Document 7MjQ4Ga3qYoXJLaax4RQ7bEOV

TO FIELD POINT OR DEPT. 4 BLDG. NO. DATE YOUR LETTER Carl YockevAvon Lake General Chpmical Plant--;-------------------- ---------- OM FIELD POINT OR DEPT. & BLDG. NO. DATE THIS LETTER J. W. Born SUBJECT Rrprlfsvillp RAD PRECISION AND ACCURACY OF VINYL CHLORIDE CROSS-CHECKS Ja mu rv -2H ^ J Q8? RETENTION CODE 801-001 PROJECT REPORT You asked me for a.statement of the limits of precision anauiwaiiy or tne er -- Goodrich method of monitoring VCM in air. That method involves adsorbing the VCM from the air as the VCM-air mixture flows through a Bendix Flasher tube at 10.0 cc per minute. The VCM is subsequently desorbed thermally and analyzed by gas chromatography. The volume of air sampled and the weight of VCM collected are accurately measured. The con centration of VCM in the sampled air is then calculated on the basis that at 25 C and 1.0 atmosphere pressure 1.0 ppm`of VCM in air equals 2.56 mg of VCM per cubic meter of air. Our BFG inter-plant cross-checks used carefully-prepared VCM samples in Bendix Flasher tubes. Prior to loading the Flasher tubes, we carefully cleaned them. We used tubes which were dedicated to the cross-check program and so were not heavily contaminated with interfering substances. We cleaned them before each loading by repeated thermal desorptions until we obtained no significant GC absorption peak. Thus, our measurements gave the results which can be expected under ideal conditions. If the sampled air contains pollutants which interfere with the GC analysis of VCM, an analyst may not achieve as good results. The VCM cross-check results for the most recent five calendar quarters appear in the following table. 0SHA requires that the 95% confidence limits be 25% or less at 1.0 ppm concentrations of VCM in air and 35% or less-at 0.5 ppm. In all but the earliest "oss-check the 95% confidence limits were within the 0SHA requirements. I hope that this information covers your needs. If not, please let me know (extension 5503). it*- John W. Born Senior R&D Chemist JWB/pf cc: E.B.Katzenmeyer, Jr. T.S.Bialke P.M.Zakriski Att. r-70 1 iff If on ^ VO? O k-u TOOoOGOZ BFG-1956-E 11 /80 UTHO IN U.S.A. Starting Date 7/20/81 4/27/81 2/2/81 10/13/80 7/14/80 3/19/80 A SUMMARY OF THE BFGOODRICH INTERPLANT CROSS-CHECK RESULTS VCM Load (Wfl) * 5.0 5.0 10.0 5.0 10.0 10.0 VCM Recovery (%) 99.3 95.8 103. 99.9 . 94.4 92.1 Standard. Dev. (%)b 13.0 14.8 10.1 8.35 10.2 20.3 95% Confideni Limits (%) c 26. 30. 20. 17. 20. f+ 33. 1 - 45. d a. These VCM loadings per Flasher tubes were the intended loadings. The actual loadings were within 10% of the target-loadings. The 10.0-yg loading approximated a concentration of 1.0 ppm of VCM in the air. b. The units of standard deviation(%) are the same as the units of percent recovery. c. The units of the limits of 95% confidence also are the same as the units of percent recovery. d. In the cross-checkstudy for this one quarter, we had higher 95% confidence limits because of less clean Flasher tubes. W C Q O 3 N