Document dQyGjadER1e8XGYnnynJGvRp0

ABD00078269 FROM UISTA TIMBERUAY ' VWo Ch*micql Company 06/24/85 11:23 P. 3 15990 N.Bwfct/t landing Rd. PoflOWc* Box 19029 Heurtonjaxai 77224 Phon* (713) 531-3200 V l March 21, 1985 \' t Central Docket Section (A-130) U, S Environmental Protection Agency 401 M Street S. V. Washington, DC 20460 Attention Docket Number A-81-21 Gentlament Vista Chemical Company ("Vista0) la taking this opportunity to comment on the 1/9/85 proposed Vinyl Chloride NESHAP's. Vista Chemical owns and operates a vinyl chloride monomer plant and, through its subsidiary Vista Polymers Inc., two polyvinyl chloride plants. On July 20, 1964 these plants were purchased from Conoco. Inc,, which had operated the plants since the Vinyl Chloride NESHAP's was promulgated. Most of the plants' staffs are former Conoco employees. These comments are thus based on eight yeare experience under the Vinyl Chloride NESHAP's. Vista has worked through the Vinyl Institute to assist EPA In this rulemaking. Representatives of Vista have met with EPA over the past two years providing technical assistance and Vista provided commente on the standard at the August 1984 NAPCTAC hearing. In previous meetings and at NAPCTAC ve expressed an opinion that the Leak Detection and Elimination (LD&E) provisions of the current NESHAP's is adequate to protect human health and the environment. Obviously, by the 1/9/85 proposal, EPA Intends to push Subpart V In addition to the existing LD&E program. We believe EPA Is overstepping its authority under the Clean Air Act by requiring compliance with Subpart V. We believe Subpart V Is being arbitrarily Imposed on an Industry without any justification. A desire by the Agency for consistency is not justification for the additional cost and administrative burden of Subpart V. The standard writers argue that proving existing LD&E programs equivalent to Subpart V requires little effort. Our experience has been that Regions make very strict comparisons when evaluating equivalencies and that an inordinate number of man-hours is required to prepare an equivalency request. Vista is skepcleal that the Regions will approve effective programs that are not as administratively tedious as Subpart V. The only justification for imposing the additional financial and administrative burden of Subpart V is for a demonstrated health effect* On page 1183 of the preamble, EPA states that the Carcinogen Assessment Group (CAG) did not propose a change in the vinyl chloride unit risk number after reviewing new health studies, The preamble goes on to say that none of these new studies had sufficient exposure information to warrant a refinement In tha unit risk factor. It is obvious from the preamble language that the basic information needed to jueelfy a more ! R.OM UISTA TIMBERWAY ABD00078270 06/24/85 11:24 P. 4 burdensome fugitive emission control program Is not available. No health justification exists, EPA Implies that Subpart V will not have a substantial effect on the VC and PVC plants. They state that this rulemaking is an administrative change leaving the Impression that no justification 1$ necessary. This reasoning Is seriously flawed. Subpart V is more than an administrative change. One of Vista's PVC plants estimates chat it will need to spend $600,000 to comply with the "open-ended valve" provisions of Subpart V, In addition, expenditures will be necessary to comply with the sensor requirements for pumps and compressors. These expenditures will not substantially reduce fugitive emissions since fugitive emissions from pumps, compressors, and open-ended valves are controlled by the existing Leak Detection and Elimination program.* Expenditures of this nature must be justified* EPA has not provided that justification. Vista takes exception to the statement on page 1193 of the preamble where EPA states: "... the proposed administrative revisions do not increase the cost of compliance with the standard..." In the same section under Regulatory Flexibility Analysis, EPA states that no small business is affected by this proposal. A small business is one that has 500 employees or less. Vista Polymers, Inc. has 300 employees and qualifies as a small business under EPA's definition. Additionally, on page 1193, EPA states that these revisions will reduce paperwork by 2.8 person-years. We have no accurate data to refute this, but we are skeptical that EPA has done a comprehensive analysis of the effect of Subpart V on paperwork. Not considered is the paperwork increase imposed on industry. The VC NESHAP's is already heavily weighted with paperwork. Inclusion of Subpart V will add to that burden. EPA has taken a reasoned approach to risk assessment in its decision to not regulate processors, landfills, and other users of vinyl chloride. Although no specifics are presented, this decision had to be based on a judgement that these sources present no significant risk. If EPA had quantitative data on VC fugitive emissions comparing existing LD&E programs with emission after implementing Subpart V, it would likely show emission changes to be less than could be realized by regulating