Document 990ba0VN4zgJ5m35nBbqo6b47

1 r <r : v 7 7' svEPA United States ' Environmental Protection Agency Office of Enforcement Office of General Enforcement Washington, DC 20460 EPA Contract No. 68-01-4141 May 1978 National Emission Standards for Hazardous Air Pollutants Inspection Manual for Vinyl Chloride OF U. SAL - i'"' n A f` % i fi I ). i n r? NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR POLLUTANTS INSPECTION MANUAL for VINYL CHLORIDE EPA Contract No. 68-01-4141 RTI Project No. 41U-3172-2 EPA Project Officer John R. Busik 3 1272 0U445 3112 Prepared for U. S. ENVIRONMENTAL PROTECTION AGENCY Office of Enforcement Office of General Enforcement Washington, D. C. 20460 May 1978 ACKNOWLEDGMENT This Inspection Manual was prepared by Mr. Michael F. Lamorte of the Research Triangle Institute. Project Officer for the Environmental Protection Agency - Enforcement Division was Mr. John R. Busik'. The Task Manager was Mr. Richard Biondi assisted by Ms. Libby Scopino. The author appreciates the many contributions made by Mr. Biondi and Ms. Scopino during the preparation of this Manual. The author also appreciates the assistance of Surveillance and Analysis and of Enforcement Division personnel of Region 6 in applying the Inspection Forms under field conditions. In addition, discussions with Mr. Ben Carpenter and Dr. Forest Mixon of the Research Triangle Institute were very helpful. Mr. R. N. Wheeler, Jr., of the Union Carbide Corporation, provided information of the more recent solvent polymerization process. Finally, many thanks are in order to Mr. John R. Lawrence, The Society of the Plastics Industry, Inc., and to industry representatives for the hospitality extended to the author during the numerous plant site visits. TABLE OF CONTENTS ACKNOWLEDGMENT LIST OF FIGURES LIST OF TABLES LIST OF INSPECTION FORMS LIST OF CHEMICAL FORMULAS LIST OF ABBREVIATIONS 1.0 INTRODUCTION 1.1 Background 1.2 Authority 1.3 Applicability 1.4 Definitions 2.0 EDC, VC AND PVC INDUSTRIES 2.1 Ethylene dichloride (EDC) 2.2 Vinyl chloride (VC) 2.3 Polyvinyl chloride (PVC) 3.0 PROCESS FLOW DESCRIPTION AND EMISSION POINT IDENTIFICATION 3.1 Ethylene dichloride--oxychlorination 3.2 Vinyl chloride 3.2.1 Hydrochlorination ofacetylene 3.2.2 Dehydrochlorination ofethylene dichloride Page ii v ix x xi xii 1 1 2 2 3 6 6 7 8 9 9 12 12 16 iii 0 0 -vi r/i O / TABLE OF CONTENTS (Continued) 3.3 Polyvinyl chloride 3.3.1 Suspension polymerization 3.3.2 Emulsion (i.e., dispersion)polymerization 3.3.3 Latex dispersion polymerization 3.3.4 Bulk polymerization * 3.3.5 Solvent polymerization 3.4 Clarifying note on the balanced oxychlorinationdehydrochlorination process 3.5 Photographs of equipment 4.0 LEAK DETECTION MONITORING INSTRUMENTATION, RECORDS AND REPORTS 4.1 Leak detection monitoring instrumentation 4.2 Leak detection monitoring recordkeeping 4.3 Routine leak detection and relief discharge recordkeeping 5.0 INSPECTOR SAFETY ' 6.0 INSPECTION PROCEDURES AND INSPECTION FORMS 6.1 Summary of compliance status 6.2 Checklist 6.3 On review of records 6.4 Pre-test equipment checklist for stack emission test 6.5 Equipment checklist for vinyl chloride concentration in inprocess wastewater, resin, slurry, wet cake and latex samples APPENDIX A: Mean value calculation APPENDIX B: National Emission Standards for Hazardous Air Pollutants - Standard for Vinyl'Chloride, October 21, 1976 APPENDIX C: National Emission Standards for Hazardous Air Pollutants - Standard for Vinyl Chloride: Cor rections and Amendments, dune 7, 1977 REFERENCES 18 18 23 26 26 29 36 38 55 55 59 59 60 61 62 62 62 63 63 86 87 103 109 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 LIST OF FIGURES Ethylene dichloride process flow diagram using * oxychlorination which involves the reaction of oxygen and hydrogen chloride with ethylene. Vinyl chloride monomer process flow diagram using hydrochlorination of acetylene. Vinyl chloride monomer process flow diagram using dehydrochlorination of ethylene dichloride. Polyvinyl chloride process flow diagram using suspension polymerization. Polyvinyl chloride process flow diagram using emulsion (i.e., dispersion) polymerization. Polyvinyl chloride latex.process flow diagram using emulsion (i.e., dispersion) polymerization. Polyvinyl chloride process flow diagram using bulk polymerization. Polyvinyl chloride process flow diagram using solvent polymerization. More recent polyvinyl chloride process flow diagram using solvent polymerization. Block diagram of a balanced oxychlorinationdehydrochlorination process. Large capacity VC reactor using the oxychlorination process is located in the tall vessel. The oxychlori nation manual vent is the open ended pipe, rising above the top of the reactor on the left side. Page 10 13 17 20 24 27 30 32 33 37 40 BAi... 0000498 v LIST OF FIGURES (continued) 3.12 The tops of medium capacity, side-by-side PVC reactors using the emulsion process may be seen. Individual RD/SRV and manual vents may also be seen mounted on top of each reactor. 3.13 Small size PVC reactor with cover removed to clean reactor in preparation for next polymerization run. Prior to removing cover the PVC and water solution were removed and transferred to stripper. "Elephant trunk" is placed through the opening to vent vinyl chloride gas through "trunk" to the recovery system in order to be in compliance with emissions standard reactor opening loss. This type reactor is used in the suspension, emulsion and latex polymerization processes. 3.14 Medium capacity side-by-side PVC reactors using solvent process. Individual RD/SRV and manual vents connected to vinyl chloride recovery system may also be seen. Lower portion of reactors comprise the heating elements to raise the contents to the tem perature required for polymerization. 3.15 The top of medium capacity, side-by-side PVC reactors using solvent process. Foreground shows a motor valve for emergency venting through to the VC monomer recovery system. 3.16 The top of a small capacity PVC stripper vessel used in the suspension, emulsion and latex processes, showing an SRV. 3.17 A stripper column is shown that removes VC from a PVC-varnish solution resulting from the solvent polymerization process. 3.18 A typical wastewater stripper column is shown that may be found in EDC, VC and PVC plants. 3.19 The top of an EDC storage tank is shown with its vent. 3.20 Cylindrical, side-by-side, above ground VC storage tanks are shown. RD/SRV vents may be seen on pipe rack above tanks. VI 40 41 41 42 42 43 43 44 44 !' '! 1.. (/ () (jO i::- LIST OF FIGURES (continued) 3.21 Spherical, above ground vinyl chloride storage tanks. RD/SRV vents may be seen perched on top of sphere. 3.22 Underground VC storage tank area. 3.23 Wash water stripper storage tank. 3.24 Upper portion of tar storage tank shown with vent. 3.25 PVC suspension resin rotary dryer anddust collector. Drying is accomplished by the application of heat and rotary action. 3.26 PVC dispersion resin spray drying takes place in the cylindrical building. Large diameter feed pipe, seen at the left of the dryer, carries drying air to the top of the dryer. Housing for the atomiza tion system is perched at topof dryer. 3.27 An EDC light ends distillation column is shown with its RD/SRV vent. 3.28 VCM column condenser and condenser vent motor valve. 3.29 The top of an EDC light ends column reflex accumulator showing the RD/SRV vent. 3.30 In the foreground the light ends dryer regeneration liquid knockout vessel is shown in balanced EDC-VC plant. 3.31 An EDC finishing column with the RD/SRV and its vent mounted on the top. (In some installations the finishing column performs the function of the light ends distillation column while in others it en compasses the functions of both the light and heavy ends columns.) 3.32 Foreground shows the insulated vent piping and motor valve from the reactor refrigerator condenser vessel in a balanced EDC osychlorination plant. 45 45 46 46 47 47 48 48 49 49 50 50 LIST OF FIGURES (continued) 3.33 Typical RD/SRV assembly with the vents from each connected to a manifold of the vent system. 3.34 Dual RD/SRV vents mounted at top of spherical vinyl chloride storage tank. 3.35 Typical pump and double mechanical seal. 3.36 Top connections on railroad tank car shown with flexible hose attached for VC loading. Smaller diameter flexible hose in the foreground is connected to recovery system. 3.37 Railroad tank car loading platform shown with pipe rack support for flexible hose VC feed and recovery system in VC plant. 3.38 Oxychlorination vent scrubber stack. 3.39 Incinerator and stacks of a PVC plant showing the platform (center stack) on which stack samples are taken to determine VC emission concentration. 4.1 Schematic diagram of continuous monitoring system for vinyl chloride emissions. 51 51 52 52 53 53 54 58 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3-8 3.9 3.10 3.11 LIST OF TABLES The potential emission points identified by the keys in Figure 3.1 in the oxychlorination process for ethylene dichloride. Hydrochlorination of acetylene product gases and their boiling point temperatures. The potential emission points identified by the keys in Figure 3.2 in the hydrochlorination of acetylene for vinyl chloride monomer. The potential emission points identified by the keys in Figure 3.3 in the dehydrochlorination of ethylene dichloride for vinyl chloride monomer. The potential emission points identified by the keys in Figure 3.4 in the suspension polymerization process for polyvinyl chloride. The potential emission points identified by the keys in Figure 3.5 in the emulsion (i.e., dispersion) polymerization process for polyvinyl chloride. The potential emission points identified by the keys in Figure 3.6 in the emulsion (i.e., dispersion) polymerization process for polyvinyl chloride latex. The potential emission points identified by the keys in Figure 3.7 in the bulk polymerization process for polyvinyl chloride resin-. The potential emission points identified by the keys in Figure 3.8 in the solvent polymerization process for polyvinyl chloride and copolymers. The potential emission points identified by the keys in Figure 3.9 in the solvent polymerization process for polyvinyl chloride and copolymers. Figure number of photographs and corresponding equipment category. Page 11 14 15 19 22 25 28 31 34 35 39 SAL. 00004S8/B IX LIST OF INSPECTION FORMS Summary of Compliance Status Page 64 Checklist 65 On Review of Records 74 Pre-Test Equipment Checklist for Stack Emission Test 80 Equipment Checklist for Vinyl Chloride Concentration In Inprocess Wastewater, Resin, Slurry,Wet Cake and Latex Samples 83 04 9 X LIST OF CHEMICAL FORMULAS Name Acetylene Chlorine Ethylene Ethylene Dichloride Hydrogen Chloride Oxygen Polyvinyl Chloride Vinyl Chloride Monomer Wa ter Formula HC = CH C12 CH2 = ch2 C1CH2 - CH2C1 HCl 2 -f H2C = CHC1 H2C = CHCl H2 xi LIST OF ABBREVIATIONS BOD COD EDC ERA Eq. FR HCL NESHAP PVC RO SOP SRV TSS VAC VC V CM Biological oxygen demand Chemical oxygen demand Ethylene dichloride U.S. Environmental Protection Agency Equation U.S. Federal Register Hydrogen chloride National Emission Standards for Hazardous Air Pollutants Polyvinyl chloride Rupture disc Standard operating procedure Safety relief valve Total suspended solids Vinyl acetate comonomer Vinyl chloride Vinyl chloride monomer 1.0 INTRODUCTION 1.1 BACKGROUND Pursuant to Section 112 of the Clean Air Act (42-U.S.C. 1857). the Administrator of the U.S. Environmental Protection Agency (EPA) has added vinyl chloride to the list