Document 3eeD7bVQ5mMkeg83RM9o86D0O

"p-" irrrsn^A ", o. n.oX PT61, SOUTH CHARLESTON, WEST VIRGINIA ?r''.'03 Copy To: Mr. 0. L. Garrison Mr. R. C. Lawrence Mr. W. R. Manning Mr. R. V. Sea ley Mr. R. N. Wheeler -- Mr. G. L. Baker Mr. R. E. Reiter FILE DO f O o** April 20, 1976 onr.m,T^o rvPt. Engineering Approval of technical manuscript for presentation at American Industrial Hygiene Conference OVE Gentlemen: Attached is a copy of the manuscript titled "Automatic Systems For Monitoring Vinyl Chloride in Working Atmospheres" authored by Messrs. G. LBaker and R. E. Reiter. This final draft of the manuscript is being forwarded to you for approval for submission to the program committee of the 1976 Industrial Hygiene Conference. If approved, the paper will be presented by Mr. Baker at the 1976 Industrial Hygiene Conference on May 21, 1976 at the Marriott Hotel in Atlanta, Georgia. The American Industrial Hygiene Association Journal also reserves first-call publication rights on all papers presented at this conference. The manuscript must be submitted prior to the start of the conference, which is May 16th. Therefore, a note of approval sent to me in Building 740, Room 1318 by May 7, 1976 would be greatly appreciated. , If you find any aspect of the manuscript questionable or objectionable, please contact me at extension 5333 or Mr. Baker at extension 4482. Thank you for assisting us in maintaining Union Carbide's position as . , a Reader, in "state-of-the-art" technology throjjghpa.tiour.yl.odustry. ,? 5 j j , JHB:ksv ` Attachment , , ^ rifc- >, J. H. Brubaker t~-'- Measurement and Control Technology CHEMICALS AND PLASTICS RECEIVED Apr 22 1976 R. N. WHEELER, JR. % ROUGH DRAFT AUTOMATIC SYSTEMS FOR MONITORING VINYL CHLORIDE IN WORKING ATMOSPHERES G. L. BAKER R. E. REITER ucc 023402 -1- ABSTRACT The drastic reduction in the allowable exposure of workers to vinyl chloride monomer (VCM) brought about by the Occupational Safety and Health Administration led Union Carbide Corporation (UCC) to develop automatic monitoring systems for use at production facilities which process VCM. This paper describes what these systems consist of, how they function, and how well they perform. The present monitoring systems evolved during a project lasting for approximately one year. During the first phase of the project, the analyzers were Installed. Each analyzer Is a chromatograph equipped with a strip chart recorder, a 20 point stream selector, and a sampling system which includes automatic introduction of a calibration sample on one stream. The measuring range Is 0-25 ppm VCM. During the second phase of the project, data acquisition and processing systems were developed and installed. These systems compute and print out raw data and shift, daily, and monthly averages for each sample point, and the averages for the entire unit. They also compute an estimated maximum employee exposure to VCM, based on the average amount of time each employee spends near each sample point. The systems are performing very well and are now considred essential operating tools at several UCC production units. UCC 023403 -2- IMTftOD'JCTION Early in 197** evidence was found indicating that prolonged exposure to high-concentrations of vinyl chloride monomer (VCM) may be harmful to human health. The Occupational Safety and Health Administration immediately wh'ch reduced the allowable 8-hour average exposure of workers from 500 ppm to 50 ppm VCM. Several months later, the,a!lowable 8-hour exposure was reduced further to 1 ppnC(2) . At Union Carbide Corporation (UCC), a high priority program was initiated to obtain automatic monitors which would provide personnel protection and also aid operating personnel in determining sources of VCM emissions at the v`nyl resins manufacturing facilities. The UCC Special Instrumentation Department (SID) conducted a survey of instrumentation available to monitor VCM, and prepared a recommendation with a cost estimate. This was completed and approved by the business manager within ten days. Within another ten days, the first monitor was installed and operating. Four more were delivered to preselected plant sites within five weeks after the project began. Even before the monitors were selected, it was felt that an automatic method of tabulating data would be needed eventually. Soon after the first monitor was Installed, this anticipated need was confirmed. Late In 197^, the decision was made to obtain data acquisition and printing capability for all the monitors. This final phase of the