Document zK1J7ELZ5gVv01LLna7K38zg

Monsanto Monnnto Comeiny 800 *.undfcargn louitvard St.Louia,Miaaogri 83186 *ona: 314) 884*1000 August 4, 1978 Office of Toxic Substances (TF-794) Environmental Protection Agency 401 "M" Street, S.W. Washington, D. C. 20460 ATTENTION: Ms. Joni T. Repasch Dear Ms. Repasch: On June 7,. 1978, the EPA published a proposed rule in 43 Federal Register beginning on page 24,802 covering Manufactur ing, Processing, Distribution in Commerce, and Use Bans of Polychlorinated Biphenyls (PCB's). This letter sets forth Monsanto Company's comments on the proposed rule. Please include these comments, with the attachment, in the official record of this proceeding. Our comments are divided into two parts. The first part provides general comments which focus on the EPA's proposed maximum concentration of 50 ppm and the justifiable need for a heat transfer system use authorization. The second part sets forth specific language and recommendations related to these matters, and to certain other matters in specific sections of the proposed rule. GENERAL COMMENTS The proposed immediate, mandatory removal from service of heat transfer systems containing fluids whose PCB concentra tion equals 50 ppm or greater is not justified in view of the lack of any significant risk to health or the environment. To require industry immediately to remove from service and drain all these systems or to replace the systems entirely to meet a 50 ppm concentration level is an arbitrary mandate and will generate a substantial, immediate risk of injury and environmental harm from spillage and result in an abrupt, massive buildup of PCB contaminated fluid without available disposal facilities or tij?e to develop orderly plans to -2- carry out the disposal. Furthermore, the EPA cites no evidence in the preamble of any environmental or health hazard to support the proposed reduction in the regulated concentration of PCB's in a PCB mixture under Section 761.2(w) from 500 ppm to 50 ppm, nor does the May, 1978 EPA Support Document/Draft Voluntary Environmental Impact Statement upon which the proposed rules are based support this reduction. Therefore, Monsanto strongly recommends the EPA withdraw this proposed reduction. Yet even with the withdrawal of the 50 ppm concentration, it is imperative that an authorization be granted for continued use of heat transfer systems containing residual PCB's. The evidence conclusively establishes that the heat transfer system use activity meets the requirements for granting an authorization to allow continued operation, and Monsanto requests the EPA to grant such an authorization. On page 42 of the EPA Support Document, the EPA states that before it grants an authorization for continuation of a non-totally enclosed use activity the following two requirements must be met: (1) the activity authorized must not pose an unreason able risk of injury to health or the environment, and (2) the ban would cause major and extensive economic disruptions. Regarding the first requirement, an analysis of potential PCB exposure from various sized PCB contaminated heat transfer systems is illustrative of the insignificant risk of injury to health or the environment posed by authorizing continued use of these systems. The highest risk area for leakage in these systems is in the pump seals, but inspection maintenance for pump seals is also emphasized because of user awareness of this risk. A pump seal failure would be rapidly detected by inspectors or automatic monitoring devices and' loss of fluid would be unlikely to exceed from 1% to 5% of the total fluid volume in the system. For example, a 500 gallon system containing 1% volume of PCBs has a total PCB content equal to 60 pounds, and a leak of 5% of the total fluid (which is unlikely) would release only 3 pounds of PCB's. Similarly, a 10,000 gallon system containing a PCB contamination of 500 ppm has a total PCB content of 60 pounds, and a leak of 5% of the total fluid (again unlikely) would release only 3 pounds of PCB's. A leakage of 3 pounds of PCB's is below the EPA's designated harmful quantity for PCB spills into navigable waters (10 pounds) under 40 CFR 118.4Vissued pursuant to the Federal AC* CCeCfcS 1 -3- Water Pollution Control Act and, in Monsanto's opinion, does not create a significant risk to health or the environment. Furthermore, the above examples are worst case situations because they assume no spill prevention program (for example, diking) or effluent control program in effect. If such programs were in effect it would be highly unlikely for any amount of PCB's approaching 3 pounds to escape into the environment. The proposed immediate, mandatory removal of these systems from service also meets the standards for the second require ment for an authorization by resulting in substantial, widespread economic disruption. Since the EPA states in the preamble and the Support Document that it lacks information necessary to evaluate the impact of the ban on the heat transfer system use activity, the following information is presented which is generally consistent with the specific information requests in 43 FR 24810 relative to hydraulic die-casting systems, and which clearly and conclusively demonstrates this economic impact: 1. Number of systems contaminated. It is estimated that when the sale of PCB heat transfer fluids was discontinued in 1972 there were 450 heat transfer systems using this fluid. This estimate is based on (a) the current level of new systems introduction, (b) the ten year period (1962-72) during which PCB heat transfer systems were specified, and'(c) the total volume of 20 million pounds of PCBs sold for heat transfer systems from 1962 through 1971 as reported by the 1972 Interdepartmental Task Force on PCBs COM-72-10419 pages 6 and 7. 2. The average liquid volume of the systems. It is estimated that about one-half of all PCB contaminated heat transfer systems have less than a 500 gallon liquid volume. Systems greater than 500 gallon volumes are estimated to average about 2,000 gallons.. These estimates are based upon a review of systems" filled in 1977 and 1978 and upon the average system size of Monsanto units converted in the 1970-72 period - a typical cross section of the industry pattern. 3. The range of liquid volume. Heat transfer systems of greater than 10,000 gallons are unusual. Small experi mental pilot plant units may contain 50 gallons. 4. Fluid top-off. Heat transfer fluids are designed to be thermally stable within the recommended temperature range of operation and to operate at low pressure. Accordingly, routine top-off is minimal. Average top- cc&kfcc A -4 off may be higher because of fluid removal from parts of the system when maintenance is performed on other components of the system such as pumps, valves and heaters or in the event a system malfunction causes a thermal degradation of the fluid. Monsanto's experience indicates that some systems may require no top-off for three to five years; other systems may require a topoff of 50% in one year if major maintenance programs requiring drainage are undertaken. However, systems are not routinely drained. 5. Efforts to reduce PCB contamination. We estimate that greater than 90% of the plants operating with PCB heat transfer systems converted to alternative non-PCB fluids in 1970-72. Some of those converting drained and refilled the system with non-PCB fluid; some drained, flushed and refilled; some drained, chemically cleaned and refilled; and other dismantled and discontinued use of liquid systems. We estimate that these efforts have generally resulted in current system PCB levels in the industry ranging from around 100 ppm to several thousand ppm. Recovery from heat transfer systems would generally be expected to be greater than 90%. Note that for the purposes of these comments the term "recovery" is defined as the percentage of total system fluid volume extracted from the system by draining. 6. Servicing fluids to remove PCB*s from systems. Following is a review and analysis of the potential success of three methods the EPA has considered for PCB removal: (a) Drain and refill. If small systems (500 gallons or less) are drained and refilled with a non-PCB fluid, a 98-99% recovery of dissolved PCB's can be expected. This service method will not, however, remove PCB absorbed by the heat transfer system walls. In the case of systems larger than 500 gallons a 90-99% recovery can be expected depending upon the complexity of the piping layout and heat exchanger design. (b) Distillation is of questionable efficiency in separating and removing PCB's from heat transfer fluid due to the similarity in boiling ranges of PCB's and non PCB heat transfer fluids. The boiling ranges for these respective fluids are as follows: PCB Therminol FR 1 617#F - 690#F Thermi.no! FR 2 644#F - 707#F QQbCt"! AC* -5- Non-PCB Therminol 55 Therminol 66 635F - 734F 643F - 668F (c) Carbon treatment experiments to remove PCB's are currently being conducted to investigate the carbon treatment removal method, but no data are presently available. 7. Present PCB levels in systems - We estimate the range of PCB concentration levels m heat transfer systems to be from about 100 ppm to about 30,000 ppm. 8. Cost of new systems - The cost of a new system could result in a large dollar expenditure compared to the much lower cost of replacing fluid. Replacing a heat transfer system may well mean totally rehabilitating a plant because these systems are integral parts of the users' plants, and replacement of heat transfer systems is impossible without replacement of reactors, coolers and heat exchangers. Such replacement would usually require a multimillion dollar expenditure. Further, the costs of lost production would be extraordinarily high in terms of lost wages, lost sales and product shortages which could impact a wide area of other manufacturing industries using finished products during the length of time required to replace each system. Thus the immediate, mandatory removal of these systems from service would result in substantial, widespread economic disruption in industries using heat transfer systems and the products produced from processes using such systems. 9. Cost of draining and refilling systems - We estimate the cost of draining and refilling systems would be from $10 million to $30 million, depending upon authorized PCB level, and based upon the model of an average 2,000 gallon system and the estimated 450 systems in the industry. For example, PCB reduction to a concentration level of 500 ppm in a 2,000 gallon system, with a current 1% PCB volume (10,000 ppm) and assuming a 90% fluid recovery, would result in a marginal removal cost of approximately $220 per pound. It is estimated that to reduce from a 500 ppm level to a 50 ppm level would require a marginal removal cost of about $30,000 per pound. Starting with a PCB content in the 2,000 gallon system of 240 pounds, draining the system (90% recovery) would leave 200 gallons of fluid containing 2 gallons of PCB's or 24 pounds. This quantity should, when the system is refilled with 2,000 gallons of flushing fluid, result in a PCB concentration of 1500 ppm. A -6 - second draining and refilling with flushing fluid should initially drop PCB's to 2.4 pounds or 150 ppm, but this amount would shortly increase as PCB's are released from the surface walls into the fluid. Once a proper concentration has been obtained to assure the concentration remains below S00 ppm, the flushing fluid would be drained and the system refilled with non-PCB heat transfer fluid. The direct cost estimates of flushing and draining, excluding costs associated with lost production and resulting disruption of supply to consumers, are as follows: (a) Flushing fluid - 2x2,000 gallons at $2.00 per gallon $8,000 (b) New fill of non-PCB heat transfer fluid - 2,000 gallons at $8.00per gallon $16,000 (c) Disposal of initial drained fluid and 2 flushing fluid changes - 6,000 gallons at $2.26 per gallon $16,000 (d) Labor andoverhead - 4 men10 to 15 days 512,000 Total $52,000 This estimate assumes that only two flushings are required and that draining and flushing proportionately reduces PCB concentration level without a major breakdown of the system involving gasket replacement or rewelding which could easily double or triple costs. Based upon this estimate, the cost of removal of 235 pounds would give a marginal removal cost of $220 per pound to achieve a residual level below 500 ppm. To achieve a level of 50 ppm it is likely that at least two further drain and flush procedures and chemical cleaning would be required. At $50,000 per procedure, the^final 1.6 pound removal would carry a marginal removal cost'of at least $30,000 per pound. .. The above information relative to environmental and health injury and economic impact is clear evidence which meets the two requirements necessary to support the grant of a use authorization for heat transfer systems containing residual PCB's. Accordingly, Monsanto recommends the proposed rules be revised to authorize for a period of 5 years the continued use and servicing of all 500 gallon capacity or less heat transfer systems used and serviced in a manner other than a totally enclosed manner to the extent these systems contain 1% or less by volume of PCB's, provided users attain*this 1% Act* C C 6 C 6 S A -7- PCB volume within one year after the effective date of the rule. For systems with greater than 500 gallon capacity, Monsanto recommends the EFA authorize for 5 years continued use and servicing of these systemsin a manner other than a totally enclosed manner to the extent they contain -a FCB concentration of 500 ppm or less, provided users attain this 500 ppm FCB concentration within 2 years after the effective date of the rule. These transition time periods would permit an orderly reduction of PCB levels during which disposal of fluids by incineration could be planned, thereby avoiding spillage and buildup of PCB contaminated fluid inventories awaiting disposal. Please note that proposed language for this use authorization is set forth as new Section 761.31(h) on pages 10-11 of these comments. Specific Suggested Changes in the Proposed Rules Sec. 761.2(w) "FCB mixture." In addition to the proposed arbitrary 50 ppm concentration, the definition includes any material, no matter how dilute in PCB concentration, if the material is contaminated by a substance containing a PCB concentration of 50 ppm or greater. For example, if a gallon of heat exchange fluid containing 500 ppm PCB were diluted to 1,000 gallons with a non-PCB flushing fluid (final concentration 0.5 ppm PCB), the resulting flushing fluid when drained would still be arbitrarily considered a PCB mixture. We recommend the definition be changed to read: (w) "PCB Mixture" means any combination of chemical substances which contains 500 ppm (0.050 percent on a dry weight basis) or greater of a PCB chemical substance. This definition includes, but is not limited to, dielectric fluid and contaminated solvents, oils, waste oils, heat transfer fluids, other chemicals, rags, soil, paints, debris, sludge, slurries, dredge spoils, and materials contaminated as a result of spills. Sec. 761.2(bb) "Manufacture ffor Commercial Purposes.'" We recommend for clarification the addition of the following subsection to the definition, which is consistent with the discussion by the EFA. set forth in column 1, 43 FR 24805: (3) As the desired product. Inadvertent manufacture as a by-product in the manufacture of another chemical is not "manufactured for commercial purposes." Sec. 761.2(cc) "PCB Sealant, Coating, or Dust Control Agent." This term is defined based on the state of the a c m ccscic -8- analytical art. In complex mixtures such as waste oils, analysis for PCB is not sensitive because of background interference. We recommend that a concentration level be set related to the toxicological properties of PCB's and the demonstrated injury risk associated with human and environ mental exposure. Such a concentration would, among other things, avoid a moving target definition depending upon the sophistication of the analytical equipment used. Sec. 761.2(dd) "Process 'for Commercial Purposes.'" We recommend the addition of the following subsection to the definition. This is consistent with the discussion in column 1, 43 FR 24805 and our proposed modification of section 761.2(bb). (3) for means other than PCB removal. Processing which removes PCB from the material to be used for commercial purposes does not constitute "processing for commercial purposes." Sec. 761.2{ff) "Significant Exposure." This term is defined based on the state of the analytical art. We believe that significant exposure should be, as discussed above relative to Section 761.2<cc), a specific concentration related to the toxicological properties of PCB's and to injury risk associated with human or environmental exposure. We recommend, consistent with the 1977 American Conference of Governmental and Industrial Hygenists publication of Threshold Limit Values for Chemical Substances in the Workplace, that this section be amended to read as follows: (ff) "Significant Human Exposure" means any exposure of humans to PCB chemical substances or PCB mixtures in excess of 0.5 mg/cu. meter TWA for a 40 hour week by an applicable analytical method. Sec. 761.1(gg) "Small Quantities for Research and Development." The definition of this term is unjustified and arbitrary. No data are presented in the preamble by the EFA that use in research has posed any measurable risk of injury to health or harm to the environment. Monsanto has no evidence of any adverse effect of PCB's in laboratory use. We suggest as an alternative replacing section 761.2(ff) with the following wording taken in part from the TOSCA Inventory Reporting Regulations, 42 FR 64576, section 710.2(y): (gg) "Small Quantities for Research and DevelQpment" means any quantity of PCB chemical substance or PCB mixture manufactured or processed only for purposes of scientific experimentation or analysis of chemical ach CGeC7l -9- research on, or analysis of, FCB's, including research or analysis for the development of a product that (1) are no greater than reasonably necessary for such purposes, and (2) are used by; or directly under the supervision of, a technically qualified individual(s ). It would also seem appropriate to include as new section 761.2(kk) the definition of "Technically Qualified Individual" based upon section 710.2(aa) of the TOSCA Inventory Reporting Regulations. Sec. 761.2(hh) "Totally Enclosed Manner." This term is defined with reference to section 761.2(ff) and, consistent with our recommended changes above to that section, we suggest section 761.2(hh) be changed to read: (hh) "Totally Enclosed Manner" means any manner that will ensure any exposure of human beings or the environ ment to FCB chemical substances will be insignificant. Sec. 761. 2 (jj ) "Applicable Analytical Method." We propose the following new definition of this term: (jj) "Applicable Analytical Method" means a method (other than the perchlorination procedure) which distin guishes FCB Chemical Substance as defined in section 761.2(t) from all other materials. The basis for exclusion of the perchlorination procedure is discussed more fully in the attached letter dated March 15, 1976 from J. Coleman Weber of Monsanto to Dr. I. E. Wallen of the EFA. Section 761(kk) "Technically Qualified Individual." We proposed a new definition for this term with the following wording taken from Section 710.2(aa) of the TOSCA Inventory Reporting Regulations, 42 FR 64576: (aa) "Technically Qualified Individual" means a person (1) who because of his education, training, or experience, or a combination of these factors, is capable of appreciating the health and environmental risks associated with the chemical substance which is . used under his supervision, (2) who is responsible for enforcing appropriated methods of conducting scientific experimentation, analysis, or chemical research in order to minimize such risks, and (3) who is responsible for the safety assessments and clearances related to the procurement, storage, use, and disposal of the chemical substance asvmay be appropriate or required X ACh CC6C7Z 10 within the scope of conducting the research and develop ment activity. The responsibilities in clause (3) of this paragraph may be delegated to another indiviudal, or other individuals, as long as each meets the criteria in clause (1) of this paragraph. Section 761.31(f) Hydraulic die casting systems-use. The same considerations that support a use authorization for hydraulic die casting systems apply to other industrial hydraulic systems as well. Therefore, we recommend that all references in this section to the words "hydraulic die casting system" be changed to "industrial hydraulic system." Section 761.31(h) Heat transfer systems - use. We recommend this new use authorization as follows: (h) Beat Transfer Sytems - Use. Heat transfer systems containing PCB mixtures may be used and serviced in a manner other than a totally enclosed manner until five years after the effective date of this rule subject to the following conditions: (1) One year after the effective date of this rule no heat transfer systems of 500 gallon capacity or less may contain fluid whose PCB mixture volume is greater than 1%. (2) Two years after the effective date of this rule no heat transfer systems of greater than 500 gallon capacity may contain fluid whose concentration of PCB mixture exceeds 500 ppm. (3) Ninety days after the effective date of this rule each person who owns a heat transfer system containing residual PCB's above the authorized levels shall report to EPA, and retain records of, the number of such systems he owns, the volume capacity of each such system, and the PCB volume or concentration, as appropriate, of the fluid contained in such systems. This report shall be sent to the Pesticides and Toxic Substances Enforcement Division (EN-342), Environmental Protection AGency, 401 M Street S.W., Washington, D. c. 20460. Each person who owns such a system shall also keep a current record of the dates of each draining or refilling and the measured PCB concentration or volume, as appropriate, of the fluid in the refilled systems on those dates for etch system. If any such system is ACM CQ6C73 U -11- sold, the transaction and the parties thereto shall he reported to EFA by the Seller. At its' discretion, EFA nay require the submission of a copy of a-person's current record. (4) Each report submitted to EFA under paragraph (h) (3) of this section shall contain the certifi cation found in 761.31(b)(2). (5) Each person who owns a heat transfer system containing residual FCB's above the authorized levels shall develop and implement a plan for the control of FCB exposures and contamination in accordance with Annex VII. Your favorable consideration of the above comments is respectfully requested. Sincerely, W. R. Corey Director, TOSCA Administration ceso1*' March 15, 1376 Dr. I. Z. 'allen 2r.vircrjr.er.tal Protection Agency Office of Toxic Substances 401 li Street S.W. Washington, DC 20460 Dear Dr. Wallen: Monsanto has a.concern over the validity of the oerchlori nation technique"used by SPA and others to measure and/or confirm polychlorinated biphenyls in er.vircnr.ental naceri la . In our investigation of the perchlorinaticn method, we have found that such chemicals as biphenyl, alkylated biphenyl', and many other substituted biphenyls, interfere with the perchlorir.ation technique. It also appears that various petroleum components may interfere. If these chemicals were present in environmental materials that were being tested for PC3*s using the perchlorination method, erroneously high PCS concentrations would be reported. A recent article in the Journal of the ACAC (Vcl.58, No.2, 1975) points out two other limitations of the perch 1er ir.acion procedure : 1. High and variable reagent blanks, which cause erroneously high findings. 2. Formation of bromoncnachlorobiphenyl, which cause low recoveries. v A copy of the article is' attached. i. ACh C08015 Or. I Wallen 2 /larch 13 197G Since these limitations can lead te significant errors in determining trace levels of PCS 's in environmental samples, we suggest that IPA carefully review the validity of the perchlortnaticn technique. results that have been obtained using this technique may not be valid. If any other information is needed, please let us hnev. Sincere!'', J. Coleman Weber .`Inniger, Product Acceptait nah cc: Dr. A. C. Traxowski Environrental Protection Agency CCC7t iOiC2: This material`na? he protected by copyright 466 JOOtXAL or t h e aoac (Vol. 5S, No. 3, 3?" Limitation on the Use of Antimony Pentachloride for Perchlorination of Polychlorinated Biphenyls WILLIAM J. TROTTER and SUSAN J. V. YOUNG Diviiion of Chemistry and Phyvcs, Food and Drug Administration, Washington, DC 10201 Two contaminant! ir* pment In rammer* dally available antimony pentachloride (SbQj) uaed to pcrdtlorinaie polychlorinated biphenyl (PCBs) to decaehlorobiphenyl (DO). DCB is found in the 5bQ* perchlorination reaction blank in which no P Q t were added. Brant nonaehlorobiphenrl (BNCB) Ij found after use of SbQ, to perchlorinate PCBi. Level of DCB found in the SbClj reaction blank from on* oui dUtributon ranted from 8 to 972 ng DCD/ ml SbCla.The relationhip of the formation of BNCB to amount of various PQ) Aroclors perehlorinated (iexamined. Polychlorinated biphenyl (PCBs) residues are extracted, cleaned up, and detected by methods similar to those used for organochlorine pesti cides. FCB residues are quantitatively deter mined by comparing the gas-liquid chromato graphic (GLC) response of the multicomponent residue and commercial PCBs (Arcelor*) or a mixture of Aroclors producing a GLC response pattern similar to that of the residue (1). This approach is limited because the multicomponent PCB residue may not have the same propor tional composition as the Aroclor or Aroclors used as the quantitation reference. Residues can be composed of mixtures of ehlorobiphenyl com ponents from more than 1 Aroclor. Metabolic and other environmental factors complicate the description of the PCB residue composition. There has been considerable work to develop methods to convert the multieomponept PCBs to a single derivative on which to base the resi due determination. Procedures have been re ported to catalytically dechloriaate PCBs with hydrogen over palladium or platinum to bi phenyl, eyclohexytbtnzrne, and birytlohexyl (2, 3). A principal disadvantage with that pro cedure is that the hydrocarbou product is determined with a GLC flame ionization detec tor, resulting in. low sensitivity. Attempts have been made to convert PCBs to the fully chlor inated deeachlorobiphenyl (DCB) (3-6). Ar mour (G) reported optimum conditions for per* chlorinating PCBs with antimony pentachloride (SbClj). The method provides a quaiitativ confirmatory procedure for PCB determination The GLC electron capture detector response : enhanced because total PCBs are manifested a s . single peak for DCB. In measuring the sirz. peak for DCB the analyst u cot faced w:* analytical judgments such as baseline correct:?' method of integration, or discrimination betwes: PCBs and non-PCB components. However, tt ! necessary to be aware that the various Aroc!:: give rise to different equivalents of DCB i-' and that the nonehlorinated biphenyl (also use u a fungicide) is perehlorinated by SbCl, : DCB. Nonetheless,' using the perchlorin.it;: derivatization can reinforce the residue vr.'-.: determined by measuring a muiticompor.rr. PCB residue. During attempts to apply the perchiorinr.r:f' derivatization in determining low residue 'eve of PCB and make use of the increased elect" capture response to DCB, 2 contaminants v.*e: indicated which led to erratic recoveries of DC- Experimental flMgeRU and Apparatus fa) Antimony pentachloride.--Hooker Chemin Niagara Falli, NY 14302 (received in piass hoc', with lead-lined cap); Mathtion Coleman S(M&S). Norwood. OH 453)2 (reaeeat grad-'B.\ (Allied Chemical). Morristown. NJ 077(reapeot grade, 99'e); Research Orgiiic-lnora::. Chemical (ROC-RIC). Belleville. NJ . 07: (MS9Cr) ; and J. T. Baker Chemical. Phillipsbur. NJ 0SS66 (Baker Analyzed Reagent). (b) Gas eAromai05ropA.--earle*Analytic (T Plaine. IL 600SS) Model 5360 wi'ti 6' x 1 mm : pla column containing 17 OV-101 on S0-1-* meih Chromoeorb W (IIP). Operating fondii:*r.i column flow, 60 ml nitrogen/min; column. 302*C detector, 202*C; injector, 225*C: pia<np (hi-r electron rapture detector with itanium 'll detector-voltaee (constant dc) adjusted m cm:one-half full *'.*ale recorder deflection for 0 7 : DCB when full scale deflection is I X IO am; (e) Mat eprefn t m t t r r Vari.in MAT i-3 Pi:* 22. Springfield. NJ 070SI) CH5*m; mw ip.-!-peter (MS) coupled to Varian Aerograph 2710 .* PLAINTIFF'S ; .^:;-(E X H i B n ; x ^ ' ACK CC6C71 A. J i T JIO T T D I * YOL' S 'C : TER C H L0R IN A T10X OF FC fl 467 ebrooatopapb via lU-flta system using WitscnBicmufi 3h u ; c Jipirmtor. GLC opentioy condi tions: 6 'x 4 mm id tikes column containing 37< OV-l oa SO-100 rcMh Chrosworb VT (HP); column Sow, 60 ml helium xia: column 2iO*C. MS opentiag conditions: electna energy, 70 ev; emission current, 300 #a; multiplier voltaye, 22 kv. Rctulu tad DitcuiMon A peak identical to that of DCB was found in the reaction blank tor the Armour perchlorination procedure (6) vr;h the described GLC operating condition*. The identification of DCB was confirmed by GLC-M5 of a hexane extract of a hydrolyzed ample of $h-G, which had not been subjected to the perchJonnation procedure. Various quantities (02-2.0 ml) of SbG. from the o commercial sources were examine'! to de termine the presence of DCB. SbG, alone was carried through the perchJonnation reaction (0) except that no CEC1 was present with SbG, in the ruction vessel. DCB was determined by GLC. Table 1 lilts the amounts of DCB found. .After perchloi mating PCBs with SbG:< a secondary peak with a GLC retention time rela tive to DCB of 1.31 was observed similar to that reported by Huekins 11 ol. (7). This later eluting peak is seen in fig. 1, the chromatogram (Torn he 02 ml SbG, (Hooker Chemical) perchlor'nation of 0.50 Aroclor 1221. This peak was found when SbG. from each supplier was used. The peak was determined by GLC-M5 to be due to bromononachlorobiphenvl (BXCB). BXCB was assumed to be a competing product with DCB arising from a small amount of SbG,Br in SbG,. so parameters relating to possible limitations of the perchlorination procedure were studied. Various quantities (0.5-10 pg) of Aroeior* 1221, 3242, 3254. and 32*0 in CHG, were perchlorinated. Recoveries of DCB and estimatee of the relative amounts of BXCB formed are given in Table 2. Calculation of the relative Table 1. DCS (n f /m l) >w i < frem vertove em euHtt ef U Q | tk O i. m l S4IO*Uf f l.I i.e l . l Av. Keeker C h fnic*l m cg *A ROC MIC l- T . i n n u 42 i t is <2 I I 0 MO 1W S I) 122 12 IS u 12 2 21 MIN FIO. ceature GLC curve tram me 0.2 ml IfeCJf (H H M r Cftam ical) eerefuertiiitiew e f OJO a AreeJer 1221: a s i n t auhriU ftt Arectar 1221 iflleete. - S u k l reg m en tt 0.11 n | DCft. Peak 2 ( ip m n a OJg ng tN C e. amounts of BXCB product formed was based on comparison of the electron capture GLC peak height of BXCB with that of a DCB reference. The amount of DCB determined in the reneiion blank was directly proportional to the amount of SbCl, used (Table 1). This indicates SbCI, was the source of the DCB and that contamination from other possible sources dur ing the perchlorination was negligible. The pro cedure for pereblorinaiinf PCBs specifies the use of 02 ml SbG,. SbG, producing S-972 ng SbO,/isl in the reaction blank would add 0.S65 ppb. based on a 3 g simple. DCB produced in the reaction blank w u as sumed to come from PC3 contamination of SbG,. In an effort to locate the origin of this contamination, SbG, bottle closures were investi gated. GLC analysis of bexane, in which tbe plastic caps were soaked for 4 days, did not reveal PCBs. Hooker Chemical, the sole d<v aesric source of 3bG,, supplied SbG, in glass bottle with lead-lined caps. This bulk supplier of SbG, indicated that the production of chlo rine in carbon anode half-cells with linseed oil or other organic binders forms certain organic compounds: however, the destructive oxidative environment in the electrolytic cells would make the production of PCB unlikely as a result of this pathway. On 'fee other hand, antimnnv metal is commonly obtained us a metallurgical by-prod uct by cartan reduction of its oxide; therefore. >CK CC6018 468 jockxal or TJtr aoac (Vol. SS, No. 3. iTM Takte 2. OC end BN CS Ira n p ereh lerln etle* a l various Areelere with 0.2 mi SbCl,* Am t, Arnelor I DCB re e d . % BNCB* DCS - BNCB* re e d . com bined % re c .% 1290 1214 1242 1221 1290 1214 1242 1221 1210 12*4 1242 1221 10 10 10 10 4 4 4 4 0.9 0.9 9.9 0.9 09 0 14 0 II 4 97 - 19 11 0 10 0 na m it 19 2 9 72 10 90 1 U 4 2 U 1 a at t 91 >4 92 79 Hooker Chem ical SbCl*. * Quantity calculated by com sariaon of electron eap* turo OLC reeponte ta BN C v s . reaponaa to DCS refer* nee standard. it u conceivable that PCBi could be associated with the antimony metal employed in the SbCl, process. No heat transfer systems containing FCBj are used in either the ehlorine or SbG, production facilities, and SbCl, does not come into contact with plastics in the manufacturing operation or in shipping containers (Hooker Chemical and Plastics Corp., 1974. private com munication) . Two parameters (various quantities and vari ous Aroclors) were studied in relationship to the production of BNCB as a competing prod uct of DCB during the perehlorination of PCBs. BNCB was calculated by comparison of the elec tron capture GLC response to BNCB vs. the response to DCB. Several factors are considered: (/) In this reaction bromination i< kinetically favored over chlorination. With perehiofiliation of lower amounts of PCBs the relative BNCB to DCB is greater because the bratr.:: ating agent is the limiting quantity tn cop.:s~ inated SbCl,. (5) Bromination occurs to a a : : degree for a given quantity of the less d o nated PCBs such as Aroclors 1221 and 2.rather than tor 1254 and 1260. This likely is 1. to a greater number of reactive sites and steric hindrance. (3) In the range of PC3s pc: chlorinated (0.5-10 /ig) in the above study, > likely that with lower amounts of PCBs nr.-, or less chlorinated Aroclors the decrease in DC recovery is principally due to the increase BNCB formed. One of the major advantages of perchlorir. tion in determining minute quantities of PC. is the inherent increase in effective GLC deb tor response. Contaminated SbCl,, as descri be, would preclude its use in many of the cases. Retocxcu (1) Offieiol Mtlhodt of Aneihjm (1073) 12th E AOAC, Washington, DC. sen. 29.0Cl-25.CC (2) Asai, R,, Gunther. R.. Westlake. W,, & lx-.: V. (1971) J . Aqt. Food Chem. 19, 3G6-G3S (3) Berg, O. W,, Diosady. P. L.. i Ree?. G. A. (1972) Buff. Environ. Contain. TosicoL 33W47 (4) Hutxioper. O. W., Safe. S., & Zitko, V. CiCT Int. J . Environ. Anal. Chem. 2. 05-106 (5) Hutxinger, O. IV, Jamieson. D., Safe, 5.. Zitko, V. .(1973) JAOAC 56. 9S3-9io (6) Armour, J. A. (1073) -MO.-iC 56. 9S7-90J (7) Huckins, J. X,, Swaaaon. J. E., & S ta le r " L. (1974) JAOAC 57, 416-417 Rmwl August SI.I0T4. I b i , p > m n i prvtmtrd et the SX'h Annual Veer..-.; the AOAC. Oct, 11-17, MTi. el W vh.ajton. DC. \ ACM CC8C7S -, S-J C Analytical Chemistry Method 73-9 Job Ho* 1048006 DETERMINATION OF POLYCHLORINATED BIPHENYLS IN THERMINOL 66 BY ELECTRON CAPTURE GAS CHROMATOGRAPHY SCOPE This procedure is intended for the determination of polychlorinated biphenyls (PCBs) in Therminol 66* PRINCIPLE The procedure is based on the fact that PCBs give a characteristic fingerprint chromatogram when analyzed by EC/GC. The Therminol 66, per se, does not interfere with the measurement but may contain some impurities with electron capture activity. This procedure measures electron capture impurities eluting at the same retention time as PCBs. If absolute confirmation is desired, GC/Mass Spectrometry techniques must be used. REAGENTS AND EQUIPMENT 1. Hexane - Fisher - "Nanograde", Cat. No. K-300. r 2. Usual laboratory glassware. 3. Hewlett-Packard 5750 Research Chromatograph equipped with a 63Ni electron capture detector (pulse - 50 y sec.). GAS CHROMATOGRAPHIC CONDITIONS Columns 2m X 1/4" 4% XE-60 on 60/100 Chromosorb W, H.P. Temperatures: Flow Ratees Injection Port - 190*C Column - 170#C Detector - 300*C Purge - 10% Argon/Methane 120 ml/min. Carrier - Helium - 60 ml/min. PROCEDURE 1. Prepare a calibration curve by injection 2; 4, 6 and 8 vl of a PCS (Aroclor 1242) solution in hexane containing about 0.5 ng/ulPlot Kg of standard vs the area of the PCB peaks on linear graph papaf. 2 Heigh 0.2g of sample to the nearest milligram into a 25 ml volu metric flask, dissolve ai^ dilute to volume with nanogrado hexane. Inject a suitable size aliquot of this solution into the c h r o m a t o graph (*v*5 ul) To insure accurate data, the sample c h r o m a t o g r a m must fall within the range of the calibration standards.* M e a s u r e the area of the PCB peaks and read the nanograms from the i calibration curve. Method 73-9 Page Mo. 2 CALCULATIONS ppm PCBs (as Arcclor 1242) (Nanograms from Curve) (Volume of sol'n, in ml) (Injection Volume, In ul) (Sample w t . , in gm) (Nanograms)(25) (injection Volume) (0.2) RECOVERY EXPERIMENTS A typical lot of Therminol 66 was analyzed by the above procedure and a chromatogram which was not typical of PCBs was obtained. Measure ment of the area where PCBs would elute gave a value of 17.1 ppm. This sample was spiked with 25, 50 and 75 ppm PCBs (Aroclor 1242) and the following data obtained* Therminol 66 + 25 ppm Aroclor 1242 Duplicate Therminol 6 6 + 5 0 ppm Aroclor 1242 Duplicate Therminol 6 6 + 7 5 ppm Aroclor 1242 35 - 17* m 18 43 - 17 26 59 - 14 45 64 - 14 50 00 1 71 f Background from PCB free Therminol 66 In practice, there will be no way to correct the background inter ference from the impurities in the Therminol 66 and so numbers below 50 ppm will probably be biased on the high side. Samples which give chromatograms not typical of PCBs should be reported as none detected, less than the calculated value. db Monsanto Industrial Chemicals Co. Applied -Sciences St. Louis, Mo* 10/73 - 0. Hicks, E. S. Tucker \ ADM 0C8G82 Analytical Chemistry Method 73-8 .Job No. 1048006 DETERMINATION OF POLYCHLORINATED BIPHENYLS IN THERMINOL 55 BY ELECTRON CAPTURE GA CHROMATOGRAPHY SCOPE This procedure is intended for the determination of polychlorinated biphenyls (PCBs) in Therminol 55. ,PRINCIPLE The procedure is based on the fact that PCBs give a characteristic fingerprint chromatogram when analyzed by EC/GC. The Therminol 55 does not interfere with the measurement. This method measures electron capture active materials eluting at the same retention time as PCBs. If absolute confirmation is desired, GC/Mass Spectrometry techniques must be used. REAGENTS AND EQUIPMENT 1. Hexane - Fisher - "Nanograde", Cat. No. H300. 2. Usual laboratory glassware. 3. Hewlett Packard 5750 Research Chromatograph equipped with a 63Ni electron capture detector (pulse - 50 y sec.). GAS CHROMATOGRAPHIC CONDITIONS Column: 2M X 1/4" 4% XE-60 on 80/100 Chromosorb W, H.P. Temperatures: Injection Port - 190* Column - 170* Detector - 300* Flow Rates: Purge 10% Argon/Methane .120 ml/min. Carrier - Helium - 60 ml/min. PROCEDURE 1. Prepare a calibration curve by Injecting 2, 4, 6, and 8 of a PCB (Aroclor 1242) solution in hexane containing about 1 ng/uiPlot ng of standard vs the area of the PCB peaks on linear graph paper. 2. Weigh 0.2g of sample to the nearest milligram into a 10 ml volumetric flask, dissolve and dilute to volume with nanograde hexane. 3. Inject a suitable size aliquot of this solution into the chromato graph (^5ul) To insure Accurate data, the sample chromatogram must fall within the range'of the calibration standards. Measure the area of the PCB peaks and read the nanograms from the calibration curve. ADM G08G63 t CALCULATIONS ppm FCBs (as Aroclor 1242) ; Method 73-8 Page No. 2 (Nanograms from curve)(Volume of sol'n, in ml) (Injection volume, in yi)(Sample weight, in gm) (Nanograms)(10) (Injection volume)(0.2) RECOVERY! 0.2g Samples of Therminol 55 were spiked in duplicate with 25, 50 and 75 ppm of Aroclor 1242 and 93.4 + 5.3% recovery was obtained. db Monsanto Industrial Chemicals Co. Applied Sciences St. Louis, Mo. 10/73 - 0. Hicks, E. S. Tucker a 0h ccec8<t ANALYTICAL TEST METHODS Nothin contained herein Is to be construed as a recommendation to use any product In conflict with any patent. MONSANTO MAKES NO WARRANTIES AS TO THE MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OF ANY PRODUCTS REFERRED TO, no guarantee of satisfactory results from reliance upon information or recommend atlons contained herein and disclaims all liability for any resulting loss or damage, whether a claim 13 based In contract, negligence, strict liability or otherwise. ADM CC6C8C VC. 5ssStf-` .r$iV\ - _,, i*.-. fSffl:-''* *-.