Document ykd09ezyrXom0ve8nGZBwJnm2

AR226-3335 Datamemo Date: September 21,2000 To: Barbara S. Larsen DuPont Science and Engineering Experimental Station, P.O. Box 80228 Wilmington, DE 19880-0228 From: Robert Bethem ALTA Analytical Laboratory Re: Development of an Analytical Procedure for the Analysis of PFOC and PFOS in Liquid Surfactants by LC-API/MS/MS INTRODUCTION A liquid chromatography atmospheric pressure ionization tandem mass spectrometry method (LC-API/MS/MS) was evaluated for the analysis of PFOC (FC-143) and PFOS (FC-95) with 3 product ion confirmation. A draft method (AIta method AMPF01) is being provided along with this Datamemo. Although the method was initially developed and evaluated for the analysis of both PFOC and PFOS, the draft provided was optimized more specifically for PFOC. The demonstrated range of reliable response for the method is 1 to 50 ppm. The method has not been rigorously validated. Experiments using standard additions to 3 different surfactants were conducted in order to evaluate the accuracy and feasibility of the analytical approach. The analytical raw data included with this report has been organized by analysis dates. A general description and summary of these results has been included with this report. METHODOLOGY The LC/MS/MS conditions developed for this method are routine and straightforward. The method was developed for 3 product ion confirmation and the instrument optimization was based on the least abundant product ion in contrast to a typical quantitative target analysis that would be optimized for the most abundant product ion. A simple reverses phase gradient was used for the separation. The success of this method seems to hinge on the ability to dilute and inject th^ surfactant products with acceptable precision and accuracy. Using a mixture of 50/50 (v/vy water/methanol as a diluent yielded very nice linearity and accuracy for PFOS but inconsistent results for PFOC. Using 100% water as the diluent seems to work much better for PFOC but the linearity and accuracy for PFOS deteriorated. Since the initial sample screening (3/29/00) indicated that there were no detectable amounts of PFOS in the samples provided, the final analysis and standard additions experiments used 100% water as the diluent (6/8/00). 000080001.att1.doc Page 1 of 5 Company Sanitized. Does not contain TSCA CBI A second limiting factor of the method is the inability to demonstrate a clean solvent blank for PFOC. In fact, the LOQ is limited by the amount of background found in the blanks. Although the PFOC background can be controlled, the background levels range from between 2% and 30%. In nearly all cases, the blanks injected at the beginning of the injection sequence contained the highest background levels with subsequent blanks showing less background as the analytical sequence progressed. The need to set the LOQ higher due to the background combined with the need to detect 3 product ions for confirmation limits the overall dynamic range for the method dramatically. For the work described here, the demonstrated calibration range was only 1-50 ppm compared to a typical API/LC/MS/MS quantitative method based on 1 product ion with a dynamic range of 250 to 1000. Results from 3 separate days of analysis are summarized on Tables 1 through 3. SAMPLE DESCRIPTION AND RECEIPT The following samples (product) were received in 1 shipment on 11/23/99 and logged into ALTA's sample control database (ALISTM) with unique sample identification numbers. ORGANIZATION OF RAW DATA The raw data provided has been organized into 4 folders. A description of the contents of these folders and the analyses conducted is summarized in the table below. Folder Date Notebook Description PFOC - Spectra and ChroMD 1/28/00 3/28/00 Vol. 100 p. 1-6 Instrument tuning, optimization for 3 ions, chromatography development, initial evaluation and range finding for product. PFOC - Alta 7709 3/29/00 Vol. 100 p. 7-9 Analysis of product. Datamemo dated April 4,2000 describing analysis and summary of work conducted to date. 50/50 water/methanol uses as the diluent. PFOC - Alta 7709 6/5/00 Vol. 100 p. 10-12 Analysis of product and standard additions using 50/'50 water/methanol. Standard addition to diluted product. PFOC - Alta 7709 6/8/00 Vol. 100 p. 13-14 Analysis of product using 100 water as the diluent. Standard additions made directly to the product prior to dilution. 000080001.att1.doc Page 2 of 5 Company Sanitized. Does not contain TSCA CBI DISCUSSION OF RESULTS 1. Need to run extensive studies to understand the nature of the carryover and/or background levels found in blanks. Probably need to evaluate different sources of reagents. We have found that background and carryover can be controlled more effectively by running isocratic methods or starting the gradient at higher organic concentrations. This should be evaluated also. 2. If a PFOS analysis is needed along with PFOC, the linerarity, precision and accuracy should be evaluate using different diluents and dilution schemes. 3. Once above issues are studies more thoroughly, a more exhaustive method validation should be conducted to determine the limit of detection, quantitation, precision and accuracy of the method. 000080001.att1.doc Page 3 of 5 Company Sanitized. Does not contain TSCA CBI TABLE 1 - Results From 3/29/00 Not Detected, The demonstrated limit of detection based on 3 ion confirmation is 1 ppm for samples diluted 1:1 signal to noise for the primary ion only, the estimated limit of detection for PFOS is 0.1 ppm. : TABLE 2 - Results From 6/5/00, ALTA ID ALTA Filename 7709_000 1 A000605112 7709J300 | A000605115 2 7709_000 | A000605117 3 Dilution PFOC ppm) STD Addition 81 45.5 Amt. Found 113 %Rec. 89% :20,000 202 182 351 91% 2.8 0.91 2.2 59% TABLE 3 - Results From 6/8/00, ALTA ID ALTA Filename Client Description 000080001.att1.doc Dilution PFOC (ppm) STD Addition Amt. Found %Rec. Page 4 of 5 Company Sanitized. Does not contain TSCA CB|