EMB PROJECT REPORT NO. 75-VCL-2 CD o AMERICAN CHEMICAL COMPANY Carson, California UNITED STATES ENVIRONMENTAL PROTECTION AGENCY Office of Air and Waste Management Office of Air Quality Planning and Standards Emission Measurement Branch Research Triangle Park. North Carolina ------- SET 2526 02 0175 Vinyl Chloride Testing Conducted at ttv American Chemical Company Carson, California, WithfSupplemental Method Evaluation? EMB Project Report No. 75-VCL-2 Prepared for: Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park, North Carolina 27711 EPA Contract 68-02-1400, Task No. 8 January 6, 1975 by Warren C. Kelly SCOTT ENVIRONMENTAL TECHNOLOGY, INC. 2600 Cajon Boulevard San Bernardino, California 92411 ------- SET 2526 02 0175 TABLE OF CONTENTS 1.0 Introduction 2.0 Summary and Discussion of Results 3.0 Process Description and Operation 4.0 Location of Sampling Points . . . 5.0 Sampling and Analytical Procedures Page Number 1-1 2-1 3-1 4-1 5-1 SCOTT RESEARCH LABORATORIES, INC. ------- 1-1 SET 2526 02 0175 1.0 INTRODUCTION The Environmental Protection Agency (EPA) has recently under- taken the task of establishing a test procedure and performance standards for vinyl chloride emissions from stationary sources. A performance standard is a standard for emissions of air pollutants which reflect the best emission reduction systems that have been adequately demonstrated taking into account economic considerations. In pursuing this task, Scott Environmental Technology, Inc., was contracted by the EPA to perform source sampling tests at the American Chemical Company, Carson, California, during the week of October 9, 1974. The purpose of the tests was to obtain background data for development of national new source performance standards in the vinyl chloride industry. Data was collected in three areas: 0 Evaluation of proposed vinyl chloride test procedure o Vinyl chloride emissions' to the atmosphere o Efficiency of emission reduction systems employed Data was collected at American Chemical Company's Vinyl Chloride Monomer Plant. Process exhaust gasses are passed through a caustic scrubber, burned in an incinerator and pass through a water scrubber prior to venting to the atmosphere. The vinyl chloride test procedure used was "Tentative Method for the Determination of Vinyl Chloride from Stationary Sources." dated September 4, 1974. Gas samples were collected before and after the incinerator and at the exit of the water scrubber. Water samples were collected at the inlet and outlet of the water scrubber. Moisture determinations were made using EPA Method 4 at each gas sampling location. Determinations of total chloride and chlorine content were made at the gas outlet of the incinerator and water scrubber and at the water inlet and outlet of the water scrubber. Oxides1of nitrogen content of gas exiting the scrubber to the atmosphere were made using EPA Method 7. SCOTT RESEARCH LABORATORIES, INC. ------- 1-2 SET 2526 02 0175 The test team was comprised of two EPA representatives and six Scott employees. Six complete tests were performed. The test equipment was set up on October 7, 1974, and tests were performed on October 7, 8 and 9, 1974. SCOTT RESEARCH LABORATORIES, INC. ------- 2-1 SET 2526 02 0175 2.0 SUMMARY AND DISCUSSION OF RESULTS This section is divided into four sub-sections, Section 2,1 presents vinyl chloride emission results as well as results from addition tests conducted at each sample point. Section 2.2 presents the results of a vinyl chloride degration study conducted. Section 