the above sources. Surely, EPA is not targeting a significant source of vinyl chloride when it goes after fugitives.' The data base for Subpart V has little to do with fugitive emission levels In PVC and VCM plants. The vinyl chloride industry is a low leak Industry, Subpart V and its predecessor, the SOCMI Fugitive NSPS, are based on refinery fugitive data. That data vaa gathered after the VC industry had already Implemented Its Leak Detection and Elimination programs. The emission reductions projected for Subpart V are not accurate for the VC industry. FROM UISTA TIMBERUAY ABD00078271 OS/24/85 11:25 Use of Subpart V could result in an Increase in fugitive emissions from VCM and PVC planes, Under Subpart V, a leak Is defined as an Instrument reading of 10,000 ppm. This is much higher than Vista's plants use as a definition of a leak. It can be argued that adoption of the Subpart V definition will result in fewer leaks being repaired than under the existing programs, i The VCM/PVC business is not a lucrative one. The VC NESHAP's does its share to erode the industry profitability. Within the domestic market this probably can be compensated for. However, the industry is under pressure from foreign producers. These producers are not subject toNESHAF's, Vista urges EPA to drop its efforts to incorporate Subpart V into the VC standard. The NESHAP's program is extremely tedious and resource intensive. We know of no other regulatory program that is as pervasive as the VC NESHAP's. The benefit of such stringent regulation has to be questioned. To add Subpart V absent justification perplexes some of us and infuriates others. Although there is no justification for Inclusion of Subpart V In the VC NESHAP's, Vista has reviewed the proposal and has the following specific comments: I, Definition of a Leak - Section 61.61(v) Item 2 within the definition-"...(2) indications of liquid dripping..." should be expanded to read "...(2) indications of liquid dripping followed by an Instrument reading of 10,000 ppm...". In pumps and compressors where the seal liquid la at a higher pressure than the process fluid, failure of the inner seal would result in seal fluid entering the process. Drips from the outer seal would not contain VC normally. Checking the drip with an instrument and repairing If the instrument reads over 10*000 ppm is reasonable. Item (4) needs to be clarified so there is no misunderstanding that this applies to sealless valves, pumps, and relief valves. We are concerned that an enforcement person in the Region or State would use 500 ppm as the definition of a leak for all equipment. Vista takes exception to the comments of the Vinyl Institute regarding the definition of "leak", Section 61.61(w). Item (3) of the definition proposed by the Vinyl Institute states that a leak is also an event defined in an approved Leak Detection and Elimination Plan, The definition of leak in LD&E Plans Is muth less than 10,000 ppm. Item (3) confuses the Issue since there will be two definitions of a leak; the definition In 6i.6l(w) and the one In the LD&E Plan. Under Item (3) of the Vinyl Institute's proposed definition, leaks found as a result of LD&E monitoring would result in the imposition of Subpart V repair and record keeping requirements. This could be tedious where the definition of "leak" in the LD&E Plan is very stringent. The effort to document small leaks as required by Subpart V is not justified. , FROM UI STA TIMBERUAY ABD00078272 OB/24/85 11:26 P. 6 i Vista suggests deleting Item (3) from the Vinyl Instltue'e proposed revision to 61,61(w), This will eliminate confusion and be more fair to those sources with a leak definition in their LD&E plan more stringent than 10*000 ppm. 2. Rupture Disks Under Relief Valves t The current NESHAP's requires rupture disks under relief valves. The proposal should be modified to recognize this as leaklees technology. No monitoring of relief valves protected by rupture disks should be required. Subpart V states chat RV's must have a no detectable level. We believe a rupture disk meets this requirement and the regulation should say so. 3. Leak Detection and Elimination Programs We believe these existing programs to be effective. EPA is apparently less confident. ?o reassure tha EPA* we suggest that plants be allowed to keep their approved customized programs to control fugitives and not do Subpart V valve monitoring. We propose an annual spot check be performed as written in 61.65(b)(8)(11)(B). This would provide EPA with an Indication of an effective program. This concept is suggested In the preamble on page 1191 under the "Leak Detection and Repair Requirements" section. 