Document zdz1zkjwvEnmawOBQYLZMZMka

CKMHILL Engineers Planners Economists Scientists AR226-2728 September 9, 1991 LMG32550.MD Mr. Michael McClusky E.I. Dupont de Nemours P.O. Box 1217 Parkersburg, West Virginia 26102 RE: Progress Report for FC-143 Method Development. Dear Sir: Since the beginning of work we have talked by telephone frequently concerning the FC-143 method development in progress at our laboratory. In brief summary, we have developed an analytical method capable of measuring FC-143 at levels below 0.1 ppb m water matrix. Further method refinement and ruggedness testing continue on schedule. Development costs to date are approximately $16K and are below projected estimates. The final costs are not expected to exceed plan. Details of the technical progress will be described below followed by a breakdown of the current expenses. TECHNICAL PROGRESS The analytical method uses liquid/liquid partitioning to extract PC-143 from a one--liter acidified water sample into an organic solvent. Method validation has already been completed using separatory funnel technique with diethyl ether as the solvent. We are hopeful that the approach can be expanded to include continuous extractors using dichloromethane solvent. Continuous extractors axe less likely to-produce emulsions with difficult samples, _and dichloromethane is a less hazardous solvent. Solid state extraction using granular activated carbon was briefly investigated but failed to offer early success. ^ Sample extracts are concentrated by boiling off the solvent using standard Kadurna Danish technique on a steam bath. A few milliliters of 1 N NaOH is mixed into the sample as a keeper before concentration. We have not yet tested the use of nitrogen blowdown to complete this procedure step where a large volume of solvent must be removed. There is some concern that the boiling action may be necessary to rapidly convert the acid form of the analyte to its salt and avoid volatility losses. Our laboratory is equipped with Zymark nitrogen blowdown units to test this option. The NaOH concentrate is transferred to a 4-dram screw cap vial and washed with dichloromethane to remove neutral and basic interferences. A centrifuge may be necessary to separate the liquid phases before discarding the organic phase. Sulfuric acid CH2MHILL Quality Analytical Laboratories 25 67FairlaneDrive, P.O. Box23054B, Montgomery, Alabama 36116 205.271.1444 Mr. Michael McClusky Progress Report for FC-143 Method Development, page 2 is then added to the aqueous concentrate to convert t h e a nalyte from the salt to the associated acid before performing a last minx extraction into ether. Early work attempted to dry the etherial extract after extraction using drying salts. Anhydrous potassium carbonate was tested for speedy drying. Deactivated anhydrous sodium sulfate was also tested. Both drying salts were found to produce measurable loss of analyte and were therefor removed from the procedure and replaced by concentration to dryness described below. One hundred microliters of concentrated ammonium hydroxide is added to the final "wet" etherial extract immediately before micro concentr a S o n on a steam bath. The final extract is quickly concentrated from approximately 10 mL to 0^5 5. ^ Ether is used to quantitatively transfer the extract . ., reaction vial. Gentle nitrogen blowdown is used to finish concentrating the sample to dryness. Acetyl chloride is carefully mixed with absolute ethanol to prepare a very drv 3% ethanolic HC1 solution. The sample residue is dissolved in 1.0 mL of the ethanolic HC1 solution and heated at 60C for 30 minutes to form the ethyl ester. Two final cleanups are performed before the esters are analyzed. rt.reaction vials hava cooled, 1,1 L of hexane and 1.0 of blank water is mixed into the sample with vigorous shaking. The ester will partition into the top liquid phase. Most of the floating phase is transferred to a 2-mL autosampler vial airea y containing approximately 1 mm of Flonsil. After shaking the vial for a few seconds, the extract is ready for analysis by ga chromatography with electron capture detection (GC/ECD). Two surrogates (perfluorononanoic acid and 11HEicosafluoroundecanoic acid) are spiked into