SB9 SB10 Guntersville Guntersville Guntersville Striped Bass Striped Bass Striped Bass Morone saxatilis Morone saxatilis Morone saxatilis SB11 Guntersville Striped Bass Morone saxatilis SB12 Guntersville Striped Bass Morone saxatilis SB13 SB14 Guntersville Guntersville Striped Bass Striped Bass Morone saxatilis Morone saxatilis 9 4.1 WaterAnalysis Water samples were initially treated with 200 pi o f 250 mg/L sodium thiosulfate to remove residual chlorine. Solid phase extraction was used to prepare the samples for LC/MS/MS analysis. A forty-milliliter portion o f sample was transferred to a Cig SPE cartridge and the cartridge was eluted with 5 ml o f 40% methanol in water. The eluate was discarded and the SPE column was eluted with 5 ml o f 100% methanol. The eluate was collected for analysis by LC/MS/MS. This treatment resulted in an eight-fold concentration o f the samples. 4.2 SedimentAnalysis For sediment samples, a 5 g portion was extracted into 5 ml methanol. The extracts were filtered and diluted to final volume o f 40 ml with ASTM Type I water. The diluted extracts were then treated in the same manner as the water samples, beginning with the solid phase extraction. Sediment and water extracts were analyzed by use o f a Hewlett-Packard HP1100 HPLC system coupled to a MicroMass Ultima MS/MS. Analysis was performed using selected reaction monitoring (SRM). HPLC conditions and MS/MS methods used for analysis and instrument parameters are given in the procedure described below. 4.3 Fish Liver Analysis Fluorochemical surfactants were extracted from fish livers using an ion pairing reagent and methyl-/er/-butyl ether (MtBE). Details o f the analytical procedure have been outlined in 3M Environmental Laboratory (St. Paul, MN) standard operating procedures o f the analysis o f fluorochemicals in tissues. The title o f the SOP is: ETS-8-6.0, "Extraction o f Potassium Perfluorooctanesulfonate or other Fluorochemical Compounds from Liver for Analysis using HPLC-Electrospray/Mass Spectrometry". Briefly, fish were thawed and livers were removed, weighed and then homogenized in HPLC grade water. An aliquot o f liver homogenates were spiked with a surrogate standard and then 0.5 M tetrabutyl ammonium hydrogen sulfate (TBA) and 0.25 M sodium carbonate/sodium bicarbonate was added to each sample. Five ml methyl-ferf-butyl ether was added to each tube and the sample was capped and shaken for 20 min. The samples were then centrifuged for 25 min at 3500 rpm and 4.0 ml o f the organic layer was 10 transferred to a fresh tube. The extracts were evaporated under nitrogen gas to dryness then reconstituted in methanol. The methanol extracts were then passed through 0.2 pm nylon filters into glass autovials for chromatographic analyses. 4.4 d a m Tissue Analysis Clam tissues were extracted with acetonitrile followed by a solid phase column cleanup o f the extract. Briefly, frozen clam tissues were thawed and the tissues were composited and homogenized in an Omni homogenizer. A 5 gram sample was weighed a 50 ml polypropylene tube and 15 ml acetonitrile was added. Samples were shaken for 30 min and then centrifuged for 30 min at 3500xg. The supernatant was passed through a column o f Florisil and activated charcoal and the eluate was collected in 50 ml polypropylene tube. The column was eluted with 15 ml methanol and both eluates were combined. Two drops o f octanol were added to the eluate and then the sample was reduced to near dryness under a flow of nitrogen gas. The sample was resuspended in 1 ml methanol and passed through a 0.2 pm nylon filter into autovials for chromatographic analysis. 5 Q A /Q C Standard operating procedures for sample collection and preparation were maintained during the entire project. Proper QA/QC samples, as required by 3M, were collected in the field (Appendix C). Two field matrix spikes and two field control samples were collected. Field matrix spikes solutions were prepared by collecting one liter o f site water and spiking with 20 or 100 pi o f a 10 ppm stock solution containing PFOS, FOSA, PFOA, and PFHS to yield 200 and 1000 ppt solutions, respectively. The same procedure was followed to spike one liter o f distilled water to prepare the field spike control samples. All samples were shipped under ENTRIX chain o f custody forms and a field book was used to document conditions and activities. A field camera was used to document all locations and to provide a visual record o f the conditions o f the site during the sampling events. In addition, ambient water quality parameters, such as temperature, conductivity, salinity, pH and dissolved oxygen were measured at each location during each sampling event. Since the method detection limit (MDL) and limit o f quantification (LOQ) are analyte and matrix specific, method and matrix blanks along with matrix 11 spikes were used to determine accuracy and precision of the extractions and final chemical determination. Water and sediment concentrations were not corrected for either matrix spike recoveries or corrected for purity of the fluorochemical standards. Fish tissue fluorocarbon concentrations were reported as wet weight and were not converted to a dry weight basis. 6 STATISTICAL ANALYSES Statistical analyses were preformed with SAS (Version 8, SAS Institute, Cary NC, USA). General linear models (PROC GLM) were used to test for differences among locations for all endpoints measured in the study. If values o f F tests indicated a significant difference, Tukey's HSD test for multiple comparisons was used to compare means of the different locations. 7 RESULTS 7.1 Water qualityparameters Water quality parameters o f river water from each location are reported (Table 3, Appendix E and F: figures la, lb, lc, and Id). Dissolved oxygen and pH were variable within the study there were no significant differences observed between values measured in waters collected from the 3M Outfall or Bakers Creeks and those from upstream or downstream locations. Conductivity and temperature measured at the 3M Outfall and within Bakers Creek (BCR1) were significantly different from the upstream and downstream sample locations. However, there was no significant difference between that from Bakers Creek 2 samples and those collected from either the upstream and downstream sampling locations. This result indicates that the effect o f the effluent from the 3M Facility on water quality parameters was limited to a small area near the 3M Outfall. 