of hazardous air pollutants (40 FR 59477) and established a national emission standard (40 FR 59532) for facilities which manufacture ethylene dichloride, vinyl chloride, and/or polyvinyl chloride. The National Emission Standards for Hazardous Air Pollutants (NESHAPs) regulations are applicable to plants producing the following: ethylene dichloride by the reaction of oxygen and hydrogen chloride with ethylene; vinyl chloride by any process; and one or more polymers containing any fraction of polymerized vinyl chloride. The regulations do not apply to equipment used in research and development if the reactor used to polymerize the vinyl chloride processed in the equipment has a capacity of no more than 0.19 m (50 gal). Research and development facilities containing a polymerization reactor capacity greater than 0.19 m (50 gal) but no more than 4.07 m (1100 gal) are exempt from all parts of the regulations except the 10 ppm limit. The proposed emission standards for existing and new plants were advanced in the Federal Register, December 24, 1975 for vinyl chloride in plants manufacturing ethylene dichloride, vinyl chloride, and/or polyvinyl chloride. Final standards (41-FR-October 21, 1976, pages 46560-46573) became effective October 21, 1976 and apply to existing and new plants [1]. EPA decided to regulate vinyl chloride because it has been implicated as the causal agent of angiosarcoma (a rare form of liver cancer) and other serious disorders, both carcinogenic and non-carcinogenic, in people subjected to occupational exposure and in laboratory animals exposed to controlled concentrations of vinyl chloride [2]. Reasonable extrapolations from these findings cause concern that vinyl chloride may cause or contribute to the same or similar disorders at present ambient concentration levels. The purpose of the standard is to set limits of vinyl chloride 1 emissions from all known process and fugitive emission sources in ethy lene dichloride, vinyl chloride, and/or polyvinyl chloride plants. This will have the effect of furthering the protection of public health by minimizing the health risks to people living in the vicinity of these plants and to any additional people who are exposed as a result of new construction [3]. This Inspection Manual contains the guidelines for the benefit of and the standardized procedures to be followed by EPA field inspectors or their designated representatives. In conjunction with the operator's testing and monitoring results and the required recording and recordkeep ing of these results, the basic enforcement tools are readily available to properly trained field inspectors. The degree to which this portion of the NESHAPs program is successful depends critically on the effective ness and efficiency with which inspectors conduct field inspections. 1.2 AUTHORITY Authority for promulgation of the NESHAPs standards and regulations of Air pollutants is contained in Section 112 of the Clean Air Act (42 U.S.C. 1857). It directs the Administrator of the U.S. Environmental Protection Agency to establish emission standards and regulations for hazardous air pollutants (40 FR 59477), and to maintain a current listing of these pollutants. 1.3 APPLICABILITY The applicability of the NESHAPs standards and regulations is specified with respect to the function of the facility, product and process by which the product is produced. There are no exemptions to the NESHAPs vinyl chloride emissions standards and regulations for production plants which employ reactors of any capacity to produce one or more of the following: ethylene dichlo ride by reaction of oxygen and hydrogen chloride with ethylene; vinyl chloride by any process; and one or more polymers containing any fraction of polymerized vinyl chloride [1]. Equipment employed in research and development of the polymerization of vinyl chloride for which the reactor capacity is not greater than 0.19 2 <-\Y- (50 gal) is not subject in any way to the vinyl chloride emission standards and regulations [1]. Equipment employed in research and development of the polymerization 3 of vinyl chloride for which the reactor capacity is greater than 0.19 m (50 gal) and less than 4.07 m3 (1100 gal) is subject only to the 10 ppm vinyl chloride emission limit into the atmosphere from each reactor, stripper, monomer recovery system, and mixing, weighing, and holding containers [1]. 1.4 DEFINITIONS The definitions of terms used in this Inspection Manual are precise and this precision is required to conduct inspections properly. The terms requiring this precision are defined below [1], (a) "Ethylene dichloride plant" includes any plant which produces ethylene dichloride by reaction of oxygen and hydrogen chloride with ethylene. (b) "Vinyl chloride plant" inpludes any plant which produces vinyl chloride by any process. (c) "Polyvinyl chloride plant" includes any plant where vinyl chloride alone or in combination with other materials is polymerized. (d) "SIiP gauge" means a gauge that has a probe that moves through the gas/liquid interface in a storage or transfer vessel and indicates the level of vinyl chloride in the vessel by the physical state of the material the gauge discharges. (e) "Type of resin" means the broad classification of resin referring to the basic manufacturing process for producing that resin, including, but not limited to, the suspension, dispersion, latex, bulk, and solution processes. (f) "Grade of resin1* means the subdivision of resin classification that describes it as a unique resin, i.e., the most exact description of a resin with no further subdivision. (g) "Dispersion resin" means a resin manufactured in such a way as to form fluid dispersions when dispersed in a plasticizer or plasticizer/diluent mixtures. 3 (h) "i atex resin1' means a resin that is produced by a polymerization process that initiates from free radical catalyst sites and is sold undried. (i) "Bulk resin" means a resin which is produced by a polymerization process in which no water is used. (j) "Inprocess wastewater" means any water which, during manufacturing or processing, comes into direct contact with vinyl chloride or polyvinyl chloride or results from the production or use of any raw material, intermediate product, finished product, by-product, or waste product containing vinyl chloride or polyvinyl chloride but which has not been discharged to a wastewater treatment process or discharged untreated as wastewater. (k) "Wastewater treatment process" includes any process which modifies characteristics such as BOD, COD, TSS, and pH, usually for the purpose of meeting effluent guidelines and standards; it does not include any process>the purpose of which is to remove vinyl chloride from water to meet requirements of this subpart. (l) "In vinyl chloride service" means that a piece of equipment contains or contacts either a liquid that is at least 10 percent by weight vinyl chloride or a gas that is at least 10 percent by volume vinyl chloride. (m) "Standard operating procedure (SOP)1' means a formal written procedure officially adopted by the plant owner and/or operator and available on a routine basis to those persons responsible for carrying out the procedure. (n) "R.un" means the net period of time during which an emission sample is collected. (o) "Ethylene dichloride purification" includes any part of the process of ethylene dichloride production that follows ethylene dichloride formation and in which finished ethylene dichloride is produced. (p) "Vinyl chloride purification" includes any part of the process of vinyl chloride production that follows vinyl chloride formation and in which finished vinyl chloride is produced. 4 O "q. '6o.''cv ition is tion tun'ng if y t edure ble iSS le (q) "Reactor" includes any vessel in which vinyl chloride is partially or totally polymerized into polyvinyl chloride. (r) "Reactor opening loss" means the emissions of vinyl chloride occurring when a reactor is vented to the atmosphere for any purpose other than an emergency relief discharge as defined in 61.65(a). (s) "Stripper" includes any vessel in which residual vinyl chloride is removed from polyvinyl chloride resin, except bulk resin, in slurry form by the use of heat and/or vacuum. In the case of bulk resin, "stripper" includes any vessel which is used to remove residual vinyl chloride from polyvinyl chloride resin immediately following the polymerization step in the plant process flow. 5 ) ()( 54 :> 56 2.0 EDO, VC AND PVC INDUSTRIES Polyvinyl chloride is a polymer employed in the fabrication of literally thousands of consumer and industrial products. The manu facturers of these products purchase the polyvinyl chloride resin from a relatively small number of producers. Polyvinyl chloride is polymerized from the vinyl chloride monomer. In turn vinyl chloride is produced.from ethylene dichloride by cracking ethylene dichloride during dehydrochlorina tion or from the hydrochlorination of acetylene. In the following brief presentation, the attributes of ethylene dichloride, vinyl chloride, and polyvinyl chloride production facilities are discussed with respect to those considerations that are pertinent to plant inspections. 2.1 ETHYLENE DICHLORIDE (EDC) The principal process for ethylene dichloride production in the U.S. is oxychlorination which involves the reaction of oxygen and hydrogen chloride with ethylene [4]. In 1974, there were nine plants using this process to produce 5.05 billion pounds of ethylene dichloride per year [6]. While refined ethylene dichloride is sold for other industrial uses, its major use is for vinyl chloride monomer production. Typically the economics of the industry dictates that an ethylene dichloride plant be in close proximity to a vinyl chloride monomer plant in which case the shipping cost of large, continuous flows of pure ethylene dichloride to a vinyl chloride plant is minimal [4]. This has led to the concentration of these plants in Texas, Louisiana and the northern states [41. In most SAL 00004?B8 6 cases, the unconverted ethylene dichloride from the vinyl chloride plant is recycled back with the crude ethylene dichloride [4]. Vinyl chloride emissions occur in an ethylene dichloride plant using the oxychlorination process from three sources: side reactions in the oxychlorination process and dissolved vinyl chloride in recycled hydrogen chloride and in recycled unconverted ethylene dichloride. Physically, the major vinyl chloride emission sources are the EDC reactor, EDC refining and the fugitive emissions that may be present. 2.2 VINYL CHLORIDE (VC) There are four processes to produce vinyl chloride monomer [6]. There are fifteen plants producing vinyl chloride having a production capacity of 6.8 billion pounds per year [6]. Two plants use the hydro chlorination of acetylene, nine use the chlorination-oxychlorination of ethylene (with oxygen from air) and dehydrochlorination, one uses the same process, except that pure oxygen is used in place of air, and three plants use the direct chlorination of ethyl.ene and dehydrochlorination [6]. The main sources of vinyl chloride emissions in plants using the hydrochlorination of acetylene are reactor condenser vent (continuous), scrubber vent (continuous), heavy ends storage vent (intermittent), and assorted fugitive sources [6]. The sources of emissions from the chlorination-oxychlorination of ethylene and dehydrochlorination processes using oxygen from air or using pure oxygen are the same. The main sources of emissions are those dis cussed in Section 2.1 and not repeated here, purification system vents (continuous), scrubber vent (continuous), loading area (intermittent), and vinyl chloride emissions from the purification process; but the loss per unit product produced is greater from the purification portion of the plant. The process using direct chlorination and dehydrochlorination is used in vinyl chloride production when the manufacturing facility has other uses for the hydrogen chloride by-product. This process also uses ethylene dichloride in a dehydrochlorination process to produce vinyl chloride. 