- SnW w vt -`is n ro v w ,. : . <5^^^^-*' vnr^+T: ****v** BURNING WASTE CHLORINATED HYDROCARBONS IN A CEMENT KILN bY L .P . MacDonald (S t . Lawrence Cement Co.) D .J. Skinner (Environment Canada) F . J . Hopton and G.H. Thomas (Ontario Research Foundation) for the Petroleum and In d u stria l Organic Chemicals D iv isio n Water P o llu tio n Control D irecto rate Environmental Protection Service FISHERIES AND ENVIRONMENT CANADA Report No. EPS A-WP-77-2 March 1377 REVIEW NOTICE Th is report hes been reviewed by the Water P o llu tio n Control D Ire cto ra te i Environmental P ro tectio n S e r v ic e , and approved fo r p u b lica t I o n ._ Approval does not n e c e ssa rily s ig n ify that the contents r e fle c t the views and p o lic ie s of the Environmental P ro tectio n S e rv ic e . Mention of trade names or commercial products does not co n stitu te endorsement for use. \ M in iste r o f Supply bnd S e rv ice s Canada 1977 Cat. No. EnA3-V77-2 ISBN 0 - f 62-00560-0 Canadian P r in t co L td . M o n tre a lQue. I ABSTRACT *. / An experimental program was ca rrie d out in 1975/76 at the S t. Lawrence Cement C o ., M ississau g a, Ontario in which waste chlo rinated hydro carbons, containing up to about 46 weight percent c h lo rin e , were burned in a ro ta ry cement k i l n . The ch lo rin ated hydrocarbons were burned in three d is t in c t phases o f in cre asin g d i f f i c u l t y o f combustion. M aterials burned Included m ixtures o f ethylene d ic h lo rld e , chlorotoluene and up to approxima- > t e ly 50 percent p o ly ch lo rin ated biphenyls (PCB). These m a te ria ls were destroyed in the cement k iln with a t le a s t 99.98 percent e ffic ie n c y in a ll cases. Emissions of high molecular weight chlo rinated hydrocarbons were not detected. Three lig h t chlo rinated hydro carbons, dlchlorom ethane, chloroform and carbon t e t ra c h lo rid e , were found In the em issions in the p art per b illio n or lower range. The q uantity of p recip itato r dust requiring d isp osal, as well as emissions of p a rticu late m atter. Increased during the t e s t. ~~ The c h lo rin e 'in p u t from the ch lo rin ated hydrocarbon waste was up to about 0 . 8 weight percent r e la t iv e to c lin k e r and t h is e f f e c t iv e ly reduced the a lk a li concentration of the c lin k e r in d ire c t sto ich io m etric proportion. A reduction in f o s s il f u e ls used w hile burning ch lo rin ated hydrocarbons was noted. \ \ ACM C C 8 C 9 C II RESUME / Au cours d'une s r ie d 'exp riences r a lis e s en. 1975*1976 la S t . Lawrence Cernent C o., i M issfssauga (O n tario ), on a b r l , dans un four r o t a t if cim ent, des hydrocarbures contenant au poids ju sq u ' 46 p. 100 de chlo re l i . La combustion s 'e s t f a it e en t r o is phases de d i f f i c u l t c r o is s a n t e . Les substances consumes comprenaent des mlanges de d ich lo ru re d 'th ylne et de chlorotolune a in s i que d 'au tre s atteig nant prs de 50 p. 100 de biphnyles p o ly ch lo rs (BPC). Dans tous le s c a s / la combustion a eu un rendement d'au moins 99,98 p. 100. On n'a dcel aucune m ission d'hydrocarbures ch lo r s poids m o l cu laire le v . Par co n tre , on a id e n t if i t r o is hydrocarbures chlo rs l g e rs, s o it le dichloromthane, le chloroforme et le ttrach lo ru re . de carbone, en con cen tration s In f rie u re s une p a rtie par m illia r d . La quantit de p o u ssires accumules dans le s p r c ip ta te u rs et q u 'il f a l l a i t lim in e r, a in s i que le s m issions de p a r t ic u le s , ont augment au cours des ex p rie n ces. La portion de ch lo re provenant des dchets d'hydrocarbures ch lo rs a a t t e in t environ 0 , 8 p. 100 du poids des c lin k e r s , ce qui a r d u it, en raiso n stoechiomtriquement d ir e c t e , la teneur en a lc a l i de ces d e rn ie rs. On a galement remarqu une consommation rduite de fu e ls f o s s ile s pendant la combustion des hydrocarbures ch lo r s. AO* C B C S 1 m TABLE OF CONTENTS ABSTRACT *. / TABLE OF CONTENTS L is t of Figures L is t of Tables EXECUTIVE OUTLINE SUHHARY RECOMMENDATIONS LIST OF ABBREVIATIONS 1 . INTRODUCTION 2. 2.1 2.2 2.3 2.3.1 2 .3 .2 CEMENT MANUFACTURE General P rin cip le s E ffe c t of -Alkat ie s The S t. Lawrence Cement Co. Relevant u n it processes - wet process k iln Relevant unit processes - suspension preheater k iln 3. CONSIDERATIONS AT THE PROGRAM PUNNING STAGE A. TRIAL ON THE SUSPENSION PREHEATER KILN 5. 5.1 5.2 5-3 5-3.1 5 .3 .2 5*3*2.1 5 .3>2.2 5 .3 .2 .3 5 .3 .2 .A 5 .3 .3 5-A 5 .A.1 TRIAL ON THE VET PROCESS KILN D iscu ssio n Organic C hloride Waste Burned Em issions Free c h lao rin e and hydrogen chtorfde Gaseous organic compounds Desorbed samples Organic so lven ts\extracted samples R esu lts obtained by the p a rtic ip a tin g la b o ra to rie s Estim ated minimum combustion e f f ic ie n c ie s P a rticu la te emissions Hass Balance on Wet K iln S ig n ific a n c e of the mass balance ADP CCfcCSi. Page i i i V vi i xi xi i i XV xv i 1 2 2 5 .6 6 6 10 12 5 15 17 22 22 22 23 25 2$ 27 28 "30 30 iV TABLE OF CONTENTS (CONT'D) 5 . ^.2 Chlorine and potassium retain ed 6 . CONSIDERATIONS ON BURNING CHLORINATED HYDROCARBON WASTES IN A CEMENT KILN 6.1 E ffe c t on Production 6 .2 A lk a li Reduction While Burning Chlorinated Hydrocarbon Wastes 6 .3 Heat Recovery from Chlorinated HydrocarbonWastes 6 .4 Cement Q u ality 6 .5 Extrap olatio n to Other K iln Types 6 . 6 Comparison o f Cement K iln Burning w ith Other Uses and Disposal Methods fo r Waste Chlorinated Hydrocarbons 7. CONCLUSIONS REFERENCES . . . ACKNOWLEDGEMENTS APPENDIX A - QUANTIFYING, SAMPLING AND ANALYSIS OF PROCESS MATERIALS APPENDIX B - ANALYTICAL DATA, CALCULATION AND DETAILS OF EXPERIMENT ON THE SUSPENSION PREHEATER KILN APPENDIX C - RESULTS AND CALCULATIONS FOR WET PROCESS SYSTEM APPENDIX D - QUALITY OF CEMENT PRODUCED ^ - APPENDIX E - EQUIPMENT DESCRIPTION AND ASSOCIATED ECONOMICS APPENDIX F - ONTARIO MINISTRY OF THE ENVIRONMENT EMISSION GUIDELINES AND ANALYTICAL SUPPORT APPENDIX G - LABORATORY ANALYSIS RESULTS FROM THE ST. LAWRENCE CEMENT FACILITY TEST (TRW Systems Group) APPENDIX H - DEVELOPMENT, CONSTRUCTION AND EVALUATION OF A . COLLECTION SYSTEM FOR LOW MOLECULAR WEIGHT HYDROCARBONS (Onta\rio Research Foundation) APPENDIX I - GC/MS/COHPUTER DETERMINATION OF CHLORINATED HYDROCARBONS AND PCB's (A ir P o llu tio n Control D ire c to ra te , EPS, Environment Canada) 31 34 3^ 3 ii 37 38 38 35 42 43 46 49 93 107 129 137 149 165 207 221 CM GCSC53 LIST OF FIGURES Figure I - Wet Process K iln . 2 Dry Process K iln 3 P r in c ip le o f Fuller-Humboldt Suspension Preheater and By-pass 4 A lk ali By-pass A .1 Schematic of the M aterial Balance A.2 Schematic of Port Locations A .3 Gas Flow D istrib u tio n a t Sampling Points A.A Grab Bag Sampling Equipment A .5 P a rtic u la te Sampling Train A. 6 Gaseous Sampling T ra in A .7 Gas Chromatographic P r o f ile from Flame Io n izatio n Detector fo r Chlorinated A lip h a tic s (WBA) Sample Feed A. 8 Gas Chromatographic P r o f ile from Flame Io n iza tio n Detector for Chlorinated A lip h a tics plus Aromatics and A l i c y c l i c s (WBB) Sample Feed A .9 Gas Chromatographic P r o f ile from Flame Io nization Detector for Chlorinated A lip h atics plus Aromatics. A li c y c l i c s and P o lychlo rin ated Biphenyls (WBC) Sample Feed A .10 Gas Chromatographic P r o f ile from Electro n Capture Detector fo r WBB Sample Feed A .11 Gas Chromatographic P r o f ile s from Electro n Capture Detector fo r WBC Sample Feed and Standard Arocolor 1242 A. 12 Gas Chromatographic P r o f ile s from Electro n Capture Detector fo r Low Molecular Weight Chlorinated Hydro carbons and for BLB and WBC Test Samples A .13 A .14 Gas Chromatographic P r o f ile s from Flame Io nization Detector fo r Low M olecular Weight Chlorinated Hydrocarbons Gas Chromatographic P r o f ile from Electro n Capture Detector fo r Impinger E x tra ct from BLA Test 3 CC6CS ACM VI LIST OF FIGURES (CONT'O) figure A. 15 / w Gas Chromatographic P r o f ile from Electro n Capture Detector for Impinger E x tra ctfromWBC T e s t .3 A . 16 ,-- -- - Gas Chromatographic P r o f ile from Electro n Capture Detector foi* Impinger E x tra ct from WBC Test 3 a f t e r Cleanup and Separation B . 1 C hlorine Level In Stage IV, June 3, 1975 ft.2 C hlorine Level in Stage IV, June 10, 1975 E . l Schematic Diagram of B a sic Chlorinated Hydrocarbon Feed System E . 2 Schematic Diagram of Chlorinated Hydrocarbon F a c i l i t i e s F. 1 F .2 R epresentative Bar Chromatogram Computer Reconstructed Bar Chromatograms fo r PCB Fuel and Aromatic Fuel plus Arocolor 1242 F .3 G. 1 G .2 G. 3 Gas Chronatogrem from GC/HS A n a ly sis o f Sample PCB Fuel TRW Sample Coding System Plan for the Combination o f ORF Solvent E x tra c ts Desorption System fo r Chromosorb 102 Tubes (TRW) H . 1 ORF T est Duct Schematic ' Page 82 83 102 103 T40 1A1 l 56 15 8 159 169 171 171 212 AC# CC6C55 vi 1 Table 1 2 3 4 5 6 7 8 9 10 11 12 A. 1 A .2 A .3 A .4 A .5 A .6 A .7 A .8 A .9 A . 10 A .n LIST OF TABLES / Composition of A lip h a t ic s (WBA) Composition of Aromatics plus Complex (VBB) Composition of Aromatics p lus PCB's (WBC) Gas Sample Volumes and Sample Concentration Facto rs Estimated K iln Em ission Concentrations (GC-EC) for S p e c if ic V o la t ile Organochlorine Compounds Estimated Minimum Combustion E f f ic ie n c ie s for Each Waste Burn Summary of P a r t ic u la t e Test Data Accumulated Mass Balance fo r C hlorine Accumulated Mass Balance fo r K?0 Average Reduction In K^O Content of C lin k e r Average Dust Discharged Recovery of Btu from Chlorinated Hydrocarbons Process M a teria ls Studied and Approximate Normal Production Q uantities Q u a n titie s of A lip h a tic Mixture Burned D aily Q u a n titie s of Aromatic plus Complex M ixtures Burned D a lly Q u a n titie s of PCS Mixture Burned D aily GC A n aly sts - System A GC A n a ly sis " System B GC A n a ly sis - System C Gravim etric Oust Analyses \ R e su lts from Leco Induction Furnace Analyses R esu lts from Atomic Absorption Analyses Least Squares Data fo r C a lib ra tio n Lin es - Page 19 20 21 Ik 25 28 29 32 32 35 36 38 kS 5k 55 57 69 69 70 85 86 87 88 ACH C C C 5 c c v iii LIST OF TABLES (COtfPO) Table */ B .I Percent Bypass Gas Required to Maintain C hlorid e Levels B .2 R e su lts from Analyses o f Dry Process K iln Raw Feeds B-3 _____ R e su lts from A nalyses of Dry Process K iln C lin k e r ' B. 4 R esu lts from A nalyses o f Stage IV Ousts 95 98 99 100 C. l R esu lts from Analyses o f C lin k e r Samples 109 C .2 R e su lts from Analyses of S lu rry Feed Samples 111 C.3 R e su lts from A nalyses of D iscard Dust 113 C .4 R esu lts from A nalyses o f Return Dust 115 X .5 Btu and C h lo rin e. Content o f Chlorinated Hydrocarbons 117 C.6 Btu, S and Cl A nalyses from No. 6 Fuel O il 118 C.7 D a ily Record o f Production and M a teria ls Consumption 120 L C.8 M aterial Balance for Chlorine 122 C. 9 D. 1 D.2 M aterial Balance fo r K^O Cements from C lin k e r Produced During B aselin e Burn Cements from' C lin k e r Produced During Aromatic p lu s Complex C hlorinated Hydrocarbon Burn ,, 125 131 132 D.3 Cements from C lin k e r Produced During P olychlorinated Biphenyl Burn F.1 HOE S p e c if ic a t io n s Applied to WasteBurns 133 150 F .2 S t . Lawrence Cement Waste Burn Experiments - Test 1 Chromosorb Adsorption Tube A n a ly sis a F .3 S t . Lawrence Cement te s te Burn Experiments - T est 2 Chromosorb Adsorption A n a ly sis 153 153 F.4 F.5 F .6 S t . Lawrence Cement t e s t e Burn Experiments - Test 3 Chromosorb Adsorption A n a ly sis Fuel Sample Id e n t if ic a t io n 6as Chromatograph Conditions 153 55 155 CM C C c C 5 7 Table F.7 F.8 G.l G.2 G.3 G.J* C.5 G.6 G.7 G.8 G.9 G. 10 G.11 G. 12 C. 13 G. 14 G. 15 G. 16 G-17 G. 18 LIST OF TABLES (COHT'D) /* y 'MS Data from GC A n a ly sis o f Sample PCB Fuel Area Counts Summary o f Samples Received from Each Waste Burn (TRW) Organic Composition o f Aromatic Waste by GC/MS Trace Metals In the C hlorinated Aromatic Waste by SSMS Organic Composition of PCI Waste by GC/MS Trace Metals In the PCB Waste by SSMS Summary o f Organic Q u a lita tiv e Survey A n aly sis o f Sample E x tra cts (TRW) Approximate Constituent Levels o f Trace Vapours Desorbed from Sorbent Tube Samples by LRMS Summary of the In te rp re ta tio n o f LRM5 Spectra fo r Trace Vapours Desorbed from Sorbent Tube Samples R esu lts and Detection Lim its from GC/FID A n a ly sis of the Concentrated Extracts R esu lts and Detection L im its from GC/ECD A n a ly sis of the Concentrated Extracts R esu lts and Detection L im its from GC/ECD A n a ly sis of the Unconcentrated Extracts Trace Metal Sem i-Q uantitative Survey o f F i l t e r Digests by 1CP0ES Lim its of Detection fo r Undetected Elements by ICPOES Concentration of Trace Metals In E fflu e n t Gas P a r t ic u la t e Matter by AAS Concentration o f Trace Metals In Aqueous Samples by AAS R esu lts o f Organic Survey A n a ly sis on C lin k e r Product.and Disposal Dust Saniples R esu lts and Detection Lim its from GC/FID A n a ly sis R e su lts and Detection Lim its from GC/ECD A n a ly sis Pa^e ISO 161 168 178 179 181 182 185 188 189 190 192 19* 195 196 197 197 199 201 201 ACM CC6CS8 X Table 6.19 LI ST OF TABLES fCONT'D) / Selected Trace Metals In SLC C lin k e r Product and D iscard Dust Semples by SSMS Page 202 v c c e c s 'i ACM xi executive outline: - Experiments using ch lo rin a te d hydrocarbon wastes In the manufa ctu re of cement appear to have transformed a d i f f i c u l t waste disposal problem Into a so lu tio n which is not only econom ically and environm entally s a t is f a c t o r y , but .has a b e n e fic ia l e f f e c t on the q u a lity of the cement. Safe d isp o sal o f the large volumes of the ch lo rin ate d hydrocarbon wastes which a re generated In Canada each year is a d i f f i c u l t environmental problem. It i s estimated co n se rv a tiv e ly that Canada generates some 25-30 m illio n pounds a n n u ally of these h ig h ly to x ic and p e rs is te n t w astes. T h is fig u re does not include many to x ic compounds such as in s e c tic id e s and polychlorinated biphenyls (PCB's). Environm entally sa fe d isp o sal is d i f f i c u l t and expensive. Disposal on land, or underground, req u ires sp e cia l and expensive precautions to prevant leaching Into waterways. The favoured method o f d is p o s a l, i n c i neration w ith recovery o f hydrogen c h lo rid e , can be very c o s t ly . Without reco very, scrubbing equipment Is required to remove hydrogen c h lo rid e from the products o f combustion to control it s em ission. T h is, in turn, u su a lly n e c e ssita te s a s a t is f a c t o r y d isposal method for the scrubber liq u id . Va luable fuel must be burned to m aintain combustion w h ile in cin e ra tin g ch lo rin a te d hydrocarbon w astes, as extremely high temperatures with long residence tim es.are required for th e ir thermal d estru ctio n . A long high-tem perature flame is required in a cement k iln to achieve the d esired product q u a lit y . During normal O p eratio n s, the thermal co n d itio n s that a re necessary e f f e c t iv e ly consume the .to x ic m a te ria ls.. The k iln a ls o co n tain s a co nsid erab le q u an tity o f time and thus has an in ten sive llscrubbIng,, a ctio n . Recovery and re-use of hydrocarbon wastes is not always f e l t to be economic. TherecyclIng o f so lve n ts is p ra ctice d when i t is econom ically and t e c h n ic a lly f s a s ib le . However, t h is e n t a ils the control of segregation, sto rag e, c o lle c t io n and ultim ate treatment of the vario u s hydrocarbon streams The report stro n g ly urges that combustible liq u id w astes, many of which a re p e r s is ta n t environmental contaminants, be combined by means of a ACP CCE IC C xi i s in g le economical .recovery system which would c o lle c t end la te r use the , I' wastes a s supplementary fuel for cement manufacture. | . " /' i-- - T h e experiments described In t h is report were monitored by \ the Department of F is h e r ie s and Environment, the O ntario M in istry of Ithe Environment, the United Sta te s Environmental P ro tectio n Agency, and the Ontario Research Foundation. For these experim ents, in d u stria l chlo rin ated hydrocarbon wastes including p o ly ch lo rin ated biphenyls (PCB's) were burned during the commercial manufacture of cement. Th is not only u t iliz e d the thermal value of the w astes, with a reduction in o i l consumption, but the r e s u lt s showed that almost a l l |the to x ic wastes were completely destroyed in the k iln . Em issions of to x ic m a te ria ls Into the atmosphere were n e g lig ib le . j Calcium ch lo rid e is o ften used In cement manufacture to reduce the a lk a l i content o f the product. The only apparent e f f e c t s that the experimental burn had on the q u a lit y of the cement were the b e n e fic ia l e f f e c t s due to the in cid e n ta l a d d itio n o f ch lo rid e ton. i i ACf* C C 6 L C 1 Jtu I SUMMARY Chlorinated hydrocarbon wastes were burned in a c a r e f u lly con t r o l l e d experimental t r i a l as a p a r t ia l fuel a t the S t . Lawrence Cement Co., M i s s i s s a u g a , O ntario. The experiment was conducted to determine whether chlo rinated hydrocarbon wastes could be burned In a cement k iln without adverse e f f e c t s on a i r p o llu tio n le v e ls . The wastes used Included a v a rie ty of ch lo rin ated hydrocarbons in the s e r ie s of program phases designed to progress from e a s ily combusted c h l o r i n a t e d hydrocarbons to those which are combusted only w ith d i f f i c u l t y . The la s t phase co n sisted mainly of polychlorinated biphenyl w astes. These m aterials were processed and formulated from chemical wastes as required for the d iffe re n t phases and supplied by Chemtrol P o llu tio n S e rv ic e s Inc. Atmospheric em ission measurements were made before, during and. a fte r the burning of each blend of ch lo rin ated w aste. Two methods of emis sions sampling were used during each phase, the method normally used for measuring em issions of p a rt ic u la t e m atter, and a sampling t ra in designed sp e c ia lly for determining em issions of organic m a te ria l. A ll samples from both systems were analyzed fo r unburned ch lo rin a te d hydrocarbons. It was concluded that the combustion e f f ic ie n c y was at le a st 99.986 screen : for the ch lo rin ate d hydrocarbons. Approximately 50 ppb of v o la t ile low m olecular weight compounds were found In the em ission samples. There were no d etectab le q u a n titie s of high m olecular weight c h lo rin a te d compounds in the stack g ases. A mass balance was c a rr ie d out on ch lo rin e and potassium. Th is showed that the ch lo rin e input as ch lo rin ated hydrocarbon was completely reacted with the process s o lid s . The a lk a li content of the c lin k e r showed a reduction which corresponded e x actly with the quantity of ch lo rin e input to the system. This agreement fu rth e r confirms the data from em ission te stin g and the mass balance. _ While burning ch lo rin a te d hydrocarbons w ith approximately liO percent c h lo rin e , a. decrease In o il consumption eq uivalent to 65 percent of the' Btu content of the ch lo rin ate d hydrocarbons was obtained. > The only d iffe re n ce s in the q u a lity o f c lin k e r produced w h ile burning ch lo rin ated hydrocarbons were the b e n e fic ia l e f f e c t s which were expected through the reduction in a lk a li content. ACM CCc 1C xiv I t mbs concluded that ch lo rin ate d hydrocarbon wastes may be used In cement k i l n s , rep lacin g other forms of ch lo rin e used for reduction of a lk a li content. A small proportion of f o s s il fuel required fo r cement manufacture Is conserved through use of these m a te ria ls. Burning ch lo rin a te d hydrocarbon wastes Is considered a valu ab le means of d estro ying p e r s is te n t and to x ic forms of p o llu ta n ts w h ile recovering useful .h e a t-v a lu e s. ACM CCS 1C3 XV RECOMMENDATIONS The experiments have shown that there is v i r t u a l l y no adverse e ffe c t on a i r p o llu tio n le v e ls by burning chlo rinated hydrocarbon wastes In a cement k i ln . These wastes Include polychlorinated biphenyls and other m a te ria ls which arcs d i f f i c u l t to d estro y. When other methods, such as in c in e ra tio n , a re used hydrogen ch lo rid e and ch lo rin e may be emitted and, i f the in c in e ra to r operation Is poorly c o n tro lle d , uncombusted m aterial may be released Into the environment. The follow ing recommendations are the outcome of the present report: (1) Burning ch lo rin ate d hydrocarbon wastes in a cement k iln is considered a valuable means o f d estru ctio n o f p e rs is te n t and to x ic m a te ria ls which a re members o f t h is fam ily of compounds. Since flame temperatures and reten tio n times a re s im ila r fo r a l l cement p lan ts reg ard less o f type o f k iln or fuel used, ch lo rin a te d hydrocarbons should be destroyed In any cement k i ln . The f e a s i b i l i t y o f doing so in a p a r t ic u la r k iln in s t a lla t io n can be determined by a tech n ica l and economic review . _ (2) Due to lack p f f a m ilia r it y w ith organic chem icals, I t is considered e s s e n tia l that in stru c tio n s on safe handling procedures be given to cement ind ustry personnel.r (3) Since problems such as p re c ip ita tio n , s o lid if ic a t io n , heat or gas release can a r is e through mixing inconpatlble waste m aterials in storage tank, i t i s considered a d v isa b le to obtain such m a te ria ls from one r e lia b le source o f supply a t any given time. v ACM CCblC't XV i f APCD _ LIST OF ABBREVIATIONS "/ Atr P o llu tio n Control D irectorate (EPS) 6LA B a se lin e A, designates em ission te s ts before waste burns BLB B a s ilIn e B, d esignates em ission te sts a f t e r waste burns DCH Dichloromethane EC E le ctro n capture (d etecto r) EPA Environmental P ro tectio n Agency (U .S .) EPS Environmental P ro tectio n S e rv ice (F is h e r ie s and Environment Canada) FIO Flame Io n iza tio n d etector GC Gas chromatography HOE (O ntario) M in istry o f the Environment HS Mass spectrometry ORF O ntario Research Foundation PCB P o lych lo rin ated biphenyl SLC St.'La w ren ce Cement Co. TRW TRW Systems Group T l , T2r T3 VBA T e st one, T est two, Test three Waste Burn A (ch lo rin ated a llp h a t ic s ) WBB Waste Burn 8 (WBA plus ch lo rin a te d arom atics and a l i c y c l i c s ) WBC Waste Burn C (WBB p lu s po lychlo rinated biphenyls) XRF X-ray flu o rescen ce \\ CM CC61C5 I 1 INTRODUCTION In Canada each year an estim ated 25*30 m illio n pounds of c h l o r i nated hydrocarbon wastes req uire disposal or d estru ctio n , 17-20 m illio n pounds being generated in Ontario [ l ] . These fig u res a re based on 5-6$ o f annual production and may be co n se rv a tiv e . Experience In Europe In d icate s that 1 0 t'6 f production Is a more r e lia b le estim ate of waste m aterial. There are waste streams from p la n ts manufacturing, processing , or using ch lo rin a te d hydrocarbons which must be disposed o f , the (netted of disposal frequently being in c in e ra tio n . Chlorinated wastes other than those d ir e c t ly from chemical p la n ts a ls o present a serio u s d isposal problem. Among these la t t e r chem icals a re p olych lo rin ated biphenyls (PCB's) and in s e c t ic id e s which may req u ire d is p o s a l. Many of these wastes a re to x ic and p e r s is te n t, and a l l pose a se rio u s d isp o sal problem. Among the methods p re se n tly used are In cin e ra tio n and land d is p o s a l. In cin e ra tio n w ith recovery of hydrogen c h lo rid e can be c o s t ly . Without reco ve ry, combustion gases must be scrubbed, thereby generating a liq u id waste req u irin g d is p o s a l. Both In cin e ra tio n me thods req u ire a d d itio n a l f u e l. Deep w e llin g and other s im ila r methods .of disposal a re environm entally unsound because of the r i s k o f water contamina tion. In cement manufacture, the k iln operates a t higher temperatures and for longer residence times than those used in In cin e ra to rs fo r des tru c tio n of these waste m a te ria ls . I t Is a ls o common p ra c tic e in the" cement Industry to add c h lo rid e s to the k iln to reduce the a lk a li concen tra tio n of the f in a l product. Use of chlorinated hydrocarbon wastes in a cement k iln would provide usefu l recovery of c h lo rin e and energy and, .a t the same t in e , so lve a se rio u s d isposal problem. _ The present research program was c a rrie d out to determine whether waste ch lo rin a te d hydrocarbons can be burned In a ro tary cement k iln without causing adverse a i r p o llu tio n . The approach taken was to analyse sta ck om issions fo t uncombusted ch lo rin ate d hydrocarbon. A mate r i a l balance on ch lo rin e was undertaken to confine the am ission fin d in g s. ACM c c e ic t V. / 2 ... CEMENT MANUFACTURE 2.1 General P rin c ip le s / While a v a r ie ty of raw m a te ria ls may be used In cement manufac t u r e , m a teria ls containing calcium , s i l ic o n , aluminum and Iron without an excess of c e rta in other elements are required. These m aterials are ground~to a fin e powder c a lle d raw meal, the chemical composition of which is c a r e f u lly co n tro lle d by proper blending of the v a rio u s m a te ria ls. Nor m a lly , blending is achieved by grinding a l l the raw m a te ria ls together (ih t e r g rin d in g ). Raw meals required fo r wet and dry processes a re s im ila r except that the raw meal fo r the wet process is In the form of a s lu r r y co n tain in g approxim ately 35% w ater, w h ile raw meal fo r the dry process co ntains le s s than 0.5% w ater. The raw meal i s fed into the k iln (see Figures 1 and 2) and is burned in the k iln to produce an interm ediate product c a lle d c lin k e r . The k iln slopes towards the burning zone and ro ta te s slo w ly , causing the raw m aterial to g ra d u ally move Into the burning zone. Reactions which occur during gradual heating In the k iln are: evaporation of free w ater, evo lu tio n o f combined w ater, evolutio n of carbon dioxide from carb o n ates, and combination o f lime with s i l i c a , alumina and iron to form the d esired compounds In the c lin k e r.- These reactio n s req uire a fin a l m a terial temperature of 1450C (2650F). Four main compounds a re present In Portland cement c lin k e r : Name Of Compound Chemical Formula Common A bbreviations Used In The Cement Industry Tricalcium S ilic a te Dicalclum S I I icate T r lc a lc iu D AIurnInate , Tetracalctum Alum lnoferrlte 3 Ca0'S102 2 Ca0"Si02 3 Ca0*Al203 4 CaO-Al203*Fe203 C3S C2S C3A - C^AF Minor compounds a re f l s o formed In c lin k e r , cannonly magnesia (MgO), p o ta sstin s u lf a t e (K^SO^) and sodium s u lfa t e (Na2S0^). T races o f other elements present In e ith e r the raw m a te ria ls or fu e l a re a ls o found In c lin k e r . Upon leaving the k i ln , the c lin k e r is ACK > 2-SLURRY FEED 7 - CLINKER COOLER 4 - PRECIPITATOR DUST SCREW 8- CLINKER 3 - PRECIPITATOR 9- FILTE R 5 - DUST RETURN CCtlC WET PROCESS KILN FIGURE 1. f yyy / k-, o ^1900 jfr, l^ijj M' >/~t" *- * < DRY feI >.** J Y iT 1 raw m ea l feed 2 STAGE I. 3 STAGE II. L STA G E 111. 5 STAGE IV. 6 KILN 7 CLINKER COOLER 6 CLINKER 9 FUEL A FILTER . B PRECIPITATOR C BY-PASS PROCESS KILN F I G U R E 2. 5 ra p id ly cooled to avoid und esirab le c r y s t a l forms of the above compounds. A fte r co o lin g , the c lin k e r Is ground and blended, norm ally by in terg rin d in g , w ith gypsum to a f in e powder. The f in a l product, c a lle d Portland cement, is the b asic Ingredient of concrete. In the burning p ro cess, co n sid erab le C02 Is driven from the raw meal. Any elem ent* not driven o f f a re Increased In the c lin k e r In propor tio n to the q u an tity o f C02 evolved. Throughout most of t h is re p o rt, the analyses have been reported on a n a tu ra l b a s is , I . e . reported r e s u lts are a ctu al concentrations present In samples as received . In some ca se s, r e s u lts have been reported on an ig n ited b a s is , I . e . on C02 fre e b a s is , and have been Ind icated as such In the rep o rt. 2.2 Effect of A lkalies In cement manufacture, reference to a lk a lie s im plies potassium and sodium oxide (K^O and Na20 ), Both a lk a lie s a re freq uently combined and reported as eq u ivalen t NajO fo r purposes of s p e c if ic a t io n . The raw m aterials a t S t . Lawrence Cement are such that Na20 is low and p r a c t ic a lly , constant (see Table A. 10). For t h is reason, only potassium oxide (K20) Is co n si dered In d e ta il in th is rep o rt. The e f f e c t of a lk a lie s on cement q u a lity has been w ell documented [2 , 3, 4 ]. While some a lk a l i may be d e sira b le fo r e a r ly strength develop ment [ 3 ] , an excess can be problem atic. The most e x te n siv e ly reported of these problems 1 the a 1k a 11-aggregate reactio n [5 , 6 ). Certain g lassy s i l i c a t e s and some dolomites re a ct slow ly w ith a lk a lie s and cause expansion and d isru p tio n o f co n crete. I t has been found by experience that cemeqt containing le s s than 0 .6 0 t to ta l a lk a lie s reported as Na20 performs s a t i s f a c t o r ily w ith such aggregates. T h is s p e c if ic a t io n Is Imposed with s u f f i c ie n t frequency In the United S ta te s that I t I s found as an optional spe c if ic a t io n under ASTM C-150 [ 7 l. Strength a tta in ed and se ttin g characte r i s t i c s a re a ls o re la te d to the a lk a li concentration of the cement [3, 83* I t Is cowman p ra c tic e In the cement Industry to add ch lo rid e s such as calcium ch lo rid e o r waste hyd ro ch lo ric a c id to the raw meal to reduce a lk a lie s [9 , 10]. ^ A lkalies a r t normally present es sulphates which a t k iln operating temperatures ere not re a d ily v o la t fllx a d , but are retained In the c lin k e r . A lk a li ch lo rtd e s e re v o la t ile a t normal k iln operating ACH CC611C 6 tem peratures. These a re evolved from the m aterial in the k iln and carried! in the gas stream to the p r e c ip it a t o r . 7The high a lk a li dust from the pre c ip it a t o r can then be d iscard ed . 2 .3 The S t. Lawrence Cement Co. The S t . Lawrence Cement C o ., M ississauga P la n t, has a nominal production ca p a city o f 1,750,000 short tons per year. The company operates two wet processes and one dry process suspension preheater k i ln . General Information on the company and the p lant has been published in two a r t i c l e s [11 12J. 2.3.1 Relevant u n it processes - wet process k iln I i Each o f the two wet process k iln s a re dumbel1-shaped A l l i s Chalmers k iln s 402* long with a diameter o f l l 1? 1, having nominal capacity] o f 1050 sho rt tons per day (see Figure I ) . The chain system in the drying zone has 57 tons o f loose hung carbon ste e l chains w ith a ra d ia tio n cu rta in at the front (flame end) o f s t a in le s s ste e l ch a in s. The chain system extends through 87' o f k iln length. The s lu r r y feed system is a bucket wheel con veyor w ith a v a ria b le speed d riv e taking s lu r r y from a constant level box. Gases from each k iln (maximum ca p a city 150,000 CFH a t 450F) pass through a s ix - s e c t io n e le c t r o s t a t ic p r e c ip ita t o r . Gases from the p r e c ip ita to rs a re exhausted v ia a common stack 554' in height w ith 1 3 ' e x it in sid e d ia meter. No. 6 fuel o il is burned in a sin g le burner at the centre of the burner pipe. For the t e s t , ch lo rin a te d hydrocarbons were fed ju s t above and to one sid e of cen tre using d iffe r e n t s iz e nozzles fo r proper atomi zation a t d iffe re n t flow ra te s. A d etailed description of the chlorinated hydrocarbon system Is given In Appendix E. 2 .3 -2 . Relevant unit processes - suspension preheater k iln The k iln (Fig u re 2) I s a 17' x 276' T ray lo r u n it normally fire d , through th ree nozzles w ith No. 6 fuel o i l . For the t e s t , ch lo rin ated hydrocarbons Were In je cte d v ia a nozzle a t the centre o f the t ria n g le | formed by the three o i l nozzles.v 1 The suspension preheater (Figure 3) c o n s is ts o f a system of | cyclo n es through which hot e x it gases from the k iln a re drawn by a fan. The raw meal passes through the system In counterflow to the g as. K iln feed I s Introduced Into the duct between the f i r s t and second stage cyclones. I t is swept w ith the hot exhaust gases into the uppermost (Stage I) cyclones CCtuu a: 1 RAW MEAL FEED 2 STAGE I 3 S T A G E Ui s t a g e nr 5 S T A G E IV 6 KILN 7 KILN EXH. FAN S PRECIPITATOR 9 DUST RETURN 10 CONDITIONING TOWER 11 BY-PASS PRECIPITATOR 12 DUST DISPOSAL | 13 DUST DISPOSAL , K WATER gggjjfjg? MATERIAL FLOW ---- > GAS FLOW > PRINCIPLE OF FULLER-HUMBOLDT SUSPENSION PREHEATER AND B Y - P A S S ' FIGURE 3. ACM c e t i l i 8 where gas and m aterial a re separated . The raw feed from the cyclone drops "/ into the_duct between the second' and tfilrd stage cyclones -and Is again suspended and separated. This procedure, being swept up w ith hot gases and then being dropped into the stream entering the*next lowest stag e, I s repeated In Stages 111 and IV before the p a r t ia lly ca lcin e d feed en ters t h e - k iln . The average reten tio n time of the system Is approximately 30 seconds. A d e ta ile d d e sc rip tio n o f the Humboldt preheater has been given by G.A. Schroth [13] from which a r t i c l e these notes have been pre pared. The raw m a terial entering Stage I is preheated to approximately 300C (6 0 0 F), w h ile the gas temperature drops from 530C (990F) to 340* (650F). At each stage, corresponding heat exhanges occur such that the m a teria l en ters the ro ta ry k iln a t approximately 800C (1475F) having been p a r t i a ll y decarbonated. The gas temperature a t the point of e x it from the k iln into the preheater i s 1040C (1900F) to 1090C (20 0 0 F). The S t . Lawrence Cement Co. preheater is a dual Fuller-Humboldt u n it w ith four stages fn each. The a lk a li bypass system (Figure 4) is an important u n it In r e la t io n to t h is study. In common w ith other p lan ts using suspension preheater system s, sp e c ia l measures have to be taken to reduce the buildup o f c h lo rid e s and a lk a lie s in the system. At the S t . Lawrence Cement Co. p la n t , a system Is In use where a fra c tio n o f the k iln exhaust gases is removed from the preheater and passed through a conditioning tower. In the conditioning tower, water Is sprayed Intu the gas stream to lower the temperature and cond ition the gases fo r p r e c ip ita t io n . C o n cu rren tly, the gas v e lo c it y Is reduced, s in c e the c ro ss sectio n o f the tower is g reater than th a t o f the bypass d u ct. The net e f f e c t o f cooling and v e lo c it y reduction Is to .d iv id e the p a r t ic u la t e matter c a rrie d in the gas stream into two f r a c t io n s . One f r a c t jo n , o f lower a l k a l i content, is separated in a condi tio n in g tower and returned to the raw meal s i l o s . The other f r a c t io n , o f { higher a lk a li content, i s co lle cte d In an e le c t r o s t a t ic p re c ip ita to r, p e l le t iz e d and d iscard ed . \ AC* Gu B 1 1- * 1 ^w***^*# ACM CCS 1 ALKALI B Y -P A SS FIGURE 4 a c o n d it io n in g t o w e r 4 PREOPITATOR 5 EXHAUST FAN 0 WATER 9 a r q u e n c h n o z z l e c o n t r o l H 5 o f ) C CONTROLLER PT PRESSURE TRANSMITTER M MODULATOR 10 3 CONSIDERATIONS AT THE PROGRAM PLANNING STAGE The h isto ry of suspension preheater k iln s In the United S ta te s! *. / in d ica tes a trend to t h is system [1AJ. Th irteen preheater k iln s were ; In s t a lle d in the years 1953-1955- Because o f inadequate knowledge about the b a sic process and e f f e c t s o f raw m a te ria ls on t h is system, numerous operating d if f ic u lt ie s were encountered. These d if f i c u l t i e s , p o ssib ly combined w ith emphasis on a lk a l i red u ctio n , caused the shut down o f s ix of these u n its . From 1956 to 1969, only two suspension preheater k iln s were in s t a lle d , one o f which has s in c e been shut down. However, w ith in cre asin g fuel co sts and emphasis on energy conservation, 22 suspension preheater k iln s have been commissioned sin ce 1970. Another in d ica tio n o f t h is trend is that f iv e o f the eleven new k iln s in s t a lle d in 1975 were p re h e a te r-ro ta ry k iln s [1 5 ]. In a d d itio n , three e x istin g ro tary k iln s were converted to suspension preheater u n its . A ll fiv e k iln s planned fo r completion in 1976 w ill be preheater k i ln s . While the situ a tio n in Canada Is s im ila r , there are fewer p lan ts and such trends are d if f ic u lt to follow . Any study concerned w ith future use of the technology developed < must take Into co n sid eratio n the suspension preheater k i l n . . Flame c h a r a c t e r is t ic s are id e n tic a l in a l l cement k iln system s, hence, demonstration that d e stru ctio n o f ch lo rin a te d hydrocarbons occurs In I one cement k iln im pjies d estru ctio n in a l l cement k i ln s . Reduction o f * a lk a l i e s In a s t ra ig h t ro ta ry k iln system by ad d itio n of c h lo rid e s ' r j (calcium c h lo rid e and h y d ro ch lo ric a c id ) has been e x te n siv e ly demonstrated.; I t was decided to conduct the experiment In the suspension j i preheater k iln d esp ite knowledge that the preheater I s prone to plugging j problems a t high a lk a li ch lo rid e le v e ls . The primary purpose o f the study was to determine whether , ch lo rin a te d hydrocarbons were destroyed In a cement k i ln . I t was considered d e sira b le to conduct the experiment in stages with m aterials o f In creasin g d i f f i c u l t y o f d e stru ctio n . A fter each stag e, normal operation was resumed, thus gluing time to analyze samples In order to determine th at no waste ch lo rin a te d hydrocarbons were em itted. For t h is stu d y. I t was agreed to use four d iffe r e n t form ulations of In d u stria l chlorinated wastes: ACM CC8115 l X i ,J i' ! 