Document xzgVEoy27Zdmvj4yezNOENxyE

fUC& and C/u*MCat- PUT IT IN WRITING To. J. D. Kramer Date 8 November 1977 Subject ACETYLENTCS PLANT INDUSTRIAL From R. J, Lomi ckv HYGIENE MONITORING PROGRAM cc.: J. A. Allbrittcn 0. Culver-. R. IV. McGinnis J. C. Novak/Trexlertown H. Stermon_______ H. L. Watsoii/Trexlertown As per the recent request to institute immediately a total Plant Industrial Hygiene Monitoring Program, the following is the proposed program to be followed at the Acetylenics production unit. The following program will list suggested locations to measure various utilized chemicals, the frequency and timing of sampling, and suggested forms to record information. The Acetylenics Plant will also institute its own Leak Detection Program to monitor organic concentrations around the various process equipment and tankage. The use of a Century Vapor Analyzer, borrowed from PVOH, will be. utilized for this program. The total Leak Detection Program will- be better defined at a later date. H.'S?.` ` *% It is suggested that we initially use a hand-held portable gas 4. ;- sampling detector pump utilized with Gastec netector Tubes to got*'. base line data on the more commonly used plant chemicals such acetone, MEK, IPE, acetylene, etc. The air sampling pump and stain tubes can detect the threshold limit values of various chemicals*in the plant operating areas to permit more, definitive air sampl ing:,to. begin utilizing the carbon absorbing sampling pumps. Shown on the attached, sketches and data forms are the various suggested locations to obtain initial air test samples using stain tubes., It is requested that we immediately begin weekly air testing using the e stain detector tubes to provide initial base line data Foil owing/ initial stain tube results, weekly sampling, utilizing the carbon:''',. absorption air detectors, can be instituted to measure I sopropyl. Ether and various carbonyls as various products are manufactured.' ' r Continued sampling in the KOH operation area should be completed* twice weekly using the Bendix pumps with midget impingers of 1.8v y liters/minute to measure for total*caustic dust. See attached * measurement data sheet for recommended sampling points in the causti> building. Immediate implementation of dust measurements at the product flaker operation should.also be instituted. Daily sampling during flaker operation should be utilized With the Bendix pumps at 2 liters/ minute to measure total dust. The Acetylenics Plant will advise the hygiene monitoring personnel of the days and times when the flaker is in operation during DO and DH flaking. AP00043671 Mr. J. 1). Kramer 8 November 1977 Pago 2 SUBJECT: ACETYLKNJCS PLANT INDUSTRIAL HYGIENE MONITORING PROGRAM A review of recent Acctylcnics Plant Industrial Hygiene Program monitoring reveals that only K0I1 dust, samples around the caustic building flakcr have been measured and obtained. It is imperative that. DO and Dll product flakcr dust, measurements and organic concentrations in the operating areas be monitored immediately. The Acctylcnics Plant supervisors will work closely with Roy McGinnis and Buddy Stermon to assure collection of the required samples at the recommended weekly frequency. Once the data is accumulated and analysed, it will he utilized as part of the monthly Acctylcnics Plant report. Shown attached are five [5) sketches detailing the plant tank farm areas, outside process area, and inside process area. On each sketch suggested sampling locations for air monitoring of organic chemicals have been selected. Also attached arc data sheets detailing other suggested sampling locations for both organic -,-=y chemicals and clust measurements. Results obtained from the air ^ f sampling can be recorded directly on the attached data sheets for;. the various sampling locations. T-" ^ Also included is the list of commonly utilized organic chemicals:- . V.T at the Acctylcnics Plant, along with established T.L.V.'s. Where/, applicable, stain tube measurements should be initially obtained." f to determine where exposure problems exist. Then carbon absorption- samples can be obtained on a weekly sampling frequency. '* < Jf there are any comments or questions, please advise. Everyone'S cooperation is appreciated. RJL/hc y\ <\ t /it/ R.'