project was completed before the end of 1975. UCC 023404 -3- Selection of Monitoring Systems The first step in obtaining automatic monitoring systems was to investigate the available methods for measuring VCM at low ppm concen trations. The Initial measuring range desired was 0-100 ppm, but it was anticipated that eventually a 0-10 ppm range might be required (the range fs now 0-25 ppm). This was an important consideration in the selection of Instruments. Other important factors considered were reliability, specificity, ease of maintenance, delivery times and cost. It was fortunate that high sensitivity was considered Initially to be one of the important requirements, because now the users desire both accuracy and reproducibility to be within i1 0.1 ppm. The investigation of methods which could meet the measuring requirements ouickly narrowed to two: infrared spectrometry and gas chromatography. Total hydrocarbons analyzers (THA's) were eliminated from serious consideration because many different organic compounds were known to be present in the vrny' resins operating areas, and a THA could not distinguish between VCM and most of these other compounds. Other types of analyzers, such as those which convert organic chlorides to hydrogen chloride and measure conductivity of an aqueous solution containing HC1 were also eliminated because of lack of spec!ficlty. Finally, Infrared spectrometry was eliminated. Low cost infrared analyzers ($3,000-$10,000) lacked the desired sensitivity and specificity. More sophisticated systems such as the E0C0M Model FMS-7200 were acceptable except for the high cost (>$100,000). Gas chromatographs were also acceptable, and cost only about $5,000 to $10,000. All these costs are for only the analyzers; stream selector, data acquisition, and installation costs are additional. UCC 023405 -4- A1though the EOCOM Model EMS-7200 was not selected for the VCM measurement, ft would be seriously considered for a problem where several pollutants were to be monitored at the same location. This system is capable of measuring several compounds at a time, even at ppb concentrations, while a chromatograph could measure only one or two with adequate resolution, therefore, it would become more cost effective where several compounds were being measured in a single area being monitored. Chromatographs manufactured by various reliable instrument manufacturers were considered for this project. However, a small, special purpose (Model 1200) chromatograph manufactured by Union Carbide for its own use was available 'or immediate delivery, and at lower cost than the commercial units. In addition, several of these were already In use at the UCC plants where vinyl resins are manufactured, so parts were stocked and instrument technicians were trained to service them. Therefore, the Model 1200 was chosen for the VCM project. The auxiliary equipment Included with each chromatograph was a 20 point e'ectromechanical stream selector, and a strip chart recorder with an adjustable high level alarm, ail mounted on a 24 by 52 inch panel. A stream switching station, Including a sample pump and stream switching valves, was also pro vided with each system. After the first system was installed and performing satisfactorily, it was found that manual tabulation and manipulation of the data from 720 analyses per day from each analyzer consumed too much time of the operating personnel. Therefore, the decision was made to obtain data acquisition and reduction systems. To provide a basis for the selection of data systems, operations personne UCC 023406 -5- representing each of the vinyl resins units described the type of data they needed. Then a survey was begun by SID to select a reasonably priced data acquisition system which would meet the requirements. A small Inexpensive microcomputer based on the Intel MCS-8 system had just been developed by Union Carbide. It was selected when the survey revealed that it was the least expensive system available at that time which met the data requirements and could be easily programmed by SID personnel. {Commercial systems can now be obtained for about $10,000-520,000, without orogramming). Operations and SID personnel collaborated to define the data re quirements more explicitly. This process was very tedious but important. The nature of a data processor is such that once the basis for a program has been established, it is very difficult to change certain aspects without going back and starting over. In spite of the fact that considerable effort was made