Document M4GV4ROjRXv08jXpBLy2Mb3ZV

1992 DIFFUSION MONITOR TECHNICAL REPORT January 14, 1993 EXECUTIVE SUMMARY Accomplishments of the Diffusion Monitor program in 1992 include validation of the replacement carbon for the Organic Vapor Monitors, resolving formaldehyde manufacturing problems, MPDA and introduction of Improved packaging for the formaldehyde monitor to reduce the blank, improving the quality control for the 3510 analysis, MPDA and introduction of the 3530 OVM (3520 with laboratory analysis) and validation of the OVM for a number of compounds. Kuraray GA Acid Washed carbon was chosen as the replacement for the Witco 964 carbon which Is no longer available. Less than a one year supply of the Witco carbon remains. A four to five year supply of the new Kuraray carbon was purchased. The carbon was ground and sieved locally (Particle Technology, Inc., Burnsville, MN). Two full-scale batches were milled successfully (December,1991 and December, 1992) A third batch is scheduled for January, 1993. Introduction of the new carbon is tentatively planned for September, 1993. The Organic Vapor Monitor was revalidated for methylene chloride to provide additional documentation as the new OSHA standard is Introduced. Many experiments were run to validate the use of our monitors in response to customer Inquiries. This work will continue as an integral part of our Q90s effort. The high and erratic formaldehyde blanks which have plagued us for many years were determined to come from outgassing of the enamel can lining at elevated temperatures. An MPDA was prepared to provide a temporary cure by first sealing the monitor in a nylon bag and also enclosing a scavenger strip in the can. Significant progress has been made in our search for a foil bag replacement for the can. We anticipate making this change in 1993. Improvements to the formaldehyde quality assurance procedure have led to fewer failing lots of wafers and better limits of detection. An MPDA was prepared to Introduce the 3530 Organic Vapor Monitor (3520 with Analysis) in June. Sales through December were *25,000. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 3H 009055 1992 Diffusion Monitor Technical Report - January, 1993 CARBON QUALIFICATION The Witco 964 petroleum-based carbon used in the Organic Vapor Monitors Is no longer commercially available. Monitor Technical Reports for *1939-1990 and 1991 describe the work that resulted in the choice of th Kuraray GA Acid Washed coconut carbon as the replacement carbon. We ordered 1500 kg (3300 lbs.) from Kuraray which was ground and sieved by Particle Technology Incorporated of Burnsville, MN. This resulted In 1350 lbs. of monitor carbon ( a 41% yield and approximately a 5 year supply). A full-scale batch of wafers was milled successfully In December, 1991 and again In December, 1992. Extensive testing of the December, 1991 lot (No. 236) was accomplished and results were described in detail in Make-Order Follow-Up Report H3500-52. Recovery and capacity data are shown in the accompanying tables. Testing of the December, 1992 lot (No. 259) is in progress. A third lot is scheduled for January, 1993. After completion of these trial lots and field testing by technical service, an MPDA will be prepared. We anticipate introduction of the new carbon In September, 1993. The usual quality control tests for OVMs and ETO were run. The ETO wafers passed the recovery tests. The manufacturing QC MEK recovery was 0.919 +/- 0.012 (n=200) compared to a Witco control lot of 0.933 +/0.012 (n=35). Our lab evaluation showed a recovery after 1 day of 0.92 dry and 0.86 wet and after 2 weeks a recovery of 0.92 dry and 0.66 wet. Witco carbon after 2 weeks showed 0.91 dry and 0.52 wet. The manufacturing QC CC14 capacity was 5.36 +/- 0.21 