2.3 presents calibration data along with the results of a subsequent reanalysis of the vinyl chloride samples, and Section 2.4 presents a discussion of the mass spectrometer analysis conducted on several samples, 2,1 SUMMARY OF TEST RESULTS A summary of the test results from sample points A, B and C (C1) are presented in Tables 2,1, 2.2 and 2.3 respectively. A second summary in terms of Kg/hr. with emission levels at points B and C corrected to 10 percent oxygen equivalent is presented in Table 2.4. Results are presented in both British and Metric Units. Test results of water samples collected at sample points D, D1 and E are presented in Table 2.5. As indicated in Tables 2,1 through 2.3, supplemental tests (moisture, chlorine and total chloride, and NO ) were conducted, once A per day, during vinyl chloride tests 1, 3 and 5. It was not decided until the second day of testing to test for chlorine and total chlorides at sample point B. SCOTT RESEARCH LABORATORIES, INC. ------- 1 ^^-^— w^ 10/7 2 10/8 3 10/8 4 10/8 5 10/9 6 10/9 2-2 SET 2526 02 0175 TABLE 2.1 Summary of Results Sample Point A Entrance to Incinerator BRITISH UNITS Run Number Date (1974) Volumetric Flow Rate - DSCFMU' 22.02 26,03 23.29 14.78 45.74 59.56 Percent Mositure by Volume 14.1 25.0 11.4 Vinyl Chloride Concentration - % 4.26 26.40 17.50 8.71 1.33 2.43 Vinyl Chloride Concentration - 71.3 441.6 292.2 145.6 22.2 40.6 x 10~6 lbs/ft3 . . . ,;:.. Vinyl Chloride Emissions 9.42 68.93 40.77 12.92 6.11 14.49 Ibs/hr METRIC UNITS Run Number 1 2, 3 4 5 6 Date (1974) 10/7 10/8 10/8 10/8 10/9 10/9 VolumetricvFlow Rate NmJ/hr.U; 37.4 44.2 39.5 25.1 77.7 101.1 Percent Moisture by Volume 14.1 25.0 11.4 Vinyl Chloride Concentration - % 4.26 26.40 17.50 8.71 1.33 2.43 Vinyl Chloride Concentration - gms/m 114.2 707.3 468.0 233.3 35.6 65.0 Vinyl Chloride Emissions, Kg/hr. 4.27 31.26 . 18.49 5.86 2.77 6.57 ' 'Dry standard cubic feet per minute at 70° F. and 29.92 in. Hg. (2) Dry normal cubic meters per minute at 21.1° F. and 760 mm Hg. SCOTT RESEARCH LABORATORIES. INC. ------- 2-3 SET 2526 02 0175 TABLE 2.2 Summary of Results Sample Point B Incinerator Exit BRITISH UNITS Run Number 1 2 3 4 5 6 Date (1974) 10/7 10/8 10/8 10/8 10/9 10/9 Percent Moisture by Volume 5.0 3.0 7.8 Chlorine Concentration - ppm U; 18.34 1.04 Chlorine Concentration -> Ibs/Hr. U; 0.2626 0.0214 Total Chloride ... Concentration - ppm 6.42 0.35 Total Chloride ,_. Concentrations - Ibs/ r 0.0945 0.0074 Vinyl Chloride Concentration - ppm <0.5 2.0 1.4 15.0 1.6 2.7 Vinyl Chloride Concentration - x 10-9 Ibs/ft3 <0.84 3.26 2.36 25.12 2.64 4.51 Vinyl Chloride __ "Emissions, Ibs/hr;- <0.013~~ -0.049... 0.037 0.390 ~ 0.060 0.101 It is assumed the chlorine concentrations are in error due to the fact they are greater than total chloride concentration (see page 12 of Appendix G-II). (2) Flow data from point C was used for calculations at point B. SCOTT RESEARCH LABORATORIES. INC. ------- SET 2526 02 0175 2-4 TABLE 2.2 (cont'd.) Summary of Results Sample Point B METRIC UNITS Run Number 1 Date (1974) 10/7 Percent Moisture by Volume 5.0 Chlorine Concentration - ppm Chlorine Concentration *• gms/hr (2) Total Chloride f . Concentration - ppm Total Chloride Concentration - gms/hr ' ' Vinyl Chloride Concentration - ppm <0.5 Vinyl Chloride Concentration - mg/m3 <1.34 ""Vinyl Chloride ' - "~ Emissions y Kg/hr... <&.006 2 3 10/8 10/8 3.0 18.34 119.12 6.42 45.87 2.0 1.4 5.2 3.8 0.022 0.017. 