by UCC to Identify all the requirements before beginning programming, an Important requirement was overlooked initially, and major reprogramnv ng was requried to overcome the deficiency. The point overlooked was that data pertaining to personnel working on a five day per week daytime schedule cou'd not be obtained from the data logged according to the regular four rotating shifts. Therefore, the systems had to be reprogrammed to allow for a ''fifth shift". This oversight cost several thousand dollars in programming time to correct. Soon after the data systems were selected for the first analyzers, five additional vinyl chloride monitoring systems, complete with data systems, were ordered. Figures 1 and 2 show one of the latest systems, including data acquisition. These new monitoring systems are functionally the same as the ucc 023407 or'g'nal systems except that the stream selection logic is provided bv microcomputer rather than hy the electromechanical stream seiector. Cent?'ai" System Oescriotlon 'n "enera1, the VCM monitoring system's operation is as oilows: Samo'e *rom each of ? sample points located throughout the product'on un't `s brouoht to the chromatograph by a vacuum pump connected to tNe s,-mp'e ''nes via a manifold and solenoid valves. Sample selection is brought about by the microcomputer seauentially energizing the solenoid valves, one every two nvnutes. T1' samp'e flows through an air cooler to drop out moisture, and tsen a 0.5 cc aliquot is 'ejected *nto the chromatograph. As soon as tse sample Ts 'njected, the microcomputer starts the next stream purging, to ensure tsat a rrosh samole is avaPable. The VCM concentration obtained by the chromatograph ;s immediately printed out on a locally mounted str'p-chart recorder. *t *s a'so stored by the m; crocomouter for printout at the end of t'^e ano'vsns o' a'' n;neteen samp'e points and check sample, and for calculat'c" of venous r\-i-n summaries. The data svmmar'cs Include the average VCM concentration anr standard dev!at!on for each sample point and 'or the entire un't, t,,n tot'1' o' the number of t:mes the VCM concentration exceeded a selected 'eve!, a-d a-' estimate of the average VCM concentratIon to which each employee was erc','c. The microcomputer can also activate alarms when: (!) the VCM concentret:o" at any point exceeds a preselected 'evel,(2) the check sample ana'ys:s 's out o' range, or, (3) the samole flow ?s insufficient.' A detailed descr'pt'or o' system's components follows. Sampiing System The goa1 was to monitor as much of a production unit as possib'e w`th each analyzer, but without an excessively long time between analyses of di''eren? areas. It was decided that there should normally be no more tsan ucc 023408 -7- ^0 minutes between checks of each individual area. The chromatograph requires two minutes per analysis, so each analyzer could monitor 20 sample points. Since one sample per cycle was check gas, this meant that 19 areas could be monitored. Production unit personnel determined the areas with the ] greatest potential for VCM leaks and it was at these locations that the s sampling stations were installed. Each sampling station (Figure 3) consists of a cartridge filter protected with an inverted 6 inch funnel as a rainguard, and each is connected to the monitor via 1/4" Type 304 stainless steel tubine. The tubing from each sample point was connected to a stream switching station, controlled by the microcomputer, which sequentially delivers sample from the 19 individual sample points, and check sample, to the analyzer. As soon as a sample is Injected into the chromatograph, the next sample is started purging through the entire system. In applications where there are lines that are too long to be satisfactorily purged within the two minute cycle time (i.e., over 200 feet), a dual pump is used. One side continuously pulls sample from all of the longer sample lines. The other side of the pump delivers the sample the remaining short distance to the analyzer. It also delivers sample directly from the shorter sample lines. Whether a single or dual pump is used, the amount of sample necessary to thoroughly flush the line is in excess of what is required by the analyzer. This excess is vented outdoors. Stream switching is accomplished through the use of 3*way solenoid valves. In the de-energized state, each solenoid valve connects a sample line to a manifold connected to the "purge" side of the pump. In the energized state, the sample line is connected to the side of the pump which