-d. l.OMJ CKY /--\ \ 'f jjht-. AP00043672 ACETYLIiNICS PLANT DATE: Zg hJcoll TIME: - fCCC DETECTOR INSTRUMENT ^Aa***-^*^ (Geriorn 1 WEATHER CONDITIONS Description) . u)u*4 ATG fT-n+'t" hstabLTsunn t.l.v. CHEMICAL MEASURED C=C ffe^ LOCATION nTANK FARM AREA TANK FARM AREA H2 TANK ]:ARM AREA if 3 TANK FARM AREA #4 BALL MILL BLDG. WASH WATER Q PHASE SEPARATOR AREA OUTSIDE PROCESS AREA in UTSJPE.PROCESS AREA II 2 INSIDE PROCESS AREA If 1 INSIDE PROCESS AREA fi2 INSIDE PROCESS AREA H3 INSIDE PROCESS AREA `M SECOND FLOOR PRODUCT STILL THIRD FLOOR PRODUCT STILL TUBE READING CONCENTRATION W (PPM) A/iru*-* tbitZ&J TEMP. C 3 TEMP.CORRECTED CONCENTRATION ;(*) (PPM) 3*toc 4t6C ' /&> 3 -t/C 3 REMARKS */oo 3 < IOC 3 <*tC6 fCC -c toe e t S' -- lee t j*/oa | *-tce \ w- top iii ^tco - ~toO a * St * ;> ... . ' -- COMMENTS: - -X.' . ^ 43ir AP00043673 ACETYLEN1CS PLANT ORGANIC MEASUREMENTS (General WEATHER CONDITIONS Description) DATE: ______________________ ________________________________ _ TIME: DETECTOR INSTRUMENT Air Rump Sainpl cr/Cnrbon OPERATOR NAME:. CHEMICAL MEASURED LOCATION T1 MI- TIME START END TOTAL TEST TIME RESULTS REMARK'S TANK 1-ARM AREA ft 3 TANK FARM AREA If 2 TANK EARM AREA it 3 TANK EARM AREA #4 BALI, MILL BUILDING WASH WATER f, PHASE SEPARATOR AREA OUTSIDE PROCESS AREA 1* 1 OUTSIDE PROCESS AREA 2 . ;; INSIDE PROCESS AREA H INSIDE PROCESS AREA 2 INSIDE PROCESS AREA fi 3 INSIDE PROCESS AREA If4 SECOND FLOOR PRODUCT STILL TUT HI) FLOOR PRODUCT STILL COMMENTS: AP00043674 ACETYLENICS PLANT DUST MEASUREMENTS (General WEATHER CONDITIONS Description) DATE; TIME: _____________________________ __ DETECTOR INSTRUMENT Air Pump Sampler/Dust OPERATOR NAME: LOCATION FIRST FLOOR CAUSTTC BLDG. AT DELTAINER CHUTE SECOND FLOOR CAUSTIC BLDG. AT BALL MILL SECOND FLOOR CAUSTIC BLDG. AT EAST POT SECOND FLOOR CAUSTIC BLDG. AT WEST POT FIRST'FLOOR AT BALL MTLL GATE SECOND FLOOR BALL MILL ROOM PRODUCT FLAKER ROOM AT DRUM PRODUCT FLAKER ROOM AT NORTH END CHEMICAL MEASURED TIME TIME START END TOTAL TIME TEST RESULTS REMARKS r COMMENTS: I* AP00043675 AP00043677 AP00043678 AP00043679 TABLE I ACETYLENICS PLANT PROCESS CHEMICALS CODE 1304-101 - 102 CHEMICAL Acetylene C* Acetone t< - 104 Butraldchvdc ^ - 107 Isopropyl Ether r -108 Ethyl Hexaldchydc ** -109 - Ill -112 -113 2-'Ethyl Hexanol Igepal AT-548 KOH - 90% Methyl Ethyl Ketonex -114 Methyl Isobutyl Ketone -116 -117 -118 -132 -183 - PolyGlycol 2000 Propargyl, Alcohol Renex 20 Ethylene Glycol Propylene Glycol DiMethy 1 I'ormamidc ESTABLISHED SAMPLING T.L.V. FREQUENCY STAIN TUBE AVAIL. Asph ix. 1000 Not list. 2S0 Not Est. Not Est. Not Est. 2 200 100 Not Est. 1 Not Es t. 100 300 10 Week]y During nil/ MB Run During OW-1 Run Weekly During EO Run S-104 Blend S-104 Blend Weekly During MP/ DO Run Duri ng S- 1 04 Run S-104 Blend EO Blending S-104 Blend S-104 Blend S-104 Blend Week 1y Yes Yes No No No No No NO Yes Yes No. No No No . No Yfcs AP0004368I A technique has bean developed to validate gas and vapor sampling methods under field insult conditions. Laboratory validation, a necessary aspect of good laboratory practices, only tests the feasibility of a sampling method. "Field validation" determines the effectiveness of a given sampling method in the actual workplace environment where potentially interfering components may exist. A simple and inexpensive means is shown for introducing a known addition spike onto the sampling device during sampling. The validation sample is subject to the same field conditions at area or personnel samples. Upon analysis, recovery of the known addition spike within specified limits, verifies validity ofthe day's samples. Insufficient known addition spike recovery alerts the hygienist that something has caused sample failure, resulting In an invalid measurement of worker exposure. Validation of gas and vapor sampling methods under field insult conditions LINDA M. CHAPMAN, BRIAN G. WARD and PAUL M. JEANNOT Monsanto Company. Department of Medicine and Environmental Health, Industrial Hygiene Section, 800 N. Lindbergh, St. Louis, MO 63166 Introduction Currently promulgated regulations are forcing industry to test the validity of sampling and analytical techniques used in industrial hygiene monitoring. Section E6 of the Emergency Temporary Standard for acrylonitrile, and the OSHA field directive no. 300, made evident that methods must be validated in order to be in compliance. Prior to 1979, NIOSH published procedures0' used laboratory generated data for a "one component" analyte system to validate their methods. Plant environments may well have multiple gas and vapor components which must be considered. These components may vary qualitatively and quantitatively from day to day, and indeed from moment to moment. Depending on adjacent processes and wind direction, one could have multiple sampling interferences one day and none the next day. These two sampling situations are quite different from each other and obviously much different from earlier NIOSH laboratory generated sampling protocols. Temperature and humidity are other factors that have been shown to adversely affect the performance of sampling devices. Field validation reveals the existence of these problems by testing the reference working curve on a daily basis. Field validation