1 at 2 11 D 2) ' 3) V molecules such as polychlorinated biphyenyls (PCB's). G enerally these are f il t e r e d , processed and blended to obtain s p e c if ic form ulations fo r control of energy va lu e, ch lo rin e content and v isc o sity . ACW CCeilt 12 k TRIAL ON THE SUSPENSION PREHEATER KILN i The f i r s t phase o f the t r i a l was done in the suspension preheater k iln to-determine whether a lk a l i reduction could be accomplished on t h is u n it by burning o f ch lo rin a te d hydrocarbons. Reduction o f a lk a lie s by ad d itio n of c h lo rid e is achieved by increasing the amount o f a lk a l i v o l a t iliz a t io n . A lk a li carbonates and ch lo rid e s are almost e n t ir e ly v o la t iliz e d at the normal operating temperature o f the k i ln . A lk a li sulphates and a lk a l i e s which a re complexed In the calcium s i l i c a t e and pluninate stru ctu res a re p re fe re n tia lly retained in the c lin k e r. The a d d itio n o f c h lo rid e Ion causes formation o f the e a s ily v o la t iliz e d a lk a li c h lo rid e . Intim ate mixing o f hot k iln gases and countercurrent co o ler raw meal causes a lk a li ch lo rid e s to condense on the raw meal thereby being returned to the k i ln . A lk a li c h lo rid e s a re trapped between the burning zone o f the k iln ( v o la t iliz a t io n ) and the bottom stage o f the preheater (condensation), causing In crea sin g concentra tio n s In the gas stream. At high concentrations o f a lk a li c h lo rid e s, s u f f ic ie n t q u a n titie s condense In Stages I I I and IV o f the preheater to cause plugging o f t h is system. To a lle v ia t e t h is s it u a t io n , the bypass is used to withdraw a fra c tio n of the gases leaving the ro tary k iln and w ith these gases a fra c tio n o f the a lk a li c h lo rid e s . As g reater q u an titie s of a lk a li ch lo rid e are v o la tiliz e d in the k iln , high w ith drawal rate s through the bypass are required. Samples o f the raw meal en terin g the k iln from the preheater can be taken to monitor the a lk a l i ch lo rid e co n cen tratio n . P r io r to the burn. I t was estim ated that eq u ilib riu m a lk a li ch lo rid e concentration a t t h is point would be reached In approximately two hours. Cases exhausted v ia the bypass have a temperature o f approx im ately I000C (1830F). To cool these gases, they a re mixed I n i t i a l l y w ith ambient a i r , than passed through a conditioning t w e r In which water spray I s used fo r fu rth e r coo ling and conditioning o f the g a se s. 1n the cond itioning tower, the c o a rse r f ra c t io n o f the d u st, which has the lower a lk a l i co n cen tratio n , s e t t le s to the bottom o f the con d itio n in g tower and Is returned to the p ro ce ss. The f in e r f r a c t io n , which has the higher a lk a li concentration, Is c a rrie d with the gas stream to the bypass p r e c ip it a t o r . Dust c o lle c te d from t h is p re c ip ita to r I s d iscard ed . I ACP CC6117 13 I n i t i a l l y , attempts to operate the bypass at the level required to compensate for ad d itio n of ch lo rid e s, resu lted in high emission ra te s of p a rt ic u la t e m atter. Weighings o f m a te ria ls showed that 50-60 tons per day of dust were c o lle c te d In the p re c ip ita to r (norm ally 6-10 tons per d a y ), w h ile k tons per day s e ttle d In the conditioning tower (normally 20 tons per d ay). The re v e rsa l o f s o lid s fra c tio n removed in the two u n its and the la rg e r amount of m aterial c a rrie d forward to the e le c t r o s t a t ic p r e c ip ita t o r was a ttrib u te d to the Increased q uantity o f ambient a i r used for coo ling the bypass stream. Th is increased gas flow gave higher v e lo c it ie s In the conditioning tower, lowered It s s e ttlin g e ffic ie n c y and re su lte d in p a r t ic u la t e entrainment in the gas stream. The burning o f ch lo rin a te d a lip h a t ic m aterial was begun on June 3 1975 w ith a m ixture having a s'p ecfflc g ra v ity o f approximately 1.2 and a c h lo rin e content of about 551 w/w. On June 5 the bypass duct between the k iln and the conditioning tower was plugged. The t r i a l was suspended w h ile the duct was cleaned . A second attempt was .begun on June 10 using the same parameters used in the June 3 t r i a l . On June 13. s im ila r buildup again caused the bypass to f a i l . I t was apparent that the a d d itio n o f c h lo rid e s to the suspension preheater k iln could not be done without major p ortions o f the bypass system being rebut I t . Due to high p a rt ic u la t e em issions, plugging o f the bypass system and the c o st o f the equipment changes envisaged to a lle v ia t e the sh o rt comings, I t was decided to continue the t r i a l In one of the two e x istin g wet k iln s . The t r i a l on the dry process k iln did show, however, that chlo rinated hydrocarbons were destroyed In t h is k iln . Chlorinated hydrocarbons used a re v o l a t i l e a t temperatures found in Stage IV (98oC), and could not have'condensed upon the m a te ria l. The expected increase in c h lo rin e co ncentration o f Stage IV dusts was confirmed by analyses i (Table 6 .6 ) . E q u i l i b r i a concentrations o f ch lo rid e a t Stage IV were obtained In approxim ately four hours (Appendix B ). Some s lig h t a lk a l i reduction was apparent. During the f iv e days on which ch lo rin a te d , hydrocarbons were burned, the raw meal feed KjO on ig nited b a sis averaged 1.471. w h ile the average c lin k e r l y ) was 1.27*. With the same quantity AC* CC6116 i 14 o f gases being removed by the bypass, but without a d d itio n o f any ch lo rin ate d m a te ria l, the raw meal feed contained on average 1.44% K^O and the c lin k e r contained on average 1 .2 & K^O. While the d iffe re n ce i s s l i g h t , the duration o f the experiment was sh o rt. Complete d e t a ils and a n a ly t ic a l r e s u lts from the experiment on the suspension preheater k iln a re given In Appendix B. i 1 i i v\ ,i a o m cceiis -*r'npTjrvvtr I m % (5 ft i V 15 *5 * TRIAL ON THE WET PROCESS KILN 5.1 . Discussion _ / The t e s t program for t h is se ctio n of the study was designed to* determine I f k iln em issions contained any chlo rin ated org anic compounds when ch lo rin a te d organic wastes were burned as supplemental fuel to reduce the a lk a li content of the c lin k e r. From a knowledge of k iln zone temperatures and gaseous product retentio n times In the k iln I t was a n ticip a te d that combustion of the waste c h lo rid e fuel would produce predominantly CO^, H^O, HC1 (hyd ro chlo ric a c id ) . Cl (c h lo rid e Ion) and a n e g lig ib le amount of fre e Cl^ (ch lo rin e gas). H2 + C12 ^ 2HC1 + *2 tPHCl)2 <Pp / Kp < V - <pci2> - 15 at 1000C ( I 8J 2F) - 68 a t 1500 C (2732F ) -150 a t 1500C (3452F ) [28] , Th is shows that amount o f fre e c h lo rin e gas decreases w ith : a) increase in temperature; b) Increase In water vapor content; c ) decrease In oxygen content; and d) decrease or removal o f h yd ro chloric a c id . The hyd ro ch lo ric a c id w i l l re a ct and the c h lo rin e w ill be retained a s a lk a l i c h lo rid e s In the process s o lid s . From a knowledge o f orthodox In cin e ra tio n systems the combustion cond itions in a cement k iln were expected to provide a very favourable means for the destruction of waste chlorinated hydrocarbons. In order to determine the e f f e c t , I f any, o f burning waste ch lo rin ated o rg an ic m aterial on a i r q u a lit y , em issions from the k iln were monitored b efo re, during and a fte r three separate periods o f burning the supplemental waste f u e ls . P rio r to each saiqpllng period, em issions were checked for the presence of AC C C t iiC 16 HCl end Cl^ w ith -d etecto r tubes a t s e n s i t i v i t y lim its o f 2 ppm end 0.5 ppm, r e s p e c t iv e ly . r/ - A mess be I*nee for c h lo rin e was c a rrie d out by an alyzin g e l l feed end process semples c o lle c te d during each waste burn fo r t h is element, to confirm the emission data. P rio r to the t r i a l on the wet process k i ln , lig h t o i l was put - Intcr th e ch lo rin ate d feed system fo r the purpose o f c a lib r a tin g and te stin g the proportioriing and metering d e v ice s. Some lig h t o i l remained in the system when ch lo rin a te d a lip h a t ic m aterial was re ce iv e d , causing the chlo rid e content to Increase as the te st progressed. While em issions were te ste d , ch lo rin a te d a lip h a t ic m aterial was burned a t d iffe r e n t ra te s w h ile attempting to compensate for the changing c h lo rid e content,- the rates being equivalent to: October 28, 1975, 0.31% c h lo rin e October 29, 1975 0.32% ch lo rin e October 30, 1975 0.63% ch lo rin e re la tiv e to c lin k e r; r e la t iv e to c lin k e r ; and re la tiv e to q lIn ker. Due to. e x ce ssiv e q u a n titie s o f ch lo rin e Input a rin g was formed, which required a k iln shutdown fo r I t s removal. A rin g Is caused by the buildup o f m aterial on the in sid e o f the k iln to such an extent that It r e s t r ic t s the flow o f process m a te ria ls. Due to production scheduling , the delay caused by the shutdown made i t necessary to drop the burning o f arom atics alone and Immediately progress to the m ixture of arom atic and complex m olecules. The conduc t i v i t y o f t h is la t t e r m aterial was too low (0.3 x 10 ^ mhos) fo r the magnetic flow meter to function properly. Control o f flow ra te s was th erefo re achieved by measuring the depth o f liq u id In the tank. While accu rate over a long period of time, sh o rt term control waf d i f f i c u l t because one Inch of liq u id In the tank I s 196 Im perial g a llo n s. V a ria b le flow ra te s were encountered during *i em ission t e s t in g , the q u an tity added each day balng: December 10, 1975 0.65 to 0.71X c h lo rin e r e la t iv e to c lin k e r ; December 11, 1975. 0.31 to 0.51% c h lo rin e r e la t iv e to c lin k e r ; and December 12, 1975 0.79% ch lo rin e r e la t iv e to c lin k e r . CM C C 6 1 2 1 17 A ring a ls o formed during t h is burn but broke away a ft e r the I sampling sequence had been completed,/thus avoiding another k iln shut down. The presence of t h is ring r e s t r ic t e d the burning o f the residual quantity of th is m aterial (aromatic plus complexes) to Interm ittent burns. Due to the above mentioned upset and requirements o f the production schedule, I t became necessary to add 13,000 g allo n s of the p olychlorinated biphenyl b?end to the 12,000 g allons o f m aterial remaining in the tank. The re s u ltIn g mixture contained a s ig n if ic a n t quantity of s o lid s and a ls o had very low c o n d u ctiv ity , thus req uiring tank measure ments as the method used for flow c o n tro l. The problem w ith s o lid s separation and settlin g'b ecam e more cr severe a s the t e s t progressed. T h is caused burner nozzle r e s t r ic t io n s i and pluggages which, in tu rn , caused Interrup tio ns and Ir r e g u la r it ie s in - t waste liq u id flows. I i For the la s t day of em ission te s t in g , the nozzle was removed I completely to permit uninterrupted flow rates of polychlorinated biphenyls. Q u a n titie s added during emission te s t s were: January 7 1976, 0.06 to 0.1AX ch lo rin e r e la t iv e to c lin k e r ; January 8, 1976, 0.13 to 0.33% ch lo rin e r e la t iv e to c lin k e r ; and, January' 9, 1976, 0.61% ch lo rin e r e la t iv e to cl in k er. 5.2 Organic C hloride Waste Burned The chemical compositions of the three waste m aterials burned, the la b e llin g used to id e n tify each burn and the tlme^perTod's during which they were fed to the k iln a re given below: Chlorinated a llp h a tlc s WBA p lu s ch lo rin a te d aroaatlcs-and a lic y c llc s V80 p lu s polychlorinated biphenyls (PCB's) WBA October 23 " November k/75 WBB December 5 " December 15/75 WBC January 3 - January 9/76 It was a n ticlp a ted yth a t correlate combustion o f these m a terials would occur in the k i ln . However, a knowledge o f the major co n stitu e n ts o f each o f the fu e ls was necessary In order to determine which tra ce components might be detected In the sanpled em issions should incomplete combustion o ccu r. Previous stu d ies on the In cin e ra tio n of waste AC* 18 ch lo rin a te d hydrocarbons { I6] had shown that trace q u a n titie s o f low m olecular weight compounds such as carboh te tra c h lo rid e (CCl^ ), chloroform (CHCl^)f dlchloromethane (CH^CI^) were detected in the combustion product g ases. I t was decided, th e re fo re , to analyze c o lle cte d em ission samples fo r the presence o f the major conponents of each composite waste fra c tio n , and for the low m olecular weight organochlorldes re fe rre d to above. Samples o f the ch lo rin a te d wastes were taken from the lin e between the tank and the k i ln . Typ ical component a n alyses o f the three composite wastes burned during the program are recorded In Tables 1, 2 and 3. The q u a n tita tiv e valu es given in the tab les a re based upon an assumed equal response o f a l l the components to a flame Io n iza tio n d etector (F ID ). While t h is assumption Is not very a ccu ra te , I t was considered s u f f ic ie n t to provide the needed compositional data. Id e n t i f ic a t io n o f components a t concentrations g re ater than one percent was achieved by use o f gas chromatography-mass spectrometry (GC-HS). Id e n ti f ic a t io n o f components a t concentrations le s s than one percent was not ro u tin e ly performed. Analyses o f samples c o lle c te d on d iff e r e n t days during a burn showed some v a r ia t io n in the concentration o f Ind ivid ual components. However, the o v e ra ll chemical composition o f the feeds remained the same. Two o f the waste burn sample feeds (WBB and WBC) were a ls o analyzed by the O ntario M in istry o f the Environment (MOE), and the TRW Systems Group. The r e s u lt s obtained by these agencies a re reported In d e t a il in Appendices F and G, re s p e c t iv e ly . The TRW Systems Group analyzed composite samples o f the d a lly feeds fo r both waste burns by GC-MS. The HOE group analysed In d ivid u a l d a lly samples fo r both waste burns using GC-FID operating c o n d itio n s. The chromatographic parameters used by the HOE group to a n a ly se fo r arom atic and PCB components precluded the Id e n t if ic a t io n o f a lip h a t ic organchlorlne compounds such as CHCl^ and CCl^. Some v a r ia t io n In the percent composition o f the waste burn samples.was apparent between th TRW and ORF r e s u lt s , e s p e c ia lly w ith resp ect to the amounts o f a lip h a t ic organochlorIne compounds present In waste burn S (WBB). The d iscre p a n cie s observed a re due to the fa c t th a t TRW received a composite sample o f a l l d a lly feeds fo r each o f the 19 TABLE 1. COMPOS H I ON OF ALIPHATICS (VBA) Sample Feed October 28, 1975 Peak / in GC P r o f ile (cf Figure A.6) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 15 17 18 19 20 21 22 23 24 25 26 27 Approximate Concentration X 17.4 1.4 0.8 1.5 3-2 16.5 6.6 1.0 10.5 0.7 0.3 1.9 1.4 1.4 2.4 7.3 15.6 2.4 0.2 3.3 0.3 1.1 0 .6 0.5 1.6 0.6 0.7 x * y Id e n tific a tio n Chloropropane & propene e th y lc h lo rid e , dichloromethane Chlorobutane 5 butene 1( 2, Dlchioroethane Carbon te tra c h lo rid e 1, 1, 2-Trlchioroethane DIchIoropropanes T etrach lo ro eth y1ene Tetrachloroethane Chlorobenzene * multi chlorinated butanes, butenes hexanes, hexenes Note: No id e n t if ic a t io n o f compounds at concentrations o f }% or le s s was attempted. \\ (: 20 TABLE 2. COMPOSITION OF AROMATICS PLUS COMPLEX (WBB) Sample Feed - pecembep 12, 1975 Peak / In GC P r o f ile (c f Figure A .7) ---------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Approximate Concentration X 2.0 0.1 0.1 1.4 1.5 3.5 0.4 1.1 1.9 0.2 0.2 0.4 0.6 0.5 0.5 52.2 0.2 0.1 0.1 6.3 2.7 0.1 1.0 8.5 4.4 6 .9\ 1.3 * Id e n tific a tio n Chloropropane - Chloroform 1, 2-DIchloroethane Carbon te t ra c h lo rid e 1, 1, 2-Trichloroethane Chlorobenzene Chlorotoluene > Hexach1o ro cy c1open ta dIene Heptachlorocyclopentene Pentachlorobenzene Octachlorocyclopentene Complex a sso c ia te d w ith Hexach1o ro cy c1opentadiene Note: No Id e n t if ic a t io n o f compounds a t concentrations o f IB or le s s was attempted. ALM CCd 12 5 1 I 4 21 TABLE 3. COMPOSITION OF AROMATICS PLUS PCB's (WBC) */ Sample Feed * January 8, 1976 Peak In GC P r o f ile (cf Figure A.8) Approximate Concentration X Id e n tific a tio n 1 3.4 2 0.3 3 l.A A 1-3 5 0.5 6 0.3 7 1.9 8 0.1 9 1.9 10 0.5 11 28.5 12 < 0.1 13 < 0 . 1 14 1.8 15 0.9 16 - 2 .8 17 0 .9 18 19 20 21 22 * 23 2A 2.2 0.3 2.1 0.9 4 6.7 3-4 5-9 25 1.6 26 12.1 27 28 5.6 4 .0 J 29 5-4 30 2.1 31 a 1.6 Chloropropane - 1, 1,-Dichloroethane Carbon te tra c h lo rid e - 1 1, 2-TrIchlo roethane - Chlorotoluene - Hexach1o ro cy c1opentad Iene + complex Dlchloroblphenyl T r Ic h 1orob 1pheny1 T e t r a chlo roblphenyl Pentachlorobipheny1 * Note: No Id e n t if ic a t io n o f compounds at concentrations o f \X or le ss was attempted. ^ V ACM CC6126 22 #' waste burns, end d iff e r e n t chromatographic columns end a d iffe r e n t detector system were used by the two la b o ra to rie s. ~ Since the study was to evaluate the burning o f waste ch lo rin a te d m a te ria ls o f v a r ia b le com position, minor d iffe re n ce s between the la b o ra to rie s 1 r e s u lt s do not a ff e c t the conclusions o f the study. __________ The agreement among a l l groups was good w ith resp ect to the Id e n t if ic a t io n o f components present in the two waste f u e ls . The presence o f ch lo rin a te d a iip h a t ic s was observed in both waste f u e ls . VBB sanples were found to c o n s is t p rim a rily o f ch lo rin a te d arom atic compounds. In p a r t ic u la r o -ch lo ro to lu en e. WBC samples were found to c o n s is t o f approx im ately 502 p olych lo rin ated biphenyls (A roclor 1242) p lu s the ch lo rin a te d arom atics and a l i c y c l l c s found In UBB sa n p le s. 5-3 Emissions Samples o f the k iln em issions before (BLA), during (WBA, WBB, WBC), and a f t e r (BLB) the burning o f waste organic ch lo rid e s were taken, using the equipment and methods described In Appendix A. The a n a ly t ic a l techniques used to determine i f any organic ch lo rid e compounds were present In the em issions samples a re discussed with respect to both gaseous and p a r t ic u la t e em issio n s. 5.3*1 fre e ch lo rin e and hydrogen ch lo rid e P rio r to each sampling p eriod , an alyses were made fo r the presence o f fre e c h lo rin e ( C l.) and hydrogen c h lo rid e (HCI) In the k iln em issio n s. *% An MSA gas sampler and MSA d etector tubes were used to determine the concentrations o f these p o llu ta n ts . In no In stan ce, e ith e r during b aselin e or waste burn sampling p e rio d s, were CI^ or HCI detected. The s e n s it iv it y lim it s using t h is a n a ly t ic a l procedure a re 0.5 ppm fo r C!^ and 2 .0 ppm for HCI. 5 .3 .2 Caseous organic compounds Em ission samples were c o lle c te d using two sampling t r a in s . The f i r s t , a gaseous sampling t ra in (Appendix H), was designed to c o lle c t any v o l a t i l e low m olecular weight Ofganlc c h lo rid e s , e .g . CCl^, CHCl^, CH^CI^ by adsorption on an In e rt adsorbent. In t h is study Chromosorb 102 was used. The second, a p a r t ic u la t e sampling t r a in (EPA Joy # 5 ), was expected gcbU 1 t i f t a 23 I to c o lle c t higher m olecular weight compounds, including any noncombusted l sta rtin g m aterials In the Implngers. . _/ Components adsorbed on the sorbent were removed by thermal desorption p r io r to a n a ly s is . A ll other c o lle c te d f r a c t io n s , from both t r a in s , were extracted with organic solvents p rio r to a n a ly s is . These included probe r in s e water so lu b les and In so lu b le s, f i l t e r s and impinger so lu tio n s. In consid ering the r e s u lt s obtained in t h is study i t is important to keep in mind the fo llo w in g : - The c o lle c t io n e f f ic ie n c y of Chromosorb 102 sorbent fo r low m olecular weight ch lo rin ated hydrocarbons was determined to be b etter than 902. The c o lle c t io n e f f ic ie n c ie s of f i l t e r s and Ice-w ater in the impingers of a Joy t r a in fo r organoch lo rid e components was not determined. - The methods of c o lle c t io n , followed by thermal desorption or so lvent e x tra c tio n , were designed to concentrate any organic compounds which may have been p resent. Such techniques allo w d etectio n and Id e n t if ic a t io n of compounds a t much lower le v e ls than would otherw ise be p o ssib le . Concentration fa c to rs fo r each type o f sample have been ca lcu la te d fo r each phase o f the burn .and a re given in Table k. For the desorbed gas samples, organic compounds from sev era l cubic feet o f k iln em ission gases were c o lle c te d on Chromosorb 102 anff then desorbed into * 500 m i l l i l i t r e s . The concentration fa cto rs are the ra tio s o f these two volumes. In the case of the solvent extracted samples, molecular weights enter the ca lcu la tio n in the conver slo n of volume to weight. A complete d iscu ssio n of these ca lcu la tio n s Is given In Section A .6.4. 5*3*2.1 Desorbed samples. The concentrations of v o l a t i l e organic c h lo rid e s c a lc u la te d fo r the k iln em issions from gas chromatography e le ctro n capture (GC-EC) are Recorded In Table 5* The r e s u lt s reported were averaged fo r each te s t s e r ie s . Dlehloromethane (DCM) was t e n t a t iv e ly Id e n tifie d as the major component of these desorbed gas samples. Others t e n ta tiv e ly Id e n tifie d were CHCl^ AO* 2k TABLE 4. GAS SAMPLE VOLUMES.AND SAMPLE CONCENTRATION FACTORS (Gaseous Sampling T ra in ) / Test - Date 1 BLA 2 BLA 3 BLA O ct. 20 (1975) O ct. 21 O ct. 22 1 VBA 2 VBA 3 VBA O ct. 28 O ct. 29 O ct. 30 1 VBB 2 VBB 3 VBB Dec. 10 Dec. I 1 Dec. 12 1 VBC 2 VBC 3 VBC Jan. 7 (1976) Jan. 6 Jan. 9 1 BLB 2 BLB 3 BLB Jan. 19 Jan. 20 Ja n . 21 Test Duration (mins) 317 261 320 362 266 285 260 317 250 268 237 228 170 247 240 Volume* Samp1ed scf 2.80 1.86 2.83 3.22 2.53 2.53 2.30 2.80 2.21 2.37 2.09 2.02 1.50 2.18 2.12 Sample Concentration Factors Desorbed Solvent** Gas Extracted 159 141 510 160 181 W*3 1*3 j 130 159 125 4 J 13*1 119 114 j 85 124 120 500-1100 600-2000 520-1500 380 Volume sampled per adsorbent tube. These va lu e s e re based on the lowest and highest m olecular weights of compounds found In the waste feed s. Fig ures given fo r both b a se lin e se rl.e s a re based upon the m olecular weight o f dichloromethane (DCM). \ 1 CM C0612S 4 1 25 TABLE 5. ESTIMATED KILN EMISSION CONCENTRATIONS (CC-EC) FOR SPECIFIC VOLATILE ORGANOCHLORINE COMPOUNDS // Emission Concentrations Test Series Diehloranethane ppb ug/n Chloroform ppb ug/m^ Carbon T e tra c h lo rid e ppb pg/m^ BLA A. 1 VBA 5. A WBB 18.0 WBC 7-7 BLB 29.0 lA .S 27.3 19.1 102.7 63.7 0.00 A 0.015 0.038 0.069 0.018 0.020 0.080 0.190 0.345 0.090 0.000A 0.0020 0.0020 0.0060 O.OOOA 0.0026 0.0128 0.0128 0.0385 0.0160 and CCl^. It was found that GC-EC p r o f ile s o f blank determ inations obtained w ith unused conditioned Chromosorb 102 were v a r ia b le , and the peaks that were present In the p r o file s possessed s im ila r retention times to the compounds o f In t e r e s t . Th erefo re, reported r e s u lt s are higher than actual concentrations In the amission g ases. The d iffe rin g concentrations fo r DCM quoted in Table 5, e s p e c ia lly w ith respect to BLB versus BLA, VBA or WBB, are probably a r e s u lt o f th is background contamina tion rather than re a l d iffe r e n c e s . I t was s t i l l noted, however, that the maximum concentration ca lcu la te d as DCM In the k iln em issions was no greater than 30 ppb even w ith t h is p o s itiv e bias from background contamination. " The HOE r e s u lt s confirm the ORF fin d in g s w ith resp ect to the d etection o f low m olecular weight ch lo rin a te d organics In the desorbed gas samples. TRW did not fin d any tra c e o f these v o l a t i l e ch lo rin ated org anics In t h e ir desorbed gas samples. However, TRW did not ro u tin e ly perform s p e c if ic a n aly ses fo r compounds present a t concentrations below 0.1 mgAi^. 5 .3 .2 .2 Organic solvent extracted samples. Oue to the much higher gas flow ra te through the p a rt ic u la t e sampling t r a in to maintain an Iso k in e tic sampling ra te , the concentration facto rs for solvent extracted samples for t h is tra in were greater than those c o lle cte d w ith the gaseous tra in . Calculated average concentration factors for each test se rie s are ACH C C6 1 3 C 26 given below: BLA WBA VBB UBC BLB / 7,000 6,000 - 1 3 ,0 0 0 7,000 - 23,000 7,000 - 20,000 6,000 For the waste burns, the two values given represent the hig hest and lowest m olecular weight sp e cie s present In the waste ch lo rin ate d hydro' carbons. C a lc u la tio n s fo r the b a se lin e s a re based on the m olecular weight o f dlchloromethane. Subtraction o f background In terfere n ce based on control blanks (Appendix A .6 .3 ) was performed on solvent extracted samples. Comparison was then made w ith the appropriate waste feed sample chromatograms. I t was concluded that no uncombusted waste fuel components were present in any o f the organic solvent e x tra ct samples a t the detection lim its of the a n a ly tic a l procedure used. Representative chromatograms together w ith d iscu ssio n are presented In Appendix A .6 .5 . TRW performed a n aly ses on portions o f a l l the organic so lven t e x tr a c t s obtained from 8LA, VBB, VBC and 6LB em ission sa n p le s. The samples were s p e c i f i c a l l y searched fo r ch lo rin ate d compounds. A lso , the Chemistry D iv is io n , A ir P o llu tio n Control D irecto rate (APCD), Department o f the fnvironm ent, Ottawa, Canada c a rrie d out an alyses on p ortions o f the so lve n t e x tr a c t s o f WBC emission samples to search fo r high molecular weight chlorinated hydrocarbons. Detailed reports o f the work undertaken by these groups a re presented In Appendices G and I, re sp e c tiv e ly . The noteworthy r e s u lt w ith resp ect to both rep o rts was that in none.of the samples analyzed by e ith e r group were any ch lo rin a te d org anic compounds d etected . Their r e s u lt s thus confirm the ORF conclusion that no ch lo rin a te d org anic resid u es were d etectab le In any o f the organic solvent e x tra cts o f the em ission samples co lle cte d during the vario us waste chlo rinated hydrocarbon b u rn s.\ 5 .3 .2 .3 R e su lts obtained by the p a rtic ip a tin g la b o ra to rie s. In gas chromatographic a n a ly s e s , program parameters define the co n d itio n s used. The choice o f d etecto r and column makes the a n a ly s is s p e c if ic to c e rta in ACh COB 131 27 -groups o f compounds, four lab o ra to rie s analyzed the em issions samples taken during th is study. One of the la b o ra to rie s, APCO, was requested to search fo r high m olecular weight ch lo rin a te d hydrocarbons. The TRW group were not requested to Id e n tify low m olecular weight ch lo rin ated compounds present a t le s s than 0.1 mg/cn^ In the sta ck gases. TRW d id , however, search for PCB's using techniques designed to detect small q u an tities of these compounds. ORF and HOE were assigned the task of detecting and id e n tify ing a l l p o ssib le ch lo rin a te d org anic s p e c ie s . A n a ly tic a l d esig n, and hence r e s u lt s , r e fle c t d ifferen ces in the analyses requested of the different laboratories. Both ORF and HOE found low m olecular eight hydrocarbons, such as dichlorom ethane, to be present at microgram per cubic meter (ug/m ) le v e ls in the em issions. That TRW detected no sueh compounds Is not co n tra d icto ry . The le v e ls fn the em ission gases ere le s s than t h e ir required d etectio n lim it s . S im ila r ly , I t was not w ith in the terms o f reference fo r the APCD work to determine these compounds. R esu lts o f the Tour la b o ra to rie s can be conveniently sunmarized. While burning ch lo rin a te d hydrocarbons, low m olecular eight ch lo rin ated compounds ere em itted a t le v e ls of a few pg/m^ (Table 5 ). None of the p a rtic ip a tin g lab o ratories, detected any high molecular weight chlorinated hydrocarbons from e ith e r a i r sampling t r a in . At detection lim its of 3 ug/m in the sta c k g ase s, p o lych lo rin ated biphenyls er.e not found. 5 .3 .2 .4 Estim ated minimum combustion e f f ic ie n c ie s . The'maxi mum value fo r to ta l ch lo rin a te d hydrocarbon content In the k iln em issions was obtained In t e s t 2 of the WBC s e r ie s . Ignoring background su b tractio n fo r the b a se lin e samples and the In terfere n ces from control blanks a maximixn value o f 40 ppb as determined. I f a c o lle c tio n e f f ic ie n c y of BOB I s assuned, a maximum level of 50 ppb In the k iln em issions is obtained. Because of the high and uncertain background le v e ls , the estim ate I s higher then actual le v e ls In the emission gases. An average m olecular eight fo r each of the three composite feeds cen be obtained from a knowledge o f the composition of the feed. Using th is Inform ation, together w ith the average fuel feed rates to the k iln end average gas volume flow ra te s In the duct, the maximum AQf* c c e 3 28 ch lo rin a te d hydrocarbon content can be determined. These values are presented in Table 6. Minimum combustion e f f ic ie n c ie s were then ca lcu la ted , / using the estim ated maximum value o f 50 ppb derived above. These fig u res In Table 6 ere considered very conservative estim ates because of the method used to c a lc u la t e them. TABLE 8. ESTIMATED MINIMUM COMBUSTION EFFICIENCIES FOR EACH WASTE BURN Vaste Compos i te Maxi nun Estim ated Organic Content In K iln Em issions Assuming no From Sanple Combustion, Chromatograms ppm g/nr ppb ug/nr* VBA 550 2.1*0 VBB *70 3-37 VBC 350 3.02 50 177.1 50 177.1 50 177*1 *8ased on the m olecular weight o f dichloromethane. Minimum Combustion Efficiency X v/v 39.930 99-989 99.986 5.3*3 P articu late emissions Suninary data fo r the p a rt ic u la t e t e s t s made during each te st period a re provided in Table 7- For each s e r ie s o f t e s t s , the p a rtic u la te loadings and em ission ra te s a re q u ite co n sisten t except fo r the th ird t e s t in the VBA s e r ie s . The very high loading obtained .probably resu lte d from some temporary m alfunction o f the p r e c ip ita t o r . I t Is n o ticeab le th a t, when the c h lo rid e wastes were burned tn the k iln , the p a rt ic u la t e em ission ra te In creased . For the VBA s e r ie s , the em ission ra te s were about four times the ra te s obtained during the b a selin e t e s t s and for the VBB and VBC s e r ie s the ra te s were twice those obtained during the baseline te sts. Increased emission ra te s were not unexpected sin c e combustion o f the c h lo rid e wastes produces^HCI and CI^, which re a ct w ith the a lk a li components In the raw feed to fotm v o la t ile a lk a li c h lo rid e s . At the p re c ip ita to r the p a rticu la te loading I s , therefore. Increased and, since the condensed a lk a li c h lo rid e s a re very fin e and have a d iffe r e n t r e s i s t i v i t y , the amount o f m aterial passing through the c o lle c t o r Increases. Another facto r Influencing emissions o f p a rticu la te matter ALP 00013 3 ji I_ _ *W| 'WMWI 1 W JW MWWMUafM TABLE 7; SUMMARY OF PARTICULATE TEST DATA Test I 1 BLA 2 BLA 3 BLA t . Date- Oct. 20/75 O ct. 21/75 O ct. 22/75 Probe Rlns Gain (g) 57.6 58.8 3S.8 * 1 UBA 2 UBA 3 UBA Oct. 28/75 Oct. 29/75 Oct. 30/75 99.0 96.8 237-5 F ilte r Gain (mg) 285.7 366.0 291.8 I262.6 1261.7 2193.8 Total Gain (mg) 363-3 626.8 327.6 1361.6 1356.5 2631.3 % of Volume Total Gain Samp1ed on FI It e r (std f t * ) 83 161.06 86 1 3 5 . 8 6 89 132.83 . 93 93 90 162.81 137.36 109.88 1 UBB 2 UBB 3 UBB Dec. 10/75 Dee. i 1/75 Dee. 12/75 163.0 06.1 165.6 626. 1 668.6 621.6 567.1 556.5 767.2 75 86 81 IO6 . 6 6 I I 7 .O6 116.26 1 UBC 2 UBC 3 UBC Jan. 7/76 Jan. 6/76 Jan. 9/76 62.2 63.1 87.8 566.3 636.5 ' 600.0 608.9 679.6 687.0 90 9.1 87 119.62 113.52 118.99 1 BLB 2 BLB 3 BLB Jan. 19/76 Jan. 20/76 Jan. 21/76 65.2 51.6 11.9 25I .7 2 6 O .I 183.5 316.9 29I.5 195.6 79 82 96 126.88 116.60 116.65 Flow rate ACFM 157.000 155.000 153.000 Average 157.000 153.000 165.000 Average 168,000 162,000 161,000 Average 172,000 160,000 167,000 Average 183,000 163,000 160,000 Average Concentration (g ra ln s/ft3 ) 0.0376 0.0683 0.0367 O.O6O9 0.1658 0.1526 0.3615 , O.2 I 32 0.0821 O.O7 3 I 0 .10 19 0 .0 8 5 7 0.0785 0.0652 0.0692 0.0776 O.0 3 8 5 O.0 3 6 6 0.0259 0.0363 Em ission Rate (Ib / h r) 20.8 25.6 19.6 21.9 83.9 86.6 200.0 122.8 6O. 3 6 0 .5 55.0 65.3 65.3 36.5 52.2 66.0 23.9 21.0 16.6 19.8 f 30 I were the k iln rin g s formed w hile burning chlo rinated hydrocarbons. K iln rin g s, by t h e ir e f f e c t on gas V e lo c it ie s through the k i l n , in crease "the amount of p a r t ic u la t e m aterial c a rrie d to the p r e c lp ita t o r . Extensive rin g formation was noted during te s t VBA. The average em ission rate for WBA was about 3 lb/ton of c lin k e r produced, compared to the. Canadian Federaf Government o b je c tiv e o f 0 .9 lb /to n. The average value for the other phases o f the t e s t was about 1.1 lb o f p a rtic u la te em itted/ton of c lin k e r produced and th at of the b a se lin e s was 0 .5 . This apparent Increase in the p a rtic u la te emissions w hile burning chlorinated hydrocarbons Is believed to be p a r t ia lly caused by the change in r e s i s t i v i t y o f the duSt en terin g the p re c ip ita to r due to i t s increased a lk a li content. A m odification in the design of the p re c ip ita to r could compensate for th is change. The higher em ission ra te s did not s ig n if ic a n t ly add to the suspended p a rtle u fa te In the ambient a i r In the v i c i n i t y of the p la n t. Computed maximum ground leve l concentrations using standard d isp e rsio n equations were le s s than 2 yg/m^ fo r the b a se lin e em ission ra te s and, during the burning of waste c h lo rid e s , did not exceed 10 lig/m^. The cu rren t O ntario standard Is 100 ug/m^. 5 . A Mass Balance on Wet K iln For the wet process k i ln , the input streams a re s lu r r y feed. No. 6 fuel o i l and, when burned, the waste ch lo rin ated hydrocarbons. C lin k e r (the product) and a portion of the p re c ip ita to r dust form the output stream s. The major portion o f the p re c ip ita to r dust is immediately returned to the p ro ce ss. Th is stream, though not required for the mass b alan ce, was monitored fo r information on k iln operation w hile burning ch lo rin a te d hydrocarbons. Balances of ch lo rin e and potassium were c a lc u la te d for both b a se lin e periods and fo r each type o f ch lo rin ated waste burned. Methods o f sampling and quantifying m aterial streams are d e ta ile d In Appendix A. D e ta ils o f a n a ly t ic a l r e s u lt s , c a lc u la t io n s dnd ta b le s o f d a ily mass balances e re given in Appendix C. 5.A.1 S ig n ific a n c e o f the mass balance In common w ith other m aterial balance experiments on large s c a le production system s, the m aterial accounting In t h is experiment showed apparent lo sse s and gains when the data were expressed In percent retention. OQbV3 5 . * 4 ** 4 31 These should not be regarded as real system lo sses or g a in s. There are random and system atic e rro rs In weighing 4 nd quantifying very large q u a n titie s o f m a te ria ls and in a n a ly t ic a l r e s u lt s . In the case o f a lk a li chlorides a c y c le is developed w ith in the k iln whereby a lk a l i ch lo rid es v o l a t i l i z e in the hotter se ctio n o f the k iln and condense in the cooler se c tio n . Should a k iln upset o ccu r, the cy c le may be broken by a lk a li ch lo rid e being retained In the c lin k e r . Such an occurence may be of s u f f ic ie n t ly short duration that c lin k e r samples taken may not be in d ic a tiv e . T h is e f f e c t Is evident from the data of October 11-1 It, 1975. A power f a ilu r e on the eleventh caused a three hour k iln shutdown. The indicated very low reten tio n o f both ch lo rin e and potassium occurred presumably w h ile the c y c le was being re -e sta b lish e d . Since the eq u ilib riu m changes when burning of ch lo rin ate d m a te ria ls begins or ends, t h is featu re can a lso be seen w ith each phase o f ch lo rin ate d waste burning. At the s t a r t o f any ch lo rin a te d burn, the mass balance for ch lo rin e In d ica te s very low reten tio n w h ile the c y c le is e sta b lish e d . At the end o f the chlorinated burn, a gain is indicated as chlo rine Is retained In process s o lid s w h ile the new e q u ilib riu m Is e sta b lish e d . Formation o f k iln ring s also caused an apparent lo ss of a lk a li ch lo rid e sin ce a considerable quantity was contained w ith in the rin g . 5 .4 .2 C hlo rin e and potassium retained The cum ulative percentage of ch lo rin e retained (Table 8) was low in a l l cases throughout the study, ranging from 50.7 percent fo r the chlo rinated a lip h a t ic burn to 92.2 percent fo r the f in a l b a se lin e . K^O (Table 9) c lo s e ly followed the pattern of ch lo rin e reten tio n , although a t a d if f e r e n t reten tio n le v e l, w ith 83*5 percent the lo re s t value found during the ch lo rin a te d a lip h a t ic burn and 97.6 percent the highest value found during the f in a l b a se lin e . The reason i s c le a r ly th at potassium c h lo rid e was being lo st In rin g formation and k iln c y c le eq uilibrium flu c tu a tio n s . The cumulative percentage retained was le s s a ffe cte d than that o f c h lo rin e by such lo sses because the quantity o f K^O was from 3 to 20 times g reater than the q uantity o f c h lo rin e , Random and system atic e rro rs could a ls o be expected to play a g reater ro le In the c h lo rin e balance fo r the same reason. ACK C C clJc Accumulated Period 7/10/75-22/10/75 23/10/75- 6 / 1 1 / 7 5 2/12/75-1*1/12/75 . 1/ 1/76- 9/ 1/76 10/ 1/76-21/ 1/76 T*1 TABLE 8. ACCUMULATED HASS BALANCE FOR CHLORINE Chlorinated Hydrocarbon Burned -- A liphatic Aromatic + Complex PCB -- Accumulated Input (lb ) <T|> 28,249 . 13*1,379 135,705 58,590 18,351 of T| Prom C hlorinated Hydrocarbon 0 83.2 85.0 82.4 0 1 Accumulated Retention (lb ) 20,768 68,088 112,6*10 41,704 16.927 1 t Retention 73.5 50.7 83.0 71.2 92.2 Accumulation Period 7/10/75-22/10/75 23/10/75- 6/11/75 2/12/75-16/12/75 3/ 1/76- 9/ 1/76 10/ 1/76-21/ 1/76 TABLE 9. ACCUMULATED HASS BALANCE FOR K20 Chlorinated Hydrocarbon Burned -- A1 Iph atld Aromatic + Complex PCB -- Accumulated Input (tons) 240.99 199.67 196.59 103.25 280.40 Accumulated Retention (tons) 229-37 166.72 186.68 92.88 276.36 \ s t Retention 95.2 83.5 95.9 90.0 97.8 CH 000137 33 While the ch lo rin e balance does show d isc re p a n c ie s. In a l l cases w h ile burning ch lo rin a te d hydrocarbons mope ch lo rin e was retained In process s o lid s than- th e .q u a n tity Input w ith m a te ria ls other than these w astes. A major portion of the ch lo rin e from the ch lo rin ated hydro carbon m aterials Is thus accounted fo r. Considering the u n certain ties Involved with rin g formation and k iln u p se ts; the mass balance for c h lo rin e confirm s the fin d in g o f the em ission measurements that a l l ch lo rin ate d hydrocarbons a re destroyed In the cement k iln . 4C* c Ce *6 3<t 6 COHStDERATIONS ON BURNING CHLORINATED HYDROCARBON WASTES -IN A CEMENT KILN / 6.1 . E ffe c t on Production / -Kiln ring formation normally decreases k iln production, apart from any loss of production caused by downtime. However, the k iln ring s formed during t h is study were a ttrib u te d to poor control o f feed rates -- -of the ch lo rin ate d waste. Ring formation through c h lo rin e ad d itio n rates higher than d e sira b le were p re v io u sly encountered at S t. Lawrence Cement w h ile adding waste h y d ro ch lo ric a c id . Adequate control would c e r t a in ly e lim in a te t h is problem and, sin c e t h is is a f a i r l y sim ple p ro p o sitio n , k iln ring formation should not be considered a d eterrent to use o f t h is waste in the k iln . During the stu d y, average production rates (Table 10) were 1038 tons per day w hile not burning ch lo rin ate d hydrocarbons and I 025 tons per day w h ile burning these m a te ria ls. There was a ls o an Increase o f approximately 20 tons per day in the q uantity o f dust discarded (Table 11) w h ile burning ch lo rin a te d m aterials. While these valu es may be p a rtly due to rin g form ation, i t is probable that dust generated would Increase due to Increased v o la t iliz a t io n of a lk a li c h lo rid e , and that t h is would have a corresponding e f f e c t on c lin k e r production. Thus, I f another form of ch lo rid e were being used for a lk a l i red uctio n, It Is u n lik e ly that any change would be detected by use of chlo rinated hydrocarbon w astes. "" 6 .2 A lk a li Reduction While Burning Chlorinated Hydrocarbon Wastes As Ind icated e a r l i e r In t h is re p o rt, Na20 is not considered in t h is study because I t 1s low and p r a c t ic a lly constant in the S t . Lawrence Cement Co. raw m a te ria ls and products. To determine the e f f ic ie n c y of a lk a li reduction, the following p oints req u ire co n sid e ra tio n . The percent KjO reported In the s lu r r y feed Is on the natural o r "as received " b a sis. To determine the quantity which would be present I f note were v o la t iliz e d In the burning p ro cess, the r e s u lt s must be c a lc u la te d on the "ig n ited b a s is " , that I s , recalculated fo r the C02 evolved from the raw m a te ria ls In the burning p ro cess. The I ACM C.C6I39 w p i w w HU I ' l l ' l W l f p TABLE <0. AVERAGE REDUCTION IN 1^0 CONTENT OF CLINKER Accumulation Period 7/IO/75-22/IO/75 23/10/75- <1/11/75 2/12/75- I6/.I2/75 C lin k e r Prod. (t/day) 1056 roso 1020 3/ 1/76- 9/ 1 / 7 6 .' 10/ 1/76-21/ 1/76 1006 1020 *Based on ch lo rin e Input. Chlorinated Hydrocarbon Cl Input With Hydrocarbon % R elative to C lin ker Prod. - A liphatic Aromatic + Complex PCD 0 0.61 0.44 0.34 0 S lu r ry K*0 (2) Natural Ignited Basis B a sis 0.92 0.92 0.90 1.62 1.62 1.62 0.91 0.91 1.41 1.40 K20 k2o C a lc u la te d * Cl Inker Reduct Ion K20 (t) (%) Reduction ( t ) 1.21 0 .8 7 0 .7 6 0.21 55 0.68 0 0.58 0.58 0.87 1.25 0.56 0.15 0.45 0 o or> d* -p n 36 TABLE 11. AVERAGE DUST DISCARDED Period `, Chlorinated Hydrocarbon Average Dust Discarded (Tons Per Day) 7/10/75*22/10/75 23/10/75* V I 1/75 2/12/75-1V12/75 3/ */76- 9/ 1/76 10/ 1/76-21/ 1/76 -- A 1 Ip h a tic Aromatic + Complex PCB -- 22.2 <2.7 7 * .0 62.5 *5.1 percent K^O In s lu r r y feed on the Ignited b a sis minus the percent K^O in the c lin k e r y ie ld s the reduction through v o la t iliz a t io n In the k iln . Sin ce the K^O reduction Is achieved by formation of potassium c h lo rid e (K C I), the reduction expected w hile burning chlorinated hydrocarbons, based on the assumption that a l l c h lo rin e Is bound into potassium c h lo rid e , can be r e a d ily c a lc u la te d . Comparison o f actual and ca lc u la te d values (Table 10) are e x c e lle n t . If ch lo rin e I s added to reduce a lk a l i e s , the reduction is s to ic h io m e tric . However, the to ta l reduction is r e la t iv e to a lk a li le v e ls in the s lu r r y and not to the b a selin e content In the c lin k e r . T h is r e s u lt was confirmed both by t h is study and by previous experience using hydrochoric acid as the source of c h lo rin e . T h is may be due to formation of potassium ch lo rid e from v o la t ile compounds sych as potassium carbonate. On average, without ch lo rid e a d d itio n , a reduction o f 0.18 percent K^O is apparent. While burning ch lo rin a te d hydrocarbons, the average reduction I s that c a lc u la te d on the b a sis of potassium ch lo rid e formation plus 0.05 percent. The extrem ely good c o rre la tio n between quantity of ch lo rin e Input by burning of ch lo rin a te d hydrocarbons and a lk a li reduction In the .c lin k e r g iv e s b e tte r proof than the mass balance that c h lo rin a te d hydrocarbons are destroyed In a cement k i l n . Otherwise, e l k a l l c h lo rid e t mV could not be formed, and v o la t iliz a t io n to the extent noted could not o ccu r. The high degree o f c o rre la tio n In d ica te s that a l l o f the ch lo rin ated hydrocarbon was destroyed. Ch CGbl^tl 37 6 .3 Heat Recovery From Chlorinated Hydrocarbon Waste C a lcu la tio n of heat recovery from these m a t e ria ls ; normally d i f f i c u l t due to the low flow ra te s involved, was made more d i f f i c u l t by the flow control problems encountered. During the arom atic p lu s complex ch lo rin a te d hydrocarbon burn, the o i l meter did not function properly and assessment o f heat recovery was made only over the f i r s t two days of burning. As near as can be e s ta b lis h e d , approximately 6$% of the heat of the ch lo rin a te d hydrocarbon was recovered (Table 12). T h is can be expected sin ce v o la t iliz a t io n of potassium ch lo rid e required energy. While the energy is recovered upon condensation, t h is occurs too f a r along the k iln to be completely recuperated. Weber [18] s ta te s " v o la t iliz a t io n of a lk a lie s consumes high grade heat in the s in te rin g zone and c a lc in in g zone above a m aterial temperature of 800C and t h is heat is subsequently released only at lower tem peratures. Thus, degradation of high-grade heat takes place". When ft Is considered that potassium ch lo rid e forms a c y c le w ith in the k iln and may be v o la t iliz e d several times before escaping to the p r e c ip it a t o r , the heat consumed by t h is process becomes co n sid erab le. fn the suspension preheater k i ln , an in d ica tio n o f the number of c y c le s of v o la t iliz a t io n o f a lk a li ch lo rid e s is given by the r e la t io n ship between the Stage IV ch lo rid e leve l and quantity of ch lo rid e input. Since In t h is stu d y, the K^O at Stage IV was a fa c to r of ten g re ater than the K^O Input, It fo llo w s that ten c y c le s of v o la t iliz a t io n occur on average in the suspension preheater k iln . The situ a tio n in the wet process k iln , however, i s d is s im ila r . There Is no convenient method fo r determining the number o f a lk a li c y c le s involved. That there w ill not be as many c y c le s as In the suspension preheater k iln is c e rta in when consid eration i s given to the d iffe re n c e s in the two p ro cesses. It Is assumed, for purpose of illu s t r a t io n , that a chlorinated hydrocarbon with 9300 Btu/lb and k2% ch lo rin e is being burned w ith an a lk a l i cy c le of three tim es. F<>r each pound of ch lo rin ate d m a te ria l, there are Q . k 2 pounds o f c h lo rin e which w ill produce 0 .8 8 pounds of potassium ch lo rid e . The heat of vaporization of potassiun ch lo rid e is 38,8k0 cal/mol [19] or 938.6 B tu /lb . The 0.88 pounds of potassium. A&* I -13 .IMl !