12 Table 3. Water quality parameters for samples collected from the Tennessee River, in the Decatur Alabama area. Values are reported as means and standard deviations Location DO (mg/L) Conductivity Temperature (pmhos/cm) (C) PH (s.u) Guntersville 7.6 (0.8) 192.8 (5.4) 29.0 8.2 (0.3) (0.4) DWTP 10.0* (1.3) 184.7 (3.4) 29.3 8.8 (0.3) (0.2) Outfall 6.5 4650* 33.2* 7.2 Bakers Creek-1 7.75 (1.4) 1843.3 * (1749.0) 30.7* (1.4) Bakers Creek-2 9.0 (0.6) 463.0 (108.9) 29.6 (0.6) Fox Creek-1 Fox Creek-2 8.1 (0.4) 8.8 (1.3) 179.0 (2.9) 162.7 (8.1) 27.8 (0.2) 27.8 (0.5) * Significantly different from Guntersville Reservoir at a = 0.05 7.7 (0.5) 8.0 (0.1) 8.2 (0.3) 8.2 (0.5) 7.2 Fluorochemicals in Surface waters Fluorochemicals were detected in all surface waters collected from the sample locations in the Tennessee River (Table 4) (Appendix F: figures 2a, 2b, 2c, and 2d). The least concentrations o f all measured fluorochemicals, were observed in samples taken at Guntersville while the greatest concentrations were observed at 3M Outfall. In increasing order, the concentrations o f PFOS, FOSA, PFOA and PFHS in surface water was as follows: GTVL < DWTP < FCR1 < FCR2 < BCR2 < BCR1 < Outfall. Concentrations o f surface water fluorochemicals at only the 3M Outfall and Bakers Creek 1 locations were significantly greater than those measured at Guntersville Reservoir. Trifluoroacetic acid (TFA) concentrations were elevated in 3M Outfall (mean =1420 pg/L), Bakers Creek l(mean = 680 pg/L) and Bakers Creek 2 (143 pg/L) samples but were less than the reporting limit o f 100 pg/L in water samples collected from the other 13 sampling locations. This result indicates that the effect o f the effluent from the 3M Facility on TFA water concentration was limited to a small area near the 3M Outfall. Table 4. Mean concentration o f fluorocarbons in surface waters o f the Tennessee River in the Decatur, Alabama area. Concentrations in means and standard deviations PFOS FOSA PFOA PFHS Location Guntersville (M8*L) 0.009 (US/L) 0.004 (pg/L) 0.0081 (Pg/L) 0.0031 (0.002) (0.002) DWTP 0.053 (0.013) 0.016 (0.025) 0.028 (0.006) 0.006 (0.006) Outfall 150.50* (7.778) 7.985* (0.148) 1900.00* (0.00) 42.70* (1.697) Bakers Creek-1 82.260* (56.672) 5.424* (3.074) 1023.60* (712.64) 22.374* (14.665) Bakers Creek-2 13.838 (2.351) 1.191 (0.244) 129.375 (31.892) 4.034 (0.575) Fox Creek-1 0.54 (0.05) 0.169 (0.022) 2.647 (0.614) 0.566 (0.102) Fox Creek-2 0.179 0.055 1.001 0.168 (0.110) (0.033) (0.441) (0.114) * Significantly different from Guntersville R eservoir at a = 0.05 1Lim it o f quantitation for all fluorochem icals water is 0.025 (ig/L. A ll fluorochemical concentrations reported below the LOQ are estim ates. Additional statistical analyses were conducted on fluorochemicals in surface waters to further examine the difference between sample locations (GTVL, DWTP, BCR2, FCR1, and FCR2) located within the Tennessee River. Results from the first analysis showed that both the 3M Outfall and BCR1 could be influencing the results o f the statistical analyses by acting in a manner similar to that o f an outlier. For instance, the concentration o f PFOS at the 3M Outfall and BCR1 were approximately 17,000-fold and 9,700-fold, greater, respectively, than the concentration observed at the Guntersville location. Removal o f the 3M Outfall and BCR1 samples from the analysis allowed for the evaluation o f the hypothesis that there were no statistically significant differences in 14 fluorochemical concentrations within the main channel and near shore locations within the river. Results from the second analysis (Table 5) show that for water concentrations o f PFOS and PFOA, only BCR2 was statistically greater than Guntersville Reservoir concentrations. Furthermore, while the concentrations o f FOSA and PFHS at BCR2 and FCR1 were significantly greater than Guntersville, they were not significantly different from the furthest downstream location. Thus, at FCR2, the furthest downstream location, the concentration o f targeted fluorochemicals was not significantly different from that observed at Guntersville Reservoir, which was selected as an upstream reference location. Table 5. Tukey's range test results for surface water fluorochemicals taken from within or near the main channel o f the Tennessee River* Site PFOS FOSA PFOA PFHS BCR2 A A AA FCR1 B B BB FCR2 B B,C B C WWTP B C B C GTVL B C BC * Sites with the same letter are not significantly different at a = 0.05 From these results, it can be concluded that the 3M Outfall does not significantly effect concentrations o f fluorochemicals farther downstream than FCR2. In addition, there were no statistically significant differences in the concentrations o f fluorochemicals within the main channel o f the river immediately below Bakers Creek as compared to those measured at the reference location. 7.3 Sediment concentrations o ffluorocarbons Fluorocarbons were detected in all sediments collected from the study sites (wet weight: Appendix F-figures 3a, b, c, d and dry weight: Appendix F-figures 4a, b, c, d). 