mg/mll of wafer thickness (n=26) compared to 7.05 and 7.52 for two Witco control lots. We found 5.6 mg/mil for GAAW vs. 7.4 for Witco in our own lab tests. The density was 1.88 +/- 0.42 for GAAW and the caliper 14.3 +/- 0.94 mils vs. typical Witco of 1.61 and 15.5. The Kuraray GAAW carbon passed the wafer contamination test. No extraneous peaks were observed. A change in the specifications will be necessary for the CC14 test and density. Our lab test for determining capacity is to spike a monitor with an excess of solvent (usually 25 uL to 30 uL), allow it to stand capped overnight, allow it to stand open overnight, desorb with CS2 and then determine the amount retained. When the capacity was determined in this manner by spiking with 40 mg of CC14, we found 41 mg on the GAAW and 32 mg on the Witco wafers, while the Witco wafers appear to pick up more CCl4 initially in the manufacturing QC test, the GAAW wafers appear to retain more when exposed to air. Similarly, when the capacity for methylene chloride was determined, Witco retained 3.6 mg while GAAW retained 7.0 mg. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER O 3H 009056 1992 Dlffusl n Monitor Technical Report - January, 1993 A six hour exposure to 50 ppm methylene chloride (2X the proposed OSHA standard) at 80 X RH showed that both Witco and GAAW In double decker monitors collected the same amount (2.1 mg). The Witco monitors showed more on the second section than the GAAW, and were just barely within our requirements for less than 50 X on the second section. The GAAW Would allow for a little more leeway in meeting this requirement. A 5 1/2 hour exposure to 100 ppm methylene chloride under dry conditions (less than 30 X RH) again showed that both Witco and GAAW monitors collected the same amount (4 mg) with approximately 10 X of the methylene chloride found on the second section of the Witco monitors compared to 2 X on the GAAW monitors. A 7 1/2 hr exposure to 141 ppm of toluene at 50X RH showed that the Kuraray carbon collected within 2X of the Witco (Witco = 6.508 +/- 0.247 vs. Kuraray = 6.373 +/- 0.216 mg). The precision for Witco was 3.8X and for Kuraray was 3.4X. Recoveries and capacities have been determined for 63 compounds as shown In Tables 1-4. The Kuraray GAAW recovery at the 100 ug level was within 10 X of the Witco recovery for 49 of the compounds, lower for 11 and higher for 3. At the 30 ug spike level some of the cellosolves and epichlorohydrin had significantly lower recoveries with CS2 for the GAAW carbon. We may have to expand the number of compounds where we recommend alternate solvents for desorption. It will be even more necessary for customers to redetermine their own recoveries than for our previous Witco lot changes, but good industrial hygiene practice requires this anyway. In summary, the Kuraray GAAW carbon can be used to prepare 0VM monitors with the same sampling rates, similar but not identical recoveries, and somewhat improved capacities as the Witco carbon which is no longer available. Recoveries of very low concentrations of some oxygenated compounds are not as good as on the Witco petroleum carbon. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 3M 009057 1992 Diffusion Monitor Technical Report - January, 1993 TABLE 1 KURARAY GAAW RECOVERIES - JAN-FEB, 1992 VS. WITCO RECOVERIES - PREVIOUS WORK (Spikes In Micrograms - CSj Elution) CMPD- WITCO SPIKE WITCO REC QAAN SPIKE Acetone Acetonltrila Benzene n-Butyl aeatata n-Butyl alcohol Butyl calloaolva 2430 Butyl eelloeolve aeatata Carbon tetrachloride Calloaolva Calloaolva aeatata 280 Chloroform Decane Dimethyl foraamlda Codecana Epichlorohydrln Ethyl aeatata Ethyl alcohol Heptane Hexane Isopropyl alcohol 3S0 2360 3930 Methyl calloaolva Methyl calloaolva acetate Methylene