Document e5VOw621YzVo0akLzE12Q7rxe

Natural Resources Defense Council, Inc. New York Office 1 * * EAST 4*ND STREET NEW YORE. N.Y. 10l6B *>* 949-0049 1350 NEW YORK AVENUE. N.W. SUITE JOO WASHINGTON, D.C. 80005 202 783*7800 Western Office 15 KEARNY STREET SAN FRANCISCO. CALIF. 941 415 4*1-6561 In Re: National Emission Standard for Vinyl Chloride Petition for Reconsideration of the Withdrawal of the June 1977 Proposed Amendments and- Comments on the January 1985 Proposed Amendments 50 Fed. Reg. 1182 (Jan. 9, 1985) Prepared on behalf of Natural Resources Defense Council and Environmental Defense Fund by David D. Doniger Senior Staff Attorney Natural Resources Defense Council March 25, 1985 100% Recycled Paper CMA 01528? .Wu- England Office; 850 boston post road * sudbury, ma. 01776 617 237-0472 Public Lands Institute: 1720 race street Denver, co. 80206 * 303 377-9740 On January 9, 1985, the Environmental Protection Agency (EPA) published a notice of proposed revisions to the 1976 national emission standard for the hazardous air pollutant vinyl chloride. Included in the notice was the withdrawal of amendments to strengthen the standard, proposed in 1977 pursuant to the settlement agreement in Environmental Defense Fund v. Train.1 The Natural Resources Defense Council (NRDC) and the Environmental Defense Fund (EDF) strenuously object to the withdrawal and hereby petition for reconsideration of it.2 The notice also announces proposed revisions principally concerning requirements for relief valve discharges, as well as certain other matters. NRDC and EDF submit that these revisions substantially weaken the current standard, and we hereby file comments in opposition to them. I. The Withdrawal of the Proposed 1977 Strengthening Amendments: Petition for Reconsideration A. Background In 1976, EPA promulgated a standard for vinyl chloride (VC) which was significantly compromised by consideration of factors that EPA is not authorized to consider under Section 112 of the Clean Air Act. EPA rejected certain requirements which would have further curbed VC emissions either on the basis that the agency 1 No. 76-2045 (D.C. Cir., filed Nov. 19, 1976). 2 Because the withdrawal appears to be a final agency action, NRDC has filed a petition for review in the Court of Appeals for the District of Columbia Circuit. NRDC v. EPA, No. 85-1150 (D.C Cir., filed Mar. 8, 1985). CMA 015290 2 deemed them not to be adequately demonstrated or deemed their cost to be too high. Certain control requirements were rejected even though they were already in use on some sources, or even though they were nothing more than reasonably forecasted improvements that the sources were fully capable of making. These technological and economic tests are inconsistent with the "ample margin of safety" requirement of Section 112(b)(1)(B). Because EPA had not complied with the requirements of Section 112, EDF filed a petition for review in the U.S. Court of Appeals for the D.C. Circuit.3 After significant discussions between the parties, the case was settled with an agreement under which EPA obligated itself to propose amendments to strengthen standard in specified ways. At the heart of the agreement was EPA's recognition of a "zero emission goal" to be approached as closely as possible, and EPA's agreement to propose an immediate strengthening revision and to undertake a review of the standard in three years leading to a further strengthening revision. As part of the compromise settlement, EDF agreed to dismissal of this specific case. EDF, however, conceded none of its legal objections to the construction of Section 112 advanced by EPA in the 1976 standard. B. The 1977 Proposed Amendments The improvements to the standard contemplated by the settlement agreement were published on June 2, 1977.4 EPA 3 See note 1. 4 42 Fed. Reg. 28154. CMA 015291 3 specifically recognized a "zero emissions goal." EPA restated its conclusion (with which we agree, of course) that VC, as a carcinogen, has no known threshold of effect and endangers public health at any level of exposure. The agency then stated: In order to assure that the standard continues to approach the only level of emissions which is known to be absolutely protective of health, namely zero emissions, EPA is proposing amendments which require more efficient use of existing control technology at existing plants, and which encourage technology to reach this goal without banning vinyl chloride.5 EPA proposed the following specific requirements: (1) To lower the emission limit for emissions from equipment used in VC and polyvinyl chloride (PVC) production from the level of 10 ppm promulgated in 1976 to 5 ppm. The proposed 5 ppm limit would take effect immediately for new sources and within three years for existing sources. (2) To eliminate the special 0.2 g kg limit for oxychlorination reactors as it applies to new reactors and require new reactors to meet the 5 ppm limit. This could be accomplished, EPA found, by using oxygen, rather than air, as the feed. (3) To strengthen PVC stripping requirements for new PVC resins by limiting residual VC after stripping to one fourth the levels allowed by the 1976 standard. (4) To require a VC emissions offset when a new source of VC emissions is located within 5 miles of an existing VC source. (5) To review and strengthen the standard within three years of promulgation of these amendments. 5 Id. at 28154. 015292 4 As we discuss below, EPA's own studies and other evidence clearly demonstrate that these requirements are achievable.6 C. The Withdrawal EPA, however, has now rejected the proposed amendments for a combination of reasons that are not legally cognizable and that ignore the evidence in the record. The January notice states that the June 1977 proposal is being withdrawn because the proposed amendments are not now considered "appropriate." The notice then reviews each of the requirements proposed in 1977 and rejects them on the basis of a combination of two types of arguments: (1) that the technology to accomplish these requirements allegedly has not been demonstrated, and (2) that the requirements are allegedly not justified on a cost-benefit basis.7 We first address the legal problems with this notice, and then the evidentiary problems. 1. Legal Violations (a) Failure to Propose the Withdrawal The withdrawal was announced as a final action, without being preceded by a proposal. This course of action contrasts sharply with the course of action followed by EPA in the case of benzene. When EPA decided to withdraw its benzene proposals for three source categories last year, the agency issued a proposal and offered the public an 6 In the years following the proposal, EDF repeatedly urged EPA to promulgate the amendments. See letter from Robert Rauch, EDF Staff Attorney, to Douglas Costle, EPA Administrator, Feb. 3, 1978; letter from Rauch to David Hawkins, EPA Assistant Administrator, Aug. 15, 1978; letters from Larry Corcoran, EDF Staff Attorney, to Costle, Apr. 18 & June 13, 1980. 