7 SAL 000049888 2.3 POLYVINYL CHLORIDE (PVC) Polyvinyl chloride is produced by a polymerization process from the vinyl chloride monomer. There are five processes used to polymerize the monomer: suspension polymerization is the most widely used and accounts for 78% of the U.S. plant.capacity; dispersion polymerization (i.e., emulsion) accounts for 13%; bulk polymerization for 6% [4]. The latex polyvinyl chloride is produced by dispersion polymerization and is sold and transported in a water suspension; solution polymeri zation is adaptable to a continuous process for copolymers and is used by one company in the U.S. [4]. Emissions may occur at any point of the processes, such as storage, reactors, strippers, mixers, weighers, blenders, recovery systems, inprocess wastewater, loading facilities, etc. [6]. 3.0 PROCESS FLOW DESCRIPTION AND EMISSION POINT IDENTIFICATION The manufacturing processes covered in the vinyl chloride emissions Final Standard are discussed in this chapter with the aid of process flow diagrams. The emission points covered in the Standard are listed and keyed on the corresponding flow diagrams for each process. 3.1 ETHYLENE DICHLORIDE--0XYCHL0RINATI0N* One of the major processes for ethylene dichloride (see Section 3.4) production in the U.S. is oxychlorination which involves the reaction of oxygen and hydrogen chloride with ethylene [4]. The oxygen may be introduced into the process in concentrated form or by an air stream as shown in Figure 3.1 [6]. The two processes are similar and will be discussed together since the emissioh sources are identical. The generic reaction equation for the process is given by CH2 = CH2 + 1/2 02 + 2HCL----------------- C1CH2 - CH2C1 + H20. (3-1 This reaction takes place in the reactor, shown in Figure 3.1, at elevated temperature. The oxychlorination process typically exhibits a 98% conversion of hydrogen chloride to ethylene dichloride per pass from the reaction represented in Eq. (3-1) [6]. The raw materials (ethylene, hydrogen chloride and oxygen/air) are made to pass through a catalyst. In the presence of oxygen, the catalyst concentrates the ethylene and chlorine allowing for the reaction, Eq. (3-1), to take place at a lower temperature. Because the reaction is highly exothermic, a water flow over the reaction tube surface is required to control the temperature in the reactor. The result is that steam evolves at the exit port of the water jacket. *While this process is usually identified as oxychlorination, it is more precisely an oxyhydrochlorination process. .9 iL 0 0004 Vo 8ciq r o Figure 3.1 Ethylene Dichloride process flow diagram using oxychlordination which involves the reaction of oxygen and hydrogen chloride with ehtylene. Table 3.1: The potential emission points identified by the keys in Figure 3.1 in the oxychlorination process for ethylene dichloride. Position Name Reactor Vent Reactor Refrigerator Condenser Vent Wastewater from Water Wash Column Wastewater from Wash/Crude Storage Wash Water Stripper Vent Wash Water Stripper Storage Vent Wash Crude Product Storage Vent Light-Ends Column Condenser Vent Finishing Column Vent Light-Ends Purification Column Vent Refined EDC Storage Tank Vent Heavy-Ends, Tar-Removed Column Vent Heavy-Ends Storage Tank Vent Key 1 2 3 4 5 6 7 8 9 10 n 12 13 Frequency Intermittent Continuous Continuous Continuous Continuous Continuous Intermittent Continuous Continuous Intermittent Intermittant Continuous Intermittent Tar Storage Tank Vent Fugitive 14 Entire Plant' 11 Intermittent Intermittent/ Continuous B A L. 0 0 0 0 T 9 B V' 2 The gas stream that exits the reactor contains ethylene dichloride in the form of a gas. In the process represented in Figure 3.1 this stream also contains ethylene, hydrogen chloride and air or pure oxygen. The stream is passed through a hot water wash column to remove impurities from the ethylene dichloride, and to maintain its gaseous state. The ethylene dichloride gas and water vapor stream are passed through a' cold water condenser prior to entering the separation tank. The separated, purified ethylene dichloride is then transported to a storage tank [6]. The water from the hot water wash and from the separation tank is passed through a stripping column to separate ethylene dichloride and waste by-products. The separated ethylene dichloride is passed on to the EDC storage tank, while the waste by-products are disposed. The potential emission points are listed in Table 3.1 and identified by the keys shown in Figure 3.1. The emission sources are listed to correspond with operational steps in the generalized hydrochlori nation process. Fugitive emission sources (pumps, pump maintenance valves, pressure relief valves, samplers, etc.) are not identified because the locations of these sources are unique for each plant. The emission from any point, whether listed or not, depends on the operating condition (batch or continuous, reaction efficiency, etc.) of the plant at the time of inspection and in some cases on the immediate past history of operating conditions [6]. 3.2 VINYL CHLORIDE 3.2.1 Hydrochlorination of acetylene In this process vinyl chloride monomer is produced by the hydro chlorination of acetylene. The reaction occurs between hydrogen chloride and acetylene at 85-141C in the reactor shown in Figure 3.2, in the presence of a catalyst, mercuric chloride on activated carbon [6]. The reaction is governed by the equation HC = CH + HCL ----------- - H2C = CHCL. (3-2) The reaction typically exhibits a 90% conversion to vinyl chloride per pass. As a result, acetylene and hydrogen chloride gases exit the reactor as well as vinyl chloride. These gases are compressed, cooled 12 O' 13 and pumped to a purification system consisting of a light-ends distillation column and a heavy-ends distillation column. In the light ends distillation column acetylene is separated and transported back into the reactor for another pass. Other volatile gases are passed to the vent gas reactor and vent condenser. Table 3.2 lists the main products and their boiling point temperatures. This shows that acetylene has the lowest boiling point temperature and is separated easily from the other compounds. Table 3.2: Hydrochlorination of acetylene product gases and their boiling point temperatures. Gas Acetylene Symbol HC = CH Boiling Point Temperature - 84C Hydrogen Chloride Vinyl Chloride Ethylene Dichloride HC1 H2C = CHC1 cich2 - ch2ci - 85 C - 13C + 83C cor prt prr lo an (b ti op The reaction represented in Eq. (3-2) is exothermic. Therefore, the heavy-ends tnat are transported to the heavy-ends distillation column from the light-ends distillation column will contain some poly merized material as well as vinyl chloride monomer. The vinyl chloride is distilled, subsequently liquefied and placed in a storage tank. The heavy-ends are transported to a temporary storage tank and ultimately incinerated. The potential emission points are listed in Table 3.3 and identified by th? keys shown in Figure 3.2. The emission sources are listed to 14 SAL 000049895 correspond with operational steps in the generalized hydrochlorination process. Fugitive emission sources (pumps, pump maintenance valves, pressure relief valves, samplers, etc.) are not identified because the locations of these sources are unique for each plant. The emission from any point, whether listed or not, depends on the operating conditions (batch or continuous, reaction efficiency, etc.) of the plant at the time of inspection and in some cases on the immediate past history of operating conditions. Table 3.3: The potential emission points identified by the keys in Figure 3.2 in the hydrochlorination of acetylene for vinyl chloride monomer. Position Name Key Frequency Reactor Condenser Vent 1 Continuous Scrubber Vent 2 Continuous VCM Condenser Vent 3 Continuous VCM Storage Vent Heavy-Ends Storage Vent 4 Continuous * 5 Intermittent VCM Loading Vent 6 Intermittent Fugitive Entire PI ant Intermittent/ Continuous * Intermittent is used in the sense that if the rate of incineration is greater than the rate of storage, emission is probably intermittent. When the reverse is true, emission will be continuous. 15 SAL 00004? 3.2.2 Dehydrochlorination of ethylene dichloride The production of vinyl chloride monomer by dehydrochlorination (removal of hydrogen chloride) involves the thermal dehydrochlorination of dry ethylene dichloride. Thermal dehydrochlorination is sometimes referred to as a cracking process. Vinyl chloride results when ethylene dichloride is placed in a cracking furnace at approximately 510C [4]. For most efficient operation, the furnace is packed with a catalyst such as pumice or charcoal. Typically, the conversion efficiency per pass is 94 to 97%. The reaction is represented by the equation C1CH2 - CH2C1----------- > H2C = CHC1 + HC1. (3-3) In an integrated ethylene dichloride-vinyl chloride plant, where the oxychlorination process is employed to produce ethylene dichloride, the hydrogen chloride by-product in Eq. (3-3) is returned to the ethylene dichloride reactor as an input crude shown in Eq. (3-1) [4,6]. Figure 3.3 shows the process flow diagram [6]. The ethylene dichlo ride is transported as a liquid (boiling point +83C, see Table 3.2) and is vaporized completely prior to entering the cracking furnace. The hydrogen chloride is generated in the cracking furnace. The gas flow exiting the cracking furnace consists of mainly vinyl chloride, but also present are ethylene dichloride, hydrogen chloride and other hydrocarbons. The quenching column uses liquid ethylene dichloride to liquify the ethy lene dichloride in the mixture while the gaseous vinyl chloride passes on to the partial condenser and condenser. The liquid ethylene dichloride is transported to the washed crude storage in the ethylene dichloride section of the plant and ultimately transported back to the cracking furnace for another pass. The vinyl chloride and hydrogen chloride gases are cooled, compressed and pumped to a purification system consisting of a light-ends distilla tion column and a heavy-ends distillation column. In the light-ends column the vinyl chloride is separated from the hydrogen chloride. The hydrogen chloride is recycled to the ethylene dichloride reactor if the oxychlorination process is used. Other volatile by-products are passed to the vent gas reactor. 16 RECYCLED TO Figure 3.3 Vinyl chloride monomer process flow diagram using dehydrochlorination of ethylene dichloride. The potential emission points are listed in Table 3.4 and identified by the keys shown in Figure 3.3. The emission sources are listed to correspond with operational steps in the generalized dehydrochlorination process. Fugitive emission sources (pumps, pump maintenance valves, pressure relief valves, samplers, etc.) are not identified because the location of sources is unique for each plant. The emission from any point, whether listed or not, depends on the operating conditions (batch or continuous, reaction efficiency, etc.) of the plant at the time of inspection and in some cases on the immediate past history of operating conditions. 