chloride Nonane Octane Propylene glycol monomethyl ether Styrene Tetrahydrofuran Trichloroethylene Toluene 1,1,1-Trichloroethane m-Xylane 0.70A 0.82T 1.01A 0.B4A 0.80T 0.73T 0.3SA 0.98T 0.80A 0.33T 1.03P 0.99A 0.89A 1.04T 0.32T 1.04T 0.89T 0.85A 0.44T 1.00A 1.02A 0.6ST 0.45A 0.44T 1.00P 0.91T 1.02T 0.68T 0.94T 0.88A 1.09T 0.90A 0.97A 1.00T 1.08A 1.04A 1.04T 1.02A 32 31 35 35 32 33 38 64 37 39 80 29 34 30 43 36 31 27 26 31 37 40 33 29 28 39 36 35 59 35 54 35 T = OATA FROM MONITOR TECHNICAL REPORTS 3-91 t 2-92 A = MONITOR ANALYSIS LAB OATA P s PINK SHEET QAAM REC 0.77 0.54 0.96 1.00 0.40 0.13 0.69 0.87 0.09 0.75 0.70 1.01 0.02 1.10 0.38 0.82 0.23 0.43 0.92 0.48 0.00 0.63 0.81 1.01 1.04 0.00 0.77 0.72 0.90 0.99 0.93 1.02 0AAM SPIKE 95 94 103 106 97 108 113 191 112 117 179 88 113 90 142 106 94 82 79 94 111 121 159 86 84 116 109 106 176 104 161 104 QAAM REC 0.84 0.63 0.95 0.96 0.47 0.45 0.90 1.00 0.47 0.89 0.94 1.05 0.14 1.09 0.45 0.96 0.31 0.79 0.94 0.49 0.05 0.81 0.98 1.04 1.05 0.23 0.82 0.88 0.99 1.00 1.03 0.97 QAAM SPIKE 786 882 975 1492 902 684 QAAM REC 0.71 1.07 0.99 1.06 0.91 1.02 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 4 3H 009058 1992 Diffusion Monitor Technical Report - January, 1993 CMPO Aoyl alcohol Butyl Collooolvo Collooolvo Dlacoteno alcohol Dloothylforoaoldo Epichlorohydrln Furfural Furfuryl alcohol Mothyl Collooolvo Mothyl format# POME Phonyl othor Prooyl nltrato TABLE 2 KURARAY GAAW vs WITCO RECOVERIES January-July, 1992 Spikes in Mlcrograms CH2CL2 Elution SPIKE 29 33 33 34 34 43 41 41 3S 33 33 39 38 KUR REC 0.91 0.B1 0.B1 0.81 0.07 o.se 0.52 0.72 o.sa 0.32 0.75 0.30 1.03 WITCO REC 0.B0 0.70 0.70 0.82 0.12 0.93 0.59 0.73 0.78 0.78 0.85 0.43 1.04 SPIKE 97 108 112 112 113 142 139 138 118 117 111 131 127 KUR REC 0.81 0.89 0.78 0.80 0.65 0.72 0.82 0.71 0.86 0.78 0.86 0.41 1.03 MITCO REC 0.85 0.91 0.85 0.96 0.74 0.95 0.71 0.78 0.83 0.93 0.92 0.52 1.05 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 5 3H 009059 1992 Diffusion Monitor Technical Report - January, 1993 TABLE 3 KURARAY GAAW VS WITCO RECOVERIES January-July, 1992 Spikes in Micrograins CS2 Elution CMPD SPIKE KUR REC WITCO REC SPIKE KUR REC Acrylonitrile Allyl alcohol n-Amyl acetate Benzyl chloride Bromochloro- methane Brorooform s-Butyl acetate t-Butyl acetate Butyl Cellosolve Chlorobenzene Cumene Cyclohexanone Cyclohexene 1,2Dichlorobenzene Dimethylformamide Epichlorohydrin Ethyl acrylate Ethyl bromide Ethyl ether Isophorone Mesitylene Mesityl oxide Methylal Methyl t-butyl ether Methylene chloride Methyl formate Methyl methacrylate Perchloroethylene PGMEA Phenyl ether n-Propyl acetate 1,1,2,2- Tetrachloroethane 29 31 34 40 72 104 31 31 33 40 31 32 29 47 34 43 33 53 25 33 31 31 31 27 2120 35 34 58 35 39 32 57 0.71 0.31 0.80 0.95 0.90 1.02 0.98 1.07 0.13 0.93 0.96 0.98 1.00 0.80 0.02 0.38 0.95 0.94 1.02 0.68 1.01 0.80 0.91 1.21 0.96 0.29 1.01 0.86 0.90 0.92 0.98 0.69 0.74 0.36 0.82 1.00 1.01 1.01 0.98 1.06 0.34 0.94 0.98 1.02 1.01 0.84 0.05 0.82 0.96 0.98 1.10 0.73 1.00 0.80 0.94 1.24 0.97 0.90 0.99 0.96 0.95 0.86 0.94 0.70 97 0.71 102 0.30 114 0.83 132 0.96 239 0.99 347 1.01 105 0.98 103 1.03 108 0.45 133 0.96 104 1.01 105 1.05 97 0.99 157 0.87 113 0.14 142 0.45 111 0.93 175 0.94 85 0.96 111 0.75 104 1.05 103 0.89 103 0.97 91 1.11 212 0.89 117 0.29 112 0.98 195 1.00 116 0.96 131 0.90 107 1.00 190 0.92 WITCO REC 0.79 0.33 0.73 1.01 0.97 1.04 1.00 1.05 0.45 1.01 1.04 1.04 1.01 0.90 0.25 0.79 0.96 0.98 1.02 0.75 1.08 0.85 0.92 1.19 0.95 0.79 1.05 1.07 0.88 0.93 1.01 0.91 ?JFIDENTIAL MATERIAL SUBJECT PROTECTIVE ORDER 6 3M 009060 1992 Diffusion Honlt r Technical Report - January, 1993 TABLE 4 KURARAY GAAW VS. WITCO CAPACITIES January- July, 1992 Spikes In Milligrams CSg Elution Compound Spike WltCO Kuraray Acetone Acetonitrile