4 5 10/8 10/9 7.8 1.04 9.71 0.35 3.36 15.0 1.6 40.2 4.2 0.177 .._ 0.027 6 10/9 2.7 7.2 0.046 It is assumed the chlorine concentrations are in error due to the fact they are greater than total chloride concentration (see page 12 of Appendix G~II). (2) Flow data from point C was used for calculations at point B, SCOTT RESEARCH LABORATORIES. INC. ------- n o w c/> W » n tc t- w o SB 2? o 2 w u> 3 n TABLE 2.3 Summary of Results Sample Point C Water Scrubber Exit to Atmosphere BRITISH UNITS Run Number ,(1) .(2) (2) 1 10/7 2598 9.06 7.6 23.09 0.3318 17.35 0.256 15.90 0.296 34.7 58.0 0.906 N/R N/R N/R 2 3 10/8 10/8 2588 2588 9.06 7.7 39.85 0.5704 38.06 0.560 68.50 1.271 24.3 34.7 40.6 58.0 0.632 0.902 30.0 27.0 50.3 45.1 0.781 0.699 4 5 10/8 10/9 2588 3719 13.47 7.7 9.28 0.1909 6.90 0.146 81.50 2.172 35.2 0.6 58.8 1.0 0.915 0.022 4.5 <0.5 7.5 <0.8 0.117 <0.017 6 10/9 6315 6.4 10.7 0.217 <0.5 <0.8 <0.017 Date (1974) Volumetric Flow Rate - DSCFM Velocity - ft/sec Percent Moisture by Volume Chlorine Concentration - ppm Chlorine Emissions - Ibs/hr Total Chloride Concentration - ppm Total Chloride Emissions - Ibs/hr NO Concentration - ppm NO Emissions - Ibs/hr Vinyl Chloride Concentration - ppm (PT.C) Vinyl Chloride Concentration - x 10-9 ibs/ft3 (PT.C) Vinyl Chloride Emissions - Ibs/hr (PT.C) Vinyl Chloride Concentration - ppm (PT.C1) Vinyl Chloride Concentration - x 10-9 lbs/ft3 (PT.C1) Vinyl Chloride Emissions - Ibs/hr (PT.C1) (1) Dry standard cubic feet per minute at 70° F. and 29.92 in. Hg. (2) It is assumed the chlorine concentrations are in error due to the fact they are greater than total chloride concentration (see page 12 of Appendix G—II.) CO w H NJ Oi O NJ Ui NJ ------- TABLE 2.3 (cont'd.) © SCOTT RESEARCH LABORATOR] M 3 Summary or Results Sample Point C METRIC UNITS Run Number Date (1974) 3 (1) Volumetric Flow Rate - Nm /hr Velocity - M/sec Percent Moisture by Volume (2) Chlorine Concentration - ppm Chlorine Emissions - gms/hr (2) Total Chloride Concentration - ppm Total Chloride - Emissions - gms/hr NO Concentration - ppm x KF NO Emissions - gms/hr 1 10/7 4411 2.76 7.6 23.09 150.50 17.35 116.12 15.90 134.27 234 10/8 10/8 10/8 4394 4394 4394 2.76 7.7 39.85 258.73 38.06 254.02 68.50 576.53 5 6 10/9 10/9 6315 6315 4.11 7.7 9.28 86.59 6.90 66.23 81.50 985.22 CO H to Ui to o o M Ui Vinyl Chloride Concentration - ppm (PT.C) Vinyl Chloride Concentration - mg/m3 (PT.C) Vinyl Chloride Emissions - Kg/hr Vinyl Chloride Concentration - ppm (PT.C1) Vinyl Chloride Concentration - mg/m3 (PT.C1) Vinyl Chloride Emissions - Kg/hr (1) 34.7 92.9 0.410 N/R N/R N/R 24.3 65.0 0.286 30.0 80.5 0.354 34.7 92.9 0.408 27.0 72.2 0.317 35.2 94.2 0.414 4.5 12.0 0.053 0.6 1.6 0.010 <0.5 <1.34 <0.008 6.4 17.2 0.098 <0.5 <1.34 <0.008 Dry normal cubic meters per minute at 21.1° C. and 760 mm Hg. (2) It is assumed the chlorine concentrations are in error due to the fact they are greater than total chloride concentrations (see page 12 of Appendix G—II). ------- © SCOTT 1 M c/> M 90 O K • f tfe. iBORATOR w SAMPLE POINT Vinyl A TABLE 2 Chloride Concentrations Before Points B and C Corrected to 10 Incinerator Inlet " Date o Run Sampled No. (1974) 1 10/7 2 10/8 3 10/8 4 10/8 5 10/9 6 10/9 NM3/ hr(D 37.4 41.2 39.5 25.1 77.4 101 VC. 4.26 26.40 17.50 8.71 1.33 2.43 vc, kg/ hr 4.27 31.26 18.49 5.86 2.77 6.57 SAMPLE POINT .4 and After Incinerator-Scrubber Percent Oxygen Equivalent •B Incinerator Outlet/Scrubber Inlet NM3/ hr(D 4411 4394 4394 4394 6315 6315 • - VC, PPM %02 0.5 10.7 2.0 10.2 1.4 9.79 15.0 9.96 1.6 10.2 2.7 10.2 vc, PPM @ 10% 00 0.5. 