3.3 POLYVINYL CHLORIDE 3.3.1 Suspension polymerization The suspension polymerization process is the most common process for polyvinyl chloride production [4]. The resin produced is sometimes referred to as suspension resin. Polymerization of vinyl chloride requires the mixing of weighed amounts of vinyl chloride (weighed amounts of a comonomer where desired to change the polyvinyl chloride properties), catalyst, water and suspending agents [4]. The raw materials are mixed in a clean glass or stainless steel lined reactor. Air is removed from the reactor by a steam jet or vacuum pump. Reaction temperature is controlled by either cooling or heating, depending on the details of the process used. The reaction is initiated by the catalyst. As the reaction proceeds, polyvinyl chloride is produced in particle form. Agitation is employed to prevent a slurry from agglomerating in the reactor and the suspending agent disperses the vinyl chloride droplets. The polymeriza tion process is allowed to continue until 85 to 90% of the vinyl chloride has polymerized; this requires approximately 6 hours [4]. If allowed to continue beyond this point, the product becomes increasingly less uniform with respect to molecular weight and, therefore, the physical properties are less uniform. The vinyl chloride residue is in vapor form in the reactor, dissolved in water, and/or trapped in the polyvinyl chloride granules. Figure 3.4 shows the process flow diagram for suspension polymeriza tion [4]. Vinyl chloride is supplied to the weighing station from the vinyl chloride source (plant or tank car) and from the recovery system. The comonomer (if required) for a batch is weighed in its own weighing station. In addition to the weighed quantities of vinyl chloride monomer 18 SAL OOO04?B9/ Table 3.4: The potential emission points identified by the keys in Figure 3.3 in the dehydrochlorination of ethylene dichloride for vinyl chloride monomer. Position Name Key Frequency Recycling of EDC to Washed Crude Storage Light-ends Distillation Column Vent Heavy-ends Distillation Column Vent Heavy-ends Storage Vent VCM Storage Vent Light-ends Column Waste Water Quench Column Vent Fugitive 1 2 3 4 5 6 7 Entire Plant Intermittent Intermittent Intermittent L1 i Intermittent __________________________ i Intermittent Continuous Continuous Intermittent/ Continuous 19 b ft i... 000049900 VENT Figure 3.4. P olyvinyl c h lo rid e process flow diagram using suspension polym erization. at and comonomer, the initiator catalyst and suspending agent are transported to the reactor. The polymerization takes place at elevated temperatures and at a pressure in the range of 5.1 to 8.0 atmospheres. The contents of the reactor (polyvinyl chloride granules, gases, water and initiator catalyst) are subjected to a stripping process to remove unreacted vinyl chloride. The stripping process may take place in the reactor or the contents may be transferred to a stripper vessel where stripping takes place. Vinyl chloride is stripped by the application of heat alone or by the application of the combination of heat and vacuum. The vinyl chloride and other volatilized chemicals drive off are transferred to the recovery system. In some plants the vinyl chloride is processed through a distillation column and passed on to the recovery system storage tank. The other gases are vented, and included will be some vinyl chloride. The remaining slurry is transported to a slurry blend tank where several batches are blended together to obtain a more uniform product [4]. The blended slurry is then pumped to a centrifuge where most of the water is removed. The wet polyvinyl chloride is then dried to remove remaining water and vinyl chloride. The resin is then transferred to storage or bagging stations. Table 3.5 lists the potential emission sources for the suspen sion process with the key referred to Figure 3.4. The Key(^T)denotes the recovery system vent through which the noncondensable gases are vented. The venting may take place manually or automatically and may be intermit tent or continuous depending on the system design and the pressures employed. Keys(?)and(?)are the weighing stations for the vinyl chloride monomer and comonomer. High pressure may build up at these stations, in which case the emission will contain vinyl chloride. The reactor emissions, Key(7), may arise when a run-away condition develops and due to residual vinyl chloride vapor that may be present when the reactor manhole is opened for cleaning. 21 11... 0 0 0 0 4 ? ? Table 3.5: The potential emission points identified by the keys in Figure 3.4 in the suspension polymerization process for polyvinyl chloride. POSITION NAME Vinyl Chloride Recovery Vent Vinyl Chloride Weighing Tank Vent Comonomer Weighing Tank Vent Reactor Vent and Opening Loss Stripper Vent Slurry Blend Tank Vent Centrifuge Vent Product Col lection Vent Silo Vent Waste Water Oryer Exhaust Bulk Loading Fugitive KEY FREQUENCY 1 Intermittent/Continuous 2 Intermittent 3 Intermittent 4 Intermittent 5 Intermittent 6 Intermittent 7 Intermittent 8 Intermittent 9 Continuous 10 Continuous 11 Intermittent 12 Entire Plant Intermittent Intermittent/Continuous 22 OOO1---ao' q' D3 Keys@(6) an<7)are all associated with the stripping, blending and drying operations of the polyvinyl chloride slurry. In each of these operations vinyl chloride emissions may occur points where emissions may occur in the storage and bagging of the resins, while Key is the emission point from the waste water treatment process. 3.3.2 Emulsion (i.e., dispersion) polymerization The emulsion polymerization process uses equipment basically similar to that of the suspension process described in Section 3.3.1 [4]. The resin produced is sometimes called an emulsion resin. Emulsion resins can be polymerized at lower temperatures and at a higher rate than suspension resins. However, emulsion resins are also more sensitive to heat and shear stresses. When subjected to either or a combination of heat and shear stresses, the resultant changes in the resin's physical characteristics may make it unsuitable for use. The resin obtained from the emulsion process is of smaller particle size than obtained from the suspension process [4]. The emulsion process flow diagram' is shown in Figure 3.5 and the cor responding keyed emission sources are presented in Table 3.6. A study of Figures 3.4 and 3.5 show that the processes are identical in the following ways: batch reactor process; water is used as the suspending medium to suspend liquid vinyl chloride; all process equipment is identical except for the dryer. In addition, the suspension process uses a centrifuge to aid in drying while the emulsion process does not. In the emulsion process soap and water are used as the emulsifier. The emulsion process differs from the suspension in the following ways: more soap is added to the slurry in the reactor which stablizes the monomer droplets and results in the absence of agglomerates; a spray dryer is used because it does not produce excessive temperature or shear stresses during the drying, while the rotary, flash, or fluidized bed dryer used in the suspension process may produce these stresses [4]. B A I... 0 0 0 0 4 9 9 0 4 23 o Figure 3.5. Polyvinyl chloride process flow diagram using emulsion (i.e., dispersion) polymerization. a. c o in 3 d) cr> 3 IT3 i_ C7l <T3 T3 O m Oci<nUj ia. TmJ os_ c r-- O u I-- +MfD-> >,-r- r~ CD > >> >> oo Cl. Cl. lO CO. CD L. C3l Table 3.6: The potential emission points identified by the the keys in Figure 3.5 in the emulsion (i.e., dispersion) polymerization process for polyvinyl chloride. POSITION NAME Vinyl Chloride Recovery Vent Vinyl Chloride Weighing Tank Vent Comonomer Weighing Tank Vent Reactor Vent and Opening Loss Stripper Vent Slurry Blend Tank Vent Spray Dryer Vent Product Collection Vent Silo Vent Process Water Bulk Loading Fugitive KEY FREQUENCY 1 Intermittent/Continuous 2 Intermittent 3 Intermittent 4 Intermittent 5 Intermittent 6 Intermittent 7 Intermittent 8 Intermittent 9 Continuous 10 Intermittent 11 Entire Plant Intermittent Intermittent/Continuous 25 Qu.. \ 06 3.3.3 Latex dispersion polymerization Latex resins are produced by the emulsion process [4]. The latex resin is polymerized vinyl chloride monomer suspended in water. It is sold and transported in this solution form. The process is identical in almost all ways to the suspension and emulsion processes. It differs in that there is no drying process step and more soap is added in the reactor than for the emulsion or suspension processes [4]. The result is a latex resin which is a colloidal suspension of polyvinyl chloride. The process flow diagram and most probable emission sources are shown in Figure 3.6 and Table 3.7, respectively. 3.3.4 Bulk polymerization The bulk polymerization process is a batch process and consists of two polymerization steps [4]. In the pre-polymerization reactor there is liquid vinyl chloride in the presence of a polymerization initiator. The reactor is of a similar design to that used in the suspension process. The conversion to polyvinyl chloride from vinyl chloride is in the range 7 to 12% [4]. This suspended polyvinyl chloride'in liquid vinyl chloride is then transferred to a larger, high pressure, horizontal-type reactor. To this is added more liquid vinyl chloride and initiator. This reactor, sometimes called an autoclave, serves as the post polymerization reactor resulting in a reaction efficiency of approximately 85 to 90% [4]. The postpolymerization reactor is more rugged and the agitation more vigorous than the pre-polymerization reactor [4]. The post-polymerization reactor must be cleaned after each batch; less frequent cleaning is required for the pre-polymerization reactor. The bulk process is similar in most aspects to the suspension process. Since there is no water or water vapor in the suspension, low temperature (-35C or -31F as opposed to 7C or 44.6F in the suspension and dispersion) condensers may be employed in the recovery system. Moreover, the drying operation is not needed, and no in-process waste water system is present [4], The remaining monomer in the post-polymerization vessel may be removed by a number of processes. The monomer may be removed by vacuum alone. Another method is to introduce steam into the autoclave and the steam and released vinyl chloride are removed by vacuum. The steam-vacuum 26 Ai... 00004990 27 iFigure 3.6. P olyvinyl c h lo rid e la te x process flow diagram using emulsion ( -e ., dispersion) p o lym e riza tio n . Table 3.7: The potential emission points identified by the keys in Figure 3.6 in the emulsion (i.e., dispersion) polymerization process for polyvinyl chloride latex. POSITION NAME Vinyl Chloride Recovery Vent Vinyl Chloride Weighing Tank Vent Comonomer Weighing Tank Vent Reactor Vent and Opening Loss Stripper Vent Slurry Blend Tank Vent Product Col lection Vent Loading Station Fugitive KEY 1 2 3 4 5 6 7 8 Entire Plant FREQUENCY Intermittent/Continuous Intermittent Intermittent Intermittent Intermittent Intermittent Continuous Continuous Intermittent/Continuous procedure may be repeated as many times as required to meet the emission standard. Those plants using the steam-vacuum process require wastewater stripping to bring the wastewater in compliance with the emission standard. Whichever method is used, the recovered monomer is placed in a temporary holding tank