every sample at the beginning of sample preparation to monitor-the success of that sample prep. Three internal standards are mixed into a measured portion of the final extract to monitor the success of data acquisition. FC-143 and the surrogates are denvatized to form esters because neither the acid nor the salt forms are amenable to gas chromatography. How should the instrument be calibrated, and how should sample results be calculated? These are issues which have several options. If pure and accurate ester standards were available for FC-143 and the surrogates, then these could be diluted and used for instrument calibration directly. Several standards were secured from three sources: PCR Research Chemicals, MTM research Chemicals of Lancaster Synthesis Inc., and FC-118 was provided by Dupont. For the perfluorooctanoic acid both the technical and the pure Mr. Michael McClusky Progress Report for FC-143 Method Development. page 3 . grades were purchased from PCR. Disturbing discrepancies were observed when various standards were compared. If calibration standards were to be prepared through the entire lengthy procedure for every batch of samples, one would worry about the variability of calibration from batch to batch. A reasonable alternative would be to prepare a calibration standard using only minimum preparation to reduce variability. This approach was taken for the validation work. The FC-118 supplied by Dupont was blown to dryness and assumed pure as a primary standard for FC-143 determination. A mixture of this FC-143 and the surrogate acids was introduced to the procedure at the derivatization step and carried from that point to the end of the preparation. Instrument calibration was performed by diluting the final preparation to create five concentrations injected into the instrument. Calibration curves were constructed to calculate FC-143 and surrogate recoveries for all subsequent samples. To test the linearity of the derivatization itself, the first five samples analyzed were five discrete samples at variable masses of FC-143 taken only through the derivatization and final cleanup steps. These five discrete partial preparations were virtually identical to dilutions of the one partial preparation diluted for calibration. Linearity of samples processed through the entire preparation is still under investigation. Ten samples were processed through the entire analytical method and are summarized in the table below. FC-143 SPIKED 0.0 ppb 0.1 ppb 0.1 ppb 0.1 ppb 0.1 ppb 0.1 ppb 0.1 ppb 0.1 ppb 0.2 ppb 0.4 ppb FC-143 FOUND 0.000 ppb 0.062 ppb 0.052 ppb 0.061 ppb 0.059 ppb 0.058 ppb 0.062 ppb 0.062 ppb 0.11 ppb 0.23 ppb ET-C9 SURR. 52% 51% 44% 49% 53% 50% 51% 54% 48% 53% ET-C11 SURR. 47% 48% 41% 46% 47% -48% 47% 50% 48% 46% These samples used laboratory blank water as the matrix. A significant bias is observed in the FC-143 data and is reflected in the surrogate recoveries. Analytical precision was good. Three standard deviations of the seven 0.1 ppb replicates was 0.011 ppb. Typical chromatograms are enclosed as well as the calibration curve described in this text. V Mr. Michael McClusky Progress Report for FC-143 Method Development. page 4 , Confirmation analysis may be accomplished by using a dissimilar analytical column. Chromatographic conditions have been developed for two dissimilar columns. A second approach to confirmation xs possible by derivatizing the acids to a different ester. The methyl, ethyl, n--propyl, and n--butyl esters have been successfully prepared. CURRENT PROJECT COSTS The table below offers a current breakdown of costs. Some figures are estimated since shipping costs or effective discounts are-not yet known. Labor - Consulting Chemist.... Labor - Senior Chemist........ Labor - Extraction Supervisor.. Labor - Associate Chemist..... Instrument Time - GC.......... Instrument Time - GC/MS....... . (129Hr) ($71.07/Hr) - $9168.03 (2Hr) ($48.93/Hr) - $ 97.86 (4Hr) ($50.00/Hr) - $ 200.00 (24.7Hr) ($35.33/Hr) = $ 872.65 (40.75Hr)($85.00/Hr) -- $3463.75 (OHr)($150.00/Hr) - $ 0.00 Reagents and supplies Analytical Standards Ordered 7/31........... . RTX-200 Capillary Column Ordered 8/1.......... Reaction Vials Ordered 8/1.................... Ethanol Reagent....... 