7 50 Fed. Reg. 1183-86. CMA 015293 5 opportunity to comment. NRDC and EDF object to the agency's refusal to do so in this case. (b) The Role of Cost Considerations. EPA opens its discussion by stating its current view of the basis for the current standard, as well as the basis which guides EPA's current actions. EPA states: The current VC standard was established based on judgments concerning the costs and benefits of the standard to society. The standard is not designed to eliminate VC exposure risk entirely. Rather, it strikes a balance between public health protection and the cost of that protection.8 This balancing of costs and benefits, however, is contrary to the requirements of Section 112 of the Clean Air Act. There are at least_three decisions of the D.C. Circuit whichreject assertions that cost considerations are relevant or permissible under a "margin of safety" statute such as Section 112. The D.C. Circuit has twice held that under Section 109(b)(1) ofthe Act (which requires EPA to set NAAQSs at the level which protects public health with "an adequate margin of safety"), standards may not be compromised by cost considerations.9 In addition, interpreting the "ample margin of safety" requirement in Section 307(a) of the Clean Water Act, the D.C. Circuit rejected the claim 8 Id. at 1183. Lead Industries Ass'n v. EPA, 647 F.2d 1130, 1148-51, 1153 (D.C. Cir. 1980), cert, denied 449 U.S. 1042 (1980); American Petroleum Inst. v. Costle, 665 F.2d 1176, 1185-86 (D.C. Cir. 1981), cert, denied 102 S.Ct. 1737 (1982). CMA 015294 6 that EPA may consider costs when establishing an ample margin of safety standard for a toxic water pollutant.10 Likewise, the Supreme Court has at least three times rejected efforts to imply the relevance of cost-benefit considerations under statutes which direct agencies to base decisions on public health or environmental protection factors.il Acknowledging the legal limitations expressed in the terms of the law and their legislative history, EPA has from time to time nonetheless tried to defend the use of cost-effectiveness or costbenefit analysis under Section 112 with the argument that Congress did not mean what it plainly said. According to this argument, the Congressional policy choice was, in EPA's view, so unwise that it should be ignored. In its place, EPA substitutes the agency's own policy choice. This sort of administrative attempt to rewrite a law with which the agency does not agree has been uniformly rejected by the courts. In all six of the cases cited above, the Court of Appeals and the Supreme Court have emphasized that neither an agency nor a court may second-guess Congressional limitations on the role of cost considerations under health and safety statutes. EPA cannot violate this restriction. 10 Hercules, Inc, v. EPA, 598 F.2d 111-12 (D.C. Cir. 1978). H Union Electric Co. v. EPA, 427 U.S. 246 (1976) (no requirement will be implied for EPA roconsider costs when reviewing a state implementation plan requirement adopted to meet a primary standard); TVA v. Hill, 437 U.S. 153 (1978) (no "reasonableness test to""be implied into the Endangered Species Act); American Textile Mfrs Inst, v. Donovan, 452 U.S. 490 (1981) (no cost-benefit test under the Occupational Safety and Health Act). CMA 015295 7 (c) Technology Demonstration Requirements. In rejecting the proposed lowering of the 10 ppm emission limit to 5 ppm, EPA states simply that based on experience to date, "10 ppmv represents the lowest level of control which has been consistently achieved."12 Before addressing this statement empirically, we must take issue with its legal premises. EPA apparently presumes that Section 112 of the Act establishes a requirement for demonstrating that a level of control has been "consistently achieved" in the past. While a demonstration requirement under Section 111, there is no such requirement under Section 112. None of the key words of Section 111 which establish such a test for new source performance standardsl3 are present in Section 112. The Supreme Court made clear in Union Electric that the omission from sections of the 1970 Clean Air Act of words relating to technological and economic showings reflects a deliberate Congressional choice to which EPA and the courts must give effect.14 Sections of the law where such requirements are lacking are "technology-foreing." In the Court's words, allowing considerations of technological infeasibility claims: 12 50 Fed. Reg. at 1184 (emphasis added). 13 An NSPS shall reflect "the degree of emission reduction achievable through the application of the best system of continuous emission reduction which (taking into consideration thecost of achieving such emission reduction and any nonair quality health and environmental impacts and energy requirements) the Administrator determines has been adequately demonstrateed for that category of sources." Section 111(a)(1)(C). 14 427 U.S. at . [8 ERC at 2146-47.] CMA 015296 8 would frustrate congressional intent. It would permit a proposed plan to be struck down as infeasible before it is given a chance to work, even though Congress clearly contemplated that some plans would be infeasible when proposed. . . . Technologyforcing is a concept somewhat new to our national experience and it necessarily entails some risks. But Congress considered those risks in passing the 1970 amendments and decided the dangers posed by uncontrolled air pollution made them worth taking.15 The Court's reasoning applies equally to a hazardous air pollutant standard issued under Section 112. EPA's apparent test -- that a control requirement must have been achieved in the past before it may be required for the future -- is also inconsistent with tyhe agency's clear technology forcing responsibility under sections such as Section 202. Section 202(a)(1), for example, instructs EPA to set a standard for a future application date which accomplishes such improvements in controls as are reasonably forecasted by then. EPA has the authority, indeed the responsibility, to project such progress and base the future standard on it. As the D.C. Circuit stated in NRDC v. EPA, the diesels case: "The Clean Air Act requires the EPA to look to the future in setting standards ...."16 And as stated as far back as the 1970 Senate Report: [EPA] is expected to press for the development and application of improved technology, rather than be limited by that which exists. In other words, standards should be a function of 15 Id. at 268-69. [8 ERC at 2151]. 16 NRDC v. EPA, 655 F.2d 318, 328 (D.C. Cir. 1981), cert, denied sub nom General Motors v. Gorsuch, 16 ERC 1616 (1981). CMA 015297 9 the degree of control required, not the degree of technology available today.17 The 1977 proposal, which contemplated setting a 5 ppm emission limit with a three year leadtime for compliance by existing sources, is precisely this sort of standard. If technology-forcing authority clearly exists under sections of the Act which are not driven by health considerations and which explicitly provide for considering costs, then certainly such authority exists under Section 112, which is driven solely by the mandate to protect public health with an ample margin of safety. EPA recognized this in 1977, but is denying it now. The withdrawal is based on an incorrect statutory construction and is therefore invalid. (2) Conflict with the Evidence When examined in accordance with Section 112's requirements, the evidence in EPA's own support documents will not support EPA's rejection of the 1977 proposed amendments. (a) The 5 ppm Standard. The bases given for withdrawing the proposed 5 ppm emission limit are: (1) that the 5 ppm limit "was not based on data for control technology different from that analyzed" in 1976; (2) that it was opposed by industrial commenters on the grounds that it would require maintaining an average level lower than 5 ppm, and that it would result in what they (and apparently EPA) view as "little" overall emission reduction; and (3) that the commenters objected to the "zero emission goal. "18 17 S. Rep. No. 1196, 91st Cong., 2d Sess. 24 (1970). 