Acrylonitrile Allyl alcohol n-Amyl acetate Amyl alcohol * Benzene Benzyl chloride B romoch1oromethane Bromoform n-Butyl acetate s-Butyl acetate t-Butyl acetate Butyl cellosolve * Butyl cellosolve acetate Carbon tetrachloride Cellosolve * Cellosolve acetate Chlorobenzene Chloroform Cyclohexanone Cyclohexene Cumene Decane Diacetone alcohol * 1,2-0ichlorobenzene Dimethyl formamide * Dodecane Epichlorohydrin * Ethyl acetate Ethyl acrylate Ethyl bromide Ethyl ether Furfural * Furfuryl alcohol * Heptane Hexane Isophorone Isopropyl alcohol Mesitylene Mesityl oxide Methylal Methyl t-butyl ether Methylene chloride Methyl ethyl ketone Methyl cellosolve * 28 4.0 28 0.1 28 1.9 28 4.1 28 30 28 27 26 16 28 26 30 7.9 29 30 26 26 26 26 26 25 27 27 28 30 40 32 28 27 29 31 28 27 30 12 26 21 28 17 26 28 26 27 28 29 26 25 30 27 26 27 30 39 27 15 28 24 29 2.4 28 7.7 29 28 28 30 27 28 26 20 28 26 27 4.8 26 27 26 24 26 5.0 27 13 33 3.6 28 12 29 24 7.7 1.0 5.1 6.1 29 26 22 26 18 31 26 27 25 27 30 41 27 31 28 21 22 21 27 27 28 25 28 27 46 20 27 6.8 12 27 29 28 24 26 9.2 27 26 10 19 7.0 18 26 7 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 1992 Diffusion Monitor Technical Report - January, 1993 TABLE 4 Cont'd Compound Methyl cellosolve acetate Methyl formate Methyl methacrylate Nonane Octane Pe rchloroethy1ene POME * Phenyl ether PGMEA n-Propyl acetate Styrene 1,1,2,2-Tetrachloroethane Tetrahydrofuran Trichloroethylene Toluene 1,1,1-Trichloroethane m-Xylene Spike 25 29 28 25 25 33 28 27 29 27 27 32 27 29 26 27 26 witco 25 0.2 25 26 26 35 26 26 30 24 26 33 8.4 28 25 20 27 Kuraray 26 0.5 28 27 25 34 27 27 30 25 27 31 15 29 26 22 27 `Desorbed with methylene chloride. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 8 3H 009062 luary, 1993 _ determined for a number nd charts. is) was introduced in June. to correct numerous errors ate new data. } SAMPLING RATES -UENT SAMPLING RATE (cc/min) H2CL2 H2CL2 32 S2 S2 S2 S2 27.9 +/- 1.3 # :S2 :s2 :H2CL2 3S2 31.5 +/- 1.1 36.7 +/- 2.5 Wltco 36.6 +/- 0.9 Kur. SAMP RATE ug (cc/min) 40.6 +/- 1.6 37.3 +/- 1-2 46.8 +/- 2.4 tnd Abscents with acetonitrile. Report - January, 1993 VALIDATION validate the sampling rate for duct d as follows: / = +/- 4.45*) RH in (CV = +/-6.8*) I rate = -3.2* n (CV = +/-2.5X) Kuraray GAAW iracy = 8.4* i are summarized in the following *8LE 6 .ORIDE VALIDATION IME EXP.SAMP.RATE NO.MON'S 38.9 +/- 3.8 6 39.3 +/- 1.7 12 ' 36.9 +/- 1.3 8 37.1 +/- 2.2 6 rate was 38.1 +/- 2.2 cc/min compared e calculated from the Hirschfelder ratory validated sampling rate listed uides is 37.9 +/- 0.3 cc/min. The X. The indicated bias in these 'A Accuracy of 12.1 X. The average ' +/-0.056. The average recovery for !8. heir polypropylene resin supplier 3. A sample of film from the new ne any possible effects on sampling posure studies were conducted. No rate was observed (see data below). LE 7 ) 078 mg 60 ug 68 ug Celgard D-209 1.810 +/- 0.053 mg 10.21 +/- 0.84 ug 10.59 +/- 0.49 ug 3M 009063 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 1992 Diffusion Monitor Technical Report - January, 1993 A sample of 3M isoporous film from John Pournoor made by laser drilling of a uniform pattern of holes In a PET film In hopes that a high density of uniform holes could allow us to achieve a higher sampling rate than is possible with the Celgard. The sample resulted in 1.193 +/-0.077 mg of toluene being collected which was 34% less than the Celgard. Unless 'a much higher density of holes is possible this material does not appear to present any advantages to the Celgard. The desire Is to find a film which can be mathematically characterized, altered and studied and which when placed adjacent to the charcoal wafer would allow an increased sampling rate monitor to be developed without adverse boundary layer