xe/EPA United States Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park NC 27711 rEMB"Report 80-DRY-10 October 1980 Air Petroleum Dry Cleaners Refrigerated System/ Condenser Emission Test Report Polly Prim Cleaners Lakeland, Florida ------- SOLVENT RECOVERY AND EMISSION CONTROL PETROLEUM DRY CLEANING INDUSTRY POLLY PRIM CLEANERS Lakeland, Florida Prepared for the U.S. Environmental Protection Agency Emission Measurement Branch Research Triangle Park, N.C. 27711 Prepared by Clayton Environmental Consultants, Inc. 25711 Southfield Road Southfield, Michigan 48075 EMB REPORT NO. 80-DRY-10 Work Assignment 37 Contract No. 68-02-2817 ------- TABLE OF CONTENTS List of Tables i List of Figures ii 1.0 Introduction 1 2.0 Presentation of Results 3 3.0 Process Description 14 4.0 Location of Sampling Points 15 5.0 Sampling and Analytical Procedures 18 APPENDICES A. FID Strip Chart Data and Field Data Sheets B. Strip Chart Data for Response Factor Determinations C. Velocity Traverse Data ------- LIST OF TABLES Table Page 2.1 Stoddard Concentrations 4 ------- LIST OF FIGURES Figure Page 4.1 Process diagram and sampling 16 locations 5.1 Stack cross-section-dryer exhaust 19 duct 5.2 Moisture sampling train 20 5.3 FID system 22 5.4 Calibration system 24 11 ------- 1.0 INTRODUCTION The U.S. Environmental Protection Agency (EPA) retained Clayton Environmental Consultants, Inc. to deter- mine the stoddard solvent emission levels from, and recov- ery performance of, a Hoyt Recovery Tumbler Dryer at the Polly Prim Dry Cleaners, Inc. plant in Lakeland, Florida. The results of this study will be used in research and development efforts for supporting New Source Per- formance Standards for the petroleum dry cleaning indus- try. This study was commissioned as Project No. 80-DRY-10, Contract No. 68-02-2817, Work Assignment 37. The testing program included determination of the following: (1) Total hydrocarbon concentrations (as stoddard) at the condenser inlet and the dryer exhaust duct during reclaim and deoddrize portions of each cycle; (2) Temperatures from the condenser inlet and outlet for both cooling water and exhaust gas; (3) Condenser water flowrate; (4) Volumetric flowrates at the dryer exhaust duct; and, (5) Moisture content at the dryer exhaust duct. - 1 - ------- The following process conditions were varied and controlled to quantify the throughput of solvent and the material mass balance within the system: condenser water flowrate and temperature, load size, and cycle duration. These conditions were recorded and will be correlated by TRW, Inc. to determine (1) the cost effectiveness of VOC removal for recovery dryers utilizing a refrigeration system to cool condenser water under high ambient tempera- tures and humidity, and, (2) to develop and substantiate operating limitations to ensure safe levels of stoddard solvent (defined as a percent of LEL) within the system, while maintaining minimal VOC stack emissions. Field sampling began July 21, 1980 and was completed August 7, 1980. The study was conducted by Messrs. N. Steve Walsh, Bruce G. Bird and Timothy J. Palmer of Clay- ton Environmental Consultants, Inc. Additional technical assistance was provided by Messrs. John R. Jernigan, and Steve Plaisance of TRW, who collected process data. - 2 - ------- 2.0 PRESENTATION OF RESULTS Table 2.1 presents all pertinent data needed to determine the thermal conductivity of the condenser and heat transfer capacity from the gas stream to the con- denser water stream, including condenser water flowrate and the temperature differential (AT) between the con- denser water inlet and outlet. Gas flowrate and tempera- ture differential between the condenser gas inlet and outlet were also determined. Elapsed time of the reclaim period and the type and quantity of metal used in the condenser coils, with the applicable heat conductivity constant, are not discussed. Condenser gas flowrates were not measured during the test program. Only condenser gas temperature differences between the inlet and outlet gas can be discussed. Con- denser gas (AT) data were not obtainable on 7/22/80 and 7/23/80 because the thermocouple had not been installed at the condenser gas outlet location. All temperature values reported in Table 2.1 are averages of individual