18 50 Fed. Reg. at 1184. CMA 015298 10 EPA then stated that the 5 ppm limit was withdrawn because it had not been "consistently achieved" on existing facilities. As stated above, this is not the legally correct test. And when the record is examined, it fully supports the achievability of the 5 ppm limit by new sources upon commencement of operations, and by existing sources within three years of promulgation. In contrast to the agency's current position, the 1977 proposal stated that the 5 ppm limit would not require the installation of new equipment. Rather, its purpose was "to force owners and operators to maximize the effectiveness of existing control systems."19 The evidence amassed by EPA since then shows that the 5 ppm level is in fact achievable both by better use of existing equipment and by new and upgraded equipment. The document "Vinyl Chloride - A Review of National Emission Standards"20 states: "Primary control devices [e.g., incineration, solvent absorption, refrigeration] are reducing emissions, in most cases, well below the 10 ppm level for exhaust gases. "21 Specifically: o A series of thermal incineration tests at one plant in Kentucky demonstrated VC levels consistently at 0.26 ppm VC or below, less than three percent of the 10 ppm limit, and less than six percent of the proposed 5 ppm limit.22 19 42 Fed. Ref. at 28155. 20 EPA-450/3-82-003. 21 i<a. at 4-1 (emphasis added). 22 id. at 4-8. CMA 015299 11 o Flares are now rated at at least 90 percent effectiveness in destroying VC, which allows a 5 ppm limit to be met for any streams under 50 ppm VC.23 o Two PVC plants have demonstrated that a double bed carbon adsorptionsystem can achieve compliance with a 5 ppm limit. While the Review states that they are effective in meeting the 10 ppm limit, it also states: "When a probe at the outlet of the bed indicates an approaching 5 ppm level (as an indicator of breakthrough), the waste stream is diverted to the other bed while the first bed is being regenerated."24 This indicates the achievability of the 5 ppm level. o Solvent absorption is available for both PVC and EDC/VC plants. The Review describes a B.F. Goodrich system, improved since the original 1976 support document was written. A 99.99 percent recovery efficiency is reported for this system. This should be sufficient to meet a 5 ppm limit on most, if not all, streams. The solvent is described as "proprietary, commercially available, inexpensive, and reputed to be low in toxicity."25 o Tenneco has developed a process of reacting VC with ozone in the presence of activated carbon which reduces VC to less than approximately 1 ppm from streams containing 10-10,000 ppm VC.56 The availability of these measures belies any claim that the 5 ppm limit cannot be achieved by new sources, or by existing ones given a three year lead time.27 As for industry's argument that the 5 ppm limit would result in "little" overall reduction in VC, these measures could reduce 23 Id. at 4-12. 24 Id. at 4-13 -- 4-14. 25 Id. at 4-14 -- 4-15. 26 Id. at 4-16--- 4-17. 27 It bears emphasizing procedure through which an existing source could show EPA that the 5 ppm limit was still not achievable and obtain an extension or a higher limit. CMA 015300 12 the estimated 680 kg/yr from a 68 Gg/yr PVC plant by half.28 Given the carcinogenic nature of VC and the requirement of Section 112 for protection of public health with an ample margin of safety, this reduction cannot legally be foregone. As for the industry commenters objections to the "zero emission goal" toward which the 1977 proposed amendments would go, the ample margin of safety provision requires no less. (b) Oxychlorination Vent Standard. The proposed 1977 amendments would have required the oxychlorination vent to meet the 5 ppm level on new sources. The 1977 proposal states that this can be accomplished by the use of oxygen as a feed instead of air. The 1977 proposal also cites studies showing that this has been found to be economical.29 The 1985 withdrawal notice cites no discussion of the use of oxygen as a feed. We can find no such discussion in the VC Review either. EPA has failed to justify the withdrawal of this proposal. (c) Stripping. The 1977 proposal would have cut the residual VC limit for new dispersion resins from 2000 ppm to 500 ppm, and the limit for other new resins from 400 ppm to 100 ppm. The proposal stated that EPA believed most new resins could meet these levels, but if one could not, a manufacturer had the option of choosing not to begin making it. EPA now states that because of difficulty administering the proposed distinction between new and existing resins, the agency chose to examine whether all 28 VC Review at 4-6, Table 4-4. 29 42 Fed. Reg. 28155, 28157. CMA 015301 13 resins could be stripped to these lower limits. (The January 1985 notice reflects no effort to change or fine-tune the "new resin" definition so that it would work.) The notice states: The review study found that resin stripping has improved since the current standard was promulgated, and that some processors can achieve lower resin residual VC levels than those required in the original standard. In certain cases, some resins can meet the more stringent levels specified in the previously proposed amendments. But because it has not been shown that all resins can at this moment be stripped satisfactorily to the lower levels, no levels lower than those specified by the current standard have been "demonstrated," and no lower level will be required for any resins, new or old.30 This is, first of all, an application of the wrong legal test. There is no such demonstration requirement. The Review study in fact shows that an impressive percentage of PVC production can already meet the lower limits, and that much of the remaining production is close.31 if EPA set this target out for new facilities, or for existing ones with a three year lead time, it clearly could be achieved simply by copying the equipment and procedures used by the leading facilities now. D. Conclusion For the reasons given above, NRDC and EDF submit that the withdrawal is not supported by the law. Nor is it supported by the evidence, when it is evaluated according to the requirements 30 50 Fed. Reg. at 1185. 