effects. ETHYLENE OXIDE In response to a customer inquiry, we compared the recovery after our recommended 1/2 hour elution time with 1, 2, 3 and 4 hours of elution and found no improvement in recovery with longer elution times. FORMALDEHYDE Formaldehyde monitors spiked with 8 ug were analyzed and the recovery determined to be 0.94. Blank analyses performed on Lot 1303 (prepared with care by M. Chouanard In 11-91 and stored at room temperature) found 1.57 ug on 11-15-91 and 1.75 ug on 3-92. Wafers prepared from Gelman A/E glass fiber had lower blanks than the standard S&S paper when aged at room and elevated (130 F) temperatures but had much lower recoveries in exposure tests. Analysis of the surface pH at CRL showed that the glass fiber pH was >10 vs. near neutral for the S&S paper. Treatment of the glass fiber by soaking in IN HC1 before treatment with bisulfite improved the recovery but then showed much higher blanks on elevated temperature ageing. This led to a number of experiments which conclusively proved that the high and erratic blanks that we have experienced in the last several years is due to outgassing of the interior coating on the can. This coating has been changed once and possibly twice without our knowledge since the can was originally qualified. Canned monitors aged at 130 F for 5 days had blanks of 8-9 ug while monitors in glass jars or nylon bags remained under 2 ug. Canned acid treated glass fiber monitors also show elevated blanks when aged at 130 F but remain under 2 ug when packaged in glass jars or nylon bags. Canned monitors exposed to 130 F for 24 days Increased to 24 ug while monitors packaged in nylon bags before canning remained about 2 ug. These experiments have led us to conclude that the high and erratic blanks which we have historically experienced are coming from outgassing of the cans at elevated temperatures which may occur in storage or shipping. An MPDA was written and implemented in May to provide a temporary fix by placing the monitor in a nylon bag and then sealing 11 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 3M 009064 1992 Diffusion Monitor Technical Report - January, 1993 the bag In a can. A one square Inch bisulfite containing paper wafer Is being placed In the can as a scavenger until such time as a replacement for the can Is found. Once the can Is replaced, we can again pursue a lower blank by changing the wafer material. Formaldehyde monitors were exposed to 1 ppm formaldehyde in air which had been dried over silica gel desiccant. Monitors without the wet-pad showed only 30% of the expected amount. This confirmed the need for the wet second section. Formaldehyde monitors were validated at 0.2 ppm and at 0.7 ppm. The quality control exposure test (QTM-2094) has been revised to use the 0.7 ppm level to bring the exposure in line with the new PEL. The accuracy and precision can be achieved only when the calibration curve standards are prepared at a comparable level. The QC test method OTM-2095 has been revised to achieve this lower detection limit. In response to a further reduction In the OSHA standard from 1.0 ppm TWA to 0.75 ppm TWA, we have demonstrated the validity of the 3M monitor over a wide range of concentrations with an average bias of 4.7% and coefficient of variation of 5.3%. The OSHA accuracy equals 15.3% for 102 monitors exposed in batches of 6 to concentrations of 0.24, 0.40, 0.60, 0.61, 1.00, 1.13, 1.90, and 1.94 ppm. MERCURY Molding of the polystyrene substrates was moved to Plastech and Koral Labs reported much difficulty In coating them. The mold was repolished to remove blemishes and proper handling and packaging of the substrates by Plastech has improved the quality of the substrates although periodic difficulties are still being