delivers the sample to the chromatograph. The microcomputer allows an operator to select one of three different modes of stream switching: ucc 023409 (1) Automat *c progression from Sample Point No. 1 through Sample Point No. 20. (2) Repeated analysis of any one desired sample point. (3) Automatic progression through all sample points while alternately selecting a single sample point, e.g., a sample progression of 1, b, 2, 4, 3, 4, etc. The 'atter mode extends the flexibility of the system because it allows a troub'e soot to be monitored more frequently and the effects of corrective act:on observed without neglecting the rest of the production unit. However, it does lengthen the tTme required to sample all the points. Chromatograph The Union Carbide Model 1200 chromatograph is a compact process analyzer eguipped with solid state circuitry, including an electronic automate zero. 't is eouipped with a flame ionization detector (3, 4) which is very sens't've to most organic compounds Including vinyl chloride. It is ourged with air and Drovided with a purge-power safety Interlock system permitting safe ope-et'on In operating areas which can sometimes be hazardous (Class 1, Group D, Oivis ion 2). A sample valve in the analyzer injects a 0.5 cc sample into a threecolumn system (Figure 4) every two minutes. The first column separates the vinyl chloride from the high-boiling compounds, such as vinyl acetate, which might be present in the sample. A column valve diverts the unwanted com pounds to vent, and at the proper time, diverts the vinyl chloride to the UCC 023410  a- second column where it is separated from most remaining interferants. A second column valve diverts t^ese interferants to vent, and at the proper time diverts the vinyl chloride to the third column, where the final senarat'on is made. (Some of the analyzers located where few interfering compounds are present, contain only two columns). The first two columns contain 20 percent Igepal CO-880 on a Chromasorb P support. The third column contains 20 percent Octoil S on a Chromasorb P support. Ca'lbratlon sample for the chromatograph Is 10 ppm VCM In a nitrogen matrix, prepared by the part'al pressure technique and kept in a steel cylinder. It can be introduced manually for calibration, and is automatically introduced as Sample Vo. 20 during normal operation. The microcomputer Is programmed to verify that Sample Vo. 20 is always In range. Data and Alarm System Each monitor is equipped with a potentiometric strip-chart recorder 'ocated on the analyzer pane', but the heart of the data system is f'e Union Carbide Model 2800 M'crocomputer, which is used in conjunction with a OS Term?Vet 300 Printer. The Model 2800 was developed by Union Carb*de or "n-bouse use. It Is designed around the Intel MCS-8 Microcomputer System. It is an 8-bit oara'lel processor with a typical instruction cycle time of 12.5 microseconds, and is capable of addressing 16,384 eight-bit words of memory. The GE TermIVet 300 prints at a rate of 30 characters per second. The result of each analysis is immediately printed out on the stripchart recorder at the same time the VCM is ''seen" by the detector. ucc 023411 -'0 deferring to Figure 5, the wide bars are the sample point identif:cation bars; the length of each ?s orooortional to the sample point number it represents. The line above the sample point identification bar Indicates the VCM concentration at that sample point. The microcomputer accepts the raw analysis data from the analyzer, performs two checks to determine that the data is valid, and then stores the data for printout and for calculating a number of statistical summar:os to provide information on probable employee exposure. These summaries are nrov;ded on work shift, daily and monthly bases (Tables I - III). One of the checks the microcomputer makes before storing a piece of data rs to see whether the sampling system flow switch is actuated (i.e., s.