readings recorded during both the reclaim and exhaust periods of each cycle. Stoddard concentrations were determined by averaging 30-second interval readings over the entire 6-minute exhaust period. This average value was then used to calculate the total mass of solvent (in pounds of stoddard per exhaust period). - 3 - ------- TABLE 2.1. STODDARD CONCENTRATIONS Date: 7-22-80 Start Time Batch Number Point in Cycle Condenser Water Temp. (F) Inlet Outlet Gas Temp. (F) Inlet Outlet Solvent Concentration (in ppm stoddard) Reclaim Maximum a Finish b Exhaust Start0 Finishd Condenser Water Flow (gal/min) Exhaust Flowrate (scfm) Emission Rates Total Ib of stoddard per period Ib/hr of stoddard 0911 0932 0944 1006 1019 1040 1054 1115 1127 1148 1231 1252 3 3 4 4 5 5 6 6 7 7 8 8 Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust 66 66 65 - 66 - 64 - 72 "• 78 78 78 - 78 - 75 - 84 ~ 113 108 109 - 107 - 108 - 110 ~ 154 155 155 152 _ 153 155 4724 4882 4882 5039 4882 5039 f f f f f f 3400 3496 4096 3936 3432 3624 2204 2348 3228 3104 2392 2080 7.59 8.92 8.34 9.44 9.42 10.5 306 306 306 306 306 306 1.94 2.01 2.60 2.60 2.01 1.89 19.4 20.1 26.0 26.0 20.1 18.9 a, 'Maximum solvent concentration Solvent concentration at end of reclaim period Q Solvent concentration in exhaust stream at start of exhaust period d Solvent concentration in exhaust s.tream at end of exhaust period Total mass of solvent emitted during exhaust period (in pounds stoddard and Ib/hr of stoddard) FID response during end of reclaim was not recorded to allow sufficient purging of the dryer exhaust duct sample line. ------- TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Date: 7-23-80 Start Time Batch Number Point in Cycle Condenser Water Temp. (F) Inlet Outlet Gas Temp. (F) Inlet Outlet Solvent Concentration (in ppm stoddard) Reclaim Maximum » Finish b Exhaust Start c Finish d Condenser Water Flow (gal/min) Exhaust Flowrate (scfm) Emission Rates Total Ib of stoddard per period Ib/hr of stoddard 0804 0831 0843 0910 0921 0948 1003 1025 1039 1100 1115 1136 1147 1158 3 3 4 4 5 5 6 6 7 7 8 8 9 9 Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust 65 66 66 66 66 65 64 78 79 79 79 78 78 78 160 162 161 165 158 158 152 159 156 160 158 155 153 141 4410 4410 5512 5197 5984 5039 5039 f 4400 f 5160 f f f 3528 4032 3196 5197 3544 3464 3888 1952 2016 2520 2676 2268 2000 2600 8.07 8.63 7.87 7.82 8.71 7.77 9.12 312 312 312 312 312 312 312 1.91 2.34 2.37 1.34 1.99 1.97 2.28 19.1 23.4 23.7 13.4 19.9 19.7 22.8 Maximum solvent concentration Solvent concentration at end of reclaim period Solvent concentration in exhaust stream at start of exhaust period a Solvent concentration in exhaust stream at end of exhaust period Total mass of solvent emitted during exhaust period (in pounds stoddard and Ib/hr of stoddard) FID response during end of reclaim was not recorded to allow sufficient purging of the dryer exhaust duct sample line. ------- TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Date: 7-25-80 Start Time Batch Number Point in Cycle Condenser Water Temp. (F) Inlet Outlet Gas Temp. (F) Inlet Outlet Solvent Concentration (in ppm stoddard) Reclaim Maximum a Finish b Exhaust Start0 Finish d Condenser Water Flow (gal/min) Exhaust Flowrate (scfm) Emission Rates Total Ib of stoddard per period Ib/hr of stoddard 0803 0945 0959 Reclaim Exhaust Reclaim 62 60 73 77 154 142 84 122 85 FID NOT OPERATING 8.05 4.87 FID NOT OPERATING Maximum solvent concentration Solvent concentration at end of reclaim period Solvent concentration in exhaust stream at start of exhaust period jj Solvent concentration in exhaust stream at end of exhaust period Total mass of solvent emitted during exhaust period (in pounds stoddard ativi Ib/hr of stoddard) ------- Date: 7-28-80 TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Start Time 0810 0836 0904 0931 0941 1007 1045 1056 1123 1134 1204 1214 1244 1253 1324 Batch Number 2 2 3 3 4 4 5 6 6 7 7 8 8 9 9 Point in Cycle Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Water Temp. (F) Inlet 65 64 64 61 65 66 64 Outlet 81 80 81 81 81 81 81 Gas Temp. (F) Reclaim Exhaust A™ Flowrate ?Ot*\^b , Ib/hr of Inlet 150 140 144 142 145 147 144 Outlet Maximum a Finish b Start0 Finish3 (gal/min) per period stoddard 90 5.24 - 89 5.87 131 90 4.47 129 130 FID NOT