31 See Review at 4-55 --4-56, Table 4-11. CMA 015302 14 and limitations of the law. We therefore request that EPA reconsider the withdrawal and publish the proposed amendments. This should be done, we submit, no later than at the time final action is taken on the new amendments proposed in January, to which we now turn. II. The 1985 Proposed Amendments; Comments The principal feature of the proposed "administrative revisions" to the standard is the change in the relief valve discharge requirements. Because the effect of this change is to substantially weaken the current standard, NRDC and EDF submit comments in opposition below. We also comment on two other changes: the amendments proposed for the reporting requirements, and the le3c . detection and elimination requirements. A . Relief Valve Discharges 1. The Proposal to Grant an Allowable Discharge Rate The current standard bars all relief valve discharges except in "emergency" situations. An emergency does not exist if the discharge is "preventable." Currently, EPA enforcement personnel examine relief valve discharge reports made by companies to determine if they were preventable through the use of available equipment or procedures. If they are found to be preventable, they are not considered emergencies and constitute violations of the standard. In recent years, enforcement of the relief valve discharge standard by EPA and the Department of Justice has picked up. CMA 015303 15 While it is not all it could be, it is a substantial effort. It has been a very successful effort so far as well. The proposal states, however, that primarily because of a "significant" use of agency resources in analyzing the preventability of discharges, EPA has chosen to revise the standard to permit an allowable number of relief valve discharges per facility per year. PVC reactor discharges would be allowed 0.35 discharges per 100 polymerization batches (with an upper limit of four allowed discharges per year for the suspension resin process). Nonreactor sources would be allowed 0.25 discharges/100 batches, with an upper limit of three per year. Continuous process units would be allowed 1 discharge per year. EDC/VC plants would be allowed four discharges per year. NRDC and EDF strongly oppose granting the industry an allowed, "free" rate or number of discharges. As a matter of principal, all preventable discharges must be avoided. They are the single largest remaining type of VC emissions, accounting for thousands of pounds per event, and tens of thousands of pounds per year at some sources. A single discharge can release tens of thousands of pounds of VC. See, e.g., the abysmal record of releases at the Formosa plant in Delaware. EPA enforcement policy states that releases are presumed preventable unless a source demonstrates otherwise. To so demonstrate, the source has to show that neither proper employee training, inspection and maintenance of equipment, proper design and operation of control equipment, nor installation of all controls needed to meet the standard would have sufficed to CMA 015304 16 prevent the discharge. If a discharge has occurred before, that fact is to be taken into account in deciding if subsequent discharges could have been anticipated and preveneted. Proper instrumentation and controls are required before a release can be deemed unpreventable. Likewise, the injection of short-stop chemicals must have been tried where possible. In addition, enforcement policy treats a release as preventable if a firm could have used a gasholder system to contain releases until they could be treated.^2 As a matter of empirical fact, the proposed discharge allowances constitute a significant weakening of the standard. EPA preventability analyses, conducted pursuant to this long standing EPA enforcement policy, have routinely resulted in the conclusion that the vast majority of releases are preventable. The very support document on which EPA relies for the proposed changes to the relief valve requirements itself indicates that the new allowable release rates will exceed what would occur under the existing standard and enforcement policy. For the purposes of developing these rates, EPA's contractor. Radian, applied the following criteria: Based on information gathered from plant visits, contacts with regional offices, 10-day compliance reports, and vendors, it was judged that certain types of discharges can be 32 See, e.g., Memorandum from Director, Division of Stationary Source Enforcement, to Lawrence Goldman, Chief, Enforcement Branch, Region I (Apr. 24, 1979), re: "Vinyl Chloride Relief Valve Discharges from PVC Reactors-Borden Chemical." See also Memorandum from Director, Division of Stationary Source Enforcement, to Stuart Roth, Attorney, Enforcement Division, Region I (Feb. 28, 1978). (These memoranda are attached as Appendix A) CMA 015305 17 prevented by reasonable measures. For example, discharges caused by operator error, due to operator negligence or failureto follow standard operating procedure (SOP), were considered clearly "preventable." In addition, operator errors resulting from insufficient training or lack of established SOP were judged clearly "preventable". . . . Another example of "preventable" discharges is discharges due to premature releases from relief devices. . . . Finally, discharges recurring for the same reasons as previous discharge incidents were considered "preventable."... Although other causes for specific discharge occurrences may be preventable, this assessment would need to be done on a case-by case basis and would require, in many cases, more information than is available in the 10day compliance reports. Thus a more refined preventability assessment can not be made with the available data. For the purposes of this analysis, only the general discharge categories described above are identified as clearly "preventable. "33 In other words, were the current enforcement policy to continue to be followed under the current regulation, more discharges would be found to be preventable. Additional control measures besides those mentioned above would be applied. Further indication that the proposed allowable discharge rates are inappropriate comes from the statements, in the Relief Valve report, that these rates would be complied with at present by the bulk of the industry.34 In other words, these levels 33 Memorandum from Reese Howie and Karen Fidler, Radian Corp., to File, re: 'Vinyl Chloride Standard - Numerical Limits for Relief Valve Discharges" (Apr. 16, 1984) at A-6, attached as Appendix A to "Vinyl Chloride: Relief Valve Study," EPA-450/3-85002 (emphasis added). 34 Id. at A-6 -- a-7. CMA 015306 15 would legitimize the current discharge rates of the industry, regardless of the ability to lower them further. The VC Review Study, in fact, describes a variety of additional control measures beyond those mentioned in the Relief Valve report. There have been significant advances in the use of instrumentation and computer controls, and in the development of short-stop chemicals and in the development and use of systems to insert them into reactors which are overpressurizing. Gasholders, about which more will be said below, are in use in at least three of the five plants surveyed for the Relief Valve report, and are capable of containing many, if not all, discharges until they can be treated by primary control devices.35 Several companies' effective preventative systems are described in the VC Review Study.36 Conoco, for example, has developed a hilling agent capable of preventing discharges under worst-case conditions.37 The only reason offered thus far for the change is the fact that the present requirements entail use of allegedly "significant" agency resources in case-by-case preventability analyses.38 To our knowledge, however, EPA has presented no 35 See VC Study, 4-18 -- 4-40; Relief Valve Report, 4-2, Table 4-1. 36 VC Review Study, 4-40 -- 4-48. 37 Id. at 4-43 -- 4-46. 