|S 3 Iu I ' * ; 1`i tf r i 38 TABLE 12. RECOVERY OF Btu FROH CHLORINATED HYDROCARBONS M aterial A liphatie Aromatic + Complex PCB Average Reduction Input With 6 0 il (10- x Btu/ton Cl inker) 0.125 0.128 0.157 Average Input With Chlorinated Hydrocarbon (10^ x Btu/ton Cl inker) 0.205 0.217 0.228 % Useful Heat From Chlorinated Hydrocarbon 61 59 69 ch lo rid e *uld req u ire 2478 Btu fo r three c y c le s . On t h is b a sis we might expect to recover: 9300 - 2478 9300 x 100% 73% o f the heat content o f the ch lo rin a te d hydrocarbon. W hile,th e above Is only an approximation and ignores concurrent heat exchange processes, It in d icates that the heat recovery of about 69% obtained in the present study is reasonable. I t should be emphasized that the add ition o f calcium c h lo rid e would req u ire a s im ila r q uan tity o f heat to v o l a t i l i z e a l k a l i e s . In the case o f calcium c h lo rid e , an In crease In fuel q u an tity would be req u ired . 6.4 Cement Qua 11 tv While the lit e r a t u r e a v a ila b le [2 , 3 10] In d ica te s that a lk a li reduction Is b e n e fic ia l w ith regard to cement q u a lit y , the p o s s ib ilit y o f unburned organic m aterial being retained tn the cement and having a d e le te rlo u t e f f e c t was considered. Cements were ground in the laboratory w ith c lin k e r s produced*whl1e burning only No. 6 f u e l .o i l as w ell a s c lin k e r s produced during the arom atic plus complex and PCB bum s. Detailed re su lts are given in Appendix D. The re su lts obtained Indicated v the only e f f e c t s to be those duevto a lk a l i reduction. 6 .5 Extrapolation to Other K lin Types Since a l l cement k iln s have In convnon the requirement that uniform ly high temperatures be m aintained, the authors b e lie v e that At* CCbI^3 39 ch lo rin a te d hydrocarbons w ill be destroyed in a l l types o f cement k iln s re g a rd le ss o f fuel used. The only** q u a li/f ic a t io n to t h is statement is that the wastes must be in jected into the burning zone. V h ile i t is a d v isab le to atomize the ch lo rin ate d w aste, there appeared to be no -d ifferen ce during the PCB burn w ith or without a nozzle fo r atom ization. For fuel economy, cement k iln s a re operated at low excess oxygen. T h is , combined w ith the high temperatures and hydrogen from the f u e l, w i l l In a l l cases ensure that hydrogen ch lo rid e (HC1) Is preferent i a l l y formed rath e r than fre e c h lo rin e . In a l l cement k i ln s , the 1inie w ill r e a d ily react w ith the hydrogen c h lo rid e . Based upon r e s u lt s o f the present stud y, no d etectab le q u a n titie s o f e ith e r compound a re expected to be emitted from any type o f cement k iln . Sin ce a lk a l i c h lo rid e is v o la t iliz e d and condensed in the gas stream as ex trem e ly-fin e p a r t ic le s . I t would be expected that em ission o f p a rt ic u la t e m atter would in crease by an amount depending on the e f f ic ie n c y of the p r e c ip it a t o r . T h is would apply to any method of c h lo rid e a d d itio n . To overcome the plugging problem caused by condensation of these s a lt s in suspension preheaters, design m odifications to these u n its need to be in s t a lle d and demonstrated. 6.6 Comparison of Cement K iln Burning w ith Other Uses and Disposal Methods fo r Waste Chlorinated Hydrocarbons Information on disposal methods used and actual q u a n titie s of chlorinated hydrocarbons requiring disposal is not re a d ily a v a lla b le in Canada. While a co n servative estim ate o f 25^30 m illio n pounds of ch lo rin ate d hydrocarbon wastes was obtained, i t seems l ik e ly that t h is q uan tity rep resents only wastes from p lan ts manufacturing ch lo rin ated hydrocarbon products. The magnitude o f the problem of disposal of these w astes In North America can be in fe rred by some of the methods used. One method makes use o f ship s designed for burning o f these wastes a t se a , t h is method being described in stu d ie s monitored by the United States Environmental P ro tectio n Agency [20, 21J . Such a method o f disposal Is expensive, requires constant monitoring of temperatures w ith in the fu rn ace, uses a d d itio n a l fuel fo r combustion and em its hydrogen ch lo rid e (HC1) which Is d isso lv e d In the ocean. 4 *0* AO Another .method o f.d isp o sa l in North America is in cin e ra tio n w ith production of hydrochloric acid [22]. _ One report from Europe [23} in d ica te s that methods of disposal .are: 1) ille g a l dumping of small q u a n titie s In b a rr e ls or other co n ta in ers on uncontrolled refuse dumps; 2) d eposition o f la rg e r q u a n titie s in b a rre ls on refu se dumps which are supposedly sanctioned for th is purpose; 3) combustion In sim ple f a c i l i t i e s without hydrogen c h lo rid e scrubbing; A) combustion of p ile s of b a rr e ls on remote beaches w ith an offshore wind; 5) dumping b a rre ls on the open se a ; 6) dumping liq u id s into the sea from moving v e s s e ls ; 7) separation o f waste m a te ria ls and recovery of useful components; 8) combustion w ith recovery of hydrochloric a c id ; and, 9) combustion on the open sea a t temperatures guaranteeing almost complete p y ro ly s is . The same report s t a t e s : 1(0nly the la s t three procedures can be considered to be not harmful fo r the environment. But procedures 7) and 8) a re p o ssib ly very expensive and In sp e cia l cases unsustainable for the producer". A lso from the same re p o rt, ft is pointed* out that combustion at sea requires extensive observation of a variety of safety procedures. A study from France [2A] in d ica te s that a v a r ie ty o f legal and Ille g a l means a re used to dispose o f ch lo rin a te d hydrocarbon w astes. Of ille g a l means: "th e discharge of what In general are In s ig n if ic a n t q u a n t it ie s I s disposed o f In drums, or by tan kers, Into waterways, former a q u a r r ie s now used fo r other purposes, or discharged w ith unsupervised w astes that reason suggests should be retained ". In comparing combustion in a cement k i ln , w ith other methods V o f d isp o sal which are considered not harmful to the environment, the follow ing p o in ts become apparent: - In cin e ra tio n of these wastes Is normally done a t a flame ACfc GC e 1 5 Jw kih A i'. V .i*i 41 temperature o f 1200C to 1560C [25] w h ile cement k iln flame temperatures are 2100C^[l8] or higher. The reten tio n time In a cement k iln flame envelope is consid er ably longer than the 0.1 seconds normally found in an In c in e ra to r. For the production of cement c lin k e r , the temperatures cite d are necessary [2 6 ], thus removing the n e c e ssity fo r constant monitoring o f temperatures as required when burning in an in c in e ra to r. To prevent operating d i f f i c u l t i e s , such as k iln r in g s , In a cement k iln , the amount of ch lo rin e Is r e s t r ic t e d to approximately 0.4 percent re la tiv e to c lin k e r. Besides m aintaining constant temperature, th is requirement ensures that s u f f ic ie n t hydrogen is a v a ila b le to form hydrogen ch lo rid e which Is re a d ily absorbed by lim e. There is always a high q uantity o f lime In the cement k iln to react w ith hydrogen ch lo rid e and thus prevent em ission o f t h is compound to the atmosphere. Burning o f these wastes in a cement k iln saves f o s s i l f u e ls , as opposed to the n e c e ssity o f using f o s s il fu e ls to ensure combustion of these wastes In an In c in e ra to r. B e n e fic ia l use Is obtained In a cement k iln of p e rsiste n t and to x ic waste m a te ria ls which normally req uire d isp o sa l. \ CCcl**6 AC* kl CONCLUSIONS . -The concept of using ch lo rin ate d hydrocarbon wastes in cement processing d e riv e s from knowledge of k iln operating temperatures and residence times tn comparison w ith in cin e ra to rs capable of destroying th ese compounds. The a ctio n of cement k iln s as "dry-lim e scrubbers" in r e l if i o n to halogens .in the k iln gases was alread y known. In a l l c a se s, In the present study, an alyses of k iln em issions ind icated the e f f ic ie n c y of combustion of chlorinated hydrocarbon wastes to have exceeded 9 9 . 9 8 p ercen t. Traces of v o la t ile Tow m olecular weight ch lo rin a te d hydrocarbons were found to be present at a maximum em ission concentration of 50 ppb above b a se lin e . A ll other s t a r t in g m a te ria ls , including p olychlorinated b ip henyls, were completely destroyed. There were no d etectab le em issions of hydrogen c h lo rid e , fre e c h lo rin e or high molecular weight chlorinated hydrocarbons. ^An'increase In to ta l p a rt ic u la t e em issions was observed w h ile burning chlo rinated hydrocarbon w astes. Th is Is apparently related to an in cre ase In dust loading to the p re c ip ita to r as ind icated by the n e c e ssity o f d iscard in g more dust w h ile burning ch lo rin ated hydrocarbons. Combustion in in cin e ra to rs designed fo r d estru ctio n of ch lo rin ated hydrocarbon wastes caused the em ission of most of the hydrogen ch lo rid e except for those in sta lla tio n s using expensive hydrochloric acid recovery or scrubbing systems. These in cin erato rs require continous monitoring of temperature p r o f ile s and use f o s s i l fu e ls to i n i t ia t e or m aintain combustion. The present study In d ica te s that useful recovery o f about 65 percent o f the heat value and approximately 100 percent of the ch lo rin e is a tta in e d by the burning of these wastes In a cement k iln . ^ A m a s s balance on c h lo rin e and the e f f e c t iv e a lk a li reduction ta derived from the c h lo rin e contained in the ch lo rin a te d hydrocarbon wastes confirms the a ir emissions data. C onsideration of the data from t h is study and examination of the general lit e r a t u r e on cement manufacturing has led the authors to conclude that a l l ch lo rin a te d hydrocarbon wastes may be used in cement k iln s without adverse e ffe c t on a i r p o llu tio n le v e ls . jtri )( : AC* -GC81<i7 *3 REFERENCES ,* / / 1. Personal communication, January 15 1975 2. N ie l, E.M.M.G., "The Influence of A lk a li-Carbonate on the Hydration o f Cement", Proceedings o f the F if t h Internatio nal Symposium on the Chemistry of Cements, Tokyo, 1968, Published 1969. 3. HcCoy, W .J. and O .L. Eshenour, " S ig n ifica n ce o f Total and Water Soluble A lk a li Contents o f Cement", Proceedings o f the F if t h In tern a tio n al Symposiun on the Chemistry of Cements, Tokyo, 1968, Published 1 9 6 9 . A. S tcin o u r, H .H ., ^ h e S ettin g of Portland Cement, A Review o f Theory, Performance and C ontro l". Portland Cement A sso cia tio n , Research Department B u lle tin 9 8 , 1958. 5- Lerch, W. "Studies of Some Methods of Avoiding the Expansion and Pattern Cracking Associated With the A lk a li Aggregate Reaction". Portland Cement A sso cia tio n , Research Department B u lle t in 31, 1950. 6. Powers, T .C . and H.H. S te in o u r, "An In terp retatio n o f Published Researches on the A Ika1 -Aggregate Reaction: P art 1 - The Chemical Reactions and Mechanism of Expansion; Part 2 - A Hypothesis Concerning Safe and Unsafe Reactions with R eactive S i l i c a In Concrete". Portland Cement A sso cia tio n , Research Department B u lle t in 55, 1955. 7. Annual Book of ASTM Standards, Part 13 (1975)- American S o ciety for Testing and M a te ria ls. 8. Lerch , W. *The Influence of Gypsum on Hydration and P ro p erties of Portland Cement P a ste s", Portland Cement A sso cia tio n , Research Department B u lle t in 12, 19^6. a 9 . Woods, H. "Reduction of A lk a lie s In Cement Manufacture", M ill Session Paper M-149, Portland Cement A sso cia tio n , Manufacturing Process Department, 1956. \ 10. Woods, H ., J . L . G i l l i la n d , J r . , J . F . W eigel, B .E . R ester, and H.A. Stevens, "Symposium on A lk a li Removal and Problems". Regional F a ll Meeting of General Technical Committee, PCA, Milwaukee, Wisconsin, AbM CCeiHc 44 Sept. 21-24 1959- M ill Session paper M-158, Portland Cement A sso cia tio n , Manufacturing Process Department, I960. 11. C oles, C .V . and D.G. Da inton, " S t. Lawrence Cement Co. Clarkson P la n t" , Cement Technology 1 ( 2 ) , 43, 1970. 12. Herod, B .C . " S t. Lawrence Cement Expands Clarkson O peration", P it and Quarry, J u ly 1968. I3> Schroth, G .A ., "Suspension Preheater System Consumes Less F u e l" , Rock Products 75 (5 ), 70, 1972. 14. G a rre tt, H.M., "The P o ten tial Promise - Prospects and P i t f a l l s in Energy Conservation by the U .S. Cement In d u stry", Paper presented a t the Cement Chemists Seminar, Portland Cement A sso cia tio n , Lincolnwood, 111., February, 1976. 15- T r a u ffe r , W.E. "Portland Cement Outlook and Review". P it and Quarry, January 1976. 16. Campaan, H ., Central Laboratory TNO Report No. CL 74/93* "On the Occurrence o f Organic C hlorid es In The Combustion Products o f an EDC Tar Burnt by the In cin e ra to r Ship Vulcanus1; A P re lim in a ry In v e stig a tio n " . O ct. 1974. 17. Gagan, E.W. "A ir P o llu tio n Em ission and Control Technology - Cement Industry", Environment Canada, Environmental Protection S e rv ice , Economic and Tech nical Review Report EPS 3"AP-74-3, 1974. 18. Weber, P. (T r a n s la tio n ), " A lk a li Problems and A lk a li Elim in a tio n in Heat-Economising Ory-Process Rotary K iln s " , Zement Kalk G ip s, (B ), 1964. 19. P e rry , J .H . , C.H. C h ilto n , and S.D . K irk p a tric k , Chemical Engineers Handbook, 4th E d it io n , McGraw-Hill. 20. Marine Environmental Monitoring of "Vulcanus", Research Burn 11, December 2-10, 1974, P re lim in a ry Report, U .S. Environmental P ro tectio n Agency, December 10, 1974. 21. Badley, J .H . , A. T e lf e r , and E.M. F re d e ric k s, "At-Sea In cin era tio n o f Sh ell Chemical Organic Chloride Waste". Technical Progress Report BRC - Corp. 13-75-F, S h ell Development Company, A p ril 1975- 3 45 22. American S o ciety of Mechanical Engineers, Research Connttee on In d u stria l Wastes Report. . l/ ic negation o f Chlorinated Hydrocarbons w ith Recovery o f HC1 a t E. I . dir Pont de Nemours t C o ., ( In c .) L o u is v ille , Ky. ASHE Industry Survey, Present S ta te o f the A r t , $ Disposal of In d u stria l Wastes by Combustion, January 1971. 23* G rassh o ff, K. (T ra n sla tio n ) " E x p e rtise Regarding the E ffe c ts of the . High Temperature Combustion of Chlorinated Hydrocarbon on Special V e ssels at High Sea s", 1973- 24. " In c in e ra tio n o f In d u stria l C hlorine Wastes on the High S e a s" ., Report from the Environmental Agency (M inistre charg de l'Environnement) o f the P o llu tio n and Nuisance Prevention A dm inistration (D ire ctio n de la Prevention des P o llu tio n s e t N uisances), France. 1974. 25. United States Environmental P rotection Agency, Permit No. 730 D008C (3) to Sh ell Chemical Company, In c. and Ocean Combustion S e r v ic e s , 6 .V . December 12, 1974. 26. Peray, K .E ., and J . J . Waddell, The Rotary Cement K iln , Chemical Publishing Co. Inc. New York, 1972. 27. Reynolds, L.H ., "P esticid e residue a n a ly sis in the presence of Polychlorobiphenyls (PCB's) Residue Reviews, 34, 27 1971- t 28. S a n to le ri, J . J . "Chlorinated Hydrocarbon Waste Recovery and P o llu tio n *t Abatement" Chem, Eng. Prog. 69 (1) 68, 1973. F v E i fit* 1I * i I 46 ACKNOWLEDGEMENTS The authors wish to thank and to acknowledge the p a rtic ip a tio n o f the follow ing people and o rg an izatio n s in t h is work. Environment Canada Hr. W.A. Neff (EPS/WPCD) fo r program coordination and invaluable a s s is ta n c e in e s ta b lis h in g d e t a ils o f the program management. Hr. J . Robert (EPS/APCD) fo r helpful advice In sampling proce d u re s, an alyses performed and review of r e s u lt s . Ontario M in istry of the Environment Dr. F. F ra n tisa k and h is s t a f f for a id in planning the program, . a n a ly t ic a l a s s is ta n c e and review o f r e s u lt s . Energy Hines and Resources Canada Dr. F.D. F r le d r ic k fo r a ss is ta n c e a t program planning stag e. S t . Lawrence Cement Co. Hr. L. Kraszewskl for program coo rd ination. U .S. Environmental P ro tectio n Agency A.W. Lindsey and J . Schaum (Hazardous Waste Management D iv isio n ) fo r program planning coordination and review o f r e s u lt s . TRW s Systems Group and Control P o llu tio n S e rv ice s In c ., who were contracted by EPA. \ ACM Cc* i5 j APPENDIX A QUANTIFYING, SAMPLING AND ANALYSIS OF PROCESS MATERIALS CCe 15b ACM f '1 ? is f . i \ V *9 APPENDIX A QUANTIFYING SAMPLING AND-ANALYSIS OF PROCESS MATERIALS While a n a ly t ic a l procedures used ere id e n tica l for both wet and dry process k i ln s , only the wet process k iln m aterial streams are considered in t h is Appendix. The dry process suspension preheater k iln system Is d e ta ile d In Appendix 6. The m aterial flow through the wet k iln production system I is illu s t r a t e d in Figure A .1. To obtain the m aterial balance, s ix m a te ria ls were q u a n tifie d , sampled and analyzed. A seventh m a te ria l, p r e c ip ita to r return d u st, was monitored to obtain any ad d itio n al operating data t h is stream might y ie ld . Table A.1 l i s t s the m a te ria ls examined with approximate r e la t iv e q u a n titie s under normal production co n d itio n s. TABLE A .1 . PROCESS HATERIALS STUDIED AND APPROXIMATE NORMAL PRODUCTION QUANTITIES M aterial Approximate Quantity S lu rry Feed 1540 dry tons/day Cl inker 1000 tons/day P re c ip ita to r Discard Dust 0 - 140 tons/day P re c ip ita to r Return Dust 350 tons/day No. 6 Fuel O il 20 gal /min Chlorinated Hydrocarbons 1 - 2 gal/min K iln Exhaust Gases 160,000 ACFM A .1 Q uantifying and Sampling S lu rry Feed and C lin k e r There a re four s lu r r y b a sin s, each o f 6000 ton ca p a city , _ Includ ing s lu r r y w ater. The product from the s lu r r y grinding m ills is ptanped Into these b a sin s. The basins a re supplied w ith a i r a g ita to rs fo r the purpose o f blending th^ s lu r r y and maintaining a uniform-suspension o f s o lid s In liq u id . S lu rry Is pumped from one basin at a time to the constant leve l box o f the s lu r r y feed system. The m aterial which o ver- CCtjJ.53 ( - - l von n -'Ci A KILN B PRECIPITATOR 5 CLINKER 6 PRECIPITATOR DUST WASTEO 7 STACK GAS SCHEMATIC OF THE MATERIAL BALANCE ,V, ciV F I G U R E A. 1 '- 1 ' I |.lM 51 flow s from the constant level box is retu rn ed .to the s lu r r y b a sin s. S lu r ry feed fo r the k iln Is taken from ,the constant level box by an A l l i s Chalmers " F e r r is Wheel" bucket wheel feeder with varTable speed d riv e . From the bucket wheel feed er, the s lu r r y flows Into a small basin with an o u tle t a t the bottom from which the s lu r r y goes d ir e c t ly Into the k iln . The s lu r r y feed samples were taken every two hours from the overfiow o f the constant level box, and blended into a d a ily composite. i A portion of the d a lly composite then was dried a t 120C fo r a n a ly s is . \ S lu rry feed q u a n titie s were obtained by m ultiplying the to tal number of revo lu tio n s o f the " F e r r is Wheel" per day by a f a c to r . Measure ment of the s lu r r y basin le v e ls w h ile feeding the k iln but not allow ing s lu r r y to be fed into the basin g ives an accurate quantity o f s lu r r y fed to the k i ln . A second method, performed every two hours, is to c lo se a va lve at the o u tle t to the k iln o f the small basin fed by the F e r r is Wheel. By determining the time to f i l l t h is b a sin , the q uantity fed to the k iln is known. ! The c lin k e r Is fed to a bucket ele v a to r by g ra v ity through a chute from the co o le r. C lin k e r samples a re taken from t h is chute a t in te rv a ls of two hours to form a 24-hour composite. The c lin k e r composite sample is mixed, reduced by "cone and q uartering" and a portion ground for analysis. In conmon w ith general p ra c tic e in the in d u stry , there is no p rovision in the p lant fo r continuous weighing of c lin k e r . The water content of the s lu r r y feed is determined th ric e d a ily , from which the follow ing c a lc u la tio n Is made: f s where SFD - quantity of s lu r r y feed on a dry b a sis SF - quantity of s lu rry feed Including water & H^O " % water in the s lu r r y feed. To obtain the q uantity of c lin k e r produced, the CO^ which is lo s t in the burning process (determined by lo ss on ig n itio n o f the dry S s lu r r y feed) is deducted and a fu rth e r co rre ctio n is made fo r the q uantity ! r! A c* CC6155 52 o f dust discarded from the system. A to tal m aterial balance may then be w ritten as fo llo w s: "/ Cl inker -where SFD q u an tity o f s lu r r y feed on dry b a sis LOI SFD X lo ss on Ig n itio n on s lu r r y feed on dry b a sis DD q uantity o f dust discarded LOI DD X lo ss on ig n itio n on dust d iscard ed . A .2 Weighing and Sampling P re c ip ita to r Dust Dust from the p r e c ip ita to r is separated into two portions by a system of screw conveyers under the p r e c ip ita t o r . The major portion i s returned v ia a conveyor b e lt Into a small holding s i l o from which the' dust is returned Into the burning zone o f the k iln (return d u s t). Samples o f the dust a re taken a t two-hour In te rv a ls from the conveyor b e lt to form a 24-hour composite. The composite i s blended and a portion taken for an aly sis'. A minor portion of the dust (d iscard dust) is fed into a second holding s i l o from which tru ck s are loaded d ir e c t ly . A sample o f t h is dust Is taken from each truck load, and a l l samples fo r each day a re blended Into a 24-hour composite. Each load o f dust Is weighed on the truck weigh .sca le s before being disposed o f. A .3 Measurement and Sampling Wo. 6 Fuel O il While I t was considered u n lik e ly that No. 6 fuel used in the p lan t would co n trib u te a su b sta n tia l quantity o f c h lo rin e , samples of o i l were taken d a lly . Fuel q u a n titie s a re continuously monitored and recorded In the production data but a re not given ir. t h is rep o rt. C hlorine content o f the o i l was determined, and the co n trib u tio n to the c h lo rin e a a ss balance due to o i l was Included. A .4 Measurement and Sampling Waste Chlorinated Hydrocarbons Sampling o f ch lo rin a te d hydrocarbons was c a rr ie d out by wi t h d r a wi n g ^ m aterial from the feed system tw ice d a lly and blending by vigorous m i x i n g . Samples were S p lit Into equal portions a fte r blending. ACM CCfilSt 53 I n i t i a l l y , I t was intended to meter ch lo rin ated hydrocarbon flow continuously. For t h is purpose, a magnetic flow meter was in s t a lle d In the lin e to monitor q u a n titie s o f these m a te ria ls. For th is type of meter to function p rop erly, It is required that the cond uctivity of the m aterial be greater than 2 x 10~^ mhos. While t h is was the case fo r the chlorinated a lip h a tic w astes, the co n d u ct!v lties of the other m aterials were le s s than 0.3 x 10 ^ mhos. As a r e s u lt , ch lo rin ated waste m aterial q u a n titie s were determined by taking measurements o f the volume of m aterial In the storage tank. Tables A .2 through A.it give d etailed information on ca lcu la te d ch lo rin a te d waste m aterial q u a n titie s . A .5 Emission SamplIng Separate sampling t ra in s were used to c o lle c t rep re sen tative samples o f k iln em issions for a n a ly s is to determine the emission ra te s o f p a rtic u la te m aterial and to determine i f organic ch lo rid e compounds were present in the gas stream passing to the sta c k . P a r tic u la te and gaseous samples were c o lle c te d from a duct downstream o f the p r e c ip ita tor c o n tro llin g dust em issions from k iln #1. The lo catio n is described below, followed byd e scrip tio n s o fsampling equipment and methodology. A .5.1 Sampling lo catio n Em issions from both #1 and 2 wet processk iln s pass through e le c t r o s t a t ic p re c ip tta to rs (Joy Manufacturing Co.) and then through se ctio n s of rectangular breeching before mixing in a axqnon sectio n of duct entering the s ta c k . The sta c k , which Is o f height 554.0 fe e t and has an e x it diameter of 13*0 f e e t , vents the exhaust gases to the atmosphere a t a temperature of about 400F and a v e lo c ity o f almost 40 fe e t per second. The rectan g u lar breeching from the p re c ip ita to r makes a 90 a bend and then angles a t about 30 upwards from the horizontal for a d ista n ce o f approximately 45 fe e t to the common header which leads into the sta c k . T h is 45 foot sectio n o f duct was considered the most su ita b le fo r In s t a lla t io n o f sampling ports a t a locatio n which would meet both Federal and P ro v in c ia l source te stin g codes. I t was subsequently decided, therefore, to in s t a ll fiv e 4" diameter ports in the v e rtic a l sid e of the duct at a p o sitio n 12 feet upstream of the bend into the i jj W li fjI*.' ii\ i.. l9l** ;l li, *iIuIJ *.i 54 TABLE A .2. QUANTITIES OF ALIRHATip MIXTURE BURNED DAILY . / Date Time From To Minutes Igpm Gallons Gellons/Day 23/10/75 09:00 12:00 24/10/75 06:00 25/10/75 06:00 26/10/75 06:00 27/10/75 06:00 28/10/75 06:00 29/10/75 30/10/75 06:00 06:00 09:30 31/10/75 06:00 12:00 06:00 06:00 06:00 06:00 06:00 06:00 06:00 09:30 06:00 06:00 1/11/75 2/11/75 3/H/75 4/11/75 0 6 :0 0 06:00 06:00 06:00 06:00 >06:00 06:00 10:00 10:00 17:30 OFF a t 17:30 180 1030 i44o 1440 1440 1 44o 1440 1440 210 1230 1440 1440 1440 1440 240 450 0.5 90 1 1080 1 1440 1440 1 1440 I 1440 1 1440 1 1440 210 2 2460 2 2880 2 2880 2 2880 2 2880 2 480 1 450 ' 1170 1440 1440 1440 1440 1440 1440 2670 2880 2880 2880 2880 930 i AC'h Ccei5e V 55 TABLE A .3. QUANTITIES OF AROMATIC PLUS COMPLEX MIXTURE BURNEO DAILY Date Time Tank Measurement / D iffe re n c e gallons minutes ' Gal /min Dec. 6 15:00 ' 12* 0" 2 loads In 60930 lb + 60690 lb 9 1. 299 SG - 7823 gal Dec. 7 Dec. 8 Dec. 9 12:15 15:00 15:30 14' 6" 14' 0" 13* 6" (7823-5886 - 1937 g al) 1937 1275 1177 1605 1177 1670 1.519 0.733 0.801 Dec. 10 12:00 12' 6" 2356 1230 1.916 Dec. 11 15:00 11' 8" 1962 1620 1.211 Dec. 12 Dec. 12 Dec. 13 09:00 15:30 13:30 11 4" I P 0" 10' 3" 785 785 1373 1080 390 1320 0.727 2.013 1.040 Dec. 14 13:20 9' 6" 1766 1630 1.235 Dec. 15 14:00 ` 8* 8" 1962 1480 1.326 Gal Ions per Day Date ' From Dec. 2 0 6 :0 0 Dec. 5 10:00 14:00 Dec. 8 06:00 Dec. 7 06:00 12:15 Oec. 8 06:00 15:00 Dec. 9 0 6 :0 0 15:30 Dec. 10 0 6 :0 0 12:00 To 13:10 12:00 06:00 06:00 12:15 06:00 15:00 06:00 15:30 06:00 12:00 06:00 Minutes 430 120 960 1660 375 1065 540 900 570 870 560 1080 Igpm (1) 1.519 1.519 1-519 1.519 0.733 0.733 0.801 0.801 1.914 1.914 1.211 Gal Ions 430 182 1658 2187 570 781 396 721 456 1665 689 1308 Total Gallons 430 1640 2187 1351 1117 2121 1997 ACM CC6159 i ii1 \ I ? 1iH 11! I8; !t P: H Ul 1i; w !!; !-ii- II!! i1n1; ri 56 TABLE A .3. (Cont'd) Date Dec. 11 Dec. 12 Dec. 13 Dec. 14 Dec. 15 From 06:00 15:00 06:00 09:00 15:30 06:00 13:30 06:00 13:20 06:00 To 15:00 06:00 09:00 15:30 06:00 13:30 06:00 13 :2 0 .06:00 06:00 Minutes - Igpm 540 900 180 390 870 450 990 440 1000 1440 1.211 0.727 0.727 2.013 1.040 1.040 1-235 1-235 1.326 1.326 Gallons 654 654 131 785 905 468 1223 543 1326 1909 Total Gallons 1308 1821 1691 1869 1909 ADH CC616C 57 TABLE A .4. QUANTITIES F PCB MIXTURE BURNED DAILY 1 Tank D iffe re n c e Date Time Measurement g allo n s minutes Gal/min 3/1/76 4/1/76 5/1/76 5/1/76 6/1/76 7/1/76 7/1/76 8/1/76 8/1/76 9/1/76 9/1/76 14:30 12: 00 09:15 14:30 09: 10 09:06 16:00 08:45 15:45 08:45 05:45 ' I P 4" 12' 6" 9 ` 8" 9* j " 8' 7' 7'11" 7* 10" 7' 9" r 8" 7' 2" 6' 8" 1962 1962 392 1766 1962 196 196 196 1177 1177 1290 1275 315 1120 1436 4l4 1003 420 1020 540 1.521 1-539 1.244 1.577 1 .3 6 6 0 .4 7 3 0.195 0.467 1.154 2.180 Gal Ions per Day Date From To Minutes Igpm Gal Ions Total Gallons 3/1/76 07:45 14:30 405 1.521 616 2030 14:30 06:00 430 1.521 1414 4/1/76 0 6 :0 0 12:00 360 1.521 548 . 221Q 12:00 06:00 1080 1.539 1662 5/1/76 6/1/76 06:00 09:15 06:00 09:10 09:15 14:30 09:10 06:00 195 315 190 1250 1.539 1.244 1.577 1.366 300 392 300 1708 " 2159 2008 7/1/76 06:00 09:06 186 1.366 254 614 09:06 16:00 a 414 0.473 196 K 16:00 06:00 840 0.195 164 t 8/1/76 06:00 08:45 165 0.195 32 1215 9/1/76 08:45 ' <5:45 06:00 08:65 15:45 06:00 08:45 18:15 420\ 855 165 570 0.467 1.154 1.154 2 .18 0 196 987 190 1243 1433 1 2 ! ii3 s-J u 3; ji I- : "I* f* tiu Hr * f 'a; 58 header, and positioned on a lin e perpendicular to the gas flo w . The port lo catio n s are shown sch e m atica lly ''in Figure A .2. As agreed upon by the O ntario M in istry o f Environment and Environment Canada, sampling was ca rrie d out a t eig ht points on each o f the f iv e duct tra v e rse s fo r a minimum of f iv e minutes per p o in t. The - sampling points were located at the centres of equal rectangular areas in the duct as shown in Figure A.3> The numbers given in each rectangle a re ty p ica l gas v e lo c it ie s measured during the study and in d ica te the d istrib u tio n of gas flow at that lo catio n. A .5-2 Sampling equipment P rio r to se ttin g up equipment fo r p a rtic u la te and gaseous sampling, grab samples of the k iln em issions were taken using the equipment depicted in Figure A .4. A gas sample was p u lled through an in -sta ck 47 nvn g la ss f ib r e f i l t e r in a s t a in le s s ste e l holder to remove p a r t ic u la t e m aterial and then through a heated teflo n lin e into a Ted lar bag. For each bag sample a clean bag was placed In the lung and the li d tig h ten ed 'to e f f e c t an a ir - t ig h t s e a l. Valves A and C were closed and B opened so th a t, w ith the pump on, the sampling lin e was purged. Valve B was then closed and valve C opened to evacuate the sampling lung, making sure that the vacuum did not exceed 5" Hg. At a vacuum o f 4" Hg, va lve A was opened and the sample bag f i l l e d with em issions from the duct. Valve A was then closed and valve B opened. The pump was shut o f f and the lin e to the pump disconnected. Valve C was opened to bring the lung to atmospheric pressure and the bag was removed from the drum and q u ick ly capped. Each bag sample was transported to the labo ratory a t ORF for a n a ly s is on the same day that the sample was co lle c te d . A .5.2.1 P articu late train . i Samples o f p a rt ic u la t e em issions were c o lle c te d using a Joy Manufacturing Company EPA (Model CU-2) t r a in , .shown sch e m a tica lly In Figure A .5- T h is equipment conforms to that recommended by both the O ntario Source Testing Code [ A .!] and the Environment Canada Code [A .2 ]. P rio r to each te st a v e lo c ity tra v e rse was conducted a cro ss the duct through each part to determine an average gas v e lo c it y . Gas temperatures were a ls o recorded a t each sample p oint. ACM ccaifcZ -'3 60 108.9 - 110.6 x' X 118.8 X 118.6 . 117.6 X 'X 114.7 X 107.8 X 103.2 'X 92.5 X __ 107.8 X J 12.0 X 113.3 X 110.6 X 110.6 X 108.9 X 107.8 X 7 8 .2 X 89.2 X 9 2 .6 X 91.9 X 94.5 X 100.2 X 102.0 X 104.9 X 72.5 X 74.2 X 7 8 .2 X 7 8 .2 X 8 2 .0 X 9 1 .2 X 97.7 X 97.7 X 42.8 X 51.9 X 66.4 X 73.4 X 7 8 .2 X 80.5 X 74.8 X 80.5 X V elo city In feet per second ,1 1 FIGURE A .3. GAS FLOW DISTRIBUTION AT SAMPLING POINTS 0An Orsat ana ly sis'w a s made to find the concentrations of C0`, CO^, ^ and In the gas stream and a gas moisture determlnatlon^was carried out using an estab lished procedure [A .1]. With the prelim inary data obtained, the Iso k in e tic sampling rate was calculated using knom standard equations lA .1]. For each p a rtic u la te t e s t , sampling was made for fiv e minutes a t each of the eight points across the duct through t h e .fiv e ports. Sp e cifie d data [ A . l j were recorded every 2 .5 minutes on a te st data . Sheet. The high flow rates encountered at the fa r w all o f the duct end across the top o f the duct necessitated use of probe nozzles -of - diameter le s s than 0.25 Inched. Nozzles of diam eter.0.219 inches and 0.188 Inches were used during c e rta in t e s ts . At the conclusion of a te st the g lass fib re f i l t e r m s / \ removed from I t s holder end placed In a labelled p e trl d ish . The v o lu m of water in the Inplngers m s measured and the contents tra n s fe rre d CCt -r* i 61 FIGURE A.4. GRAB BAG SAMPLING EQUIPMENT c 0 8Vl>5 62 PROBE CP= HEATED AREA n STACK WALL ,,FILTER HOLDER THERMOMETER CHECK VALVE REVERSE-TYPE pitot Tu b e v pitot r MANOMETER (OPTIONAL) 6C IMPINGERS ICE B A T H VACUUM ( v FIGURE A.5. PARTICULATE SAMPLING TRAITI /y ACM ccaiec to polyethylene b o t t le s , which were then la b e lle d . The probe and nozzle were rinsed w ith d Is t ilie d - w a te r as the in sid e o f the probe was c a r e f u ljy brushed. The rin s in g s were c o lle cte d in a polyethylene b o ttle . The glassware between the end of the probe and top of the f i l t e r holder was washed w ith d i s t i l l e d water and these washings added to the probe r in s e . The bottom-of the f i l t e r holder and glassware between f i l t e r and lmpingers was a ls o washed and the contents added to the impinger catch. A fter f il t e r i n g the probe r in s e , drying and weighing the in so lu b le p a r t ic u la t e m a te ria l, weighing the reconditioned f i l t e r , and determining the so lu b le portion of p a rtic u la te s in the probe rin se by aliquot evaporation, the to tal weight of co llected p a rticu la te m aterial was obtained. The m oisture removed from the sampled gas was found by adding the gain in weight of the s i l i c a gel to the volume of a d d itio n a l water c o lle cte d in the impingers. The volume of gas sampled was corrected to standard conditions and the p a rt ic u la t e loading and em ission rate c a lcu la te d by use o f appropriate equations [ A .1]. A .5 .2 .2 Gaseous t r a in . It was o r ig in a lly Intended to sample for gaseous organic ch lo rid e compounds using the EPA-type p a rt ic u la t e t r a in ,, w ith s p e c if ic absorbents rep lacing water in the impingers. Due to the high flow rates (0 .5 - 1.0 cfm) required fo r is o k in e t ic sampling, however, i t was considered that the c o lle c tio n e f f ic ie n c y of any absorbent for v o la t ile organic compounds would be very low. An experimental program was, th e re fo re , devised and c a rrie d out a t ORF to evaluate c o lle c tio n methods and develop a s u it a b le sampling t r a in for gaseous org anic c h lo rid e s. D e ta ils o f t h is stud y, performed under co n tract to the O ntario M in istry o f the Environment, a re given In Appendix H. Data obtained which are relevan t to t h is sectio n are described below. A p a r t ic u la t e t r a in was se t up In the laboratory and heated a i r was p u lled through the complete system at a flow r a te of 0.5 cfm. Water was placed in the f i r s t (mpinger and solvents such as x y len e, toluene, and decane, containing a few ppm o f CHCl^, CH^Cl^ or CHC^CI^CI, were placed In the second Impinger. The th ird Impinger was le f t empty and the fourth contained s i l i c a g e l. A fter a period o f three to four hours, the time a n ticip a te d for a p a rtic u la te t e s t , the impingers were 61* examined w ith resp ect to solvent volume and organic ch lo rid e concentra tio n . In e l l c a se s. I t was found'that' the lo ss o f ch lo rid e compounds was co n sid erab le. For the polar so lv e n ts, such as toluene or xylene, the ch lo rid e concentration was retained but the to ta l lo ss o f solvent p lu s c h lo rid e s was in the range of 50$-70$. For the non-polar so lv e n ts, su ch .as decane, the volume lo ss of solvent was n e g lig ib le but the ch lo rid e lo ss was in the range o f 60$-90$. Even I f e f f ic ie n t c o lle c tio n of tra c e organic c h lo rid e s from the gas stream by s p e c if ic so lven ts were p o s s ib le , these compounds would not be retained In so lu tio n at gas flow ra te s o f 0 .5 cfm or hig her. Recent stu d ie s have shown that c o lle c t io n and reten tio n o f tra c e org anic compounds i s p o ssib le using s p e c if ic adsorbent m a teria ls [A .3 ]. In ert m a te ria ls, such as the Chromosorbs, are considered to have ce rta in advantages over activ ated carbons In that water vapour present In the gas does not hinder organic compound adso rption, and adsorbed compounds a re r e a d ily removed by thermal d esorption. E x p e ri ments using the p a r t ic u la t e t r a in w ith Chromosorb 102 o r a ctiv a te d carbon In the th ird implnger were c a rrie d o u t, passing heated a i r containing a few ppm o f CHCl^ through the system a t a flow rate of 0.5 cfm. N either adsorbent removed more than 70$ of the CHCl^ i n i t i a l l y and, a f t e r about t h ir t y m inutes, almost a l l of the CHCl^ was passing through the system. I t was concluded, th e re fo re , th at e f f ic ie n t c o lle c t io n and re te n tio n o f v o la t ile organic compounds was not p o ssib le % w ith gas flow r a te s o f 0.5 cfm or higher and the p a rt ic u la t e t r a in could not be used fo r t h is purpose. A fte r fu rth e r stu d ie s In the laboratory w ith both Chromosorb 102 and a c tiv a te d carbon, the sampling t ra in fo r gaseous organic compounds depicted tn Fig u re A .6 was co n stru cted . T ests made w ith an a i r stream containing 15 ppm o f CHCl^ showed that a lt h e r adsorbent would remove e t t e r than 95$ of the CHCl^ over a sampling period of four hours. Chromosorb 102 was se le cte d on the b a sis o f the advantages p revio u sly mentioned. A kl itm f ib r e g la s s f i l t e r was placed d ir e c t ly behind the probe n o zzle to remove p a r t ic u la t e m aterial a t the duct gas temperature. The sample gas was passed through midget fmplngers containing water and ACW CC t i c e - & 65 : FIGURE A.6. GASEOUS SAMPLING TRAIN I ! ACM 0C8169 I *6 c a u s tic soda (5* s o lu t io n ), r e s p e c t iv e ly , to remove any HCl and Cl^ present in the k iln em issio n s. A fter the impinger, the gas was f ilt e r e d through fourXhromosorb adsorbent traps placed in p a r a lle l'a t a flow rate co n tro lled by rotam eters. A flow ra te o f 250 ml per minute was se le cte d fo r each tra p , g iving a to ta l flow of one l i t r e per minute through the " tra fn .