OPERATING FID NOT OPERATING V "*** *-"• ~ii~ *l~ _ 122 90 5.19 129 89 10.3 125 90 124 Maximum solvent concentration Solvent concentration at end of reclaim period CSolvent concentration in exhaust stream at start of exhaust period Solvent concentration in exhaust stream at end of exhaust period 6Total mass of solvent emitted during exhaust period (in pounds stoddard and Ib/hr of stoddard) ------- TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Date: 8-4-80 Start Time Batch Number Point in Cycle Condenser Water Temp. (F) Inlet Outlet Gas Temp. (F) Inlet Outlet Solvent Concentration (in ppm stoddard) Reclaim Maximum a Finish b Exhaust Start c Finish"3 Condenser Water Flow (gal/min) Exhaust Flowrate (scfm) Emission Rates Total Ib of stoddard per period Ib/hr of stoddard 1 GO 1 0829 0857 0905 0933 0941 1009 1034 1103 1113 1141 1154 1224 1240 1308 1321 1350 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust 64 64 64 66 65 66 65 65 83 84 85 87 89 91 83 80 137 142 142 140 143 146 138 146 87 126 126 89 129 89 131 89 132 91 130 89 132 89 130 5810 5794 5878 5676 5911 6549 7372 6700 5571 5646 5496 4846 4653 4878 5614 4855 5048 5085 5309 4108 4076 4750 5350 3864 2954 3290 3402 3241 2856 2792 3270 2510 5.05 5.06 5.03 5.06 5.03 5.06 5.05 5.04 306 306 306 306 306 306 306 306 2.56 2.72 2.77 2.51 2.26 2.35 2.79 2.08 25.6 27.2 27.7 25.1 22.6 23.5 27.9 20.8 Maximum solvent concentration Solvent concentration at end of reclaim period Solvent concentration in exhaust stream at start of exhaust period Solvent concentration in exhaust stream at end of exhaust period Total mass of solvent emitted during exhaust period (in pounds stoddard and Ib/hr of-stoddard) ------- TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Date: 8-5-80 Start Time Batch Number Point in Cycle Condenser Water Temp. (F) Inlet Outlet Gas Temp. (F) Inlet Outlet Solvent Concentration (in ppm stoddard) Reclaim Maximum a Finish b Exhaust Start c Finish d Condenser Water 'Flow (gal/min) Exhaust Flowrate (scfm) Emission Rates Total Ib of stoddard per period Ib/hr of stoddard 0719 0747 0812 0841 0856 0925 0941 1009 1020 1049 1058 1126 2 2 3 3 4 4 5 5 6 6 7 7 Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust 63 66 68 68 68 68 79 82 83 84 84 83 145 142 139 142 140 142 88 127 91 129 92 129 92 130 92 128 92 130 4413 5595 5678 5995 6328 5928 4378 4809 4941 5340 5738 5340 4278 4742 4875 5140 5538 5107 2620 2918 3217 3648 3880 3549 5.15 5.20 5.19 5.19 5.19 5.18 306 306 306 306 306 306 2.29 2.52 2.68 2.96 3.14 2.89 22.9 25.2 26.8 29.6 31.4 28.9 Maximum solvent concentration Solvent concentration at end of reclaim period °Solvent concentration in exhaust stream at start of exhaust period Solvent concentration in exhaust stream at end of exhaust period Total mass of solvent emitted during exhaust period (in pounds stoddard and Ib/hr of stoddard) ------- TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Date: 8-6-80 Start Time 0727 0756 0815 0844 0857 0926 0939 1008 1028 1 1056 1109 o 1138 1152 1 1220 1233 1300 Batch Number 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 Point in Cycle Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Condenser Water Temp. (F) Inlet 64 68 70 71 70 70 72 72 Outlet 79 83 86 86 85 86 87 87 Gas Temp. (F) Inlet 137 140 141 143 142 144 142 142 Outlet 87 128 91 129 94 127 94 128 93 131 94 129 95 129 94 130 Solvent Concentration (in ppm stoddard) Reclaim Maximum a Finish 7094 4330 8309 5262 8259 5096 8276 4962 8393 4896 9358 4862 9425 5162 9392 5096 Exhaust Startc Finish d 3963 1732 5129 1798 4996 1898 4496 1765 4430 1665 4163 1565 4130 1565 4563 1465 Condenser Water Flow (gal/min) 5.15 5.03 5.38 5.15 5.21 5.17 5.39 5.20 Exhaust Flowrate (scfm) 301 301 301 301 301 301 301 301 Emission Total Ib of stoddard per period 1.59 1.97 1.93 1.72 1.61 1.47 1.38 1.43 e Rates lb/hr of stoddard 15.9 19.7 19.3 17.2 16.1 14.7 13.8 14.3 Maximum solvent concentration Solvent concentration at end of reclaim period Solvent concentration in exhaust stream at start of exhaust period Solvent concentration in exhaust stream at end of exhaust period eTotal mass of solvent emitted during exhaust period (in pounds stoddard and lb/hr of stoddard) ------- TABLE 2.1. STODDARD CONCENTRATIONS (CONTINUED) Date: 8-7-80 Start Time Batch Number Point in Cycle Condenser