38 EPA also states that the current standard leads to "uncertainty" for the industry, as it does not know exactly what discharges EPA will consider preventable. This appears to us to be a highly salutory uncertainty; the goal of the current standard is to come as close as possible to eliminating relief valve discharges and the uncertainty about risk of being in violation of the standard tends to stimulate greater efforts to prevent CMA 015307 19 analysis to support this resource concern. Moreover, even granting that the case-by-case analyses consume some resources, we submit that EPA cannot legitimately resolve its resource problem by weakening the standard. Rather, if EPA lacks the resources to carry out its duties, its obligation is to tell Congress what it needs to do the job right. 2. The Failure to Require Gasholders NRDC and EDF submit that EPA should be strengthening the standard to specify use of all available equipment and procedures that can prevent discharges or, should they occur, prevent or minimize the escape of VC to the atmosphere. Foremost among the control measures not now required by the standard is a gasholder system. As mentioned above, EPA's support documents show that such a system is feasible and is fact in use on a number (at least three) of facilities. Just this month, the State of Delaware entered a consent agreement with Formosa Plastics under which that company has become obligated to install such a system at its plant.39 EPA might advance two reasons for not requiring the use of a gasholder system on each plant. First, EPA might deem the cost to be too high. As demonstrated above, however, cost considerations may not be considered under Section 112. Moreover, it is difficult for NRDC and EDF to accept such a contention from EPA's regulation-writing side in light of the fact that the agency's discharges. 39 Delaware v. Formosa Plastics Corp., C.A* 84C-DE-73, Para. I (B) TDe. Sup- Ct, consent order filed Mar. 13, 1985). CMA 015308 ' 20 (and the states') enforcement side has proved capable of achieving such requirements in consent orders. The second reason EPA might advance is a contention that some facilities may be able to prevent discharges almost entirely without such systems. To this we respond that no facilities will be able to eliminate such discharges entirely; there will continue to be "emergencies" that are not judged "preventable." There is no reason to allow these emissions to reach the atmosphere, even if the discharge valve releases themselves are unpreventable. For these releases there remains a vital purpose served by gasholder systems. NRDC and EDF therefore request that EPA reevaluate its position on gasholder systems and related equipment, and promulgate regulations requiring their installation and use. B. Leak Detection and Elimination NRDC and EDF do not support the proposed adoption of the generic leak detection and repair program promulgated in June 1984 in place of the current leak detection and elimination programs for VC, as it appears that the effect of this change will be to weaken the standards. EPA states that the regulated firms will have the choice whether to stick with their current programs or to adopt the generic one; that option will predictably be exercised only by companies which find their current programs weaker than the generic one. Firms with stronger requirements will likely switch to the generic program. One place where the weakness of the generic program versus the current ones can be seem is the definition of a "leak." The CMA 015309 21 generic program, and the proposed revisions to the VC standard, would define a "leak" as a portable monitor reading of 10,000 ppm or more. But current leak detection and elimination programs developed by the companies and approved by EPA incorporate much lower "leak" definitions. The VC Review Study reports that the "leak" definition under the current programs ranges from 300 ppm to ppm.40 These levels are from three to 0.05 percent of the proposed "leak" definition. There is no justification for such a change. We request that EPA prepare and air for comment a point-by point analysis of the current leak detection and elimination programs as they compare to generic program proposed for adoption. C. Reporting Requirements NRDC and EDF object to the proposed changes in reporting requirements. In particular, we object to the proposed repeal of the 10-day requirement for reporting relief valve discharges, and the proposed substitution of a quarterly reporting requirement. Frankly, we are surprised EPA would consider such a reduction in the reporting obligation at this time, in the wake of the Bhopal catastrophe. The 10-day reporting requirement serves to give EPA, state officials, and the public contemporaneous knowledge of releases of this carcinogenic compound. Such notice could air in the prevention of repeat incidents. It also serves the public's right 40 4-72, Table 4-14. CMA 015310 22 to know what it is forced to breathe. Finally, allowing a longer interval for reporting these releases would tend to promote a perception that a reported release is merely "past history." We submit that taking these sometimes enormous releases with the seriousness they warrant requires immediate notice. Ain fact, we urge EPA to tighten the 10-day requirement by requiring notice to EPA within 24 hours. Such a requirement would be more consonant with the requirements and spirit of the CERCLA reporting requirements, about which more is said belcw. NRDC and EDF also object to the proposed repeal of the requirement to report all test results, not just those that are in excess of standards. An affirmative requirement to report all results promotes accuracy and completeness. It also would permit EPA to build a data base from which to evaluate possible revisions of the substantive requirements of the standards during the periodic reviews mandated by Section 112. D. CERCLA Reporting Requirements The Vinyl Institute has filed comments questioning the , applicability of CERCLA and its reporting requirements, which are discussed in the proposal.41 In our view, the releases of VC from relief valve discharges are clearly reportable under CERCLA, except to the extent that they are the result of genuine emergencies. Any non-emergency, preventable release is not permitted by the current standard (and should not be under any revised standard). Such a release is not a "federally permitted 50 Fed. Reg. at 1193. CHA 015311 23 release" and therefore is not exempt from the CERCLA reporting requirements. D. Conelusion For the reasons explained above, NRDC and EDF object to the proposed changes in the relief valve discharge standards, the failure to propose and promulgate gasholder systems and other effective controls, the proposed changes in the leak detection and elimination requirments (such as the definition of a "leak"), and the reporting requirements. Ill. Conclusion NRDC and EDF request that EPA re-examine the withdrawal of the proposed standard for the reasons given in our petition for reconsideration contained in Part I of this submission. We also request that EPA reconsider and modify its proposed changes in the VC standard in conformity with our comments set forth in Part II. CMA 015312 `i i *-s 24 APR 1979 memorandum cf' rt<fc*.:r SUBJECT: Vinyl Chloride Relief Valve Discharges from PVC Reactors-Boraen Chemical FROM: Director Division of Stationary Source Enforcement TO: Lawrence M. Goldman, Chief, Enforcement Branch Region I This is in response to your memo of March 29, 1979, requesting guidance on what