Document zdQ87qeav5qO3ybkGMo3321r6

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY Office of Air Quality Planning and Standards Research Triangle Park, North Carolina 27711 stmjtcr. Alternate Vinyl Chloride Test Methods Resubmitted by date: Nearer 9, 1991 Union Carbide room: William firlmley* Test Support Section Emission fbasurement Branch, ESED (MO 19 TO: Director, Surveillance and Analysis Division, Regions 11, III, IV, VI, IX In accordance with requests from Regions IV, IX, and Union Carbide, we have reviewed the vinyl chloride test items which Union Carbide has reslAmitted for consideration as alternate procedures. This review has beam dome In collaboration with the Quality Assurance Branch of the Envlroamontal Monitoring and Support Laboratory. The Items resubmitted were as follows: 1. Method 106, Section 4.3.2. Alternate chromatographic colwn Stainless Steel, 6.1 m x 3.2 mm, packed with 20 percent Tergltol E-3S on 60/99 mesh Chromasorb W AW. 2. Method 107. Entire method; attached. Based on the additional supporting information supplied by Union Carbide, both Items are now judged acceptable as alternate procedures. Because of widespread Interest, the alternate Method 107 (draft copy attached) will, with some refinement, be proposed as such In the Federal Register. Acceptance of the enclosed draft should satisfy Union CarbideV needs for the present. They will be expected to fellow tho method as It Is finally promulgated, of course. For your Information, we also expect to propose changes In both Methods 106 and 107 so that It will be possible for the method utor to perform a self appraisal to determine acceptance of a particular celimm as an alternate procedure. This change should greatly reduce the number of requests for alternate procedure approvals. Should you have any questions, please contact me. My FTS phone Is 629-5276; commercial phone is 919 541 5276. Enclosure cc: Robert Ajax (MD 13) Richard Biondi (EN 341) Alvin Chun, Region IX John Clements (MD 77) Joe Knoll (MD 77) James Wu, Region IV Susan Wyatt (MD 13) UCn 2. tonga and Sensitivity The lower limit of detection of vinyl chloride In dry PVC resin Is 0.2 ppm. For resin solutions, latexes and wet resin this limit rises Inversely as the nonvolatile (resin) content decreases. With proper calibration the upper limit may be extended as needed. 3. Precision and Reproducibility ^ A comparison of precision and reproducibility between the alternative method and Method 107 using a suspension resin homopolymer sample showed the alternative method had a standard deviation of 0.95 percent at the 340 ppm level and yielded mean results ten percent higher than Method 107. Method 107 In the same test had a standard deviation of 1.09 percent. A standard sample of latex containing 181.8 ppm vinyl chloride analyzed ten times by the alternative method showed a standard deviation of 7,5 percent and a mean error of 0.21 percent. A sample of vinyl chloride copolymer resin solution wes analyzed 9M times by the alternative method and showed a standard deviation of 6.C percent at a level of 35 ppm. The Increase In standard deviation on the latex and the resin solo*, tion samples Is due to the correction of the analysis from a gross bastfj to a contained dry solids basis. Both the latex and resin solution f * samples contained acetaldehyde along with other monomers and solvents. UCC 102032