encountered. ANALYSIS LABORATORY A second GC was obtained and installed last year for 0VM analysis to handle overflow and provide backup for the HP5880. Calibration of the new instrument was completed and it is now being used routinely due to erratic response of the HP5880. The computer program was updated to perform the calculations for 3520/3530 OVMs with backup. The report fits on the existing report form. The 0VM analytical method was rewritten to include analysis of 3520/3530 monitors, analysis of standards with every set of samples, and to reflect the present capillary column parameters. The method Is being reproduced for use by technical service to answer customer requests. Several compounds were added to the analysis list and the Monitor Record Form was revised to include the new compounds as well as the 3530 Monitor Analysis. 3M 009065 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 1992 Diffusion Monitor T chnlcal Report - January, 1993 The most recent lot of CS2 which we received from Aldrich has a very high benzene content (49 ug/mL) which prevents us from getting acceptable benzene analyses. I have evaluated another grade of CS2 which has a <1 ug/mL spec but which Is 4 times the cost and have recommended use of the Aldrich 34,227-0 CS2 when benzene free CS2 Is 'needed. Tests were run In the pilot plant to determine the limit of quantitation for formaldehyde In an effort to reduce the minimum level which we report. We found that we are able to see 0.5 ug/3 mL sample and we are now reporting at this level rather than the 3.5 ug level used In the past. This will allow customers that submit blank monitors to receive an actual number on their report which they can subtract from the samples to achieve a lower detection limit. In the past almost all monitor blanks would be reported as "less than 3.5 ug". Our new packaging in nylon bags In cans should prevent blanks from reaching this level. Several sets of quality control samples were submitted to the analysis lab. The most recent set was also sent to 3M Colombia and 3M Venezuela at the request of Paul Olson. No reports have as yet been received from Colombia or Venezuela. I discussed the PAT program statistics with NIOSH. The true value and standard deviation are determined from the mean of the results of 70 labs which have historically had consistently excellent results. An outlier is defined as a result which is greater than +/-3s from the mean. Typical relative standard deviations for organic solvents in the PAT Program are +/- 4-6*. Three standard deviations would then give you a range of 85 to 115 * of the amount expected. I determined the average per cent recovery for the recent set of three spiked samples that I sent to the analysis lab as shown below. TABLE 8 COMPOUND AVE X. EXPECTED PP REC COEFF R&D REC COEFF March: Isopropanol Toluene p-Xylene 104 92 97 0.51 1.06 1.02 0.61-0.66 1.04 1.04 December: Toluene 1,1,1-TCE Isopropanol 100 108 115 1.06 1.04 0.51 .1.04 1.04 0.61-0.66 3H 009066 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 1992 Diffusion Monitor Technical Report - January, 1993 NEW MONITORS CARBON MONOXIDE Quantum CO Tests showed that significant exposure of the chip resulted from CO entering the monitor other than through the silica gel entrance slot. Sealing the slot and taping the outside of the monitor body reduced the sampling rate but did not eliminate the CO exposure. After a meeting with Mark Goldstein and Michelle Oum, Quantum Group, it was decided that no further evaluation would be done by us until they have overcome several sensor chip manufacturing problems and we have developed marketing Information. THERMAL DESORPTION - NEW SORBENTS I have demonstrated that the UOP Abscents silica molecular sieve material can be used to