(REPORT NUMBER) AIR POLLUTION EMISSION TEST Bensenville, Illinoi: U. S. ENVIRONMENTAL PROTECTION AGENCY Office of Air and Water Programs Office of Air Quality Planning and Standards Emission Standards and Engineering Division Emission Measurement Branch Research Triangle Park, N. C. 27711 ------- Emission Testing Report". EMB Project No. 74-VAR-3 STRESS EN-REUTER Bensenville, Illinois Project Officer: Carl D. Bell Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park, North Carolina 27711 January 1974 ------- TABLE OF CONTENTS Page Number(s) I. INTRODUCTION 1 II. SUMMARY AMD DISCUSSION OF RESULTS 2-5 TABLE I - Inlet to Catalytic Afterburner .... 4 TABLE II - Outlet to Catalytic Afterburner ... 5 III. PROCESS DESCRIPTION 6-9 Figure 1 - Process Equipment 7 TABLE III - Process Data 8 TABLE IV - Batch Process 9 IV. CONTROL EQUIPMENT DESCRIPTION 10 V. SAMPLING AND ANALYTICAL PROCEDURES 11-13 Figure 2 - Samnling Equipment. 13 VI. SAMPLE PORT LOCATION ..._... 14 VII. APPENDICES •• A . . 15-21 FIELD DATA SHEETS 16-17 SAMPLE CALCULATIONS 18-19 TEST LOG , . . . 20 TEST PARTICIPANTS 21 ------- INTRODUCTION Under the Clean Air Act, as amended, the Environmental Protection Agency is given the responsibility of establishing performance standards for stationary sources that contribute significantly to air pollution. A performance standard is established using the best emission reduction systems which have been shown to be operable and economically feasible. In order that realistic performance standards can be set, accurate data on pollutant emissions must be gathered from the stationary source under consideration. The Stressen-Reuter Paint Company in Bensenville, Illinois, was considered a well-controlled stationary source in the paint and varnish industry and was, therefore, selected by the Office of Air Quality Planning and Standards (OAQPS) for the emission testing program. The emission testing was conducted at the inlet and outlet of a catalytic afterburner which served as a pollution control device for a cooking kettle at the plant. These sites were sampled for total hydro- carbons, NO , methane, and 0?. The sampling was conducted by personnel /\ t. from Scott Research Laboratories and the Environmental Protection Agency on September 26, 1973. 1 ------- SUMMARY AND DISCUSSION OF RESULTS Because of an unforeseen and unfortunate series of occurrences, the originally-planned three day test ended up being only.one working day making only one test run on only one batch of Kettle production. This series is as follows: 1. Testing equipment was delayed in shipping by almost two working days. All of day was lost. 2. Plant personnel advised that there will be no third day. The one and only batch would be the evening of the second day. 3. Outlet stack was actually two concentric stacks for recycling of the spent afterburner gas. The directions of flow in the two concentric stacks were in the opposite direction. 4. The equipment testing the inlet to the afterburner was rendered non-functional early on in the test. An attempt was made to use one set of equipment to test both sides of the control device. 