15 Table 6. Mean concentration of fluorochemicals in sediments taken from the Tennessee River in the Decatur Alabama area" Wet weight sediment concentrations Dry weight sediment concentrations Location PFOS FOSA PFOA PFHS PFOS FOSA PFOA PFHS Guntersville (Pg/kg) 0.18 (Pg/kg) 0.08 (Pg/kg) 0.08 (Pg/kg) 0.08b (Pg/kg) 0.43 (Pg/kg) 0.19 (Pg/kg) 0.19 (pg/kg) 0.19 (0.07) (0.01) (0.01) (0.20) (0.04) (0.04) (0.04) DWTP 0.98 (0.42) 0.15 (0.06) 0.09 (0.02) 0.08 (0.01) 1.72 (0.83) 0.26 (0.11) 0.16 (0.04) 0.14 (0.04) Outfall 5930.00* (254.56) 1200.00* (42.43) 1855.00* (106.07) 135.00* (4.24) 12600.00* 2555.00* (565.69) (91.92) 3950.00* (226.27) 287.50* (9.19) Bakers Creek 1 1298.67* (751.89) 283.30* (176.21) 892.00* (345.80) 42.10* (22.05) 2275.67* (142.76) 503.17* (364.31) 1548.17* (774.63) 74.6* (48.57) Bakers Creek 2 192.00 (191.27) 54.09 (36.33) 237.61* (283.01) 11.46* (8.08) 272.64 (279.05) 75.91 (53.03) 338.43* (410.79) 16.18 (11.91) Fox Creek 1 2.58 (1.37) 1.70 (1.09) 1.81 (0.77) 0.22 (0.11) 4.51 (2.88) 2.99 (2.21) 3.11 (1.68) 0.38 (0.23) Fox Creek 2 0.93 0.44 0.80 0.08b 2.02 0.95 1.73 ( 1Q) (0.05) (0.22) (0.22) (0.10) (0.46) * Significantly different form Guntersville Reservoir at a = 0.05 * Concentrations given as means w ith standard deviations in parenthesis b Lim it o f quantification o f all fluorocarbons in sediment is 0.20 pg/kg wet weight. A ll values reported below the LOQ are estimates. 0.17b 16 Least concentrations o f fluorochemicals in sediments were observed at Guntersville Reservoir while the greatest concentrations were observed in sediments collected in the vicinity o f the 3M Outfall (Table 6). Concentrations o f fluorochemicals in sediment decreased in the following order: GTVL < DWTP < FCR2 < FCR1< BCR2< BCR1< OUTF. For PFOS and FOSA, only concentrations in sediments at the 3M Outfall and BCR1 locations were significantly greater than those collected from the reference location in Guntersville Reservoir. In addition, concentrations of PFOA and PFHS in sediments at the 3M Outfall, BCR1 and BCR2 were significantly greater than concentrations in sediments from Guntersville Reservoir. However, the fluorochemicals in sediments at BCR2 were not significantly different from concentrations measured at DWTP, FCR1 or FCR2. When sediment concentrations of targeted fluorochemicals were evaluated on a dry weight basis, only concentrations at the 3M Outfall and BCR1 were significantly different from those measured at Guntersville Reservoir. As was done with the surface water data, statistical analyses were conducted that examined only the Tennessee River channel sample locations (GTVL, DWTP, BCR2, FCR1, and FCR2). Results from the second set o f analyses showed that only the BCR2 concentrations o f targeted fluorochemicals were greater than those observed at Guntersville Reservoir reference location on both a wet and dry weight basis. Thus, while concentrations at the Outfall and within Bakers Creek were statistically greater than the reference location, concentrations in sediments downstream o f the 3M Facility were not statistically greater than those measured at the upstream reference location. The results o f both o f these analyses indicated that that there is little accumulation o f the target fluorochemicals in sediments downstream o f the 3M Outfall. Sediment trifluoroacetic acid (TFA) concentrations were not significantly different between the sampling locations within the main channel o f the Tennessee River. TFA concentrations for all sample locations except the 3M Outfall were less than the reporting limit o f 500 pg/kg, wet weight. The 3M Outfall sediment TFA concentration averaged 17 510 pg/kg. This result indicates that the effluent from the 3M Facility only alters sediment concentrations in the immediate vicinity o f the 3M Outfall, but does not significantly affect sediment TFA concentrations within Tennessee River. 7.4 Fish Mean length, whole body weight and liver weight data for all fish species collected from the Tennessee River in the Decatur Alabama area are given (Table 7, Appendix G). Table 7. Mean length, weight, and liver weight for fish collected from Sample ID LM1 Location Guntersville Length (cm) 26.5 Weight (g) 195 3M Outfall 32 454 SHAD 3M Outfall 34 352 (7.6) (160) WP1 3M Outfall 16.8 76 (0.36) (7.1) CAT Guntersville 32.5 376 (2.12) (112) 3M Outfall 40.5 (12.0) 659.2 (531.2) GAR Guntersville 58.4 (5.25) 765 (170) 3M Outfall 48.3 460.7 (5.3) (170.4) SB Guntersville 23.3 168 (2.17) (55.6) --val:-u-e-s--a-r-e-m---e-ans with standard deviations (in parentheses) Liver wt (g) 1.32 2.92 1.95 (0.985) 0.42 (0.04) 5.6 (0.18) 6.8 (3.8) 6.4 (4.3) 3.6 (1.9) 1.01 (0.34) Due to the small sample size and large amount o f variability in the endpoints, there were no statistically significant different between the sites for species common to both the Outfall and Guntersville locations. Physiological condition factors were calculated for each species and are given (Table 8, Appendix G). Statistical analysis o f both condition factor and LSI showed no significant differences between sites for the species collected from both locations. The lack o f any statistical significance was influenced by large amount variability in the data and by the 18 small sample size. The small sample size was due the fact that only a few species were collected at both locations. Furthermore, replicate fish samples were only collected for catfish and gar at both Guntersville and the Outfall location while largemouth bass, only a single fish was collected at both locations. Table 8. Species and location specific condition factor (K) and liver somatic index (LSI) for fish collected from the Tennessee River in the Decatur Alabama area. Species Location K2 LS? Largemouth Bass 3M Outfall 1.3855 2.92 Shad White perch Catfish Largemouth Bass 3M Outfall 3M Outfall 3M Outfall Guntersville 0.8207 (0.1224) 1.6146 (0.0481) 0.8724 (0.0102) 1.0478 0.4772 (0.2877) 0.5463 (0.0043) 1.1935 (0.3923) 0.6769 Catfish Guntersville 1.0743 (0.1142) Gar Guntersville 0.3815 (0.0180) Striped Bass Guntersville 1.2094 (0.1207) T T -------- ------------ Data presented as mean and standard deviation 2 Condition factor (K) was calculated as K= (W t x 100)/L3 3 Liver som atic index (LSI) was calculated as LSI= (Liver wt/total wt) xlOO 1.5826 (0.5186) 0.8649 (0.6137) 0.6241 (0.1902) 7.4.1 Concentrations o fPFOS, FOSA, PFOA, AND PFHS infish livers. Fluorocarbons were detected in all livers removed from fish collected at both the Guntersville and Outfall locations (Table 9, Appendix H). The least liver concentrations for all measured fluorocarbons were observed in fish collected from the Guntersville location while the greatest concentrations were observed in Outfall fish. Average for PFOS, FOSA, PFOA, and PFHS liver concentrations for all fish collected from the Outfall were approximately 19-fold, 340-fold, 109-fold and 710-fold greater than those observed in Guntersville fish, respectively. 