Document MGoEjoaV7eR9xE6RQZ0eMbrr7

E I5P "? ---, %-s^jjSii ' .-,!*>>>? '-v - ' ';- i-^:r;'':- *-<* ; cV.'v V-*J ' v* * : '-*f'* *$. V '- '-V jX/. >.'- ^MMHHMHa .V-.>u"s ..; -,*;-*: ,*r:Pt>'.v/i w* 'f*i^*rtv Wweeadnneessadaayy .- . - -s*.v^a^sf. January 9, 1985>w.tk, ` ! iwi / vy * W** . *r * ----.fcr.lw-V... v... . .- .- ...,'- `i* w>;< :'te\W:^i-:.: : i?.. .. .: " . : _* i si - - ' -v * .- .w- *.. . . tA-.c .......... l .p ,, . . -.a-* _.3Y'`'v'.Vfeii> *K/*; <iz>- ;.;j; ^*u- ; ' , ^ *. 1 : ''`u:. trt. * * . - -- - - - y >-.-- T**'* 'cJV-V.i ,5 i.'v .;fA'-v.-v *.; V'^- '-vvV.'^PtW!^ 1*"TT? ;"; A; :~ ^ : *' - i*.*-' -'i-V' f-~! 'f*/ *' */*-, * ' 1^ yv.,,.. .. ' V ... ^ jir iv * , > >s `* . ;:i . . . ;: ;V >vi' ;ls* 1 .'Oi '* *"' I '-V. .; . ri-.* ' " ` . - Part \f . nail V ^ -. *.'; - ' 1_ -1-1* ,,-_ . s': r-' V .vS'Vi . *i ' '^'sv- ,, -^V- : Environmental Protection Agency 40 CFR Part 61 7' ' National Emission 'Standards for Hazardous Air Pollutants; Vinyl Chloride; . Proposed Rule and Notice of Public ..' . Hearing ;. > y -.'&i'-r -/ r/.-V-r* ;v.r v *: *: >.v. r % tt'VL* '-1 *,*. IX'r*'/;* < V.VyT' ._ ..*> ;V-. < jL^Vvt'.r-r.'*'- ' ' ''-tA* *. V' ? 1 .+'* .".** . -3*: .v : .* *:i j *, * v* * AP00003607 Federal Register / Vol. 80. No. 6 / Wednesday, January'9, 1985 ( Proposed Rule* 1183 " with VC by requiring reasonable control Regulating Airborne Substances Posing' ' Assetiment Group reviewed new health meaiurei. As etated in the preamble to . Risk of Cancer.'* This proposal - the proposed standard (40 ft 59532, - addressed several lames which were ' December 24,1675), there Is no known ' central to the proposed VC -......- threshold level of effects for VC amendments. It also articulated the . studies that have become available since the standard was promulgated. This review included a study of the .-.vr? estimated carcinogenic strength of VC os' Therefore, the only approach that would EPA's conclusion that Section 112 4oes (the VC unit risk number) and focused _-a - eliminate health risks associated with . not express an intent to eliminate totally on whether this number should be.; ,,-ad VC would ban its production and use, all risks from emissions of airborne changed to reflect new infonnaton. ^s-^a -v This approach was not selected. Rather, . .Carcinogens. The EPA'a selection of the Since the current standard was' .T.-rs ' -' an approach was'seleeted to minimize . level of control for a hazardous air promulgated, new occupational studies jj* the health risks associated with VC by .. use of reasonable control messure. On November 19,1976, the Environmental Defense Fund (EDF) "1 petitioned the United States Court of KMutant emission standard would not have confirmed qualitatively that liver;jj?i based bh a policy that requires zero . and brain cancer incidence are v .' ay-'. -. emissions of carcinogens. This policy is asociated with population exposure to,,,, j consistent with the basis for other atmospheric VC, However, none of '.? recent actions under Section 112. For these new studies have sufficient ~ Appeals for the District of Columbia ; Circuit to review the standard. .On ' 'J.. March 24,1977, the EDF and the EPA ' 7 moved to dismiss the proceedings on the basis of a settlement agreement requiring the EPA to propose example, standards for benzene from coke ovens and leaks from equipment components in benzene service are not based on a zero emissions policy but rather on a reasonable level of control. which considers emissions and health exposure information to warrant a '' -i. ff*; - refinement of the quantitative can rissk estimate., -iky,x-t _ (, a Findings and Coochisioas of the Review, Study amendments which would require increased efficiency of existing control equipment require more stringent control ofnew sources, and prohibit increases in emissions within the vicinity of an existing source due to new construction. The preamble to the proposed amendments was to state that the EPA's policy for regulating carcinogens under Section 112 of the Clean Air Act would Include a general goal of eliminating emissions of carcinogens and that the EPA would risks. .- ... '- . The findings and conclusions of the the EPA believes it is not appropriate VC review study are presented In the ' ??? to leave the proposed amendments to following subsections. Tbs first the VC standard in effect or to , subsection discusses the need and basis ' promulgs te amendments bated on the for the current standard. The eecond'. ,.? v. \ proposed amendments. Therefore, the .. -. subsection addresses the level of control .',* ; Jons Z1977, proposal is withdrawn. As " ' described in the following section of this required by die current standard. The third subsection identifies source ?. '-hLy*~T^ - notice, the EPA began a review study to. -.categoriesnot covered by the; current'. m - , " obtain additional technical data and v Standard and evaluates the cost information and to detennle appropriateness of regulating these .-> !&- ' whether other amendments to the . - sources. 1 standard are needed. New amendments Initiate a review of the VC standard 3 . : developed as a result ofthe review v"V (1) Need aidBoth for Current Standard years after the promulgation of the . amendment*. study are proposed in thi* notice,, . - ^ a The current VC staru^ard was' -'-'.V- On June 2.1877, the amendments were Review of VC Standard established baaed on judgments proposed (42 FR 26154). Many comments pertaining to policy, technological Early in 1880 the EPA began e review. of the VC standard. The primary '' - concerning the costs andbenefits of standard to society. The standard is the not ... feasibility, and procedural aspects of the purpose of the review was to investigate designed to eliminate VC exposure risk . ..... - proposed amendments were received. the adequacy and appropriateness of the entirely. Rather, it strikes abalance .. Review of these comments indicated that additional technical data and cost standard in light of policy decisions, health studies, control technology ' i 4>etween public health protection and the cost of that protection. Data , . ^ information were required before the developments, Bnd enforcement end proposed amendments, or revisions of compliance experience which have . . the proposed amendments, could be - -. occurred since the standard was first (evaluated before the current standard _ ' was established) strongly indicate that"? VC causes or contributes to the < promulgated. , promulgated. The review consisted ofa development ofangiosarcoma, other Meanwhile, the EDF Sled e petition ' screening study of: (1) Existing and new - with the EPA requesting the . > ` 'control technologies, (2) sources not cancers, and various noncarcinogenic '? ; disorders .in people with occupational - i . eetabliehment of a comprehensive .SV; regulated by the standard, and (3) '* ? exposure and in animals with .o^-rv program for regulating airborne carcinogens under Section 112 of the Clean Air Act The aspects of the EDFi petition concerning the development of . enforcement and compliance experience since promulgation of the standard. 'Information and data evaluated during ' this study were obtained through experimental exposure to VC Although ' no doss-response data are available at * tiie concentrations ofVC found in the - > ambient air, the EPA concluded when . 'i' standards under Section 112 were " literature searches, plant visits, and the standard was established that any . 'V similar to those proposed in the June Z * interviews with industrial ~ atmospheric concentration of VC poses... 1877, amendments to the VC standard. representatives and EPA regional ;. some public heath risk. To eliminate the . . Baaed on the similarity of the proposed : personnel involved in enforcement and risk of VC exposure entirely, a complete amendments and tha EDFs requested , _ surveillance of the VC-emitting prohibition of all VC emissions would ;, comprehensive program for regulating - industries. The information and data are be necessary. This would require the - 't> - airborne carcinogens, the EPA believed presented in a document that may be closure of ths entire industry and result that it should not take final action on the obtained a* described in the aDOSEtaIS iii serious, adverse economic impacts. proposed VC amendments until after it section of this preamble. Decisions .Furthermore, the EPA concluded at the ' had acted on the EDF# petition. based on this review are summarized in time the current standard was :., '< * : On Ootober 10,1876 (44 FR 58642), the the next two sections of this preamble. ; established that a complete prohibition ' EPA proposed "Policy and Procedures . As another aspect of the review of the of all VC emissions would not be for Identifying. Assessing, and VC standard, the EPA's Carcinogen desirable or necessary. The EPA AP00003608 ..lies :*f- ^Federal Register / Vol. BO. Wo.'6 / Wednesday, January 0,1535 / Proposed Rule* --,, volumt of inart substances in the vent include blend tanks,'dryers, centrifuges, processors meeting the more stringent ^f stream and, consequently, the cost for ,- * storage allot, begging operations; anq _ - limits proposed previously are stripping t \v >< supplemental fuel required for -. ' sny sources following foe stripper. f these resins to this low level to offset inoineretion. Comments received on this Control of these emissions is based on ' ` amissions from those resins which are V proposed amendment focused primarily either stripping the PVC resin to e ' more difficult to strip. Without this on the high expense and large energy specified (based on retin type) residual ability to average the emissions and 7,*^ f; requirements associated with die n-*i VC level ((,, 400 ppm for suspension, reductions among resins, these , ** The review shu^Hdentified no control' bulk, solution, and latex resins:'and processors might not achieve the current. `,44...., 2,000 ppm for dispersion resins] or' -' '. standard.Exemptingteeingrades. ,;}> technology for oxychlorination vents at controlling the emissions from all -J-'!"'; known to be difficult to strip isnot .1^. ci n EDC/VC plants thathed not been > sources following the stripper with a / & feasible because these resins cannot tin- considered during the development of ' control device. The 1677 proposed <' * readily be-defined. For foe foregoing .j-..,' the original standard. Additionally, the -' amendments would have required "new reasons, foe SPA has concludedthat .'7... EPA reevaluated the cost ofretrofit reslns"`to be stripped to lower levels there is no demonstrated level of control Incinerator controls and reached the':- - (i.e,, 100 ppm tor suspension, bulk, : which could significantly and - ame conclusion drawn in the * V ~'~ solution, and latex mine; and 500 ppm 1 consistently reduca residual VC levels r.' development of the original standard. ` for dispersion resins). When the in resin* to levels below that required , As before, the high cost associated with1 amendments were proposed, foe EPA -- by foe current standard. Therefore, foe . incinerating oxychlorination vents at believed that tome resins could meet the EPA is not Investigating further foe . existing EDC/VC plants makes this level proposed limits; whereas, for other' -* control of foe combined sources after of control unreasonable. Thus, the ..... retins foe manufacturer would have - , Stripping. v;!* 4-"*--^ currant standard of 0.2 g/kgEDC is' been required to develop improved considered still to be the most .r;-x: stripping technology or not to produce Equipment Leaks. Because little was , known about leak detection and reasonable level of control for existing . foe retin. elimination programs for control of oxychlorination vents. In addition, the industry comment* Stated that most'' equipment leaks from components to VC review study concluded that significant,. dispersion, copolymer, and bulk resin* ' -service, specific requirements forteese new construction or modification of ' would sufferdegradation ifmore EDC/VC plants is not expected. At this ; . stringent emission limits were imposed. time, only one new EDC/VC facility is ' Additionally, the commenter* noted the reportedly planned. (BP Goodrich has v inherent difficulties in defining e "new _ programs were not included in foe . . . current standard. Instead, each plant - ; - was required to institute and implement a formalized leak detection and plans to construct an EDC/VC facility in resin." Information submitted by elimination program Incorporating both . Convent, Louisiana.) Oxychlorination commenters indicated that minor - vents at new EDC/VC plants will be adjustments to resin compositions are * regulated by the proposed standards of . made routinely, and completely new ' performance for air oxidation processes -. resins are rarely, if ever. made. As a (40 CFR Pert 60 Subpart III) or by the .. result of these comments, foe EPA .: - r BACT or LAER requirements of new .. i concluded that it it impossible in many ' source review regulations applicable in cates to distinguish between new end specific locations to a level comparable existing resins end still hive any resins to that achievable through the use of covered by foe proposed amendments. incineration. Because the Further, the proposed amendments did , technologically achievable level of ' not address wbst levels of control could control is assured through the current be achieved by improved stripping requirements, the EPA concluded that technology. For these reasons, the EPA investigation of additional control (be* , chose to evaluate whetherhigher levels incineration) was not required for of control are achievable for all resins. , oxychlorination vents. . or only for some special classes of . fixed-point monitor and a portable monitor. Plant-specific program* were ' -:( ; subject to approval by the *'4 Administrator. Coniequently, due to site-specificdifferences among plants, ' aa well as variatiana in leak definition* . and monitoring practices, differences in control of equipment leeks among foe plants have resulted. Since foe standard was promulgated, foe EPA hae obtained - more information pertaining to foe - control of equipment leaks from ' components in VC service. With the - information obtainedform foe development of other standards, an Reactor Opening--0.02g/kg PVC Product The current VC standard . restricts emissions during polymerization reactor openings. The' resins. ....... effective leak detection and repair The review study found that resin ' stripping technology has improved since the currant standard was promulgated. ,, program based on use of a portable monitor can now be specified for - --equipment covered by this program. The, : standard was based on reactor purging and on a reduction in the frequency of reactor openings. An increased level of control was not proposed in the 1077 and that tome processors can achieve lower resin residual VC levels than , those required in foe original standard. In certain cases, soma resins can meet . .. specific leak detection and repair " f. requirements ere discussed to the : f-. Administrative Revision* section of this - ' ' preamble.' .7,1 ' amendments. (The level of control the more stringent levels specified in foe Relief Valve Discharge Standard. provided by the current standard, previously proposed amendments.' .,.. Sources ofVC emissions covered by this 0.02 g/kg of PVC product reduces VC ' However, other processors standard include discharges from relief emissions to about 1.36 Mg per year for.' ` manufacturing resins of differing grades - vplves on pressure vessels, transfer e model PVC plant.) During the review' end characteristics can only marginally ' lines, snd other equipment in EDC/VC , of the standard, no technology was - comply with foe original standard. . and PVC plants. The standard is based identified that would provide additional Because of foe wide variation in resin ", ' on emission control by a combination of . VC reductions beyond the level of the ' i currant standard. Therefore, the EPA is grades and characteristics, it cannot be ' - equipment and process modifications. . . concluded that, even though a particular and operational procedures, found in not investigating further foe control of resin made by one company can meet a plants during development of foe - ... f.'