1.99 1.4 14.9 1.6 2.8 vc, kg/ hr 0.005 0.022 0.016 0.171 0.026 0.044 NM3/ hrCD 4411 4394 4394 4394 6315 6315 SAMPLE POINT C Scrubber Outlet VC, VC, PPM @ PPM %00 10% 00 34.7 10.5 ... 36.4 24.3 10.2 24.8 34.7 9.79 34.0 35.2 10.3 36.2 0.6 10.2 0.6 6.4 10.2 6.5 VC kg/hr 0.398 0.278 0.396 0.402 0.010 0.105 S3 Ui IV) a* o NJ •vj Ui (1) Dry normal cubic meters per hour at 21.1° C. and 760 mm Hg. ------- 2-8 SET 2526 02 0175 TABLE 2.5 Summary of Results Sample Points D, D' and .E ^ ' Vinyl Chloride Runsv ' ^ HCL Run Sample Point 1 2, 3,4 5,6 Date (1974) 10/7 10/8 10/9 10/8 10/9 Total Chlorides - ppm D 6205 475 548 365 1095 D' 1825 1643 1825 2008 3103 E 9673 10,403 14,235 8213 5475 No chlorine was detected in the water samples, No vinyl chloride was detected in the water samples. Sample Points D and D' - water to scrubber Sample Point E - scrubber discharge water (2) Integrated Samples SCOTT RESEARCH LABORATORIES. INC. ------- 2-9 SET 2526 02 0175 2.2 VINYL CHLORIDE DEGRATION STUDY Several samples collected underwent a degration study. This involved analyzing each sample in its original sampling container at periodic intervals. Results of each analysis are presented in Table 2.6. 2.3 CALIBRATION DATA AND SUBSEQUENT SAMPLE REANALYSIS 2.3.1 Calibration Data Each day vinyl chloride samples were to be analyzed, five vinyl chloride standards were made up (Section 7 of Appendix G-I). Concentrations of the five standards were 1, 10, 100, 1000 and 5000 ppm. Available also as calibration standards were three high pressure cylinders of vinyl chloride in nitrogen which had previously been analyzed for vinyl chloride by the National Bureau of Standards (NBS). The results of this analysis by NBS were.Cylinder 1- 0.883 t 0.044 ppm; Cylinder 2 - 40.6 ± 1.2.ppm; and, Cylinder 3 - 874 i 15 ppra. These cylinders were periodically analyzed during the vinyl chloride analyses; however, all vinyl chloride results reported in Section 2,1 are based on the calibration bag standards. Three discrepanices were noted during the first set of vinyl chloride and calibration analyses: (1) hang up could occur if samples of markedly different concentrations were analyzed back to back. This was attributed to the lack of a heated sample loop which was corrected jfor the subsequent reanalyses reported in Section 2.3.2, (2) An incorrect amount of nitrogen was being metered into the calibration bags. It was determined that 8.19 liters of nitrogen were being used as opposed to 5.0 liters as specified in the test procedure. All data presented in the report have been corrected. This discrepancy was corrected for. the reanalyses reported in Section 2.3.2. SCOTT RESEARCH LABORATORIES, INC. ------- 2-10 SET 2526 02 0175 (3) A "Chromosorb 102" GC column was not available during the first day of analysis (10/8/74). As a substitute, a "Porapak" column was used which gave adequate separation, An analysis was conducted to determine the advantage, if any, in using a least mean square fit of the calibration data instead of the 5000 ppm or 1000 ppm (NBS) calibration standard to reduce sample results. The term M (slope) in the least mean square fit equation Yi * mXi was evaluated where Y = nominal standard concentration and Xi= area of standard peak (counts) . Three sets of data were analyzed: 10/14/74 Bag Standards Xi Yi 1 1553 10 1458 "100 7800 1000 64786 5000 384,303 M - 46.64 Y5000/X5000 = 46.92 'Difference = +0,6% 10/21/74 Bag Standards Xi 1 1 10 I •'.. 10 10 . 