5. Halfway through the first outlet traverse, the actual exit of the stack was tested to try to gain a more representative flow rate, as the only port was made at the interface of the opposite flows. Tables I and II seem to show some increases in some gas constituents across the afterburner. This would definitely indicate problems with the testing method. Some error could have been due to the switching back and forth from inlet to outlet, using one set of instruments. Other reasons for the difference might include the fact that the numbers do not represent the exact same times but are staggered back and forth. At best, any conclusions derived from examination of this data should be considered highly questionable. ------- It is regretful that no more than a rough idea of gaseous concentrations of alkyd batch process can be derived from these four days. It was totally a misfortunate experience, serving as an excellent example on what to be aware of on a presurvey and test. The concentration values for the inlet and outlet may serve some value as background information on this type of varnish process. However, in evaluation of this data, one should keep in mind the technical problems encountered and, therefore, the questionable validity of the results. ------- TABLE I Stressen-Reuter Paint & Varnish Inlet to Catalytic Afterburner ppm (by vol) Time THC NQ2 NO..2 CH4 ' C02 ' Percent 02 ' 16:45 17:00 17:15 17:30 17:45 18:00 19:00 19:45 20:45 20:50 21:55 22:45 22:50 606 1393 1725 2427 1995 576 592 268 140 A 0.2 0.3 0 0 1.3 1.4 0.4 0 0 60 40 111 586 2011 1613 20.7 20.7 21.2 1 Wet basis 2 Dry basis 3 Grab samples takai at each specific time. ------- TABLE II Stressen-Reuter Paint & Varnish Outlet to Catalytic Afterburner* ppm (by vol) 1 2 2.23 23 23 Time THC NO NOM CH. ' C09 ' Percent Og ' X *r t c. 75 1842 18.2 18.5 18.5 90 4«9 18.2 18.2 18.5 18.6 18.5 134 39118 18.8 18.4 *Probe moved from port to actual stack outlet due to inner stack with opposite flow at port. 1 Wet basis 2 Dry basis 3 Grab samples taken at each specific time. 18:45 19:00 19:15 20:00 20:15 20:30* 20:45 21:00 21:30 21:45 22:15 22:20 22:30 22:45 23:00 360 376 384 344 328 296 288 232 240 224 212 — 244 248 240 6.0 4.5 3.5 1.2 4.1 4.1 3.6 5.4 5.8 7.4 4.9 7.2 6.2 5.6 7.3 7.4 7.2 7.0 8.4 ------- PROCESS DESCRIPTION Various ingredients are mixed and cooked in a batch fashion and in a closed, but not sealed, kettle for approximately 6-8 hours at 400-500°F. For this alkyd-type varnish we might expect around 160 pounds of hydrocarbon emission per ton of product. Although attempts were made to do velocity checks and assign flow to the stack, an actual pollutant mass rate could not be accurately assigned because of the types of problems which were encountered. This cooking kettle is heated by a circulating hot oil jacket. Process temperatures were taken at the inlet and outlet of this oil jacket as seen in Table iv and Figure 1. The ingredients mixed in this batch are seen in Table ------- 7 ~~7"77~~7 a»r g OO. O»'lufi«i fiir 7 7—7 Part FdfJ \Scrubloer i««, ------- TABLE III Stressen-Reuter Company Bensenville, Illinois Process Data Product RV-1637 - an alkyd base for ink or paint Batch #BW-66 Prepared in Kettle #K-4 Total Batch Wt.: 19,831 pounds Net Weight (percent): 18,903 Formula Ingredient Code Linseed Oil 0217 Trimethylolpropane S-238 Lithium Acetate Dihydrate D-126 Isophthalic Acid D-120 Phthalic Anhydride D-21 Triphenyl Phosphate S-90 Antifoam Manufacturer Tenneco Lithoca Amoco Tenneco Monsanto TOTAL Weight, Ibs. 12,265 3,000 5 3,586 941 28 6 19,831 The linseed oil and trimethylolpropane were added at 0630 COT and brought up to 320°F and 1330 CDT. The following table describes the remaining steps in process. Emissions were measured between 1645 and 2300 CDT. *A polyester resin type of varnish binder. 