. reactor openings. particular level, any other resin or standard. An exact combination of 47- ' Combined SonnetAfter Resin r similar resins produced by another . modifications end operational .. I'.xY Stripping. The sources of VC emissions company could also meet that level procedure* was not specified. Instead, a . covered under the current standard Furthermore, in some cases these performance standard (i.e., an emission AP00003609 T Federal Register / Vpl. SO, Wo. B / Wednesday. January fl. 1985 / ProposedRuJea' m transfer lines, and other equipment in' .- of decreased discharge* over the hardware end operational procedures ^ EDC/VC and PVC pleats. The control compliance period. Following an initial implemented by each ofthe plants alongi ,J'_ techniques considered as the but* of drop tn relief valve discharges after the . with the attitudes adopted toward, ..`-lif. the standard involve a combination of standard went into effect the frequency ` preventing relief valve dischargee' '>. ,^v< 'U equipment modifications, process r-i.-.;- .t and quantity ofrelief valve discharges represent the types of control measure* modifications, and operational by EDC/VC plant* have decreased that the standard intended. In particular.. procedures. An exact combination of. slightly or remained relatively constant ' the EPA ooncluded that the low modifications and operational ,.> ^ - GeneralBatitforNumericalLimits. frequency of discharges by the visited; >' procedures was not specified in the ->t In selecting the proposed numerical ' plants was indicative oftheir degree jjv current standard; rather, a performance * limits, EPA first evaluated in detail the " effort to prevent relief valve discharges. standard (Le, an emission standard) t, recent performance (1981 to 1983) of five.: Consistent with the goal ofthis "V.v >" was established because different - ....... PVC plants and one EDC/VC plant proposed revision, the EPA decided that combinations of the modifications and .; These plants ware chosen based an ' an alternative numerical emission limit ^ procedures were expected to be equally ' discussions with EPA Regional Office based on performance resulting under ,,,> -: effective in controlling relief valve - personnel and industry and were .'1 the current standard could be revised in ..^Tif discharges. . ...-i:%. -T intended to represent plants with good .' a format that would be easier to ; --'l Bssed on B years of enforcement and. relief valve discharge records. In . ; understand by snforcemenl and Industry' T/l/V *- compliance experience, the EPA has ; general, the EPA's evaluation of these V. personnel . vj*-=-? concluded that the relief discharge --i.v plants indicate* that each has adopted ,, - The EPA investigated two basic ways ..ttj standard hat resulted in: (1) Significant the combination of equipment' \ ' ''i ^ of expressing relief valve discharge [k1' reductions in the frequency and quantity operational procedures and attitude performance for PVC plants. One format - of VC discharges from relief valves, (2) toward prevention ofrelief discharge* is bised on mass emissions, for significant use of agency resources to intended by the current standard, and. -j example, tha pounds of VCdischarged / evaluate individual discharges for \ that their resulting performance is . . ' per million pounds of PVCproduced (lb preventabillty. and (3) uncertainty on consistent with compliance with the VC/MM lb PVC). Baaed oh a review of 3 - t&i the part of producers regarding whether current standard. The EPA'a evaluation methods used by industry to determine they comply with the standard. - 7 found that a few discharges may : j the amount of VCdischarged from relief Additionally, the EPA learned some of continue to occur from tome plants that valves, the EPAwas unable to Identify a - VC and PVC believe that this part of the comply with the standard. This ' . .. . sufficiently accurate method for .r... .,? ;, ' current standard applies only to . ' observation is consistent with the measuring discharge quantities from discharges through safety relief valves ' .expectation held by the EPA when the' ,. . relief valves. At present, produeera-are and that discharges through other standard was written. . <... required only to estimate discharge pressure relief devices, such as rupture In order to revise the standard in i quantities for reporting purposes. ' disks or manual or automatic vent ,;>..; terms of numerical limits representing v .. Demonstration of compliance with a lb . . valves, are not covered. This , V. compliancewith the current format of . .. VC/MM lb PVC limit would require ..-c-'j' interpretation is not compatible with the the standard, this evaluation separated i- producer* to measure the amount of VC - intent behind the current standard. To PVC end EDC/VC plants. For plants, discharged during an incident. Because . provide more efficient enforcement by relief valve discharge performance data 1 suitable measurement method was not decreasing the burden of individual were further separated by source .. Identified, the EPA decided not to <- preventabillty assessments on the EPA, (reactor vs. nonreactor) and by resin redefine relief valve discharge .'-,' - and to provide a better understanding to type The EPA then reviewed the. . -/ V - perfonnance by PVC plants tn a lb VC/ . plant operators of the goal of the .; , performance of 2S additional PVC plants . MM lb PVC format. standard, the EPA is proposing to 5 and 12 additional EDC/VC plants. The Another format is Basdd on the reformat the standard for relief valve EPA reviewed this large set of plant* to . frequency (i~ number per unit time) of discharges and to define the emission - - ensure that the level of performance discharge from occurrences. No method ' points covered by this standard to demonstrated by the evaluated plants . for measuring the amount of VC ~ include appropriately all pressure relief could be achieved by all PVC and EDC/ discharged from relief valve* is needed - devices. As discussed more completely VC plants. because only the occurrence of a release < In the following sections, the EPA is . The numerical Hmits presented hi th* . , is required for this format The >;. proposing to change the format of the "V.. Findings section of this preamble are 1 1 occurrence of a discharge can be I"-'?1'.* `. numerical limits in the standard to '' based on an evaluation of the number of dete_r_m__in_e_d b,y monitor_i_n_g_p_ro__c_e_s_s reflect the number of discharges that - discharges representing the . . parameter* as well as inspecting relief 'v`; occur from those plants complying with demonstrated performance level *r * - ' valve performance reports. Thus, of the' ` die format of the current standard.. associated with compliance with the ' .'-- two basis ways of expressing relief- , The EPA found in the review study provisions of the existing standard. - " valve performance that were that efforts by all EDC/VC and PVC FormatforNumericalLimits. The ' considered, the EPA selected a format " producers to comply with the standard EPA visited the five PVC plants ; j.-. based on the frequency of discharges! . are reflected in their preformance (in ' 1 evaluated in detaiL As expected, the ' Based on this decision, the EPA then .. terms of size and frequency of EPA found differences in the z.; ' considered how the format would be ' ., discharges) since die standard went into 1 combinations ofhardware'and applied to PVC and EDC/VC plants. At"-''; effect, in general, a reduction in the operational procedure* associated with PVC plants, the frequency of discharges reported frequency and size of relief control of relief valve discharges of each from polymerization reactors and valve discharges by PVC producers has ofthe plants. Furthermore, no exact ' associated process equipment may be" -. occurred since 1978. A further decrease relationship we* found between the *v ^ related to the fact that a batch process is'' in relief valve discharges by the PVC effectiveness ofspecific hardware items used to produce most types of PVC. For : ' Industry occurred between 1980 and ' and operational procedures and 1 v batch PVC production processes, the ; ;.': 1981. Performance by the EDC/VC prevention of discharges. In the EPA's opportunity for dischargee is related to industry exhibited a less marked trend Judgment, die various combination* of the number of times a new ~ -V' ' 1. ' *. mm*n' :-r fr AP00003610 Federal Register / Vol BO, No.'6 / Wednesday, January 9. 1985 / Proposed Rules V- .1189 discharge frequencies indicated that "polymerization batch" consists of each operating history, reliefvalve discharge " " plants with the lowest polymerization. sequence of charging VC and other performance by certain plants is ry.^ natch frequencies typically experience - materials to the reactor, heating reactor, expected to be much better than the' ~- . i ii about one discharge in a 12-month - contents, polymerization ofreactor respective limits. For example, toms J'v period. The EPA-concluded that for most contents, and removal (to* blowdown)' new suspension resin PVC plants plants a 12-month reporting period of reactor contents. Any batch that is ' produce about 5.000 batches during a,'y (rolling every 6-montns) was both aborted following charging of VC to the 12-month compliance period. One'end -^:,- suitable and appropriate for determining reactor is nonetheless counted as a w two discharges at one of these plants'v'-ff? compliance with the proposed numerical polymerization batcb in assessing limits. For plants producing only a small - . during a compliance period would result In a discharge performance of032 and -. ..ii? amount of a particular resin (Le, low ..at.-: bulk resin, a single "polymerization - 035 discharges per 100 batches, vyscrf'.k^-' ' number ofpolymerisation batches), an '' batch" includes both prepolymerization respectively. The second discharge .. *. apparent violation of the standard may. ' and postpolymerization reactor ;- - during the compliance period would be --ii' result from a single discharge .. v,. operations. ;'?* '.VVy' f a violation of(he proposed 0.035 occurrence during a 12-month Discharge frequency can be recorded discharges per 100 batches limit despite, 'compliance period as described below. in two ways. Discharge frequency can V the fact that die Bret discharge would ' For a PVC plant producing a single ' be recorded on the basis of discharge ' result in performance well below the resin type to meet the numerical limit for events (involving discharges from one dr limit These types of plants were " reactor discharges (l.eH 0.035 . dischaiges/100 batches), It must experience and average of no more than more relief valves) or on Individual ^ relief valve discharges. Inmost cases, " plants currently report discharges ' '' ' considered in selecting the proposed /**.;' ^< limits and reporting procedures for relief Jg* valve discharge*. The result that plants '-'.V one discharge per 2358 polymerization individually when they occur from relief of this type must perform well below the batches over the preceding 12-month . valves on separate equipment However, .limits in the standard in order to be in.^t*-^ period. An average reactor discharge ' frequency exceeding one discharge per certain equipment such as .-. *,. \- ; compliance is consistent with the- polymerization reactor* that are - " proposed limits, which were selected to' 2,858 batches would bs a violation of the equipped with multiple relief valves may represent an upper boundary on the . i" standard. However, if the plant made experience dischargei simultaneously , number of allowable discharges less than 2,858 polymerization batches / from more than one relief valve. Most intended by the standard. The EPA X- over the 12-month compliance period, a i single discharge occurrence would be ah apparent violation of the standard (lie., J plants currently report such multiple discharges from a single piece of ; equipment as a single discharge. Thus, expects that plants using the best technology and procedures should be .ilk. able to perform better than the proposed V the discharge frequency per UOO batches. die performance levels serving as the . limits. - would exceed 0035). Because ...r..; _. basis for the numerical limits represent Reporting Requirements. The current ^, x- `.'Nr insufficient batches were made, the j.' Individual discharges end not multiple . Standard for relief valve dischargei - ' ' reported discharge frequency per 100 " discharge events except when they requires producers to report discharges , :\$i . 'batches would not correctly reflect the occur from a single piece of equipment. - '.within 10 days of the incident. The EPA performance by that plant in comparison For determining compliance with the | proposing to eliminate the 10 day to other plants complying with the numerical limits, discharge frequency is reporting requirements and to require .. standard. In rectifying the undue to be recorded on the basis of individual reporting of all dischargee on a quarterly .compliance burden posed on plants with discharges except when simultaneous ' basis. Although compliance is to be " . -.'small numbers of batches by the discharges occur from relief valves on .i determined on a semiannual basis, . discharge/100 batch format and the the same piece of equipment .. .,. ! quarterly reporting of discharges is' ^ ' .selected 12-month compliance.period, .. A relief valve discharge in considered appropriate because violations of the ' :'..the EPA Is proposing to add additional . to be any venting through a pressure - standard may occur well before the end ' ^''provisions affecting the number of batches used to calculate the discharge :Z frequency. For PVC plants producing ' relief device to prevent or relieve an . - of the 6-month period. Quarterly overpressure condition from equipment 'reporting notifies enforcement personnel in VC service that results in emissions ' of potential violations and violations ' - - ' less than 2358 batches of a particular of VC directly or indirectly to the . that have already occurred prior to the ' resin, the minimum number of 2,858 ' atmosphere. In determining whether or .' end of the compliance period ao that .' batches will be used when determining not a relief valve discharge results in corrective action* can take place sooner ' compliance with the numerical limits.'" .. emissions to the atmosphere, the - following the end of the compliance PVC plants producing more than one '' - controlling factor is the ultimate period. Information to be included in the 1. ! resin type must demonstrate compliance disposition of the gases. Venting to a ' semiannual report for individual relief ' - . '* separately for reactor discharges manifold or header system that hw valve discharge* i* to be reduced to : 'occurring from different resin production ultimately discharges to the atomsphere include only the date, time, source, - * r" ' .r, ' ' processes. Only the relief valve . constitutes a relief valve discharge. If ; * cause and estimated amount of each : ' ' . discharges and polymerization batches die manifold or header discharges gases discharge occurrence. The semiannual ' specific to each resin type are - through a control device meeting the 10 report will also iniqude information on considered for determining compliance. ' ppmv VC emission limit the venting compliance status. "' ''"'XI However, for determining compliance > does not constitute a relief valve In addition, plant* will now be with the standard for nonreactor discharge. .V required to maintain relief valve discharges, the total number of For purposes of reporting compliance discharge records for 8 years, because of polymerization batches (regardless of 1 status with the limits, plants will be the potentially significant increase in the. , resin type) are counted. . ,- required to calculate their discharge per time perfodbetween a discharge ' . To determine the number of - batcb frequencies with sufficient occurrence and reporting of the . polymerization batches produced for ' precision to demonstrate that discharge. purposes of assessing compliance, the - performance is either equal to, beiow of Effective Data ofRevision. The 'Vvff-'"'. following guidelines apply. A ,' in excess of the limits. Based on ' . current standard as written will remain AP00003611 - . .