and recycled back to the pre-polymerization reactor through a filter [4]. The process flow diagram and the potential emission points are given in Figure 3.7 and the keyed emission points listed in Table 3.8 [4], 3.3.5 Solvent polymerization The solvent polymerization product is considered a speciality product and is a small segment of the total PVC industry [5]. However, it serves a large number of important needs that usually involve thin PVC coatinqs, such as for the food and beverage industries [5], The early developed process is shown in Figure 3.8 and the more recently developed process is shown in Figure 3.9 [5]. The processes are similar in most aspects, differing more in the technological developments of recent years than in the basic process operations. The early process i.s described below followed by the points of difference. The early process flow diagram is shown in Figure 3.8 and the cor responding source emission points listed in Table 3.9. The emission points in Table 3.10 correspond to the process flow diagram shown in Figure 3.9. The comonomers, initiators and solvents-are continuously, introduced into the reactor. The process flow shows that the VCM, comonomer and initiator are introduced into the reactor along with the solvent, usually n-butane [4]. The comonomer is almost always vinyl acetate. The continuous process provides a degree of turbulence among the constituents in the reactor that results in a copolymer conversion efficiency approaching 100%. A continuous copolymer stream is drawn off from the reactor and filtered. The filter cake is passed on to a flash evaporator where it is dried and the monomers recycled. The solvent is drained from the filter and recycled back into the reactor along with the recovered monomers. During any one pass of the solvent stream, some solvent is lost to the process; therefore a solvent make-up stream is also required [4], 29 0 0 0 0 4 o 9 -i NITROGEN co o Figure 3.7. Polyvinyl chloride process flow diagram using bulk polymerization i yam Ui.ny duik polym erization. Table 3.8: The potential emission points identified by the keys in Figure 3.7 in the bulk polymerization process for polyvinyl chloride resin. POSITION NAME KEY FREQUENCY Pre-Polymerization Reactor Vent Post-Polymerization Reactor Vent Monomer Holding Tank Vent Recycle Condenser Vent Pressure Reducer Vent First Bag House Vent Second Bag House Vent Reject Vent 1 2 3 4 5 6 7 8 Intermittent Intermittent Continuous Continuous Continuous Continuous Continuous Continuous Fugitive Entire Plant Intermittent/ Continuous 0Q'JV... '-' >"-.i 31 3 (( i 'i 4 UJ ro BAGGING a SHIPPING BAGGING & SHIPPING Figure 3.8. Early polyvinyl chloride process flow diagram using solvent polymerization \ r 8 A L 0 0 0 0 4 ? 9 ;j. 4 33 Figure 3 .9 . More recent p o ly v in y l ch lo rid e process flow diagram using solvent polym erization. Table 3.9: The potential emission points identified by the keys in Figure 3.8 in the solvent polymerization process for polyvinyl chloride and copolymers. POSITION NAME KEY FREQUENCY Receiving Tank Vent VCM Storage Vent Reactor Vent Flash Evaporator (Stripper) 1 2 3 4 Continuous Continuous Continuous Continuous Solvent and Monomer Recovery Bag Filter 5 Continuous 6 Continuous Bag Filter-Screen 7 Continuous Grinder Filter 8 Continuous Grinder Filter-Screen 9 Continuous Storage Silo Bulk Loading 10, n 12, 13 Continuous Intermittent Fugitive Entire Plant Intermittent/ Continuous 34 Table 3.10: The potential emission points identified by the keys in Figure 3.9 in the solvent polymerization process for polyvinyl chloride and copolymers. POSITION NAME Reactor KEY FREQUENCY -- 1 Continuous Monomer Condenser 2,3,4 Continuous Resin Drying 5 Continuous Silo 6 Continuous Solvent and Vinyl 7 Continuous Acetate Condenser Fugitive Entire Plant Intermittent/ Continuous i 0 Cci oo 35 Vinyl chloride emissions from the reactor area in a continuous process are relatively lower than from batch processes [4]. There is some evidence to suggest that the vinyl chloride is more easily removed from the resin than in other polymerization processes. The most important difference between the processes is that the earlier process produced a slurry as a result of polymerization while the later process produces a solution. The stripping operation from solution gives a lower VCM concentration and typically produces less emissions. Moreover, inprocess water comes into contact with the polymerized material after the stripping operation. This avoids the requirement of an inprocess water stripper, and the inprocess water storage and treatment equipment should not give any emissions. Finally, drying is accomplished by means of hot air rather than flash evaporator. 3.4 CLARIFYING NOTE ON THE BALANCED OXYCHLORINATION - DEHYDROCHLORINATION PROCESS Oxychlorination and direct chlorination are the two major processes used for ethylene dichloride production [4}. EDC plants operate a balanced process which consists of a vinyl chloride plant and a direct chlorination plant [4], A block diagram of the balanced process is shown in Figure 3.10. Typically ethylene dichloride refining is common to both the direct chlori nation and to the oxychlorination plants. Therefore, the ethylene dichloride crudes are refined through the same equipment. In Figure 3.10 the EDC refining equipment is shown to be part of the EDC oxychlorination plant. The crude oroduced from the oxychlorination plant may contain vinyl chloride monomer. Therefore, the common receiving point of the monomer and all down stream parts of the direct chlorination plant are subject to the Standard. That the crude from the oxychlorination plant may contain vinyl chloride arises because of recycled ethylene dichloride and because the recycled hydrogen chloride used is a by-product of the monomer cracking. In Figure 3.10 this is shown by the flow of HCl and EDC from the dehydrochlorination to the oxychlorination reactors and EDC refining, respectively. 0 c>- 36 r . le 1 ss ;d f de 'he de frogen 1 VCM Figure 3.10. Block diagram of a balanced oxychlorinationdehydrochlorination process. 37 SAL 0000499. 1 The economics of the HVC industry demands that large producers of EDC employ the balanced oxychlorination - dehydrochlorination type plants [4], From Eq. (3-3) it is seen that for each vinyl chloride molecule produced in the cracking of EDC, one HC1 molecule evolves as a by-product. This forms the HCl stream in Figure 3.10. However, in the oxychlorination reaction, governed by Eq. (3-1), two HCl molecules are required for each EDC molecule produced. The HCl by-product provides at a maximum one-half the EDC required by the cracking furnace. Therefore, the direct chlorination process must supply at least one-half of the EDC required in a balance type plant. Presently, 95% of the EDC annual production rate is produced in balance type plants. 3.5 PHOTOGRAPHS OF EQUIPMENT The technology employed in the EDC, VC and PVC industries is of a relatively higher technical level than Inspectors normally encounter in other plant ins actions subject to the Clean Air Act. In addition, specific processes differ for the same general product class from plant to plant. In many cases, if not in most, the equipment .used was constructed according to unique specifications. The operation of the equipment may be different, in which case the materials, size, shape and their placement in the plant may also be different. Typically, the Inspector Is not able to draw from experiences gained of previous inspections to the degree that is commonly done in other industries. To assist the Inspector in conducting complete and efficient plant inspections, photographs of equipments are shown in Figures 3.11 to 3.39 from which VC emissions are more likely to occur. From photographs of general types of equipment such as reactors, strippers, storage vessels, etc., it will be less difficult to determine the function of more specialized equipment types. No attempt has been made to present photographs of dif ferent types and sizes of equipments because this would require in excess of 200 photographs. Table 3.11 gives the figure numbers and the equipment category of those selected for reproduction in this Manual. 0't 38 Figure number of photographs and corresponding equipment category. Figure Number 3.11 to 3.15 3.16 to 3.18 3.19 to 3.24 3.25 and 3.26 3.27 to 3.32 3.33 and 3.34 3.35 3.36 and 3.37 3.38 to 3.39 Equipment Category Reactors Strippers Storage Vessels Dryers Distillation Vessels RD/SRV Double Mechanical Seal Railroad tank car loading and unloading Incinerators 0 0 () 0 4 9 9~ 39 Manual Vent Figure 3.11 Large capacity EDr reactor usingthe oxychlori*' nation process is located in the tall vessel. The oxychlorination manual vent is the open ended pipe, risinq above the top of the reactor on the left side. EDC Reactor amvm RD/SRV PVC Reactor AH'- y Figure 3.12 The tops of medium capacity, side-by-side PVC reactors using the emulsion process may be seen. Individual RD/SRV and manual vents may also be seen mounted on top of each reactor. 40 :ity EOc i Tori- ' ited in ' oxy;nt is "ising ~eactor e O) > on c f- cj > (0 . 4- ro i Q_ 3 >> c ai E -o C HI JZ HI Of r- -- HI 4-9 o > > on HI i-- r-- .E 0) LO X3 o HI -M _e sz I C 4-> .o 4-> >> HI i-- _Q -o o on O 4-9 1 Hi on 4- ro ai -l-> S- Kl TU on o ra Cl on i-- <-- on (U E 4-9 i_ on a) c > O c OJ o c ra o QJ E >> o o E 4-9 > 4-> L. u on C r-- r- CL E i_ o o CJ on GJ O o Cl rO 4-9 4-9 4-> 4-1 Q. C c to (J QJ C. <o a> HI >> rO J= o o > > on a> 4-9 4- r-- i_ E o g-- >n HI *o 3 on 03 s_ <4- on 0J 3 HI o r- i- -o on C > (O f-- a> z: 1c-- ro E O (J c o u 3 cr on HI i-- o OJ 3 -a i_ 4-9 4-9 S- *a- e S- r-- on ro OJ O on QJ S- a Cl 4--9 S- m O > if c3 +J OS c. S- QJ 4-> <D U CO o HI E ro lO "'v. 3 Hi i_ 3 ai o j= O r-- HI CD 5- a: o _1 HI Cl C o 4-> r-- 1 C a 4-9 3 HI <o r-- H> > on i I-- !-- >Mo o i--i IS) E C- >> on S_ O S- c s- ai 4-> Q) oE ai ai i_ s_ E QJ > -1-9 >> 09 O on U t-- >>4-9 S- c CJ on <o o u r-- rO OJ QJ >-- QJ Cl QJ o 3 Cl C S_ E i- > Cl Q. HI QJ X O a r-- CL Hl HI HI CJ 4-9 c s_ o -C -C 0-4-9 X ro 4-> 4-> 4-9 > ro xz QJ on QJ r-- 4-9 r-- 4-9 C CJ JZ O 3 r-- > o 4-9 4-9 on T2 S- Cl 4-> OJ c o JC s O ro s_ 4- HI a Oi-X c J-- o .C QJ 3 C ra 4-9 c 4-9 s_ O 3 oo c. s_ U c o <o S_ QJ 4-9 CL on on OJ 4-> Q> 4- 4-9 r E cm I-- i_ ro > 00 OO3 i_ O c o x: o <-- E CJ rO O rO QJ 09 HI > Cl S_ CJ 3 c on O- HI On 4-9 ro O >r-- C * J_ c r-- s_ - c Hl CL r" "O o.x: HI c o fsi > c 4-> XZ QJ !- r- c Q rt3 on Cl on V) r- r- on O o c QJ -a ra 4-9 HI r-- $- S- QJ - 09 S- CL on e-- O > jit S- O on OJ ro 4-> o o c OJ H) 4-> 3 on E O 4-9 E 3 "O -o CJ on iS) CO rO QJ S- r-- i- ra QJ HI S_ U 4-4 o HI HI O &_ o O S_ JC o ro l-- QJ 4-9 c 4-> i- C S- i- C JC -o C- iO o_ rO o i- C ro HI 5 E <o - c f-- Cl g-- HI -O o 0J U c Hi >>4-9 C -O u c o i-- c on rO QJ 4-9 3 o 3 r- LU >v 4-> IS) ro cm 4-9 S_ 4-9 = > on on 3 M SAL 000049922 41 Figure 3.15 The top of medium capacity, side-by-side PVC reactors using solvent process. Foreground shows a motor valve for emergency venting through to the VC monomer recovery system. SB- O v- Q 0 42 Figure 3.17 A stripper column is shown that removes VC from a PVC-varnish solution resulting from the solvent polymerization process. Figure 3.18 A typical wastewater stripper column is shown that may be found in EDC, VC and PVC plants. V Figure 3.19 The top of an EDC storage tank is shown with its vent. Figure 3.20 Cylindrical, side-by-side, above ground VC storage tanks are shown. RD/SRV vents may be seen on pipe rack above tanks. 