24/40 Stoppers....... Acetyl Chloride...... Analytical Standards.. Reaction Vial Septa... Miscellaneous Supplies Communication........... .. 274.00 450.00 78.92 150.00 126.88 100.00 75.00 130.24 100.00 200.00 Total................................................ $ 15487.33 The next written report will accompany the final analytical method as well as summarize complete costs for the project. Jewell W. Smiley Consulting Chemist Gas Chromatography Supervisor cc: Craig Vinson/LMG Herb Kelly/LMG David Smith/LMG Analysis of Perfluoro-Octanoic Acid Samples Dave DuBois, Betsy McCord and Liz Lozada 10/2/2000 Summary and background: We received 4 samples from M. Kaiser to analyze for chemical structure. We analyzed them by 19F and 1H NMR. One sample was also analyzed by 13C NMR. Samples: Notebook 88570-8-1 88570-8-2 88570-8-3 88570-8-4 LI MS 68425 71402 71404 71405 Description 3M FC-118 20% PFOA Miteni RM 350T 20% PFOA MDF Recovered C-8 20% PFOA Dynan 30% PFOA GUS-NR-69641 Results: Sample 8-1 was typical of the 4 samples. It contains the following groups: CF3(CF2)x (CF3)2CF(CF2)x 3 types of CF3(CF2)xCF(CF3)(CF2)y Thus there are 4 types of CF3CF (-71 to -75 ppm; -184 to -188 ppm) groups and 4 additional types of -CF3 groups (-82 to -8 5 ppm), as shown in the about structures. There are a large number of signals in the -1 1 0 to -130 range which are CF2 groups. There was an unidentified signal at -6 3 ppm which might be -C F 2CI, CF2I, CF2Br, CH3ORf, or possibly a vinylidene CF3 or a tBu CF3. Sample 8-3 was almost identical to 8-1 except that it contained a small amount of CF3CFHCOOH (-76, -196 ppm). This material had sharper NMR lines. The 1H NMR showed a doublet of quartets at about 5.2 ppm that supports this assignment. The j values observed agreed with that of an authentic spectrum of A. Sieved. Samples 8-2 and 8-4 were almost identical to 8-1. Quantitative Results, relative mol%: Sample (ppm) 8_1 8_2 B_31 8_4 CF3(CF2)x (CF3)2CF(CF2)x -83.5 -73.9 78.9 8.61 79.0 7.9 72.4 10.2 83.0 7.4 CF3(CF2)xCF(CF3)(CF2)v c f 3c f h c o 2h A, -71.7) B. -71.9' C ,-72.3 -75.8,-195.5 4.3 j 3.3 4.4 0.0 3.9 3.5 4.7 0.0 5.1 4.4 5.7 1.2 3.3 2.3 3.6 0.0 pentad -63.3 0.7 1.0 1.0 0.5 Experimental: 19F NMR spectra were obtained on a 400 MHz Varian Inova spectrometer on a 5 mm probe on samples run neat with a D20 containing capillary for lock. Spectra were externally referenced to a capillary containing F-11 (0 ppm). Spectra were acquired at 30 C using an rf pulse of 2 usee (pw90 = 14 usee), a spectral width of 90 kHz, an acquisition time of 1.28 sec and a recycle delay of 20 sec. A total of 256 transients were acquired. 1H (with water suppression) and 13C (10 mm) spectra were obtained on the same instrument using standard conditions. Acknowledgement: We would like to thank A. Sievert for his help and advice with the 19F NMR spectra. *rl i G A o G ^.cf2 ^ cf2 ^,cf2 ^ F OPJ ^CFa vCF2 VCF2 ^0 CF3 D CF / cf3 c f xf2 CFZ CFj O B ?F3 hE CF:3 CCF,4^ CAF\ ,ifcF^ N 0 '' E JB ' L\2^3___ i___ 1 -- _________ :__LjJaJj>_ UUl. (..\ i...i r..| ~ r n i.| i i 1 1 j i i i i | n i t | i i i i ) i n i 'p r n i } " r t r i " | n i i ; i... n D JJU i j i i \ i y r -r -r r i i i i ",| " " n i " r r | r i n [i i i i| -60 -80 -100 -120 -140 -160 -180 -200 ppm Figure 1. 19F NMR of Sample 88 5 7 0 -8 -1 ,3M FC-118 20% PFOA Figure 2. 1:9 F NMR of Sample 88570-8-2, Miteni RM 350T 20% PFOA G A. G F cf3 c f 2\ c f^2 , c f 2s. c f^2 , c f 2 o O DCF3 l-'^ ,C F 2 ^ C F 2 '''C F ""CFz ''C fJ / _ CF3 H CF3 "CFH O J BH x3 l |,,,"T"""i...i1" i i i I.... r i" " |......1......................... i 'I1.....................1 "j -60 -80 -100 jd J u l D A -U i 1 i i j................ i......................... M i l i " ...i i" i i............i......................... i ........................ ( T " - | - r T i i i -120 -140 -160 -180 -200 i-- n r r i-- r i ........................i .........................r f i T ' i 'i..}.......... ppm Figure 3. 19F NMR of Sample 88570-8-3, MDF Recovered C-8 20% PFOA Figure 4. 