- The to ta l flow over each t e s t period was a lso monitored by a wet t e s t meter placed a f t e r the pump. Impinger so lu tio n s and adsorbent traps were maintained a t Ice temperature to e ffe c t a better c o lle c tio n of c h lo rid e compounds. A fte r each te s t the in -sta c k f i l t e r and impinger so lu tio n s were stored in la b e lle d c o n ta in e rs. The adsorbent traps were removed, capped and la b e lle d . The probe was rin se d and the rin sin g s placed in a polyethylene b o ttle . A ll samples were then tran sferred d ir e c t ly to the ORF laboratory for a n a ly s is . The gaseous sampling t r a in was operated sim ultaneously w ith the p a rtic u la te sampling tra in for each t e s t . A .6 Organic Chloride Analyses The methodology used to c o lle c t k iln em ission samples, in b a se lin e or waste burn t e s t p erio d s, provided four d is t in c t types of samples for a n a ly s is o f ch lo rin a te d organic compounds. These sample types and the a n aly ses required were: . - chlorinated waste feeds for compositional a n a ly sis; - grab bag samples fo r any ch lo rin ated organic s p e c ie s ; - - Chromosorb adsorbent samples for volatITte low m olecular weight org anic compounds; end, - so lve n t e x tr a c ts o f f i l t e r s and so lu tio n s fo r noncombusted waste components. A .6.1 Sample preparation Waste feed. Samples o f the prospective VBA feed m aterial were supplied In advance of the te s t burn for a n a ly se s. When the t e s t period a r r iv e d , however, In s u f f ic ie n t q u a n titie s of some components were a v a ila b le to sake up the s p e c if ic blend of waste feed. The conposltlon of m aterial a c t u a lly d e liv e re d to SLC was, th e ra fo re , d iffe r e n t from the sample sup p lied to ORF In advance. In order to obtain a true compositional a n a ly s is o f the m aterial being supplied to the k i ln , and to determine ACH CC617C 67 feed co n sisten cy , a liq u o ts o f the waste feed were c o lle c te d on s p e c if ic */ sampling days during the burn. Crab bag. Em ission samples c o lle cte d in Ted lar bags were analyzed d ir e c t ly for gaseous ch lo rin ate d hydrocarbons by in je c tin g syring e samples taken from the bag Into a gas chromatograph. A fter d ire c t a n a ly s is , the contents o f each bag were drawn through a g la ss tube containing Chromosorb 102 adsorbent by means o f a small pump. Any trapped organic compounds were then therm ally desorbed from the adsorbent tube Into an evacuated gas j a r . The gas j a r contents were then subjected to GC a n a ly s is . Adsorbent tubes. Chromosorb 102, being re a d ily a v a ila b le in amounts required fo r t h is stud y, was se le cte d as a su ita b le adsorbent fo r low m olecular weight organohalides. Experiments were performed in the laboratory to confirm the s u i t a b i l i t y and e f f ic ie n c y o f t h is adsorbent for the c o lle c t io n o f v o la t ile organohalides from an a i r stream. Thermal desorption o f adsorbed components into an evacuated gas j a r was se le cte d as a method fo r preparing samples fo r GC a n a ly s is . I n i t i a l tubes were prepared with fre sh Chromosorb 102 from the b o ttle without co n d itio n in g . It was determined a f t e r the f i r s t b a se lin e te s t period that conditioning o f the adsorbent would be necessary to reduce the amount o f bleed produced on heating which.tended to produce excessive background noise - during GC a n a ly s is . ^ Adsorbent tubes were made up by packing Chromosorb 102 in g la ss co n tain ers 11.5 cm In length and 11.0 nm in sid e diam eter, using s ila n iz e d g la s s wool plugs a t e it h e r end fo r support. Tubes fo r the WBA and WBB burns ware preconditioned by heating to 200C and passing a stream of nitrogen a t 40 mlVain through them for four hours. Chromosorb 102 used In tubes fo r the VBC and BLB t e s t s was extracted with acetone in a Soxhlet fo r 18 hours and then treated a t 240C w ith nitrogen fo r 12 hours p rio r to use. vV Adsorbed components on the Chromosorb 102 a f t e r each t e s t were ` removed by thermal desorption Into an evacuated gas j a r o f 500 ml c a p a city . The adsorbent tube and gas J a r were connected by Teflon ccsn 1 AC* 68 tubing and the tube heated to a fix e d temperature o f 170C using heating tqpe control led. by a v a r ia b le transform er. When the fixed temperature was reached the stopcock' o f the gas j a r between the j a r and the tube was opened, and heating of the tube continued fo r a fu rth e r 15 minutes. The stopcock was then clo se d , the jar.rem oved from the connecting tubing and taken o u tsid e the laboratory to f i l l up with clean a i r . Previous stu d ie s o f therm al.desorption using t h is removal technique In the laboratory have shown that reco veries in excess of 90% may be expected fo r adsorbents w ith adsorbed organohalides. So lu tio n and f i l t e r e x t r a c t s . Pentane o r hexane was used as an e x tra c ta n t fo r org anic compounds from a l l f i l t e r s and probe rin s e in so lu b le f r a c t io n s , using a Soxhlet apparatus. The same so lve n ts were used to e x tra c t probe rin s e s o lu tio n s , and aqueous sodium hydroxide and Implnger s o lu t io n s , using a 1iquid/1quid p a rtitio n in g procedure. The e x tra c ts were then d ried over anhydrous ^$0^, f i l t e r e d , and concentrated by evaporation using a combination of ro tary and Kontes tube h e a te rs. The concentrate was made up to a known sm all volume w ith so lven t for GC a n a ly s is . Pentane was used as e x tra cta n t for BLA, WBA and VBB te s t sam ples, sin c e I t s high v o l a t i l i t y would minimize sample lo ss o f low m olecular weight compounds during evaporation. Hexane was used as e x tra cta n t fo r VBC and BLB t e s t samples. A .6 .2 Sample a n a ly s is A ll samples were analyzed by gas chromatography w ith flame Io n iz a tio n ( F I ) o r e le ctro n capture (EC) d e te cto rs. Gas chromatography - m s s sp actro m etrlc (CC-HS) a n a ly s is was performed on waste feed samples to confirm the Id e n tity of major components. The va rio u s co nd itio ns and columns usad a re sumnarlzed In Tables A .5 to A .7* The mass spectrom eter used was an AEI MS-30 Instrument equipped w ith en e le ctro n bombardment Ion so u rce. Between the GC end the MS the In te rfa c e Is o f a l l g la s s design w ith a s i 1leone m olecular membrane. GC columns s i m i l a r to those described In Tables A .5 to A .7 were used a t approximately the same Instrument co n d itio n s. ADM T 69 TABLE A. 5. CC ANALYSIS - SYSTEM A Column Co 1imn Temperature In jecto r Temperature Oetector Temperature Oetector Flow Rates Range and Attenuation Chart Speed - Chromosorb 102 (80/100 mesh) 6' x 1/8" SS - 180C - 215C - 215C - FID and EC - at 60 ml/rain A ir and H2 adjusted for maximum se n sitiv ity - As required - As required TABLE A .6. GC ANALYSIS - SYSTEM B Co 1umn Column Temperature In jecto r Temperature Detector Temperature Detector Flow Rates Range and Attenuation Chart Speed \ 15t SE 30 on Chromosorb W (AW; HDMS; 60/80 mesh) 12* x 1/6" SS - 60C isothermal for 20 minutes then progranvned at 10C/min to maximum temperature ** - 190C - 230C - FID - N2 at 60 ml/min Air and Hj adjusted for maximum se n sitiv ity - As required - As requ i red CM CC173 70 Col unn TABLE A .7. Column Temperature in je cto r Temperature Detector Temperature Detector Flow Rates Range and Attenuation Chart speed .GC -ANAlyYSIS - SYSTEM C **% SE 30/6* QF 1 on Chromosorb U (HP; 60/80 mesh) 6* x 1/8" SS - 200C - 250C * 230C - Lin ea rized EC - at 25 ml/min - As required As required These co n d itio n s a re used for ro u tin e PC8 a n a ly se s. The same parameters were used fo r WBB waste feed except that an isothermal temperature of 155C was m aintained. A .6.3 Blanks and standards For tra ce a n a ly tica l studies i t Is necessary to avoid contamination o f samples a t a l l stages o f sample c o lle c t io n and p rep aratio n . Though extreme care was taken during the study i t was not p o ssib le to p re *e x tra ct a l l hardware and chemical reagents used. Oue to the high s e n s i t i v i t y of the EC detector to many compounds, some o f the blank e x t r a c t s , th e re fo re , gave complex GC-EC p r o f il e s , which had to be subtracted as background from the sample chromatograms. The follow ing blanks were obtained and analyzed fo r t h is purpose. 1) P a r t ic u la t e t r a in : (a) P a rtic u la te f i l t e r + probe rin se f ilt e r + Soxhlet thimble. (b) Water n ) Gaseous tra in ? (a) In -stack f i l t e r + probe rin se f il t e r + Soxhlet thimble (b) Water \ (c) Aqueous NaOH m ) Solvents: (a) Pentane (100 ml concentrated to 1 ml) (b) Hexane (100 ml concentrated to 1 ml) ac* CC817*: r t Tjpyjflp: 5 71 tv) Polyethylene containers: (a) Solvent e x tra cts (100 ml concentra/ ted to 1 ml) Standard gas samples o f vario u s components o f In t e r e s t , such as CCl^, C H C 1 C H j C l j and 1,2-d Ich lo ro eth an e, were made by In je c tin g 50 pi a liq u o ts o f these compounds into an evacuated 500 ml gas j a r through a s ilic o n e septum. Clean a i r was then at towed to en ter the j a r to a tt a in atmospheric p ressu re . The sealed gas J a r , th e re fo re , represented nom inally a 100 ppm standard. Gas standards of lower concentration were made by d ilu t io n o f the 100 ppm standard using a s im ila r procedure. These standards were used lim e d la te ly a ft e r preparation and were not retain ed fo r use on the follow ing day. Solution standards were made by weighing a ccu ra te ly known amounts o f the components o f In te re s t and d isso lv in g them in a known volume o f pentane. A .6.4 Concentration factors Desorbed compounds from the adsorbent tubes were concentrated in to a 500 ml gas j a r . For each p a r t ic u la r t e s t , th e re fo re , the concentra tio n fa cto r would be the volune o f em ission.passed through the tube divided by 500 ml. The volumes sampled fo r each tube varied from 1.5 cubic fe e t to 3-22 cubic fe e t during the complete t e s t program giving a. range of concentration fa c to rs from 85 to 181. Thus, a concentration fo r a p a r t ic u la r component o f 1 ppm In the gas J a r sample would mean a concentration o f about 10 ppb in the k iln em ission, alssuning a 100 p ercen t c o lle c t io n e f f ic ie n c y , and subsequent desorption o f the adsorbent. Most o f the extracted samples were concentrated into a 2 ml voliana o f so lv e n t. The follow ing example I l lu s t r a t e s how the concentration fa c to rs were determined fo r va rio u s components d etected. Assume that C I^ C Ij mos detected a t a concentration o f 1 ppm in the so lven t e x tr a c t . ' T h is corresponds approximately to a weight o f 2 ug o f the compound in 2 ml o f so lv e n t. The m olecular weight of CH^Clj I * 85- T h ere fo re, 2 ug o f CH2C12 a t 70C, the temperature a t which the em ission sample was measured, occupies a volune o f: 2 T x 2k l i t r e s , or --------- r- x -- -- cubic f e e t . 85 x 10 85 x 10 28.3 AOh 008115 72 For te s t VBA 1, fo r example, a volume q f 142.81 s c f was sampled. Thus, the concentration o f CI^ C^ In the k iln em ission i s : * 2 - 7 x 24 x 0 6 ppm or 0. .1,,4 ppb. 8 5 x 1 0 ' 28.3 142.81 The concentration fa cto r is th erefo re 10^ - 7140. 0.14 Facto rs were ca lc u la te d fo r feed compounds o f lowest and highest molecular weight. A .6,5 Sample chromatograms A n a ly sis fo r low m olecular weight organohaIdes and noncombusted ch lo rin a te d organic compounds present In any c o lle cte d samples was made using GC techniques o n ly , by s e le c tin g appropriate GC parameters and using reten tio n time d ata. No s p e c if ic cleanup or separation procedures were performed in order to segregate components o f In te re s t from p o ssib le A in te rfe rin g compounds because, although GC p r o f ile s obtained were q u ite complex, very low concentrations o f organic compounds were evident from the peak heights obtained fo r the attenuations used. Estim ates o f s p e c if ic confound concentrations in waste burn samples were made by su b tra ctin g blank and b a se lin e le v e ls and comparing w ith a standard fo r th a t compound. Waste fee d s. GC-FID p r o f ile s fo r the three waste feed m a teria ls burned in t h is study a re given in Fig ures A .7 to A .9. '` Id e n t ific a t io n o f the main components In the re sp e ctiv e waste feeds a r e presented in Tab les 1, 2 and 3 In Section 5 o f the report* - I t was a n tic ip a te d that the le v e ls , I f any, o f uncombusted components In the stack em issions would be very low. Therefore GC-EC a n a ly s is was used to*evaluate the organic solvent e x tra cts o f the v a rio u s Implnger samples from the sampling t r a in s fo r the presence of any uncombusted components. Fig ures A .10 and A .11 represent GC-EC p r o f ile s fo r d ilu te d waste feed m aterial fo r WBB and WBC, r e s p e c t iv e ly . WBB a ls o contained approxim ately 5 QB o-chlorotoluene which has a week response to the EC d e te cto r. Th erefo re, in t h is Instance the so lven t ^ 7 e x tra c ts were a ls o analyzed by GC-FID In order to determine whether any uncombusted o-chlorotoluene was p resen t. ACF* CC i71 ACM FIG U R E A. 7 GAS CHROMATOGRAPHIC PROFILE FROM FLAME IONIZATION DETECtOR FOR CHLORINATEO AUPHATICS |WBA| SAMPLE FEED i .n il mmm m ,i it*. V. , lit i;!.' U .ft<.L f i g u r e a . e GAS CHROMATOGRAPHIC PROFILE FROM FLAME IONIZATION DETECTOR FOR CHLORINATED ALIPHATICS PLUS AROMATICS AND ALICYCLICS IW B B ) SAMPLE FEED AA J "*C o 3t F I G U R E A .9 GAS CHROMATOGRAPHIC PROFILE FROM FLAME IONIZATION J DETECTOR FOR CHLORINATEO ALIPHATICS PLUS AROMATICS, " A LICYCLICS AND POLYCHLORINATEO BIPHENYLS |W B C | ' SAMPLE FEED ala* ACh CCfclfcC ^ Miciitii.41 /..I , F I G U R E A.10 g a s c h r o m a t o g r a p h i c p r o f i l e s f r o m e l e c t r o n c a p t u r e n n T B C T O n POR W B D -SA M P LE F E E D * . . . ,J.- rnrnmmmmm *10 Pi. W w -J to 0- STAMDARD AROCLOR 12(2 6 OC 5Y ST EM C 7- mmmm m W BC-ARO M ATICS * P C B s SA M PLE F E E D JAN. 0 th 1976 GC SYSTEM C 1p i |2.73 10<MJ/>lI| INTENSITY C 3 1 1x256 -r- --T* - 6 8 10 12 0 4 6 8 T P 12 T~ ri T IM E (m ini 7 TIM E (min) r o S F I G U R E A.11 0A5 CHROMATOGRAPHIC PROFILES FROM ELECTRON CAPTURE n' DETECTOR FOR STANDARD AROCLOR 12(2 AND SAMPLE FEED WBC viiiM / ?r!S75iB? JLIU U r J A U L a U - * .J J L I 'P . !.,u m J . i . . l .L li . . u i t -- m : -------- J 78 Standards. Fig ure A . 12 shows the GC-EC p r o f ile for the low m olecular weight ch lo rin a te d hydrocarbons, e .g . CH^Ci^, CHCl^ CCl^ and 1,2-d ichJo rethene. Figure A . 13 shows the GC-FID p r o f ile for the same compounds to in d ica te the g re ater s e n s it iv it y of the EC d etecto r. A GC-EC p r o f ile fo r 5 ng o f standard A roclor 12**2 Injected into the column Is -shown In Figure A . 11. Waste Burn Samples. Fig ure A .12 shows GC-EC p r o f ile s for desorbed gas samples for BLB and WBC te st samples. The main d iffe re n ce between the samples is In the in te n s ity o f the peaks o f in t e r e s t. The GC-EC p r o f ile s fo r a l l the solvent e x tra c t samples showed the presence of small peaks. When the p r o f ile s were compared to those obtained from the BLA samples and the vario us control blanks ( c f . Section A .6 .3 )* no peaks p e c u lia r to the waste burn samples were detected. BLB samples became contaminated w ith WBC feed m aterial and were d iscard ed . In most in sta n c e s, the EPA impinger e x tra c ts showed the la rg e s t response to the EC d e te cto r. Fig ures A. 14 and A .15 show GC-EC p r o f ile s fo r the org anic so lve n t e x tra c ts o f the EPA Impinger samples obtained fo r BLA-T3 and W8C-T3, r e s p e c t iv e ly . P r o f ile s obtained fo r the two samples a re q u ite s im ila r . There i s a large EC response o ccurring as an unresolved peak in the e a rly portion of the chromatographic p r o f ile . T h is lim ite d the sample s iz e that could be used for In je c t io n , and r e s t r ic t e d the use o f high s e n s i t i v i t y se ttin g s on the gas chromatograph. It Is normal p ra c tic e when an alysin g fo r PCB's to perform cleanup and sep aration procedures In order to segregate the PCB's from In te rfe rin g components and thus f a c i l i t a t e th e ir a n a ly s is . Such a cleanup and sep aratio n procedure was performed on WBC-T3 using F l o r i s i l adsorbent (27) in order to determine whether the e a rly unresofved peak (Fig u re A .15) could be renoved. A fter cleanup, i t was p o ssib le to use a la rg e r a liq u o t fo f In je c tio n and higher s e n s it iv it y se ttin g s on the chromatograph. The GC-EC p r o f ile o f the cleaned up W8C-T3 sample is shown In Fig u re A .16. The GC-EC p r o f ile ihown In Figure A .16 was obtained on an org anic so lven t e x tra c t of a sample that should contain PCB's I f any were present In the k iln em issio n s. The p r o f ile shows the presence o f a few small peaks. When compared w ith the Standard A roclor p r o f ile (Figure A . I I ) , AC* CC6162 \ oa> F I G U R E A.12 GAS CHROMATOGRAPHIC PROFILES FROM ELECTRON CAPTURE <T 1 DETECTOR FOR LOW MOLECULAR WEIGHT CHLORINATED HYDROCARBONS AND FOR BLB AND WBC TEST SAMPLES 80 /"N FIGURE A.13 GAS CHROMATOGRAPHIC PROFILE FROM FLAME IONIZATION < } DETECTOR FOR LOW MOLECULAR WEIGHT HYDROCARBONS auk cctifa* 81 9 GC SYSTEM C 8 injection 7 6 5 4 3 2 1 x 64 5 v 10 IJ J i ! I* 15 20 TIME {mini A .K CAS CHROMATOGRAPHIC PROFILE FROM ELECTRON CAPTURE DETECTOR FOR IMPINGER EXTRACT FROM BLA TEST 3 Jr i Hit I ii ijii r. fi* i\l i. If * F tn <9 ' oO;`-: ^ '1 ill 82 i 3 (OJ 4 `J FIGURE A.IS GAS CHROMATOGRAPHIC PROFILE FROM ELECTRON CAPTURE DETECTOR FOR IMPINGER EXTRACT FROM WBOTEST 3 B3 91900 wnv FIG U R E A. 16 GAS CHROMATOGRAPHIC PROFILE FROM ELECTRON CAPTURE DETECTOR FOR IMPINGER EXTRACT FROM WBC TEST 3 AFTER CLEANUP AND SEPARATION 84 and taking into account that the e x tra c t represents a >7000 concentration f a c t o r . I f any o f the peaks In Figure A ..16 represent PC8 components the le v e ls must be extremely low. A .7 Analysis of Process Solids For the purpose o f obtaining a mass balance during the experimental ch lo rin a te d hydrocarbon burn, I t was required to c a rry out chemical a n a ly se s of c lin k e r , s lu r r y feed and d iscard dust fo r ch lo rin e and potassium. Sulphur was determined on process s o lid s and the return dust samples were analyzed fo r changes which may have a ffe cte d k iln operation. X-ray flu o rescen ce (XRF) a n a ly s is has been e x te n siv e ly used fo r determ ination o f su lp h u r, potassium and ch lo rin e in cement, cement raw m a te ria ls and re la te d m a te ria ls , and was selected for use In t h is study. A .7.1 A nalytical procedures Samples were prepared by grinding 10 grams o f sample for 30 seconds In a SPEX Shatterbox w ith tungsten carbide co n ta in ers and pucks. From t h is m aterial a 1.25 inch diameter p e lle t a t 8 tons pressure was made In SPEX aluminum sample caps. For the X-ray flu o rescen ce method to be a p p lie d , there must be a lin e a r r e la tio n s h ip between elemental concentration (over the f u ll range of concentration required) and the measured flu o rescence in t e n s it y . The slope of the lin e representing t h is re la tio n sh ip (hen can be used d ir e c t ly as a c a lib r a t io n constant, u su a lly In the form o f the r a t io , counts per second:percent elem ent. When a c a lib r a tio n lin e Is generated by known a d d itio n s o f the t e s t element to a m aterial being analyzed ("sp ik in g "), the lin e Intercep t g ives the percent o f the element present in the o rig in a l sample p rlp r to any a d d itio n s. In addition to lin e a r it y , re p ro d u cib ility o f the ca lib ra tio n constant Is a p r e re q u is ite I f I t I s to be applied a c ro ss a range of m a te ria ls having some degree o f m atrix v a r ia b i l i t y . Both conditions were examined p r io r to the f in a l development o f the a n a ly t ic a l procedure. The c a lib r a t io n curves for ch lo rin e content o f c lin k e r and raw m e a l-slu rry feed were prepared by ad d itio n o f standard lithium i-s ' aoa c c e i c t % " i I \ s * ! 85 ch lo rid e in tcohoi so lu tio n to samples of these m a te ria ls. Oust samples were analyzed by the standard Vathard njethod for c a lib ra tio n o f the c h io r ne"curve and the standard g ra vim e tric method of p re c ip ita tio n w ith barium for c a lib r a tio n of the sulphur curve (Table A .8 ) . For c a lib ra c i n of curves fo r sulphur In c lin k e r and raw m eal/slu rry feed, determ inations were made on a Leco'Ind uction Furnace Model 523CS with le co Automatic T it ra t o r Model 517 (Table A .9 ). For c a lib ra tio n of the potassium curve, a l l potassium r e s u lt s , as w ell as those fo r sodium, were obtained from analyses on a Perkin-Elm er Model 460 Atomic Absorption Spectrophotometer in the flame em ission mode (Table A .10). Examination of these samples yielded lin e a r c a lib ra tio n re la tio n sh ip s between X-ray flu o rescen ce count rates and percent element. The le a st squares computations fo r each se t of data gave the slo p e s, in tercep ts and reg ression (c o rre la tio n ) c o e ffic ie n ts lis t e d in Table A. 11. TABLE A .8 . GRAVIMETRIC OUST ANALYSES Oust Type Discard 11 11 11 11 h 11 11 11 11 H Return ' 11 it Date 19/10/75 11/11/75 31/11/75 V l 2/75 5/12/75 6/ 12/75 7/12/75 5/ 1/76 6/ 1/76 9/ 1/76 I V 1/76 16/ 1/76 5/12/75 13/ 12/75 1/ 1/76 IB / 1/76 X Cl 2.00 2.08 5.97 2.89 7.11 4.89 6.18 4.71 3.72 it. it3 0.46 0.90 2.11 4.34 2.10 1.00 , % SO3 - 3.17 2 .3 6 6 .5 3 12.12 7.27 6.90 5 .0 8 4.23 4.18 4.65 4.32 7-35 4.22 <1.99 5.66 4CP CGfilcS ; ; [ 86 TABLE A .9. RE5ULTS FROM LECO INDUCTION FURNACE ANALYSES ' Hater la) Cl inker it 11 it ** ii ii ii ii ii S lu rry Feed M ii ii ii ii ii ii ii ii ii .. Date/ 19/10/75 21/10/75 23/10/75 31/10/75 1/11/75 5/H/75 8/12/75 31/12/75 1/ 1/76 11/ 1/76 20/ 1/76 13/10/75 20/10/75 27/10/75 9/11/75 6/12/75 9/12/75 13/12/75 16/12/75 31/12/75 3/ 1/76 8/ 1/76 21/ 1/76 X SO. (T e t ll S as) 0.93 1.3* 0.75 0.30 0.16 0.66 0.93 1.39 l.*3 1.06 1.42 0.65 0. *2 0. *1 0.45 0.52 0.48 0.60 0.55 0.43 ' 0.67 0.46 0.67 ACK r 87 TABLE A .10. RESULTS FROH ATOMIC ABSORPTION ANALYSES Mater r i l Date t k2o X Na20 D iscard Oust ii ii 11/11/75 31/H/75 V I 2/75 6.36 10.30 7.14 0.32 0.39 0.36 ii 6/12/75 8.23 0.44 ii 7/12/75 9-96 0.45 5/ 1/76 8.12 0.43 ii 6/ 1/76 6.43 0-34 n 9/ 1/76 7.9* 0.39 t ii I V 1/76 4.64 0.35 ii 16/ 1/76 4.18 0.34 Return Dust 5/12/75 7.79 0.42 [ ii 13/12/75 5-54 0.41 ii 1/ 1/76 6.02 0.41 Cl inker 19/10/75 1-32 0.23 I i i 21/10/75 1.61 0.25 ii 23/10/75 1.00 0.20 n 31/10/75 0.54 0.21 l ii 1/11/75 0.40 0.21 i i i 5/11/75 1.01 0.21 ti . 8/12/75 0.64 0.20 ii 31/12/75 1.18 0.21 n 1/ 1/76 1.31 0.24 ii 11/ 1/76 1.10 0.22 ii S lu r ry Feed 20/ 1/76 13/10/75 1.45 0.93 0.26 0.19 ii 20/10/75 0.93 0.22 ii 27/10/75 0.69 0.18 ii 9/11/75 0.86 0.20 ii 6/12/75 0.91 % 0.19 n 9/12/75 0.90 0.22 n 31/12/75 0.90 0.18 ii 3/ 1/76 0.91 0.22 n 8/ 1/76 0.93 0.19 ii 21/ 1/76 0.94 0.22 i n * f.' k. <i F!.. it ; i.v : \ t6.f;i i if .U I i COBV^1 1 88 TABLE A .11. LEAST SQUARES DATA FOR CALIBRATION LINES M aterial Raw Meal + S lu rry Feed C lin k e r Oust Element Correlation C o e ffic ie n t Slope Potass 1tan (as K^O) Chlorine Sulphur (as S0^) Potassium (as K^O) Chlorine Sulphur (as S0^) Potassium (as K^O) Chlorine Sulphur (as SO^) 0.9172 0.9995 0.9913 0.9829 0.9999 0.9917 0.9893 0.9548 . 27128 5260 739 24980 4781 400 3166 1927 278 Intercept -0.01 0.060 0.27 -0.01 0.011 -0.23 -0.12 -0 .t4 -0.10 A .8 Determination of Heat Value. Chlorine Content and Sulphur Content In No, 6 Fuel The heat value was determined using standard methods on a Parr Calorim eter Model 1241 equipped w ith P arr oxygen bomb Model 1108. Sulphur was determined using the standard barlun p r e c ip ita tio n method on the washings from the'bomb. Where c h lo rin e content was req u ire d / the washings were analyzed fo r c h lo rin e by the standard Vojhard method. A .9 Determination o f Heat Value and Chlorine Content In Chlorinated Hydrocarbons The heat value was determined on a P arr calo rim eter model 1241 equipped w ith P arr oxygen bomb model 1108 follow ing standard operating procedures fo r determining heat valu e of o i l . Due to the c o rro siv e nature o f the combustion prod ucts. I t i s recommended that the ch lo rin a te d m a te ria ls be d ilu te d w ith standard o i l . I t has been found th a t, w h ile co rro sio n o f the In t e r io r o f t(i bomb Is e x te n siv e , burning these m aterials without d ilu t io n g iv e s s im ila r r e s u lt s . .Time between weighing and Ig n itin g Is extrem ely Important due to the v o la t ile nature o f some of these compounds. AGP CCfci9i iewn mo i m Mu 89 A fter Ig n itin g and determining the heat value of the ch lo rin ated M a te ria l, the In te r io r o f the bomb is thoroughly washed w ith d i s t i l l e d w attr into a volum etric fla sk . Aliquots of the proper s iz e for the ch lo rin e content expected were taken. N it r ic acid is added to allow a More d is t in c t end p o in t, and the c h lo rin e is determined by the standard Volhard t it r a t io n . REFERENCES A . 1 O ntario M in istry of the Environment, Source Testing Code. January, 1973* A .2 Standard Reference Methods for Source T e stin g : Measurement of Em issions o f P a r t ic u la t e s from Sta tio n a ry Sources. Environmental P ro tectio n Se rv ice Report EPS l-A P -7 k -l, A ir P o llu tio n Control D ire cto ra te , Environment Canada, February, 197k. A .3 P e lliz a n , E .D ., J . E . Bunch, and B.M. Carpenter. Env. S c i . Technology 9 (6 ), 552-560 (1975). I i | \ I r El i e: \ ^j Pi t( APPENDIX B ANALYTICAL DATA. CALCULATION AND DETAILS OF EXPERIMENT ON THE SUSPENSION PREHEATER KILN 33 APPENDIX B / _ ANALYTICAL DATA, CALCULATION AND DETAILS OF EXPERIMENT ON THE SUSPENSION PREHEATER KILN B .l Theoretical ' The e x c e lle n t a r t i c l e by Weber c ite d in the body o f the report d escrib e s In d t a il the problem o f a lk a li reduction in a suspension preheater k i ln , and was used In preparing the follow ing notes. At m aterial temperatures above.800C in the rotary k iln , a lk a l i e s (K^O and Na^O) a re v o la t iliz e d from the m aterial being burned and a re c a rrie d along w ith the e x it gases to the k iln in le t . These condense a t gas temperatures below 900C. In p a rt, they a re recaptured by the m a te ria l, p a r t ic u la r ly in the preh eater, and a re thus c a rrie d back to the sin te rin g zone, so that an in te rn a l a lk a li c y c le is e sta b lish e d The a lk a l i e s a re In part a ls o p re c ip ita te d a t the tra n s it io n from the preheater to the k iln . Coatings thus formed in te rfe re w ith m aterial and gas flows and, in the most severe c a s e s , completely, plug the system. In preheater k i ln s , the raw meal Is heated in the f in e ly divided condition In the gas stream and these p a r t ic le s a ct as condensation nuclei for the a l k a l i e s . Between 81 and 971 o f the K^O v o la t iliz e d in the k iln is trapped in the preheater. The g re ater the a lk a li c y c le , and therefore th e `Concentrt ion of a lk a l i In the g a s, the more a lk a l i w i l l condense on the bottom ducts and cyclone sta g e s. I f the a lk a li is very v o l a t i l e , a c c re tio n s a re l ik e ly to occur In the preheater. However, the a lk a li c y c le can be reduced.by the p ro v isio n o f a bypass, that Is by drawing o f f a portion o f the gas a t the bottom,of the preh eater. (See Figure 3 Section 2 of the rep o rt.) Of a lk a l i which e n te rs the k iln system w ith the raw meal, a proportion ( c j) w i l l be v o la t lllz b f l and the remaining (1- e ( ) w i l l be discharged w ith the c lin k e r . I f the proportion of gas withdrawn through the bypass Is V, then o f the voi at 111zed a lk a l le s , the proportion c^ V w ill be removed by the bypass. The remaining portion c , (l-V ) w i l l return with the feed, thereby giving r is e to the In te rn a r cy cle in the k iln . ACM C0dlS5 3k The alka li in the internal cycle has a different v o l a t i l i t y than a l k a l i from the raw meal which has not yet been v ol a t il i ze d . By defining total a lk a li including raw meal a l k a li into the kiln by K, then the quantity of a lk a li from the cycle alone is K-l. Of this quantity, (K-1)c ^ v is removed by the bypass. Equilibrium is attained when the total quantity of a l k a l i input with raw meal and fuel is equal to the quantity output by the bypass and retained with the clinker. This is most ea sil y determined by considering only a 1ka11 in the cycle. Entering the cycle is the v o l a t i l e raw meal a l k a l i e^i leaving the cycle are a proportion in the clinker (K-! ) (l-c^) and a proportion through the bypass [cj (K-1) By setting the quantities input and output from the cycle as equal, the equation may be solved to determine the amount of gas the bypass must remove to keep the a lk a li cycle from becoming excessive. B. 2 Experience at St.~ Lawrence Cement At St. Lawrence Cement, samples are taken of the material between the Stage IV cyclone and the kiln inlet (see Figure 3). this material being called, not quite appropriately, "Stage IV dust". From experience at St. Lawrence Cement, it was known that a chloride content of 2.51 chlorine in this material (natural basis) is excessive. The quantity of gas required to be withdrawn by the bypass to maintain lower chloride levels was calculated on the assumption that the chloride v o l a t i l i t i e s were 0.99 and 1.0 (completely v o l a t i l e ) . These quantities are given in Table 8.1. While the bypass was designed to remove the quantities of gas required for the program, attempts to reach this level resulted In excessive emissions of particulate matter from the bypass precipitator. These were attributed to excessive gas flows. To a l l e v i a t e this problem, measures to increase cooling water and decrease the quantity of ambient a i r were undertaken, with limited success. A C * CCfclSe 95 TABLE B.1. PERCENT BYPASS GAS REQUIRED TO MAINTAIN CHLORIDE LEVELS I / Chloride addition relative _to c l inker . X k20 in c l i nicer Bypass percent required to ma inta i n Stage IV Cl a t : 9 1.56* or 2 .0 * nat. Ign. 2.01 or 2.56$ nat. ign. 2-5% o r 3-2$ nat- ign. 0 1.31 2.1 1.6 1.3 0.05 1.24 A .8 3.6 2.9 0.10 1.18 7-1 5.5 A.A 0.15 1.11 9.6 7.5 6 .0 0.20 1.04 12.1 9 .A 7 .0 0.30 0.91 17-1 11.7 10.6 0. A0 0.77 22.2 17.2 13.8 B.3 Equation of Approximated Time for Equi librium It was e s s e n t ia l to the success of the program to determine the time required to a tt a in the eq u ilib riu m s t a t e . To obtain an approxi mat ion o f the time req u ired , and r e a liz in g that it is only an approxima tio n sin ce reten tio n time In k iln and absorption phenomena influence t h is tim e, the follow ing approach was taken: a dt where: - C ,RM ^*cl Inker ^ bypass dt - C lRM - C ,clinker C l.bypass m change In q uantity with time a ch lo rln e/tim e In with raw meal chlorine/tlm e out with clin k e r ch lo rine/tlm e out through bypass (pounds/minute) (popnds/minute) (pounds/mInute) (pounds/minute) C lR.Mu and Cl c lin k. e r a re approximately constant, hence they can be combined a s : c - C1ORMM - C l c l.in. .k e r,, ACM 96 nd C ,byp.s* where: Q(t) V Then: VQ (t) / q uantity In t r a n s it io n chamber a t time ( t ) In pounds bypass valve proportion. & dt I c l - VQ ( t ) Solving for time g ives: In [I - VQ ( t ) ] Q* t - ------- ----------------V Qp minutes. Since In 0 Qp must be taken as a very smalt d ista n ce from the true equilibrium . The sampling sequence a t Stage IV was planned from t h is model. ii B.4 Sampling and A n a ly tic a l Methods A n a ly tic a l methods were those described In Appendix A. Samples of c lin k e r , raw meal feed, bypass p re c ip ita to r dust and conditioning t tower dust were taken a t two hour in t e r v a ls . These were then blended into 24-hour composite sa n p le s. On the b a sis of the c a lc u la tio n in Section B.3* samples were taken from Stage IV at 3 to 10 m inute, in te rv a ls when attempting to s t a r t the chlo rinated burn. These la t t e r samples were analyzed in d iv id u a lly . B.5 Discussion In both cases d escrib ed , the bypass was operated to withdraw 9 percent o f the gases from the preheater tower. T h is was the maximum a tta in a b le due to the em ission ra te of p a rt ic u la t e m atter. On the f i r s t attempt to burn chlo rinated hydrocarbons, the quantity to y ie ld 0.24 percent c h lo rin e r e la t iv e to c lin k e r was fed s t a r t in g a t 09:3&* June 3. 1975- Plugging o f the duct between the k iln and the cond itioning tower caused a bypass shutdown on June 5, 1975- It was not p o ssib le to m aintain feed o f ch lo rin a te d m aterial to the k iln u n less the bypass was functio ning . On June 10, 1975 a second and s im ila r attempt was started i I ACP CC61<56 MtfUBI , . W ... IH IliW W M * r f 1J,)t ^} \ 97 at 11:20. The bypass again fa ile d due to a ccre tio n on the duct on June 13, 1975. On the second attem pt,^ am pling of Stage IV m aterial was continued fo r a longer period on the b a sis of the f i r s t set of re su lts. It was r e a liz e d w ith the f a ilu r e that rw jor re v isio n s to the bypass system would'be required to continue t h is experiment on the suspension preheater k iln . B.6 A nalytical Results A n a ly tic a l r e s u lt s are given in Tables 8.2 through 8 .A. The raw meal feeds showed a gradual In crease in both the content of ch lo rin e and KgO. The r e s u lt s from the c lin k e r a n a ly s is (Table B.3) showed that the burning o f ch lo rin a te d hydrocarbons resu lte d In a reduction in the K^O content (June 3~5, 1975 and June 10-12, 1975). The r e s u lt s o f the an alyses from Stage IV dusts (Table 8 .4 ) show th a t the ch lo rin a te d hydrocarbons were destroyed In the burning p rocess. Chloride content o f these samples Increased'w hlle burning ch lo rin a te d w astes. T h is find ing could not have re su lte d without breakdown of the hydrocarbons sin ce the raw m aterial Is at too high temperature a t th is positio n in the preheater for the chlorinated hydrocarbons to condense upon i t . By. the methods described in Section 8 .3 , the approximate time to eq u ilib riu m was ca lcu la te d to be about 90 m inutes. I t was a ls o expected that an ad d itio n al time o f approximately 20 minutes would be required to tra v e l through the k ilq ,. The actual curve Is le s s steep and e q u lllb rlu n Is reached in about four hours. For the June 3 1975 burn, the curve was extrapolated to the ca lcu la te d eq u ilib riu m (Fig u re 8 .1 ) . On June 10, 1975 samples were taken fo r a longer perio d . Although there a re points above the ca lcu la te d curve (Fig u re 8 .2 ) , the data from June 11, 1975 show that eq uilib rium was reached; the higher valu es found in d ica te random sampling or analytical erro rs. ACM 008155 t i !! Si l A [i jr f r IF 98 TABLE B .2. RESULTS FROM ANALYSES OF DRY PROCESS KILN RAW FEEDS / Date X Cl X so, (Total S^as) X K20 90/5/75 31/5/75 1/6/75 2/6/75 3/6/75 6/6/75 5/6/75 6/6/75 7/6/75 8/6/75 9/6/75 10/6/75 11/6/75 12/6/75 13/6/75 16/6/75 15/6/75 0.067 0.069 0.073 0.071 0.076 0.082 0.079 0.09 A 0.087 0.093 0.088 0.084 0.092 0.084 0.104 0.102 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.62 0.60 0.60 0.58 0.60 . 0.60 NO SAMPLE 0.60 0.60 0.62 0.92 0.95 0.93 0.93 0.93 0.95 0.96 0.98 0.97 0.97 0.97 0.95 0.98 0.96 1.00 1.00 A0* CCiCC.i ~ 99 TABLE B .3. RESULTS FROM ANALYSES OF,DRY / PROCESS KILN CLINKER Date 30/5/75 31/5/75 1/6/752/6/75 3/6/75 6/6/75 5/6/75 6/6/75 7/6/75 8/6/75 9/6/75 10/6/75 10/6/75 10/6/75 10/6/75 ' 10/6/75 10/6/75 10/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 11/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 12/6/75 13/6/75 16/6/75 15/6/75 ' Time Average n it ii i ii ti ii ii ii ii ii 12:00 14:00 16:00 18:00 20:00 22:00 . Average 00:00 02:00 04:00 06:00 08:00 10:00 14:00 16:00 18:00 20:00 22:00 * Average 00:00 02:00 04:00 06:00 08:00 10:00 * 12:00 14:00 16:00 18:00 Average ii ii x ci 0.03 0.03 0.03 0.21 0.03 0.05 0.04 0.04 0.07 0.05 0.06 0.04 0.06 0.04 0.04 0.05 0.05 0.05 0.06 0.04 0.06 0.07 0.06 0.06 0.05 0.05 0.05 0 .06. 0.05 0.05 0.07 0.07 0.05 0.05 0.05 0.05 0.10 0.05 0.06 0.05 0.05 0.05 0.05 0.04 % so2 1.05 1.15 1.15 1.16 1.10 M9 1.23 1.34 1.33 1.31 1.50 1.22 1.38 1.00 1.07 1.30 1.03 1.11 1.29 1.18 1.36 0.98 0.96 1.10 1.08 1.18 1.08 1.12 1.23 1.01 1.28 1.07 1.12 1.11 1.10 1.11 1.25 1.11 1.23 Ml 1.14 1.29 1.40 1.35 ,, % KjO 1.27 1.29 1.31 1.30 1.24 1.30 1.33 1.44 1.48 1.43 1.63 1.29 1.25 0.97 0.87 1.1 0 0.96 1.05 1.24 1.27 1.25 1.00 0.99 1.10 1.08 *1.15 M3 1.14 1.21 1.06 1.29 1.15 1.16 1.15 1.16 1.18 1.36 1.15 1.28 1.17 1.18 1.40 1.52 1.41 1 \ f t 1 `1 1 \ 1 - 1 1 |. |: !i ] ! 1 1 i <aIIti ii If { ( | i: 1: i' 1 \1>!. 1 IItI Ii 1!i 0 ], r. !I f; l! i1 !1i: 1 if ll r <*! 100 TABLE B .4 . RESULTS FROH ANALYSES OF STAGE IV DUST5 Date 30/ 5/76 2/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 3/ 6/75 6/ 6/75 6/ 6/75 5/ 6/75 6/ 6/75 Time , - 09:00 09:06 09:12 09:18 09:24 09:30 09:38 09:44 09: *7 09:50 09:53 : 09:56 09:59 10:02 10:05 10:08 10:11 10:16 10:23 10:32 10:41 10:54 * 10:59 11:04 11:09 11:14 11:19 11:24 11:29 11:34 11:39 11:44 .11:49 11:54 11:59 12:04 12:09 12:14 14:00 15:30 09:00 14:00 - x X Cl 0.67 0.82 0* 86 0.84 0.81 0.85 0.82 0.86 0.90 0.97 0.96 1.02 1.20 1.21 1.21 1.22 1.34 1.44 1.51 1.44 1.56 1.79 1.83 1.85 1.95 2.08 2.03 2.00 2.03 2.30 2.17 2.21 2.22 2.29 2.29 2.32 2.38 2.49 2.28 2.34 2.47 2.84 1.92 1.59 0.97 1.14 x so. (Total S as) 1.14 1.30 1.23 1.25 1-19 1.22 1.18 , 1.30 1.34 1.32 1.35 1.38 1.29 1.31 1.34 1.30 1.46 1.38 1.52 1.29 1.41 1.37 1.34 1.32 1.46 1-59 1.35 ,, 1.39 1.41 1.47 1.36 1.37 1.43 1.51 1.57 1.66 1.47 1.51 1.46 1.44 1.45 1.58 1.42 1.28 1.44 1-17 x k2c 1.77 2.26 2.11 2.03 2.03 2.14 2.06 2.17 2.15 2.21 2.38 2.39 2.37 2.42 2.51 2.55 2.68 2.66 2.95 2.91 2.94 3-38 3.43 3.35 3.74 3.85 3.72 3.66 3.70 4,15 3.93 4.04 4.00 4.18 4.11 4.24 4.36 4.45 4.12 4.21 4.44 5-07 4.17 3.4B 2.54 2.71 ACH CC62 C . tMiklNik.H 101 TABLE i . 4. (Cont'd) Dace 9/ 6/75 10/ 6/75 10/-6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/7S 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 . 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 10/ 6/75 11/ 6/75 12/ 6/75 Time - 09:00 09:10 09:20 09:30 09:40 09:50 10:00 11:15 11:25 11:35 12:05 12:25 12:35 12:45 12:55 13:05 13:15 13:25 13:35 13:65 13:55 14:15 14:25 14:35 I4;45 14:55 15:05 15:25 15:35 15:45 - - . X Cl 1.66 1.09 1.17 1.19 1.15 1.16 1.16 1.14 1.11 1.20 1.46 1.91 1-99 2.15 2.25 2.31 2.45 2.37 2.32 2.62 2.51 2.63 2.57 2.87 2.74 2.85 2.77 2.79 3.02 3.28 3.09 2.58 2.42 X so3 (Total 5 as) 1.20 1.18 1.22 1.36 1-39 1.29 1.27 1.29 1.29 1.34 1.41 1.49 1.46 1.64 1.59 1.66 1-79 1.71 1.47 1.67 1.54 1.69 1.69 1.81 1.67 1.70 1.77 1.52 2.41 1.59 1.62 1.51 1.49 t KjO 3-79 2.64 2.93 3.07 2.92 3.01 2-99 2.95 2.91 3.04 3.22 4.02 4.07 4.50 4.54 4.65 4.98 4.75 4.48 5.10 4.83 5.12 5.02 5*53 5.18 . 5.35 5-23 5.23 5.66 6.07 5.78 5.23 5.13 \ ACM CC2C3 CHLORINE t CHLORINE LEVEL IN STAGE IV 3/6/1975 FIGURE B.1 1>.I o '} ^ * m m m m ig q M I lw li'- ilf f ln 11 11"**rclllllliaWllMIMIII IIIIWIIIWiiM : * r i `i* CHLORINE % 1 n t nn CHLORINE LEVEL IN STAGE IV 10/6/1975 m ot\> in F I G U R E B. 2 >***. ta *i** APPENDIX C RESULTS AND CALCULATIONS FOR WET PROCESS SYSTEM \ \ ACP C C b i C t 107 APPENDIX C *. / - RESULTS AND CALCULATIONS TOR WET PROCESS SYSTEM C . 1 Mass Balance Experim entation C .I.l Results of analyses of process m aterials Analyses o f process m a te ria ls fo r the mass balance were c a rrie d out a t the S t. Lawrence Cement Co. p la n t. A n a ly tica l r e s u lt s are given in Tables C.1 to C .6. C .1 .2 Calculation of m aterial balances A d a ily record of production and m aterial consumption I s given in Table C.7 for the period October 7, 1975 to January 21. 1976. Produc tio n was disrupted during t h is period by a k iln shutdown fo r the removal of a k iln ring. The program was fu rth er disrupted by formation o f a r in g , not req u irin g k iln shutdown, but time to reduce the q u an tity of arom atic p lus complex ch lo rin a te d hydrocarbon in the tank and d i f f i c u l t y In scheduling d e liv e r ie s . From the d a ily composite a n a ly s is o f each m aterial fo r each element, and the to ta l d a ily q uantity of each m a te ria l, d a lly elemental q u a n titie s (as pounds or tons per 25 hour period) were determin ed by the re la tio n s h ip : x i j M1J X 100 where X 1J C IJ to ta l d a lly weight o f element x in m aterial i on day j concentration of element x in the composIte sample m aterial I on day J . These data ere then used to form the separate in d iv id u al elemental balance accounts given in Tables C.8 and C-9- Due to process flu ctu a tio n s and the very low concentrations In the case o f c h lo rin e , l i t t l e s ig n if ic a n c e should be attached to Individual d a ily balance r e s u lt s , except as they Indicate the responsiveness of the system to major changes In Input o f ch lo rin a te d hydrocarbons. S ig n ific a n t assessment o f the 108 balance is a v a ila b le by consid ering prolonged periods of plant operation. Summary balances fo r t h is purpose are given in the body of t h is report (see Section 5*^ Tables 8 and 9)* M a te ria ls entering the mass balance c a lc u la tio n s (see Figure 1, Section 2) were s lu r r y feed,' No. 6 fuel o il and ch lo rin a te d hydrocarbon m a te ria ls as input streams w ith c lin k e r and d isca rd dust forming the output stream s. The return dust Is returned almost iim e d ia te ly to the system and does not form a part of the mass balance. r i /"V fX \ ADM CCEC 109 TABLE C.l. RESULTS FROM ANALYSES OF CLINKER SAMPLES Oate 7/10/75 8/10/75 9/10/75 10/10/75 11/10/75 12/10/75 13/10/75 1V 10/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 2V10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 1/11/75 2/11/75 3/11/75 V I 1/75 5/11/75 6/11/75 7/11/75 8/11/75 9/11/75 2/12/75 3/12/75 V I 2/75 5/12/75 6/12/75 7/12/75 8/12/75 9/12/75 10/12/75 11/12/75 12/12/75 13/12/75 1V I 2/75 t Cl 0.033 .. 0.032 0.034 0.042 0.032 0.029 0.030 0.031 0.031 0.031 0.035 0.029 0.032 0.042 0.030 0.029 0.030 0.030 0.029 0.031 0.031 0.034 0.032 0.039 0 . 031) . 