19 Average liver concentrations for PFOS, FOSA, PFOA, and PFHS for all fish collected at the Guntersville site were 1036, 22.0,22.1 and 3.79 pg/kg wet weight, respectively. Table 9. Liver concentrations o f selected fluorocarbons in fish collected from the Tennessee River in the Decatur, Alabama area. a___________________________ Species G u n tersv ille PFOS FOSA PFOA ____ (Pgfcg)______ (Pgfcg)__ (gg/kg) PFHS (gg/kg) Catfish 87.3 20.3 18.8b 3.75b (22.2) (5.0) Gar 249.3 20.95 18.8b 4.58 (185.9) (16.4) (1.65) Striped Bass 1330.0 (763) 23.19 (10.0) 24.7 (17.6) 3.75b Largemouth Bass 643 18.8b 18.8b 3.75b O u tfall Catfish Gar Shad White perch Largemouth bass 10093.5 (5,777.8) 13892.0 (3073.1) 9195.2 (13252.2) 49285.0 (37657.7) 27143.0 14521.5 (2090.9) 5,089.0 (3,739.2) 4,420.0 (4047.3) 17403.0 (14679.5) 2456.0 252.9 (277.3) 1,934.5 (652.7) 96.9 (49.0) 1628.5 (1228.2) 3.75 30.2 (28.6) 256.5 (89.9) 129.1 (145.3) 2069.5 (1941.0) 11.6 a Concentrations are reported as m eans and standard deviations. b R esults was reported as the lim it o f quantitation: 18.8 pg/kg for PFOS, FOSA, PFOA and 3.75 pg/kg for PFHS 0 Estimated value. Differences in liver fluorocarbons concentrations were also observed between fish species collected at the site. Liver concentrations for the selected fluorocarbons increased in the following manner: PFOS: FOSA: Catfish < Gar < Largemouth bass < Striped bass (Largemouth bass)< Catfish < Gar < Striped bass 20 PFOA: PFHS: (Catfish = Largemouth bass) < Gar < Striped bass (Striped bass = Catfish < Largemouth bass) < Gar The fish in parentheses indicate liver fluorochemical concentration below the LOQ. Except for PFOS, there were no statistically significant differences in fluorochemical liver concentrations between fish species collected at Guntersville. For PFOS, there was a 160-fold difference between catfish that had the least concentration and striped bass with the greatest liver concentration. FOSA, PFOA, and PFHS liver concentrations in fish from Guntersville were all less than their respective LOQs. The significance o f these results and a true understanding o f the magnitude o f the observed differences at this site are unknown at this time due to the variability in the data and small sample size. Average liver concentrations for PFOS, FOSA, PFOA, and PFHS for all fish collected from the Outfall location were 15,692, 7,195, 581, and 384 pg/kg wet weight, respectively. As was observed at the Guntersville site, differences in liver fluorochemical concentrations were also observed between fish species collected at the site. Liver concentrations o f the selected fluorochemicals at Outfall site increased as follows: PFOS: FOSA: PFOA: PFHS: Shad < Catfish < Gar < Largemouth Bass < White Perch Largemouth Bass <Shad< Gar < Catfish < White Perch Largemouth Bass < Shad < Catfish < White Perch < Gar Largemouth Bass < Catfish < Shad < Gar < White Perch Except for PFOA, there were no statistically significant differences in species specific liver concentrations o f the selected fluorocarbons for fish from the Outfall. For PFOS, FOSA, and PFHS there were 5.4, 7.1 and 178-fold differences between the least and greatest liver concentrations. However, due to the small sample size and large amount of variability observed in the data, these differences were not statistically significant. For PFOA liver concentrations, there was a 516-fold difference between the least and greatest concentration. However, only shad and gar liver concentrations were significantly different. Largemouth bass was not included in the analysis due to the fact that there was 21 only a single value for this species. The magnitude and significance of these differences is unknown at this time due to the variability in the data and small sample size. 7.4.2 Wholefish concentrations o fPFOS, FOSA, PFOA, and PFHS Fluorochemicals were detected in all fish collected from both the Guntersville and Outfall locations (Table 10, Appendix I). Table 10. Whole body concentrations (wet weight) o f selected fluorochemicals in fish collected from the Tennessee River in the Decatur Alabama area. a PFOS FOSA PFOA PFHS Species (pg/kg) (Pg/kg) (Pg/kg) (Pg/kg) G u n tersv ille Catfish 7.5b 7.5b 20. l b 7.5b Gar Striped Bass Largemouth Bass 15.1 (15.2) 65.6 (23.2) 230 7.5b 10.4 (0.96) 8.76 20. l b 8.03b 8.03b 7.5b 7.5b 7.5b O u tfall Catfish 1190 (170) ND 119.6 11.4 (140.6) (5.52) Gar 1863 ND 154.5 35.5 (1679) (40.3) (25.0) Shad ND ND 54.0 38.9 (48.2) (60.2) White perch ND ND 278 405.5 (69.3) (129.4) Largemouth bass ND 557 20. l b 7.5b Concentrations are reported as m eans and standard deviations. b R esults w as reported as the lim it o f quantification:7.5 pg/kg for PFOS, F O SA PFHS, 8.03 pg/kg for PFOA N D signifies that the compound is present but not quantified due to the data not m eeting quality control criteria. 22 The least fluorocarbon concentrations were observed in fish collected from the Guntersville location while the greatest concentrations were observed in fish from near the Outfall. Mean PFOS, FOSA, PFOA and PFHS concentration for all fish collected from the Outfall were approximately 23-fold, 2-fold, 9-fold and 12-fold greater than those measured in Guntersville fish, respectively. Mean whole body concentrations for PFOS, FOSA, PFOA and PFHS for all fish collected from the Guntersville site were 59.1, 9.43, 11.7 and 7.50 pg/kg wet weight, respectively. Species differences in whole body concentration at the site were also observed with concentrations increasing as follows: PFOS: FOSAA: PFOA: PFHS: Catfish < Gar < Striped bass < Largemouth bass Catfish = Gar < Largemouth bass < Striped bass Largemouth bass = Striped bass < Catfish < Gar All species equal due to being at LOQ Due to the small sample size, there were no statistically significant differences between whole body concentrations for the species collected at this site. Furthermore, except for PFOS, there was less than a 3-fold difference between the least and greatest fluorochemical concentration in the species being evaluated in this study. For PFOS, there was a 30-fold difference between the least (catfish) and the greatest (largemouth bass) concentrations indicating that there is a potential for differences in the bioaccumulation o f PFOS by fish species at this site. Mean whole body concentration for PFOS, FOSA, PFOA and PFHS for fish caught at the Outfall were 1332.0, 20.1, 106.4, and 86. 