-imolo flow is insufficient). if it is, on alarm is actuated and the word "FLOW" is printed out instead of raw data for that sample point, and the value is not included in calculating the statistical summaries. The other check is oerformed following each analysis of the check sample, if r^e result obtained for the check sample does not lie betwee" two orese'ected values, an alarm is actuated and the message "CHECK C-AS" will appear on the next printout. Subsequent analyses values will be printed out but not stored Eor use in calculating the statistical summaries until the result for the check sample returns to an acceptable value. A printout of results occurs upon completion of analysis of all twenty sample points. The time of day is printed out with each line of data. At the end of each work shift, day and month, a summary of the data Is printed (Table 1). ucc 023412  M. I ,, EncH shift summary (Table l) consists of the average of the past rinbr hours' analyses for each point, a total of the number of times the overall averano VCM concentration exceeded a preset value, and a calculated average exposure by joh-tynn. The job-type average-; are time-weighted averages for up to 16 d;cferent job types, which are based on the estimated average percent of time a worker doing a particular type of job spends in the vicinty of each sample no'nt be *ng analyzed. They are calculated from the eight-hour averages. The daily summary (Table 11) consists of two sections: 1) a summary o* the oast 2h hou>-s in the form of averages for each point and for the entTre unit, and the number of times each point and the entire unit exceeded the preset 'imit; and 2) a month-to-date summary which consists of the average and standard deviation of all data accumulated to date for the current month, the total number of times each point exceeded the preset limit, and the average number of times per day each point exceeded the limit. The monthly summary (Table !!!) consists of six sections with the same cornat, one for each of the rour shrfts, one for day shift personnel, n"d one for the unit as a whole. Each section consists of a summary for the ent're month and contains the average analysis value and standard deviation for each sample point and for the entire unit, the total number of t:res each sample point exceeded the preset limit, the average number of times per day each point exceeded the limit, an estimated average exposure and standard deviation by job-type as described in the 8-hour summary, ard finally, a calculated probability that the employees of each job type were exposed to a concentration greater than the preset level. j / ucc 023413 -12- ,r the amount of VCM In the sample is above either of the two adjustab'e alarm levels preset in the microcomputer, an alarm is actuated. !n order to he'n 11'meet the regulations of Aoril 1, 1976 and to eliminate the "lag time" nvo'vod when the control room operator has to relay alarm Information to the r,eid operators, some of the systems are utilizing an additional output circuit of the microcomputer to act'vate warning lights located In each of the monitored areas. The light stays on until subsequent analysis of that samp'e polrt shows the VCM concentration has returned to a safe level. Consideration has been "`von to ut:!'zing this same circuitry to activate an annunciator which would automat-ca11y play a recording announcing which area Is unsafe. An alarm can a1so be actuated if the flow switch in the sampling system detects a low flow {e.g., a n'ygged sample line) or if the check sample analysis is beyond the preset values. Cystem Performance The performance of the monitoring system was checked to see how var:ous nnyTroemental factors a"ected its response. Extensive tests were -un with one of the analyzers to determine Its reproducibi11ty under both steady nrd unsteady-state ambient temperature conditions, and to determine the Tinterference efrect of the many different compounds sometimes present in the ambient a:r o^ the production units where VCM is monitored. Under controlled ambient conditions, with the measuring range set to 0-10 ppm, the readings from a 5 ppm VCM calibration sample reproduced within 0.05 ppm for 24 hours (Figure 6). A sample containing no VCM and another conta;n:ng '.0 ppm reproduced slightly better. !p other tests, with the measuring range set to 0-25 ppm, the ' ppm sample reproduced within 0.1 ppm for 24 hours. ucc 023414 -13- The average VCM readings changed slightly as the ambient temperature changed. The results are shown In Figure 7. Ambient temperature changes produced another effect which might in some c!rcumstances have a small effect on the results. A rapid change in the temoerature (3-4C/mln.) of the sample tubing near the analyzer can change the readinn by as much as 7,0 nhrccnt of the VCM'concentration. A sudden increase of temperature Increases the normal response from the sample, whHe a sudden decrease of temperature decreases the response. This apparentl results from a small amount of adsorption of sample on the tubing walls. Th`s ;s not believed to be a serious problem for the existing applications, but If the effect occurs, It con be reduced by insulating the sample tubing near the analyzer to prevent rapid temperature changes. Operating personnel at the vinyl resins units were questioned concerning what compounds might be present In the atmosphere at the units. Synthetic samples of these compounds and others which were known to elute near the VCM elution time were prepared and introduced into the analyzer to determine whether any of them would Interfere with the VCM analysis. Methyl chloride oroduced the greatest interference of any of the compounds. 