constitutes a "preventable" relief valve discharge under 40 CFR 61.65(a) anc on what steps a source can be expected to take in order to prevent rue cr rdcv'''r're of relief valve discharges. When a source reports a relief valve discharge we should presume that the discharge was not due to an emergenc but was preventable and is therefore a violation of Section 61.65(a). The source will then have the opportunity to demonstrate otherwise. In order for a discharge to be considered an emergency, the source would have to demonst:at that it could not reasonably have been expected to antioioot the discharge and then to prevent or contain it. Re a minimum., the source would have to demonstrate that the discharge could not have been prevented by implementing any of the following procedures: 1) employee training programs inciudin on emergency procec-res. 2) equipment inspection and maintenance prog: a 3) proper design-end"operation of process and control equipomeenntt, and 4) installation and operation of all control equipment needed to comply with the vinyl chloride standard. CMA 015313 2. One factor which should be taken into consideration in determining whether a discharge could have been anticipated and therefore prevented, is the plant's prior history with respect to discharges. If the source has previously experi enced one or more discharges, some or all of which were caused by the same factor, the source is expected to have taken corrective measures designed to prevent or contain future discharges. In summary, a relief valve discharge is a violation of Section 61.65(a) if it could have been anticipated and preventive measures could have been taken or if the discharge could have been prevented by properly training employees or by properly operating, maintaining anc inspecting equipment. In particular, your memo requested guidance on what constitutes o preventable operator error. As is the case with other discharges, we should presume each discharge to have been preventable and then provide the source with the opportunity to demonstrate otherwise. As a minimum, the source would have to demonstrate that operators were veiltrained initially and had received refresher training courses, as necessary, to cover both normal and upset conditions. Refresher courses should have been provided particularly after the occurrence of ar. initial relief discharge caused by operator error. The source should be able to provide documentation as to the cates training was offered and the operations covered. The final issue to be addressed in this memo concerns the types of actions a source should be expected to take, after experiencing a relief valve discharge, in order to prevent future discharges. The preamble to the proposed vinyl chloride standard lists, on page 5953S, several measures a source can reasonably be expected to take to prevent relief valve discharges. Measures which can be taken to prevent discharges from FVC reactors include, but are not limited to, the following: 1) properly instrumenting the reactors to detect upset conditions, 2) injecting chemicals to stop the polymerization reaction during upset conditions, 3 3) venting the reactor contents tc c csrholder during upset conditions and ultimately tc a recover., system, 4) providing employees with improved training or preventing and handling upset condi trcr.s, and 5) maintaining a backuo source of mower. FR 59539, December 24, 1975.* See 4C It is apparent from this discussion that in developing the vinyl chloride regulations we envisioned the use of gasholders to prevent or contain relief valve discharges. Therefore, Borden Chemical and any other PVC manufacturer can reasonably be expected to install gasholders to prevent discharges if other preventive measures, implemented in a timely manner, fail. In fact some PVC manufacturers have already installed gasholders for this purpose. Should you have any further questions on this issue, please contact Libby Scopino at 755-2564. Edward E. Reich cc: Susar. Wyatt, ESED Marsha Spink, Region 1 Marcus Kantz, Region II Peter Schaul , Region III Leon Folsom, Region IV Bruce Varner, Region V Martin Brittain, Region VI Paula Eissor., Region IX t* 4ll i CMA 015315 ( I'MTr.? HT/;res E.hvinchniu:,'": al pgg WASHING TON. D C. 2C .10 GZ >-1. CFFi:: C- t:->T.5-::':-T sur-orG "KG"; TC: Relief Valve Discharges PPG Tncilitv at Guavanilla, Puerto Rico Director Division of Stationary Source Enforcement Stuart Roth, Attorney Enforcement Division, Region II This is in response to your memo of January 1C, 1C7C, requesting guidance on enforcement options available tc or event relief valve discharges of VCK from vinyl chloride storage tanks at the Guayanilla PPG plant. V.:e believe that the most effective approach to elimi- u- . a. ^ ,.c.. .1 -- -- Vi uJ t.y w 4. ^ * i w w w*. j- *- w. - efforts. v;e feel that this position is justified by the preamble to the proposed standards which states, with respect to relief discharges from equip;..ant other man \ t t c j j4 y 5 V I l \ I I ft l ) I i i Cne possible option which coo Id-sat isfy the above requirea-er.ts and which combines options listed ir. ycur nemo, v.-ould be installation of additional refrigeration units designed to servo as backups for the existing units ir, combinetion with a cos holding tank which could contain any releasor, resulting from total re f_r i g e r = t i c r. failure cr free ether equipment failures or deficiencies, including any releases w.-shic':; -o-c--c--u---r during t.ne time it takes tc switch over frc~ a r a 1 f cr. cn : u n-ing refri; C. _ ,, V, . . W...W ~ 'w' usebum unit. Cf course , ?PC '..'ill ho responsib~leo tFoo*- c2C*C 1 C IT i T;C n-j r C- *. tc t CMA 015316 9 u r~ i In your sent you request clarification cf t: tubes an "emerge :y relief discharge". As stated ir. nemo of December 15, 1S77, the preamble to the props: vinyl chloride regulation describes an emergency c;s c:, n r g as one which cannot be avoided by taking; preventativ C* P/j [2 ~ surcs, such as those caused by natural disasters (40 rn 5953!), December 24, 1975). Natural disasters includ q h u r r i - car.es, tidal waves, earthquakes, etc. Discharges wh ich could have beer, prevented will not be considered "er. ere; enev discharges". Finally, you request guidance on the en f cr cenevnt approach to be used to require installation of equipment designed to minimize the amount and frequency cf discharges. As you knew, an administrative order (Section 113(a) order) is an inappropriate method to remedy a HECHAP violation unless the order requires immediate compliance with the standards. In some circumstances, a waiver of compliLr.ee nay be issued which requires the installation of equipment necessary for ccmpl 'i ance, wi th the standard if the equipment can be operational and the source can be in compliance by October *21, 197S. Pote that the issuance of a waiver of compliance is discretionary, and consideration should be given to both the source's good faith and whether vc belicw. the terms of the waiver will be complied wish and final com.pl i i.r.ce ultimately achieved. In addition tc ar.y rc:y ire* m.ents designed to protect public health during the period cl the waive: (4C CFP. C 61.11 (b ) ( 3 ) ) and any cth-.