Document VjbKQ6L203q4ddxZ9Xv10KKaK

SOUTH CHARLESTON PLANT UNION CARBIDE CORPORATION CHEMICALS AND PLASTICS P. O. BOX 8004. SOUTH CHARLESTON. W. VA. 21303 RECEIVED JUL 9 1974 H. V. HOJPc.< July 5, 1974 Mr. Don R. Goodwin Director, Emission Standards and Engineering Office of Air Quality Planning and Standards U. S. Environmental Protection Agency Research Triangle Park, North Carolina 27711 Dear Mr. Goodwin: Per your request of Union Carbide Corporation, dated May 30, 1974, data on the latexes containing vinyl chloride produced by Jennat ' Corporation is being supplied. Jennat Corporation is a wholly-owned sub sidiary of Union Carbide Corporation operating latex plants at Torrance, California, Tucker, Georgia, And Somerset, New Jersey. The 1974 planned production of latexes containing vinyl chloride is as follows: Plant Torrance, Calif. Tucker, Georgia Somerset, N. J. Somerset, N. J. Product Milt 31 UCAR-1000 UCAR-1000 UCAR-508 Total MM Gross Lbs. 4.0 6.0 3.0 2. 0 15.0 Our total 1974 Latex production of these three locations will be approximately 150 MM pounds, so the vinyl chloride types will be only 10% of the total. Product Vinvl Chloride Monomer Content Product Average Value UCAR-508 60 UCAR-1000 300 Standard Deviation t 40 t 130 Milt 31 600 + 160 Weighted averag has d on pr duction volumes is approximately 400 ppm. N'S-'k ' A-.1 : * * - t-f I* ' UuCcCc 032613 W - Mr. Don R. Goodwin EPA -2 July 5, 1974 (3) Solid Waste Generally, these wastes are disposed of by an authorized commercial contractor to a land fill. Although our solid waste is in the order of 50 cubic yards per plant per month, the vinyl chloride containing portion of this is quite minor - probably only two yards or less. VCM and VC1 polymer content has not been measured, but the VCM would probably be less than 1 ppm, and the VC1 polymer 3-4%, (4) Liquid Waste This is in two categories. (a) Water washes and rinses run 2M to 5M gallons per day per plant into a local municipal sewer for the total production output. VCM and VC1 polymer jcjjntent has not been measured, but we would expect no VCM and no appreciable VCL polymer in these streams^ (b) Scrap latex and cleanings in drums run 80 to 100 drums (4, 000-5, 000 gal.) per month per plant for the total production output. This waste is 50% total solids (50% water), and is disposed of by an authorized commercial contractor to a land fill. Since the VC1 latexes are relatively scrap-free, the portion of this scrap in the total would be even less than 10% (estimate 5%). VCM and VC1 polymer again have not been measured but would probably approximate the concentrations listed in Item (1). The polymer content would vary with the product, but would range from 9 to' 18%. The VCM content would average rAMrppm. j (5) Material Balance Indicates an over-all VCM efficiency of ^%. blance is indicated belowt * A typical -4 -* 4- . 44, ucc 032614 t, ,n* 7 ^ JJL Mr. Don R. Goodwin EPA -3- Input Vinyl Chloride Monomer Output Product (PVC) Product (VCM) Heel returned in Truck (VCM) Fugitive (VCM) Waste (VCM & PVC) Total Accounted July 5, 1974 Pounds 100.00 94. 00 0.30 1. 50 4. 10 0.05 99.95 Attached is a report showing process details, plant layout, and other pertinent information. If we can be of further service, please make your needs known. Very truly yours, , fi' f]< r\ R. N. Wheeler, Jr. RNWJr/ra Attachment. be: Mr. G. J. Hanks, Jr., 511 Mr. H. V. Hooper, Jr.-*-- Dr. A. B. Steele, NYO-28 * Mr. G. J. Triplett, NYO-46 ucc 03261 VINYL CHLORIDE PROCESS AND EMISSION DATA IN LATEX MANUFACTURE by W. P. Miller K. Volkers June 20, 1974 Jennat Corporation, wholly owned subsidiary of Union Carbide Corporation currently utilizes vinyl chloride monomer in the production of latex products at plants in Somerset, New Jersey, Tucker, Georgia, and Torrance, California. All such products are relatively new and total production volume is small. The latexes produced are interpolymers containing less than 50% polymerized vinyl chloride. The attached data is a best estimate from a limited base of information and experience. The data supplied respond to the general request to Mr. R. N. Wheeler by Mr. Don R. Goodwin of May 30, 1974 (copy attached). Process and Emission Data Process Description (See Fig. 1): 1. Vinyl chloride monomer is received by tank truck and/or tank car and is pumped to underground storage tanks. 2. Vinyl chloride and other monomers are trans ferred from storage and metered into a monomer mix tank where they are mixed. 3. Warm water, catalysts and stabilizer solution are pumped to the reactor from mix tanks. 4. An initial monomer charge is added to the reactor and the reaction initiated. 5. At specified temperatures, continuous monomer and catalyst feeds are started and continued until all the monomers are fed. 6. A cook-out time is followed by addition of "post-add" catalyst and a further cook-out period to reduce residual monomer. 