100 100 1000 5000 Yi 29,648 27,735 115,736.58,542 116,414 122 ,.314 123,774 100,504 5,278,314 :M = 643.4 Y5000/X5000 =644.5 Difference = +0.2% SCOTT RESEARCH LABORATORIES, INC. ------- 2-11 SET 2526 02 0175 10/21/74 Bag Standards Xi 1 50 50 1000 1000 Yi 1187 50,338 49,769 1,001,680 1,007,272 M P 1004 ) = 1007 Difference = + 0.3% Y1000/X1000 = 1007 As a result of these small differences, only the 5000 ppm bag calibration standard was used to reduce vinyl chloride results, 2.3,2 Subsequent Sample Reanalysis As a result of the large variation in the initial calibration data and the discovery that the sample loop should be heated, it was decided that a reanalysis of each sample should be conducted with additional calibration data being collected. Each sample was assessed with respect to both a second set of calibration bag standards as well as the NBS calibration standards. Nitrogen blanks were run during samples and calibrations of distinctively different concentrations to eliminate any hang up of vinyl chloride. Samples were either retained in their original sample container, or a vacuum flask was filled with sample. The results .of the second analysis were considered to be more consistent and subsequently used in the data presented in Section 2.1. Results of all analyses conducted are presented in Appendix A. SCOTT RESEARCH LABORATORIES, INC. ------- o o w W SO n 68 O o M p Sample No. 1C 5A 5C 6A 6C Date of First Analysis (1974) TABLE 2.6 Summary of Degration Results Day 0 10/9 10/10 10/10 10/10 10/10 » — — .BiB™^— 22.4 12,800 3.1 17,700 3.4 CO « H t-o I/I N> O tsa Oi Sample Analysis (ppm) Day 2 Day 4 Day 5 696.0(1) 16.8 8,500 17:7/23. 15,300 0.0 Day 11 Day 18 Day 19 16.6 11,000 10.4 15,300 3.1 18.0 3.7 M NJ (1) May be-results of no sample loop heating. (2) Results of two analyses. ------- n O M O CB O 33 > O 2 M P TABLE 2.7 Calibration Bag Standards (Integrator Counts) Second Analysis 5000 ppm Date . Counts xlO"3 11/15 1,920 1,950 11/20 2,020 2,020 11/21 2,200 2,230 11/24 1,940 1,940 1000 ppm Date Counts xlO~3 11/15 453 448 450 462 420 422 11/20 466 470 11/21 434 425 11/24 384 390 100 ppm Date Counts xlO"3 11/15 56.7 55.8 49.8 49.8 11/20 52.9 57.9 53.9 11/21 54.2 54.2 11/24 48.6 47.0 10 ppm Date Counts xlO"3 11/15 45.8 47.1 11/20 .423 1.464 .523 1.640 .994 .994 11/21 6.200 6.000 11/24 5.380 S.fi7n 1 1 ppm Date Counts. xlO"3 11/15 11.5 10.8 11/20 0.312 0.312 11/21 0.340 0.755 0.330 11/24 0.483 0.567 0.600 C/I M ho o o M tn NJ I ------- 2-14 SET 2526 02 0175 2.3.3 Subsequent Standards Investigation As a result of inconsistencies noted in calibration standards during testing, an investigation was conducted to attempt to resolve these inconsistencies. In preparation of this investigation, the following action was taken: °A new cylinder of pure vinyl chloride was obtained. °The calibration bag evacuation technique was checked to verify complete evacuation was occuring. °New calibration bags were obtained. "The flow meter used to meter five liters of nitrogen was calibrated. \ °A bag, after being filled with five liters of nitrogen, was evacuated through a wet test meter and found to vary an average of less than 0.4 percent from five liters. A total of 19 tests were conducted. Each test consisted of repeated injections of one of three standards: °NBS high pressure cylinder of either 40.6 ppm or 874 ppm vinyl chloride in nitrogen standard injected directly into chromatograph. °NBS standards transferred into tedlar bag prior to being injected into chromatograph. °50 and 1000 ppm vinyl chloride standard in nitrogen prepared in accordance with procedure. Summary results of the 19 tests conducted are