0.039 - 0.032 0.026 0.031 0.031 0.037 0.029 0.035 0.031 .0.029 0.036 0.033 0.029 0.028 0.038 0.040 0.034 0.035 0.032 0.032 0.031 x % so3 0.88 0.79 0.86 0.62 0.93 0.80 0.80 0.86 1.27 0.90 1.11 0.99 0.97 1.03 1-32 0.99 0.77 0.92 0.59 1.06 0.82 1.16 1.27 0.88 0.36 0.18 NO SAMPLE 0.26 0.30 0.64 0.70 0.71 0.68 0.72 1.13 1.39 1.02 1.23 MB 1.18 0.91 0.71 0.67 0.70 0.92 0.75 0.73 2 k2o 1.13 1.12 1.14 0.88 1.27 1.06 1.09 1.15 1.47 1.16 1.31 1.24 1.21 1.28 1.51 1.28 1.06 1.20 0.83 1.24 1.00 1.31 1.37 0.96 0.49 0.36 0.48 0.56 1.00 1.18 1.20 1.15 1.19 0.94 1.28 1.01 1.16 0.90 0.84 0.76 0.56 0.36 0.48 0.54 0.39 0.48 ACM CCB2CS Z""", V__ TABLE C.1 (COMT'O) 110 Cate 31/12/75 1 / 1/76 2/ 1/76 3/ 1/76 6/ 1/76 5/ t/ 7 6 6/ 1/76 7/ 1/76 8/ 1/76 9/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/ 1/76 16/ 1/76 15/ 1/76 16/ 1 / 7 6 17/ 1/76 18/ 1 / 7 6 19/ 1/76 20/ 1/76 21/ 1/76 X Cl 0.029 0.029 0.029 0.033 0.030 0.035 0.031 0.029 0.030 0.028 0.028 0.030 0.030 0.029 0.030 0.029 0.030 0.037 0.028 0.028 0.029 0.029 X SO3 X K2 0 1 .40 1.48 1.49 1.12 1.68 1.14 0.70 1.14 0.88 0.66 0.91 0.92 0.96 M3 1.14 1.12 1.38 1.36 1.20 1.06 1.62 0.94 1.20 1.34 1.37 0.77 1.42 0.78 0.58 1.26 0.83 0.46 1.05 1.12 1.15 1.35 1.28 1.22 1.65 1.42 . 1.27 1.15 1.46 1.12 ACM GC621G - rdb J5L 2? in TABLE C.2 RESULTS FROM ANALYSES OF SLURRY FEED SAMPLES fate 7/10/75 8/10/75 9/10/75 10/10/75 H/10/75 12/10/75 13/10/75 14/10/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 24/10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 . 1/11/75 2/11/75 3/fl/75 4/11/75 5/11/75 6/11/75 7/11/75 8/11/75 9/11/75 2/12/75 3/12/75 4/12/75 5/12/75 6/12/75 7/12/75 8/12/75 9/12/75 . % Cl 0.080 0.056 0.052 0.058 0.060 0.066 0.050 0.044 0.046 0.043 0.051 0.055 0.046 0.043 0.038 0.040 0.040 0.050 0.043 0.055 0.042 0.047 0.055 - 0.050 0.047 0.043 0.043 0.044 0.042 0.042 0.044 0.042 0.050 0.046 0.042 0.041 0.040 0.040 0.052 0.041 0.040 \ (Total t as) 0.48 0.48 0.45 0.46 0.46 0.45 0.50 0.44 0.44 0.43 0.43 0.43 0.44 0.42 0.41 0.40 0.40 0.40 0.41 0.62 0.43 0.41 0,43 NO SAMPLE 0.42 0.42 0.42 0.43 0.44 0.41 0.43 0.41 0.40 0.43 0.48 0.48 0.47 0.52 0.51 0.53 0.51 0.51 % k2o 0.92 0.91 0.91 0.91 0.92 0.92 0.93 0.92 0.93 0.93 0.93 0.93 0.93 0.93 0.92 0.93 0.93 0.93 0.92 0.93 0.91 0.92 0.93 - 0.91 0.91 0.92 0.93 0.92 0.93 0.92 0.92 0.91 0.94 0.90 0.90 0.89 0.91 0.90 0.91 0.91 0.89 t ir t 1t i 1 1i1I1 1 * l | J|11 1 j 1f ;; I\ !1 i: I1 *1 r[-l * 11 <2 * a > 07 .! ; *i i? 1 { r a1 1 TABLE C.2 (CONT'D) Date -LO/.12/75 11/12/75 12/12/75 13/12/75 14/12/75 31/12/75 1/ 1/76 2/ 1/76 3/ 1/76 4/ 1/76 5/ 1/76 6/ 1/76 7/ 1/76 8/ 1/76 9/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/ 1/76 14/ 1/76 15/ 1/76 16/ 1/76 17/ 1/76 18/ 1/76 19/ 1/76 20/ 1/76 21/ 1/76 X Cl 0.041 0.039 0.050 0.042 0.048 0.059 0.042 0.045 0.043 0.046 0.038 0.038 0.046 0.039 0.046 . 0.046 0.046 0.040 0.041 0.044 0.042 0.048 0.041 0.054 0.039 0.040 0.041 112 t so, (Total S as) 0.54 0.52 0.55 0.58 0.54 0.46 0.46 0.46 0.46 0.46 0.46 0.46 0.47 0.46 0.45 0.46 0.44 0.44 0.44 0.46 0.47 0.50 0.47 0.46 0.44 0.43 0.43 % k2o 0.91 0.91 0.91 0.90 0.91 0.90 0.91 0.92 0.91 0.92 0.91 0.91 0.91 O. 8 9 O. 9 2 0.91 O.9 1 O.9 1 0.92 0.9 I 0.91 0.91 O. 9 0 0.91 0.91 O.9 I 0.92 I 1 ii t( j j Ii 7 i I 113 TABLE C.3. RESULTS FROM ANALYSES OF DISCARD DUST Date 7/10/75 8/10/75 9/10/75 10/10/75 n/10/75 12/10/75 13/10/75 14/10/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 24/10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 1/11/75 2/11/75 3/11/75 4/11/75 5/11/75 6/11/75 7/11/75 8/11/75 9/11/75 2/12/75 3/12/75 4/12/75 5/12/75 6/12/75 7/12/75 8/12/75 9/12/75 % CL 0.99 1.01 0.89 0.96 0.96 1.35 2.51 1.64 2.40 1.88 1.86 1.77 - 2.08 1.41 3.08 5.11 4.28 4.01 4.39 3.89 . 6.79 6.62 5.85 5.82 3-09 1.35 1.07 .0 .7 5 0.83 5.33 4.14 2.64 7.14 4.76 5.75 6.24 4.76 t SO3 (Total S as) 7.09 7.22 7.28 7.48 7.39 7.59 10.34 6.65 NO DUST DISCARDED 7.35 6.00 6.02 6.12 NO SAMPLE. NO DUST DISCARDED 6.23 5.65 6.05 5.56 5.49 5.22 5.15 4.84 NO SAMPLE NO SAMPLE 3.62 3.27 3.10 2.81 4.08 4.12 4.11 3.76 4.13 5.84 6.37 6.51 . 12.66 6.2B 6.41 6.97 5.41 X K2 0 7*37 7.45 7.34 7.66 7.48 8.26 11.27 7.65 3.16 7.32 7-22 7.13 7.83 6.70 8.60 10.07 9.28 8.63 8.87 8.21 * 9.62 9.38 8.57' 8.34 7.46 5.83 5.47 4.67 5.42 9.36 8.78 7.8l 16.28 9.10 9.88 10.70 8.24 ACH C06Z13 TABLE C.3 (CONT'D)' Date To/12/75 11/12/75 >2/12/75 13/12/75 16/12/75 31/12/75 1/ 1/76 2/ 1/76 3/ 1/76 6/ 1/76 5/.1/76 6/ 1/76 7/ 1/76 8/ 1/76 9/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/ 1/76 16/ 1/76 15/ 1/76 16/ 1/76 17/ 1/76 18/ 1 / 7 6 19/ 1/76 20/ 1/76 21/ 1/76 X Cl 5.35 5.52 5-53 5.68 - 1.98 - <.55 4.58 4.03 - 3 .O6 4.50 1.20 1.11 1.01 1.20 0.76 O. 9 2 O.8 O O. 7 8 0.69 0.72 0.76 0.86 2 S03 (Total S as) 5.5I 4.90 5.62 NO SAMPLE 5.54 NO SAMPLE 6.19 NO SAMPLE NO SAMPLE 4.62 4.62 6.70 NO SAMPLE 5 .3 I 4.51 5.43 5.27 5.16 5-37 4.94 4.89 4.41 4.50 4.28 5.01 6.81 6.28 % k2o 8.27 8.20 8.66 8.65 6.88 - 7.37 7.38 6.89 7.25 7.60 6.07 5.79 5-54 6.10 5.16 5.15 4.62 . A .66 5.32 5.19 4.84 \ ACP. CC6l<t 115 TABLE C,4. RESULTS FROM ANALYSES OF RETURN DUST Da te 7/10/75 8/10/75 9/10/75 10/10/75 11/10/75 12710/75 13/10/75 16/10/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 26/10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 1/11/75 2/11/75 3/11/75 6/11/75 5/11/75 6/11/75 . 7/11/75 8/11/75 9/11/75 10/11/75 2/12/75 3/12/75 6/12/75 5/12/75 6/12/75 7/12/75 8/12/75 9/12/75 t Cl 0.80 0.73 0.54 0.73 1.59 1.36 1.29 2.46 2.29 2.15 2.33 1.81 1.86 1.76 2.56 1.53 1-52 3.82 3-41 3-87 3.48 3.02 3.51 ' 6.22 4.66 6.37 3.96 4.00 1.27 1.00 0.79 0.54 .8.44 - 4.00 2.52 2.69 2.58 1.00 0.79 0.58 3.68 \ (Total S as) 6.25 6.26 5.36 7.26 7-05 7-79 6.21 8.16 6.91 6.80 7.20 5.61 6.51 5.86 6.56 6.00 5-37 5.36 4.87 4.84 4.48 6.75 4.24 NO SAMPLE 2.58 2.84 3.22 2.47 2.70 3.60 3.67 3.77 3.54 3.30 NO SAMPLE 5.65 5.97 6.51 6.96 3-69 3.81 3.57 4.68 i % K2 6.22 6.25 5.15 7.54 8.20 8.90 7.02 9.83 8.54 8.32 8.73 6.71 7.73 7.08 8.44 6.92 6.60 8.63 7.75 8.21 7.47 6.91 7.23 6.49 7.07 9.05 6.28 6.69 5.20 4.91 4.92 4.50 3.89 10.70 9.63 8.58 8.33 4.94 4.93 4.61 6.59 ACM CC8215 TABLE t . 4 (CONTl D) Dace -10/12/75 n / 12/75 12/12/75 13/12/75 IV I 2/75 31/12/75 1/ 1/76 2/ 1/76 3/ 1/76 V 1/76 5/ 1/76 6 / 1/76 7/ 1/76 8/ 1/76 9/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/ 1/76 IV 1/76 15/ 1/76 16/ 1/76 17/ 1/76 18/ 1/76 19/ 1/76 20/ 1/76 21/ 1/76 X ci 2.51 2.36 3-69 6.06 3.02 1.56 1.78 1.77 2.42 3.55 2.37 2.56 1.35 1.76 2.30 1.06 0.93 0.80 0.79 0.78 0.78 0.65 0.58 ` 0.69 0.58 0.72 0.87 116 X SO3 (Total S as) 3.68 3 . 15 4.22 4.07 4.06 5.09 5.14 4.70 4.62 4.48 3.36 3.43 3.23 4 .9 1 . 3.71 4.74 5.05 5.14 4.82 4.85 4.41 4.27 4.11 4.19 4.56 4.47 4.24 % k2o 4.60 4.22 5.94 6.05 5.16 5.60 5.99 5.65 5.82 6.33 4.29 4.84 3.58 5.56 4 .9 I 5.05 5.^6 5.37 5.10 4.93 4.59 4.33 4.10 4.22 4.7O 4.82 4.83 H7 TABLE C.5. Btu AND CHLORINE CONTENT OF CHLORINATED HYDROCARBONS (SAMPLES FROM LIKE TO KILN) M aterial Date - Btu/lb % Chlorine At ip h a tic 24/10/75 27/10/75 28/10/75 3/11/75 4/11/75 13/11/75 18/11/75 Average 13400 11390 10750 8900 8590 8340 8410 9970 Aromatic + Comp 1ex 10/12/75 11/ 1 2/75 12/12/75 13/13/75 15/12/75 Average 9530 9500 8820 9320 9310 9300 P olychlorinated 3/ 1/76 . 11380 Biphenyl 4/ 1/76 11590 5/ 1/76 11880 6/ 1/76 12170 7/ 1/76 12070 8/ 1/76 12050 9/ 1/76 Average 12000 11880 21.52 32.76 39.40 42.23 43.52 42.38 43-38 37.88 40.56 45.91 44.16 41.80 40.48 42.58 36.16 37-75 34.90 33.90 33.19 34.93 33-90 34.97 V isc o sity (Centpoise i 23C) S p e c ific G ra v ity - 1.13 40.8 1.27 17.0 1.18 * f > CCc2 17 118 r\ TABLE C.6. Btu, S AND Cl'ANALYSES FROM NO. 6 FUEL OIL Date ' 7/10/75 8/10/75 9/10/75 10/10/75 14/10/75 15/10/75 16/10/75 17/10/75 20/10/75 21/10/75 22/10/75 23/10/75 24/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 3/11/75 4/11/75 5/11/75 6/11/75 7/11/75 lO /lt/7 5 11/11/75 12/11/75 2/12/75 3/12/75 4/12/75 5/12/75 8/12/75 9/12/75 10/12/75 11/12/75 12/12/75 15/12/75 30/12/75 2/ 1/76 5/ 1/76 6/ 1/76 7/ 1/76 8/ 1/76 9 / 1/76 12/ 1/76 9/ 1/76 Btu/lb 17809 17222 17978 17942 18106 18008 17920 17988 17924 17877 17984 18189 17701 17901 17862 17970 17917 17889 18135 18077 18047 18079 18036 17979 17782 .17719 17700 17676 17714 17648 17679 17732 17695 17596 17655 17504 17678 17565 17498 17433 17630 17900 17719 17990 17896 % s 2 .AO 2.42 2.34 2.26 2.26 2.28 2.31 2.30 2.24 2.24 2.09 2.23 2.22 2.24 2.22 1.95 2.29 1.77 1.62 1.57 1.55 1.5* 1.56 1.26 1.89 2.06 2.54 2.28 2.56 2.58 2.55 2.52 2.58 2.50 2.56 2.64 2.59 2.41 1.61 x 2.07 2.28 2.47 2.42 1-96 1.95 . * C! Average 7/10/75 - 22/10/75 X C hlorin e - 0.0285 Average 23/10/75-12/11/75 X Chlorine 0.064 Average 2/12/75-15/12/75 t Chlorine 0.047 Average 30/12/75-9/1/76' X Chlorine 0.030 ACM c c e 2 TABLE C.6 (COHT'D) Date 14/ 1/76 15/ 1/76 16/ 1/76 19/ 1/76 20/ 1/76 21/ 1/76 22/ 1/76 Btu/lb 1-7935 17952 18025 17762 17846 18005 18160 19 %S 1.54 2.00 1.98 1.92 1.89 1.5* 1.42 X Cl Average 12/1/76-22/1/76 % Chlorine - 0.038 \ ACA CCfl19 120 TABLE C.7. DAILY RECORD OF PRODUCTS AND MATERIALS CONSUMPTION Date SLURRY FEED (tons/24 hr) CHLORIN. HC (gsl/24 hr) CLINKER (tons/24 hr) DISCARD DUST (tons/24 hr) 7/10/75 8/10/75 9/10/75 10/10/75 11/10/75 12/10/75 13/10/75 I V ! 0/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 24/10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 1/11/75 2/11/75 3/11/75 V I 1/75 5/11/75 6/11/75 7/11/75 8/11/75 9/11/75 1653 1676 1681 1658 1585 1325 1632 1604 1639 1646 1661 . 1682 1669 1669 1661 1664 1681 1686 1669 1732 1656 652 .1621 1633 1647 1658 1673 1679 1679 1692 1693 1698 1670 1701 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1170 1440 1440 1440 1440 1440 1440 2670 2880 2880 2880 2880 930 0 0 0 0 0 1043 1041 1073 1027 1029 845 1036 1047 1064 1058 1065 1081 1079 1071 1078 1046 1082 1079 1067 1112 1061 1040 1021 1032 1027 1029 1057 1030 1010 1044 1054 1058 1030 1056 . 40.44 63.44 24.84 65.66 0 18.94 6.96 23.00 0 12.90 17-86 15-32 4.93 15.97 0 45.36 11.96 19.72 20.96 16.42 18.32 42.58 41.36 37-17 56.52 63:32 38.92 80.66 107.72 72.82 60.46 59.90 73.50 64.26 . 2/12/75 1622 430 1038 20.58 3/12/75 1627 v' 0 1042 18.30 V I 2/75 1644 0 1039 37.72 5/12/75 1653 1640 ' 1028 61.40 6/12/75 1670 2187 1056 38.26 7/12/75 1655 1351 1074 0 8/12/75 1666 1117 1043 51.36 _y 9/12/75 1664 2121 984 129-78 ACM CCticC 121 TABLE C.7. (CONT'D) Date 10/12/75 11/12/75 12/12/75 13/12/75 14/12/75 15/12/75 31/12/75 1/ 1/76 2/ 1/76 3/ 1/76 4/ 1/76 5/ 1/76 6/ 1/76 7/ 1/76 8/ 1/76 9/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/ 1/76 I V 1/76 15/ 1/76 16/ 1/76 17/ 1/76 18/ 1/76 19/ 1/76 20/ 1/76 21/ 1/76 SLURRY FEED (ton$/24 hr) 1676 1669 1669 1661 1653 1658 1606 595 1608 1630 1629 1627 1616 1612 1615 1619 1621 1627 1636 1639 1618 1633 1596. 1606 1635 1608 1602 1629 CHLORIN. HC (gal/24 hr) 1997 1308 1821 1691 1869 1909 0 0 0 2029 2210 2168 1985 604 1230 1367 0 0 0 0 0 0 0 0 0 0 0 0 CLINKER (tons/24 hr) DISCARO OUST (tons/24 hr) 994 999 1024 985 956 980 1024 1035 1039 1017 1005 970 1034 944 1045 1027 1020 1011 1032 1040 1028 1046 1026 979 1026 1020 991 1016 -, 127.66 113.76 79.12 125.96 157.68 130.30 24.72 0 6.11 55.43 70.90 116.98 19.86 137.39 4.54 32.10 43.92 61.20 40.52 32.36 42.80 19.10 13.38 85.76 47.62 32.22 66.56 55-75 ACf* G G 6 2 *1 TABLE C.8. MATERIAL BAL DATE 7/10/75 8/10/75 9/10/75 10/10/75 11/10/75 12/10/75 13/10/75 14/10/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 24/10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 50/10/75 31/10/75 1/11/75 2/11/75 3/H/75 4/11/75 5/11/75 6/11/75 CH10Rlife INPUT (lb/24 h r) SLURRY 16 CHLORINATED FEED OIL HYDROCARBON 2,645 1.877 1,748 1.923 1.902 1,801 1,632 1,1i12 1,508 l,4 lfi 1,69* 1,850 1,535 1.435 1,262 1.331 1.365 1,686 1.635 1,905 1,391 1,553 1,783 1,502* 1,667 1,558 1.639 1,444 1,678 1.t|21 1,!|22 81 81 82 81 78 '67 01 60 81 81 81 81 81 81 80 81 185 183 183 190 182 182 183 181 186 179 179 178 179 184 182 0. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2,865 3.502 3,959 4,416 4,873 6.611 , 6,662 12,316 13.284 13,284 13,284 13,284 13,743 6,574 0 * Cnlcul ated from average data TOTAL 2.726 1,958 1,830 2,004 1.980 1,868 1,713 1,492 1,589 1,497 1,775 1,931 1,616 1.516 1,342 1,412 4.375 5,371 5,577 6.511 6,446 8,146 6,608 13.999 15,117 15,021 14,902 14,906 15.400 6,179 1,604  ' j ANCE FOR CHLORINE CHLORINE RETAINED (lb/26 hr) CLINKER DISCARD DUST TOTAL 688 666 730 836 658 .490 622 649 660 656 746 627 690 900 647 607 649 647 619 689 658 707 653 805 698 803 676 659 566 647 653 801 1,281 442 1,261 0 511 369 754 0 619 672 570 174 575 0 1.887 337 1,215 2,142 1,606 1.469 3,738 3.218 3,970* 6,036* 8,599 5,153 9.637 12.539 4,500 1,632 1.689 1.967 1,172 2,097 658 1,001 971 1.603 660 1.275 1,618 1,197 664 1,675 647 2,494 966 1,862 2,761 2,095 2,127 4,445 3.871 6.775 6,736 9,402 5.829 10,096 13,105 5.167 2.285 i TOTAL X j RETAINED | i 1 54.6 99.6 64.0 104.6 33.2 53.6 56.7 94.0 41.5 85.2 79.9 62.0 53.5 97.3 48.2 176.6 22.5 34.7 69.5 32.2 33.0 54.6 65.0 36.1 44.5 ' 62.6 39.1 67.7 8 5 .I 83.3 142.4 TABLE C.8 (CONT'D) DATE 7/H/75 8/11/75 9/11/75 2/12/75 3/12/75 V I 2/75 5/12/75 6/12/75 7/12/75 8/12/75 9/12/75 10/12/75 11/12/75 12/12/75 13/12/75 l i * / 12/75 31/12/75 1/ 1/76 2/ 1/76 3/ 1/76 V -l/ 7 6 5/ 1/76 6/ 1/76 7/ 1/76 8/ 1/76 9/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/ 1/76 CHLORINE INPUT (1b/2* hr) SLURRY #6 CHLORINATED . FEEO O IL HYDROCARBON 1.696 1.603 1.701 183 181 * 182 0 0 0 1.692 12* 1.367 . 33 1.3*8 33 1.322 32 1.336 1,721 r .366 3 13 1*2 1.331 1.376 132 30 1.302 3* 1.669 13* 1.395 133 1.587 13 2,325 0 0 8,868 11,626 7 .3 0 6 6,0*0 11.670 10,287 7.626 10.213 8.977 10,107 1.895 1.360 1.667 1,*02 1,699 1.236 1,228 1.568 1,260 1.556 1.556 1.562 1.309 1,366 85 87 87 83 82 83 83 85 83 81 109 109 109 108 0 0 0 8.657 9.8** 8 ,9 2 8 7.950 2.366 5,070 5.668 0 0 0 0 TOTAL 1.677 1.586 1.883 3.961 1.500 1,*8l 10,322 13.293 9 .1 5 8 7,568 12.933 11.791 9.062 12,016 1 0 ,5 0 5 11.825 1.980 1.627 1.536 10,1*2 11,625 10,2*7 9.261 3.999 6,613 7.103 1.665 1.671 I,*I8 1.652 CHLORINE RETAINED (1b/2* hr) CLINKER DISCARD DUST TOTAL 783 1,282 2 ,0 6 5 597 1,102 1.699 739 1.067 1,806 6** 2 .1 9 6 2,838 . 60* l.55 2.119 708 1 ,9 9 2 2,700 678 8,768 9,**6 612 3.662 6,256 601 0 601 793 6,*10 7,203 787 12,355 13,162 676 13,660 16,336 699 12,559 13.258 655 8,751 9.606 630 1 * ,208 l*,838 593 1 7 ,9 0 6 18.699 596 979 1.573 600 0 600 603 2*2 8*5 67 6 .6 3 6 5.105 603 6.652 7.055 679 10,715 11,396 6*1 1.601 2 ,2*2 5*8 10,991 11.539 627 278 905 575 2,889 3. * 6 * 571 1.056 1.625 607 1,359 1,966 619 818 1.637 603 777 1.380 TOTAL t RETAINED 123.1 07.3 95.9 72.0 1*1.3 182.3 91.5 32.0 6.6 95 .6 101.6 121.6 1*6.3 78.3 1*1.2 156.* 7 9 .6 *2.0 55.1 50.3 61.8 111.2 2*.2 288.5 1*. 1 * 8.8 976 117-6 101.3 95.0 ALh 3 TABLE C . (CONT'D) DATE tA/ 1/76 IS / 1/76 16/ 1/76 17/ 1/76 18/ 1/76 19/ 1/76 20/ 1/76 21/ 1/76 CHLORINE INPUT (tb/2<l hr) - SLURRY 16 CHLORINATED FEED OIL HYDROCARBON 1,1*21 1,372 1.532 1,317 1,776 1.256 1,282 1.336 - 108 108 107 108 108 107 107 109 0 0 0 0 0 0 0 0 TOTAL CHLORINE RETAINED (1b/24 h r ) ' CLINKER DISCARD OUST TOTAL 1.532 1,480 1.639 1.625 1.876 1.361 1.389 1,665 . 617 607 616 72*i 575 571 575 589 650 351 214 1.338 657 464 1,012 959 1,267 958 830 2,062 1,232 1,035 1.587 1.548 TOTAL t RETAINED 1 82.7 64.7 50.6 144.7 65.7 76.0 114.2 107.1 n T O orr M t ,tf k 125 TABLE C.9. MATERIAL BALANCE FOR I^O OATE "K?0 INPUT (tons/24 hr) SLURRY FEED 7/10/75 8/10/75 9/10/75 10/10/75 11/10/75 12/10/75 13/10/75 14/10/75 15/10/75 16/10/75 17/10/75 18/10/75 19/10/75 20/10/75 21/10/75 22/10/75 23/10/75 24/10/75 25/10/75 26/10/75 27/10/75 28/10/75 29/10/75 30/10/75 31/10/75 1/11/75 2/11/75 3/11/75 4/11/75 5/11/75 6/11/75 7/11/75 8/11/75 9/11/75 15-21 15.25 15.30 15.09 14.58 12.18 15.18 14.76 15.24 15.31 . 15.45 15.51 15.52 15.52 15.28 15.48 15.63 15.68 15-35 16.11 15-07 15.20 15.08 15.02* 14.99 15:09 15.39 15.61 15.45 15.74 15.58 15.62 15.20 15.99 2/12/75 3/12/75 4/12/75 5/12/75 6/12/75 7/12/75 . 8/12/75 9/12/75 10/12/75 11/12/75 14.60 . 14.54 14.63 15.04 15-03 15.06 15.16 14.81 15.25 15.19 v K?0 RETAINED (tons/24 hr) CLINKER OUST TOTAL 11.78 11.66 12.23 9.04 13.07 9.13 11.29 12.04 15-64 12.27 13-95 13.40 13.06 13.71 16.28 13.39 11.47 12.95 8.86 3-79 10.61 13* 62 13.99 9.91 5.03 3.70 4.44* 4.94 5.66 10.44 12.44 12.70 11.84 12.57 2.98 4.73 1.62 5.03 0 1.56 0.78 1-76 0 1.18 1.31 1.11 0.35 1.27* 0 3-55 0.80 1.70 2.11 1.52 1.58 3-78 3.40 2.91* 4.42* 6.09 3.65 6.91 8.88 5.43 3-52 3.28 3.43 3.48 14.76 16-39 14.05 14.07 13.07 0.69 2.07 13.80 15.64 13.45 15-26 14-51 13.41 14-98 16.28 16.94 12.27 14.65 10.97 15.31 12.19 17-40 17.39 12.82 9.45 9.79 8.09 n .85 14.54 15.87 15.96 15.98 15.27 16.05 9.76 13-34 10.49 11.92 9-50 9.02 7.72 5.51 3.58 4.80 1-93 1.61 2.94 10.00 3.48 0 5-50 0.69 10.56 9.33 11.69 14.95 13.43 21.92 12.98 9.02 3.22 16.20 14.14 14.13 TOTAL % RETAINED 97.0 107.5 91.8 93.2 89.6 87.8 79.5 93.5 102.6 87.8 98.8 92.8 86.4 96.5 106.5 109.4 78.5 93-4 71.5 95.0 80.9 114.5 115.3 85.4 63.0 64.9 52.6 75.9 94.1 100.8 102.4 102.3 100.5 100.4 80.1 102.1 91.8 145.7 86.4 59.9 87.2 109.4 92.7 93.0 * C alcu lated from average data. At* CC6225 126 TABLE C.9 (CONT'D) DATE K?0 INPUT (tons/24 fir) SLURRY FEED 12/12/75 13/12/75 14/12/75 31/12/75 1/ 1/76 2/ 1/76 3/ 1/76 4/ 1/76 5/ 1/76 6/ 1/76 7/ 1/76 8/ 1/76 5/ 1/76 10/ 1/76 11/ 1/76 12/ 1/76 13/1/76 14/ 1/76 15/ 1/76 16/ 1/76 17/ 1/76 18/ 1/76 19/ 1/76 20/ 1/76 21/ 1/76 15.19 14.95 15.01* 14.45 14.51 14.79 14.83 14.99 14.80 14.70 14.67 14.37 14.89 14.75 14.80 14.89 15.08 14.72 14.86 14.52 14.45 14.88 14.63 * 14.58 14.99 * C alcu la ted from average d ata. K?0 RETAINED (tons/24 hr) CLINKER DUST TOTAL TOTAL % RETA 1NED 5-53 3.94 4.59 12.29 13.87 14.23 7.83 14.27 7.57 6.00 11.89 8.67 4.72 10.71 11.32 11.87 14.04 13.16 12.76 14.88 13.90 13.03 11.73 14.47 11.38 6.69 10.71* 13-64 1.70* 0 0.42* 4.01 5.22 8.63 1.37 0.93* 0.33 2.44 2.66 3.54 2.24 1.97 2.21 0.98 0.62 4.01 2.14 1.71 3.45 2.70 12.22 80.4 14.55 18.23 97.3 121.2 13.99 13.87 96.8 95.6 14.65 11.84 99.0 79-B 19.49 130.0 16.20 109.4 7.37 50.1 21.82 148.7 9.00 . 62.6 7.16 48.1 13.37 14.87 14.11 90.6 100.4 94.8 16.01 106.2 15.37 13.74 104.4 92.5 15-50 106.7 17.91 15-17 13.44 17.92 123.9 101.9 91.9 122.9 14.08 93.9 a c n cceiit: * APPENDIX D DUALITY OF CEMENT PRODUCED \ CG82*1 au* 129 APPENDIX D QUALITY OF CEMENT PROOUCED While remote, the p o s s ib ilit y of n o n -v o latile chlorinated hydrocarbons not being destroyed and remaining with the c lin k e r was considered, i t is common knowledge that e x ce ssiv e q u a n titie s of organic m aterials can d e trim e n ta lly a ff e c t s e t t in g , a i r en train in g and compressive strength c h a r a c t e r is t ic s of cement. A fte r preparing the d a ily composite c lin k e r sample fo r a n a ly s is , the q uantity o f sample which had not undergone s iz e reduction was retain ed . These were combined into a composite of several d ays, three such composites being prepared for each of the arom atic plus complex chlo rinated hydrocarbon burn, the PCB burn and the f in a l b a se lin e burn. Each composite, a f t e r being crushed to approximately on e-h alf Inch, was blended w ith the c o rre c t amount of the normal production gypsum used at S t. Lawrence Cement. The clinker-gypsum blends were ground in a laboratory b a ll m i l l . The grinding was timed and samples were taken at su ita b le in te rv a ls during grinding to determine surface area. The cements were produced to conform to the requirements of Symbol 10 cement a* designated in the Canadian Standards A sso ciatio n (CSA) Standard A5 "Portland Cements", and Type I cement as defined in the American S o ciety fo r Testin g and M a teria ls (ASTM) designation C150, "Standard S p e c ific a tio n fo r Portland Cement". Each cement was tested In conformance to the follow ing ASTM methods: C185, Test fo r A ir Content of H ydraulic Cement M ortar;, C109, T est fo r Compressive Strength of H ydraulic Cement Mortars (using 2-1n.'cube specimens); C359* T est fo r F a ls e Set of Portland Cement (Mortar Method); C20A, T est fo r Fineness of Portland Cement by A ir P erm eability Apparatus; C 151 Test fo r Autoclave Expansion of Portland Cement; C191, T est fo r Time o f Setting o f H ydraulic Cement by V ica t Needle. ACM CCtcZfc 130 ,r~\ Tq ensure that d iff re n c e s in q u a lity were not due to unexpected d iffe re n ce s in cement com position, a chemical a n a ly s is was performed on each cement te ste d . Procedures followed were those normally used for control purposes a t S t . Lawrence Cement. The technique used fo r determ ination of the concentration of each oxide was fu sio n with lith iu m tetrab o rate followed by x-ray fluorescence a n a ly s is . T h is method Is convnonly used for cement a n a ly se s. With the exception of lower K^O content in the cements from c lin k e r s produced w hile burning ch lo rin ate d hydrocarbons, there were no s ig n if ic a n t d iffe re n ce s In t h e ir chemical compositions (Tables D .1, 0.2 and 0 .3 ) . D iffere n ces in p h ysical c h a r a c t e r is t ic s found by te stin g these cements were: S ettin g Time - The water required to obtain normal co n sisten cy was g re ater and se ttin g time sho rter fo r b a s e lin e c lin k e r cements. Shorter time of se t and higher water requirement were a ttrib u te d to the higher a lk a li content of the baseline c lin k e r. F a lse Set F a lse se t was severe on the cements from c lin k e r produced during baselTne burning. There m s no in d ic a tio n of f a ls e se t In cements from c lin k e r produced w hile burning chlo rin ated hydrocarbons. The improvement was due to lower a lk a l i content of [ the cement. In t h is regard, the determination of f a ls e se t by ASTM Method C451* 'T e s t for Fa lse Set o f Portland Cement (Paste Method)" would not have shown as dramatic an improvement. This la t t e r te st takes into account the water requirement of cements, and therefore compensates for the g re ater quantity of water required by the higher a lk a li cements. Compressive Strength - Higher i n i t i a l and lower u ltim ate stre n g th s,a re c h a r a c t e r is t ic s of higher a lk a li content cements. The o n ly apparent d iffe re n ce s in cements from c lin k e r s produced w ith and without burning of ch lo rin ated hydrocarbons were those expected because o^ the d iffe re n ce s in a lk a li content. *i> C Z bZt t It L- 131 TABLE D .l. CEMENTS FROM CLINKER PRODUCED DURING BASELINE BURN Sample Number 7 Chemical Tests Loss on Ig n itio n SI02 (*) Al-.O- " Fe^ol " CaO 11 hgO " SO3 " K2 0 " Free CaO ft) C3 S (*) c2 s 11 C3A " Ci,AF '* ft) 1.52 .20.69 5.93 2.20 63-42 2.52 2.7* 1.15 0.44 50.1 21.6 12.0 6.7 Physical Tests Fineness ^ Blaine (an /g) Passing 200 mesh ft) 3406 96.4 Settin g Time N.C. Penetration {mm) N.C. Water ft) In it ia l (minutes) Final (minutes)- 10.0 22.5 115 225 False Set Q Temperature F Penetration (mm) 3 minutes 5 minutes 8 minutes 11 minutes Remix 70 50 50 21 16 50 Compressive Strength 1 day (p s i) 3 day (p s i) 7 day (p s i) 2 6 day (p si) A ir Content Water (X) Flew (t) A ir (*) 2050 2910 3480 4130 V v 73.0 8 0 .6 7.2 Autoclave Expansion ft) 0.07 89 1.51 20.59 5.90 2.22 63.39 2.52 2.79 1.23 0.52 50.8 20.8 U -9 6.7 1.43 20.53 6.00 2.24 63.35 2.52 2.89 1.22 0.54 50.0 21.2 12.1 6.8 1 3579 97.6 3607 98.8 9.5 2 3 .O 106 216 9.0 23.0 106 221 73 73 37 " 34 7B 44 43 50 50 2010 2990 3610 4470 . 1910 3210 3880 4530 73.0 8 6 .8 7.6 0.09 73-0 8 6 .0 8 .3 0.05 Aver. I .49 20.60 5.94 2.22 63.39 2.52 2 .8I 1.20 0.50 50.3 21.2 12.0 6.7 * 3530 97.6 9.5 22.8 109 . 221 72 40 22 10 8 50 1990 3040 3660 4380 73.0 84.5 7.7 0.07 ACM CC8E3C *{ ' 132 TABLE 0 .2 . CEMENTS FROM CLINKER PRODUCED DURING AROMATIC PLUS COMPLEX CHLORINATED HYDROCARBON BURN Sample Number Chemical T ests Loss on Ig n itio n S 1O2 (?) A120 , (?) Fe203 (*) CaO \%) HgO (?) SOj (?) *20 (t) Free CaO (!) C3S (?) C2S (?) c3a (?) Ci,AF (?) ($) Physical Tests Fineness ,, B la in e (cm /g) Passing 200 mesh ($) Settin g Time N.C. Penetration (mm) N.C. Water ( !) In it ia l (minutes) Fin al (minutes) False Set Temperature F Penetration (mm) 3 minutes 5 minutes 8 minutes 11 minutes Remix Compressive Strength 1 dey (p si) 3 day (p si) 7 day (p si) 28 day (p s i) A ir Content Water (t) Flow (?) A ir (t) Autoclave Expansion ($) 4 1.63 20.37 5.85 2.26 63.19 2.51 2.81 0.66 O. 5 0 51.8 19.A 11.7 6.9 3626 98.0 10.0 22.0 116 238 70 50 50 50 50 50 1900 3520 6590 \ 5900 - 71.0 80.0 9.2 0.03 5 1.86 20.51 5.79 2.23 62.99 2.69 2.69 0.53 0.59 50.8 20.6 M .6 6.8 6 1.63 20.83 5.89 2.22 63.56 2.52 2.89 O. 6 9 0.48 49.4 22.5 1*1.9 6.7 3562 97-6 3561 97-2 10.0 22.0 135 250 9.5 22.0 135 250 72 . 73 50 50 50 50 50 50 50 50 50 50 1640 3200 4360 6050 ' 1890 3180 4160 5660 72.0 90.5 9.0 0.04 72.0 81.3 6.7 0.02 Aver. 1.71 20.57 5.84 2.26 63.25 2.51 2.80 0.63 0.52 50.7 20.8 11.7 6.8 3583 97.5 9.8 22.0 129 246 72 50 50 50 50 50 1810 3300 6370 5800 71.6 83.9 8.3 0.03 ACM CC8 231 133 TABLE D.3. CEMENTS FROM CLINKER PRODUCED DURING POLYCHLORINATED BIPHENYL BURN. Sample Number Chemical T e sts Loss on Ig n itio n (2) " SiO, (i) A l,0 , (1) Fe20 , (%) CaO T?) *gO (%) S03 () k2o (%) Free CaO (4) c3s (*) C2S (%) C,A U ) Ci,AF (*) Physical Tests Fineness ^ Blaine (an /g) Passing 200 mesh- (%) S ettin g Time N.C. P enetration (mm) N.C. Water (%) In it ia l (minutes) Final (minutes) False Set ^ Temperature F Penetration (mm) 3 minutes 5 minutes 8 minutes 11 minutes Remix Compressive Strength 1 day (p s i) 3 day (p si) 7 day ( p s i) 28 day (p si) A ir Content Water (%) Flow (*) A ir (*) Autoclave Expansion (2) \ 1 1.46 20.51 5 .9I 2.16 63.24 2.53 2.94 1.01 0.53 50.4 20.9 12.0 6.6 3388 56.0 10.0 22.0 153 273 70 50 50 50 50 50 1980 3080 3770 4620 71.0 88.0 7-7 0.05 2 1-58 20.72 5.86 2.18 63.27 2.56 2.64 0.74 0.57 50.0 21.8 11.9 6.6 3 Aver. 1.48 20.87 5.89 2.16 63-65 2.58 2.61 0.96 0.69 50.3 22.0 12.0 ' 6.6 1.51 20.70 5.89 2.17 63.39 2.56 2.73 0.90 O .6 O 50.2 21.6 12.0- 6.6 3579 97.0 11.0 22.0 125 250 71 50 50 50 50 50 1800 3000 4000 5110 71.0 84.5 7-9 O. 0 6 3593 96.8 9-0 2 2 . 112 235 71 50 50 50 50 50 1990 3260 4090 5080 ! 71.0 82.2 9-2 0.01 8 3520 96.6 10.0 22.0 130 253 71 `50 50 50 50 50 1920 3110 3950 4940 71.0 B4.9 8.3 0.06 l[ APPENDIX E EQUIPMENT DESCRIPTION AND ASSOCIATED ECONOMICS f ii ACM CC 6 23 3 APPENDIX E EQUIPMENT DESCRIPTION AND ASSOCIATED ECONOMICS In order to burn waste ch lo rin ate d hydrocarbons in a cement k iln , they must be blended to be compatible w ith the handling and storage system and to achieve uniform feed to the k iln . A storage tank and feed system with asso ciated control devices must be in s t a lle d a t the k iln s i t e . For the program to be econom ically v ia b le , the p ric e stru ctu re must be favourable to the o rig in a to r o f the wastes and to the cement producer. The waste d isposal company must be able to generate s u f f ic ie n t revenue to cover shipping, handling, blending, disposal of non-fuel fra c tio n s and re ce iv e some p r o f it . The economics from the point o f view o f the cement producer a re discussed in Section E .2 . E. 1 Equipment D escrip tio n The system in s t a lle d a t the cement plant can v a ry. A simple .in f le x ib le in s t a lla t io n with manual co n tro ls and a used r a i l ca r for storage would cost approximately $25,000. At the other extreme, a j sophisticated system with large storage ca p acity , corrosion re sista n t constructio n, well instrumented could cost $200,000. I Althoug; there a re general s im ila r it ie s between cement k i ln s , the flow ra te of waste ch lo rin ated hydrocarbon depends on the percentage i of ch lo rin e in the w aste, the decrease desi-e.*1 in the potassium and sodium le v e ls of the fin ish e d product and the production ca p acity of the k iln . Thus, fo r the optimum system, the design, en g in eerin g , 1 in s t a lla t io n and operation must be compatible with these param eters: Two systems a re described h e rein ; one is the actu al system in s t a lle d a t the S t. Lawrence Cement Co. k iln and the other has been somewhat a r b i t r a r i l y selected as a b a sic system. Th is b a sic sy ste m 1 assumes that a q u a lit y co n tro lle d blend of chemical wastes w i l l be used. The fuel product would be a low v is c o s it y , single-phase liq u id , non-; c o rro siv e or m ild ly co rro siv e to feerbon s t e e l, not hig hly v o l a t i l e , having an approximate composition o f: 60,000 B tu/g allon, AOt C l , and s p e c if ic g ra v ity o f 1 .2 . Fu els with higher or lower Btu and ch lo rin e can be prepared fo r those a p p lic a tio n s where other compositions a re optimum. ACMi CCB23* 138 At these compositional values, a feed rate of I Igpm into a 1000 ton per day k iln would provide an a lk a li reduction of O .k SX as K20. The Btu-'s provided would be a small percentage of the total heat requirement of the k iln . With waste streams with a lower chlorine le v e l, larger volumes can be used to make up a greater fraction of the heat requirements, in practice up to 10-151 of total Btu Input. Other than capacity and the provision that the k iln be rotary, It makes l i t t l e difference what the primary fuel or fuels are. Wet or dry fuels can burn waste chlorinated hydrocarbons, with consideration given to the increase in dust loading to the dust handling equipment. E . 1.1 entaiI: Cost of basic system The basic system with minimum component cost estimates would Equipment Cost: Tank, 10,000 gal carbon steel Feed Pump, Centrifugal, 7 .5 hp Flame Arrestor Vent Scrubber, Activated Carbon Tank Level Indicator Grounding Tank Berming Unloading Piping Piping, Tank to Kiln, 100' Electrical-Combination Starter Concrete Pad, S ite Preparation Miscellaneous, Hardware ^Painting and NFPA Code Marking Instrumentation Nozzle Assembly Installation: \ Total System Cost, Installed $ 6,000 1,200 100 1,000 2,500 ^00 1000 500^ 1,000 1,500 2,500 2,500 ^00 200 $22,800 ^000 $26,800 ALh CC6235 139 .A schematic representation o f the system is shown in Figure E . l . Operating Cost: Maintenance, k% of investment/year Operating Labour, Based on 330 days/year Operating 2k hr/day 1 hr per s h if t 0 $8.00 E l e c t r i c i t y , 7*5 hp, 8000 hr 3/kwh $1,080 8,000 1,800 Startup o f the system should be achieved with minimum e ffo rt once the system is in s t a lle d and tested for le a k s. Allowance of $1,500 fo r I n i t i a l startu p should be adequate. Safety equipment fo r personnel p ro tectio n and a f i r e exing uisher a t the unloading/pumping s ta tio n can be provided for about $250. E .2 .2 System at the S t. Lawrence Cement Co. plant The actu al system used for the te st (Figure E .2 ) was designed l to contain co rro siv e m a te ria ls as w ell as those with low f la s h p o in ts. So p h isticated equipment Is used to control the feed to the k iln and the system includes traps to avoid vapours being emitted to the atmosphere The storage tank received two lin e r s of "Furoglass" membrane and one lin e r of 2 iMacid proof b rick . Tracks were unloaded via a ce n trifu g a l pump through f i l t e r s into a 55.000 gal storage tank. Fumes exhausted from the storage tank w h ile .unloading tru cks were returned to the truck by a return vent lin e . Fumes emanating from the tanks by normal vapour d iffu sio n or expansion due to temperature were removed by sodium hydroxide and charcoal f i l t e r s . E ith e r o f two ce n trifu g a l pumps was used to feed the k iln s at a pressure o f approximately 120 p s i, the pressure maintained by returning an amount o f m aterial through a su ita b le o r i f i c e to the storage tank. Each k iln was co n tro lle d sep a ra tely using an electro-pneum atic control v a lv e , magnetic flow me^er and tra n sm itte r. The tubing Inserted Insid e the pipe sleeve located on top of the burners was made from tantalum. The tubing was threaded so that a titanium n o z z le , o f co rre ct s iz e to atomize the q uantity o f liq u id input to the k i l n , up to a maximum ca p acity of 4 g allo n s per m inute, could be attached. j ADt* I. I I I T> n x ,, <D--- (P LP 'l ii.l 3 TO KILN 4 RETURN LINE 5 VENT PIPE > 6 LEVEL GAUGE 7 FILLER PIPE F I G U R E E.1 SCHEMATIC DIAGRAM OF BASIC CHLORINATED HYDROCARBON FEED SYSTEM # * (B) |iww^KiiijH. f JHl>* * n n 'i i> W M O 1 w**W AC* CC6236 1 STORAGE TANK 2 KILN FEED PUMPS 3 FILTERS 4 FLOW METER 5 CONTROL VALVE 6- TO KILN HI 7 fLOW RECORDING CONTROLLER 6 TO KILN II 9 TO KILN 1 10 FILTER 11 TO ODOR CONTROL SYSTEM ______ 12 BACK TRAP 13 CAUSTIC SCRUBBER -14 CHARCOAL FILTER 15 TRUCK f i g u r e E.2 SCHEMATIC DIAGRAM OF CHLORINATED HYDROCARBON FACILITIES I .142 i i[ The total- in s t a lle d cost of th is system was over $200,000, comprising the following items: Equipment and In s t a lla t io n : Tank Acid B rick -Tank Lining Teflon- Lined Pumps andF i l t e r s Teflon* Lined Piping Odour Controt Equipment Foundations Dike Pump House S it e C learin g and Leveling Instrumentation Painting - ' Sales Tax ^ Engineering Consultant Travel Expense TOTAL $25,000 35,000 12,000 45,000 5,000 10,000 1,000 10,000 3,000 20,000 5,000 5,000 18,000 8,000 10,000 $212,000 Operating Cost: Maintenance and Operating Labour (cost per year) E le c t r ic it y (23c/hour for 8,000 hours) TOTAL $25,450 1,840 $27,290 Due to the p o te n tia lly dangerous nature of these f u e ls ,, both w ith regard to personnel sa fe ty and environmental co n sid e ra tio n s, an exten sive tra in in g and In stru ctio n program was ca rrie d out, 1 E .2 Economics of Burning Waste Chlorinated Hydrocarbons It is not p o ssib le to 'g iv e a d e f in it e statement on the economics c f burning ch lo rin ated wastes for a l l circum stances. Some cement1plants 'Trademark, E . l . du Pont de Nemours. -,L W-4'iW *39 '* V V I [ I ( AT T\ r\ 143 req u ire a reduction in the a lk a li oxide (K^O * Na^O) content of the < fin ish e d product. In th is ca se , ch lo rin e must be added to the process m a te ria ls, and is a v a ila b le from d iffe re n t sources in d iffe r e n t geographical a re a s. Waste hyd ro chlo ric acid from the ste e l industry may be used where a v a ila b le ; in other areas cement plants purchase calcium c h lo rid e . If reduction of a lk a li is not req u ired , then addition of ch lo rid e would cause the generation of g reater q u a n titie s of dust, thereby increasing dust disposal co sts. Another fa c to r making economic comparisons d i f f i c u l t is the j d iffe re n t waste streams a v a ila b le . In some c a se s, the ch lo rin ated wasti e streams may be s u ita b le for p u rify in g and fu rth e r use. Other waste i streams, such as PCB's and in s e c tic id e s banned for environmental reasons, present a serious disposal problem. Obviously, these d iffe re n t m aterials would not a ll cost the same for use in a cement k iln . However, to present some models, the p ric e stru ctu re used by Chemtrol P o llu tio n S e rv ic e s, he. for blended noncorrosive wastes su ita b le for use in the b a sic system described above has been adopted. At p resen t, Chemtrol plans to charge 80 percent of the fuel value and, where a p p lic a b le , 30 percent of the ch lo rin e value for these blended " fu e ls " . In the examples, the follow ing assumptions have been made: The k iln is a wet process type producing 1,000 t^n- per day vyith a fuel requirement equivalent to 5,150,000 Btu per ton of production. E.2.1 Process requiring addition of chlorine i! I Nj Assume in t h is case that the fuel cost is $1 per m illio n Btujand that it is necessary to reduce the content of the c lin k e r by 0.45 percent. This would require approximately 6,800 pounds of ch lo rin e per day. A 32 percent so lu tio n of calcium ch lo rid e weighing 13 pounds and costing $oil4 per Imperial gallon ($0.053 per pound of ch lo rin e ) is used. In th is ca se, 2,585 g allo n s would be required at a total cost of $359- This could be replaced by >rTro1 Fu el*" containing a nominal 70 percent ch lo rin e and 65,000 Btu per Imperial g a llo n . At a weight o f 13 t pounds per g a llo n , 750 g allo n s would-be required each day. Using the Trademark, Chemtrol Pollution Services Inc. r- AC* \ * i I 144 JfC above mentioned p ric e s tru c tu re , the cost per g allon would be: -9q.1i lrbb CC1l2 xx SIfOa-0c 5,^1 xx ]0300 65,000 Btu x $ 1.00 1,000,000 Btu 80 100 - $0-11*5 - $0.052 TOTAL $0.20 Thus, the 750 g allo n s would co st $150 and would supply 48.8 mi 1Mon Btu. '| The to ta l d a ily tang ible savings then become Cost of Calcium Chloride Cost o f Normal Fuel Replaced Le?s Cost of 'Trol Fuel" Net Savings Per Day $359 $ 49 $150 $258 E.2.2 Process not req u irin g add ition of ch lo rin e The 1,000 ton per k iln and a s p e c if ic heat consumption|of 5,150,000 Btu per ton of c lin k e r produced and a fuel co st equivalent to $1.00 per m illio n Btu is again taken as the model. In t h is case, I however, no c h lo rin e is required and the waste stream contains le ss than m 5 percent c h lo rin e . Th is m aterial is a v a ila b le a t $0.80 per m illio n Btu and is used a t a replacement rate of 10 percent of the Btu requirement or 515,000 Btu per ton of c lin k e r . The fuel normally used on the k iln but replaced by th is waste stream would co st $0,515 per ton of c lin k e r or $515 per day. The d a ily sa v ' ngs can be ca lcu la te d a s : Cost o f Normal Fuel Replaced Cost of 'Trol Fuel" Net D a ily Savings $515 ill 2 $103 Th is saving can be negated by the cost of d iscard ing the i additional dust co lle cte d in the p re cip ita to rs (See Section 6 .1 ), * factor which w ill vary with each cement p la n t. E .2 .3 Total economic co n sid eratio n s I A> C C fi^ The above models are not intended to be a ll in c lu s iv e .| Due to the many v a ria b le s involved, each possib le a p p lica tio n must be assessed based upon economic fa cto rs e x istin g in the lo c a le . At the time the system i i 145 was in s ta lle d a t the S t. Lawrence Cement C o ., It was necessary to use ch lo rin e In two wet k iln s each having a. cap acity o f 1,000 tons per year. For the period January through September 1976, maintenance on the syston amounted to $"18,760- Thus, the fig u re o f $27,290 for y e a rly operating and maintenance cost Is considered a ccu ra te . Under these circum stances, the economic p ic tu re over the period o f a year becomes: Total Savings Using 'T ro l Fuel" (600 k iln operating days $258 per day) Maintenance and Operating Cost In te re st on $212,000 6 lit / y e a r Met $154,800 27,290 23.320 $104,190 Th is estim ated saving , which would have given e x c e lle n t payback on the system, has not beer, r e a liz e d . Low a lk a li requirements have been fa llin g into disfavour because such requirements are not compatible with the goal of energy conservation. With the requirement fo r ch lo rin e a d d itio n , savings must be based only on fuel co sts which. I f a l l fa cto rs I are favourable, would to ta l for the two k iln s : ^ Total Savings Using "Trol Fuel" (600 UM 1 operating days @ $1D3/day) Maintenance andOperating Costs In te re st on $212,000 l l t / y r Net $61,800 27*290 23.320 $1M 30 ' Even t h is modest sun has not been r e a liz e d , fo r two reasons: the cost o f d iscard in g ad d itio n al d u st, and the unsteady supply o f these wastes. I t Is obvious from t h is d iscu ssio n that p lan ts req uiring the add ition o f c h lo rin e e re in e b etter p o sitio n to make burning o f chlorinated wastes a ttr a c tiv e than those plants not requiring ch lo rin e . A lso , a le s s so p h istica te d system than that In s t a lle d a t S t . Lawrence Cement Co. is necessary I f the only consid eration Is the fuel valu1e of these wastes. ACM cca2 APPENDIX F ^ ONTARIO MINISTRY OF THE ENVIRONMENT EMISSION GUIDELINES AND ANALYTICAL SUPPORT \ ACM c c a z * * 3 - -- ... * APPENDIX F ONTARIO MINISTRY OP.THE ENVIRONMENT EMISSION GUIDELINES AND ANALYTICAL SUPPORT F . 1 Emission Guidelines A ll in d u stria l sources of p o te n tia lly s ig n ific a n t emissions to the atmosphere in O ntario come under the ju r is d ic t io n of the P ro vince's Environmental P ro tectio n Act. The Act req u ires such sources to operate under a C e r t if ic a t e of Approval which s p e c ifie s the acceptable operating cond itions and em ission rates fo r p o te n tia l a i r contaminants. The M in istry of the Environment required that the experimental waste chlorinated hydrocarbon burn at S t. Lawrence Cement C o ., M ississauga meet ce rta in conditions sp ecified in a C e r tif ic a te of Approval. Conditional approval was granted for each of the three stages of the waste burn (At 6, C) under the terms of the te n ta tiv e g u id e lin e s for the allow able em ission rates of organic chloride. P asq u i11-G ifford d isp e rsio n c a lc u la tio n s based on the g u id elin e em ission rates from the wet k iln were used to p red ict the w orst-case ground level concentration of residual organic chloride (-hour average). Table r .1 shows the s p e c if ic a t io n s which were applied to each *" stage of the waste burn, along w ith the measured em ission ra te s. The le ss strin g e n t g u id e lin e for the f i r s t two waste burns was met e a s ily . Because of the larg e percentage of hig hly p e rs is te n t poly chlorinated biphenyls anticip ated in the fuel for the th ird waste burn, a much more strin g e n t g u id e lin e was given. In the most c r i t i c a l ca se, the th ird burn, the ca lcu la te d a approximate flu e gas concentration of resid u al c h lo rid e , 50 yg/nr (as C l ) , backed on the stated g u id e lin e , corresponds to roughly 5 ppb (v/v) of A roeior 12^2 or 17 qpb (v/v) o f dichloromethane (at ambient tem perature), the predominant in d iv id u a lly id e n tifie d component o f the flu e gas. These values compare w ith the detection lim it fo r PCB's o f 3 W9/m3 or 0.3 ppb in the flu e gas (ambient temperature). Since no PCB was found in the flu e gas samples and the methy lene chlo rid e concentration only s lig h t ly exceeded the very stringent g u id e lin e for organic ch lo rid e in th is stag e, the g u id e lin e should be considered to have been met without question. ISO TABLE F . l . HOE SPECIFICATIONS APPLIED TO WASTE BURNS Waste Burn A B C "1 0.5 0.5 0.005 2 99.5 99.5 99.995 3 99.990 99.989 99-988 A 5000 5000 50 5* 23-150 1 6 - ISO 88-150 67 0.2 0.2 0.002 0.001-0.006 0.001-0.006 O.OOA-O.OO6 Column Id e n t if ic a t io n : 1. G uideline emission rate fo r resid u a l organic ch lo rid e (g C l/ s e c ) . j 2. Estimated required percent destruction or retention of chlorinated fu el. 3- Estimated minimum percent combustion of ch lo rin ated fuel from te s t data (Table 6, Section 5*3). A. Approximate ca lcu la te d flu e gas resid u al ch lo rid e concentration based on g u id e lin e em ission ra te tug 5. Approximate measured flu e gas organic ch lo rid e concentration based on dichloromethane and-data from Table 5* Section 5.3 (ug C l / m ) . The upper lim it of the range corresponds to 50 ppb organic Cl as CHj C Ij .