7. The product is partially cooled and transferred to a cooling tank. 8. The product is then transferred directly to storage or via an intermediate bl nding tank. 9. The product is pumped to tank trucks for d livery to custom rs, ^ 032616 -2 - A. process Plow Sheet: Attached Figure 1 shows all major equipment and the amount of product produced. B. Emissions: No known continuous emissions of vinyl chloride exist. All intermittent sources of emission are classified as fugitive emissions and discussed in Section D. C. Control Devices - No vinyl chloride emission control devices are currently in service. The known fugitive emission sources defined in items 2, 3, 4, 5, 6 of Section D could possibly be collected and incinerated to reduce their concentration to 10 ppm. In the absence of any specific data on this subject,no further discussion of this possi bility is indicated. D. Fugitive Emissions - Known sources resulting from line or vessel blow down, reactor ventilation, storage and transfer operations within the plants are estimated and summarized below. Please refer to the simplified process flow sheet (Fig. 1). Unknown sources of fugitive emissions; for example, from leaks through valve stems, pumps seals and miscellaneous sources, are believed to be of a similar order of magnitude as the previously estimated known sources. E. Emission Control Procedures - Several process design features have been incorporated into the plants to eliminate or control emissions: 1) Storage and use of vinyl chloride in closed vessels designed to withstand the pressure of the fluid without venting during transfers, 2) Special mechanical seal design on transfer pumps to preclude leakage of vinyl chloride liquid or vapor in case of a primary seal failure. These design features are very effective in controlling emissions. A process design feature that could be used to further reduce losses due to hose disconnections when unloading tank car or tank trucks would be the use of dry break hose disconnects. These devices effectively eliminate the small pockets of trapp d liquid or vapor created by hose connections. ucc 032617 Known Fugitive Vinyl Chloride (VCM) Emissions from Latex Manufacturing Points (See Fig. 1) Description Emission*3 Composition Amount of VCMa Lbs/Wet Lb. of Latex Produced Cone. ppm 1. Disconnect vinyl chloride Pure VCM tank truck or tank car un loading 2. Monomer mix tank blowdown between product changes VCM contain ing mixture 3. Reactor blowdown between product changes or clean ing cycles 4. Latex transfer to cooling tank on each batch 5. Latex transfer to blend ing tank on each batch VCM contain- , ing mixture ' VCM contain ing mixture VCM contain ing mixture 6. Lat x transfer to storage tank on each batch VCM contain ing mixture 7. Tank truck loading on each shipment latex VCM contain ing mixture 1.6 x 10-6 1.5 x 10"4 1.6 x 10"5 1 x 10-6 1 x 10"6 1 x 10"6 2 x 10-8 Pure (100%) 260,000 10,000 600 600 600 10 Lbs/Hr 12 Minutes Time Duration of Each Emission Period 5 110 10 8 30 0.04 0.04 0.04 0.0007 35 35 w 35 35 a. Points 1, 2 were estimated from line or tank sizes and operating procedures. Points 3, 4, 5, 6, and 7 were estimated from a minimum number of concentration measurements coupled with transferred amounts. b. Complete composition of emission is unknown for VCM containing mixtures. 032618 -4 - Control and Economic Data A. Investment Cost No specific emission control devices are present in the existing plants to control vinyl chloride emissions and therefore investment cost is zero. Special emission control process design features listed above were incorporated into the original plant design and did not result in significant additional investment or installation costs. General commitment for Installation of suitable devices when and if needed has been obtained. B. Operating and Maintenance Costs The operating and maintenance costs are not affected by the process design features and operating procedures for vinyl chloride handling in this process. Ambient Air Data A. Results of a minimum number of analyses are shown below and on the simplified plant layout (Pig. 2) attached. All analyses have been performed by the Union Carbide carbon adsorption tube method 38C-9C1-R2. Weather conditions at the time of measurement are not known. Sample Points (See Fig. 2) Ambient Air ppm VCM a. Fence line near monomer <1 b. Monomer storage area during VCM unloading A to-100 -fO c. Reactor area in Process Bldg. < 1 hf d. Product filtration area in Process Bldg. <1 e. Product storage area <1 f. Laboratory area <1 g. Front of Office Bldg. <1 B, Information necessary to estimate ambient air con centrations for the fugitive emissions cited previously are shown in the attached table. ucc 032619 Information Necessary to Estimate Ambient Air Concentrations1 SOURCE Tank Truck, Tank Car Unloading Hose Disconnects Hfx Tank B1 cwdown x. Reactor Blowdown J Cooling Tank Venting < JBlend Tank Venting_/ Storage Tank Venting Tank Truck Loading STACK HEIGHT FEET CONFIGURATION DIAMETER IN. GAS FLOW RATE CFH TEMPERA TURE "C H.A. Open N.A. N.A. Ambient IS . 3 30 26 12 Gooseneck Gooseneck Gooseneck Gooseneck Gooseneck Truck Vent 3 45 I t 100 2l 10 2^ \550 2 10 20 10 Ambient 30 `A ! 60 j 80/_ 30 30 GASl VELOCITY FT/SEC N.A. 15 306 7.6 COMPOSITION % VCI BY VOLUME . hi 100 . j J6 & 1b .06 e 418 .06 O 7.6 .06 < .08 .001' 1. Exit gas velocities to the atmosphere from vents - . vjt?f ' Jt,1', i\ ucc 032622