*v*x Federal Register / VoL SO, No. 6 / Wednesday, January 9,1985 f Proposed Rules ', " -- - " 1191^ ' detection and repair requirement* on . ZJO percent or lets. Any proceet unit in - that achieve* that standard may be effective existing plan*. .-i - which the percentage of leaking valve* applied. Section 112(e)(2) define* the Leak Detection andRepair .'jr U found to exceed ZO percent will be ' following conditiont under which it W * *~jsk Requirements. The EPA established leak required to comply with the provisions not feamole to prescribe or enforce an detection and repair requirement* (40 of Subpart V. 't -< irzfi-fz p-v.. \ ; - emitdon standard: (1) If the pollutants CFR Part 61 Subpart V) for certain +>;. .the Subpart Vrequirements for ' /' cannot be emitted through a conveyance equipment in volatile hazardous air ''*' valves are bated on a leak detection designed and Constructed to emit or vi.V:SfJjs pollutant (VHAP) service on June 6, -V.!:-'. and repair program that requires (l) : capture the pollutant; or (2) if the 1984. These requirement* were1 monthly monitoring for valves in gat/ v ' application ofmeasurement established in conjunction with the final - vapor and light liquid service, (2) an ' methodology is not practicable due to' ^ . atandard for benzene equipment leaks.`v1*:initial attempt at repairing these valves ' technological or economic Hmitationsr*V;-'<? The requirements of Subpart V generally within 5 days after detection of a leak. Section 112(e)(1) allows that if an apply to piumps, compressors, pressure ' (8) repair of lealdng valves within 15 -'' emission standard is not feasible to relief devices, sampling connection ' days after detection of the leak unless " prescribe or endorce, then the EPA may --.-S** systems, open-ended valves or lines, >... repair would require a process unit ~~\r . Istead promulgate a design, equipment valves, flanges and other connectors, - v shutdown, and (4) repair of valves . work practice, or operational standard.; -' --2. and product accumulator vessel*. These during the next process unit shutdown or combination thereof. - requirements reflect the level of control ` after repair is delayed until a process '4 The EPA has reviewed the design, that the EPA considers reasonable for . unit shutdown. Valves found not to leak equipment, work practice and , , equipment covered by developing for 2 successive months can be ..'-"-`.r-j.; operational requirements contained in . standards for VHAP. The EPA is monitored quarterly until leaks are >V>. the current VC standard. The only therefore proposing to add VC to the list detected. Monitoring of equipment to sources covered by foe current standard '.r| of substances covered by Subpart V. detect Isakt is conducted in accordance . with ana of the requirements for which a $T-. Subpart V would substantively affect with Method 21 and a leak is defined as performance standard (Le* an emission only valves and flanges in VC service a measured organic concentration equal standard] U feasible are pressure relief -;:x within this industry. All other equipment to or greater than 10,000 parts per devices. As discussed below, the EPA is - in VC service are already required by million by volume (ppvm). For a - setting a "no detectable emissions'* limit >> the VC standard to comply with - complete description of the leak -'.s for these sources. For the other sources. equipment and work practice standards detection and repair requirement!, see ' the EPA is reinstating those .... +..<. consistent with those in Subpart V. For Subpart V (49 FR 23498, June 6,19S4). ' t requirements as set forth in the currant vT?,. > example, pumps and compressors . In addition, Subpart V contains - standard. --y. lii-., meeting the dual mechanical seal standards for other types of equipment '' The EPA selected the use of rupture ift requirement* pf the current VC standard (e-8, flanges, and open ended valves or ' disks as ths basis for the current. ~ v,will be in compliance with the Subpart lines).Standards for flanges include , standard for pressure relief devices. .u .:jy$Z V requirements. In addition, the ' monitoring with a portable instrument.' When the integrity ofruptures disks is " -i , sampling connection system* ` -`v* -{ nnder prescribed procedures within 8 maintained, equipment leak* through the '. requirement! of Subpart V are . : days of observing evidence of a ' relief device are eliminated. Rupture ? essentially the same as the current - potential leak by visual, audible or other disks normally maintain their integrity \ standard. Tha use of rupture discs for means. Open-ended valvea or lines are - unless an overpressure occurs. After the controlling leaks from pressure relief - required to be capped, blinded or fitted occurrence ofan overpressure, i...^ ' `x devices, as required by the VC standard, with a second valve. These provisions ' - replacement of the rupture disk once itji is consistent with the "no detectable - are not expected to significantly affect again eliminates equipment leaks of VC -x y. emissions" requirement included in ' producers with these types of equipment through the pressure relief device. .. t.. jrt*. Subpart V. Requirements for controlling , in VC service. The equipment and - For emission control techniques that . ? ' leaks from pressute'relief devices are procedures employed as normal practice eliminate equipment leaks, such as the -' described In more detail later in this by these producers or as a result of the ' use of rupture disks, a "no detectable jh section. Thus, Subpart V will affect current VC standard are expected emissions" limit is feasible. An . jV- primarily valves and flanges in VC j generally to ensure compliance with js. service by requiring a specific ;<-'. Subpart V, '-- instrument reading of less than 500 parts ;v' per million by volume (ppmv] above a .] monitoring schedule, leak definition end , Pressure ReliefDevices. The EPA background concentraUon based on repair provisions. proposed and promulgated the work Reference Method 21 can be used to : ' > Compliance with the provisions of - practices, equipment, design and - indicate whether equipment leaks have -x. Subpart V will be used to determine V operational standards In the current ' ^ been eliminated; that is, that the compliance with the portable monitor standard before explicit legal authorityequipment has "no detectable '.tuf*6v, v;. -leak detection and elimination existed in Section 112. These ".'emissions." requirements in foe current VC standard requirements are found in i 61.85(b). In - ftbe "no detectable emission" limit '* . 'i ' (40 CFR Bl.SS(b)(8)(i<)). and therefore, s August of 1977, Congress amended ^ 'ewould not apply to discharge* through " v- the current standard is being revised to Section 112 to allow the use of these Hhe pressure reliefdevice during ....... reflect this change. However, process requirements. Section 112 of foe Clean overpressure relief. (These releases are' `units within VC and PVC plants in -Air Act requires that an emission > covered under ii 61.64(a) and 81.65(a).) which the percentage of leaking valves . standard (l.e a performance standard)' *The standard would specify, however, is equal to or less than 2ff percent are ' be established for control of a ' ('(hat foe reliefdevice be returned to* ~ considered by the EPA to be effectively hazardous air pollutant unless, in the 'state of "no detectable emissions".': ,-i'- - 7 controlling VC emissions from leaking judgment of foe EPA it is not feasible to %lthin 5 days after such a discharge. ''' ' ~ ' V valves. For these process units, foe prescribe or enforce such a standard. An Tbs standard would further require an .existing leak detection and elimination emission standard allows for some annual test to verify the "no detectable program will continue to be allowed flexibility in complying with foe kmlssions" status of foe pressure relief while foe percentage of leaking valves is standard, since any control technique ' "`device* and a test after each over t AP00003612 federal Register / Vol SO, No.8 / Wednesday, January 8, 2985 / Proposed Rule* M93 Other. 1b additionlo the revisions . increase or decrease in the overall ... ^.-revisions do not increase the cost of described above, areview of the . -y, - recordkeeping and reporting currently compliance with the standard:-'-'* Moordkaeplng and reporting requirements of the current standard practiced by individual plants. "(tW'.yJ 'The EPA concluded that the current Public Healing . was performed to identify way* to ease recordkeeping requirements, as -vnu - If requested, a public hearing will bit recordkeeping and reporting burden on -specified to 40 CFR 6171, are still ;s(s _ held to discuss tha proposed revislorrito t'* plants and to identify any additional appropriate. However, the EPA is `--.-i. ~ tiis VC atandard In accordance with recordkeeping and/or reporting needs. . proposing to extend the current ,' sections 112(b)(l){B) and 307(d)(5) ofthe The EPA identified two areas where the recordkeepingrequirements for all , Clean Air Act Persons wishing tomake reporting burden on plants could be -reporting activities from 2 to 3 yeare. - .reduced. The current reporting The net Impact of the revised requirements foe residual VC monomer ^recordkeeping and reporting - - '` Oral presentations on fire proposed -- revisions should contact tha EPA atthe- addresa given In tiie admecsm section "t-t specification! and reactor opening "requirements proposed by the EPAis .; j .-ofthls preamble. Oral presentstioni'Wtll .Vi measurements require that results of all estimated to be a decrease in a - - be limited to 15 minutes each. Any rt; - compliance teits be reported in paperwork burden of about 27 person-' ' member of the publicmay file a written semiannual reports. The EPA is .v;: . yeare. ................. - - --, . statement before, during, or within SO: " proposing to allow plants to report only . It should be noted that all'L ,r V -days after the hearing. Written" test results that show exoeedences of - - Comprehensive. Environmental 7 statements should be addressed to me. V yf-' the respectiv( standards. If no ? . Response, Compensation, and Liability' Central Docket Section address gtveiTin v-.' . exceedences occur, plants will be ! . required to indicate that fact In tiie - semiannual report. This type of exception reporting la currently allowed for demonstration of compliance with Che 10 ppmv standard for process vents. The second area is the requirement to. report relief valve discharges within 10 Act (CERCLA) Section 101(14) .' '- 'the addresses section of this preamble. . hazardous substances such as vinyl -j A verbatim transcript ofthe hearing' 7rr 'chloride are subject to reporting . and written statements will be available requirements under Section 103(a) of CERCLA. CERCLA requires that persons' for pubfic inspection and copying during normal working hours at the EPA'e. - in charge of vessels or facilities from which hazardous substances have been released in quantities (RQe) ., Central Docket Section in Washington * r-y D.C. (see addresses section of this days of their occurrence. Th EPA is proposing to allow plants to report relief valve discharge occurrences on a immediately notify the National Response Center (NRC) ofthe release* The toll-free 24-hour telephone number . Docket. itll V Tie dockeVis an organized and ,'*& &T , quarterly basis rather than within 10 of the NRC is 600-424-8802 and in days of their occurrence. Furthermore, . . Washington. D.C. metropolitan area It is ,. the reporting requirements for relief .. - (202) 426-2875. (See CERCLA Section valve discharges have been streamlined 103 and 48 FR 23592. May 25.1683.) - by dropping the need to report actions - taken and implemented preventive r? measures for each discharge. Information on the date. time, source. cause end estimated amount of individual relief valve discharge will be . included with the semiannual reports along with information on compliance ..^status. v..-1-wtr.i c).. .; wv-1 c . Additional semiannual reporting 1>- - requirements heing added for PVC - - - Vinyl chloride was assigned a ' statutory 1 pound reportable quantity -' under Section 101(14) until adjusted by regulation, and ia presently undergoing assessment for both chronic toxicity and carcinogenicity. Its RQ will be adjusted pending the outcome of these reviews by ihs Office of Emergency and Remedial Response. Federally permitted releases tinder CERCLA (See CERCLA Section r 101(1) and 46 FR 23552) are not subject to-CERLA notification requirements or complete ftte of all the Information 7!.-, 1 V-c. submitted to or otherwise oonsldered.by the EPA In the development of this ' proposed rulemaking. The principal - lir-'rie# purposes of the docket are: (1) To allow v ; interested parties to identify and locate 'documents so that they can affectively participate in the rulemaking process and (2) to serve as the record in esse of judicial review (except for interagency . : .- review materials [| 307(d)(7(A)J). -~^ .v - /.. "bWecillaneoE* ietS '^rr . In accordance with section 117 of the V.-. Act, publication of this proposal was . -J producers are the number of reactor liabilities. However, releases of '- * preceded by consultation with v-r ,V openings and the design.capacity * . number of polymerization batches for' hazardous substances that are not ' TT appropriate advisory committees, i-tf rubjeot to a permit or control regulation independent experts, and Federal. -,-.j .' V. each resin type. This requirement will must be reported^ r ' department* and agencies. The .' .vr' ' provide general information to facilitate Administrator will welcome comments' " review of industry-wide compliance Regulatory Flexibility Analysts -. on all aspects of the proposed ,.-7 - . .status during past-reporting periods. ' -The Regulatory Flexibility Act ofI960 "regulation, including health, and . Specific recordkeeping and reporting requires that adverse effects of all ,ry;> ' ' economie'and technological issues.'. 1. - ^ requirements are included at part of the Federal regulations upon small. : i-i-ji ' ;",Ths information collection . -. .-revisions to the leak detection and ^ . businesses be identified. According to requirements in this proposed rule have ~ i repair requirements. The recordkeeping the current guidelines of the Small > been submitted forapproval to the ~7 . requirement* include preparation of an . Business Administration (SBA), a email Office ofManagement and Budget '.?sjT'T ' Initial log to.record equipment > business that produces or processes VC (OMB) under the Paperwork Reduction *J'. component identification, physical i 'is one that hat .500 employees or less.: Act of 1880,44 U.S.C. 3501 et teg. ' : tagging of equipment components which Currently, none of the existing ' <f>- Comments on these requirements shoidd . ' leak, and maintaining a record of -r..; producers or processors that are - ? ' be submitted to the Office of 7 equipment leaks and repair action. > affected by the standard are estimated ' Information and Regulatory Affalrs'bf ' Included in the reporting requirements to be email by this definition. Since none OMB, marked "Attention: Desk Officer '- are the number of equipment leaks and of the companies meets the SBA' for EPA",as well as to the EPA docket V V the repair status of leaking components. definition of email business, no -described above. The final rule will : 1- . Depending on the particular leak -- .' - -regulatory flexibility analysis is - respond to any OMB or public detection and repair program in place, required. Even if an analysis were <r - comments on tha information collection'' . . these requirements may represent an _ required, the proposed administrative requirements. -j- .'h.