-r require:-cats general. applicable, any such waiver should elude- v f. r v detailed incremental dates for design and in:-ml 1 a t ion cf the enuinmer.t The waiver should be structured .sc that the first increment will become due- as soon after . ssuance as possible. This will ensure prompt action while still allowing E?A sufficient time to act should p?c fail tc comolv. Violation of the- wrivc-r requirements erne, cf course, resu.t ir, :r.e c: encemanu o: il cr criminal action. However, if the source has not.-requested a waiver to illov: instill t ion of the requires eg uipnor,, if the source has r.ot acted in good faith, or if the Region believes it is impossible for the equipment to be installed and the source to be in com,piiante by October 21, i?"c, civil artier should be conn meed seeking injunctive relief for t -gi-ditious installation c: the necessary control ecuIg~:rt. Ir. light c i me r,i.;:r;:.5 nature- of the pollute.-.! t> 4,1 I,' . v 4 L b k v 1 , ^ ,Ct " ^ ^ '' 10, we ::,a : t ion , vc car, air -: s mu r l : i- vr c 1 ci r . 51 i- ' 1r i CMA 015317 '2'J rwcil'-' any further as a i a t '.nee J n this : clease fce.1 free to contact Cojfj rprr.i';cr:l; \ 7 3 5 -- 2 5 7 C ; regard: any legal questions or Lien/ Scuyino (7SS-21C'd) regarding any technical questions. uIul:.. O- O ' /j h Edward Z. Rarer. cc: Susan Wyatt, ESED t'.arcus Rente, Region II CMA 015318 956 613 FEDERAL REPORTER, 2d SERIES sioner declines to define a substance as a "food additive," though it comes within the strictly literal terms of the statutory defini tion, he must state the reasons for exercis ing this limited exemption authority. In context, a decision to apply the literal terms of the statute, requires nothing more than a finding that the elements of the "food addi tive" definition have been satisfied.*7 In the case at hand, the Commissioner made specific rulings that the component element of the definition was satisfied with, respect to acrylonitrile beverage containers having an RAN level of 3.3 ppm or more. These rulings were premised on a projection, based on an extrapolation from reliable data, of migration of acrylonitrile monomer in then-undetectable amounts. In light of the supplementary submission made in response to the post-argument inquiry of this court, we find that the determination can be made for the 3.3 ppm RAN contain ers with an appropriate degree of confi dence, and with the support of the required quantum of evidence.1* [12,13] Turning to the safety element of the definition, the Commissioner deter mined that the scientific community had insufficient experience with acrylonitrile to form a judgment as to safety. Based on this lack of opinion, the Commissioner made a finding that acrylonitrile was not general ly recognized as safe within the meaning of the statute. The Commissioner acted with in his discretion in making such a finding, but we note that the underlying premise may be affected, perhaps weakened, per haps strengthened, with time and greater experience with acrylonitrile* This find ing on the safety element will be open to reexamination on remand at the discretion of the Commissioner. He would have lati tude to consider whether acrylonitrile is generally recognized as safe at concentra tions below a certain threshold, even though he has determined for higher concentrations that in the view of the scientific community acrylonitrile is not generally recognized as safe. V [14] Petitioners also made a claim of discriminatory treatment--that the Com missioner is applying policies in the peti tioners' case that have not been applied in other similar circumstances. However, there is no claim that the Commissioner was motivated by discriminatory intention to bring the petitioners before the agency and to focus on their product. Petitioners came before the agency in the ordinary course. Once the Commissioner undertook scrutiny, he shifted the lens of his microscope to a higher power--but that is no ground for objection, so long as the final action re mains within the legitimate scope of discre tion. The decision of the Commissioner is af firmed in part, and in part is remanded to provide the opportunity for reconsideration. So ordered. O 5 WY NUMBER SYSTE*l'> f~r n r - r i f 27. Absent a showing of bad faith or other ex traordinary circumstances, a court will not con sider meritorious the claim that the Commis sioner has abused his discretion in declining to exercise his exemption authority for de minimis situations. This is an area of decision by its nature committed to the informed discretion of the Commissioner. 28. See note 20 supra. On judicial review, thi court must be satisfied that the Order of the Commissioner is based "upon a fair evaluation of the entire record." 21 U.S.C. 5 348(f)(2). (gX3) (1976), The Commissioner applied the "component" part too automatically, and in the future must support his decision with more than a conclusory reference to the diffusion principle of the second law of thermodynamics. 29. Like the "component" element of the defini tion, the "safety" element may at times call for more rigorous examination. Thus, the Com missioner has discretion in determining when the statute applies to a given substance, but substances that do fall within its term should be so identified. VG t E Uni D Whil Heart, L Institute dlscrimin trict Cot George L and plair. peals. A1 Judge, sit statistical fade cast an indivi diacrimin; tistical ev parity be ployees ir tions at o an emplo; female ei units was facie cast Revet MscK and filed 1. CivU R Title of present ment disci has been r prove tha sion on leg illegitimat persuasion must be g denoe that i nt evidence has been te time has been prond that the record ra-; in this instance, cord did not reflect ive judgment: FDA istrative proceedings it dictated by an or- o concerned that the 'e reached his deterf that he was constrictly literal terms :ctive of the public ^derations. As we ' latitude inherent in 0 avoid literal applir definition of "food : situations u^Pint of the Cem ent no public health 1 general doctrine of serves the public inotection of the conthis case production ;e containers was dei when this court isA order, and in any ited pending further rned that the record toward a precipitate 1 factors bear on our text of the decision, on as to basis. Anhe beverage containnmissioner was charon level well below t It was offered by data generated in re inquiry, petitioners no ation does occur from " container (RAN level nditions of its intended Ja ote 16 supra ; i, 5 supra, at 11. hstitute v. Federal ReVpp.D.C. 311. 