Water Temp. (F) Inlet Outlet Gas Temp. (F) Inlet Outlet Solvent Concentration (in ppm stoddard) Reclaim Maximum3 Finishb Exhaust Start0 Finishd Condenser Water Flow (gal/min) Exhaust Flowrate (scfm) Emission Rates Total Ib of stoddard per period Ib/hr of stoddard 0706 0734 0754 0822 0834 0902 0924 0952 1009 1038 1048 1150 1 1 2 2 3 3 4 4 5 5 6 6 Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust Reclaim Exhaust 70 73 73 75 74 74 86 88 90 89 87 87 140 143 142 142 139 142 94 129 95 129 96 129 97 129 95 123 95 126 7510 7593 7976 9358 7693 8526 4862 4696 5046 5362 4463 4779 4862 4696 5046 5362 4463 4779 1499 1499 1682 1699 1365 1365 5.18 5.20 5.15 5.20 N/A 5.23 306 306 306 306 306 306 1.56 1.48 1.73 1.73 1.26 1.24 15.6 14.8 17.3 17.3 12.6 12.4 Maximum solvent concentration Solvent concentration at end of reclaim period CSolvent concentration in exhaust stream at start of exhaust period Solvent concentration in exhaust stream at end of exhaust period eTotal mass of solvent emitted during exhaust period (in pounds stoddard and Ib/hr of stoddard) ------- Prior to each test, both the condenser water flow- rate and temperature were set and regulated closely at selected time intervals, using separate controls. Addi- tionally, the volume of solvent reclaimed (ml) and wet (solvent laden) load size for each batch, along with clothing fabric type were recorded in conjunction with these data by TRW, Inc. The ambient temperatures and humidity of the air surrounding the dryer system varied little from day to day, but increased noticably from the early morning hours to mid-afternoon. Cost effective operation of the solvent recovery tumbler dryer should be determined from these settings as they affect the con- sumption of all utilities. Consideration should be given to achieving a safe level of stoddard solvent (percent LEL) within this closed system during the reclaim period of the cycle. Condenser water inlet temperatures were regulated by a thermostat on the refrigeration unit, calibrated for a range of 55 to 80F. This control provided good response at mid-range (60 to 75F), however, average temperatures at the condenser water inlet outside this interval were not recorded. An attempt to regulate condenser gas inlet temperatures at 90F +5-degrees between 7/28/80 and 8/7/80 was less effective as condenser water temperatures were increased. Condenser water flowrate was controlled at the - 12 - ------- condenser inlet by a manually actuated valve which pro- duced a good response over the entire range of flowrates. Dryer exhaust gas flowrate was held constant using a constant speed fan which controlled the flow to 306 +6 scfm. Solvent concentrations at both the condenser inlet and dryer exhaust duct locations appeared to vary pro- portionally to all variables previously discussed. In addition, the wet (solvent laden) weight and type (fabric material) of the pre-dryer load had obvious consequences. Experience with field measurements and obvious trends in the response of the uncontrolled process condi- tions to the controlled process conditions discussed previously (Section 1.0) indicate some probable correla- tions. Though several combinations of the data presented in Table 2.0 could be evaluated statistically, full under- standing of the integrity in the measurement methods employed and the components of the dryer system and their basic function dictate the particular conditions and statistical technique chosen to evaluate the correlation. To satisfy the basic objectives of the test program, reclaim efficiency of the solvent recovery tumbler dryer should be evaluated by taking into consideration variables affecting cost, safety, and stack emissions. - 13 - ------- 3.0 PROCESS DESCRIPTION (To be provided by TRW). - 14 - ------- 4.0 LOCATION OF SAMPLING POINTS Figure 4.1 depicts the sampling locations with respect to the process. Velocity traverse data sheets are presented in Appendix C. DRYER EXHAUST DUCT The dryer exhaust duct was accessed through only one 1/2-inch sampling port for velocity traverses, since the other port was inaccessible due to the lack of working space behind the dryer. The port was located 5-feet downstream and 2-feet upstream from the nearest disturb- ance. This provided adequate upstream/downstream