-Wiifc:r*an-n r'.. <#tU AP00003613 i. Federal Register / Vol. BO, No. 6 / Wednesday. January B, 1985 / Proposed Rulei 1195 ty** Emission alsndsrd tar polyvinyl (2) The reactor opening lost from each oWortdopient*. r-:>. reactor U not to exceed 0.02 (vinyl An owner or operator of a polyvinyl chloride/kg (0.00002 lb vinyl chloride/lb) chloride plant shall comply with the . ofpoly vinyl chloride product except ai requirement* of this section and 101.65. provided in paragraph* (f)(1) and (f)(2) (a) Reactor. Hie following :--- of this section, with the product. " requirement* applyto reactors: 5 ' determined on a dry solid* basis. This (1) Hie concentration of vinyl chloride requirement does not apply to in each exhaust gas stream from each prepolymerization reactors in the bulk reactor to not to exceed 10 ppm (average P100** Tbi requirement doe# apply to -y.' By revising paragraph (a) to 161.85.'T' .as follows: .... .. . - ^ ..... 4-ljy 16145 Emission standard for ethylene ..ii ", dtcMortde, vtnyt chtoride and polyvtnyt D(!r chlorideplanta ....... ' An own^- or operator of an ethylene. ^ . ' dichloride, vinyl chloride, and/or Kj'*/' .' .. ( polyvinyl chloride plant ahaU-comply v with the requirements of tide section.-I for 3-hour period or aa determined in portpolymeriiation reactor* in the bulk - (a) Reliefvolve ditchargee. (1) /V accordance with 161.67(g)[l)), except as P"***1- where the product means the provided in parsgraph (a)(2) of this ' gross product of prepolymerization and Polyvinyl chloride plants (suspension, ^ dispersiom latex, and bulk processes).# ; r sectionand f 61.65(a). s'- i- postpolymerization. vtv i (1) Reactor. Hi* numbsr of discharge* ' ' v>'. * VV- to the atmosphere from relief valve* On' .... ' (b) Stripper. Hie concentration of .' vinyl chloride in each exhaust gas stream from each stripper is not to v ` 5. By revising paragraph [e] introductory text and adding-paragraph (a)(3) to 181-64 as follows; --- , polyvinyl cldoride reactors in vinyl .. -j;f. : chloride service is not to exceed the "'. following limits except as provided-in exceed 10 ppm (svsrage for S-hour period or a* determined in accordance with i 61.87(g)(1)), except aa provided in f 41.44 Emission ataiidatti tor potyvtnyi' paragraph (s)(l)(iil) of this section. For all reactors producing suspension resins eMorhleplants. . r.r. -*.1- . l.:: * J/ *. *..**'***: within a PVC plant, the number of rellef i 61.65(a).'This requirement does not apply to equipment that has been opened, is out of operation and met the valve discharges Is not to exceed (U035 j (e) Sourcesfollowing the-etripperfe). discharges per 100 polymerization _i>\> The following requirements apply to batches nor 4 discharges perysar. For : -j ", requirement in 161.65(b}(6)[l] before -emissions of vinyl chloride to tiie all reactors producing dispersion and .j}. being opened. ' (c) Mixing, weighing, andholding ,containers. The concentration of vinyl . chloride in each exhaust gas stream Jrom each mixing, weighing, or holding container in vinyl chloride service which . precede* the itripper (or the reactor if the plant hat no stripper) in the plant J;' " process flow is not to exceed 10 ppm (tvertgt for 3-hour period or at determined In accordance with ' ' * ' ' atmosphere from the combination of all sources following tbs stripper(s) [or the Teactorfs) is the plant has no stripperfs)] 'in the plant process flow including but not limited to, centrifuges, concentrators, blend tanks, filters, .-- dryers, conveyor air discharges, baggers, storage containers, and lnprocess ; wastewater, except a* provided in r?-y i -' paragraph (f) of this seotiont . tv.---.v - ; latex resins within a PVC plant, tha. number erfrelief valve discharges is not. to exceed 0.035 discharges per 100 poKmerization batches. For all reactor* including prepolymerizatibA and .. postpolymerization reaclura, produdng bulk resins within a PVC plant the numbsr of relief valve discharge* is not to exceed 0.035 discharges per 100 ' t polymerization batches.. ;v, ; t- .* - (it) The number of discharges to the ' ~ i. ' . ...j ';'r . 161.67(g)(1)), except a* provided in ' '' (3)The provisions of this paragraph atmosphere from reliefvalves on - 161.65(a). This requirement does not apply at all time* including When off- - equipment (excluding polyvinyl chloride .apply to equipment that has been specification or other types of resins are reactors) in vinyl chloride service is not - .-opened, is out of operation, and met the made. v to exceed 0.025 discharges per 100 . requirement in 161.65[b)(6)(i) before .. being opened. y - - -.=. -.t? (d) Monomerrecoveryeyetem. The 6. By adding paragraph (f) to f 61.64 . at follows: ," polymerization batches nor 3 discharge's . per year except as provided in _ paragarph (a)(l)(iiij of foil section. ''i'' concentration of vinyl chloride In each 181.64 Emission standard tor potyvtnyt ' (iil) The limits specified in paragraphs :i. exahust gas stream from each monomer chloride plant* (a)(l)(i) and (a)(l)(U) of this section may recovery system is not to exceed 10 ppm be exceeded when only one relief valve. * (average for 3-hour period or aa ,^determined in acconlance with 4\ 161.67(g)(1)), except at provided in " I (f) Reactor, usedas stripper. When a . .'discharge to foe atmosphere occurs nonbulk resin reactor is used as a during foe 12-monfo period preceding' ' ^ stripper tills paragraph may be applied foe close of foe 6-month reporting ' * 161.65(a). This requirement doe* not in lieu of i 61.64 (a)(2) and (e)(1): ' period. ,i ; . .-...apply to equipment that has been 'Z "opened, is out of operation, and met the " "requirement in 181A5[b)(8)(t) before... jf 'beingopened.. > ; n:, * a ' > 4 ------ ,-i; 4. By revising existing paragraphs ~ . r f 61.64(a)(2) and by removing (s)(3) as : .' (1) The weighted average emissions of vinyl chloride from reactor opening loss and all aources following the reactor used as a stripper from all grades of " polyvinyl ohloride resin stripped in the reactor on each calendar day may not : exceed: ,:j (2) Polyvinyl cdilorfde plants (solution and other continuous PVC production processes)! The number of discharge* to the atmosphere from relief valves on all equipment in vinyl chloride aervice is . ' not to exceed l ditchaig* per year. (3) Ethylene dichloride and vinyl .j.u v ' j. ' follows: . - >. (i) 202 g/kg (0.00202 lb/lb) ofpolyvinyl chlorideplants. The number of - | I1JM Emission standard tor polyvinyl . aMorlda plants. An owner or operator of a polyvinyl r chloride plant shall comply with the requirements of this ssction and f 61.65. . b... (a) Reactor. The following # . ?. requirements apply to reactors: ;. 4- ' * * chloride product for dispersion polyvinyl chloride resins, excluding latex resins, with the product determined on a diy solids basis. fii) 0.42 g/kg (0.00042 lb/lb) of polyvinyl cblorids product for all other . polyvinyl chloride resins, including latex resins, with the product determined on a dry solids basis. discharges to foe atmosphere from relief valves on equipment in vinyl chloride aervice i* not to exceed 4 discharge* per year. '-,,-r (4) Each relief valve discharge that contributes to n relief valve discharge ' frequency in excess of any limit 12 prescribed In ptragarphs (a)(1), (a)(2) and (a)(3) of this paragraph constitutes AP00003614 Federal Reciter / Vol. BO. No. 8 / Wednesday, January 9, 1885 / Proposed Rulei 1197 14 it* fc -! i Et.i mk 0. By revising paragraph (b)(4) to determined in accordance with the 14. By tevialng paragraphs (g)(3) ' 161.65 as follows: . . - following: ' Introductory text (gX3)(i), and (g)[3)(iii) ' ' . ' ~ 1*125 Emission standard lor ethylene dlehlortde, vinyl chlortds and polyvinyl eWorld* pianta . . (A) A performance test as specified in paragraph (b)(8)(ii)(Q of thia section shall be conducted initially within SO days of the effective date of these of { 6127 as follows: 1*127 Emission Wets. v* * M' -- * (b)Fugitive' em. lssion so-u. rces.' (4) Leaksfrom reliefvalves. Vinyl chloride emissions due to leaks from * each reliefvalve on equipment in vinyl chloride service shall comply with 161.242-4 of Subpart V of this part regulations, annuity and at times requested by the Administrator. ' (B) For each performance teit a minimum of 200 or 90 percent of the total valves in VOC service (at defined in i 60.481 of Subpart W of Part 60) within the process unit shall be randomly elected and monitored within 1 week (g)* * * _.... : (3} When a stripping operation is used to attain the emission limits in 191-84 (e) and (f), emissions are to be - ^ ;*' determined using Test Method 107 a* .... J. : follows: .. i. (t) The number of strippers (or .I..1' i'l*-* reaotors using as strippers) and samples ' ~- V 10. By revising paragraph (b)(7) of 161.65 as follows: . , . by the methods specified in i 61245(d) of Subpart V of this part If an instrument reading of 10,000 ppm or end the types and grade* of resin to be sampled are to be determined by the .' . Administrator for each individual plant 1*1.4* Emission standard for ethylene dtehlortde, vinyl chloride and polyvinyl chloride pianta. mater is measured, a leak ia detected. ' The leak percentage shall be determined by dividing the number of valve* in . at the time of the lest baaed on the plant's operation. : v (U)* v-4*. ' VOC service for which leaks are (b) F'ugitive emission sources. (7) Samples. Unused portions of . samples containing at least 10 percent by weight vinyl chloride are to be returned to the process or destroyed in a control device from which the concentration of vinyl chloride in the exhaust gas does not exceed 10 ppm. Sampling techniques are to be such that sample containers in vinyl chloride are purged into a closed process system. - detected by the number of tested valves . In VOC service. (C) If a leak is detected, it shall be material processed by each stripper (or " reactor used as a stripper) is to be determined on a tby solids basis and by - . repaired in accordance with 181242-7 (d) and (e) of Subpart V of this part (D) The results of the performance test a method submitted to and approved by.. i the Administrator.. v ft ft ft. ft shall be submitted in writing to the Administrator in the first semiannual . . 15. By revising paragraph (g)(5) introductory text and adding paragraph report following the performance test as (g)(6) to 161.67 at follows'. ' part of the reporting requirements of ,, \ *81.70. . . .,.............................. . 1*127 Emlaalan tests. ,, . .. (E) Any process unit in which the ' .' 11. By revising paragraphs (b)(8) introductory text, (b)(8)(U), and (b)(B)(v) to f 61.65 as follows: 101.8S Emission standard for ethylene dlchlorlde, vinyl ehlorldt and polyvinyl chloride plants. percentage of leaking valves ia found to be greater then 2.0 percent must comply with all provisions of Subpart V of this part within 90 days. ft ft ft V * (v) It contains a plan of action to be taken when a leak is detected consistent (g) * (5) Tbs reactor opkning lost for which an emission limit is prescribed in * 6124(a)(2) Is to be determined. The number of reactors for which the determination la to be specified by the Administrator for each individual plant (b) Fugitive emission sources. .ft ft ft ft ft * *. . (8) leak detection and elimination. Vinyl chloride emissions due to leaks from equipmsnt in vinyl chloride service '; are to be minimised by instituting and. * implementing a lead detection and ' repair program consistent with the > 'provisions of Subpart V of this part The 1 'program ia to be implemented within 90 . >' days of the effective date of these .: regulations, unless a waiver of . compliance is granted under *61.11. .. Approval of program will be granted by the Administrator provided he finds: -. (ii) It tncludss a reliable and aoourate portable hydrocarbon detector to' be used consistent with the provisions of . Subpart V of this part An owner or operator is exempt from ( 61242-l(d), " II 61.242-7 (a), (b) and (c). f 61246 and with Subpart V of this part 12. By revising * 61.66 as follows: at the time of the determination based on the plant's operation. 4 ft ft ft ft 16128 Equivalent equipment and procedures. (6) For a reactor that is used as a stripper, the emissions of vinyl chloride Upon written application from an from reactor opening loss'ana all owner or operator, die Administrator , sources follovdng the reactor used as a. may approve uee of equipment or , - stripper for which on emission limit 1* .. procedures which have been prescribed in * 61.64(f) are to be demonstrated to his satisfaction to be determined. The number of reactors for equivalent in terms of reducing vinyl .. . which die determination is to be made is chloride emissions to the atmosphere to to be specified by the Administrator for.. those prescribed for compliance with e each individual plant at the time of the specific paragraph of this eubpart determination based on the plant's *-? - 13. By revising paragraph (f) of i 61.67 as follows: operation. '. , (l) For each batch stripped in the ' > > ' reactor, the following measurements are . 1ft6127ft Emfitssionft tests. . ft to be made: . - v. ...i. - (A) The concentration (ppm) of vinyl (f) The owner or operator shall retain' chlorids in resin after stripping, - 4 at the plant end make available, upon 1. measured according to paragraph (g)(3) request for inspection by the of thia section; ^ ' - . 161247 of Subpart V of this part for any Administrator, for a minimum of 3 years, - (B) The reactor vacuum (mm Hg) at . '. process onlt in which the percentage of records of emission test results and end of strip from plant instrument; and < leaking valvat it demonstrated to be other data needed to determine. (C) The reactor temperature (*C) at" .. " equal to or leas than U0 percent as emissions. end of atrip from plant instrument 'v' ?- .IA. AP00003615 Federal Register / Vol. BO, No. 6 / Wednesday. January 8, 1985 f Proposed Rules' 1*1.70 Reporting. vinyl chloride content in die polyvinyl vinyl chloride concentration, as - -no (a)(1) The owner or operator of any source to which this aubpart applies shall submit to the Administrator on September IS and March 15 of each year a report in writing containing the 1 * or vchloride resin. . () * . v. ` fill) The vinyl chloride content in each ample is to be determined by Test determined in this paragraph, in excess ; of the limits prescribed in 161.61(e). The ' vinyl chloride content found in each sample required by paragraphs (c)(2)(l) and (c)(2)(U) of this section shalTbe information required in paragraphs, (9). (d) and (e) of this section and on December 15 and June 15 of each year a Method 107 as prescribed in f 61.67(g)(3). (iv) [Reserved] - 1 .. . averaged separately for each type of '*; resin, over each calendar day and weighted according to the quantity of * report in writing containing the ' information required in paragraph (e) of this section, except as provided in . . paragraph (a)(2). ................... tv) The report to the Administrator by each grade of resin processed by the die owner or operator is to include-a atripperfs) that calendar day, according V record ofany 24-hour average resin ' to the following equation: ' (2) In the case of an existing source . that submit* semiannual reports on an - approved fixed schedule other than . September 15 and March IS. the . approved semiannual reporting schedule J,etV,+ . rt-Pa.lfu. o, TGT 1 . ~ " - shall be used to report the information required in paragraphs (e), (d) and (e) of '.C-C r this section. In addition, the information where Include a statement that excess . :! ' required in paragraph (e) of this section rt--z4-hour average concentration of type, Ti emissions have not been detected. will be reported exactly 3 months resin in ppm (dry weight basis). -V following the semiannual reporting dates. ~ (3) The first report it to be submitted following the first full 3 month reporting period after the initial report it submitted. Q--Total production of type 7i reein over the 24-hourperiod, in kg. Tt "Type ofresin: rU.. .at wherein it tote] number of resin types produced during the 24-hour period. Af*Concentration ofvinyl chloride In one sample of grade G resin.in ppm. . 