20 ppm methyl chloride produced a response equivalent to 1 ppm VCM. High concentrations of toluene were also found to cause some interference. 1000 ppm toluene produced a response equivalent to 6 ppm VCM. This response to toluene was not immed'ate it showed up approximately 40 minutes after it was sampled. The complete results of the tests are shown in Table IV. ucc 023415 - ' b- "'ests were performed on two different sample lines instated :n a o'.-nt Ana was Ah? feet of 1/V `rype -3r stainless steel; tine other, 125 'act or V eo'yethy'nne. Mo reduction in the response to a 1 ppm VCM sample was r'nrpct w''ns '>vV"d through the sta'aless steel line. Also, the fu11 response was cbta'n^d w'th'n the two m>ute analysis cycle time, "owever, a loss o' abcu "'i nrrrrn1: or ``hr response "o the same 1 pnm sample occursed when it was ou''ed though the oo'yethy'one line, even after several minutes. ''Vse test 'esu'ts a^d the poor mechanical properties of polyethylene i_.vnr. j i` st;r:ed the secommendat ion not to use it for sample lrnos in spite o' substantia''y 'ewer installation costs. in general, t^e "on'to-s 'nave required little maintenance and stay "on-line" greater than o' the time. Periodically, a mfcroswitch or e'ectromechanical nelay has had to be replaced in the ^jjomatograph, and so'enoid va'ves have had to be rcolaced Tn the sampling systems. A number o' oroblems were encountered''With the printers shortly after each was s^a'-gnr' un, but once t'^nse were corrected, the printers have not n^nvl-nr a n^oap deal of maintenance. ucc 023416 -:5- System Ut?!ization The most important serv'ce the VCM monitors have performed is In he'p'ng reduce VCM emissions by 'ocating sources of leaks. it has helped engineers determine the perpetual trouble spots, which were remedied by equicmenr^edosign or replacement, and it st'll helps shift personnel locate leaks due to equipment malfunction. Whenever the VCM concentration exceeds an acceptable value at any sample point, one of the shift personnel goes^nto the vicinity or the sample point with proper respiratory equipment and a portable organic vapor ana'yzcr to locate the leak source. 1n most cases, the estTmated average VCM exposure of each employee calculated by the microcomputer agrees aulte well with the data obtained by carbon-tube and Sipoin pump worn by the employee. At one of 'JCC's vinyl res'ns units, agreement has been so good, they have done away with requiring personnel to wear carbon tubes, and rely solely on the monitor's data. Thrs emp'oyee exposure calculation has also been put to use in a s'Jght!v d;"erent way. At one plant 'ocatlon, personnel are trying to determine :E one of their buildings can be de-regulated. They have entered all the samrj'e po'nts in that build:"g as one "job", so the microcomputer prints out an easily referred to set of data on the building, along with the printout of employee exposure data. ucc 023417 -i$- CONCLUSION The VCM monitoring system? which have been started up and debugged have oerformed very well. The data have be(?n obtained and logged on permanent '`ecords w'thout consuming valuable operator time. The systems have bean valuable In rapidly detect'ng VCM emissions into the working environment so that Immediate corrective action can be taken and employee exposure minimized. These or equivalent monitoring systems are essential to the vinyl rcs;ns units :n meeting the existing regulations affecting personnel exposure to VCM. The systems are very sensitive, accurate and relatively maintenance free. VCM concentrations can be read to the nearest 0.1 ppm, and very few compounds arfoct the VCM readings. Several built-in system checks, such as sample flow and the automatic introduction of calibration gas help to ensure that the data obtained is accurate and reliable. Union Carbide considers the VCM monitoring systems to be completely successful, and preferred to other approaches for dealing with the VCM ""'sum and employee exposure problems. Similar systems would be useful ;n monitor'am other volatl'e compounds wh'ch are suspected to be harmful to human health. ucc 023418 o *"onc0<5 red. P.eg'st. 39(67): T 23^2-1 23hh ('97b) . "od. Peg's*. 