322, 551 4 MONSANTO CO. v. KENNEDY 955 OUM4I3F44M7 (1S7S) petitioners in the hearing as available as a sion principle, a mere finding of any contact result of ongoing technology, but the time whatever with food. Petitioner's conten constraint imposed by judicial mandate pre tion on this point is sound. vented the agency from scheduling the kind of administrative consideration that would ordinarily have been provided. For the component element of the defini tion to be satisfied, Congress must have intended the Commissioner to determine with a fair degree of confidence that a IV Pretermitting various issues that should await conclusion of the remand proceedings, we turn to certain other important ques tions that are presented by the record, that have been fully briefed and argued, and that are ripe for resolution.11 The statute requires a demonstration of safety precedent to FDA approval of any "food additive." B The statutory definition of "food additive" which triggers that re quirement contains a two part test. First, the component element of the definition states that the intended use of the sub stance must be reasonably expected to re sult in its becoming a component of any food,14 Second, the safety element of the definition states that the substance must be not "generally recognized [as] safe under the conditions of its intended use." * substance migrates into food in more than insignificant amounts. We do not suggest that the substance must be toxicologically significant; that aspect is subsumed by the safety element of the definition. Nor is it necessary that the level of migration be significant with reference to the threshold of direct detectability, so long as its pres ence in food can be predicted on the basis of a meaningful projection from reliable data. Congress has granted to the Commissioner a limited but important area of discretion. Although as a matter of theory the statuto ry net might sweep within the term "food additive" a single molecule of any substance that finds its way into food, the Commis sioner is not required to determine that the component element of the definition has been satisfied by such an exiguous showing. The Commissioner has latitude under par ticular circumstances to find migration "in [7-9] Petitioners are concerned that the significant" even giving full weight to the Commissioner has determined, or will deter public health and welfare concerns that mine, that the component element of the must inform his discretion. !i definition may be satisfied solely by that [10,11] Thus, the Commissioner may de application of the second law of thermody termine based on the evidence before him namics called the diffusion principle: any that the level of migration into food of a I t - ^ - two substances that are in contact will tend particular chemical is so negligible as to to diffuse into each other at a rate that will present no public health or safety concerns, ir^j be determined as a function of time, temp- even to assure a wide margin of safety. erature, and the nature of the substances. This authority derives from the administra Congress did not intend that the component tive discretion, inherent in the statutory requirement of a "food additive" would be scheme, to deal appropriately with de min satisfied by a mere recitation of the diffu- imis situations.** However, if the Commis- 22. The issues fully ripe for decision at this time Volkswagenwerk, A. C. v. FMC, 390 U.S. 261. include questions of statutory interpretation 276-77, 88 S.Ct 929, 19 L.E<L2d 1090 (1968): that will be pertinent to the proceeding on remand. United Glass & Ceramic Workers v. Marshall, 189 U.S.App.D.C. 240, 242, 584 F.2d 398, 400 13. See 409(c)(3)(A), quoted in note 3, supra. (1978); Marine Space Enclosures, Inc. v, FMC, 137 U.S.App.D.C. 9, 16, 420 F.2d 577. 584 24. See note 2 supra. (1969). Cf. Ingraham v. Wright, 430 U.S. 651. 674, 97 S.Ct. 1401, 51 L.Ed_2d 711 (1977). 23. Id. Sniadach v. Family Finance Corp., 395 U.S. 337. 342, 89 S.Ct 1820, 23 LEcL2d 349 (1969) 24. See, e. g., FPC v. Texaco, Inc., 417 U.S. 380, (Harlan, X, concurring). v 399, 94 S.Ct 2315, 41 L.Ed.2d 141 (1974); rt CMA 015320 * 954 613 FEDERAL REPORTER, 2d SERIES ever, the Commissioner would have* latitude to issue a statement of policy based upon the results of the proceeding or remand that would specify what in his review was an acceptable RAN level. This would serve a technology-forcing objective.1* [4,5] FDA opposes petitioners' post-ar gument motion for remand, asserting that the proffered new evidence will not affect the Commissioner's order insofar as that order precludes manufacture of beverage containers with RAN levels equal to or greater than 3.3 ppm--the type of container already tested. FDA points out that the material submitted in response to this court's inquiry affirmatively supports the validity of the Commissioner's findings and conclusions.1* FDA contends that a peti tion for modification of the regulation, or a similar procedure, would be the appropriate vehicle for presentation of any new evi dence indicating that migration ceases when RAN levels fall below a certain threshold. As a general rule, courts defer to admin istrative agency orders closing the record and terminating proceedings. The rule has applicability in cases involving scientific matters notwithstanding the possibility that advances and experiments will yield new material data. Indeed, the importance of finality as a matter of administrative neces sity may be magnified by the possibility-- indeed probability--of advance in at least some areas. Procedures for rehearing or modifying orders are generally available to provide appropriate relief from any hard ships or other harm.11 The general rule of finality applies in the usual case because the courts trust the ad ministrator's ability to make a reasoned judgment that sufficient evidence has been submitted, that adequate time has been pro vided for rebuttal, and that the record should be closed. However, in this instance, the closing of the record did not reflect unfettered administrative judgment: FDA conducted these administrative proceedings under a time constraint dictated by an or der of this court [6] The Court is also concerned that the Commissioner may have reached his deter mination in the belief that he was con strained to apply the strictly literal terms of the statute irrespective of the public health and safety considerations. As we discuss below, there is latitude inherent in the statutory scheme to avoid literal appli cation of the statutory definition of "food additive" in those de minimis situations that, in the informed judgment of the Com missioner, clearly present no public health or safety concerns. In the usual case, the general doctrine of necessity and finality serves the public in terest in immediate protection of the con suming public. But in this case production of acrylonitrile beverage containers was de ferred voluntarily even when this court is sued a stay of the FDA order, and in any event it is now prohibited pending further proceedings. Finally, we are concerned that the record reflects a momentum toward a precipitate determination. Several factors bear on our judgment One is the text of the decision, with its lack of precision as to basis. An other is the fact that the beverage contain er evaluated by the Commissioner was char acterized by a migration level well below the agency's initial limit It was offered by ally before him in the remand proceeding or any subsequent proceeding. 19. The submission by Monsanto is that no mi gration can be expected to occur from contain ers with RAN levels lower than 0.1 ppm. Salame Affidavit, note 16 supra, at 1) 10. There is a further indication that the manufacture of beverage containers with RAN levels of less than 0.1 ppm is technologically feasible. Mon santo Memorandum, note 15 supra, at 7. 20. In view of the new data generated in re sponse to the Court's inquiry, petitioners nc longer contest that migration does occur from Monsanto's "Cycle-Safe" container (RAN level of 3.3 ppm) under the conditions of its intended use. See Salame Affidavit, note 16 supra: Monsanto Memorandum, note 15 supra, at 11. 21. See Investment Co. Institute v. Federal Re serve System, 179 U.S.App.D.C. 311, 322, 551 F.2d 1270, 1281 (1977). CMri 01532