22. By adding paragraph (c)(4) to ' '( 161.70 as follows: ' i' 161.70 Reporting ' . . ' 5/ <' (0* * -. r~-- IB. By revisingparagraph (c)(1) of P--Production of grade G resin represented ' (4) In polyvinyl chloride plants for ( 61.70 as follows: by the sample. In kg. f 61.70 Reporting. Gt-Grada ofresin: t&, Gt, G and G.' n--Total number ofgrades of resin produced . during the 24-hour period. $(l) Tha owner or operator shall Include In the report a record of the - vinyl chloride content of emissions for If no 24-hour average resin vinyl chloride concentrations in excesm of the ' limits prescribed in f 61.64(e) are each 3-hour period during which average measured, the report shall state that no emissions are in excess of the emission excess resin vinyl chloride limits in i 61.62 (a) or (b), f 61.63(a), or concentrations were measured. r 161.64 (a)(1), (b), (c), or (d), or dining (vi) The owner or operator shall retain which average emissions are in excess . at the source and make available for 'S' of the emission limits specified for any inspection by the Administrator for a V; control system to which reactor minimum of 3 years records of all data -``emissions are required to be ducted in needed to furnish the Information i 61.64(a)(2) or to which fugitive required by paragraph (c)(2)(v) of this emissions are required to be ducted in section. The records are to contain tha J 61.65 (b)(i)(ii). (b)(2), (b)(5), (b](6)(ii), or following information; it. (b)(9)(ii). If emissions in excess of die (A)* * * '". emission limit* era not detected, the * } report shell contain a statement that no '; excess emissions have been detected, r * The emissions are to be determined in ,1/-.accordance with $ 61.68(e). . (B)*.*'1 '..v;;.-. .-.sia--. * * * 21. By revising paragraph (c)(3) of 161.70 as follows: : ' 2a By revising paragraph (c)(2) | *1.70' Reporting. , '' introductory text, removing paragraphs * e * g. ' |c)(2)(tv). revising paragraph (c)(2)(ili) t and revising (c)(2)(v) and (c)(2)(vi) . (C)* * - which stripping in the reactor it used to attain tha emission level prescribed in ' f 6164(f), the owner or operator shall include in the report a record of the .^iY, vinyl chloride emissions from reactor '&.& opening loaa and all source* following- - the reactor used as a stripper. (t) One representative sample of s. polyvinyl chloride resin is to be taken from each batch of each grade of resin - 1 .immediately following the completion of the stripping operation, and identified Jhj. - by-resin type end grade and the date . and time the batch, is completed. The . , corresponding quantity ofmaterial processed in each stripper batch is to be 1 recorded and identified by resin type and grade and the date and time the - batch it completed. -.. - (ii) The vinyl chloride content in each sample i* to be determined by Test ' . Method 107 as prescribed in 161.67(g)(3). - ` (tit) The combined amission from - reactor opening loss and all sources -M ~ - following the reactor used as a stripper ' are to be determined for each batch tripped in a reactor according to tite : < introductory text to S 61.70 as follows: f 61.70 Reporting. >* . '* V (3) The owner or operator shall include in the report a record of any emissions from each reactor opening In excess of the emission limit* prescribed procedure prescribed in 161.67(g)(6). (iv) The report to the Administmtorby the owner or operator is to include a record of any 24-hour average combined '(c) ` * in i 61.64(a)(2). Emissions are to be reactor opening loss and amissions from ' (2) In polyvinyl chloride plants for - - determined in accordance r which a stripping operation is used to with161.67(g)(5), except that emissions ~. attain the emission level prescribed in for each reactor are to be determined. If ~~ f 61.64(e). the owner or operator shall emission* in excess of the emission - -. Include in the report a record of the limits are not detected, the report shall all source* following the reactor used as a strippers* determined in this paragraph, in excess of the limit* - .\i prescribed in 161.64(f). The combined ' - reactor opening loss and emissions from AP00003616 1 > Federal Register / VoL 50. No. 6 / Wednesday, .January 9, 1985 / Proposed Rules initially and remain unchanged unless significant change* to the design ., capacity occur. 24. By revising paragraph (a) introductory text of f 01.71 as follows. f 01.71 Recordkeeping. (a) The owner or operator of any . source to which this subpart applies.. . shall retain the following information at the source and make it available for inspection by the Administrator for a minimum of 3 years: ' `if* 25. By adding the words "vinyl chloride" to the definition of the term' "volatile hazardous air pollutants" in 161.241 of Subpart V as follows: 161.241 Definition*. *# "Volatile hazardous air pollutant" or "VHAP" mean* a substance regulated under this part for which a standard for equipment leaks of the substance has been proposed and promulgated. Benzene is a VHAP. Vinyl chloride is a VHAP. (Sec. 112 Clean Air Act of 1976] {PR Doc. 85-509 Filed l-S-85: a-45 am] 1201 f. ... .? * f \-f\- * .. , * * *' . . >rs: \ % * 4 AP00003617 7-'SI "87 907 Current Developments Litigation D.C. CIRCUIT REJECTS VINYL CHLORIDE RULE, BUT ACCEPTS USE OF COST IN AIR ACT STANDARDS Id setting standards for hazardous emission* from iadustrial sources, the Environmental Protection Agency moat base its review primarily on health concerns, but the agency can alio consider cost and technological feasibility In setting emission limits, an 11-judge panel of the UJS. Court of Appeals for the District of Colombia Circuit ruled July 28. However, the agency's use of cost and feasibility con straints in withdrawing a 1977 proposed standard for vinyl chloride emissions was improper under the Clean Air Act, according to the D.C. Circuit The agency improperly substi tuted technological limitations for health concerns as the primary consideration in withdrawing the proposal, the appeals courtaaid {NRDC v. EPA, CADC, No. 85-1180). The en banc panel vacated a November 1986 holding by a three-judge panel of the court that allowed EPA to with draw vinyl chloride standards to consider whether they imposed "unreasonable" regulatory costs (CADC, 28 ERC 1105; Current Developments, Nov. 14,1986, p. 1169). EPA originally issued the standard in 1876, requiring emission reductions of 95 percent, bat the agency opted to propose stricter standards in 1977 in response to litigation over the rules. In 1985, EPA withdrew the proposal, citing concern that no control technology was demonstrated that could reduce emissions below the 1978 standards. The Natu ral Resources Defense Council challenged the action. Judge Robert H. Boric determined In 1986 that the Air Act requires EPA to provide "an ample margin of safety to protect public health," but the agency can use its "discretion and judgment to bear on scientific uncertainty" in preparing emission limits. EPA Must Ascertain `Safa' Laval But Bork wrote on behalf of the panel in the July 23 ruling that EPA failed to make a preliminary determination as to what constituted a "safe" level of vinyl chloride emissions. Congress' requirement that the standard provide for an "ample margin of safety" iqeant the agency had to make an initial finding of what'll "safe," according to Bark. Cost and technological feasibility factors do not enter into the safety determination, Bork said. Once a safe level is assured, EPA may use oost and technological considerations in determining what is an "ample margin" to provide In the tandard, Bork said. On a review of Congress' Intent in Section 112 of the Air Act, Bork said the legislative history of the hazardous air pollutant provision is "ambiguous" as to whether cost and technological feasibility be considered. The D.C. Circuit and U.S. Supreme Court have ruled on the issue of cost and technological feasibility eoeeidenitions, Bork said, but the cases dealt with provisions other than s*c" *** toother sections, Congress indicated, through statutory language or legislative history, that cost and feasi- review wu precluded, although no such indication appears with Section 112, Bork said. On the other hand, the 1977 amendments to the Air Act on hazardous air pollutants did not represent Congress' "ratifi cation" of EPA'a approach to the vinyl chloride standards in 1976, which included consideration of casts and technologi cal feasibility, Bork said. Be continued that "we cannot be certain that Congress was aware of the content of the vinyl chloride regulation*." Because there is no "dear congressional intent to pre clude" cost and feasibility limitations, EPA can consider anch factors. Both concluded. NRDC Raisos Soientiflc Uncertainty NRDC argued that the agency should prohibit emissions where EPA cannot develop a safe threshold level for expo sure to emissions, but the D.C. Circuit rejected the argu ment Coagross indiestod in using the language "ample margin of safety" that EPA had "peat latitude in**"g its responsibility" nndsr tbs Air Act, according to Bork. Emphasizing the likelihood of scientific uncertainty about the dangen posed by hazardous air pollutants, Bork indicat ed that Congress provided EPA with discretion in dealing with the uncertainty involved with each emission standard. If EPA were required to impose a rerolevel of omissions in cases where safe threshold limits were questionable, it would force the court to conclude that "Congress mandated massive economic and social dislocations by shutting down entire industries," which would bo an unreasonable conclu sion, Bork said. The agency is not sure bow tbe ruling will affect further rulemakings under tbe Air Act or other statutes, according to Earl Salo, an EPA assistant general counsel who worked oo the case. EPA may and up looking at whether other standards incorporated consideration of "safe" levels before technological and cost concerns were applied, Salo told BNA July 29. Ak Pollution BILL WOULD REPEAL EPA SANCTION POWERS, BUT SPECIFY CONTROLS IN NON-ATTAINMENT AREAS A bill introduced July 28 by Rep. Henry A. Waxman (DCalif) would replace construction bans for cities that cannot attain the federal air quality standards for ozone and carbon monoxide by Dec. 21 with mandated control measures. In an approach similar to that taken In S 1351 in the Senate, Waxman'i bill would place citiM into three classes, based on the extent to which they exceed air pollution standard*, and require different control strategies for each class. Citiaa with moderate attainment problems would be exempted from Clean Air Act sanctions the bill would not repeal. Under tbe measure, no sanctions would apply for simple failure to meet ambient air quality standards. Instead, sanction* would be imposed on areas that fail to adopt the control measures mandated for their non-attainment classes. Failure to meet air quality standards, however, would cause a state to be moved into the next.class. 7-01*7 tfMfWWUftt Biporfr AP00003618 CURRENT DEVELOPMENTS I i-K-flt 1169 around the point o( exposure to hazardous substance and alter the size ol an exposed population "until you get what ever level you think you want to clean up to," the staff member charged. We have a clearly developed position on our side that the ACL process is illegal under RCRA and "we take it on [under superfund] In that context, too." be said. Continuing Fight Foreseen The law states that ACLs may not be used instead of applicable laws or standards "if the process assumes a point of human exposure beyond the boundary of a facility, as defined at the conclusion of the remedial investigation find feasibility, except" in three specific instances. However, the staff member acknowledged that the issue will be a "continuing fight" because the language does not unambiguously eliminate the use of ACLs where a maxi mum contaminant level for a pollutant has been established. "It's our position that it is inappropriate where an MCL exists; it's the interpretation we're pressing for and it's now apparent EPA is going to take a different interpretation," he said. Quarles: 'Cadillac Cleanups' John Quarles, an attorney with Morgan, Lewis & Bockius, which represents many potentially responsible parties at superfund sites, told BNA Nov. 10 that strict adherence to Drinking Water Act standards will be prohibitively expen sive or unachievable in many instances, and result in what he called "Cadillac cleanups." "My interpretation of the statutory language is that it would permit ACLs at superfund sites if the required crite ria demonstrating the absence of risk could be met," Quarles, a former EPA Deputy Administrator, said. In an Oct. 17 memorandum to clients, Quarles said that in discussions with Thomas, the Administrator made it dear to him "that he regards that authorization to set ACLs, and particularly the explicit congressional debate on this subject before the ACL authority was written into the statute, as having great significance." (See related item in this issue.) Quarles said the RCRA cleanup program has been logjammed by disagreement over the applicability of ACLs and that the issue has the potential to snarl the superfund program as well. Litigation APPEALS COURT UPHOLDS EPA WITHDRAWAL OF PROPOSED VINYL CHLORIDE EMISSIONS RULE Withdrawal by the Environmental Protection Agency of proposed national hazardous emission standards strictly limiting vinyl chloride under Section 111 of the Clean Air Act was upheld Nov. 4 by a federal appeals court. The U.S. Court of Appeals for the District of Columbia, In a suit brought by the Natural Resources Defense Council Inc., ruled that Section 112 authorizes EPA to consider feasibility factors in withdrawing proposed standards that would have strictly limited vinyl chloride emissions, with a goal of zero emissions (NRDC v. EPA, No. 15-1150). EPA issued vinyl chloride standards in 1976, requiring emission reductions of 95 percent The Environmental De fense Fund challenged the adequacy of the standards, and EPA settled the litigation by proposing stricter standards in June 1977. However, EPA withdrew the proposed standards in January 1985, claiming they would not be feasible to implement. Following the agency's withdrawal decision, NRDC peti tioned the appeals court to reinstate the proposed standards, charging that the agency's decision would return vinyl chlo ride regulation to the 1976 standards, which NRDC claimed were inadequate to protect public health. In upholding EPA's withdrawal action, the appeals court concluded that agency decisions in setting emission stan dards under Section 112 need not be restricted to consider ations of public health. Court Cites 'Unreasonable' Cost* The appeals court observed that the agency, in withdraw ing the 1977 proposal, said the standards would have im posed "unreasonable" regulatory costs and that no control technology has been demonstrated for reducing emissions below that required by the 1976 standards. Judge Robert H. Berk, joined by Judge Harry T. Edwards, dealt with Section 112's language requiring EPA to set emission standards for hazardous air pollutants that will provide "an ample margin of aafety to protect public health." According to Judge Bork, the language Implied that EPA can bring "discretion and judgment to bear on scienti fic uncertainty" in deriding on vinyl chloride standards. Judge Bork said that limiting the agency's consideration to health-based factors alone could not have been the intent of the Act, because "deriding how much uncertainty to allow from a strictly health-based perspective would always lead to the same answer--none." A review of the history of the 1970 Air Act amendments proved to be "ambiguous" in terms of factors to consider in etting standards for hazardous air pollutants. Judge Bork said. Dangers Ssid Uncertain in Trace Amount! If evidence "positively demonstrated" that a given sub stance endangered public health in trace amounts, EPA might be able to prohibit any emissions of that substance, according to thm circuit judge. However, an area of uneertainty exists concerning health effects of vinyl chloride at trace levels. Judge Bork commented. EPA