33 (' S'O : 35390-3529? ':74). Cn'cote, H. 7.: !or> Production and Pemoval in Flames, '"esented a* ".'"o F7ghth Symposium ('ntemat Tonal) on Combustion, ?, sadena. Sa'TJ7orn`a, August 7-3 - Septer'ber 2, 'S6C. ^ <-0 r"b" rg t J. D., Co ' ' awnv, V. S., and D. T. L. Jones: 1 "",',e Mechanism or ,,"soon^e o*7 r'aro 'citation detectors." ISA Proceed' ngs '9GT Into "atronn1 Cns Chromatography Sympos!um. ucc 023419 ucc 023421 FIGURE 3. TYPICAL SAMPLING STATION ucc 023422 jlwywypwEMff'.11. in* wnitoiofr fiM r * M ucc 023423 S ilflN IW -1W11 30 3 IS is 30 PPM VCM PBOM CALIBRATION IAMHI 3 30 3 w 3o 3 3* 3 10 34 3* 3s 33 STRIAM 1 ft 111 ft ' ' ------------ --- ' "Hi---------- A...... A---------- r * ft--------- A 1 ft'....... i&o CHABT DIVISIONS FIGURE 5. ANALOG OATA PRESENTATION Iiccf 023424 * TIM ! FIGURE 6. VON MONITOR REPRODUCIBILITY -- ~ ucc 023425 70 *0 SO O = 40 > 5 30 5 PPM VCM ANAIYZIR SAMOI 0 - 10 PPM 0-100 CHAU DIVISIONS 20 I PPM VCM 10 0 PPM VCM 10 20 30 40 50 TIMPIRATURI C* FIGURE 7. EFFECT OF AMBIENT TEMPERATURE ON ANALYZER RESPONSE ucc 023426 TABLE 1 FORMAT FOR RAW DATA AND SHIFT SU-MARY VINYL CHLORIDE BY CONTINUOUS JWLTIPOINT CAS CHROMATOGRAPH 1C -ANALYSIS EXPERIMENTAL UNIT--NO TOWN, USA 06/39/25----------------------------------------------------------------------------------------------------------------- --TIME-- A SHIFT SAMPLE POINT -=afc-a2ft.--.^fc."=ate -22- --3Q--_4.L. -J5- -4&_ -ig 00U3------ ------CLrQ--M- 0.0 0,0--QwQ- -9.0- 0,0 -0.0- Q0 -0.0 Q.O -0r&- -Qi-Q-- OtO--9.0- --0r8---8*6- -6.-6- 0-0 -40. 00*53 Ot *32 02H2 00 6.6 0.0 00 0.0 0.0 00 o.o 0.0 00 0.0 0.0 nn 0.0 0.0 0o 0.0 0.0 0,0 0.0 n.n 00 0.0 n.n 02*52 03*32 0.0 p.o 04*12 0.0 04*52 0.0 05*32 0.0 06*12 _ 0.0 06*52 0.0 07*32____ ___ Q.Q 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Q.O 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 O.B 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 Q.Q.. 0.0 0.0 n.n 0.0 0.0 0.0 0.0 0.0 0.0 0.0 n.n 0.0 0.0 0.0 0.0 0.0 Q.O 0.0 0-0 0.0 0.2 0.0 0.1 0.0 0.0 8 HOUR AVERAGES A06/30/75 ____Q..Q Q.Q._Q42_ Q.O Q.l Q.Q .0.0 -Q.Q NUMBER OF TIMES EXCEEDING 5. 0 PPM* 0 0o 0.0 0.0 0.0 n.n 0.0 Q.O 0.0 o.o 0.0 0.0 nn 0.0 n,n 0.0 n.n 0.0 0.0 0.0 Q.Q 0.0 0.0 00 0.0 n.n 0.0 n, n 0.0 0.0 0.0 0.0 0.0 o.n nn 0.0 nrn 0.0 no 0.0 0.0 0.0 0.0 0.0 0.0 0. 0 - Q.Q -Q4L._CLO na 0.0 0,0 0.0 o, o 0.0 0.0 0.0 0,0 0.0 Q.O Q.Q nn 0.0 0.0 0.0 0,0 0.0 Q.O 0.0 0.0 0.0 0-0 Q.Q /v n 0.0 0.0 0.0 n, D 0.0 0,0 0.0 0.0 0.0 0-0 Q.Q nn 0.0 o.n 0.0 0,0 0.0 0-0 0.0 o.o 0.0 0.0 Q.O AA 0.0 o.n 0.0 n. n 0.0 0.0 0.0 0,0 0.0 0-0 0.0 nn 0.0 0.0 0.0 0,0 0.0 0.0 0.0 o.o 0.1 0,0 0.0 n n in o.o 1331 0-0 to 0.0 1Qf n. o ip 0.0 in*. 0.0 |f)f 0.0 lQr n.n ipj 0.0 1Q^ n.n |fy, o.o in. AVERAGE EXPOSURE BY JOB TYPE JOBOO* jQ^Q____ J0B01*21.3___ J0B02* 0.0 J0B10*19.3 JOB 11 * 0.0 JOB 12*14.8 J0R03*20. 3 JOB!3*12. 8 J0B04 *13.7 JOBI4 * 12.9 JORQ5125 .0 JOB) 5* 15 .2 .10R0AU2.9___ -1QB07I25-A JOB 16*12.8 JOB!7* 0.0 B SHIEI 08*12 08*52 09*32 0.0 0.0 0.0 0.0 0.0 0.0 Q.Q 0.1 .0.0- 0.0 Q.Q -_Q..Q_ Q.Q Q.Q Q.Q 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Q.O 0.0 0.0 Q.Q 0.0 0.0 Q.O 0.0 0.0 0.0 o.o 0.0 10*12 10*52 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 o.o 0.0 Q.Q 0.0 0.0 0.0 0.0 0.0 11 *32 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12*12 12*52 0.0 0.0 0.0 0.0 0.0 0.0 0.2 o.o 0.2 0.0 0.0 Q.Q Q.Q Q.Q Q.Q 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 13*32 0.0 0.1 0.0 0.0 0.0 0.0 -041 -Q41- Q.l -Q..-Q- Q.Q Q.Q -Q.Q- n.n -o.n. 1 14*12 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0. 1 . i 14*52 15*32 0.0 0.0 0.0 0.0 0.0 0.0 Q.O Q.&. o.o Q.O 0.0 -Q41. 0.0 Q.Q 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0. 1 0.0 0.0 0.0 0.0 0.0 0.0 8 HOUR AVERAGES BQ6/30/75 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 * 0.0 0.0 0.0 0.0 0.0 NUMBER OF TIMES EXCEEDING__ 5.0 PPM*_____ 0 * 0.0 0.0 0-0 0.0 Q.O 0.0 0.0 0.0 0-0 0.0 0.7 0.0 0.1 Q.O 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0. 1 0.0 0.0 0.0 0.0 n.n 0.0 0-0 0.0 Q.O 0.0 0.0 0.0 0.0 0.0 0.0 o.o 0.0 JJL. o.o m 0.0 100.0 If1)TM 0.0 10- o.o 0.0 100.0 JO 0.0 ioT o.o irtT 0.0 IQ. 0.0 IQTf. '% AVERAGE EXPOSURE BY JOB TYPE JOBOO* 0.0 J0B01*21.6 JOB02* 0.0 JOB03*20. 6 JOBIO* 19.3 JOBIH Q.O JOB12 15.2- J0BI3H3.0 JOB04*13.8 JOB05*25 .4 J0B06*13. 