presented in Table 2.8 with complete results presented in Appendix D. One calibration bag previously filled with 874 ppm vinyl chloride in nitrogen was evacuated and refilled with nitrogen. Analysis of the bag contents after 14 hours at room temperature indicated a vinyl chloride content of approximately 1 ppm. SCOTT RESEARCH LABORATORIES, INC. ------- Run // 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 TABLE 2.8 Summary of Standards Investigation Standard Type and Concentration (ppm) NBS NBS Bag Bag 40.6 874 40.6 A0.6 874 N2(Bottle) 40.6 N, 40.6 874 40.6 40.6 874 40.6 40.6 874 Average GC .. Area (Counts) 50 50 1000 111,989 2,637,763 117,533 118,616 110,782 2,686,327 2905=1 ppm 0 121,208 163,167 160,511 3,335,252 2,858,007 113,156 119,963 117,582 120,736 2,776,128 2,749,745 Number of Injections' 6 5 3 2 4 3 2 3 3 3 3 4 2 5 A 2 2 2 Max. Deviation From Average (%) 1.3 6.4 8.5 4.0 5.6 6.4 48.6 ,8 ,8 6.3 1.8 3.9 0.8 10.3 2.0 1.7 0.7 1.1 Counts per ppm 2758.3 3018.0 2894.9 2921.6 2728.6 3073.6 2985.4 3263.3 3210.2 3335.3 3270.0 2787.1 2954.8 2896.1 2973.8 3176.3 3146.2 N> (Ji NJ O 10 Ui S3 Corrected to 760 mm Hg. pressure and 25° C. ------- 2-16 SET 2526 02 0175 2.4 MASS SPECTROMETER ANALYSIS Following gas chromatograph analysis of each A sample, the suspected vinyl chloride peak was analyzed using a mass spectrometer for positive identification of vinyl chloride. In each case, the suspected vinyl chloride peak was positively identified as vinyl chloride and composed of only vinyl chloride. Vinyl chloride levels in the B, C and Cf samples were too low for mass spectrometer analysis. SCOTT RESEARCH LABORATORIES, INC. ------- 3-1 SET 2526 02 0175 3.0 PROCESS DESCRIPTION AND OPERATION A total of 21 vents are combined to make the waste gas stream. Two of the largest of these vents are from the EDC direct chlorination reactor and the ethylene chloride reactor Csee Figure 3.1), Upsets in the operation of these reactors can be detecetd by the air loading on the pressure control valves. Data point A is the air pressure on the EDC reactor vent." The valve is air to close, 3-15 pounds per square inch gauge (PSIG). Data point B is the air pressure on the ethylene chloride reactor vent. The valve is air to open, 3-15 PSIG. The combined streams were measured at J with a type 710, two inch Ellison Annubar. Pressure differential was measured in inches of water with a Vol-0-Flo magnetic pressure cell. A sample of the inlet gas stream was made at point 1 and the sampling rate at point 1 was proportional to the flow measured at J. The incinerator-boiler was a modified Dixon Firetube (20' x 6') rated at 10,000 pounds of steam an hour (Ibs/hr) at 250 pounds per square inch gauge. During this test the steaming rate was set at 7,500 Ibs/hr. As the heat content of the waste gas went up and down, natural gas was added automatically to maintain this steaming rate. It was, therefore, possible to detect major changes in the composition of the vent gas by observing the natural gas valve position and the draft air valve position. The natural gas valve position was recorded as data point E (as percent open) and the draft air valve position was recorded as data point F (as percent open). The boiler steaming rate was recorded as data point C, and the boiler pressure was recorded as data point D. A sample of the flue gas was taken at point 2. It was not possible to measure flow at this location. The flue gas scrubber was a packed column, 54 inches in diameter and 40 feet high, The hot gas was cooled