dis- tances from disturbances and allowed the determination of a representative velocity profile of the dryer exhaust. Velocity pressures and temperatures were measured at each of six sampling points. Stoddard solvent concentra- tions, exhaust gas temperature, and moisture content were also monitored from this same sampling port. CONDENSER GAS INLET AND OUTLET The condenser inlet was accessed through a single 1/2-inch port carefully positioned in the condenser bypass damper housing. This sampling location was the best site possible, however, it did not provide adequate upstream/ downstream distances as required by EPA Method 1 due to internal dryer parts. Therefore, volumetric flowrates were not measured. Only stoddard solvent concentrations and temperatures were monitored at this location. Temper- atures were recorded at both the condenser inlet and outlet locations. - 15 - ------- Dryer (a) Exhaust duct (b) Condenser water inlet (c) Condenser water outlet (d) Condenser gas inlet (e) Condenser gas outlet (f) Totalizing meter (g) Steam line to heater (h) Reclaim to separator Figure 4.1. Process diagram and sampling locations. ------- CONDENSER WATER INLET AND OUTLET The water temperature was monitored at both loca- tions with an iron constantan (I/C) Type-J thermocouple attached to a calibrated pyrometer and installed in the condenser water lines. Condenser water flowrates were measured by a positive displacement flow totalizing meter positioned upstream of the condenser in conjunction with the elapsed time of the reclaim portion of the cycle. - 17 - ------- 5.0 SAMPLING AND ANALYTICAL PROCEDURES DRYER EXHAUST DUCT Velocity Traverse Exhaust gas velocities were measured in accordance with the procedures outlined in the U.S. Environmental Protection Agency's Standards of Performance for New Stationary Sources, 40CFR60, amended through August 17, 1977, Reference Methods 1, 2, and 4. During a preliminary velocity traverse, the 6-inch duct was divided into 6 equal annular areas at whose mid- points exhaust gas velocities and temperatures were measured, in accordance with EPA, Methods 1 and 2. Velocity pressures were measured at each sampling point using an S-type Pitot tube and inclined 0 to 10-inch water gauge manometer. Temperatures were measured with an iron-constantan (I/C) thermocouple attached to the Pitot tube and to a cali- brated pyrometer. Exhaust gas flowrates were calculated from the single port velocity traverse data. A diagram of the stack cross-section is presented in Figure 5.1. A modified moisture sampling train was run simul- taneously with the velocity traverse. The train consisted of preweighed portions of silica gel' in 18" x 5/8" O.D. glass tubes connected to a calibrated limiting orifice and dry gas meter followed by a vacuum pump (Figure 5.2). - 18 - ------- N 6-inches I.D. This was port not sampled Point 1 2 3 4 5 6 Distance (Inches) 5.7 5.1 4.2 1.8 0.9 0.3 Figure 5.1. Stack cross-section - dryer exhaust duct. - 19 - ------- o I Stack wall Silica gel tube Orifice PVC tubing Inclined manometer Thermometers J L PVC PVC o Dry gas meter Vacuum pump Figure 5.2. Moisture sampling train. ------- FID SYSTEM Stoddard solvent concentrations were measured with a Ratfisch/lPM Model RS5 and Beckman Model 400 hydrocarbon analyzer equipped with a flame ionization detector and recorded continuously on a strip chart recorder (Figure (B\ 5.3). Two 1/8" I.D. heated Teflon^ sample lines (approximately 50-feet in length) were connected to sep- arate 2-way valves and a diaphragm pump (used to purge the lines) to minimize the FID response time when sampling /e\ locations were switched. Heated Teflon^ sample lines then connected these valves to another 2-way valve and filter assembly. This assembly was then connected to a "T" fitting with an on/off toggle valve. The purpose of the toggle valve was to actuate the flow of calibration gases entering the diaphragm pump following the "T" fit- ting. In turn, the pump forced the sample gas and cal- ibration gases into the FID at a regulated pressure and flowrate. Temperatures were measured with a Type-J thermocouple and recorded every 5-minutes for the duration of a cycle. The strip chart and field data appear in Appendix A. CONDENSER GAS INLET AND OUTLET Stoddard solvent concentrations