^, 6. C alculated point of impingement -hour average ch lo rid e concentration based on g u id elin e emission rate (ug Cl/m^). 7. Approximate ca lcu la te d point of Impingement -hour average ch lo rid e concentration based on dichloromethane and data froft) Tables 5 and 6, Section 5-3 (ug Cl/m^). ___________________________ \ Since no PCB's were detected In the flu e gas samples* none o f the resid u al organic ch lo rid e values In column 5 should be a ttrib u te d to PCB's in any of these sta g e s. 1 4 151 If the concentration of resid u al organic ch lo rid e in the flu e gas (column 7, Table F.1 ) were taken to represent PCB's In a `worst case" ( r e c a ll that no PCB's were d e te cte d ), the ca lcu la te d point of impingement concentration by the Pasqui11-G iffo rd method Is approximately equal to typ ica l measured ambient a i r concentrations of PCB's In urban a re a s. There would appear to be l i t t l e cause for concern when the emission g u id elin es are met in t h is process co n fig u ratio n . The above d iscu ssio n r e la te s only to the w e t-k iln te s ts a t S t. Lawrence Cement. If the same g u id e lin e s were applied to another k i ln , for example the lo w e r-le v e l, m u lti-sta c k dry k iln at S t. Lawrence Cement, the same emission rate as measured in the wet k iln te s ts would lead to a s ig n if ic a n t ground level co n cen tratio n . For example, even i f the strin g e n t g u id e lin e for PCB em issions were met In t h is hypothetical case, the maximum ground level concentration is ca lcu la te d by v ir t u a l source methods (applied to the dry k iln stack config uratio n) to be 0.07 ug/m^ (as A roclor 1242) which is 7 to 70 times th$ ty p ica l measured ambient a ir concentration of 0.001 to 0.01 ug/m3. The point is that some caution should be exercised in applying the r e s u lts of these t e s t s to another stack configuration. In the case of the th ird waste burn, i t was estimated that the ~ to ta l organic vapour concentration in the flu e gas during the te s ts was about 50 ppb, of which about 30 ppb was found to be dlchloromethane. Since none of the components of the o rig in a l fuel mixture were detected, however, the maximum re sid u a l concentration of Aroclor \2kt, fo r example, could be only about 1 ppb or le ss (noting the approximate detection lim it of 0.3 ppb). For speculatio n purposes o n ly , In t h is worst ca se , 1 ppb A roclor 1242 In the flu e gas (at ambient temperature) would correspond to about 11 ug/nr PCB (or about 5 ug/m as ch lo rid e) and about 0.1 kg of A ro clo r 1242 emitted per day (24 hours of continuous o p eratio n ). In r e a l i t y , t h is quantity is ah overestim ate of the to tal em issions and Is probably not s ig n ific a n t. F.2 Gas Chromatographic A n aly sis of Process and Emission Test Samples F.2.1 Chromosorb 102 adsorption tube a n a ly sis D uplicate Chromosorb adsorption tubes fo r each of the waste burns were analyzed by the A ir Q u ality Laboratory, Laboratory Services CCbZ**6 AC* 152 Branch, Ontario M in istry of the Environment for support and cross-check of a n a ly tic a l r e s u lt s obtained by the*O ntario Research Foundation. Each adsorption .tube sample represented 60 l i t r e s of emission g as. A n aly sis was done a fte r desorption at 120C into an evacuated g la ss ve ssel by in je c tin g a 1 ml gas sample into a gas chromatograph. T h is Instrument was equipped with a 9 'foot x 1/8 Inch Chromosorb 102 column held a t a temperature of 180C, and a Sc^H e le ctro n capture d etecto r. The r e s u lts from these analyses are sunvnarized in Tables F.2 to F .lt. A ll Chromosorb tube samples, including those from "B aselin e B", were found to contain small but measurable amounts of s ix or seven compounds and tra ces of a fu rth er s ix to eight compounds. These 'tneasurable" compounds were almost c e r t a in ly c h lo rin a te d , low-molecular weight compounds which, in the t a b le s , have been designated 'taajor" and 'talnor", according to the s iz e of t h e ir gas-chromatographic peaks. Of these compounds, only chloroform had been te n ta tiv e ly id e n tifie d and q u a n tita tiv e ly determined. A ll other compounds separated^from each sample were estimated together by re la tin g the sum o f the gas chromatographic peak areas with those obtained from a chloroform standard. T h eir concentrations are lis t e d under "Other Compounds". The r e s u lts shown in Tables F .2 to F.4 In d icate that the to tal of a l l chlo rinated compounds in the Chromosorb adsorption tube samples corresponds to concentrations only in the low parts per b illio n range in the stack emission gases. The agreement between these r e s u lts and those of the Ontario Research Foundation and TRW Is only q u a lit a tiv e . The g e n e ra lly small amounts (ppb range) o f organic ch lo rid e vapours found by ORF are confirmed but trends from stage to stage are not reproduced. Such disagreement is merely in d ic a tiv e that the lim it s of s e n s it iv it y of the combined sampling and a n a ly t ic a l methods are being approached. F.2 .2 Fuel feedstock an aly sis Samples of the waste ch lo rin ated hydrocarbon fu e ls used in waste burns B and C were analyzed* in the laboratory of Professor F.W. Karasek, Department o f Chem istry, U n iv e rsity of Waterloo under a research grant from the A ir Resources Branch, Ontario M in istry of the Envi ronment. ACM CC82<t7 153 TABUE F .2 . ST. LAWRENCE CEHENT WASTE BURN EXPERIMENTS - TEST l CHROMOSORB ADSORPTION TUBE ANALYSIS Waste Burn Experiment A B C Baseline B Blank. (Chromosorb) "Major" "Minor" Other Total Compounds Compounds Chloroform Compounds Compounds ppb ppb ppb '3 4 1.192 7.121 8.313 1 5 0.058 0.646 0.704 2 4 0.329 2.909 3.238 2 5 0.055 2.487 2.542 0 0- -- TABLE F .3 . ST. LAWRENCE CEMENT WASTE BURN EXPERIMENTS - TEST 2 CHROMOSORB ADSORPTION TUBE ANALYSIS Waste Burn Experiment A B C. Basel ine B B 1ank (Chromosorb) "Major" "Minor" Other Total Compounds Compounds Chloroform ` Compounds Compounds ppb ppb ppb 4 3 1.462 8.827 10.289 A* 1 1.464 5.270 6.734 3 4 0.924 12.089 13.013 0 1 0.022 0.443 0.465 0 0- -- TABLE F .4 . ST. LAWRENCE CEMENT WASTE BURN EXPERIMENTS - TEST 3 CHROMOSORB ADSORPTION TUBE ANALYSIS Waste Burn Experiment A B C Baseline B Blank (Chromosorb) "Major" "Minor" Other Total Compounds Compounds Chloroform Compounds Compounds ppb ppb ppb 50 33 3 \8 34 00 .2.356 0.481 0.560 0.030 15.773 1.916 2.089 5.077 * 18.129 2.397 2.649 5.107 AUM CC fci<r 6 IS** F .2 .2 .1 Preparation of fuel sam ples. A ll samples were d ilu te d ten times with Burdick and Jackson " D is t ille d In G la ss" methylene ch lo rid e . A 0.1 ml a liq u o t of the fuel was made.up to 1.0 ml with the methylene chloride. Methylene ch lo rid e was chosen for the solvent based on previous fuel samples analyzed in th is laboratory. F .2 .2 .2 A n a ly sis of fuel sam ples. A ll samples were surveyed by gas chromatography using a 10* x 2 mm i . d . g la ss column w ith a s p e c ia lly prepared, high reso lu tio n column packing, referred to as Ap-20M. A s im ila r ly prepared so lu tio n of A roclo r \2k2 was a ls o chromatographed for compart son. The GC conditions used are shown in Table F .6 and were c lo s e ly monitored by the use of a HP $830 A D ig ita l Gas Chromatograph using a flame io n izatio n d e te cto r. The area of the solvent peak was rejected to f a c i l i t a t e comparison of GC data. v Mass sp ectral data were obtained using a Perkin-Elm er 900 GC interfaced to a HITACHI RMU-6 magnetic mass spectrometer v ia a Biemann - Watson effu sio n sep arato r. The spectra were in it ia t e d and counted manually. A ll spectra were obtained with 70 eV io nizing voltage a t 250C. Samples I and 9 were analyzed using GC/HS techniques as above. F .2 .3 R esu lts and D iscussion Samples 8 to 10, the aromatic ch lo rin ate d f u e ls , displayed only seven compounds with in d ivid u al integrated area percentages g reater than 1% and only nine with s im ila r area percentages greater than 0.5%. There appear to be only three major components: one with c^. 73% a re a , one with ca. 4.5%f and one with ca^. 6.8%. This appeared to be the case for a ll the aromatic f u e ls . A rep resen tative bar chromatogram is shown in Figure F . l . The peak.at four minutes in the arom atic fuel was ca_. 73%- Samples 1 to 7 the PCB f u e ls , displayed several peaks In addition to those of the aromatic f u e ls , samples 8 to 12. Comparison of the add itional peaks with a chromatogram of A roclor 12k2 Ind icates that thsse peaks are due almost e n t ir e ly to PCB compounds (see Figure F . l ) . To i l l u s t r a t e th is observation, the GC data from the aromatic fuel was added to the GC data from A roclor 12A2 and the computer program ACM C G b i ^ S a 155 TABLE F.5- FUEL SAMPLE IDENTIFICATION Sample No. Comments 1 Chemt ro1 Line Sample (PCS' s ) , Jan. 3/76, 10:00 AM 2 11 1 II II , Jan. A/76, 12:00 3 11 11 II II , Jan. 5/76 k 11 11 II II , Jan. 6/76 5 11 11 II II , Jan. 7/76, 9:00 AM 6 11 11 1 II , Jan. 8/76, 8:A5 7 11 11 II II , Jan. 9/76* 8 Chemt ro1 Line Sample (Aromatics), Dec. 10/75** 9 11 1 n 11 , Dec. 11/75** 10 11 11 11 11 , Dec. 12/75** 11 11 11 11 11 * . Dec. 13/75** 12 11 it n 11 v , Dec. 15/75** 'Sample appeared to have water in i t . ^"Sample had reacted w ith l i n e r of sample vi al cap. I TABLE F. 6. GAS CHROMATOGRAPH CONDITIONS HP 5830-A GC TEMP 1 TIME RATE TEMP 2 TIME `2 50 k min A/mi n 2^0 50 mi n 0 0 0<0M INJ TEMP FID TEMP FLOW RATE CARRIER GAS ATTENUATION \ 300 28 ml/m HELIUM 28 256 ACM CC 6 SC 1 0 Vr * n o cP \T iI % TOTAL AREA FIGURE F I REPRESENTATIVE BAR CHROMATOGRAM ^ |t|t) ( ".Ml*****;*.,. *<1. ir V i - '' \ f t -V \ 157 treated the data as though i t orig in ated from a sin g le sample. This computer-produced bar chromatogram, t it le d "AROM + CLOR", was then plotted in comparison w ith the bar chromatogram of the PCB fuel sample as shown in F i g u r e F .2 . The data were not subjected to any manipulation other than stra ig h t combination of the data. Inspection o f the two p lo ts shows such a clo se correspondence of GC peaks between the computer reconstructed mixture and the actual PCB fuel mixture that there is a reasonable assurance they are the same. A GC/HS a n a ly s is was conducted on sample 1, one of the PCB f u e ls . A rep re sen tative gas chromatogram with num erically id e n tifie d peaks is shown in Figure F-3- Hass spectra were obtained for these num erically id e n tifie d peaks. The probable id e n tity , m olecular weight and/or degree of ch lo rin a tio n is shown in Table F .7 . Since a l l of the peaks a re found in the PCB fuel and only some of the peaks are found in. the aromat i c ^ u e l . an estim ate o f the d ilu tio n or mixing of these compounds can be derived. During the gas chromatograph a n a ly s is , a l l v a ria b le s were held constant; thus, absolute Integrated . area counts are in d ica tiv e of changes in re la tiv e concentrations. These r e s u lts are shown in Table F .8 . The u n its are area counts and the valijps it shown are 1/10 of output v a lu e s. Peak #3 is absent in the aromatic fuel and in A roclor 1242 but present in the PCB f u e l. The mass spectrum of peak #3 suggests cr ichlorobenzene. " Peak #19 is present in small amounts in the chlo rin ated aromatic f u e ls , absent in A roclor 1242 but present in increased concen tra tio n s in the PCB f u e ls . A d e f in it iv e mass spectrum o f t h is peak was not obtained but was suggestive of lack of halogenation and is most lik e ly an uhsaturated hydrocarbon of high molecular w eight. F .2 .4 Conclusions The aromatic fuel is p rim a rily o-chlorotoluene and the PCB fuel Is a mixture of A roclo r I24,and the aromatic f u e l. Area data for peak 1 indicated an i n i t i a l d ilu tio n of approximately o ne-th ird (by comparison of area counts for aromatic fu e ls to area counts for sample 1) which increased to approximately f o u r - f if th s by the end of sampling data. i ADP . CCtZS* % TOTAL AREA \ Y ) j FIGURE F2 COMPUTER RECONSTRUCTED BAR CHROMATOGRAMS FOR PCB FUEL AND AROMATIC FUEL PLUS AR'oCHLOR 1242 160 The r e s u lts of the U n iv e rsity of Waterloo group w ith respect to id e n tific a tio n of components of the waste burn B and waste burn C fu e ls are in agreement with those of ORF and TRW. TABLE F .7 . MS DATA FROM GC ANALYSIS OF SAMPLE PCB FUEL (Reference to Figure F .3 .) Peak No. 1 2 3 k 5 6-8 9-12 13-16 17 18 19 Comments MW 126, 1 ch lo rin e - sp ectra suggest o-ehlorotoluene MW 160, 2 ch lo rin e s - spectra suggest aromatic compound dichlorotoluene MW l80, 3 ch lo rin e s - sp ectra suggest t r 1chlorobenzene MW appears to be 216 with^jf ch lo rin e s - p o ssib le id e n tity C6HJ*C lA MW appears to be 216 w ith A ch lo rin e s - spectra very s im ila r to peak #1* MW 222 2 ch lo rin e s - sp ectra suggest dichlorobiphenyls PCS1s HW 256 3 ch lo rin e s - PCB's MW 290 A ch lo rin e s - PCB's MW-32A 5 ch lo rin e s - PCB's MW uncertain - no suggestion of halogenation 1 i- t j ? i % 4 p * * j - % - ACM CCA255 TABLE F.8. AREA COUNTS (x IOk) Peak No. 1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 17 18 19 8 10/12/75 3860 80.5 - 2*3.9 3(3.1 - " / - - - 7.8 9 11/12/75 3976 82.5 - 247.6 3(2.1 - - - - 8.6 10 12/12/75 37(9 78.8 - 240.6 3(0.9 - - - t - - 8.4 h 13/12/75 399* 82.5 251.4 373.3 - - - - - 8.2 12 15/12/75 3842 80.4 245.1 3 (3 .* - - - - - 8.6 i 3/1/76 1200 24.9 70.4 74.2 IS*. 1 158.2 7 (.9 340.4 465.6 82.7 217.7 9*3.9 265.0 197.0 287.0 I9 I .3 102.0 52.1 I77.5 2 *i/l/76 1142 23-1 23.8 72.3 148.1 154.8 75-7 333.7 454.8 80.6 211.8 518.0 271.3 183.8 292.4 O I.) 107.5 51.5 168.9 3 5/1/76 9 1 3 .* I 9 .O 38.0 64.5 155.0 160.6 77.0 348.4 4 5 I .2 84.9 212.4 503.6 275.5 179.7 302.5 186.8 134.8 5*.9 143.8 * 6/ 1/76 770.6 16.1 48.2 60.6 I 5 O. 2 156.5 66.5 328.8 *3 1 . 3 84.3 203.3 *81.8 258.3 177.0 284.3 180.2 143.2 50.1 120.5 56 7 7/1/76 8/1/76 9 / 1 / 7 6 766.8 753.0 7 0 2 . 4 16.1 I 5 . 7 14.6 50.9 61.7 153.8 *9 . 7 58.3 1'57 .8 60.4 152.0 . 157.6 160.2 161.1 167.8 88.6 89.4 95.1 335.2 *39.3 ^ 85.6 . 336.7 .440.6 85.2 351-3 *57.2 89.1 207.9 489.1 265.9 208.9 49I .2 2 6 B. 6 2 1 7 .I 507.6 282. 1 178.6 292.B 178.3 29*.5 186.7 309.8 178.6 188.4 I9 0 . 5 149.1 151.4 167.5 51.3 *9 .5 50.2 120.5 120.9 I I 5 . 9 1 WOV f APPENDIX G LABORATORY ANALYSIS RESULTS FROM THE ST. LAWRENCE CEMENT FACILITY TEST by D.G. Ackerman, J.F. Clausen and C.A. Zee TRW Systems Group k v ACM C C 8 2 5 7 165 APPENDIX G LABORATORY ANALYSIS RESULTS FROM. THE ST. LAWRENCE CEMENT FACILITY TEST G. 1 Summary A n a ly sis of the flu e gas samples for organic composition gave the following r e s u lts : - Hydrocarbons were not detected In the samples of flu e gas with d etection lim its ranging from 0.6 mg/m^ to 0 . 0 0 9 mg/m^, - Low molecular weight chlorinated organic compounds such as methylene c h lo rid e , chloroform , and carbon, te tra ch lo rid e were s p e c i f i c a l l y searched fo r by gas chromatography with electro n capture detection (ECO) and were g en erally not detected in most of the samples. Four samples indicated the possib le presence of some of these compounds, hut a l l at le v e ls of le ss than 3 0.1 mg/m of flu e gas. - P o lych lo rin ated biphenyls were searched fo r by GC/ECD and GC/M5 and were not found in any samples a t the GC/HS detection lim it of 3 ug/m^ of flu e gas. Trace metal concentrations in the flu e g ases, as determined in the samples taken with the EPA sampling t r a in , were a l l le ss than 10 yg/m^, with the exception of lead In the.WBC t e s t s . The em ission of lead during i the WBC te s ts averaged 0.12 mg/m . % A n a ly sis of the c lin k e r product and d iscard dust samples for organic composition gave the following r e s u lts : - Hydrocarbons were not detected in so lve n t e x tra cts of any of the c lin k e r product or d iscard dust samples. The level of d etection by GC/ECD was 5 ug/g of sample or lower. - P o lych lo rin ated biphenyls were not detected In any of the c lin k e r or dust samples by GC/ECD. Detection lim its for PCB's in the samples wdre t y p ic a lly 0.04 ug/g or b e tte r. The inorganic ch aracter of the c lin k e r products and e le c t r o s t a t ic p r e c ip ita to r d iscard dusts was changed very l i t t l e I f at a l l by the addition ACM CC6256 166 of Che chlorinated wastes to the SLC wet k iln process. Trace metal levels in the U6B and WBC test samples did not d iffe r sig n ific a n tly from the'baseline samples. Analytical techniques selected for an alysis of flue gas constituents which include vapours, condensables and particu late matter were gravim etric, IR, LRMS, GC, GC/HS, SSMS, ICPOES, and AAS. These a were selected on the basis of s e n s itiv ity and s e le c tiv ity c r it e r ia to i enable id en tificatio n of species in the flue gas at concentrations on the order of 0.1 mg/tn^, representing the threshold level of the most toxic species as defined by OSHA and other occupational health and safety organizations. Detection lim its for many of the techniques extend to ug/m^ lev e ls. However, sp e c ific analyses to Identify compounds 3 below the level of in te rest, 0.1 mg/m , were not routinely performed. The techniques used were both q u a lita tiv e and quantitative in nature, with an intended accuracy range of plus or minus a factor of two to three. * G. 2 introduction In cooperation with Environment Canada and the Ministry of Environment of Ontario, the United States Environmental Protection Agency supplemented the Canadian studies through participation In analyses of *> samples acquired from various streams during test burns of chlorinated i hydrocarbons in the St. Lawrence Cement Co. wet process k iln . These analyses were performed by TRW Defense and Space Systems Group, Redondo Beach, C a lifo rn ia , under contract to the U.S. EPA. Samples were provided to TRW through the courtesy of the Ontario Research Foundation and St. Lawrence Cement Co., of Ontario, Canada. Emphasis in the EPA sponsored work was directed toward analyzing for input waste residual compounds and by-products in the clin k er (product), dust from the e le c tro s ta tic p recip ita to rs, and the various component samples from the EPA Method 5 and ORF sorbent trap tra in s . These analyses were limited to the two test burns performed using chlorinated aromatic and polychlorinated biphenyl waste blends. In addition, background anal y tic a l work was aecompllshed\on samples acquired during the two baseline tests in which the k iln was fired using only residual o il as a fuel. Analyses were also performed on the two waste blends. This work was done to supplement the Canadian laboratory resu lts as well as to acquire ACM C C b i b S 167 ad d itio nal evaluation data in accordance with U.S. EPA protocol used for ongoing chemical waste in cin era tio n te st programs In the United S ta te s. A b rie f economic a n a ly sis was made by S t. Lawrence Cement Co. using data from both th e ir own operations and from Chemtrol P ollutio n Services Inc. (Appendix E ). G .3 A nalysis Techniques The purpose of the analyses performed on the samples from SLC was to id en tify and q u an tify: - known hazardous species present from the waste as determined by p re te st a n a ly sis of a sample of the waste m a te ria l, - secondary decomposition products ( e .g ., incomplete combustion products which are p re d ic ta b le ), and - other sp ecies found to be present but which are not predicted or otherwise expected. A complete l i s t of the samples received -by TRW for a n a ly sis is given in Table G .l . This table shows what portion o f the to tal sample co llected by the tra in s was sent to TRW as w ell as the resp ective burns and test numbers for the received samples. The coding system used to uniquely id e n tify ech sample is shown in Figure G .l . These codes are ^ used in subsequent tabulations of data and r e s u lts . The f i r s t step in the a n a ly tic a l approach involved various extractio n and preparation procedures to separate the organic and Inorganic constituents and/or to concentrate the samples In a su ita b le form for a n a ly s is . The e x tra cts and concentrates were then analyzed by the techniques to be described in th is se ctio n . G.3.1 Ex tra ctio n s and sample preparation G .3 .J .1 Solvent e x tra cts received from ORF. With a few exceptions, solvent e x tra ctio n s for organic* species were performed by ORF using pentane or hexane. Subsequently, a liq u o t s were received by TRW and combined according to the plan shown In Figure G .2. These combinations were performed because: \ 1) the aliq u o ts represent r e la t iv e ly small gas sample volumes, thus making constituent concentrations very 1m and d i f f i c u l t to measure, and 1 AC* CCe^tC TABLE G.l. SUMMARY OF SAMPLES RtCEIVED FROM EACH WASTE BURN ,i t It i. i I9Z.900 J t -.a h h i t o u r , r i w . m m j -'j s i l i ic h is Rui ti r.jni|il uni 1i .i in Ir >nn (.oniponenl lx tra ile d Any '.pccidl Preparation lest Ho. 1 ir s l fl.isel inp on Primary Fuel - 6LA Waste Burn on Chlorinated Aromatics - WttB Standard ll'A Nelhnd !i Ira in - LPA 1 i 1t e r / Insolubles - 1 Probe R in sr - PR Caustic Impinger - NAOII Water Impinqcr - HpO CiNiibiried Impingers - 1 Extracts of A ndilied Solutions - AE Waste Burn on PCB Blend - W8C Second Baseline on Primary Fuel - BIB ORF Desfgned "Sorbent" Traln-ORF / r FOR EXAMPLE, ALL OF THE EXTRACTED SAMPLES FROM ONE WASTE BURN WOULD BE COOED AS FOLLOWS; I l , ' 1?. nr 13 i f sample is tram only one test Blank i f sample .is a combination of a ll three tests WBB-OAF-FE Waste Burn B , ORF T r a in , Combined F i l t e r E x tra c ts from a ll three te sts WBI-ORF-PRE A NAOKE - Watte Burn B , ORF T r a in , Combined Probe R in se and C au stic Implnger E x tra cts fro m /til three te s ts WBB-ORF-NA0H-AE-T3 u Waste Burn B , ORF T r a in , E x tra c t o f the a c id if ie d c a u s tic Implnger so lu tio n from T est 3 HB8-0RF-H20E Waste Burn B , ORF T r a in , Combined Water Implnger E x tra c ts from a l l three t e s t s ' WB8-EPA-FE A PRE - Waste Burn B . EPA T r a in , Combined F i l t e r and Probe R inse i E x tra c ts from a l l three t e s t s WBB-EPA-IE Waste Burn B , CPA T r a in , Combined Implnger E x tra c t s from a ll three te sts o T 1 FIGURE G .l. TRW SAMPLE COOING SVSTEH voO> /?9 3 170 2) a ll three tests of each waste were performed at one nominal operating condition. In general, following this plan, the'organic concentrate obtained for each extracted sample type ( e .g .t f i l t e r s , impingers, so lid s , e tc .) from each of the two sampling trains were the combination of a ll three tests of each waste. Exceptions to this plan were the BLA and aqueous samples for which only one of the three test samples were received (see Table G.1). At no time were EPA train samples combined with ORF tra in samples. Small aliquots (2-5 ml) were f i r s t taken of the "as received" extracts and set aside for analysis of v o la tile compounds that would be lost in the next step which was to concentrate the remaining solvent sample using Kuderna-Danish evaporators and a steam bath. The solvent extracts of the probe rin se s, in addition to being combined as in the plan shown In Figure G.2, were also combined with the sample (EPA t r a i n - f i 1t e r , ORF t r a in - f ir s t 1impinger) whose juxtaposition in the sampling train and sim ila r physical ch a ra c te ristic s permit the combination. The rinses of the EPA train were made of the glass probe lin e r in front of the f i l t e r . However, with the ORF train which only has a short nozzle in front of the in-stack f i l t e r , rinses were made of the 14 feet of probe and tubing between the f i l t e r and the f i r s t liquid impinger. The resulting solutions were filte re d through standard f i l t e r paper and then extracted by ORF. G-3.1.2 S o lid s, aqueous solutions, and f i l t e r s . The so lid samples were prepared for organic analyses by extraction in a Soxhlet apparatus for 24 hours with d ls tille d - In - g la s s grade pentane. These extracts were concentrated with Kuderna-Danish evaporators to a 10 ml volume. Prepara tion of the so lid samples for inorganic analyses consisted of a lowtemperature plasma ashing to remove possible organic Interferences. Other samples for inorganic an alysis also required some degree of preparation. Aliquots of the aqueous Impinger and probe rinse samples were taken and a cid ifie d w ftfc n itric acid to s t a b iliz e any metals present. Each o f the insoluble residues obtained by filt e r in g the probe rinses was combined by ORF with it s matching p articu late f i l t e r for treatment as one sample. These in so lu b les/fi 1ter samples were plasma ashed and then --i AC* GC8263 171 TC ANALYSIS OF ORGANIC CONCENTRATES FIGURE G.2. PLAN FOR THE COMBINATION OF ORF SOLVENT EXTRACTS VACUUM GAUGE FIGURE G.3 . DESORPTION SYSTEM FOR CHROMOSORB 102 TUBES 172 extracted by reflu x in g constant b o ilin g aqua regia over the each sample for two hours. The re su ltin g a d d d ig e sts were made to 50 ml for analysi s.- G .3 .1 .3 Sorbent tubes. The Chromosorb 102 sorbent tubes from the ORF sampling tra in were prepared fo r a n a ly s is by a q u a n tita tiv e desorption technique. The apparatus used for the desorption Is shown sch em atically in Figure G-3- The sorbent tube Is attached with a minimum length of Tygon tubing to a 20 ml g la ss sample bulb and Is then heated In a small furnace to 185 10C. . With va lve A closed and va lves B and C open, the sample bulb was evacuated and iitvnersed in liq u id n itro gen. Valve A was then opened for 30 minutes to allo w the contents o f the sorbent tube to tra n sfe r to the sample bulb. Valves A and C were then clo se d f the LNj f la s k removed, the sample bulb allowed to e q u ilib ra te to room temperature, and the pressure recorded. The volume of the e n tire manifold system. Including each sample bulb, was c a r e f u lly c a lib ra te d and on the average was 68.8 cc. With th is information and the recorded pressure (assuming the temperature to be co n sta n t), the volume of the m aterial desorbed was ca lcu la te d according m to the ideal gas law: where: PT Vl p.v. 22 P.I * the pressure measured, In mm Hg V| * the volume o f the sample bulb and m anifold, *v68.8 cc ?2 m 760 mm Hg V2 m the ca lc u la te d volume of desorbed m aterial a t 1 atm. The desorbed m aterial was recovered from the manifold by re immersing the sample bulb in LN^. Valve B was then closed and the sample bulb removed from the desorption system for a n a ly s is w hile s t i l l at negative pressure. G.3-2 A n a ly tica l methods \ The extracted and prepared samples were analyzed by various methods. The s p e c if ic analyses selected depended to some extent on the samples' forms, which were: ACH C C c2 t5 A Q 173 For Inorganics - Solids - Aqueous and acid solutions For Organics * A liquots of the neat solvent ex tra cts - Concentrates of the solvent e x tra cts - Desorbed m a te ria ls from the sorbent tubes G. 3-2.1 Analyses for inorganics . The inorganic composition of the so lid samples was determined by spark source mass spectrophotography (SSMS) with e le c tro n ic d e te ctio n , which w ill detect elements present down to a concentration of 1 ppm. This SSMS technique has an accuracy from 100-500% enc w ill survey the sample for a l l the elements except gases ( i . e . , H, N, 0, He, Nef A r , K r, Xe, and Rn) and Hg because of it s high v o l a t i l i t y . Other elements with ap p reciab le v o l a t i 1i t i e s t. such as Be and Se, have f a r reduced accu ra cies In t h is a n a ly s is . The aqueous and acid so lu tio n s were f i r s t surveyed fo r metals by in d u ctively coupled argon plasma o p tic a l em ission spectroscopy (ICP0ES). The ICP0ES an alysis determines 32 elem ents, including most of the to x ic eiements of in terest in the program, down to ppb le v e ls , with an accuracy of 100-200%. The purpose of th is survey was p rim a rily to check that the netals in the test samples were present tn approximately the same amounts re la tiv e to each other as they were in the waste m a te ria l. Those elements which, from the r e s u lts e ith e r of the ICP0ES survey or o f the a n a ly sis of the waste m a te ria l, seemed to be present at p o te n tia lly to x ic le v e ls , were determined q u a n tita tiv e ly by atomic absorption spectrometry (AAS). The s e n s it iv it y of t h is method v a rie s from approximately 1.0-0.001 ppm for the elements which were determined, w ith an accuracy between 10-50%. G .3 .2 .2 Analyses for o rg a n ics. Organic constituents of the te st samples were determined by a combination of the following techniques: - infrared spectrometry (IR ); - gas chromatography (GC) with e ith e r an electro n capture (EC) or flame io n izatio n detector (FID ); - low reso lu tio n mass spectrometry (LRMS); - combined gas chromatography/mass spectrometry (GC/MS). ACh C ecity m The a liq u o ts of the neat solvent e x tra cts were analyzed only For v o la t ile organochlorine compounds. EC/CC was used for t h is a n a ly sis and the samples were compared to standards of CHCI^, and CCI^ as w ell as the o rig in a l waste m a te ria l. The column was 183 cm x 0.635 cm o .d ., g la s s , 1.5% 0V 17 and 1.95% QF-1 on Chromosorb WHP. The concentrates of the solvent e x tra cts were analyzed to id e n tify the less v o la t ile compounds that would not be lo st by evaporation of the e x tr a c t. An a liq u o t of each sample was f i r s t taken and evaporated in a cle a n , 7 ml weighing b o ttle at ambient conditions to remove the so lve n t. The residue was then weighed on a micro balance*to 0.001 mg and scanned by IR as an evaporated thin film . The sample was returned to the weighing b o ttle and stored for use in the LRMS a n a ly s is . To obtain an LRMS sp ectra, the sample was rinsed into the g la ss ve ssel fo r the s o lid s probe In le t system and the solvent evaporated again at ambient co n d itio n s. The IR and LRMS analyses y ie ld q u a lit a t iv e inform ation ^bout the c la s s e s or types of compounds ( e .g ., hydrocarbons, phenols, POM's, e t c .) present as w ell as an idea of the complexity and to x ic nature of the concentrated sample. The s e n s it iv it y of the LRMS so lid s probe technique for s p e c if ic compounds in an organic m atrix can be estimated even though the mass ^ spectrometer is not s t r i c t l y considered a q u a n tita tiv e instrument. At ty p ica l mass spectrometer operating pressures of 10 ^ t o r r , the r e la t iv e v o l a t i l i t y of organic m a te ria ls at normal soTids probe temperatures O0Q-250C) is not an important fa c to r. That i s , the.,v o la tl 1 t ie s of m aterials of in te re st are s u f f ic ie n t ly high fo r adequate detection. The so lid s probe of the mass spectrometer is in such clo se proximity to the io n ize r that sample d iffu sio n does not s ig n if ic a n t ly reduce s e n s it iv it y . Previous experience and contacts w ith other la b o ra to rie s using sim ila r equipment in d ica te that i f 10 micrograms of an org anic compound is present in a s o lid s probe along with other m a te ria l, i t w ill be detected to the extent that the ten or more strongest fragment peaks w ill be recorded. Since the typical^ weight of sample placed In the s o lid s probe is 1 m illig ra m , the required weight percentage o f a given organic compound to be detected in an organic m atrix Is approximately one percent. Therefore, any compounds not detected by LRMS were assumed to be present at le ss than one percent le v e ls . f1 II 175 i i The organic compounds in the e x tra ct concentrates were separated and q u an tified by' GC using both flame io n izatio n and electron^ capture d e te ctio n . Columns and GC conditions are given below: > Flame io n izatio n d etection * - Varian i860, dual d if f e r e n t ia l FID , - Columns: d u al, l 83 cm x 2 mm i . d . , s t a in le s s S t e e l, 3-5% OV-17 on 100/120 mesh Chromosorb WHP j - Temperatures: column, ambient for 5 minutes, then ambient to 275C at 10C/min; d e te cto r, 275C; in je c t o r , 250C. - Flow ra te : helium c a r r ie r at 30 ml/min; hydrogen1at 30 m l/m in.; a i r , 300 ml/min. [ - Attenuation: 1 x 1 0 ^ a/mv I Electron capture detection - Tracor MT-150, ^ N i sin g le ECO - Column: 183 cm x 0.4 cm i . d . , g la s s , 1.5% 0V-17 and 1.95% Q.F-1 on 80/100 mesh Chromosorb WHP - Temperatures: column, 200C; d etecto r, 225C; in je cto r 225C. - Flow ra te s : p re -p u rifie d c a r r ie r through column a t 60 m l/m in.^ d e t e c t ;* purge a t 40 ml/min. j - P o la risin g vo ltage: 14V; bucking range -2 x 10-8 input atten u atio n , 102 ; output atten u atio n , X2 to X64. i1 G.4 A n a ly tica l Data ' j. Data obtained from the vario us analyses performed w ill presented in th is section in the following order: j Chlorinated Hydrocarbon Wastes Tested I be Chlorinated Aromatics - P olychlorinated Biphenyls (PCBs) | * Samples Obtained from the EPA and ORF Sampling T ra in s v - Organic Constituents - Inorganic Characterization I i Solid Residues and E fflu e n ts from the SLC K iln Process f - Clinker Products J - E le c tro sta tic P recip itato r Discard Dusts [ l ADK t .. CCbAtt ill <1 It i. E! I*. J. i (: , i: ^; ii t! 1, \ 'j, i; 176 G.4.1 Chlorinated hydrocarbon wastes tested Samples of the ch lo rin ate d wastes burned a t S t . Lawrence Cement were taken by ORF from the liq u id waste feed tank on each day of te stin g . Composites o f these were sent to TRW. The analyses used to ch a ra cte rize the w astes and determine the expected compounds of in te re s t in the test burn samples were: - thermal content - gross heat of combustion; - viscosity; - sp ecific gravity; - lo ss on ig n itio n (L01); - C, H, N, S, and halogens; - infrared spectroscopy (IR ); - gas chromatography/mass spectroscopy (GC/MS); and, - spark source mass spectroscopy (SSHS). G .4.1.1 Primary f u e l . In add ition to the two w astes, a sample of the primary f u e l, bunker "C" o i l , was a lso received . Of the analyses lis te d above only the gross elemental determination (C, H, N, S , and halogens) and a trace metal scan by X-ray fluo rescence (XRF) were performed on the o i l . P hysical p ro p erties were not determined sin ce they are f a i r l y standard, and accurate compound id e n t ific a tio n was not necessary sin ce the background of compounds contributed by burning the bunker "C" o il as fuel can best be determined by a n a ly s is of the combustion te st samples from the b a se lin e burn. The elemental determination on the o il gave the'fo llo w in g re su lts: - 86.87% C - 9.89% H - 0.36% N - 2.62% S - 0.05% Cl (to ta l halogens as c h lo r in e ). ACP CC62CS *\ \ 177 The XRF seen detected s ix elements at the le v e ls lis t e d below: Element Si P n; v* Ca Ti - Approximate Concentration (ppm) >500 50-500 5-50 5-50 <25 <5 .*1.1.2 Chlorinated a ro m a tics. The arom atic waste was a dark brown, low v is c o s it y liq u id that was v is u a lly free of sediment. The measured physical c h a ra c te ristic s were: - thermal content - 5170 keal/kg (9310 B tu / lb ); - v is c o s it y - 1.09 cen tisto k es a t 3C ( 100F) ; - s p e c if ic g ra v ity - 1.281 at 16C (6 0 F ); - LOI - 99.98%. Elemental an alyses performed showed the follow ing composition: - Uk.00% c - }.kS% H - 0.028S N - 0.0191 s - ^9-10t Cl (to tal halogens as -chlorine). A n a ly tic a l techniques used to determine the organic composition included IR and GC/HS. % The IR spectrum indicated the waste to be composed p rim a rily of aromatic hydrocarbons as well as a lip h a t ic alkanes and alkenes with a strong response in the 600-800 cm ^ region which can correspond to C-Ct bonding. No in d ica tio n of any other fu n ctio n al groups such as phenols, e th e rs, or a amines was found. Comparison of the waste spectrum w ith spectra from the Sad tler in d ices fo r the compounds found by GC/HS, showed the p rin c ip le 3 Element is p o t e n t ia lly to x ic - OSHA TLV of <1 nvg/m fo r an eight-hour exposure. AC* CC6i ? c I \ \ i .1. V t* 8a 1 r1 11 i1 I I 178 co n stituent to be o-chlorotoluene, the most c h a r a c t e r is t ic peak o f which is a strong, sharp peak at 750 cm \ GC/H5 Chromatographic separation was ca rrie d out on a Finnigan GC/HS system .using two columns. One was Chromosorb 101, temperature programmed from 30 220C at 10C/min, and the other was 0V--17 temperature programmed from 30 - 275C at 10C/min. The compounds shown in Table G.2 were id e n tifie d and t h e ir concentrations ca lcu la te d based on r e la t iv e peak areas found from both columns and assuming equal response to the to ta l ion monitor of the Finnigan instrument. These r e s u lts r e f le c t the fact th a t, in the course of performing the te sts at SLC, the waste feed tank was not emptied between w astes. Thus, the m aterial fo r the "arom atic" burns contains '03% ch lo rin ated a lip h a t ic s ca rrie d over from the previous burn (as a cost savings to the program, TRW did not plan to analyze samples from the ch lo rin ate d a lip h a tc s'fe u rn ). The primary constitu ent of the aromatic waste is chlorotoluene (52.5%) with the remainder made up of three other chlorinated arom atics; d ichlorotoluene, octachlorocyclopentene, and octachloronaphthalene. TABLE G.2. ORGANIC COMPOSITION .OF AROMATIC WASTE BY GC/MS . Compound Acetone Methy(acetate Oichloromethane Chloroform Carbon T e tra ch lo rid e Dichloroethane Trichloroethane Tetrachloroethane Trichloroethylene Toluene Chlorotoluene Dichlorotoluene \ Dimethyl Benzene (Xylene) Octachlorocyclopentene C |0Clg (Octachloronaphthalene) Estimated Concentration (% w/w) 1.6 0.3 0.4 11.1 13.9 3-1 2.5 . 0.4 1.0 1.0 52.5 4.6 0.9 3*7 3.0 i ACM C G 271 .(i _ 179 S5HS Trace metals In the chlorinated aromatic waste were determined by spark source mass spectroscopy (SSMS). The waste was f i r s t ashed by a sulphatsd dry method. A very small amount of ash, 0.01*$, was obtained of whicn the major elements were 8a, Fe, Na, P, S, S i, and Ti (the sulphur most lik e ly represents the contribution from the H^SO^ added during ashing). Other elements detected by SSMS down to 1 ppb are shown in Table G.3> This an a ly sis was performed to determine what p o ten tially toxic elements might have been present in the waste at high enough concentrations to warrant q uantitative examination of the flu e gas test samples. Given the sem i-quantitative nature of the data, the SSMS re su lts should not be construed as a q uantitative characterizatio n of the waste m aterial. TABLE G.3. TRACE METALS IN THE CHLORINATED AROMATIC WASTE BY SSMS Ei ement A pproximatea Concentration (ppm) - v' Element Approximate9 Concentration (ppm) A1 Ca Cu Ma Zn K Crb Pbb Mn Hgb*C Sr Pt B Cob Ni U MO 0 .7 -1 .5 0 .7 -1 .5 0 .7 -1 .5 0 .7 -1 .5 0.7-1.5 0.6 o .v 0.3 0.07 0.07 0.06 0.05 0.05 0.05 0.03 0.01 0.003 ` Sn 0.008 Zr 0.008 yb 0.007 Asb 0.006 A9. 0.003 Cdb 0.002. Ce 0.002 La 0.002 Sbb " 0.002 Beb 0.001 Bi 0.001 Ge 0.001 Li 0.001 Rb 0.001 Seb 0.001 Sm 0.001 Yb 0.001 aThe accuracy of th is technique ranges from 100 to 500 percent. P o t e n t ia lly toxic metals - 0SHA JLV of <1 mg/m* for an eight-hour \ exposure. cHg was determined by a highly quantitative atomic fluorescence technique. ADM CCEZ7Z 180 G .it .1 . 