9 pg/kg wet weight, respectively. As observed at the Guntersville location, differences in whole body fluorochemical concentrations were also observed between species caught on site. Whole body concentrations o f fluorochemicals in Outfall site increased as follows: 23 PFOS: FOSA: PFOA: PFHS: Gar < Catfish (Shad, White perch, Largemouth bass) Largemouth bass (Gar, Catfish, Shad, White perch) Largemouth bass < Shad < Catfish < White perch Largemouth bass < Catfish = Shad < White perch The species listed in the parentheses were not quantified during the analyses and could not be evaluated in this report. As with the Guntersville fish, small sample size precluded any statistical analysis of the data to evaluate differences between species at the Outfall location. Thus, while these data show that there are differences between locations and between fish and locations, the small sample size precludes an evaluation o f the significance and magnitude o f these differences. 7.5 Fluorochemicals in Clam Tissues Fluorochemical concentrations were also measured in clam tissues collected at both sites (Table 11, Appendix H). PFOS concentrations did not differ between the sites and ranged from 15.6 pg/kg to 14.1 pg/kg wet weight at Guntersville and the Outfall, respectively. In contrast, there was a 42.8-fold difference in FOSA clam tissue concentrations with 25.1 pg/kg and 1074 pg/kg, wet weight, being measured at Guntersville and Outfall locations, respectively. There were no significant differences in the concentration o f PFOA and PFHS in clam tissue collected at each of the two sites. The results o f the clam analyses reflect the areas from which the samples were collected. Due to substrate problems, Outfall clams were collected at the Fox Creek location. This location had surface water and sediment fluorochemical concentrations that were statistically similar to those observed at the Guntersville location. Thus, these results indicate that for at least clams, the environmental concentrations o f fluorochemicals downstream o f the 3M Outfall have returned to background levels within in the Tennessee River. 24 Table 11. Tissue concentrations o f selected fluorochemicals in clams collected from the Tennessee River in the Decatur Alabama area. 1 Location PFOS (pg/kg) FOSA (pg/kg) PFOA (gg/kg) PFHS (Pg/kg) Gtintersville 15.62 25.1 4.382 0.9462 Outfall 14.12 1074 8.422 0.372 (13.7) (1358) J7-1SL. - 1 Concentrations presented as means and standard deviations. (0-53) 2 Lim its o f quantitation: 18.8 gg/kg for PFOS, FOSA, PFOA, 3.75 gg/kg for PFHS. Concentrations reported below LOQs are estim ated values 8 REFERENCES Hansen, K.J. and H.O. Johnson. 2000. Determination o f perfluorooctanoate sulfonate (PFOS), perfluorooctane sulfonylamide (FOSA), and perfluorooctanoate (PFOA) in water by liquid-solid extraction and high-performance liquid chromatography/tandem mass spectrometery (HPLC/MS/MS). 3M Environmental Laboratory method. Number ETS-8-154.0. 3M Environmental Laboratory, St. Paul, MN. LIMs Report numbers: E00-2361, E001958, E01-0520. 25 9 APPENDICES Appendix A. Structure and chemical characteristics o f selected fluorochemicals monitored in water, sediment and biota o f the Tennessee River in the Decatur Alabama area. Appendix B. Water and Sediment locations within the Tennessee River study areas Appendix C. Water, sediment, clams, and fish sampling protocols Appendix D. Centre Analytical Laboratories analytical report for the characterization o f fluorochemicals in water and sediment. Appendix E. Water quality data collected at sample locations. Appendix F. Topographic maps o f sample locations in the Tennessee River study area. Maps include water quality data, water or sediment fluorochemical concentrations for sample locations. Figure 1. Water quality parameters Figure 2. Fluorochemicals in surface waters Figure 3. Fluorochemicals in sediments, wet weight Figure 4. Fluorochemicals in sediments, dry weight Appendix G. Physiological data for fish collected from the Tennessee River. Appendix H. Fluorochemicals concentrations in fish livers and clam tissues collected from the Tennessee River in the Decatur Alabama area. A ppendix! Whole body fluorochemical concentrations in fish collected from the Tennessee River in the Decatur Alabama area. 26 Centre Analytical Laboratories, he. 3 0 4 8 Research Drive S tate C ollege, PA 16801 w w w .c e n tre lab .c o m Phone: (814) 231 -8032 Fax: (814) 231 -1253 or (814) 231-1580 Analysed Report Fluorochem ical C haracterization o f W ater and Sedim ent Sam ples M SU -Entrix (Tennessee River) FA C T-G EN -037 (E00-1958) C entre A nalytical Laboratory R eport No. 0 2 3 -0 1 4 0 (R evision 2) Revision Date 6/28/01 Testing Laboratory Centre Analytical Laboratory, Inc. 3048 Research Drive State College, PA 16801 3M Environmental Laboratory Contact Kent R. Lindstrom Bldg. 2-3E-09 P.O. Box 33331 S t Paul, MN 55133-3331 Phone: (651)778-5352 Requester Dale Bacon Ph.D. 3M Environmental Technology & Safety Services Bldg. 2-3E-09 P.O. Box 33331 St. Paul, MN 55133-3331 PAGE 10 F S 1 Introduction Results are reported for the analysis of a series of water and sediment samples received by Centre Analytical Laboratories, Inc. (Centre) from the 3M Environmental Laboratory. The samples were collected from the