will not reveal specific causes ofsampling method .failure, but will indicate when failure occurs. By validating an area sample in the same general area that personnel samples are taken, one may assume all samples in that area can be considered valid. To implement this program, one needs a device to introduce a vapor spike onto the sample tube and an adept person to spike the tube. We believe that this procedure far exceeds OSHA monitoring method requirements. If worker exposure is to be determined, both sampling and analytical validity must be ensured. This technique begins to tie together the integ rity of both the sampling and the analytical procedures. Chain of custody from the field through the laboratory process can be assured using field validation. validation equipment To implement field validation one needs a means to introduce a vapor spike. A glass gas bulb fitted with a suitable septum works well. We recommend that the internal volume of the gas bulb and the flow rate allow at least 10 volume changes per hour. To calculate the maximum internal volume necessary for a long duration sample, i.e. greater than 4 hours, do the following: volume ofbulb(cc) = flow rate used in sampling (cc/roin,) x 60 min./hr. 10 air volume ehanges/hr. Smaller internal volumes may be used affording increased transfer rates. This becomes necessary when validating sampling procedures for short term exposure level (STEL) monitoring. All other validation equipment is identical to that required in the sampling method. procedure developing a calibration curve Standard solutions of analyte in the range equivalent to expected levels in the workplace air must first be prepared. Recommended levels are 0.2x, 0.5x, lx, 2x, and Sx the permissible exposure level (PEL). The NIOSH solvent flush technique121 is strongly recommended for all syringe injections. We also recommend a minimum of 6 injections per concentration in order to obtain good statistical data. These individual concentration responses are then averaged. Average instrument response vs. pg in the standard 2 standard deviations ( 2a) data are determined at each point. Figure 1 illustrates a calibration curve which is used in conjunction with the reference working curve used for field validation. A daily test of the calibration curve should fall within 2o at any point. Failure to do so requires regeneration of the curve. Copyright I ISO. Amoaican Induttnol Hygiyno Allocation asa Am. tmt. Hrt Assoc J (41) September, 1980 AP00043682 IB 42 RECOVERED Figure 1 -- Calibration curve for Benzene. developing a reference working curve This curve is generated by vapor spiking a sampling device using the gas sampling bulb and then analyzing to determine tig recovered.*1* In developing the working curve, it is important to set the air flow at the level that actual sampling is performed (10%) because the working curve can varyas a function of flow rate. Repetitive runs are made to enable means and standard deviations to be calculated. The analytical "pg recovered" Is obtained from the calibration curve, Figure l.The "tigactuaf is the known addition spike. These values are plotted to generate a reference working curve in Figure 2, not desorption efficiency as advocated by NIOSH. NIOSH protocol states that a method fails when the desorption efficiency is less than 75%. The working curve allows an extension of the linearity at lower limits and an improved coefficient of variation at the 95% confidence level (CVt) with no compromise of data quality. For example, the data for acrylonitrile sampling generated by OSHA*4* using a modified NIOSH procedure'51 utilizing desorption efficiency was replotted on a reference working curve. By using the reference working curve approach, the method showed a linear response to exposures of an order of magnitude lower than the approach used by OSHA, while exceeding the actual performance of the method required by the OSHA acrylonitrile standard.*61 field validation Assemble 2 sample trains as in Figure 3. Calibrate the flow with a loosely packed charcoal tube between the soap film flow meter and the sampling device. It is extremely important to keep both sample train inlets as close to one another as possible when sampling. After 25% of the total sampling time has elapsed, the bulb is spiked with an aliquot of standard. It is essential to wait before spiking, because the sample tube should be exposed to the environmental matrices before the validation test begins. In practical application field validation is a means of determining whether the sampling and analytical integrity have been maintained. Therefore the sampling device must be exposed to the environment before the known addition spike is added. This should reveal whether the analyte remains adsorbed or whether it is lost due to matrix conditions. The spike recovered should be 100% 25% of theoretical. If the validation is not within these limits, then something in the environment interfered with the method causing it to fail. field data Table I contains field validation data obtained while sampling for acrylonitrile.