4. Valve Monitoring - 61.65(b)(8)(li)(E) The preamble states in several places that it is EPA's intent to require a more standardized valve monitoring program and not to change other fugitive emission controls. However* the proposed language in the standard goes beyond this. Vista believes the Intended purpose stated in the preamble 1$ reasonable. If Subpart V is triggered* only the sections applicable to valve monitoring should be required. Implementing all of Subpart V would require capital expenditures to comply with a number of requirements. Although the LD&E portions of the VC NESHAP's are substantially the same as Subpart V they are not precisely the same. We believe fugitives are adequately controlled under both programs. As proposed* Section 61,65(b)(1)(11)(E) requires monitoring in compliance with the valve monitoring of Subpart V within 90 days upon finding greater than 2% of valves leaking. We believe the goal to keep leaks repaired can still be met by allowing the plant to retest its valves within 30 days; if the plant falls the retest* Subpart V valve monitoring must be adopted for one year. Allowing a quick retest would give the plant an opportunity to evaluate its shortcomings and correct them quickly whether they be actual leaks* Instrument or operating problems* or other random errors.. We believe most plantB would quickly solve problems in order to avoid implementing Subpart V valve monitoring, Plante should also be allowed to revert back to the provisions of 61.65(b)(8)(11)(B) after a period of time under Subpart V valve monitoring. We suggest that a year under Subpart V valve monitoring should convince most plants to keep better control of valve leake. Considering the above* we suggest 61.65(b)(8)(ii)(E) be revised to read: FROM UISTA TIMBERUAY ABD00078273 06/24/05 11:27 " (E) Any process unit in which the percentage of leaking valves Is found to be greater than 2,0 percent must retest in accordance with this Section within 30 days. If, after retest* the percentage of valves leaking Is greater than 2,0 percent, the process unit must, comply with all valve monitoring provisions of Subpart V for one year." * 5. Product Accumulators - Subpart V Section 61.242-9 This section should be modified to specify that vents from process accumulators in VttAP service must be controlled. As written* this section could be interpreted to mean that all fugitive sources on a product accumulator must be controlled, We don't believe this is EPA's intent, 6. Multiple Relief Valve Discharges 61,65(a)(4) Additional language should be included in 61.65(a)(4) to assure the multiple relief valve discharges that occur because of one overpressure event be considered only one discharge, E?A states in Che preamble on page 1189: "...for determining compliance with the numerical limits, discharge frequency is to be recorded on the basis of individual discharges except when simultaneous discharges occur from relief valveB on the same piece of equipment," Typically PVC reactors are outfitted with several relief valves set to relieve at Increasingly higher pressures. This is considered good design practice since the amount of a release can be better minimised. If necessary more than one of the relief valves could discharge, but not necessarily exactly at the same time since they could be set to relieve at different pressures. EPA appears to recognize this possibility in the preamble quoted above. However, the requirement that discharges be simultaneous should be eliminated. Rather, multiple releases that occur because of a single over pressure event should be considered one release. We suggest that 61.65(a)(4) be revised by adding the sentence: "Multiple relief valve discharges from the same vessel or equipment occurring because of a single overpressure event are considered to be one discharge." 7. Slurry Stripping Calculation - Section 61.67 We believe that the constant K used in the equation to calculate C in 61.67(g)(6)(iii) should be 1,002 (1.002 x I09)y not 1.002. Also, in 61.67(g)(6)(11)(B), PPVA should be PPVC. 8. Fugitive emissions - Section 61.65(b)(8) There is a typographical error in 61,65(b)(8)(ii)(A). The reference should be (b)(8)(ii)(B) not (b)(8)(ii)(C). 9. Reporting * Section 61.70(f) Reporting of design capacity required by 61.70(f) should not be required. This Is proprietary information and should not be in the public record. Plants that strip in reactors using tha test FROM U I STA TIMBERUJAY ABD00078274 06/24/85 11:29 P. 0 proposed in 61.67(g)(6) should not be required to provide this information since It is not relevant for compliance determination. Other plants should be allowed to provide this Information In a separate confidential document. 10, Emission Standard For EDO Plants - Section 61,62 This standard is unclear as to whether tank~area vents are included under this definition, 11. Reactor Opening Loss - Section 61,67 (g)(6) The calculational procedure to determine reactor opening loss ' may necessitate the Installation of more accurate Instrumentation, In that case, sources subject to Section 6lTi>7"'Cg) (6) will likely need some time to select, order, and Install Che additional Instruments. It should be made clear that waivers of compliance are available as allowed by Section 61.11, We urge EPA to consider these comments and contact us if additional information or clarification is necessary. Your decision on Subpart VI is important to our company. We remain ready to discuss these comments and fugitive emission controls at your convenience. Sincerely, / ^'J* C. Ledvina, Director Environmental Activities ajo/003