1 JI1B14U341- J0BJ5X15.2___ J0R16H3-0 J0B07*26.1 JOBI7I 0.0 ucc 023427 f TABLE II DAILY SUMMARY FORMAT WITH NON-FACTUAL DATA VINYL CHLORIDE BY CONTINUOUS MULTIPOINT GAS CHROMATOGRAPHIC ANALYSIS EXPERIMENTAL UNIT--NO TOWN, USA SUMMARY FOR 06/3QyiS SAMPLE POINT ENTIRE 21 22 23 24 28 2t\ 27 28 20 30 31 ,1? 33 34 38 3 A 37 3ft 30 An ifhi it 24 HOUR AVERAGES 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.3 0.0 NUMBER OF TIMES EXCEEDING 5.0 PPM (EACH POINT) OOP_____0_____Q____0-____ a____ CL____ 0_____L____ a____Q____ Q____ Q____ Q____ Q____ Q____ Q____ Q___________ Q__ month-to-date SUMMARY AVERAGE FOR MONTH 0.4 0.3 0.1 0.1 0.5 2.6 0.4 0.4 0.3 1.0 0.1__ 0.3- 0.2 _ 0.3 0.2 2.2 0-3 1.3 n,,? 10. 1 0.6 STANDARD DEVIATION OF AVERAGE 1.7 1.6 0.5 0.5 1.6 6.4 0.9 1.1 0.6 2.9 0.4__ Q.B Q.3 -0.9 -J-.L. 6.3 0-9 3.7 0,,9 0.6 2.8 NUMBER OF TIMES EXCEEDING 5.0 PPM < EACH POINT) 14 8 0 1 20 137 10 7 3 38 ____ 0_____5 ___ 6____ 2____ &_ 106 7 78 ft 460 AVERAGE NUMBER OF TIMES PER DAY EXCEEDING 5.0 PPM (EACH POINT) 1 1_____ Q____ Q_____2____LL_____I_____ CL___ Q____ 4_____ 0.____0_____Q____ 0____ 0_ -13. ___ Q____ 2_ _L __ 57 ooo % TABLE III AVERAGE FOR MONTH MONTHLY SUMMARY FORMAT WITH NON-FACTUAL DATA' 0.3 0.3 0.2 Q.l 0.4 2.4 0.3 0.3 0.2 0.6 nT 1 o.? 0.2 0,3 Ot 2 ?,? n, 3 1 r 3 0. 3 in,, 2 0.5 STANDARD DEVIATION OF AVERAGE 0.7 1.2 0.3 0.3 0.0 5.1 0.6 0.4 0.4 1.1 0.2 n.6 0.4 O, 5 1 .6 4.3 0.7 2.ft 1,2 n 5 2. i NUMBER OF TIMES EXCEEDING 5.0 PPM (EACH iPOINT) 1 2 0 0 2 48 1 0 0 3 0 0^ _0 O 1 35 1 31 5 1 30 AVERAGE NUMBER OF TINES PER DAY EXCEEDING 5.0 PPM (EACH POINT) Q_____0 0 0 0 6 ___ 0_____Q____ Q_____Q_____ Q____ Q_____Q____ Q____ Q____ 4____ fl____ 3____ L __ LA_ AVERAGE EXPOSURE BY JOB TYPF____________________________: JOBOO* 6.4 JOBOI*33.7 J0B02* 2.0 J0B03*25.9 J0B04*17.0 JOB05*45.3 J0B06*21.2 JOB)Q*22.5___JQfll It 0.4___ JOB)2124.3___ JDB13H6.5___ J0BI4I2Q.2___ J0B15H7.7____JQR14H6.4 J0B07t47.2 JOBl7i 3,6 STANDARD DEVIATION ON AVERAGE EXPOSURE JflBQQl 3.3___ J0B01 i 2.0___ JQBQ21 3.6___ JOBQ3I 4.9___ JORfMl 4.6___ JQRQ^i 4.3 JOB 10* 4,4 JOB! I* 0.7 JOBI2* 2.7 JOBI3* 4.2 J0BI4* 7.1 JOBI5* 4.2 .iQRQ4i a.a JOB16* 6.6 ioro7 7.3 JOBI7* 6.6 PER CENT PROBABILITY OF EXPOSURE GREATER THAN__ l.a PPMt________________________________________________ _ JOBOO*9I*9 JOBOI*99.9 JOB02*5I.7 JOB03*99.9 JOB0499.9 JOB05*99.9 JOB06*99.7 JOB07*99.9 JOBIQA99.9 JOB-11 * 2.5___ JOB 12199.9___ JQBI.3199.9___ I0B14I99.3____ JQBI5l99.9____lQRl4Qft.7 jori7i40 3 UNIX AVERAGE FOR MONTH 0.4 0.3 Q.l Q.l 0.5 2.6 Q.4 Q. 4 0.3 1.0 0.1 0-3 0. ? 0.3 0.2 2.2 0.3 1.3 0.2 in.i STANDARD DEVIATION OF AVERAGE _____U2__ 1.6 0.5 0.5 1.6 6.4 0.9 1.J__ Q.A 2.9. 0.4 0.8__ Q.,7 0.9 1.1 6.3 0.9 3.7 0.9 0.6 NUMBER OF TIMES EXCEEDING 5.0 PPM (EACH POINT) 14 8 0,1 20 137 IQ_____2_____3 38 Q 5 A. 7 5 106 7 78 R AVERAGE NUMBER OF TIMES PER DAY EXCEEDING 5.0 PPM (EACH POINT) _______ 1_____ I_____ Q____ Q_____2____LZ_____ 1_____ 0_____Q____ 4_____ Q____ 0____ SL____ Q_____0____L3_____0_____9 ____1____ n.6 2.5 440 57 AVERAGE EXPOSURE BY JOB TYPE______________________________________________________________________ . JQBOO* 6.9 JOBOI*35.1 J0B02* 2.1 J0B03*26.1 J0804*17.2 JOBO5*47.1 JOB06*22.0 JOB10*22.7 JOB111 0.4 J0B12I24.5 JQBI3*17.2 JOB14*20.2 J0B15*17.7 J0R16t17.0 J0B07*49.2 JOR171 4.0 023429 STANDARD DEVIATION ON AVERAGE EXPOSURE JQBOO* 7.4 JOBOI* 7.8 J0B02* 4.8 J0BQ3* 5.4 JOBIO* 5.9 JOB 11 * 1.1 J0B12* 5.2 JOB13 6.5 JQBQ4I 6.1 JOB 14* 6.9 JOR051 5.9 J0R06* 3,9 J0BI5* 5.5 JOB 1'* 2.5 J0R07* 7.0 JOB 17* 2.1 PER CENT PROBABILITY OF EXPOSURE GRFATER THAN__ LA PPM*__________ JOBOO *74.9 JOBO1*99.9 JOB02 *51.7 JOB03*99.9 J0B04I99.3 JOBIO99.9 J0B11*10.9___ JDBI2I99.9 JOB 13199.1 ___JQBI4199.5 J0B05I99.9 JOB06*99.9 JOB07 *99.9 J0B15I99.7___ JOR 16l99r 0 -JOB! 71-B5. 1 % >* TABLE IV Response of VCM Monitor to Compounds Other Than VC1 (Analyzer Range - 25 ppm Full Scale) Compound Concentration, ppm Response, as ppm VCM acetaldehyde acetone butadiene n-butyl acrylate 1-butyne chlorine ethyl chloride ethylene oxide hydrogen chloride isobutane isobutylene isopropanol methanol methyl acetate methyl chloride methyl ethyl ketone n-nonane 1-pentene toluene trichloroethylene vinyl acetate methane ethane ethylene acetylene propane propylene aliens propyne n-butane 1-butene cis-2-butene trans-2-butene 25 10,000 100 1,000 100 100 100 100 100 100 1,000 1,000 1,000 1,000 20 1,000 1,000 100 1,000 1,000 1,000 1,200 Total Done None None 1* None None None None None None 0.2 None None None 0.6 None None None 6* None None 0.6 * Interference does not appear immediately, but occurs approximately 20 samples later. ucc 023430