Tennessee River and are part of 3M Project E 00-1958. The Centre study number assigned to the project is 023-014. Specific fluorochemical characterization by liquid chromatography / tandem mass spectrometry (LC /M S/M S) was requested for all samples. A total of 66 samples were received for analysis. The samples were prepared and analyzed by LC/M S/M S for the following list of fluorochemicals: Table 1: Target Analysis Compound Nam e Perfluorooctane Sulfonate Perfluorooctane Sulfonvtamide Periluorooctanoate Perfluorohexane Sulfonate Acronvm PFOS PFOSA POAA PFHS T he analytical methods used for water samples were validated by Centre. The validation protocol and results are on file with Centre. The methods w ere modified for the sediment samples, however the procedures have not been fully validated for this matrix. D ata presented here is the highest quality data available at this time. 2 Sample Receipt The samples were submitted in individual plastic containers and were not preserved. Sixty-six individual sam ple containers were received. Samples were received on 7/26/00. The samples were collected between 6/19/00 and 6/22/00. Chain-of-custody information is presented in Attachment C . 3 Holding Times T he analytical method used was validated against a maximum holding tim e of 14 days. Samples w ere received after the 14-day holding tim e. However, it should be noted that field fortifications in water and other matrices have shown acceptable recoveries at 100 and 1000 ng/L for periods longer than 14 days. PACE2 OF6 4 Methods - Analytical and Preparatory 4.1 LC/M S/M S 4.1.1 Sample Preparation for LC/MS/MS Analysis W ater samples were initially treated with 200 uL of 250 mg/L sodium thiosulfate solution to remove residual chlorine. Solid phase extraction (SPE) was used to prepare the samples for LC/M S/M S analysis. A forty-milliliter portion of sample was transferred to a C i8 S PE cartridge. The cartridge was first eluted with 5 mL of 40% methanol in water solution. The etuate was discarded and the SPE column was then eluted with 100% methanol. A 5 mL portion of methanol was collected for analysis by LC/M S/M S. This treatm ent resulted in an eight-fold concentration of the samples prior to analysis. For the sediment samples, a representative portion of sample (5 grams) was first extracted into 5 m l of methanol. The extracts were filtered and diluted to a final volume of 4 0 mL with Type I water. The diluted (tracts were then treated in the sam e m anner as the w ater samples, beginning with the solid phase extraction. 4.12. Sample Analysis by LC/MS/MS In HPLC, an aliquot of extract is injected and passed through a liquid-phase chromatographic column. Based on the affinity of the analyte for the stationary phase in the column relative to the liquid mobile phase, the analyte is retained for a characteristic amount of tim e. Following HPLC separation, ES/M S provides a rapid and accurate means for analyzing a wide range of organic compounds, including fluorochemicals. Electrospray is generally operated at relatively mild temperatures; molecules are ionized, fragmented, and detected. Ions characteristic of known fluorochemicals are observed and quantitated against standards. A Hewlett-Packard H P1100 HPLC system coupled to a Micromass Ultima M S/M S was used to analyze the sample extracts. Analysis was performed using selected reaction monitoring (SR M ). W ater samples were extracted on 8/2/00 and 8/3/00 and analyzed by M S /M S between 8/3/00 and 8/17/00. Sediment samples were extracted 8/4/00 and 8/10/00 and w ere analyzed ' by M S /M S between 8/8/00 and 8/17/00. T he H PLC and M S/M S methods used for analysis and instrument parameters can be found in Attachments O and E. 5 Analysis 5.1 C alibration A 7-point calibration curve was analyzed at the beginning and end of the analytical sequence for the compounds of interest. The calibration points were prepared at 0, 25, 5 0 ,1 0 0 ,2 5 0 , 500, and 1000 ng/L (ppt). The instrument response versus the concentration was plotted for each point Using linear regression with 1/x weighting, the slope, y-intercept and correlation coefficient (r) and coefficient of determination ( r ) were determined. A calibration curve is acceptable ff r >.0.985 (r2 > 0.970). Calibration standards are prepared using the sam e SPE procedure used for samples. P A G E 30F6 Cafibration check standards were analyzed periodically (every three to five sample injections) throughout the analysis sequence. Compliance is obtained if the standard analyte concentrations are within +/-20% of the actual value. For the results reported here, calibration criteria were m e t 5.2 Blanks Extraction blanks were prepared and analyzed with every extraction batch of samples. The extraction blanks should not have any target analytes present at or above the concentration of the low-level calibration standard. For these samples, the extraction blanks were compliant. Instrument blanks in the form of dean methanol solvent were also analyzed after every highlevel calibration standard, and after known high-level samples. Again, the blanks should not have any target analytes present at or above the low-level calibration standard. For the samples presented here the instrument blanks are compliant. 5.3 Surrogates Surrogate spikes are not a component of the LG/M S/M S analytical method. 