44 0000*^ c.^\- T Vents Figure 3.21 Spherical, above ground vinyl chloride storage tanks. RD/SRV vents may be seen perched on top of sphere RD/SRV's 2^3 Underground VC Storage Tank Area Figure 3.22 Underground VC monomer storage tank area is shown. o0o Figure 3.23 Wash water stripper storage tank with its vent mounted at the top of the tank in an EDC-VC plant. Drying Pipe Figure 3.24 Upper portion of tar storage tank shown with vent. 46 r 2 top Figure 3.25 PVC suspension resin rotary dryer and dust collector. Drying is accomplished by the application of heat and rotary action. i Figure 3.26 PVC dispersion resin spray drying takes place in the cylindrical building. Large diameter feed pipe, seen at the left of the dryer, carries drying air to the top of the dryer. Housing for the atomization system is perched at top of dryer. 47 At.. 0 0 0 0 4 ? ? 2 8 Vent Figure 3.27 An EDC light ends distillation column is shown with its RD/SRV vent. VCM Column Condensor Figure 3.28 VCM column condenser and condenser vent motor valve. Motor Valve Vent 48 I lueid 3A-303 pa:>ue|.eq ul UMoqs si [dsssA ino^joou^ pinbLL 110142.1011363.1 u3vOp spuo 4l|6ll 9Ml punoa63Joj. aq; ui qe* aunfiLj 4U0A AMS/Oy 941 6ULM04S uo;e[nujnoDe xa[j.au uiunioD spu3 ;q6i.[ 3Q3.UB ^0 do; aiu $Z' 3jn6Lj EDC Finishing Column Vent Motor Valve Figure 3.31 An EDC finishing column with the RD/SRV and its vent mounted on the top. (In some installations o the finishing column performs the function of the X light ends distillation column while in others it X encompasses the functions of both the light and heavy <= ends columns.) Figure 3.32 Foreground shows the insulated vent piping and motor valve from the reactor refrigerator condenser vessel in a balanced EDC oxychlorination plant. r >cc CO OC 51 H1- Ov'nJ.'.j. .-!|'''"1 'v'.' O' Common M anifold / \ . RD/SRV $ F igure 3.33 T y p ic a l RD/SRV assembly w ith the vents from each connected to a m anifold o f the vent system. F igure 3.34 Dual RD/SRV vents mounted a t to p o f spherical vinyl chloride storage tank. 1 Figure 3.35 Typical pump and double mechanical seal Fiqure 3.36 Top connections on railroad tank car shown with flexible hose attached for VC loading. Smaller diameter flexible hose in the foreground is connected to recovery system. -aa> u+-> oj c J <s> -cZo3 o aC>Ti so a> .c 4-> C 0a0> O c --01' -*EO->J arX--at t- 0>00 J>-> i- >QJ 01 i1---1) a-T3 OO sO_J +O> Figure 3.37 Railroad tank car loading platform shown with pipe rack support for flexible hose VC feed and recovery system in VC plant. 53 h L 0 0 () 0 4 9 Q "-j 4 1 Figure 3.39 Incinerator and stacks of a PVC plant showing the platform (center stack) on which stack samples are taken to determine VC emission concentration. 54 -.o leak detection monitoring in^stirkuumhfennttaatttionun, ^records, and reports A requirement of the Standard is that an EPA .-approved leak detection and elimination program be operational. This requirement incl d installed continuous leak detection monitorinaa s'vjp^'t-osmm, ^rou*t.ine ,leak detection monitoring with a portable hydrocarbon detector and a 1eak elimination plan. The Standard also requires recordkeeping (recording and retention for at least two years) of data relating to leaks detected by one of several ways. Conducting a meaningful inspection for the determination of com pliance of vinyl chloride emissions in typical EDC-VCM-PVC plants presents a number of unique challenges. The first is that the technology used requires a relatively high decree of expertise. The second is that the plant area, from raw materials to finished product shipping, is measured in acres rather than in square feet and may extend from below ground level to several hundred feet above ground. Added to these is the difficulty in making a definitive determination of some equipment with respect to the specific process(es) or function(s) it serves. Therefore, the NESHAPs Inspector needs to resort to complementary methods, in addition to normal inspection procedures, to make a determination of compliance. To circumvent these unique challenges, a NESHAPs inspection, of necessity, will need to rely on the in-plant continuous monitoring instrumentation and the records obtained therefrom. In this chapter, the typical continuous monitoring system, the resultant records and reports are described and discussed. All are required to be in compliance. 4.1 LEAK DETECTION MONITORING INSTRUMENTATION The Standard promulgated on October 21, 1976 requires that continuous monitoring detection and measurement instrumentation be permanently installed in a plant in which vinyl chloride may be emitted to the atmosphere. 55 C' O C'O .<?. o o Typically, the instrumentation consists of a vinyl chloride or hydrocarbon measurement instrument, mini-computer, computer program for data acquisition and data reduction, data terminal and assorted 1/4" stainless steel tubing, solenoid valves, vacuum pumps, etc. The measurement instrument may be a gas chromatograph or, if the owner/operator assumes that all hydrocarbons measured are vinyl chloride, an infrared spectrometer or flame ion detector or an equivalent or alternative method. The monitoring instrumentation is based on area (i.e., volume) sampling. However, some plants will also monitor fugitive emission sources such as pump seals, agitator seals, couplings, etc. The typical instrumentation system uses one measurement instrument with a system of tubing that serves to draw air samples from an area of the plant into the detector air sample chamber where a measurement of vinyl chloride concentration is made. The concentration value is then transmitted to the computer memory to be printed out on the computer terminal at a later time. The time interval between measurements is 1 to 3 minutes. Usually, each monitoring point is measured in sequence and the sequence is unchanging. When all points associated with a measurement instrument have been measured for vinyl chloride emissions and transmitted to the computer memory, the computer program provides for each measurement to be printed in tabular form, Each measurement is identified with respect to the time of the measurement and location within the plant. Depending on the computer program, average vinyl chloride emissions for each point on the basis of shift, day, week and month may also be printed out and become a part of the record. In some instances, the data terminal does not print any measurement made unless the vinyl chloride concentration is greater than some defined level, such as 5 ppm. Plants that employ this system will record each measurement for each point on the measurement instrument's printer. It, too, becomes a part of the required recordkeeping. Usually the concentration is measured at each point at least every 25 minutes. Each plant sets its own threshold level for the purpose of defining a leak. In most cases, two consecutive measurements from the 4V 56 c,{V- same monitoring point equal to or greater than the concentration threshold value are used in the definition of a leak. The concentration threshold level is the definition of a leak for the leak detection monitoring system. This definition requires the approval of the EPA Administrator and it is set at a level compared with the vinyl chloride background concentration. When two consecutive measurements at a point indicate a leak, plant personnel assigned to the "Leak Detection Patrol" investigate with a portable instrument the region of the plant in which the monitoring point is located. The monitoring system is required to be calibrated daily by one of two methods, described in 61.68(c). Some plants reserve one of the points in the sequence of point measurements for calibration. Thus, the instrument is calibrated iin each sequence of measurement. Some plants may use more than one measurement instrument when a large number of points are being monitored. The number of points under observation by an instrument ranges from 9 to 19, while the total number in a plant ranges from 9 to 76. However, EPA approval is required with respect to the position and minimum number of points. A schematic of a monitoring system is shown in Figure 4.1 for which there are n-air sample inlets. When a solenoid valve is activated, it allows an air sample to be? drawn by a vacuum pump from a point in the plant into the detector air sample chamber. The pump operates continuously, evacuating the manifold of the previous air sample so as not to influence the vinyl chloride concentration measurement of the next air sample. In addition, the volume of the air sample drawn prior to actual measurement is sufficient to effectively purge the detector air sample chamber of any residue from previous air samples. The switching of the solenoid valves may be accomplished by one of two methods: a mechanical or electronic timer where the sequence is unchanging; a computer-control 1ed system where the sequence may be changed according to a program. Those installations using timers usually activate one solenoid at an.y one time. Computer-controlled systems may have sophisticated programs where.one or more solenoids may be operated to more quickly assess the emission(s) in one or more plant areas. 57 i:> i':'i L_ 000 049933 Probe Location 1 Figure 4.1: Schematic diagram of continuous monitoring system for vinyl chloride emissions. ,,`V .-vO'- 58 4.2 LEAK DETECTION MONITORING RECORDKEEPING The owner/operator is required to record data, retain the records on location for a minimum of two years and to make available, upon request of an EPA representative, those records obtained from the continuous leak detection monitorinq system. Specific data of the detected leaks will be recorded and retained which pertain to the location within the plant, vinyl chloride concen tration and the date and approximate time of measurement. 