*7* The data shown are averages of Amarlcan Industrial Hygiuta Atsoeiatian JOURNAL (41) 9/80 C3I AP00043683 I0-- Uj 3 z nt several sampling and analytical determinations. All samples were obtained in the same operating unit utilizing an identical sampling and analytical procedure. Using the 25% criteria. Held samples taken during field validation trial 2 are considered valid. The reason for failure in trials one and three is not known; but the unit in which the samples were taken is subject to local high humidity conditions on a random basis. The industrial hygienist was able to better interpret the exposure data because of the use of field validation. A new sampling strategy to avoid high humidity operations was adopted based on this data. discussion limitations of this techniqus Since the field validation procedures require the use of a glass bulb, it is not recommended for actual personnel samples. This limits the use ofthis procedureto area samples only. However, if several people work in a non-transient area, a validated area sample under the day's worst environmental conditions would in principle validate all samples taken in that area. We recommend field validation be done on a daily basis because random day-to-day environmental changes have been observed as the cause of (32 method failure as illustrated in the field data shown in Table I. Field validation will apply only when the analyte is completely vaporized. Heating the glass bulb may become necessary to obtain complete vaporization and mass transfer of some compounds. Maleic anhydride is an example of such a compound. Once calibration and reference working curves have been developed, they become good indices of method problems. GAS BULB AP00043684 Sample no. 1 Ambient Environ. Condition# Jan., 1978 Tamp. 21 C no. 2 July. 1978 Temp. 30 *>C no. 3* July, 1978 Temp. 30 C TABLE t Field Validation Data Background Control* Field Spike Background plue Spike* 12.0 HQ 8.0 hO 22.5 hO 44.0 hq 16.0 hq 62.1 ho 10.0 |ig 8.0 HO 14.8 HO 6p!ke Found* 10.5 hO 18.1 HO 4-1 HO Spike Recovered (vg.| 131% (invalid) 113% (valid) 51% (invalid) All values are "ho actual" obtained from the reference working curve. `This sample had a steam unit operating in the near vicinity. All samples are 20L volume with charcoal lot no. 106 obtained from SKC, Inc.. R.D. 1. no. 395 Valley View Road, Eighty Four, PA 15330. For example, problems with the sampling device are detected through the reference working curve and problems with the analytical device are detected through the calibration curve. Field validation is not the answer to all the sampling problems we encounter Field validation will not reveal the specific causes of method failure, but it will indicate when failure occurs. This becomes very important ifexposure is to be accurately determined and evaluated. Monitoring data without appropriate validation can be dangerously mis leading. "False negatives" may cause an area of concern to be overlooked, while "false positives" may cause unneces sary and expensive corrective procedures to be implemented. Performing field validation of vapor and gas samples avoids such errors and assures the quality a defensibility of monitoring data. references 1. U.8. Dept, of Health, Education and Welfare: MOL Manual of Analytical Methods. 2nd Ed. Vol 1, 2. 3 (197' 2. U.S. Dept, of Health, Education, and Welfare: MOL Manual ofAnalytical Methods. 2nd Ed. Vol. 2, S19, (197' 3. Deifrich, M.W., L.M. Chapman and J.P. Mieure; Sar pling for Organic Chemicals In Workplace Atmospher with Porous Polymer Beads. Am. Ind. Hyg. Assoc. J. 33:38 392(1978). 4. Exhibit ISA, OSHA Acrylonitrile Hearing, March, 1976. 5. Federal Register. 45:45818-45819. October 3, 1976. 6. Federal Register. 43:45810. October 3. 1978. 7. Sheridan. D.L.: Monsanto Intercompany eommunlqi between D.L. Sheridan and P.M. Jeannot. (1979). Joint Occupational Health Conference The Joint Occupational Health Conference, sponsored by the American Academy of Occupational Medicine end th American Academy of Industrial Hygiene, will be held at the Hyatt Regency Hotel in San Francisco, CA, October 28 31, 1980. This year, for the first time, the Society of Toxicology is also a participant. For a copy of the program anregistration forms write or call the American Academy of Industrial Hygiene. 475 Wolf Ledges Parkway, Akron, 01 44311 (216) 762-7707. American Industrial Hmene Association JOURNAL (41) S/80 63 AP00043685