is allowed to use its discretion to "protect against dangers [the agency] cannot know" by setting standards as strict as possibla, "given both available technology and the requirement that the cost of reduction not be grossly dispro portionate to the level achieved," according to Judge Bork. In that range of uncertainty, be added, EPA's consider ations of feasibility in setting standards "seem natural, perhaps inevitable, choices." By emphasizing available tech nology, the agency ensured maximum regulation "without the economic and social displacement that would accompa ny the closing of an industry or any substantial part of an industry." Distent Emphasizes Public Health Concern* Judge J. Skelly Wright dissented, claiming the "clear decision" by Congress was that Section 112 emission stan dards "should reflect public health considerations alone." The agency's discretion under Section 112 must account for an "ample margin of safety," without any reference to "feasibility considerations, either technological or econom ic," Judge Wright maintained. Without "solid statistics for harm," be said, "the costs of regulation will always dominate" an agency's decision on the extent of regulation necessary. The elimination of "entire industries" is not likely under a stricter standard, Judge Wright said, commenting that the 11-14-86 Environment Reporter 001*-*ai1/*M0+.IO ? AP00003619 1170 ENVIRONMENT REPORTER majority's fur of industry disruption was "greatly exaggerated." Because many air pollutants are carcinogens, Judge Wright uid, they can pose a danger at any concentration, and Congress was aware that these substances would re quire "the strictest regulation without regard to cost, even if the extent of that danger might be unclear at the third decimal point." EPA `Pleased' with Ruling Earl Salo, EPA acting assistant general counsel, told BNA Nov. S he was "very pleased" with the federal appeals court decision. "We are not especially surprised," he uid, "since we have bun interpreting the statute this way for a long time. What is unusual here is that it took IS years before we got a case right on point" Salo uid the 197$ standards, which are the ones in effect following the appeals court ruling, approximate the best technologically achievable reductions of vinyl chloride. Litigation COURT UPHOLDS EPA EMISSION STANDARDS, REJECTS AGENCY TIMING FOR IMPOSING LIMITS Federal limits on nitrogen oxide and particulate emissions from heavy-duty motor vehicles were upheld by a federal appeals court Nov. 7, but the court ruled that the Environ mental Protection Agency failed to follow requirements under the Clean Air Act concerning the time that the emis sion standards would take effect. EPA improperly required engine manufacturers to com ply with a nitrogen oxide standard Issued in 1985 by the 1988 model year, according to the U.S. Court of Appeals for the District of Columbia Circuit. The Air Act requires a fouryear lead time between when a standard is issued and when it takes effect, the court determined {NRDC v. Thomas, No. 8S-1294). Although EPA was "gratified" that the court upheld the majority of the agency's standard-setting actions, there was some disappointment with the D.C. Circuit's reading of requirements for lead times under the Air Act according to Nancy A. Ketcham-Colwill, an attorney in EPA's Office of General Counsel. Parties Challenging Emission Standards The D.C. Circuit rejected challenges to the nitrogen oxide and particulate standards raised by the Natural Resources Defense Council Inc., which was joined by six environmental groups and four individuals. In addition, the Engine Manu facturers Association, joined by five engine manufacturers and a trade association, challenged the strictness and timing of the standards. EPA issued emission standards for heavy-duty trucks in March 1985, requiring a reduction of nitrogen oxide emis sions to 6 grams per brake horsepower-hour (g/BHP-hr) and particulate emissions to 0.6 g/BHP-hr by the 1988 model year (Current Developments. March 15,1985, p. 1910). The agency was under a court order to issue the stan dards, due to an NRDC suit against EPA for failing to prepare standards in time for the 1985 model year (NRDC v. Ruckelshaus, 21 ERC 1953; Sept. 15,1984, p. 792). Inadequate Lead Time Provided Even though the nitrogen oxide standard was supposed to be in effect by the 1985 model year, EPA cannot use its delay in issuing a standard as a reason to circumvent the four-year lead time mandated in the Act the appeals court determined. Therefore, the ctandard cannot go into effect until the 1990 model year, the court ruled. In addition, the agency incorrectly imposed nitrogen oxide standards for light duty trucks that are also regulated as heavy-duty vehicles, according to the D.C. Circuit The agency regulates light-duty trucks that weigh between 8,000 and 8,500 pounds as heavy-duty vehicles, the court noted. EPA understood the statutory requirements for a fouryear lead time for imposing emission limits, but the agency had hoped the court would recognize' that EPA wanted to minimize delay in carrying out the 1985 standards, Ket cham-Colwill told BNA Nov. 12. Balancing Said Required In Setting Standards The appeals court rejected the environmental. group's challenge to the technological adequacy of the standard for nitrogen oxide emissions, holding that EPA did not have to set a standard based on the achievements of the leader in technology for the motor vehicle industry. Congress intended EPA's standard setting to be a balanc ing process, accounting for industrywide factors such as cost noise, energy, and safety, according to the D.C. Circuit. Congress did not express, with "laser-like clarity," require ments for the use of one manufacturer's engine design as the baseline for setting emission standards, the appeals court observed. Although there is reference to the use of technological leaders in the legislative history of the Air Act, EPA pro vided a reasonable approach in determining that the emis sion standards were to be set at a level "geared to what the more progressive manufacturers could achieve,'', the ap peals court said. The agency must also consider the other factors, including cost, the D.C. Circuit noted. Emission Averaging Upheld EPA's willingness to let engine manufacturers average mission levels among engine types to meet standards was also challenged by the environmental groups. The agency should have required each engine type to meet an emission standard, and the averaging of emissions by a manufacturer could lead to minimal penalties for non-complying engine types, according to the groups. The appeals court agreed with the agency's argument that averaging of emissions will provide more flexibility to manufacturers to allocate costs among different engine operations. At the same time, the court Mid, averaging will ensure that standards are met Non-compliance penalties will not be undermined by emission averaging, according to the D.C. Circuit, because a manufacturer can still be penalized if an entire fleet of engines fails to meet a standard. California's Challenge Rejected California challenged EPA's decision to base nitrogen oxide limits on the ability of diesel engine manufacturers to comply with the requirements. Although the appeals court agreed that it might have been more appropriate for EPA to Mt stricter standards based on gasoline vehicle compliance, the court said it could not review the state's objections. The state failed to raise its complaints during the public comment period when the limits were proposed, and was precluded from raising the objections in court according to the D.C. Circuit. Engine Manufacturers' Challenges The engine manufacturers complained that EPA based particulate emission standards on its prediction as to future 11-14-86 Copyright C1966 by The Bureau of Nstionsl Affairs, Inc., Washington, D.C. OOtS-SSII/M/tO-JO i f' AP00003620 120:1086 FEDERAL REGULATIONS Method 107--Determination or Vnrvi Chloride Content or Inprocess Wastewatdi Samples, amd Vinyl Chlo* ride Content or Polyvinyl Chloride Rum. Slurry. Wet Cake, and Um Sam ples Introduction Ptrformtnee of this method should not be attempted by persons unfamiliar with the operation of a ias chromatograph (OC). nor by those who art unfamiliar with source sampling, because knowledge beyond the scope of this presentation is required. Care must be exercised to prevent exposure of sampling personnel to vinyl chloride, a car cinogen. 1. Applicability and Principle 1.1 Applicability. This method applies to the measurement of the vinyl chloride mon omer (VCM) content of Inproeess wastewater samples, and the residual vinyl chloride monomer (RVCM) content of poly vinyl chloride (PVC) resins, wet cake, slurry, and latex samples. It cannot be used for polymer In fused forms, such as sheet or cubes. This method is not acceptable where methods from Section 304(h) of the Clean Water Act. 33 U.S.C. 1351 et seq. (the Pederal Water Pollution Control Amendments of 1072 tii amended by the Clean Water Act of HIT) an required. 1.1 Principle. The basis for this method relates to the vapor equilibrium that is es tablished at a constant known temperature in a eloaed system between RVCM, PVC min. water, and air. The RVCM in a PVC retin will equilibrate rapidly In a closed vessel, provided that the temperature of the PVC resin is maintained above the glass transition temperature of that specific min. (1.2 amended by 32 FR 20398, June 1, 1987] 2. Range and Sensitivity The lower limit of detection of vinyl chlo ride will vary according to the sampling and chromatographic system. The system should be eapable of producing a measure ment for a 50-ppm vinyl chloride standard that is at least 10 times the standard devi ation of the system background noise level. (2. revised by 52 FR 20398, June 1.1987] 3. Interferences The chromatograph columns and the cor responding operating parameters herein de scribed normally provide an adequate reso lution of vinyl chloride; however, resolution Interferences may be encountered on some sources. Therefore, the chromatograph op erator shall select the column and operating parameters best suited to his particular analysis requirements, subject to the ap proval of the Administrator. Approval is automatic provided that the tester produces confirming data through an adequate sup plemental analytical technique, such as analysis with a different column or OC/ mass spectroscopy, and has the data avail matograph backflush capability may be re able for review by the Administrator. quired. 4. Precision and Reproducibility An Interlabomlory companion between seven laboratories of three resin samples each split into three parts yielded s stand ard deviation of 2.63 percent for a sample with a mean of 2.09 ppm. 4.16 percent for a (ample with a mean ol 1.55 ppm. and 5.30 percent for a sample with a mean of 62.6t ppm. (6.3.1 reviled by 52 FR 20398. June l, 1987] 6.3.3 Chromatographic Columns Stain less steel 1 m by 3.2 mm end 2 m by 3.2 mm. both containing 50/60-mesh Porapak Q. The analyst may use other columns provid ed that the precision and accuracy of the analysis of vinyl chloride standards are not 8. Safety Do not release vinyl chloride to the labo ratory atmosphere during preparation of standards. Venting or purging with VCM/ air mixtures must be held to a minimum. When they are required, the vapor must be routed to outside air. Vinyl chloride, even at low ppm levels must never be vented inside the laboratory. Altar vials have bean ana lysed, the gas must be vented prior to re moval of the viol from the instrument turn table. Vials must be vented through a hypo dermic needle connected to an activated charcoal tube to prevent release of vinyl chloride into the laboratory atmosphere. The charcoal must be replaced prior to vinyl chloride breakthrough. impaired and he has available for review in formation confirming that there is adequate resolution of the vinyl chloride peak. (Ade quate resolution it defined as an area over lap of not more than 10 percent of the vinyl chloride peak by an lnterferent peak. Calcu lation of area overlap is explained in Appen dix C. Procedure 1: "Determination of Ade quate Chromatographic Peak Resolution.") Two 1.S3 m columns, each containing 1 per cent Carbowax 1500 on Carbopak B, have been suggested (or samples containing acet aldehyde. 6.3.3 Thermometer, o to 100*C, accurate to 0.1'C. [6.3.3 amended by 52 FR 20398. June 1. 1987] 6. Apparatus 5.1 Bampling. The following equipment Is required: 5.1.1 Olass bottles. 60-ml (2-csl capacity, with wax-lined screw-on tops, for PVC sam ples 6.1.2 Glass Vials Headspace vials, with Teflon-faced butyl rubber sealing discs lor water samples 6.3.4 Integrator-Recorder. To record chromatograms. [Former 6.3.4 and 6.3.5 removed, former 6.3.6 revised and redesignated as 6.3.4, and new 6.3.5 added by 52 FR 20398. June 1, 1987] 6.3.3 Barometer. Accurate to 1 nun Hg. 6.3.6 Refulfttors. For required ru cylin [6.1.2 revised by 52 FR 20398, June 1, der*. 1987] [Former 6.3.7 end 6.3.8 removed and for 6.1.3 Adhesive Tape. To prevent loosen mer 6.3.9 and 6.3.10 redesignated as 6.3.6 ing of bottle tops and 6.3.7, respectively by 52 FR 20398, 6.2 Sample Recovery. The following June 1. 1987] equipment Is required: [6.2 revised by 52 FR 20398. June 1. 6.1.7 Headspace Vial Fre-Praturizer. Nltrogen pressurized hypodermic needle Inside 1987] protective shield. (Blueprint available from Test Support Section, Emission Measure 6.2.1 Glass Vials. Headspace vials, with butyl rubber septa and aluminum caps Sili cone rubber is not acceptable. 6.2.2 Analytical Balance. Capable of de termining sample weight within an accuracy ment Branch. Office of Air Quality Plan ning and Standards, Environmental Protec tion Agency. Mall Drop 19. Research Trian gle Park. KjC. 27711.) 7. ReopenIs of z 1 percent. Use only reagents that are of chromato 6.2.3 Vial Sealer. To seal headspace vials graphic grade. 6.2.4 syringe. 100-al capacity. S3 Analysis The following equipment is required: g.3.1 Headspace Sampler and Chromato 7.1 Analysis. The following items are reauired for analysis: 7.1.1 Hydroten. Zero grade. 74.2 Hitrogen or Helium. Zero grade. graph. Capable of sampling and analyzing a constant amount of headspace gas from a sealed vial, while maintaining that vial at a temperature of M*C O.t'C. The chromato graph shall be equipped with a flame Ioniza tion detector. Ferkin-Elmer Corporation Models P-40, F-42. F-45. HS-5. and HS-100. and Hewlett-Packard Corporation Model 1B3S5A have been found satisfactory.' Chro [7.1.2 reviicd by 52 FR 20398, June 1. 1987] 7.14 Air. Zero grade. 7.1.4 Water. Interference free. [7.1.4 added by 52 FR 20398, June 1, 1987] Environment Reporter [Appendix B, Method 107] 142 r AP00003621 HAZARDOUS EMISSIONS 120:1089 1.2 Calibration. The following Itezni are the tap, and immediately tighten a cap on required for calibration: the bottle. Wrap adhesive tape around the 1.3.1 Cylinder Standards C4). Oas mix- cap and bottle to prevent the cap from loos ture standards (to-. SOO-, 2000- and 4000- ening. Place an identifying label on each ppm vinyl chloride in nitroccn cylinders). bottle, and record the date. time, and The tester may use cylinder standards to dl- sample location both on the bottles and in a redly prepart a chromatograph calibration log book. All samples should be kept refrig curve as described in 8ection S.2, IX the fol erated. lowing conditions are met: (a) The manufac turer certifies the sas composition with an (8.1.1 amended by 52 FR 20398. June 1. accuracy of 1 percent or better (see Sec 1987] tion 1.2.1.1). (b) The manufacturer recom mends a maximum shelf life over which the 1.1.2 Water Sampling. At the sampling (as concentration does not chance by great location fill the vials bubble-free to over er then 8 percent from the certified value, flowing so that a convex meniscus forms at (c) The manufacturer affixes the date of (as the top. The excess water is displaced as the cylinder preparation, certified vinyl chloride sealing disc Is carefully placed, with the concentration, and recommended maximum Teflon aide down, on the opening of the shelf life to the cylinder before shipment to vial. the buyer. Place the aluminum seal over the disc and 1.3.1.1 Cylinder Standards Certification. the neck of the vial, and crimp Into place. The manufacturer shall certify the concen Affix an Identifying label on the bottle, and tration of vinyl chloride in nitrogen in each record the date, time, and sample location cylinder by (a) directly analyzing each cylin both an the vials and in a log book. All sam der and (b) calibrating his analytical proce ples must be kept refrigerated until ana dure on the day of cylinder analysis. To cali lysed. brate hit analytical preoedure. the manufac turer shall use, as a minimum, a 3-point cali (8.1.2 amended by 52 FR 20398, June l, bration curve. It is recommended that the 1987] manufacturer maintain (1) a hlsh-eoncentration calibration standard (between 4000 and >000 ppm) to prepare his calibration curve by an appropriate dilution technique and (2) low-concentration calibration standard (between SO and 500 ppm) to verify the dilution technique used. If the difference between the apparent concentration read from the calibration curve and the true concentration assigned to the low-eoneen- trstion calibration standard exceeds 5 per cent of the true concentration, the manu facturer shall determine the source of error and correct it. then repeat the 3-point cali 5.2 Sample Recovery. Samples must, be run witbin 24 hours. 