8 ------- TABLE IV Batch Process Time [COT] 1200 1215 1330 1530 1730 1745 1800 1815 1830 1845 1900 1915 1930 1945 2045 2100 2130 2145 2200 2215 2230 2245 2300 Batch Temp. -80 100 320 455 480 480 480 455 435 425 430 440 440 450 475 480 485 490 495 500 500 505 510 Hot Oil Inlet Temperature (5F) 355 525 535 540 545 540 535 535 530 535 535 535 535 535 535 535 540 540 542 545 550 547 Hot Oil Outlet Temperature (5F) 325 500 515 525 530 525 515 515 505 510 515 515 510 515 515 520 520 522 525 528 530 530 Activity Increase temp to 32C Increase temp to 32C Add S-238 Add D-126 Add D-120, D-21, and S-90 Increase temp. End emission samplir ------- CONTROL EQUIPMENT DESCRIPTION As the reaction takes place in the kettle, gases are given up and vented into a scrubbing chamber, pulled by a fan. From there, they are channeled to the inner stack of the afterburner which is going down into the afterburner. The opening at the top of the inner stack, level with the sampling port, allows a draft of recycled exit gas to go back through the catalyst. These gases then go through a preheater to 800°F and pass through the series of platinum plates and are combusted in the afterburner. Combustion air is added here as well as the dilution air near the exit of the stack (see Figure 1). The gases are then pumped out the stack and released. 10 ------- SAMPLING AND ANALYTICAL PROCEDURES The gases emitted from this process were measured with several continuous monitor type devices (see figure 2). The hydrocarbons at both the inlet and outlet were measured with a Scott Model 215 heated flame ionization detector. The hydrocarbon analyzer was spanned with a propane in air standard. The flame formed when pure hydrogen or hydrogen diluted with are inert gas, burns in air contains an almost negligible number of ions. Introduction of mere traces of hydrocarbons into such a flame, however, produces a large amount of ionization. This effect is the basis of the flame ionization method. The ppm was printed directly onto a stlrip chart. A Scott Model 125 chemiluminescence analyzer and thermal converter were used for nitric oxide and total oxides of nitrogen. EPA Method 7 was not used due to the low concentrations of 190 and the need for a A 10-hour continuous sample. PPM for both NO awd NO were printed A directly onto a strip chart. A Scott Model 150 paramagnetic analyzer was used for oxygen. The analyzer was spanned with an oxygen in nitrogen standard and the readout was in percent. Carbon monoxide, carbon dioxide, and oxypn were measured using an orsat analyzer according to Method 3 of the December 23, 1971, Federal Register. Samples were taken about ewery two hours. 11 ------- Grab samples were collected in glass flasks and returned to Scott Research Laboratories for methane analysis. The analysis was performed on a Perkin-Elmer Model 900 gas chromatograph using a molecular seive column at 120°C. The calibration gas used was a methane in air standard. As a check against the very low Orsat readings, CCL was also analyzed in the flask samples by gas chromatography using a silica gel column and a thermal conductivity detector. Inlet and outlet stack velocity patterns and temperatures were measured with an "S" type pi tot tube and an fron-constanton thermocouple and calibrated pyrometer. Moisture was not tested but was assumed to be 5% at the inlet and 15% at the outlet, based on previous experience. As will be noticed in Figure 2, the FID and 02 analyzer were working on a wet basis. Due to the trap before the chemiluminescence meter, nitrogen constituents were given on a dry basis, as was considered the Orsat analysis. These corresponding concentrations are seen in Tables I and II. 12 ------- Kettle JD _O S- o Inlet N, Vacuum Pump Pre- leater Catalytic Afterburner NO Thermal Converter To atmosphere ^Outlet Heated line Filter holder (heated) Flame [lonization! Detector C& 1 Figure 2. Sampling Equipment Stressen-Reuter, Bensenville, Illinois Knalyzer t ------- V SAMPLING PORT LOCATION The inlet sampling