create a double decker monitor for alcohols. I have successfully desorbed methanol and ethanol from the wafer with acetonitrile solvent. I have determined the sampling rates and recoveries shown below (Table 9). I have also shown that it is possible to thermally desorb methanol from these wafers although much work remains to be done to make thermal desorption a quantitative method. A thermal desorber was obtained on loan from the CRL Industrial Hygiene lab for these Initial tests. Pursuing this approach will require investing in an instrument of our own. I have also shown that it is possible to desorb some organics from the charcoal wafer although the efficiency will need to be improved. I feel that many other compounds will be shown to work successfully with this silica molecular sieve material or similar materials and I think that it would significantly enhance our product line to have a monitor with these characteristics. A significant effort should be allotted for work in this area during the coming year. TABLE 9 Compound Recovery Exp.Sampling Rate Calc.SamplIng Rate MeOH EtOH 0.83 0.85 53.7 +/- 2.4 (+/-4.5X) 55.6 CC/min 46.8 +/- 2.4 (+/-5.1X) 44.7 cc/min In contrast, the EtOH recovery from Witco carbon with CS2 is 0.41 and MeOH doesn't work on carbon at all. MISCELLANEOUS EVALUATION OF DOW XPR-1531-D-01116-H3 FIBERS The Dow fibers were tested for MEK recovery, Carbon Tetrachloride capacity, and Carbon Tetrachloride activity. Results (described below) were compared to carbon-Teflon wafers used in our Organic Vapor Monitors. I do not see any particular advantage to using the Dow fibers 3M 009067 CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 1952 Diffusion Monitor Technical Report - January, 1993 in a monitor application at this time, however further testing with dynamic organic vapor challenges may show an application to respirators. Larry Brey suggested that a mixture of the fibers and chopped BMF be pressed Into a disc which could be challenged using our breakthrough 'test procedures. Test Data MEK Recovery: Approximately 160 mg of fibers was spiked with 97 ug of MEK in CS2. After sitting overnight the fibers were eluted with CS2 and the extract analyzed by GC. The ratio of the amount to the amount spiked is the recovery and was found to be 0.71 +/- 0.03 for 3 samples compared to a typical value for our carbon-Teflon wafers of 0.93. Carbon Tetrachloride Capacity: Approximately 160 mg of fibers was spiked with 32 mg of CC14, allowed to sit overnight, the vial opened and again allowed to stand overnight. The fibers were then with CS2 and the extract analyzed by GC. The amount is considered to be the capacity and was found to be 28.3 +/- 0.9 mg for 3 samples compared to 32 mg for our carbon-Teflon wafers. Carbon Tetrachloride Activity: Approximately 300 mg of fibers was placed in an aluminum pan in a desiccator containing a saturated CC14 atmosphere. The weight gain was determined periodically (30, 60, 135, & 195 minutes). The maximum weight gain for a carbon-Teflon wafer was achieved In 30 minutes (minimum time measured). Previous experiments have shown that the maximum weight is gained in less than 10 minutes. The maximum weight for the Dow fibers was achieved at 135 minutes although 89% was gained in the first 30 minutes. The weight gain divided by the amount of carbon or fiber times 100 is the CTA. The CTA for the carbon in the carbon-Teflon wafer was 74 while that for the fibers was 121. MANUFACTURING MOVE TO PLASTECH I participated in the evaluation of a proposal from Plastech, Corn., Rush City, MN to manufacture the monitors except for mercury. The proposal includes making the wafers, assembly and packaging the monitors and performing the quality control tests. MOLD REDESIGN I worked with Paul Flannigan to resolve a number of problems with molding parts by redesigning the molds for the elution cap, main body and primary cup. The design changes are complete and fabrication is in process with an anticipated introduction in April, 1993. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 15 3M 009068 5 1992 Diffusion Monitor Technical Report - January, 1993 PACKAGING The lining of the can causes high formaldehyde blanks and the cans are becoming more difficult t get. The size used f r formaldehyde Is no longer available. A number of foil laminates and polymer films were *screened for applicability (Table 10). The screening tests (Tables 11 and 12) Included: OVM blank; OVMs exposed to saturated vapors of benzene, methylene chloride, MEK mixture, and carbon tetrachloride; 555 ppm ETO; saturated formaldehyde vapor; formaldehyde blank. Our first choice, a Prlntpack nylon/foil extrusion laminate, has an unacceptable lead time of 58 weeks. Our next choice. Graphic Packaging Vitek, is now undergoing extensive aging tests. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 3M 009069 1992 Diffusion Monitor Technical Report - January, 1993 TABLE 10 ALTERNATIVE PACKAGING MATERIALS PRINTPACK FOIL - Extrusion Laminated 60 ga. blax Nylon/prlmer/7# PE/0.0003 Foll/7# PE/1 mil 5X EVA e PRINTPACK METALLIZED POLYESTER - Machine debug for the 5000 respirator 48 ga. met. PET/2 mil PE TORAYFAN METALLIZED POLYPROPYLENE KAPAK SILVER FOIL - CADPAK N 60 ga. Nylon/15# PE/0.00033 Foil/0.0022 LDPE KAPAK white FOIL - With adhesive strip LPS A - VF-52 PET/PE/Foll/PE to ACLAR/PE/Foll/PE LPS B - VF-52 PET/PE/Foil/PE to VF-62 Nylon/Saran/PE TEDLAR - Tursso Group - Polyvlnylfluoride VITEK - Graphic Packaging 92 ga. PET/7.2# Surlyn/0.001 Foil/14.4# Surlyn SURLYN - 3M Medical Products 1 mil biaxial PP/14.3# LDPE/0.00035 Foll/26# Surlyn #1652 WHITE - Graphic Packaging 48 ga. PET/10# White LDPE/0.00035 Foll/21.6# EAA CASCADE - Graphic Packaging 48 ga. PET/Ink/10# White LDPE/0.00035 Foil/7.5 EAA/2-1/2 LLDPE LAMC0R - Cast Nylon/PE/Foil/PE OMEGA - Graphic Packaging 48 ga. PET/14# LDPE/0.00032 Foil/adheslve/2 mil LLDPE CAN NYLON BAG ELUTION CAP CLOSURE CAP - No ports PRIMARY CUP CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 17 3M 009070 > 1952 Dlffusl n Monitor Technical Report - January, 1993 PKG OVM BLANK CAN + NYLON + ELUTION CAP + CLOSURE CAP + PRINTPACK FOIL + PRINTPACK METALLIZED POLYESTER +/- TORAYFAN METALLIZED POLY PROPYLENE + KAPAK SILVER +/FOIL KAPAK WHITE FOIL LPS A LPS B +/+ TEDLAR +/- TABLE 11 ALTERNATIVE PACKAGING TESTS JUNE-JULY, 1992 OVM OVM BENZENE CH2CL2 OVM MEK OVM CCL4 FORMALDE MIX HYDE +/-,+/+ ,+/- + - + - - --- +/- + + + +/- + * + ,+ - +/- + ,+ +/- +/- - + -- - +/- +/- +/- -- + = No permeation or background observed. +/- = Small amount or variable results observed. - = Large amount observed. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 3M 009071 1992 Dlffusl n Monitor Technical Report - January, 1993 PKG CAN NYLON ELUTION CAP PRIMARY CUP PRINTPACK VITEK SURLYN WHITE CASCADE LAMCOR OMEGA OVM BLANK + + + + + + -- TABLE 12 ALTERNATIVE PACKAGING TESTS NOV-OEC, 1992 OVM OVM BENZENE CH2CL2 ET0 +/- + + + ,+ +/+ +/- + ,+ +/+++ - +/-,+/- + -- FORM VAPOR + + + + +/+ + +/- FORM BLANK* 1.31 1.50 1.92 1.49 1.94 1.65 1.29 + = No permeation or background observed. +/- = Small amount or variable results observed. - = Large amount observed. * = Blank after 15 days at 100 F. Formaldehyde monitors, after 5 days at 100 F, gave >90% recovery for all tested packages - Cascade, Vitek, White, Surlyn, Prlntpack, Can, Can (Room Temp), and Lamcor. Formaldehyde monitors, after 5 days at 100 F, showed no difference In blanks between Vitek bags and a nylon bag Inside the Vitek bags. CONFIDENTIAL MATERIAL SUBJECT TO PROTECTIVE ORDER 19 3M 009072