8.2.1 Resin Samples. The wefehl of the resin used must be between 0.1 and 4A trims. An exact weight must be obtained (1 percent) tor each sample. In the esse of suspension resins, a volumetric cup can be prepared for holding the required amount of sample. When the eup is used, open the sample bottle, and add the cup volume of resin to the tired sample vial (tared. indudins septum and aluminum cap). Obtain the exact sample weight, add lOOp) or about two equal drops of watsr, and Immediately aeal bration. the vial. Report this value on the dau 1.2.12 Verification of Manufacturer's sheet; it Is required for calculation of Calibration Standards. Before using, the BVCM. In the esse of dispersion resins, the manufacturer shall verify each calibration cup cannot be used. Weigh the sample in an standard by (a) comparing it to gas mixtures aluminum dish, transfer the sample to the prepared (with OO mole percent vinyl Chlo tared vial, and accurately weiah it In the ride) in accordance with the procedure de vial. After prepreuurlzatlon of the samples, scribed in Section T.l of Method 1M or by condition them for a minimum of 1 hour in (b) calibrating it against vinyl chloride cyl the 90* C bath. Do not exceed s hours. inder Standard Reference Materials Prepressurization la not required If the (SRM's) prepared by the National Bureau sample weight, as analyzed, does not exceed of Standards, if such SRM's are available. 0.2 gram. It It alio not required if eolution The agreement between the Initially deter of the prepreuurlzatlon equation yields an mined concentration value and thi verifica absolute prepressurlzation value that la tion concentration value must be within S within 30 percent of the atmospheric pres percent. The manufacturer must reverify all sure. calibration standards on a time interval con Note Some aluminum vial caps have a sistent with the shelf life of the cylinder center section that must be removed prior standards sold. to placing into sample tray. If the cap if not removed, the Injection needle will be dam 17.2.1.2 amended by 52 FR 20398. June 1, aged. 1987) I. Procedure (8.2.1 amended by 52 FR 20398. June I. 1987] 8.1 Sampling. >.1.1 PVC Sampling. Allow the resin or slurry to flow from a tap on the tank or silo until the up line has been well purged. Extend and fill a SO-ml sample bottle under >.2.2 Suspension Resin Slurry and Wet Cake Samples. Decant the water from a wet cake sample, and turn the sample bottle upside down onto a paper towel. Wait for the water to drain, place approximately 0.2 to 4.0 grams of the wet cake sample In a tared vial (tared. Including septum and alu minum cap) and seal Immediately. Then de termine the sample weight (El percent). All samples, welshing over 0.2 gram, must be prepressuriaed prior to conditioning for 1 hour at 90'C. except as noted In Section >3.1. A sample of wet oake It used to deter mine total solids (TS>. This is required for calculating the BVCM. [8.2.2 amended by 32 FR 20398, June 1, 1987] jls Dltpenton Resin Slurry and Oeon Latex Samples. The materials should not be Uttered. Sample must be thoroughly mixed. Using a tared vial (tared. including septum and aluminum cap) add approximately eight drops (0.2S to 0.SS g) of slurry cr latex using a medicine dropper. This should be done immediately after mixing. Seal the vial as soon as possible. Determine sample weight (1 percent). Condition the vial for 1 hour it 00 *C in the analyzer bath. Deter mine the TS on the slurry sample (Section IW [8.2.3 amended by 52 FR 20398, June 1. 1987) S.2.4 Inproeess Wastewater Samples. Using a tared vial (tared. including septum and aluminum cap) quickly add approxi mately 1 cc of water using a medicine drop per. Seal the vial as soon as possible. Deter mine sample weight (1 percent). Condition the vial for 1 hour at 90'C in the analyzer bath. (8.2.4 amended by 52 FR 20398, June t. 1987] (.3 Analysis. 8.3.1 Preparation of Equipment. Install the chromatographic column and condition overnight at ISO* C. In the first operation. Porapak oolumns must be purged for 1 hour at 230* C. Do not connect the exit end of the column to the detector while conditioning. Hydro gen and air to tha detector must be turned off while the column Is disconnected. >.3.1.1 Flow Rate Adjustments. Adjust flow rates as follows; a. Nitrogen Carrier Oas. Set regulator on cylinder to read >0 psig. Set regulator on chromatograph to produce a flow rate of 30.0 ce/min. Accurately measure the flow rate at the exit end of the column using the aoap film flowmeter and a stopwatch, with the oven and column at the analysis temper ature. After the instrument program ad vances to the **B~ (backflush) mode, adjust the nitrogen pressure regulator to exactly balance the nitrogen flow rate at the detec tor as was obtained In the "A" mode. b. vial Prepressurtser Nitrogen. After the nitrogen carrier Is set. solve the following 2-12-33 [Appendix B, Method 107] Published by THE BUREAU OF NATIONAL AFFAIRS INC.. Washington. DC. 20037 143 f AP00003622 120:1090 FEDERAL REGULATIONS equation and adjust the pressure on the vial prepreasurlzer accordingly. When: TiAmblent temperature. 'K. T.-CondltlonLn* bath temperatun, `K. r.sOax ehromatocraph absolute dosing preaiurt (analysis mode), k Pa. P.i-Water vapor pressure 90' C (ISM mm Hg>. P..-Water vapor pressure Q 22*C (19.8 mm Es>. 7M mm He per k Pa. 10 k Pa-Factor to adjust the prepressurised pressure to slishtly leu than the dosing pressure. Because of gauge errors, the apparatus may over-pressurite the vial. If the vial pressure Is at or higher than the dosing pressure, an audible double injection will occur. If the vial pressure is too low, errors will occur on resin samples because of inad equate time for head-spsee css equilibrium. This condition can be avoided by running severs! standard gas samples at various pressures around the calculated pressure, and then selecting the highest pressure that does not produce s double Injection. All simples and standards must be pressurized for go seconds using the vial prcprexsurlzer. The vial is then placed into the 90* C condi tioning bath and tested for leakage by plac ing s drop of water on the septum at the needle hole. A clean, burr-free needle Is mandatory. c. Burner Air Supply. Set regulator on cyl inder to read SO psig. Set regulator on chro matograph to supply sir to burnsr at a rate between 250 and ZOO cc/mln Check with bubble flowmeter. d. Hydrogen Supply. Set regulator on cyl inder to read 30 psig. Set regulator on chro matograph to supply approximately 35 a S cc/min. Optimize hydrogen flow to yield the most sensitive detector response without ex tinguishing the flame. Check flow with bubble meter and record this flow. 1.3.1.3 Temperature Adjustments. Bet temperatures as follows: a. Oven (chromatograph column). 140' C. b. Dosing Line. 150' C. c. Injection Block. 1T0` C. d. Sample Chamber, Water Temperature, 90' C 1.0' C. 1.2.1J Ignition of Flame Ionization De tector. Ignite the detector according to the manufacturers instructions. I.l.t.4 Amplifier Balance. Balance the amplifier according to the manufacturers Instructions. I.3.S Programming the Chromatograph. Program the chromatograph as follows: a. I--Doting or Injection Time. The normal setting is 9 seconds. ' b. A--"Analysis Time.'' The normal set ting is approximately TO percent of the VCM retention time. When this tuner termi nates, the programmer Initiates baekfluahIng of the first column. e. B--Baekflushlng Time. The norms! set ting 1* double the "anslytis time." CL W--Stabilization Time. The normal get ting la 0.5 min to 1.0 min. e. z--Number ol Analyses Per Sample. The normal setting it one. 9JJ Preparation of Sample Turntable. Before placing any sample Into turntable, be certain that the center section of the alu minum cap has been removed. The num bered sample vials should be placed In the corresponding numbered positions In the turntable. Insert ssmples in the following order Position 1 and a--Old 3000-ppm standards for conditioning. These are necessary only after the analyzer has not been used for 34 hours or longer. Position 3--50-ppm standard, freshly pre pared. Position 4--500-ppm standard, freshly pre pared. Position 5--3000-ppra standard, freshly prepared. Position 8--4000-ppm standard, freshly prepared. Position 7--Sample No. 7 (This is the fim sample of the day. but is given as 7 to be eonslstent with the turntable and the inte grator printout.) After all ssmples have been positioned. Insert the second set of 50-, 500-, 2000-. and 4000-ppm standards. Ssmples, Including standards, must be conditioned In the bath of 90* C for 1 hour (not to exceed 5 hours). [8.3.3 amended by 52 FR 20398, June I, 1987] 3.3.4 Start Chromatograph Program. When all samples, Including standards, have been conditioned at 90* C lor 1 hour, xtart the Analysis program according to the manufseturer's instructions. These Instructions must be carefully followed when starting and stopping a program to prevent damage to the dosing assembly. 3.1.5 Determination of TS. For wet cake, slurry, resin solution, and PVC latex sam ples, determine TS for each sample by accu rately weighing approximately 3 to 4 grams of sample in an aluminum pan before and after placing In a draft oven uos to 110* C). Samples must be dried to constant weight. After first weighing, return the pan to the oven for a short period of time, and then re weigh to verify complete dryness. The TS are then calculated as the final sample weight divided by Initial sample weight. 9. Calibration Calibration is to be performed each 9-hour period the chromatograph la used. Alterna tively. calibration with duplicate 50-. S0O-, 1,000-, and 4,000-ppm standards (hereafter described at a four-point calibration) may be performed on a monthly basis, provided that a calibration confirmation test consist ing of duplicate analytes of an appropriate standard it performed once per plant shift, or onet per chromatograph carrousel oper ation (If the chromatograph operation is lest frequent than once per shift). The crite rion for acceptance of each calibration con firmation test It that both analyses of 500ppm standards (2,000-ppm standards if dis persion resin (excluding latex resin) samples are being analyzed] must be within 9 per cent of the most recent four-point calibra tion curve. If this criterion is not met then m complete four-point calibration must be performed before sample analyses can pro ceed. [9. revised by 52 FR 20398, June 1,1987] 9.1 Preparation of Standards. Calibration standards are prepared as follows: Place 100pl or about two equal drops of distilled water in the sample vial, then fill the vial with the VCM/tUtrogen standard, rapidly seat the septum, and seal with the alumi num cap. Use a (4-in. stainless steel line from the cylinder to the vial. Do not use rubber or Tygon tubing. The sample line from the cylinder must be purged (Into s properly vented hood) for several minutes prior to filling the vials. After purging, reduce the flow rate to 500 to 1000 ec/min. Place end of tubing Into vial (near bottom). Position a septum on top of the vial, press ing it ggalnst the (t-ln. tilling tube to mini mize the size of the vent opening. This is necessary to mlmlmlze mixing air with the standard In the vial. Each vial is to be purged with standard for 90 seconds, during which time the filling tube is gradually slid to the top of the vial. After the 90 seconds, the tube la removed with the septum, simul taneously sealing the vial. Practice will be necessary to develop good technique. Rubber gloves should be wom during the sbove operations. The sealed vial must then be pressurized for SO seconds uslns the vial propressurizer. Test the vial for leakage by placing s drop of water on the septum at the needle hole. Prepressurization of stand ards Is not required unless samples have been prepressurized. [9.1 amended by 32 FR 20398, June 1, 1987] 9.2 preparation of cnromatograph cali bration Curve. Prepare two vials etch of 50-. too-. 2.000-. and 4,000-ppm standards. Run the calibra tion samples in exactly the same manner as regular samples. Plot A., the integrator area counts for each standard sample, versus C,, the concentration of vinyl chloride In each standard sample. Draw a straight line through the points derived by the least squares method. [9.2 amended by 32 FR 20398, June 1. 1987] Environment Reporter (Appendix B, Method 107] 144 HAZARDOUS EMISSIONS ,,Ml| 120:1091 >0. Calculations 10.1 Response Fsctor. if the calibration curve described in Section (.2 pisses through sero, sn avenge response factor. Rt. may be used to fscilltate com. putatlon of vinyl chloride sample con* centralions. To compute R first compute a response factor, R* for each sample ss follows: A, K. -- la. 107-1 R.-=Rejpease factor, area eoonts/ppm. A,mChromatograoi ares counts ofvinyl chloride for the sample, area counts. CcCoacentrstioB of vinyl chloridela the standard sample, ppm. Volume of vapor phase, cm*. IX ~ 09653 8um the Individual response tsetors, and calculate Re If the calibration curve does not pass through sero, use the ealihntion curve to determine each simple concentra tion. flO.I revised by 52 FR 20398, June 1, 1987; corrected by S3 FR 36972, Septem ber 23, 1988] 10.2 ResMusI Vinyl Chloride Monomer Concentration. (&~) or Vinyl Chloride Mon omer Concentration. Calculate CU In ppm or mg/kg u follows: [10.2 revised by 52 FR 20398. June 1, 1987] AsPa TH V + *p <TS> T2^0- TS) tJi rvc f *1 Eq. 107-2 Where: A,-Chromatogram ares counts of vinyl ehloride for th* sample. P,-Ambient atmospheric pressure, mm He. R,-Response fsctor In arcs counts per ppm VCM. T,-Ambient laboratory temperature, 'K. M.-Moleculir weight of VCM, 62A g/mole. V,-Volume of the vapor phase, cm'. R-Oas constant, (62260 cm*) (mm Hg)/ (moleX'K). m-8ample weight, g. K,-Henry'i Law Constant for VCM is PVC 60* C, 6.62xlO',g/g/ma Hg. TS-Total solids expraoed as a dteimal fraction. Ts - Eauilibrium temperature. *K. K.-Henry's Law Constant for VCM in water 0 B0" C. 7x 10" g/g/mm Hg. V.-Vial volume, cm*. IJ6. Density of PVC at 90'C. g/em1. t.M$3-Denslty of water at WC, g/cm*. m(l-TSl 1.36 0.9653 Results calculated using these equations represent concentration based on the total sample. To obtain results based on dry PVC content, divide by TS. 11. Bibliography [II. head revised by 52 FR 20398, June 1, 19871 1. B.F. Goodrich. Residue) Vinyl Chloride Monomer Content of Polyvinyl Chloride Resins. Latex. Wet Cake. Slurry and Water Samples. B.F. Ooodrich Chemical Group Standard Test Procedure No. ioos-E. BP. Goodrich Technical Canter. Avon Lake, Ohio. October g. 1979. Z Berens. A.R. The Diffusion of Vinyl Chloride In Polyvinyl Chloride. ACS--Divi sion of Polymer Chemistry. Polymer Pre prints IS (21:197. 1974. 