in a quench section of the scrubber by city water. Two Vee Jet 3/8 UNC 50 60 nozzles sprayed the water into the gas stream. A sample of this steam was taken at point 5, SCOTT RESEARCH LABORATORIES, INC. ------- 3-2 SET 2526 02 0175 and the pressure on the nozzles was taken as data point G. From information provided with the nozzles the total flow from the two nozzles was estimated to be 15 gallons a minute, The scrubbing water for the column was taken from a waste water sump^ A level control device maintained a level in the sump by regulating the water to the top of the scrubber (valve a on diagram). During periods of high waste water flow valve b also opened and allowed water to bypass the column directly to the sump to waste treatment. Valve a was air to open 3-15 PSIG, valve b was air to open 6-30 PSIG. The air loading to these two valves was recorded as data point I, the pressure on the discharge of the pump as data point H. The pH of the waste water stream was recorded as data point K, and a sample of the water stream was taken .at point 4. Flow at point 4 was estimated by the plant engineer to be 75 gallons per minute. The water effluent from the scrubber was sampled at point 6. The exit gas from the scrubber was measured by traverse and sampled at point 3. SCOTT RESEARCH LABORATORIES, INC. ------- AMERICAN CHEMICAL INCINERATOR TEST OCTOBER 7, 8, 9, 1974 Data Points A etc. Sample Points 1. etc. ro O o 19 other vents EDC vent a air natural 8as TS- steam to plant A to atmosphere Boiler Fire Box chloride vent two Vee jet nozzles city water CO U) to waste treatment Figure 3.1 waste water from plant ------- 4-1 SET 2526 02 0175 4.0 LOCATION OF SAMPLING POINTS The exhaust gas from the vinyl chloride monomer operation passed through a caustic scrubber, incinerator and water scrubber before exiting to the atmosphere. To facilitate testing, an exhaust stack was fabricated and installed on the exit of the water scrubber. Exhaust gas was sampled at four points: o Sample Point A - Inlet to Incinerator » Sample Point B - Inlet to Water Scrubber » Sample Point C - Exit to Water Scrubber « Sample Point C'- Exit to Water Scrubber Sample Points C and C1 were located at the same sample location and were intended to determine variability in sample collection and testing. In addition to the exhaust gas samples collected, three water samples were collected: o Sample Point D - Inlet to Scrubber - by product process water from monomer plant o Sample Point D'- Inlet to Scrubber - city water o Sample Point E - Water Exit to Water Scrubber Figure 4.1 details each point sampled during testing. Figure 4,2 shows the location within the stack of sample points, C and C1 , SCOTT RESEARCH LABORATORIES. INC. ------- r> o 50 W t/i M SO n D5 t" 03 O SB O 2 M y N4 z rt 9 Sampling Point Stack Caust Lc scrubbers Air Natural Gas Process Water To Atmosphere i a Ul N> o o 31.5 se- T 72" city water N> water out From Process Incinerator Figure 4,1 Sampling Point Location Vinyl Chloride Monomer Plant American Chemical Company Carson, California Water Scrubber ------- a- H to Ui ro o to o H1 U) Figure 4.2 Cross Section of Stack (Sample Point C and C') ------- 5-1 SET 2526 02 0175 5.0 SAMPLING AND ANALYTICAL PROCEDURES Samples for the determination of vinyl chloride content were collected in accordance with "Tentative Method for the Deter^ ruination of Vinyl Chloride From Stationary Sources" dated September 4, 1974, a copy of which is presented in Appendix G-I. Two-hour integrated samples were collected proportional