5.4 M atrix Spikes Matrix spates were prepared for every field sam ple using all compounds of iite re s t Matrix spike recoveries are given in Attachment B. Field spikes were submitted with one sedim ent and one surface water sam ple. Field spike recoveries are also included in Attachment B. The sediment sample showed no recovery of the field spike, indicating that there was a problem with the sediment field spiking procedure. Field control samples spiked at 200 ppt and 1000 ppt were submitted. T he field control samples showed recoveries between 7 0 and 130% . The results are also included in Attachment B. 5.5 Duplicates All field samples were analyzed in duplicate. Results are given along with the sample results in Attachment A. 5.6 Laboratory C ontrol Sam ples For LG/M S/M S analyses, Milliq water was spiked with all compounds of interest at 2 5 and 250 ' ng/L during each extraction s e t All recoveries for all compounds were between 70-130% in each LCS. Results are given along with the raw data In Attachments 0 and E. 5.7 Sam ple R elated Com m ents There are no other sam ple related comments for this data s e t 6 Data Summary Please see Attachment A for a detailed listing of the analytical results. Surface water results are reported in parts per trillion (ppt) (ng/L). Sedim ent sam ple results are reported in parts per billion (ppb) (u g k g ) on both an as-received and dry-weight basis. PAGE4 OF6 7 Data/Sample Retention Samples are disposed of one month after the report is issued unless otherwise specified. All electronic data is archived on retrievable media and hard copy reports are stored in data folders maintained by Centre. 8 Attachments 8.1 Attachment A: Results 82 Attachment B: Matrix Spike Recoveries 8 .3 Attachment C: Chain of Custody 8 .4 Attachment D: LC/M S/M S Raw Analytical Data (Surface W ater Samples) 8.5 Attachment E: LC/M S/M S Raw Analytical Data (Sedim ent Samples) 9 Signatures Date Other Lab Members Contributing to Data Karen Smith RAGE5 O F6 SECTION A ANALYTICAL REPORT Centre Analytical Laboratories, Inc. 3048 Research Drive, State College PA 16801 814-231-8032 FAX 814-231-1253 I A nalytical R e su lti GEN-037 Surface W aters 3M Sam ple Identification SW-FB01 SW-13 SW-13 Dup SW-14 SW-14 Dup SW-15 SW-15 Dup SW-16 SW-16 Dup SW-17 SW-17 Dup SW-18 SW-18 Dup SW-18 Field Duplicate SW-25 SW-25 Dup SW -26 SW-26 Dup SW-01 SW-01 Dup SW-02 SW-02 Dup SW-03 SW-03 Dup SW-04 SW-04 Dup SW -05 SW-05 Dup SW -06 SW-06 Dup SW-06 Field Duplicate SW-11 SW-11 Dup SW-11 FMS1 SW-11 FMS2 SW-FSCS1 SW -FSCS2 SW-12 SW -12 Dup SW -12 Field Duplicate SW -27 SW-27 Dup D ate C ollected 6/20/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 PFOS (ng/L) NQ 60.2 65.2 67.0 64.0 49.7 59.7 53.5 529 41.2 68.4 329 33.2 40.6 156000 145000 77300 78300 131000 154000 66700 67700 11900 13300 14500 10900 14800 18200 11900 11900 15200 565 497 687 1350 223 993 565 574 588 592 579 PFOSA (ng/L) ND NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ 7880 8090 6960 6700 7930 8830 4400 4840 1120 1160 1300 878 1430 1610 1020 988 1140 151 141 316 1050 219 717 184 188 190 181 188 POAA (ng/L) ND 30.7 32.0 30.1 32.1 25.6 30.3 28.6 29.2 NQ 420 NQ NQ NQ 1900000 1900000 58200 57200 1700000 1600000 748000 753000 117000 126000 107000 102000 162000 187000 104000 104000 143000 2050 1920 2040 2470 192 903 46 3030 2990 3410 3510 PFHS (ng/L) ND NQ NQ 26.4 NQ NQ NQ NQ NQ NQ NQ <25 <25 NQ 43900 41500 12100 11500 35300 38800 17000 17700 3070 3460 4250 3210 4130 5030 3920 3800 4470 446 410 584 1090 194 889 637 654 635 658 658 Limit of Detection (LOD) for the procedure is appoximately 2.5 ng/L for PFOS, PFHS, and PFOSA and 7.5 ng/L for POAA Limit of Quantitation (LOQ) for the procedure is 25 ng/L for all com pounds ND - Compound not detected NQ-Compounddetectedata levelbetweentheLODandLOQ. Resultisnotquantifiable. ND < LOD < NO < LOO P lease re fe r to th e reve rse sid e to r o u r sta n d a rd term s a n d co n d itio n . ANALYTICAL REPORT Centre Analytical i Laboratories. Inc. 3048 Research Drive, State College PA 16801 814-231-8032 FAX 814-231-1253 I A nalytical R esu lts GEN-037 S urface W aters I 3M Sam ple Identification SW -07 SW-07 Dup SW -08 SW-08 Dup SW -09 SW-09 Dup SW -10 SW-10 Dup SW -19 SW -19 Dup SW -20 SW-20 Dup SW-21 SW-21 Dup SW-22 SW-22 Dup SW -23 SW-23 Dup SW-24 SW-24 Dup SW-FB02 D ate C ollected 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/22/00 PFOS (ng/L) 509 464 319 315 97.6 98.8 116 109 NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ ND PFOSA (ng/L) 149 143 99.4 96.3 36.1 35.4 34.4 31.2 ND ND NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ ND POAA (ng/L) 2220 2050 1630 1430 562 559 925 882 ND ND ND ND ND ND ND ND ND ND ND ND ND PFHS (ng/L) 520 474 324 305 86.3 83.3 110 96.9 ND ND ND ND ND NO ND fO ND ND ND ND ND Limit of Detection (LOD) for the procedure is appoximately 2.S ng/L for PFOS, PFHS, and PFOSA and 7.S ng/L for POAA Limit of Quantitation (LOQ) for the procedure is 25 ng/L for all compounds ND - Compound not detected NQ - Compound detected at a level betw een the LOD and LOQ. Result is not quantifiable. ND < LOD < NQ < LOQ P lease re fe r to the reve rse sid e fo r o u r sta n d a rd term e an d co n d itio n s. ANALYTICAL REPORT SSSS&k. Centre Analytical i vrSv8B>. Laboratories, Inc. 3048 Research Drive, State College PA 16801 814-231-8032 FAX 814-231-1253 A nalytical R eaulta GEN-037 Sedim ents 3M Sam ple Identification D ate C o lectad PFOS (ug/Kg) (a s received) PFOSA (ug/Kg) (a s received) Sed-11 Sed-11 Dup Sed-12 Sed-12 Dup Sed-12 R ed Dup Sed-13 Sed-13 Dup Sed-14 Sed-14 Dup Sed-1S Sed-15 Dup Sed-16 Sed-16 Dup Sed-17 Sed-17 Dup Sed-18 Sed-18 Dup Sed-18 R ed Dup Sed-2S Sed-25 Dup Sed-27 Sed-27 Dup 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/20/00 6/20/00 1.56 1.62 3.47 2.85 4.80 1.62 1.38 0.43 0.506 1.53 1.30 0.452 0.574 1.03 0.896 0.824 0.816 1.38 5750 6110 3.63 3.54 0.818 0.848 2.48 2.09 3.50 0.255 0.225 NQ NQ 0.248 NQ ND ND NQ NQ NQ NQ NQ 1170 1230 1.55 2.90 Limit of Detection (LOD) for the procedure Is appoxknately 0.08 ug/Kg for all compounds Limit of Quantitation (LOQ) for the procedure Is 0.20 ug/Kg for a l compounds ND - Compound not detected NQ - Compound detected at a level betw een the LOD and LOQ. Result is not quantifiable. ND < LOD < NQ < LOQ POAA (ug/Kg) (as received) 1.17 1.18 2.34 2.31 2.32 NQ NQ ND ND NQ ND ND ND ND ND NQ NQ NQ 1930 1780 2.49 2.76 PFHS (ug/Kg) (a s received) NQ NQ 0.278 0.295 0.310 NQ ND ND ND ND ND ND ND ND ND ND ND ND 132 138 0.330 0.320 P le a s* re fe r to the reverse sid e to r o u r stan dard term s and co n d itio n s. r #S5eBBi ANALYTICAL REPORT Centre Analytical 1 Laboratories. Inc. 3048 Research Drive, State College PA 16801 814-231-8032 FAX 