19F NMR of Sample 88570-8-4, Dynan 30% PFOA GUS-NR-69641 Analysis of Perfluoro-Octanoic Acid Samples Dave DuBois, Betsy McCord and Liz Lozada 10/2/2000 Summary and background: PFOS chemicals are used in a wide range of products from fire-fighting foams, coatings for fabrics, leather, and some paper products, to industrial uses such as mist suppressants in acid baths. DuPont produces fluorochemicals based on the perfluoroctanyl iodide chemistry such as Telomer B alcohol. If DuPont can replace materials made with 3M's PFOS with materials based on DuPont's telomer chemistry, there would be a large potential replacement market for us. Our knowledge of persistence, bioaccumulation, and toxicity (PBT) of DuPont materials is incomplete based on the possible applications in the marketplace. CTO Dr. Joseph Miller established a team to quickly address the issues. The team's charter is to provide the science and technology basis to proactively address SHE concerns related to our products. Subteams on Toxicology, Exposure, Environmental Fate and Effects, and PFOA alternatives are working on solutions. Key to each of these teams is the Analytical subteam, which will need to provide product analysis via existing or newly developed methods. We received 4 samples from M. Kaiser to analyze for chemical structure. We analyzed them by 19F and 1H NMR. One sample was also analyzed by 13C NMR. Samples: Notebook LIMS 88570-8-1 68425 88570-8-2 71402 88570-8-3 71404 88570-8-4 71405 Description 3M FC-118 20% PFOA - Miteni RM 350T 20% PFOA MDF Recovered C-8 20% PFOA Dynan 30% PFOA GUS-NR-69641 i Results: Sample 8-1 was typical of the 4 samples. It contains the following groups: CF3(CF2)x (CF3)2CF(CF2)x 3 types of CF3(CF2)xCF(CF3)(CF2)y (CF3)3C(CF2)X Thus there are 4 types of CF3CF (-71 to -75 ppm; -184 to -188 ppm) groups and 4 additional types of -CF3 groups (-82 to -8 5 ppm), as shown in the about structures. There are a large number of signals in the -1 1 0 to -1 3 0 range which are CF2 groups. ' There was a signal at -6 3 ppm which was assigned to (CF3)3C(CF2)3 group. Its chemical shift and multiplicity correspond to values found in the literature. Sample 8-3 was almost identical to 8-1 except that it contained a small amount of CF3CFHCOOH (-76, -196 ppm). This material had sharper NMR lines. The 1H NMR showed a doublet of quartets at about 5.2 ppm that supports this assignment. The j values observed agreed with that of an authentic spectrum of A. Sieved. Samples 8-2 and 8-4 were almost identical to 8-1. The 1H NMR spectrum of 8-4 showed a singlet at 3.4 ppm, which may be methanol; spiking experiments are underway to verify this. Quantitative Results, relative mol%: j Sample CF3(CF2)x (CF3)2CF(CF2)X | (ppm) -83.5 -73.9 i 8_1 78.9 8.6 ; 8_2 79.0 7.9 j 8T 72.4 10.2 I 8_4 83.Q 7.4 CF3(CF2)xCF(CF3)(CF2)v c f3cf-h c o 2h (CF3)3C(CF2)3 A, -71.7 B. -71.9 C. -72.3 -75.8,-195.5 -63.3 4.3 3.3 4.4 0.0 0.2 3.9 3.5 4.7 0.0 0.3 5.1 4.4 5.7 1.2 0.3 3.3 2.3 3.6 0.0 0.2 Experimental: 19F NMR spectra were obtained on a 400 MHz Varian Inova spectrometer on a 5 mm probe on samples run neat with a D20 containing capillary for lock. Spectra were externally referenced to a capillary containing F-11 (0 ppm). Spectra were acquired at 30 C using an rf pulse of 2 usee (pw9Q = 1 4 usee), a spectral width of 90 kHz, an acquisition time of 1.28 sec and a recycle delay of 20 sec. A total of 256 transients were acquired. 1H (with water suppression) and 13C (10 mm) spectra were obtained on the same instrument using standard conditions. Acknowledgement: We would like to thank A. Sieved for his help and advice with the 19F NMR spectra. Go F ^C F 2 ^C F 2 ^CFa CPs ^CFa x CF2 'sCF2 'O O DCF3s u .C F2 -CF CF CF2 "CPs "O / cf3 yO^uc/Sr 3 o Ti G A. O CF3 J CF3 -\ I ^F2x f 2 xf2 E B x3 jU a IaJ JLik CF ' D JUL n~~T~r"1...11..1...'"I..1"1I I I I >...1..1..T1...1...1....'"-I...I I ! I I I.... I l l I I I I I I I"...I-)..I I I I I....I...|"t ITl I I...|-H"1..I l T|-|....I-I I I I I I I I I ) 1...r I [ l i l i i -60 -80 -100 -120 -140 -160 -180 -200 PPm Figure 1. 19F NMR of Sample 88570-8-1, 3M FC-118 20% PFOA ! Figure 2. 19F NMR of Sample 88570-8-2, Miteni RM 350T 20% PFOA i'l Figure 3. 19F NMR of Sample 88570-8-3, MDF Recovered C-8 20% PFOA Figure 4. 19F NMR of Sample 88570-8-4, Dynan 30% PFOA GUS-NR-69641 a j i! For Sample 88570-8-1 Compound: NMR mol% 3M Fax mo!% ,CF, ,CF, CF, 'CF, 'CF, 78.9 80% 8.6 8.5 CF, CF3 c. / cf3 ,CF2 CF, ` o o- 12.0 11.3 0.2 0.3