to the gas stream in Tedlar bags sealed in 55 gallon fiber barrels. The barrels were evacuated which, in turn, would cause the bag to expand and fill. Four samples were collected as described in Section 4.0. Dupli- cate samples (C and C') were collected at the exit of the water scrubber using one glass and one stainless steel probe each equipped with a glass wool filter to remove particulate matter. Sampling after the incinerator (Point B) was accomplished using a stainless steel probe inserted into the gas stream at an elbow of an 18-inch diameter pipe. Sampling before the incinerator (Point A) was accomplished at a liquid knockout trap installed in a four-inch diameter pipe Immediately prior to entering the incinerator. No sample probe was used at this sampling location, and a small glass bulb water trap was installed between the liquid knockout trap and the sample line. Moisture determination was made at all three sampling points using EPA Method 4 as described in "Determination of Moisture in Stack Gases", Federal Register, Volume 36, No. 247, Tuesday, December 23, 1971. With the concurrence of the project officer, no heated sample line was used at sample point A. No moisture sample was collected at point C', and only one moisture sample was collected at" each sampling point per day. Moisture samples were collected for1 approximately one hour at each sample point. Velocity measurements were made at point C at the beginning of each test day in accordance with EPA Method 1, "Sample and Velocity Traverses for Stationary Sources" and EPA Method 2, "Determination of Stack GasiVelocity and Volumetric Flow Rate" as described in the above referenced Federal Register. SCOTT RESEARCH LABORATORIES, INC. ------- 5-2 SET 2526 02 0175 Flow rate data was also determined at sample point A via an Annubar installed in the process line, A description of the Annubar and its use is contained in Appendix G-III. The Annubar serial number was 53442, Model 715-316-SS and had a 2.067 inch inside diameter, An NO determination was made once per day at point C in X accordance with EPA Method 7, "Determination of Nitrogen Oxide Emissions From Stationary Sources" as described in the aforementioned Federal Register. The total chloride and chlorine content of the exhaust gas before and after the water scrubber (points B and C) was determined once per day in accordance with "Determination of Hydrogen Chloride and Chlorine in Stack Gas", U. S. Department of Health, Education and Welfare, April 1968, a copy of which is contained in Appendix G-.H. Water samples were also collected of the two scrubber water inlets and one outlet. Two samples were taken per day. One sample coincided with the EPA Method 4, Moisture Sampling, a sample of water being collected every fifteen minutes during the one-hour mositure test. The other water sample was integrated over all the vinyl chloride testing conducted each day. A water sample was collected every one half hour during each of the two-hour vinyl chloride tests. Each water sample was analyzed for vinyl chloride using a chromatic- graph and for total chloride and chlorine in accordance with Appendix G-II. Molecular weights of sample gas at points B and C were determined using mass spectrometer analysis for 0~ and C0_ content and subsequent molecular weight calculations. Weights at point A were determined by tare weighing an evacuated flask, filling the flask with sample gas to a known pressure and reweighing the' flask. SCOTT RESEARCH LABORATORIES, INC. ------- |