point was located about two feet after the water scrubber in an eight-inch by eight-inch rectangular duct. The velocity at this point was constant but somewhat lower than expected. The outlet port was locfv^d in the side of the exhaust stack. The top of the inner stack was located level with this sampling port creating great difficulty in measuring the correct velocity because the flow in the inner stack was strongly downward. Midway through the test, the probe was moved to the stack exit where it was felt more reliable velocity and gas readings could be obtained. The outer stack had a diameter of 34" while the inner stack's diameter was 18". The stack exit had a diameter of 27". 14 ------- APPENDICES 15 ------- FIELD DATA 16 ------- ORSAT ANALYSES Run 1 Run 2 • , Actual Date: 9/26/73 Sampling Time (24-hr Clock): 1840 CDT Sampling Location: Fume Scrubber Inlet Sample Type (Bag, Integrated, Continuous): Tedlar Bag Analytical Method: Orsat Date: 9/26/73 Sampling Time (24-hr Clock): 1904 CDT Sampling Location: Fume Scrubber Outlet _j Date: 9/26/73 Sampling Time (24-hr Clock) : 2055 CDT Sampling Location: Fume Scrubber - Ttil p t* J.11 JL C I* Sample Type (Bag, Integrated, Tedlar Bag Analytical Method: Orsat Date: 9/26/73 Continuous) : Sampling Time (24-hr Clock) : . 2055 CDT Sampling Location: Fume Scrubber - Stack Outlet Sample Type (Bag, Integrated, Continuous): Tedlar Bag Analytical Method: Orsat Gas co2 02 (Net is actual 02 Reading Minus Actual CO Reading) CO (Net is Actual CO Reading Minus Actual 02 Reading) C°2 02 (Net is Actual 0- Reading Minus Actual CO Reading) CO (Net is Actual CO Reading Minus Actual 0 Reading) CO £» 0_ (Net is Actual 0. 2 '2 Reading Minus Actual CO Reading) CO (Net is Actual CO Reading Minus Actual 02 Reading) CO £- 0 (Net is Actual 0 Reading Minus Actual C02 Reading) CO (Net is Actual CO Reading Minus Actual 0 Reading) Reading 99 79 79 99 79 79 99 79 79 99 79 79 .8 .4 .4 .9 .3 .2 .8 .3 .3 .6 .4 .2 Net 0.2 20.4 0 0.1 20.6 0.1 0.2 20.5 0 0.4 20.2 0.2. Actual Reading 99 79 79 99 79 79 99 79 79 99 79 79 .7 .2 .2 .7 .2 . .1 .8 .3 .3 .4 .3 .1 . Net 0.3 20.5 0 0.3 20.5 0.1 0.2 20.5 0 0.6 20.1 0.2 Avg. Net Volume 0. 20. 0 0. 20. 0. 0. 20. 0 0. 20. 0. 25 45 2 55 1 2 5 5 15 2 ------- SAMPLE CALCULATIONS 18 ------- Sample Calculations Orsat - MM Inlet Outlet 18:40 CDT 28.86 19:04 28.85 20:55 28.85 20:55 28.89 AVERAGE 28.85 28.87 Ml* d Ac s MW P .95 64" 28.30 30.58 *Assumed .85 653" (port) 572" (stack exit) 27.23 30.65 19 ------- TEST LOG September 25, 1973 September 26. 1973 8:00 a.m. 10:00 11:45 12:00 4:25 p.m. 5:30 6:00 6:30 7:00 8:30 10:15 11:00 12:00 Equipment lost in shipment so test delayed by one day. Pick up equipment and take to plant. Equipment unloaded at plant. We find out that plant is getting ready to start the final batch of the week so we will set up and test as long as possible. Plant begins firing afterburner. Begin heating kettle - 4 to 390°F. Still not sampling due to many equipment problems and a shortage of tubing. Inlet is ready but outlet is still causing problems. Plant is adding an ingredient to kettle causing an increase in flow and much particulate. Inlet is plugged and will not be serviceable for the remainder of the test. Hope to use outlet equipment for the inlet also. Outlet is running okay. Moved probe to stack exit where hopefully opposite flow of inner stack will not affect numbers. Kettle reaches 500°F. IMothing else will be added or changed in the kettle. Sampling completed. Leave plant. 20 ------- TEST PARTICIPANTS Scott Research Tony Souza - Crew Chief Four (4) technicians Environmental Protection Agency Doug Bell - Project Officer 21 ------- |