at the condenser gas inlet were measured using the FID system discussed above. Temperatures were measured at both locations with a Type-J thermocouple every 5-minutes for the duration of the cycle - 21 - ------- Chart Recorder FID with sample pump From condenser inlet Purge pump exhaust Calibration gases Purge pump From dryer exhaust duct Figure 5.3. FID system. ------- CONDENSER WATER INLET AND OUTLET Condenser water flowrates in gallons per minute (gpm) were determined by dividing the total volume of water indicated on the totalizing meter by the elapsed time of the reclaim portion of the cycle. RESPONSE FACTOR DETERMINATIONS The instrumental response factor determination incorporated the following components: a compressed gas supply of zero air «1 ppm total hydrocarbon) , monitored and regulated by a dual stage regulator; a stainless steel needle valve; a calibrated dry gas meter equipped with a bimetallic thermometer at the outlet with ports before and after the gas meter to obtain its internal gauge pressure on a 0 to 1.0-inch inclined water manometer; a molecular sieve; a "T" fitting containing a system for sample injection; a Greenburg-Smith midget impinger (50-ml capac- ity) placed in a heated oil bath; a heat traced "T" fit- ting containing a 1/8" O.D. iron-constantan thermocouple; a heat-traced two way-valve which allowed the gas sample ® to pass into a Tedlar bag contained in a heated barrel; a two way valve allowed the gas sample to pass to and from the Tedlar^ bag inside the heated barrel; sample line equipped with a temperature monitor; a Teflon pump equipped with a bimetallic thermometer at the outlet; and a Beckman Model 400 Hydrocarbon Analyzer having a flame ionization detector (FID). Figure 5.4. displays the instrumental response factor calibration system. - 23 - ------- Two-stage regulator Syringe idget impinger /T\Molesieve j filter Hydrocarbon free air cylinder Dry gas meter 0.40ft / Tedlar bag Hot plate to Temperature Readout Pressure Readout Heated Teflon lines Heated metal barrel H.C. Analyzer Recorder Figure 5.4. Calibration system. ------- Two hours were allowed for the entire system to heat up and reach an equilibrium. It was then checked for leaks at all branches of the 3-way valve. The system was then purged with zero air (dilution gas) prior to each sample injection until no measurable amount of hydrocarbon could be detected from the system by the FID. During this purging period, the flowrate of the zero air through the system was set to approximately 0.1 cubic feet per minute (cfm). This flowrate was kept constant for all subsequent response factor calibrations by uti- lizing the on/off valve positioned before the dry gas meter to stop the gas flow. Using the equations and the molecular weight of the stoddard solvent (provided by TRW, Inc.), two approximate volumes of stoddard sample were calculated for giving a response of 50 and 125 ppm propane. Carrying out the calculations, volumes of 1.2 and 2.9 microliters (yl) of stoddard solvent would produce the instrumental response relative to a propane concentration of 50 and 125 ppm, respectively. The hydrocarbon analyzer was calibrated directly by using a 85 ppm propane NBS traceable standard and zero air gases. Initial data was collected from the calibration system components and recorded: - 25 - ------- (1) Dry gas meter volume (2) Dry gas meter outlet temperature (3) Oil bath temperature (4) Impinger outlet temperature (5) Barrel temperature The stoddard sample was injected into the impinger and not more than 20-seconds later the dilution gas was introduced. The dilution gas was allowed to flow by opening the 3-way valve to the 0.4 cubic feet (ft ) (R) TedlaxS' bag and the on/off valve. During dilution, the • internal pressure of the dry gas meter was recorded. After 0.35 ft of diluted gas, it was turned off by closing the on/off valve and then the 3-way valve. The dry gas meter volume was read and recorded. The 3-way valve was opened to allow the TedlarM bag gas sample to flow through to the FID. Auxiliary temperatures of the sample line leading to the Teflon"-' pump and of the sample line at the pump outlet were recorded. Strip chart data was recorded for each stoddard volume calibrated and appear in Appendix B. - 26 - ------- |