3. Berens. A.R. The Diffusion of Vinyl Chloride in Polyvinyl Chloride. ACS--Divi sion of Polymer Chemistry, Polymer Pre prints is (21:203.1974. 4. Berens. AJt,, L-B. Crider. CJ. Tomanek, and J.M. Whitney. Analysts for Vinyl Chloride In PVC Powders by Bead--Space Gas Chromatography. Journal of Applied Poly mer Science. 79:3169-3172.1975. 6. Mansfield, RA. The Evaluation of Henry's Law Constant fZp) and Water En hancement In the Perkin-Elmer Multifraet r-40 Gas Chromatograph. BP. Goodrich. Avon Lake. Ohio. February 10,1973. - Method 107A--Determination or Vinyl Chloride Coxteut or Solvents, RbirSolvent Soldtior. Polyvinyl Chloride Ream. Reals Stnir, Wet Resin, and Late* Samtlzs Introduction Performance of this method should not be attempted by persona unfamiliar with the operation of a gas chromatograph (GCl or by those who are unfamiliar with source campling because knowledge beyond the scope of this presentation is required. Care must be exercised to prevent exposure of sampling personnel to vinyl chloride, a car cinogen. 1. Applicability and Principle 1.1 Applicability. This Is an alternative method and applies to the measurement of the vinyl chloride content of solvents, resin solvent aolutiona. polyvinyl chloride (PVC) nsin. wet cake slurries, latex, and fabricat ed resin samples. This method Is not accept able where methods from Section 304(h) of the Clean Water Act. 33 VAC. 1251 et seq.. (the Federal Water Pollution Control Act Amendments of 1272 as amended by the Clean Water Act of 1977) are required. 1.2 Principle. The hula for this method lies in the direct inleetion of a liquid sample Into a chromatograph and the subsequent evaporation of all volatile material into the carrier gas stream of the chromatograph, thus permitting analysis of all volatile material Including vinyl chloride 2. Range and Sanritivitv The lower limit of detection of vinyl chlo ride in dry PVC ruin la 0.2 ppm. For resin solutions, latexes, and wet resin, this limit rises inversely as the nonvolatile (resin) con tent decreases. With proper calibration, the upper limit may be extended as needed. A Interference! The chromatograph columns and the cor responding operating parameters herein de scribed normally provide an adequate reso lution of vinyl ehloride. In esses where reso lution interferences are encountered, the chromatograph operator shall select the column and operatine parameters best suited to his particular analysis problem, subject to the approval of the Administra tor. Approval is automatic, provided that the tester produce* confirming data through an adequate supplemental analyti cal technique, such as analysis with a differ ent column or GC/mass spectroscopy, and has the data available for review by the Ad ministrator. 4. Precision cad Reproducibility A standard sample of latex containing 181.8 ppm vinyl chloride analyzed 10 times by the alternative method showed a stand ard deviation of 7A percent and a mean error of 0.21 percent. A sample of vinyl chloride copolymer resin solution was analyzed 10 times by the alternative method and showed a standard deviation of gA percent at a level of 35 ppm. S. Safety Do net release vinyl chloride to the labo ratory atmosphere during preparation of standards. Venting or purging with vinyl ehloride monomer (VCM) air mixtures must be held to minimum. When purging Is re quired. the vapor must be routed to outside air. Vinyl chloride, even at low-ppm levels, must never be vented Inside the laboratory. 10-7-68 [Appendix B, Method 107A] Publithad by THE BUREAU OP NATIONAL AFFAIRS INC.. Washington. D.C. 20037 S3 F AP00003624 120:1092 FEDERAL REGULATIONS g. Apparatus 1.1 BampUng. The following equipment It required: 1.1.1 Glass Bottle*. l8-o* wide mouth wide polyethylene-lined, eerew-on tope. 6.1.2 Adhesive Tape. To prevent loosenin* of bottle tops. (.2 Sample Recovery. The following equipment It required: 6.2.1 Glut Visit. 20-ml capacity with polycone terew caps. (.24 Analytical Balance. Capable of weighing to 0X11 gram. 6.2.2 Syringe. SO-mieroUter site, with re movable needle. (.2.4 Fritted Glee* Bparter. Fine porotity. 6.2.6 Aluminum Weighing Dishes. (.2.6 Sample Roller or Shaker. To help dltMlve sample. 6.3 Analysis. The following equipment It required: (.3.1 Oat Chromatograph. Hewlett Pack ard Uode) 5720.1 or equivalent. 6.3.2 Chromatograph Column. Stainless steel, 6.1 m by 2.2 mm, packed with 20 per cent Tergitol E-S5 on Chrometorb W AW 60/80 mesh. The analyst may use other col umns provided that the precision and accu racy of the analysis of vinyl chloride stand ards are not Impaired and that he his avail able for review information confirming that there Is adequate resolution of the vinyl ehlorlde peak. (Adequate resolution is de fined as an area overlap of not more than 10 percent of the vinyl ehlorlde peek by an Interfering peak. Calculation of area overlap it explained In Apendix C. Procedure 1: "De termination of Adequate Chromatographic Peak Resolution.") 64.3 Valeo Instrument Six-Port Rotary Valve. For column back flush. 64.4 Septa For chromatofiaph injection port. 64-5 Injection Port Liners. For chro matograph used. 64.6 Regulators. For required gas cylin ders. (4.7 Soap Film Flowmeter. Hewlett Packard No. 0101-0113 or equivalent. 6.4 Calibration. The following equipment it required: 6.4.1 Analytical Balance. Capable of weighing to =0.0001 g. 6.4.2 Erlentneyer Flask With Glass Stop per. 125 ml. 6.44 Pipet*. 0.1. 04.1.6.10. and 50 ml. (.4.4 Volumetric Flasks. 10 and 100 ml. 7. Rtaetnts Use only reagents that are of chromato graph grade. 7.1 Analysis. The following Items are re quired: 7.1.1 Hydrogen Gas. Zero grade. 7.1.2 Nitrogen Oaa Zero grade. 7.14 Air. Zero grade. 7.1.4 Tetrahydrofuran <THF). Reagent grade. Analyte the THF by Injecting 10 microliters Into the prepared gat chromatograph. Compare the THF chromatogram with that shown In Figure 107A-1. If the chromato- tram Is comparable to A the THF should be sparged with pun nitrogen for approxi mately * hours usini the fritted fiats sparg er to attempt to remove the interfering peak. Reanalyze the sparged THF to deter mine whether the THF la acceptable for use. If the scan la comparable to B. the THF Should be acceptable far use in the analysis. Intsrferina pek Time, minutes Figure 107A-1 7.1.5 N. N-Dimethylacetamlde (DMAC). Spectroeraphie grade. For use In place of THF. 7.2 Calibration. The foliowine item la re quired: 7.2.1 Vinyl Chloride M. Percent. Ideal Gat Product* lecture bottle, or equiva lent. For preparation of standard solu tions. g. Procedure S.l Sampling. Allow the liquid or dried resin to flow from a tap on the tank, alio, or pipeline until the tap has been purged, nil a wlde-mouth pint bottle, and immediately tightly cap the bottle. . Place an identifying label on each bottle and record the date, time, sample loca tion. and material. 6.2 Sample Treatment. Sample must be run within 24 hours. 6.2.1 Resin Samples. Weigh 6.00 0.01 g of THF or DUAC In g tend 20-ml viL Add 1.00 0.01 g of realn to the tared vial con taining the THF or D14AC. Close the vial tightly with the screw cap. and shake or otherwise anute the vial until complete so lution of the realn is obtained. Shaking may require several minutes to several hours, de pending on the nature of the regin. g.2.2 Suspension Resin Slurry and Wet Resin Sample. Slurry must be filtered using a small Buchner funnel with vacuum to yield a wet resin sample. The filtering procfa must be continued only at long aa a steady stream of water la iHny from the funnel. Excessive filtration time could result la some lorn of VCU. The wet resin sample ii weighed Into a tared 20-ml vial with THF or DMAC at described earlier for realn sta ples (*4.1) and treated the same aa the resin sample. A sample el the wet resin it used to determine total solidi as required for calcu lating the Ktidual VCM (Section 5.3.4). 6.2.3 Latex and Reals Solvent Solution*. Sample* must be thoroughly mixed. Weigh LOO 0.01 g of tha latex or main-solvent so lution into a 20-ml vial containing 940 a 0.01 g of THF or DUAC as for the resin templet (8.2.1). Cap and shake until complete solution la obtained. Determine the total solids of the latex or realn solution sample (Section 8.3.4). 8.2.4 Solvents and Non-vlacous Liquid Samples. No preparation of these samples is required. Tha neat samples are injected di rectly into the QC. 84 Analysis. (4.1 Preparation of OC. Install the chromatographic column, and condition over night at 70' C. Do not connect the exit end of the column to the detector while condi tioning. i Environment Reporter [Appendix B, Method 107A] 04 r AP00003625 HAZARDOUS EMISSIONS 12&1C83 KJ.1.1 new Rate Adjustments. Adjust rerun ene or more samples from the preced known concentration. Draw a straight line the now rate as follows: ing day to test stability and precision prior through the points derived by the least a. Nitrogen Carrier Ok. set regulator on to starting on the current day's work. quant method! cylinder to read 60 pelf, set "niimm now 8.3.4 Determination of Total Solids (18). 10. Calculation* controller on the chromatograph using the For wet reein, resin solution, and PVC latex 10.1 Response Factor. From the calibra soap Aim flowmeter to yield a flow rate Of aamplea determine the TS for each earaple tion curve described in Section 0.2. select 40 ce/min. by accurately weighing approximately 3 to 8 the value of C, that corresponds to H, for b. Burner Air Supply. Set regulator on the grams of sample into a tared aluminum pan. each sample. Compute the response factor, cylinder at 40 paig. Set regulator on the The Initial procedure is as follows: K* for each sample as follows: chromatograph to supply air to the burner a. Where water is the major volatile com to yield a flow rate of 250 to too ee/mln ponent: Tare the weighing dish, and add 3 R. -- so. IOTA-1 using the flowmeter. to 8 grams of sample to the dish. Weigh to H, c. Hydrogen. Set regulator on cylinder to read eo pais. Set regulator on the chromato graph to supply 20 to 40 ec/mln ueiiif the the nearest milligram. b. Where volatile solvent is the major volatile component: Transfer a portion of where: R-Chrometograph respowsr factor, ppaf flowmeter. Optimise hydrogen flow to yield the most sensitive detector response with out extinguishing the flame. Cheek flew the sample to a 30-ml screw cap vial and eap immediately. Weigh the vial to the nearest Ci-Concenlration of vinyl chloride ia the milligram. Uncap ths vial and transfer a 3- itiodud impli, ppBL with flowmeter and record this flow. to 5-gram portion of the ample to a tared H,-Ptak height of the standard sample, nun. d. Nitrogen Back Flush Gas. Sat regulator on the chromatograph using ths aoep film flowmeter to yield a flow rate Of 40 ee/min. AS.1.2 Temperature Adjustment!. Bet temperature as follows: a. Oven (chromatographic column) at 70* C. b. Injection Port at 100* C. e. Detector at 200* C. 5.3.1.2 Ignition of name Ionization De tector. Ignite the detector according to the manufacturer's instructions. Allow system to stabilize approximately 1 hour. 8.3.1.4 Recorder. Set pen at aero and 'start chart drive. 3.3.1.5 Attenuation. Bet attenuation to yield desired peak height depending on sample VCM content. 8.3.2 Chromatographic Analyse!. a. Sample Injection. Remove needle from 50-mlcroliter syringe. Open sample vial and draw 50-mlcroliten of THP or DUAC sample recovery solution into the syringe. Recap sample vial. Attach needle to the sy ringe and while holding the syringe vertical ly (needle uppermost), eject 40 mierollten into an absorbent tissue. Wipe needle with tissue. Now Inject 10 mierollten into chro.matoeraph system, itepsst the injection until two consecutive values for the height of the vinyl chloride peak do not vary mare than 6 percent, flee the average value for these two peek heights to compute the simple concentration. b. Back Plush. After 4 minutes has elapsed after sample Injection, actuate the back flush valve to purge the first 4 feet of . the chromatographic column of solvent end other high boilers. & Sample Data. Record on the Chromato graph strip chart the data from the sample lftbftle d. Elution Time. Vinyl chloride elutes at 2A minutes. Acetaldehyde elutes et 3.T min utes. Analysis Is considered complete when chart pen become! stable. After 5 minutes, reset back flush valve and inject next sample. 3.3.3 Chromatograph Servicing. a. Septum. Replaoe after five sample in jections. b. Sample Port liner. Replace the sample aluminum weighing dish. Recap the vial and rewelsh to the nearest milligram. The vial [10.1 corrected by 53 FR 36972, Septem weight lots Is the sample weight. ber 23, 1988] To continue, now place the weighing pan in a 130' C oven for 1 hour. Remove the dish and allow to cool to room temperature in a desiccator. Weigh the pan to the near 16.2 Residual vinyl chloride monomer concentration (Cn,) or vinyl chloride monomer concentration in reein: tit 0.1 mg. Total solids Is the weight of ma CW-WHJR, terial In the aluminum pan after heating di Eq. 107A-2 vided by the net weight of sample added to Where: the pan originally times 100. C*.--Concentration of residue! vinyl chloride f. CaflOrgffon Ofliu Chromatograph 6.1 Preparation of Standard! Prepare a 1 percent by weight (approximate) solution of vinly chloride in THP or DUAC by babbling vinyl chloride gas from a cylinder Into a monomer, ppm. H,-Peak height of sample, mm. R,-Chromatograph response factor. [10.2 corrected by 53 FR 36972, Septem tared 138-ml glass-stoppered flask contain ber 23, 1988] ing THP or DUAC. The weight of vinyl chloride to be added should be calculated prior to this operation, Le., 1 percent of the weight of THP or DMAC contained In the tared flask. This must be carried out In a laboratory hood. Adjust the vinyl chloride flow from the cylinder ao that the vinyl 103 Samples containing volatile material, Le* resin solutions, wet resin, and latexes: C*,- RR,000) ------------ TS EQ.101A-3 where: chloride dissolves essentially completely In TS--Total solids ia the sample, weight the THP or DUAC and is not blown to the fraction. atmosphere. Take particular care not to vo lethe any of the solution. Stopper the flask [10.3 corrected by 53 FR 36972, Septem and swirl the solution to effect complete ber 23, 1988] mixing. Weigh the stoppered flask to near est 0.1 mg to determine the exact amount of vinyl chloride added. Plpet 10 ml of the approximately 1 per cent solution into a 100-mi glass^toppered 10.4 Samples of solvents and in process wastewater Cgvc RR ------- Be- X07A-4 volumetric flask, and add THP or DUAC to fill to the mark. Cap the flask and Invert 10 Where: ____ to 30 times. This solution contains approxi 0.888-Specific cravlty ofTHF. mately 1,000 ppm by weight of vinyl chlo (10.4 corrected by 53 FR 36972, Septem ride (note the exact concentration). Plpet SO*. 10-. 5-. 1-. 0.5-. and 0.1-ml all- quote of the approximately 1,000 ppm solu tion into 10 ml glass stoppered volumetric flecks, Dilute to the mark with THP or DUAC. cap the flasks and invert each 10 to 30 time! These solutions contain approxi mately 800, 100, 80, 10, 8, and 1 ppm vinyl ber 23,1988] 11. Bibliography 1. Communication from R. N. Wbeeler. Jr* Union Carbide Corporation. Part 61 Na tional Emissions Standards for Hazardous Air Pollutants Appendix B. Method 107--Al ternate Method. September IS. 1877. chloride. Note the axaet concentration of Uxtkop tee--DsOTurrnaTloH op each one. These standards are to be kept under refrigeration in stoppered bottles, FsancviATS srn> Osseous Axsexic Emissions and must be renewed every 3 months. 8.3 Preparation of Chromatograph Cali [Method 108 added by 51 FR 28025, bration Curve. Obtain the OC for each of the six final so August 4. 1986] port Uner with a clean spare after five lutions prepared in Section 3.1 by using the ample injections. procedure in Section 8.1.2. Prepare u chart 1. Applicability and Principle c. Chromatograph Shutdown. If the chro- plotting peak height obtained from the 1.1 Applicability. This method applies to metotragph has been shut down overnight. chromatogram of each solution versus the the determination of Inorganic arsenic (As) 10-7-88 [Appendix B, Method 108] Published by THE BUREAU OF NATIONAL AFFAIRS INC.. Washington, D.C. 20037 85 r AP00003626