814-231-1253 I A nalytical R esu lts GEN-037 S edim ents 3M Sam ple Identification D ateC oB ected PFOS (ug/Kg) (a s received) PFOSA (ug/Kg) (as received) Sed-01 Sed-01 Dup Sed-02 Sed-02 Dup Sed-03 Sed-03 Dup Sed-04 Sed-04 Dup Sed-05 Sed-06 Dup S e d -06 Sed-06 Dup Sed-06 Field Dup Sed-07 Sed-07 Dup Sed-08 Sed-08 Dup Sed-09 Sed-09 Dup Sed-10 Sed-10 Dup Sed-19 Sed-19 Dup e/19/oo 6/19I/D0 6/19/00 6/19/00 8/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/21/00 6/21/00 1400 1720 I860 2070 309 433 465 478 69.9 85.5 64.7 83.6 97.3 1.01 0.872 0.936 1.12 0.92 0.912 0.864 0.832 NQ NO 270 317 431 513 73.6 95.2 107 96.9 54.6 56.9 19.3 23.1 20.8 0.598 0.553 0.491 0.471 0.468 0.447 0.373 0.393 ND ND Limit ol Detection (LOO) for the procedure is appoximately 0.08 ug/Kg for all com pounds Limit of Quantitation (LOQ) for the procedure is 0.20 ut^Kg for ail compounds ND - Compound not detected NQ Compound detected at a level between the LOD and LOQ. Result is not quantifiable. N D <LO O <N O <LO Q POAA (ug/Kg) (a s received) 1010 862 1270 1230 453 527 654 643 105 135 36.7 47.7 41.9 0.880 0.848 0.710 0.740 1.08 1.07 0.582 0.612 ND ND PFHS (ug/Kg) (a s received) 36.5 39.6 67.4 69.4 16.9 22.8 24.5 21.9 722 7.79 5.99 6.56 6.24 ND NQ ND ND ND ND ND ND ND ND P le a s a n te r to the reve rse s id e fo r o u r sta n d a rd term s a n d co n d itio n s. ANALYTICAL REPORT 3048 Research Drive, State College PA 16801 814-231-8032 FA X814-231-1253 A nalytical R esu lts GEN-037 S edim ents 3M Sam ple Identification Sed-20 Sed-20 Dup Sed-21 Sed-21 Dup Sed-22 Sed-22 Dup Sed-23 Sed-23 Dup Sed-24 Sed-24 Dup D ate C ollected 6/21/00 3/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 PFOS (ug/Kg) (a s received) 0.217 0.318 NQ NQ NQ NQ NQ NQ 0.266 0.257 PFOSA (ug/Kg) (a s received) ND NQ ND ND ND ND ND ND NQ NQ POAA (ug/Kg) (a s received) ND ND ND ND ND ND ND ND NQ NQ Limit of Detection (LOD) for the procedure is appoidmately 0.08 ug/Kg for a l compounds Limit of Quantitation (LOQ) for th e procedure is 0.20 u^K g for an compounds ND - Compound not detected NQ Compound detected at a level betw een the LOD and LOQ. Result is not quantifiable. NO < LOD < NQ < LOQ PFHS (ug/Kg) (as received) ND ND ND ND ND ND ND ND ND ND P lease re fe r to the reverse s id e fo r o u r sta n d a rd term an d co n d itio n . ANALYTICAL REPORT A nalytical R esu lts GEN-037 S edim ents 3M Sam ple Identification Sed-11 Sed-11 Dup Sed-12 Sad-12 Dup Sed-12 Field Dup Sed-13 Sed-13 Dup Sed-14 Sed-14 Dup Sed-15 Sed-15 Dup Sed-16 Sed-16 Dup Sed-17 Sed-17 Dup Sed-18 Sed-18 Dup Sed-18 Field Dup Sed-25 Sed-25 Dup Sed-27 Sed-27 Dup Dato C ollacted 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/20/00 6/20/00 PFOS (ug/Kg) (dry w eight) 2.05 2.13 6.46 5.31 9.14 2.72 2.32 0.572 0.673 2.35 2.00 0.657 0.834 2.78 2.42 1.36 1.34 2.28 12200 13000 6.54 6.38 PFOSA (ug/Kg) (dry w eight) 1.07 1.12 4.62 3.89 6.67 0.429 0378 NQ NQ 0.382 0.235 ND ND NQ NQ NQ NQ NQ 2490 2620 2.79 5.23 POAA (ug/Kg) (dry w eight) 134 135 4.36 430 4.42 NQ NQ ND ND NQ ND ND ND ND ND NQ NQ NQ 4110 3790 4.49 4.97 Limit of Detection (LOD) for the procedure is appoximately 0.08 ug/Kg for aH com pounds (as received) Limit of Quantitation (LOQ) for the procedure is 0.20 ugK g for all com pounds (as received) ND - Compound not detected NQ - Compound detected a t a level between the LOD and LOQ. Result is not quantifiable. ND < LOD < NQ < LOQ PFHS (ug/Kg) (dry w eight) NQ NQ 0.518 0.549 0.590 NQ ND NO ND ND ND ND ND ND ND ND ND ND 281 294 0.595 0.577 P leas re fe r to th e reverse a id e fo r o u r sta n d a rd term s a n d co n d itio n s. ANALYTICAL REPORT 3048 Research Drive, State College PA 16801 814-231-8032 FAX 814-231-1253 A nalytical R esulta GEN-037 S edim ents 3M Sam ple Identification Date C o lectad PFOS (ug/Kg) (dry w eight) PFOSA (ug/Kg) (dry w eight) POAA (ug/Kg) (dry w eight) Sed-01 Sed-01 Dup Sed-02 Sed-02 Dup Sed-03 Sed-03 Dup Sed-04 Sed-04 Dup S e d -05 Sed-05 Dup Sed-06 Sed-06 Dup Sed-06 Field Dup Sed-07 Sed-07D up S e d -08 Sed-08 Dup Sed-08 Sed-09 Dup Sed-10 Sed-10 Dup Sed-19 Sed-19 Dup 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/19/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/20/00 6/21/00 6/21/00 2150 2640 3660 4070 472 662 671 690 94.6 116 86.9 115 133 1.36 1.18 2.04 2.44 1.96 1.94 1.89 1.82 NQ ND 415 487 848 1010 113.0 146 154 140 73.9 77.0 26.5 31.7 28.3 0.806 0.745 1.07 1.03 0.998 0.953 0.816 0.860 ND ND 1550 1320 2500 2420 693 806 944 927 142 183 50.4 65.5 57.1 1.19 t.1 4 1.55 1.61 2.30 2.28 1.27 1.34 ND ND Limit of Detection (LOD) for the procedure is appoximateiy 0.08 ug/Kg for aH compounds (as received) Limit of Quantitation (LOQ) for the procedure is 0.20 ut^Kg for a l com pounds (as received) ND - Compound not detected NO Compound detected at a level between the LOD and LOQ. Result is not quantifiable. ND < LOD < NQ < LOQ PFHS (ug/Kg) (dry w eight) 56.1 60.8 133 137 25.8 34.9 35.4 31.6 9.77 10.5 823 9.01 8.72 ND NQ ND ND ND NO ND ND ND ND P lease re fe r to th e reverse s id e fo r o u r sta n d a rd term s a n d co n d itio n s. ANALYTICAL REPORT 3048 Research Drive, State College PA 16801 814-231-8032 FAX 814-231-1253 A nalytical R esu lts GEN-037 Sadlm anta 3M Sam ple Identification Date CoUected PROS (ug/Kg) (dry w eight) PFOSA (ug/Kg) (dry w eight) POAA (ug/K g) (dry w eight) Sed-20 Sad-20 Dup Sed-21 Sed-21 Dup Sed-22 Sed-22 Dup Sed-23 Sed-23 Dup Sed-24 Sed-24 Dup 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 6/21/00 0.658 0.817 NQ NQ NQ NQ NQ NQ 0.600 0.580 ND NQ ND ND ND NO NO ND NQ NQ ND ND ND ND ND ND ND ND NQ NQ Limit of Detection (LOO) for th e procedure is appoximatety 0.08 ug/Kg for a t com pounds (as received) Limit of Quantitation (LOQ) for the procedure is 0.20 ug/Kg for ail com pounds (as received) ND - Compound not detected NQ - Compound detected a t a level betw een the LOD and LOQ. Result is not quantifiable. ND < LOD < NQ < LOQ PFHS (ug/Kg) (dry w eight) ND ND ND ND ND ND ND ND ND ND Pleaao re fe r to th e reverse M o to r o u r sta n d a rd torm a a n d co n d itio n s.