4.3 ROUTINE LEAK DETECTION AND RELIEF DISCHARGE RECORDKEEPING The owner/operator is also required to record data, retain the records on location for a minimum of two years and to make available, upon request of an EPA representative, records obtained of leaks detected during routine monitoring with a portable hydrocarbon detector and for relief discharges from reactors. Specific data of each leak detected during routine monitoring will be recorded and retained which relate to location within the plant, vinyl chloride concentration, date and time of measurement, cause of each leak, and the action taken to repair or eliminate each leak. SAL 0000499 59 5.0 INSPECTOR SAFETY Prior to inspecting an EDC-VC, VC, or PVC plant, an inspector should check the plant's records to determine whether any leaks have occurred in the past several days and whether any leaks are currently being experienced. Before entering a plant the inspector should check to assure he has the proper safety equipment, including safety shoes, safety glasses, hard hat, and a respirator specified for use with vinyl chloride. Any safety regulations and plant emergency responses should be noted by the inspector. OOO'f C-, (-:\\- 60 6.0 INSPECTION PROCEDURES AND INSPECTION FORMS The plant's compliance with the Standard will be determined through periodic inspections by an EPA representative, in addition to data required in the semi-annual reports on stripping and reactor opening loss and on emission tests. The inspection entails a physical inspection as well as a review and assessment of the plant's records. In the following presentation, a number of forms have been developed for the purpose of aiding the inspector to make a definitive inspection in the most efficient and expeditious manner. If the forms are used properly, all major facets in the Standard will be covered. In fact, the questions and statements that make up the forms mirror the major facets of the Standard. Moreover, the questions and statements have been couched so that answers and responses may elucidate or reveal noncompliance conditions. Ideally, all questions and statements on all forms will be answered completely during the regular or periodic inspections. However, some plants may excel in complying with one or more sections of the Standard. An inspector having become aware of this fact may elect not to pursue questions and statements relating to that portion of the Standard. The frequency of the inspections is not recommended in this document, because this is a function of personnel being available on a regular basis. Besides, experienced inspectors are the best judge of the required inspection frequency. The forms described below are self-explanatory and provide a sequence that an inspector may choose to follow during an inspection. Questions which may be answered prior to the actual inspection (or verified after the inspection) are identified on the forms by an asterisk. Suggested sources for this data, from records on file at EPA Regional Offices, include plant Standard Operating Procedures, the Leak Detection Program and Initial, Semi-annual and other required reports. High potential 61 r leak sources and any violations can be identified from previous inspection reports. If this data is recorded on the forms prior to the inspection, it should be confirmed during the inspection and any inconsistencies noted. Similar notations should be made on the forms when data collected during the inspection is later compared with data from Regional Office records. 6.1 SUMMARY OF COMPLIANCE STATUS This form serves to summarize on one page the status of the plant with respect to the major facets of the Standard. The questions and statements serve to determine where emissions occur and at what levels, the emission control devices used for each emission source and the waivers that have been issued. This form will be most useful to personnel of the Enforcement Division having responsibility to make a determination of compliance. 6.2 CHECKLIST This form contains 29 question and statements requiring responses. They are grouped under the following categories: General, Leak Detection Monitoring System, Stack Emission Monitoring System, Portable Instrument, Leak Detection and Elimination, Discharges to the Atmosphere, Fugitive Emissions, Reactors and Furnaces, Control Devices, Stack Emissions, Inprocess Wastewater and Reactor Opening Loss. The responses require the inspector, in most cases, to actually view and verify. In the case of instrumentation, there is provision for the inspector to determine the accuracy of the instrumentation and/or system calibration. It is strongly recommended that a calibration be performed on one or more monitoring points of the leak detection monitoring system. In doing so, the inspector automatically checks the integrity of those monitoring points. It may be more practical for the plant's personnel to provide the responses to No. 23 rather than the inspector. However, the inspector should be present during the time the sample is being prepared and data obtained. It is strongly recommended that the inspector observe in its entirety the plant's procedure to determine the reactor opening loss. 6.3 ON REVIEW OF RECORDS Due to the maturity of the PVC industry and the economic environment 62 t P 1 3 * that results, EDC, VCM and PVC are typically large capacity plants. One plant may cover many acres and extend up to 100 to 200 feet above ground level as well as below ground level. It may be physically impossible for an inspector to inspect all parts of a plant in person. It is strongly reconmended that the central point of the inspector's focus be placed on the records that the plant is required to maintain, and particularly the leak detection monitoring system. Therefore, the review and assessment of the plant's records-is an important aspect of every inspection. The items in the form provided for the review of records is designed to elucidate compliance at the major or potential emission points. 6.4 PRE-TEST EQUIPMENT CHECKLIST FOR STACK EMISSION TEST The stack emission test required in the Standard, Test Method 106, is very clear and precise on the equipment and apparatus to perform the test. This form is designed to ensure that both inspector and plant personnel are reminded of the entire equipment and apparatus list required. It also serves to document alternative or equivalent equipment or procedures. 6.5 EQUIPMENT CHECKLIST FOR VINYL CHLORIDE CONCENTRATION TN INPROCESS WASTEWATER, RESIN, SLURRY, WET CAKE AND LATEX SAMPLES Test Method 107 is also clear and precise on the equipment and apparatus required to analyze samples in a head space analyzer. This form is designed to aid the inspector and plant personnel in conducting analytical tests that conform with the Standard. SUMMARY.OF COMPLIANCE STATUS (SCS) NESHAPS INSPECTION FOR VINYL CHLORIDE EMISSIONS COMPLIANCE Inspection Date_ Inspector's Name Firm's Name______ Firm's Address PARAMETER OR ITEM *1 Emission Sources *2 Where ducted? Atmosphere or Control Device *3 Emission Control Device Description In Existence Date to be Installed *4 Frequency of Emissions Cont./Intermit./Emergency *5 Applicable Regulations *6 Emission Standard ppm by vol. or ppm by wt. or gm. per Kg. *7 Estimated Emissions ppm by vol. or ppm by wt. or gm. per Kg. *8 Monitoring Requirements *9 Waiver Applications *10 Date Waiver Issued EMISSION SOURCE STATUS *11 Remarks on Compliance Emissions & Emission Tests Waivers Recordkeeping * Data may be available from records on file at EPA Regional Offices. 10 Data may be a v a ila b le from records on f i l e a t EPA Regional O ff! vaoii CHECKLIST NESHAPS INSPECTION FOR VINYL CHLORIDE EMISSIONS COMPLIANCE Inspection Date__________________ _ Inspector's Name________________________ GENERAL 1? Firm's name_________ __________________________________________________________ 2* Firm's address___________________ ________________ ,___________________________ 3* Process designation/product identification (check one or more): EDC: oxychlorination balanced N.A. VCM: hydrochlorination dehydrochlorination 0 N.A. PVC: suspension Q dispersion 0 latex N.A. bulk O solution 0 copolymer 4. Rated and average annual reactor/cracking capacity EDC: Desiqn* ; Normal Max* VCM: PVC: Desiqn* _* Design ; Normal Max* ; Normal Max* ; Actual ; Actual ; Actual LEAK DETECTION MONITORING SYSTEM 5* Permanent leak detection monitoring system [61.65(8) (i) J a) Total number of points monitored: b) Number of measuring instruments:; T.ype(s) ______________________ c) Time interval to cycle all points:_____ d) Definition of a Leak: _____________________ e) Lower detection limit (LDL) of instrument_____________________ f) Measurement sequence: 0 unchanging 0 program controlled g) Data reduction (check one or more) 0 none hourly 0 shift 0 daily weekly monthly other (specify)_______ * Data may be available from records on file at EPA Regional Offices. q ^ [ () () 0 0 4 / V 4 4 65 r 6. Calibration of leak detection monitoring system [61.65(8)(iii)]: a) List instruments and data from monitor points which were tested for calibration. (Attach sheets if more space is required.) Date & Time Location in Plant Instrum. Iden. Cal. VCM Concen. Instrum. VCM Percent Action Concen. Deviation Required b) Calibration method used (check one or more) None PI Paragraph 61.65(8)(iii)(A) - Test Method 106 - 5.2.1, 5.2.3 Paragraph 61.65(8)(iii)(B) Other__________________________________________ _______________________ STACK EMISSION MONITORING SYSTEM 7* Emission (i.e., stack) monitoring system a) List emission (i.e., stack) sources Location in Plant Ducted Processes Continuous or Sampling, Other Number of Monitoring Points b) Lower Detection Limit (LDL) of instrument 66 I 8. Calbration of emission monitoring system: a) List instruments and data from monitor points which were tested for calibration. Date & Location Instrum.. Cal. VCM Instrum. VCM Percent Action on Time in Plant Iden. Concen. Concen. Deviation Required red PORTABLE INSTRUMENT 9. Calibration of portable hydrocarbon detector(s): ) a) List instrument(s) which were tested for calibration Date & Time Location in Plant Instrum. Iden. Cal . VCM Concen. Instrum. VCM Percent Action Concen. Deviation Required b) Calibration method used (check one or more) None Paragraph 61-65(7)(1ii)(A) - Test Method 106-5.2.1, 5.2.3 Paragraph 61.65(7)(iii)(B) Other_____ _____________________________________________________ _ 67 Li } I... () () () (' A 9 p p r LEAK DETECTION AND ELIMINATION 10* Leak detection and elimination process [61.65(8)] a) What is the frequency of the "Leak Detection Patrol"?. b) Check typical points and equipment which are required to be routinely checked for leaks by the "Leak Detection Patrol" in the following: reactor storage - heavy cracking furnace stripper EDC purification recovery system mixing weighing holding tank storage - water 1ight ends col. heavy ends col. wastewater col . water wash col. water quench col. wash water stripper separation tank dryer n blending tank condensers storage - raw RD/SRV storage - finished Others: control devices pumps compressors agitator(s) loading unloading flanges valves fi1 ter strainers centrifuges holding bins silos c) What is the average elasped time between the determination of a leak by personnel of the "Leak Detection Patrol" and the plant's personnel taking corrective action for the purpose of eliminating the leak? Small leaks: Large leaks: d) What is the average elapsed time between the monitoring system alarm becoming activated and the plant's personnel taking corrective action for the purpose of eliminating the leak? Small leaks: Large leaks: DISCHARGES TO THE ATMOSPHERE 11. List any SRV's which do not use RD's or vent to recovery system or to gas hold tank. Location in Plant Comnents 68 l-', 1 12. If a pressure gage is located between rupture disc (RD) and safety relief valve (SRV), list points in the plant where the gage indicates a higher than normal pressure, typically 0 to 5 psig. (Note: While in common use in plants, a pressure gage between RO and SRV is not a requirement of the Standard.) Identify those RD/SRV points which are in PVC plants by check (/) in column 3. RD/SRV Identification Location In Plant PVC Service Inspector's Comments. 13? List vents, other than emergency types, which are vented to the atmosphere and not connected to a recovery system, and which are suspected of having had short and/or long periods of emissions exceeding the limits specified in Sections 61.62, 61.63, 61.64 and 61.65. Vent Identification Location In Plant Location of Nearest Monitoring Point(s) FUGITIVE EMISSIONS 14. Investigate and witness the- plant's standard operating procedure [61.65(c)] for fugitive emission sources [61.65(b)(1), (b)(2), (b)(5), (b)(6) and (b)(7)] and list any possible deficiencies. Pertinent Fugitive Emission Source Location In Plant Possible Deficiency 69 SAL 000049950 15. Are plant personnel following the established standard operating procedures? Ves No 16. If the response to the above is "NO", list pertinent fugitive emission source, location in plant and action required where the established standard operating procedure is not being followed. Pertinent Fugitive Emission Source Location In Plant Action Required 17? List equipment, location in plant, identification in the process of any pump, compressor and agitator which is not equipped as seal-less or with a double mechanical seal or double outboard seal, and does not duct emissions through a control system, or maintain proper pressurization between seals or equivalent. Equipment Location In Plant Identification In Process Inspector's Comments 18.* Incoming raw material received by: Pi peline Truck Rail 19* Finished product material shipped by: Pi peline Truck Rail 70 Barge Barge ^ ,-.Q *-A' I STACK EMISSIONS 23, List plant's performance specifications of stack emission continuous monitoring instrumentation: a) Stack identification b) Mean value, ______________ , calculated from a series of absolute measurements made by using the equipment specifications and procedure of reference Test Method 106 (see APPENDIX A); c) Number of measurements used to calculate the mean value d) Accuracy, percent of the mean value; e) Calibration error, gas mixture value; ______ percent of each calibration f) Zero drift (2 hr.), ________ percent of calibration span; g) Zero drift (24 hr.), _______ percent of calibration span; h) Calibration drift (2 hr.). percent of calibration span; i) Calibration drift (24 hr.), percent of calibration span; j) Response time. (Time required from the insertion of a known vinyl chloride concentration gas sample into the stack and the stack instrument indicating a value 95% of the known vinyl chloride concentration). 24. If more than one stack was tested, and if the other test data were significantly different from the above data, use attach sheets to provide the information obtained from other stack tests. 25. Were the stack emission tests made under conditions of maximum production rate? Yes No If the response is No, give the percent of maximum production rate under which the stack emission tests were made:percent. 26. Were all stack samples analyzed within 24 hours? Yes No If the response is No, give elapsed time, to the nearest hour, from taking sample to its being analyzed:hours. 0000A'7 ? 72 27* List deviations or substitutes from reference Test materials and procedures which are required in rnna ^od '106 equipments emission test. acting the stack _ .......... , Equi pment/Maten al s/Procedure Identification ueviacion or oubstitutes 1 From Test Method 106 Inspector's Camments ute INPROCESS WASTEWATER an; 28. Identify any inprocess wastewater stream and location in plant which is mixed with another water stream prior to the reduction of vinyl chloride concentration to 10 ppm or less. an; Inprocess Wastewater Location Identification Inspector's ion Stream Identification In Plant of Process Step Comments Je jn REACTOR OPENING LOSS 29. Briefly describe the plant's procedure to determine the emission due to opening the reactor. 73 SAL (X 0 04 74 ON REVIEW OF RECORDS i NESHAPS INSPECTION FOR VINYL CHLORIDE EMISSIONS COMPLIANCE Review Date_________________________ Reviewer's Name____________________ Firm's Name_____________________________ Firm's Address_______________________________ '_________________________________________ EMISSION (STACK) 1. Review of hourly Dyes Dno average from continuous stack emission. 2* Number of stack emissions (continuous emissions for 1 hour or more) which exceeded limits specified in Sections 61.62, 61.63 and 61.64: 3. Were there additional emissions which were not properly and accurately reported in the appropriate Semi-annual Report? Q Yes ONo 4. If the response to No. 3 is Yes, list-date, time, emission point location, duration, estimated integrated emission, and the cause or causes of emission. (Attach sheets if more space is required.) Date & Time Location In Plant Duration Est. Integrated Emission Cause or Causes Action Taken LEAK DETECTION SYSTEM 5. Review of (check one or more) 0hourly Q shift Qdaily monthly average from leak detection system. weekly * Data may be available from records on file at EPA Regional Offices. 74 6. Number of times one or more monitoring point* exceeded the value for the leak definition nation rLi,uc^tel!mIss10n (See CHECKLIST 5 (d): levels 7. List date, time, monitoring point location, duration, estimated intearated emission and the cause or causes of emission of those leaks in Nn * ateG (Attach sheets if more space is required.) * b* Date & Time Location In Plant Duration Est. Integrated Emission Cause or Causes Action Taken 1 ATMOSPHERIC DISCHARGES 8* List information of emergency discharges to the atmosphere. Date & Time Location in Plant Duration Est. Integrated Emission Cause or Causes Inspector's Comments 75 r 9. Determine from plant records whether temperature, pressure, flow rate(s) and/or other process variables and/or if equipment failures ' (i.e., reduced flow rate of cooling water, defective temperature controller, etc.) gave rise to the necessity of the emergency discharge(s) from PVC reactors. List the date, time, location in plant and the condition(s) which appear to have produced the need for an emergency discharge(s), Date & T ime Location In Plant Conditions 10. List the dates and time when similar or equivalent conditions existed as in No. 12 and for which an emergency discharge was not reported. Date & Time Location In Plant Conditions 76 e(s). 11. Review the leak detection reco d the location(s) and down stream !n50ths "nltorf,,, points nearest wofheeremecrogenndciytiodnisscheaxrisgtee.d Which anno 3 on thro,s\e\ dafoc * ^est emission levels of the nearest monitorino nn^lQn Plant and til 12. List the date, time, location in plant from the tabulation in No. 14 where the owner/operator appears not to be in compliance with the NESHAPS vinyl chloride standard and its amendments. Date & Time Location In Plant Inspector's Comments 1A'u1 ',,J (.J. 0 4 9 9 '; o 77 REACTOR OPENING LOSS 13? Review analytical records of vinyl chloride concentration in reactor vapor space to determine "reactor opening loss" of reactor (and stripper where applicable), prepolymerization and post polymerization vessels. List date, vessel identi fication, and batch identification where emissions exceeded standard. Da te Vessel Identification Batch Identification Inspector's Comments RESIN, SLURRY, WET CAKE AND LATEX SAMPLING 14? Review analytical records of vinyl chloride concentration in polyvinyl chloride resin, slurry, wet cake and latex to determine the weighed average residual vinyl chloride concentration in all grades of poly vinyl chloride resin processed through the stripping operation on each calendar day. List date, vessel identification, and batch identifi cation where emissions exceeded standard. Date Vessel Identification Batch Identification Inspector's Comments --o .^ 0ov!' avA-- 78 RECORDKEEPING 15. Review and assess the recordkeeping as re the listing below, and comment on each with1r6d 1n the stan(Jard in form and ease of reference. aspect to completeness. ;:;AL 000049960 79 EQUIPMENT PRE-TEST CHECKLIST FOR (TEST METHOD STACK 106) EMISSION TEST NESHAPS INSPECTION FOR VINYL CHLORIDE EMISSIONS COMPLIANCE Pre-Test Meeting Date_____________________ Inspector1s* 3N4am5e__________________ Finn's Name_____________________________ Firm's Address 1. Probe a) Is probe made of stainless steel, pyrex glass, or teflon tubing? b) What is temperature of stack? c) Does probe have glass wool plug? Yes No Sample line a) Is sample line made of teflon? Q Yes No b) Is a new unused piece used for each series of bag samples? Q Yes Q No 3. Quick connects a) Are 2 male and 2 female connects used? Yes No b) Are they made of stainless steel? Yes 0 No c) Does the pair for the bag have ball checks? 0 Yes 0 No d) Are they assembled as required? 0 Yes No 4. Rigid container a) Is container leak proof? 0 Yes 0 No Unknown b) Does it have a cover to protect contents from sunlight? 0 Yes 0 No 5. Sampling bags a) What material are bags made of? ______ b) Are bags of 100 liter capacity? 0 Yes 0 No 0 Unknown .,o6v on ,-.^0^ ' ST 6. Needle valve Will needle valve allow proper adjustment of sample flow? Yes No 7. Vacuum pump a) Is pump of the leak-free type? Yes Q No b) Does pump have a minimum capacity of 2 liters per minute? 0 Yes 0 No 8. Charcoal tube Does a charcoal tube follow pump to prevent admission of vinyl chloride to atmosphere? 0 Yes No 9. Flow meter Does the flow meter have a capability of measuring flow range from 0.10 to 1.00 liter per minute? 0 Yes 0 No 10. Pitot tube and manometer a) What type of pitot; tube is used? b) Is pitot tube attached to probe? 0 Yes No c) Will an inclined manometer be used? 0 Yes No 11. List substitutes for equipment and materials required in Test Method 106 in conducting stack emission tests. * 12. Is plant in compliance with Test Method 106? Yes No 13. If the response is No, list action items which must be completed prior to Test date in order to conduct stack emission tests: Action Item 1:____________________ : 2: 3: 14. Test Date: SAL 00004 82 `"'-"sYuV'f . RES ^ ' CHECKLIST FOR VINYL IN PROCESS WASTEWATER, WET CAKE AND LATEX SAMPLES (TEST METHOD 107) NESHAPS INSPECTION FOR VINYL CHLORIDE EMISSIONS COMPLIANCE Inspection Date Inspector's Name_______________________ Firm's Name______________ ______________________________________ _ Firm's Address___________________________ 1. Sample bottles a) Are the sample bottles of 60 ml (2oz) capacity? Yes No b) Do the bottles have waxed lined screw on top? Yes No c) Do the bottles have electrical tape or equivalent to prevent loosening of bottle tops? Yes No 2. Vials a) Are vials of 50 ml capacity? Yes No b) Are they equipped with sealed Teflon faced Tuf-Bond discs for water samples? Yes No c) Are they equipped with seals and caps, Perkin-Elmer Corporation No. 105-0118 or equivalent? Yes Q No Unknown 3. Analytical balance a) Is it capable of weighing reproducibility to +0.001 gram? Q Yes Q No Unknown 83 0000^99 b) What is the weighing span in the region of weight that is used in Test Method 107?_____ 4. Syringe a) Is its capacity 100 Yes No Unknown b) Is the model Precision Series "A" No, 010025 or equivalent? Yes No Unknown 5. Vial Sealer a) Is the Model, Perkin-Elmer No. 105-0106 or equivalent? Yes No Unknown 6. Gas Chromatograph a) Is the Model, Perkin-Elmer Model F-40 head space analyzer No. 104-0001 or equivalent? Yes No Unknown b) List substitutes used for the following: 2 m x 3.2 mm stainless steel column; - contains n.',* carbowax on carbopak A (or Carbopak B) Perkin-Elmer No. 105-0133 or equivalent: 7. Thermometer a) Range 0 to 100C, with accuracy + 0.1C, Perkin-Elmer No. 105-0109 or equivalent. Yes No 8. Sample Tray Thermostat System a) Perkin-Elmer No. 105-0103 or equivalent. Yes No 9. Septa a) Sandwich type, for automatic dosing, 13 mm, Perkin-Elmer No. 105-1008 or equivalent. Yes No 84 10. Integrator - Recorder a) Hewlett-Packard Model 3380A or equivalent. Yes No 11. Filter dryer assembly a) Perkin-Elmer No. 2230117 or equivalent. Yes No 12. Soap Film Flowneter a) Hewlett-Packard No. 0101-0113 or equivalent. Yes No ! ! ii 000049966 BAL 85 APPENDIX A: MEAN VALUE CALCULATION The mean value calculated from the reference method (Test Method 106) test data measurements is used as a norm to assess the stack emission continuous monitoring instrumentation. The mean value of the data set is calculated according to the following expression X 1 n (A-l) where X- = The i--absolute measurement obtained from reference Test Method 106, n ^ = Sum of the n- absolute measurements, i=l n * Number of absolute measurements, J = Mean value. QO i}\Vk; 86