3 P o lych lo rin ated biphenyls (P C B 's). The PCB waste was a medium brown, low v is c o s it y liq u id with fin e suspended p a r t ic u la t e that tended to flo a t to the top and cake out on the sid es of the co n tain er. The measured c h a r a c t e r is t ic s were: - thermaK content - 6,710 kcal/kg (12,083 B tu /lb ); - v is c o s it y - 5.87 cen tisto k es at 38C (10 0 F ); - s p e c ific g ravity - 1.196 at I6C (6 0 F ); - 101 - 99.90% Elemental analyses performed showed the follow ing composition: - 57-9<*% C - 5.23% H - 0.018% N - 0.12% S - 33-60% Cl (to ta l halogens a i^ c h lo rin e ). The a n a ly t ic a l techniques used to determine the organic composi tion of the waste were IR and GC/MS. _I_R The IR scan showed strong peaks at 1100 cm ^ and 1450 - 1480 cm ^ which are c h a r a c t e r is t ic of b iphenyls, and a peak at 750 cm * in d ic a tiv e of remaining o-chlorotoluene in the waste. Other small peaks in the spectrum matched those in the aromatic waste IR scan w^ich su b stan tia tes the GC/MS r e s u lts discussed below. GC/HS A n a ly sis of the PCB waste was performed by the same GC/MS methods and co n d itio n s as fo r the arom atic w aste, and the compounds that were id e n tifie d are shown tn Table G.4. The PCB waste composition shewed a the e ffe c t of previous waste m aterial le f t in the feed tank. It was composed of 12% ch lo rin a te d a lip h a t ic s , 33% ch lo rin ated arom atics, and 45% PCBs. There were 22 separate PCB compounds id e n tifie d including several isomers of each of the m ultiple chlorinated biphenyls. SSM5 Trace metals in the PCB waste were determined by SSMS. The waste was f i r s t ashed by a sulphated dry method. A small amount of ash, 0.15%, 181 TABLE-G.ii. ORGANIC COMPOSITION OF PCS WASTE BY GC/MS Compound Water Acetone Methylacetate Methanol Chloroform Carbon T e t r a c h lo r id e Ethanol Oi ch i or oet h ane T r i chloroethane HexachIoroethane Tr i c h l o r o e t h y 1ene Tetrachloroethylene Xylenes Toluene Chlorotoluene Dichlorotoluenes Trichlorotoluenes Octachlorocyciopentene Chlorobi phenyl Dichiorobiphenyls T r i c h 1orob i phen y1s Tetrachlorobiphenyls Pentachlorob(phenyls He x ac h1o r o b i p h e r y l " Heptachlorob i p h e n y I s . O c t a c h l o r o b i phen y1s D i- n -o c ty l or D i- e t h y l Hexyl P hthala te C 1qC 18 " Oc t a c h l o r o n a P h t h a l e n e Estimated Concentration {% w/w) 0.7 1.3 0.5 0.4 1.4 6.8 5.2 0.6 0.7 0.2 1.6 0.2 2.3 1.0 17.2 2. A 3.3 4.0 0.5 8.4 15.5 11-15 2-6 1-2 0.5 0.2 4.0 2.5 was obtained of which the major elements were Fe, Na, P, P t, and S I. Other elements detected by SSMS down to 1 ppb are shorn In Table G.5* This a n a ly sis was performed to determine what p o te n tia lly to x ic elements n i g h t have been present In the waste at high enough concentrations to warrent q u a n tita tiv e examination of the flu e gas test.sa m p les. Given the sem i-q u a n tita tiv e nature of the data, the SSMS re s u lts should not be construed as a q uantitative ch aracterizatio n of the waste m aterial. G .4 .2 Samples obtained from \he EPA and ORF sampling tra in s Samples obtained by the two t ra in s from the WBB, WBC, and BLB te s ts were received for a n a ly s is . A l i s t of the samples derived by ORF ACP c c e * 7 ^ ... 182 -^J.f '> TABLE C .5 . TRACE METALS IN THE PCB WASTE BY SSHS Element A* pproxi.mate a Concentrt ion (ppm) E 1emen t Approximate3 Concentration (ppm) K 7.n Cu S Mg A) Ni MBaob Ti B Ca Wb PCbrbb - vb Ag Mn $bb ASnsbb ZCorbb 7-15 7-15 7 5 5 k k 3 2 2 1 1 1 0.8 0.8 0.7 0.6 o .s 0.5 0.5 0.2 0.2 0 .1 .. Seb rb c HCvdb Th La Nb Bi Li Rb Ta Au Nd Hf ^m. Be Oy Ga Ge Pr Sc U t 0.1 0.07 0.06 0.05 0.02 0.01 0.01 0.009 0.009 0.006 0.00 A 0.003 0.003 0.002 0.002 0.001 0.001 0.001 0.001 0.001 0.001 0.001 aThe accuracy of t h is technique ranges from 100 to 500 percent > hpotential ly to x ic meta 1s - 0SHA TLV o f <1 mg/m^ for an eight-hour exposure cHg was determined by a hig hly q u a n tita tiv e atomic fluorescence technique. from each t ra in was presented in Table G .I . A ll o f these samples were scheduled to be analyzed u n t il, in e a rly A p r il, ORF's review of th e ir a n a ly t ic a l data brought them to conclude that the solvent e x tra cts of the BLB samples had been contaminated w ith PCB waste m aterial [ G .l] which was confirmed by the analyses at TRW. The contamination appeared to have occurred before the s p l i t of tt^e samples Into i a liq u o t s . In view of t h is development, a n a ly s is of the BLB solvent e x tra ct samples was discontinued and e x tra c ts from a BLA te s t were sent as replacements. However, there was no evidence of a contamination problem with the remaining BLB samples. Thus, the f i l t e r s , probe rin se in so lu b le s, : - i `- 83 aqueous so lu tio n s, and. Chromosorb 102 sorbent tubes from the BLB te sts were analyzed for a background reference to the corresponding WBB and WBC samples. All of the te st samples were coded fo r ease of refe ren ce . The coding used was presented in Figure G .1, which a lso showed the combination of the probe-rinse samples w ith other appropriate samples. The reasons and methods for combining the probe rin se s w ith e ith e r the c ilt e r (for the EPA tra in ) or impinger (for the ORF tra in ) samples, as well as the preparation and a n a ly t ic a l techniques for a l l the samples were described in Section G.3* Where co n stitu en ts found in the te s t samples a re reported as concentrations in the e fflu e n t gas, the sample gas volunes needed for the c a lc u la tio n s were taken from Section 5-3 of the re p o rt. Volunes for the ORF tra in were taken from Table 1*, and volumes fo r the EPA tra in from Table 7. G.^.2.1 Organic c o n stitu e n ts. Samples f o r v h e a n a ly s is o f organic composition were in the*form o f: - a liq u o ts of the neat solvent e x tra c ts; - concentrates of the solvent e x tra c ts; - residues from the evaporation of the e x tra c t co n cen trates; and - desorbed m a te ria ls from the sorbent tubes. These samples were f i r s t surveyed for th e ir q u a lit a t iv e nature by g ra v i m etric, infrared spectrometry ( IR ) , and low reso lu tio n mass spectrometry (LRMS) techniques. Q u an titative determ inations and a n a ly s is , for s p e c if ic compounds, such as ch lo rin a te d hydrocarbons and PC B's, were then performed by gas chromatography (GC) and combined gas chromatography/mass spectrometry (GC/HS)^ The r e s u lt s obtained from a l l of these an alyses a re presented in the follow ing paragraphs. The methods themselves a re described In Section G.3> Qua 1it a t iv e d ata. The purpose of the q u a lit a t iv e or survey a n a ly sis was to id e n tify any m aterial in the samples which were not present in the o rig in a l waste, and therefore were not expected. In ad d itio n , th is survey a n a ly s is searched for secondary or inconplete combustion products which would r e s u lt from the waste being converted to compounds ACf* CC 184 ocher chan CO^, H^O, and HC1. The q u a lit a t iv e data w i l l be described in two groups: V) E x tra ct concentrate residues and 1) m aterial desorbed from the sorbent tubes. I) Ex tract concentrate residues - The r e s u lt s of these q u a lita tiv e analyses on the re sid u e s, obtained by evaporating a liq u o ts of the solvent e x tra c t con cen trates, are summarized and presented In Table G.6. This table includes: - The id e n t ific a tio n of the e x tra c t samples according to the coding shown in Figure G . l . Th is coding shows how c e rta in re late d samples have been combined. - The Sample/Aliquot Ratio shows the to tal volume of e x tra ct prepared by ORF in the numerator and the volume received by TRW in the denominator. One-fourth of each e x tra c t was sent to TRW. - The amount of residue found fit TRW's portion of the to ta l e x tra c t is expressed as m illigram s (mg). Values have been corrected for appropriate blanks. - The Total Residue is that amount (in mg) o f residue found in ** TRW's p o rtio n , m u ltip lie d by the Sample/AlIquot R atio. This value is in d ic a tiv e of what may be found in the total e x tra ct i f analyzed by the same procedures. inspection of the 1R and LRHS data revealed that the only m aterials detected in the evaporated residues of the various extracts (except fo r the contaminated BLB samples) were s ilic o n e s , hydrocarbon o i l s , fa tty a cid s or f a t t y acid e s t e r s , and p h th a llc a cid e s te r s . These same m a te ria ls were a ls o present in the blank and control samples; th e ir concentrations vary from sample to sample as estimated by LRHS. These m a te ria ls / as c la s s e s of compounds, are not to x ic and would not be considered an em issions hazard even I f they were present In much higher co n cen tratio n s. These m a teria ls are not believed to have come from v the combustion gas. Samples front the EPA t ra in did not show Increased le v e ls of these m a te ria ls co n siste n t with more than ten -fo ld sample volumes. In the a n a ly s is o f trace o rg a n ics, i t becomes very d i f f i c u l t to completely avoid these ubiquitous o i l s , p la s t ic is e r s , lu b rica n ts, antioxidants, ........-- - , . ACM CCti2 7 7 I 185 TAeLE G.6. SUMMARY OF ORGANIC QUALITATIVE SURVEY ANALYSES OF SAMPLE EXTRACTS Total Extract^'"''"^ Samele Identification ^ ^ T R U Aliquot (ml) 3dsei ine Burn A BLA-0RF-FE-T3 BIA-0RF-PRE*NA0HE-T3 6LA-ORf-H2OE-T3 BU-EPA-PRE*FE-T3 BLA-EPA-IE-T3 Basel ine Bum B BlB-ORF-FE BlB-ORF-PRE+NAOHE BLS-0RF-H20E blb-epa-pre*fe 8LB-EPA-IE Uaste Burn B WBB-ORF-FE 88-ORF-PRE+N40HE HSB-ORF-H2 OE H8B-ORF-NAOH-AE-T3 wbb-epa-pre*fe -SB-EPA-IE Waste Bum C WBC-ORF-FE WBC-ORF-PRE+NAOHE WBC-ORF-HjOE W8C-0RF-NA0H-AE-T1 WBC-EPA-PRE+FE WBC-EPA-IE 200/50 200/50 100/25 300/75 100/25 600/150 600/150 300/75 900/225 300/75 600/150 400/100 200/50 N/AC 900/225 300/75 600/150 600/150 3X/75 N/Ac 900/225 300/75 Survey Residue Found In TRW Aliquot (mg) 3.?3 N/Dj N/D0 0.12 2.09 s. 0J4 N/06 0.50 0.48 0.04 0.07 N/Db N/Db 0.28 1.50 11.18 0.58 N/Db 14.83 0.23 1.12 6.25 Total Residue (mg, corrected for Aliquot Factor) 12.92 N/0 (<1.00 N/D (<2.68) 0.48 8.36 1.36 N/D (<1.00) 2.00 1.92 0.16 0.28 N/0 {<1.00) N/D (<2.68) 1.12 6 .X 44.72 2.32 N/D Ul.OO) 59.32 0.92 4.48 2 5 .X #There H no significant difference In the results between U *t run*. The materials In the residues are not believed to have come from the coefcustlon gas. bN/0 Not detected at levels higher than the blank. CN/A Not applicable; the total caustic ls*1nger contents from one test ere used. :P I. I ACM C C 8 7 8 I 186 e t c . , that can enter the te st m atrix in very small amounts d espite the most ca refu l and exten sive preparation of equipment and se le c tio n of reagents. One cannot be sure-whether the source of these m a terials is the hardware, the. reagents, the samples them selves, or a combination of these. A large number of blank and control samples might have distinguished the source, but due to the n o n c ritic a l nature of the m a te ria ls, sucn an examination was c le a r ly not warranted. tt was pointed out that the survey a n a ly s is of the BLB samples known to contain PCB's I G .l] d id , In f a c t , reveal t h e ir presence. Evidence of PCB,s was c le a r ly shown in the LRMS data for the follow ing BLB sam p les: - ORF t r a in , combined probe rin se e x tra cts and c a u stic impinger e x tr a c ts ; - ORF t r a in , water impinger e x tr a c t s ; - EPA t r a in , combined probe rin se e x tra ct and f i l t e r e x tra c t. Since these PCB's were found only as estimated minor co n stitu en ts in samples whose total weights were le ss than one m illig ra m , the a b il i t y of the mass spectrometer to detect these small q u a n titie s was c le a r ly e sta b lish e d . The l RMS procedures can u su a lly detect 10 micrograms, or about one percent of a ty p ica l I m illigram sample. The presence of a s p e c if ic compound in acT organic m atrix is not normally detected by IR with adequate c e rta i nty unless present a t 10 percent or higher. The s e n s it iv it y and detection lim its of these techniques were discussed In Section G .3-2. It is stressed that chlorinated species w ere'specifical1y searched for in a l l o f these residue samples and none were found except for in the contaminated BLB samples described above. 2) M a teria ls desorbed from the sorbent tubes - The Chromosorb 102 sorbent tubes from the ORF sampling t ra in were therm ally desorbed to recover the c o lle c te d sample according to the procedure described in Section G.3-1The desorbed gases were then analyzed q u a lit a t iv e ly by LRMS using a gas in le t system. The co nclusions drawn from reduction and In te rp re tatio n of the LRMS data were that a lV of the sorbent tubes contained e s s e n t ia lly the same types of m a te ria ls ; however, the r e la t iv e amounts of these co n stitu en ts did appear to vary from sample to sample. ACM GCb27 187 Although the LRMS is a q u a lita tiv e technique, estimates can -3 nade through inspection of the data to provide an indication as to hether a species is present as a trace, minor, moderate, or major component. Such a sjmmary for the six selected sample sorbent traps that were analyzed <5 presenceJ in Table G.7. A n unused sorbent trap was a lso analyzed by the s3ne procedure to estimate background contributions by the Chromosorb 102. Jnfortunateiy a leak in the system resulted in the loss of sample and, Therefore, usable data. Only carbon dioxide was seen in the mass spec data at greater than trace lev e ls. The v o la tile compounds removed from the sorbent traps and collected in the gas sample bulbs were largely the oxygen, nitrogen, carbon dioxide, and water that are the usual species found in combustion cases. The remainder of the compounds detected were those that could indicate incomplete combustion, or possibly the formation of other compounds. The levels at which these remaining compounds could be oresent were estimated on the basis of LRMS si nstrument response for key peaks attributed to these trace residual organics. These estimates indicated that the Cj to Cj hydrocarbons present as a major component of these trace residual organics could be present In the sampled combustion gasss in the I to 10 ppm range. The minor or trace components of these trace organics were believed to be. present at very much lower levels. A summary of the LRMS peak patterns observed In the spectra of the desorbed gases and the compounds assigned to these p atte rn s is q '.en in Table G.8. Q ua-titative data. Quantitation of the compounds detected and Identified i" the test samples was performed by GC and GC/MS techniques. I; Extract concentrates ~ The chromatography with flame ionization . detect ion was performed fchiefly to analyze organic sp ecies. It should be noted that the FID is quite se n sitiv e to chlorinated hydrocarbons as long as chlorinat'on is not complete. At the electrometer settings used, the s e n s it iv it y , or the minimum detectable q uantity, was 0.002 g/ul as benzene, naphthalene, or Aroclor 1232. The chief re su lt of the analyses using the FID was that none of the concentrated samples showed peaks other than those found In the ADM C Ct^cC <f '> ) TABLE G .7. APPROXIMATE CONSTITUENT LEVELS OF TRACE VAPOURS DESORBED FROM SORBENT TUBE SAMPLES BV LRMS Sorbent Tube Sample WBB-T1 WBB-T2 NBC-Tl WBC-T2 BLB-T1 BLB-T2 Cl - 5 Hydrocarbons Minor Major Major Major Major Major Nitromethane Trace Minor Minor Minor Moderate Minor ; a Constituent Level NO. NO and c P o ssib ly Ethanol Benzene Methyl Chloride Trace Major Major Major Moderate Moderate Trace Moderate M inor, Moderate Major Moderate NDb Moderate NO NO NO NO Substituted Benzene NO Trace Trace trace Trace Trace Methyl Siloxane Major Minor Minor Trace Trace Trace aThe range of tra ce to major le v e ls represents 0.1-10 ppm concentrations In the flu e gas. ^ND - Not Detected. I > . 4 * * I* T9 ? POO * 3 V 189 TABLE G.8. .SUMMARY OF THE INTERPRETATION OF LRUS SPECTRA FOR TRACE VAPOURS DESORBED FROM SORBENT TUBE SAMPLES Peak Pattern (Peaks at Atomic Mass U n i t s (AMU)) Assignment CO 15 16 17, 18 27, 2 9 . <>i. * 3 , 55, 57 u , 16. 28. 32 kk 30. i*6, 61 30, A6 50, 51, 52 50, 52, 15 91 1*7., 207, 281 Methyl group, CH^ Methane, CH^ Water C*--CPc hydrocarbons Ni trogen/oxygen C02 P o ssib ly nitromethane CH^NO^* NO, NOj* and/or ethanol Benzene ** Methyl ch lo rid e CH^Cl Substituted benzene ring ( e .g ., toluene) Methyl silo xa n e -The 30 AMU peak i.as very large and i t is believed that n i t r i c oxide, nitrogen dioxid e, ethanol and nitromethane may have a l l contributed to i t . solvent co n tro ls. Consequently, a " le s s than" value was assigned in terms of benzene fo r the a n a ly sis of hydrocarbons, or A roclor 1232 for the a n a ly sis of ch lo rin a te d o rg a n ics. The r e s u lts are given in Table G.9. The values presented in Table G.9 were derived In the following manner. The sample labeled WBB-EPA-FE+PRE, fo r example, totaled 225 ml as received . A 213 ml portion of th is sample was concentrated to 10 ml in a .Kuderna-Danish evaporator. The chromatogram of the concentrated sample showed no peaks other than so lv e n t. Since the aromatic waste contained considerable chlorinated hydrocarbons, quantitation as Aroclor 1232 was considered appro priate. T h u s, WBB-EPA-FE+PRE contained <0.002 wg/wl o f sp ecies as Aroclor 232. The amount of m aterial in the e n t ir e received sample volume of 225 nl was determined. T h i s value was m u ltip lie d by fo u r, sin ce only one-quarter ACM C C 62 6 l /1 TABLE G .9. RESULTS AND DETECTION LIMITS FROM GC/FID ANALYSIS OF THF CONCENTRATED EXTRACTS Sample ID BlA-EPA-FE*rAE-TJ -IE-T3 -0AF-FE-T3 -H20E-T3 -PAE+RA0HC-T3 HB9-EPA-FE+PRE -IE -OAF-FE -H20E . -PAE+NAONC M8C-EPA-FE+PAE -IE -OAF-FE -HjOE -PAEMA0HE ng/ul as Halocarbon" 0.052 0.26 0.26 0.061 0.21 ND(<0.0038)e 0.89 0.12 0.26 0.053 0.019 (lo s t) ND(<0.0038) NOf<0.0038) 0.26 Volta of Sanple (m l)- 75 25 50 25 50 225 75 150 50 100 225 75 150 75 150 Aliquot Factorb Fraction of Saaplec Volume of Sample Gas, Std. o3 * Concentration In Flue Gas (mg/*3) 67/20 21/10 46/10 21/10 42/20 213/10 63/10 138/10 42/10 42/10 213/10 63/10 138/10 63/10 138/10 i 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 1/4 3.76 3.76 0.32 0.32 0.32 9.62 9.62 0.83 0.83 0.83 9.97 9.97 0.73 0.73 0.73 0.001 0.003 0.04 * 0.09 0.06 N0(<0.00002) 0.004 0.006 0.02 0.003 0.00008 (lo s t) HD(<0.00002) NO(<0.00002) 0.015 , Noliam of sta p le received by TMf. A liq u o t Factor: fo r e n q > U 67 *1 of the 75 a l BLA-EPA-Fe*PAE sa^ tle received was concentrated to 20 ! fo r a n a ly s is . 'fr a c tio n of S tap le: A ll sta p le s received by TAM were 251 o f the to ta l s ta p le . ^ h a Instrument was c a lib ra te d w ith A rpclor 1232. However, the A roclor peak pattern was not found In any s a p l e chromatogram. High s e n s it iv it y GC/KS ()u g /a i of flu e gat) did not detect any chlorinated species In samples WBC-ORF-FE and HK-0AF-K2QC. *ND'* Not detected. Values 1n parentheses In d icate the detection H a l t s . 1 -, 191 of the whole sample was divided by the volume, in m^, of gas sampled. A concentration value of <0.009 mg/m^ in the flu e gas was thus derived for the sample WBB-EPA-FE+PRE. SFnce the minimum detectable quantity Is constant and sin ce both the volume of gas sampled and the volume of e x tra ct v a ry , t h is method of c a lc u la tio n gives d iffe re n t values of mg/m3. It must be stre ssed that these analyses showed no sp ecies other than so lve n t. Since the ECD is much more s e n s it iv e to po lychlo rin ated species than the FID, the an alyses using e le ctro n capture d etectio n were performed to determine po lych lo rin ated compounds below the level that could be detected by the FID a n a ly se s. At the electrom eter se ttin g used on the electro n capture d e te cto r, the minimum d etectable q u a n titie s for ty p ica l species of in tere st were: - A roclor 1232: 0.0038 ng/ul - Naphthalene: 0.066 ug/yl - Benzene: 0.11 pg/ul. ^ The most noteworthy re s u lt of the GC analyses using the ECD was that none of the concentrated e x tra c ts showed peak p atterns corresponding to those given by A roclo r 1232 (the PCB mixture most s im ila r to that burned in waste bum C) or A roclor 12^2. The q u an tified ECD re su lts are gi ven i n Tab Ie G.10. Some of the concentrated sample e x tr a c t s , e . g . , WBB-EPA-FE+PRE, shewed no peaks other than so lve n t. Host of the e x tra c ts showed several very small peaks. A fter q u a n tita tio n , a l l but one of the "samples were below the level of in t e r e s t , 0.1 mg/m^f and no fu rth e r e f f o r t was expended. One sample, WBC-0RF-FE, showed a number of peaks, some o f which were large. Another sample, WBC-ORF-H^OE, showed a large number of small peaks. Both of these samples were analyzed by GC/HS. The GC/HS an alyses of the f i l t e r e x tra c t indicated one vanishing small hydrocarbon peak, a small phthalate e ste r peak, and 10 peaks which were various t r lm e t h y ls iIy 1 compounds (TMS). The GC/HS analyses of the water e x tra ct showed extremely small hydrocarbon peaks and 16 peaks th^t were t r im e t h y ls i1y 1 compounds. Because these THS compounds were most lik e ly the r e s u lt of some contamina tio n , the peaks appearing in the chromatograms were considered spurious. ACH CCtttH TABLE G.10. RE SUL I S AND DETECTION L I M I T S FROH f.C/ECD A N A L SIS OF THF CONCENTRATED EXTRACTS Settle ID ug/w1 as Renzene Htaiglo/cualrbaosnd B.A-EPA-FC+PRE-T3 -IE-T3 -0RF-FE-13 -#?ot-rj -PKHiMHE-ra kM-tPA-F$*PllE -IE -Otf-FE -M20E -PRE+NAO* WC-EPA-FE+PK -IE -ORF-FC -H20C -PtCNA0HE ND(<0.002)* NDi <0.00? NO <0.002 NO <0.002 NO <0.002] - - * - * - - - NDI <0.002] NO <0.002 NO 0.00? NO <0.002 ND| <0.002] NDI <0.002) NO 0.002) NO <0.002) HD <0.0071 NO[<0.002) Volin of AlIquoi Fraction of Sample, ml* F acto r, Sanplec 75 67/20 1/4 25 21/10 1/4 50 46/10 1/4 25 21/10 1/4 50 42/20 1/4 225 213/10 1/4 75 63/10 1/4 150 138/10 1/4 50 42/10 1/4 100 2/10 1/4 225 213/10 1/4 75 63/10 1/4 150 138/10 1/4 75 63/10 1/4 150 138/10 1/4 Volune of Sanple Gas, Std. n3 i Detectable t ln lt s of Concentration in Flue Gas, ng/n3 3.76 3.76 0.32 0.32 0.32 9.62 9.62 * 0.83 0.83 0.93 9.97 9.97 0.73 0.73 0.73 N0(<0.05) MO(<0.03) NO <0.3) HD <0.3 NO[<0.6) ND <0.009) NO <0.01) ND < 0.l) NO <0.1) NO <0.1) NDi[<0.009) HD <0.01) HO <O.I) ND <0.1) NO[<o.iJ H ilv * of smj1o rtcclw d by TW. k Aliquot Factor: fo r i i H p l e 67 nl o f thw 75 a l BLA-EPA-Fe*PRE saa^le received was concentrated to 20 nl fo r l n a l y s l t . 'fr a c tio n o f saa p la. A ll sanples received by TW were 251 o f the to ta l taa p le . dThe I n s t r u m t was ca lib ra te d w ith benzene o r A roclor 123?. Hm ever, the A roclor peak pattern was not found In any iw p l ch ren eto g r. Nigh s e n s it iv it y GC/NS (3yg/3 o f flo e gas) did not detect any chlorinated species In sanples dBC-ORF-FE and WC6-0RF-H20C. S o Not detected, values In parentheses in d ica te the detection U n it s . 53?900 W3? 1 193 These GC/MS a n a l y s e s d i d not d e t e c t any c h l o r i n a t e d s p e c i e s in e i t h e r sample ( t y p i c a l s e n s i t i v i t y o f 3 ug/m3 of the f l u e g a s ) . 2) A iiq u o tro f neat extracts - There was concern that the process of concentrating the sample e x tra c ts, as received in pentane and/or hexane, might cause the loss of re la t iv e ly low boiling halocarbons such as chloroform which might 'be expected in the samples. Therefore, the retained portions of the as-received extract samples were chromatographed on the Tracor instrument. The column was operated Isothermally at 70C to increase resolution of low boiling compounds. The re su lts are given in Table G.1T. Host of the samples showed no peaks other than solvent. Of those samples showing peaks other than solvent, a l l quantified below the level 7 of in terest (0.1 mg/m ) , and no compound id en tifica tio n by GC/MS was performed. 3) M aterials desorbed from sorbent tubes - The sorbent traps were desorbed as discussed in Section G.31< Portions of^the desorbed vapours contained In the sample bulbs were chromatographed isothermally at 70C with electron capture detection. The samples were q u a lita tiv e ly sim ila r to the sorbent trap blank (unused chromosorb 102) and the baseline B (fuel o il burn) samples. The waste burn B samples each contained a peak which might be CH C l: UBB Test 1, 0.0001 mg/m3 ; and WBB Test 2, 0.0002 mg/m3 . The waste burn C samples were e s s e n tia lly the same, q u a lita tiv e ly and q u an titatively as the blank and baseline B samples. G .4.2.2 Inorganic ch aracterizatio n . Inorganic elemental concentrations in the flue gas were determined by a n a ly sis of the p articu late f i l t e r s and the aqueous probe rinse and impinger samples. The f i l t e r s (including the filt e r e d probe rin se insolubles) were a l l acid digested. Out of these 2 0 dig ests ( 1 8 test samples and two b lanks), three samples were prepared for a trace element survey a n a ly sis by combining equal aliq uo ts from each of the three tests of WBB, UBC and BL8 using the f i l t e r samples from only the EPA sampling tra in . The EPA tra in samples were selected on the basis of 1 ) representing larger sample gas volumes, and 2 ) having been taken by a method sp e c ific for accurate and representative p articu late sampling. The re su lts of the survey an a ly sis are shown in Table G.12. The blank ACM CCA266 4 TABLE G.ll. RESULTS AND DETECTION LIMITS FROM GC/ECD ANALYSIS OF THE UNCONCENTRATED EXTRACTS ng/pl as Halocarbonc Volume of Samp fe Fracti onbof Samp 1e V o 1ume of SampleGas. Std. nr Concent rat ion in Flue-Gas, mg/m- BLA-EPA-FE+PRE-T3 -IE-T3 -0RF-FE-T3 -H2OE-T3 -PRE-NA0HE-T3 ND(<O.OO30)d n d (<o .0 0 3 8 ) ND(<0.0036) 0.0051 ND(<0.0038) 75 25 50 25 50 41#. l i i 3.76 3.76 0 .3 2 0 .3 2 0.32 ND(<0.0003) WD(<0.0001) ND(<0 .002) 0.0016 ND(<0.002*<) WBB-EPA-FE+PRE -IE -ORF-FE -H20E -PRE+NAOHE ND(<0.0038) O.liO ND(<0.0038) ND(<0.0038) ND(<0.0038) 225 75 150 50 . 100 h h i h h 9 .6 2 9 .6 2 0.83 0.83 O.8 3 ND(<0.0O0M 0.012 . ND(<0.003) ND(<0.0009) ND(<0.0018) WBC-EPA-FE+PRE -IE -ORF-FE - h2oe -PRE+NAOHE ND(<0.0038) 0 .1 0 ND(<0.0038) 0.15 ND(<0.0038) 225 75 150 75 150 u i i i i 9.97 9.97 0.73 0.73 0.73 ND(<0.0003) 0 .0 0 3 0 ND(<0.0031) 0 .0 6 1 ND(<0.003I) g Volume of sample received by TRW. * ^Fraction of sample. A ll samples received by TRW were 25% of the total sample, cThe Instrument was calib rated with Aroclor 1232. However, the Aroclor peak pattern was not found In any sample chromatogram. High s e n s it iv it y GC/MS (3 wg/m^ of flu e gas) did not detect any chlorinated species In samples WBC-ORF-FE and WBC-ORF-H20E. ND - Not detected. Values tn parentheses indicate detection lim its . TABLE G. 12. TRACE METAL SEM I-QUANT I TAT I VE*3 SURVEY OF FILTER DIGESTS BY ICPOES Element A1 Ba B Ca Cd Cr Co Cu Fe Pb Mg Mn Ni P K Si Ag Na Sr Ti V Zn WBB 0.078 0.01 1 0.018 2.2 0.0006 0.003 NDC 0.003 0.18 0.047 0.076 0.003 0.003 0.014 14. 0.082 0.0003 1.8 0.004 0.002 0.0006 0.014 Average Concentration*3 (mg/m^) WBC BLB 0.51 0.014 0.084 1 1. 0.002 0.006 0.001 0.007 1.1 0. 12 0.38 0.018 0.006 0.055 32. 0.051 0.003 5-5 0.013 0.008 0.002 0.042 0.26 0.012 0.038 6.5 0.001 0.004 NDC 0.003 0.73 0.044 0.21 0.013 0.002 0.044 6.7 0.093 0.0004 2.4 0.0008 0.004 0.001 0.027 aAccuracy estimated to be a f a c t o r of 2 or b e t t e r . C a l c u l a t e d based on average sample gas volumes o f : WBB - 3.2 m| WBC - 3-3 m BLB - 3-4 nr Not co rrected for f i l t e r co n t ri b u t io n CNot detected (<0.0003 ppm) (an a ci d dig est of unused f i l t e r s ) sample was not surveyed, so the values reported in t h is table are uncorrected for the f i l t e r background. Of the 32 elements that can be determined by the ICPOES a n a l y s i s , ten were not detected in the f i l t e f digest samples. These ten elements w ith t h e i r lower dete ction l i m i t s are l i s t e d in Table G.13 along with a c a lc u la t io n of the average detectable lim it for each of these elements in the flue gas. ADM CCtccc *96 TABLE G .I 3 . LIMITS OF DETECTION FOR UNDETECTED ELEHENTS BY ICPOES | EIemen t Au As Be Eu Mo Se Te Sn W U ICPOES Detection Lim it (ppb) 5 ko 1 15 11 60 65 50 90 80 Average Detectable3 Lim it In Flue Gas (mg/m3) 0.00008 0.0006 0.00002 0.0002 0.0002 0.0009 0 .0 0 10 0.0008 0 .001k 0 .0 0 12 1 aBased on an average sample gas volume of 3 * 3 cubic meters. The re su lts of th is survey indicated only one element, lead, was present at p o te n tia lly toxic lev e ls in the stack. While not a problem with tne stack configuration in the present study, the concentration anc the effect of thermal dispersion should be considered when extrapolating these resu lts to other k iln s . To be sure of an accurate measurement of the levels of lead and certa in other toxic metals which the waste analysis indicated might be present at le v e ls of In te re st, a quantitative determina tion by AA5 was performed on fiv e elements. The re su lts from th is analysis are presented in Table C .lA . The AAS re su lts were corrected for background levels as determined by an alysis of the f i l t e r blank sample. It should be noted that back ground contributions were not only from trace contaminants In the f i l t e r m aterials but also in some cases from matrix e ffe c ts in the mixed acid solutions. AAS was a lso used to analyze for selected elements in the ! aqueous probe rin se and impinger samples. Table G.I5 l i s t s the re s u lts' from these analyses and shows that, in general, nothing sig n ifica n t was' found. No background or blank solutions were av ailab le so the BLB resu lts should be used to correct for background e ffe c ts . The re su lts for these ! aqueous samples were le f t in ppm because the total sample volumes were unknown. ACM c c e ^ e s fi f i =_x: L I \ t i v _ ,,/ i 197 TABLE G .I4 . CONCENTRATION OF TRACE METALS IN EFFLUENT GAS PARTICULATE MATTER BY AAS ra -sE. I Waste Burn WBB Train EPA WBC EPA BLB EPA Test 1 2' 3 1 2 3 1 2 3 Concentration of Element Ba Cd Co Cr <0.008a <0 . 0 0 8 <0 . 0 0 8 <0 . 0 0 1 <0 . 0 0 1 0.001 <0 . 0 0 2 <0 . 0 0 1 <0 . 0 0 2 <0 . 0 0 9 <0 . 0 0 2 <0.004 0.013 <0 . 0 0 8 0.027 0.001 0.002 0.003 <0 . 0 0 2 <0 . 0 0 2 <0 . 0 0 2 <0 . 0 0 0 <0 . 0 0 8 0.003 <0.007 <0 . 0 0 8 <0 . 0 0 8 0.001 0.002 0.001 <0 . 0 0 2 <0.003 <0 . 0 0 2 0.001 <0.007 <0 . 0 0 2 Pb 0.028 0 .0 12 0.079 ,0 . 0 9 6 0.103 0.153 0.040 0.062 0.015 a " i " , a less than or equal to sign, indicates those elements which were detected but not s 1 gn i f i c a n tiy above background levels a TABLE G.15. CONCENTRATION OF TRACE METALS IN AQUEOUS SAMPLES BY AAS Train/Componen t EPA-Probe Rinse Impingers Waste Burn vs., Basel ine WBB WBC BLB a WBB WBC BLB Concentration of Element (ppm) Pb Cr Cd Co NO 1-9 NO 0.03 NO NO NO 0 .0 1 NO . 10 .2 ND ND NO NO ' NO NO 0.02 NO NO NO ND 0.05 NO NO cce^ 0 * 198 Analysis of the f i i ter acid digest, probe rin se , and impinger samples from the ORF train by .AAS confirmed the presence of those elements found i-n the EPA train samples. However, since the ORF train was not operated with the intent of co llectin g an accurate p articu late sample ( i . e . . iso k in e tic, traverses, e tc .) the AAS re su lts for this train were not calculated out to concentrations in the flue gas. G.w.3 program: Solid residues and efflu en ts from the SLC k iln process Of the four types of so lid samples collected during the test - ci i nker product; - clinker fin e s; - e le c tro s ta tic p recip itato r discard dust; and - cement mixes, analyses were performed only on the cl yiker product and discard dusts for the following reasons. The clin k e r fines were recovered from a ir blown tnrough the clin k e r product and were then returned to the c lin k e r storage. The cement mixes were simply clin k e r product with gypsum added. Thus, neither of these m aterials represented unique samples. The following paragraphs describe the re su lts of analyses performed on the clin k er products (CP) and discard dust (OD) samples. If sig n ific a n t amounts of toxic m aterials were found in the analyzed samples, further tests on the remaining samples would have been conducted. However, th is did not prove to be necessary. G.A.3.1. Organic co n stitu en ts. Portions of the clin k er product and discard dust samples were extracted with pentane using a Soxhlet extractor and preextracted paper thimbles. A blank sample was prepared by running a pentane extraction on an empty, precleaned thimble. In addition, a doped control sample cons'isting of ^ 3 0 grams of cl inker^ to which 0 . 9 mg of a known PCB mixture had been added, was also extracted for a n a ly sis. This doped sample represented a PCB concentration ^30 ppm In the clin k er product. These samples were a 11 analyzed both q u a lita tiv e ly and quantita tiv e ly by the same techniques described in Section G .A .2 .1. Q ua!itative data. Aliquots of the pentane extracts were evaporated and the residue was weighed and scanned by both IR and LRMS. These results AC* CCS251 199 TA3LE G .16 . RESULTS OF ORGANIC SURVEY ANALYSIS ON CLINKER PROOUCT AND DISPOSABLE DUST SAMPLES Sa.no !e 1den t i f ca t on LB-CP BLB-QD WBB-CP WBB-DD WBC-CP WBC-DD PCB Doped Control Weight of Extracted Sample (q) Weight of Residue in Extract (mg) Q u alitative Nature of Residue 3 0 .2 7 0 2085 <1.16"" 1.31 29. ^ 5 I 8 . 9 A8 21.816 33.112 ' I .30 1.16 ,* <1.16 -w <1.16 , Hydrocarbon o i l s , phthalate e ste rs, traces of fatty acids JL and silic o n e s 2 9 .9 9 8 * <1.16 A ll of the above compounds plus polychlorinated biphenyls The residue obtained from the blank thimble^and pentane sample was 1.16 mg. Sample values varied with some being less than 1.16 mg due to the v a r ia b ilit y of the background. Over th is range, the d ifferen ces in residue weights are not s ig n ific a n t. are summarized in Table G .16. In most cases, the amount of weighable residue found in the extracts did not exceed that found in the blank. The m aterials id entified by IR and LRMS are in d icative of the low level contamination by greases, o i l s , soaps, e t c ., that often accompanies trace organic a n a ly sis. Hydrocarbon o i l s , phthalate e ste rs, fatty acid s, and silic o n e compounds were found In a ll the extract residues including the blank. No-indication of the chlorinated species was found in any of the test samples. Polychlorinated biphenyls were c le a rly detected in the PCB doped control sample. The LRMS spectrum contained peaks at 220, 222, 290, 292, and 29k AMU. The PCB's present at 30 ppm In the doped clin k er were e a s ily found and i t is certain that much lower le v e ls could have been detected. It is estimated that the mass spectrometer can detect the strongest peaks from a spec ifVeal Iy searched for compound when present at the 1 0 ng level in a 1 milligram organic residue from the c lin k e r. This re su lts in a lower lim it of detection by th is q u a lita tive I t I' 1 ,I1>1< K! ; I I ACM CC&2S 1 200 survey technique of about 1 ppb-in the clin k e r or dust. The IR survey data was not useful in detecting PCB's at th is le v e l. Quantitative data. The clin k e r and dust extracts were analyzed by GC with both FID and ECO detection. The re su lts are given in Tables G-17 and G.18, resp ectiveIy. Using the FID, no peaks other than solvent were found, and the resu lts are expressed as minimum detectable q u a n titie s. Three of the samples showed r.o peaks other than solvent when analyzed with the ECD. The three samples which had peaks other than solvent using the ECO were quantitated at well below the level of in te re st. To ascertain the recovery of PCB's from the clin k e r and dust samples, a 30 gram sample of clin k e r was doped with 0.9 mg of Aroclor 1232, extracted and concentrated in the same fashion as the c lin k e r and dust samples. The ECO chromatogram of th is a r t i f i c i a l sample matched the pattern of a neat sample of Aroclor 1232, and the recovery of PCB's from the doped sample was calculated as 25%. This recqyery factor has been applied to the cl inker and dust samples in Tables G.17 and G.18. (The recovery factor is the reason why the values for the samples in Table G.17 exceed several ppm. If the minimum detectable quantity using the FIO Is 0.002 ug/ul, then a fte r applying the recovery fa c to r, the concentration in th^sample cannot exceed 0.008 ug/ul* For a 10 ml extract volume and a 30 g clin k er sample, the composition is thus <2 . 7 ug/g.) G.it.3.2 Inorganic ch a ra cte riza tio n . The CP and DO samples were also examined for trace metal content by spark source mass spectrometry (5SMS). The major constituents of both types of material were determined to be aluminum, calcium, iron, magneslun, sodium, sulphur, and s ilic o n . Other trace inorganic elements which were detected down to a 1 ppm concentration are liste d in Table G.19. It should be remembered that SSMS data are sem i-quantitative. The values reported should be considered as being a accurate within 5 0 0 % of the true value. The trace metal character of the c lin k e rs and dusts did not change sig n ific a n tly from one waste burn to the other. The r e la t iv e ly low feed rate of the wastes, compared to that of the precursor material and the primary fu e l, appeared to make any contribution of trace metals from the wastes n e g lig ib le. ACP CC6293 201 TABLE G.17. RESULTS AND DETECTION. LIMITS FROM GC/FID ANALYSIS Sample ID ug/vil asa Halocarbon3 Volume of Sample (ml) ug/g of . Sample Material BLB-CP W83-CP BC-CP 3LB-DD U6 S-DD WBC-DD ND(<0.'002)b ND(<0.002) ND(<0.002) ND(<0.002) ND(<0.002) ND(<0.002) io 10 10 10 10 10 N0(<3.0) ND(<3.3) ND(<4.5) no(<3.9) ND(<5.1) ND(<2.9) 3 The instrument was calibrated with Aroclor 1232. However, the Aroclor peak pattern was not found in any sample chromatogram. bND Not detected. Values In parentheses indicate detection lim its. TABLE G. 18. RESULTS AND DETECTION LIMITS FROM GC/ECD ANALYSIS Sample ID BLB-CP WBB-CP WBC-CP ELB-DD VBB-DD VSC-DD P+PCB ng/ul asfl Halocarbon" 0.059 0.11 ND(<0.004)b ND(<0.004) ND(<0.004) 0 .0 6 0 22.4 Volume of Sample (ml) 10 10 10 10 10 10 10 ug/g of Sample Material 0.02C 0 .04c ' - ND(<0 . 0 0 2 ) ND(<0.002) ND(<0.002) 0.02C 7.5 aThe instrument was calib rated with Aroclor 1232. However, the Aroclor peak pattern was not found in any sample chromatogram. bND - Not detected. Values In parentheses indicate detection lim its. cNo chlorinated species were detected in the q u a lita tiv e survey of these samples by LRMS. ACM CC294 * 202 TABLE G.19. SELECTED TRACE METALS IN SLC CLINKER PRODUCT AND DISCARD DUST SAMPLES 8Y SSMS Element As B Ba Be Ce Co Cr Cu Cs Dy Ga Ge La Li Mn Mo Nb Nd Ni Pb Pr Rb Sc Se Sm Sr Th Ti U V Y Zn Zr WBB CP DD <1.5 30 N70Dbb 10 ND 7 1 ND 1 1 ND 7 ND 100 1 3 10 7 ND 3 7 ND ND ND 700 1 700 ND 30 3 30 30 3 30 70 ND 30 ND 10 10 10 ND 1 ND 7 MQOO 100 1 3 10 3 70 3 100 1 10 ND 300 7 700 1 30 3 * 30 30 Concentration (ppm)a WBC BLB CP DD CP DD <1.5 30 300 3 10 too 1 ND 70 30 ND 3 30 10 37 ND 30 31 3 ,,i ND ND 30 10 30 100 300 300 11 73 10 10 7. 3 10 10 7 too 3 300 N ND ND 3 11 700 700 77 700 700 33 30 30 7 10 70 100 100 70 <2.9 30 70 7 10 1 30 7 ND 1 3 l 10 100 300 3 3' 10 10 ND 3 ; 10 7 "1 1 700 3 700 1 70 3 30 70 1 30 30 ND 10 ND 10 3 10 ND 10 3 7 100 100 1 3. 7 3 3P 1 70 3 .1 ND 100 l 700 ND 30 3 30 10 3 SSMS data g e n e ra lly ranges w ith in 500$ accuracy. bND - Not Detected (<1 ppm). ACM C C 8 2 S 5 203 REFERENCES G .l Communication, Gordon Thomas, Ontario Research Foundation to Arnold Grant, TRW Systems, 7 -A p ril, 1976 x \ C C ci^ :k APPENDIX H DEVELOPMENT, CONSTRUCTION AND EVALUATION OF A COLLECTION SYSTEM FOR LOU MOLECULAR WEIGHT HALOCARBONS by F .J . Hopton and G.H. Thomas Ontario Research Foundation Mississauga, Ontario CC62S7 D 207 APPENDIX H DEVELOPMENT, CONSTRUCTION AND EVALUATION OF A COLLECTION SYSTEM FOR LOW MOLECULAR WEIGHT HALOCARBONS H. 1 Summary Numerous experiments have been performed to determine the e ffe c tiv e n e ss of absorbents and adsorbents to remove and re ta in low m olecular weight halocarbons from a flowing gas stream. I n i t i a l stu d ie s using a p a rtic u la te sampling t r a in , requiring a high gas flow rate of at le a st 0.5 cfm, ind icated that n eith er absorbents nor adsorbents would e f f ic i e n t l y c o lle c t compounds such as chloroform (CHCl^) present In the gas stream at low concentrations of a few p arts per b i l l io n . An in e rt adsorbent, e .g . Chromosorb 102, was found, however, to e f f ic i e n t l y remove and re ta in CHCl^ at much lower gas flow ra te s of 0.2 - 1.0 11tres/m inute. A c o lle c tio n system for low m olecular weight halocarbons was constructed and evaluated at low flow rates over sampling periods of up to four hours duration. R esu lts obtained from several t e s ts Indicated that the system had a c o lle c tio n e f f ic ie n c y of b etter than 90% over the longest time period . Desorption of adsorbed halocarbons was re a d ily achieved by thermal treatment of the adsorbent. H. 2 Introduction mr Late in 1975 an experimental program to burn waste ch lo rin ated organic compounds in a ro tary cement k iln was c a rrie d out at the S t. Lawrence Cement Company (SLC) p lan t In M ississaug a, O ntario. The program, sponsored by Environment Canada, was designed to obtain information on - the use of waste chlorinated hydrocarbons as a supplemental fuel to the k iln ; - the reduction of a lk a li concentration of the c lin k e r by the ch lo rin e contained in the waste m aterials added; and, - the e f f e c t of em issions from the k iln on a i r q u a lit y . In order to c o lle c t the required inform ation, process and em ission samples were c o lle c te d during periods when d iffe re n t composite wastes were burned in the k iln . The em ission samples were analyzed for tra ce chlo rin ated organic compounds. ACM C C 6 2 9 b 208 During p relim in ary d iscu ssio n s between members of the Ontario M in istry of the' Environment (HOE), the Ontario Research Foundation (ORF) and SLC p rio r to in it ia t in g the study, some concern was expressed about the sampling methodology for tra ce ch lo rin ated organic compounds. Since a major program requirement was to obtain the p a rtic u la te emission rate during each b u rr, *i t was hoped that the EPA Method 5 sampling t r a in , used to c o lle c t p a rt ic u la t e m a te ria l, could a lso be used fo r the c o lle c tio n of chlo rinated organic compounds. However, I t was not known whether absorbents or adsorbents were more e f f ic ie n t for c o lle c tio n of the organic compounds, or whether e ith e r c o lle c t io n medium would be e f f ic ie n t at the high gas flow ra te s required For p a rtic u la te sampling. HOE, th erefo re, requested that ORF complete a laboratory investigation to evaluate suggested procedures, and to develop an optimum method for the sampling and c o lle c tio n of tra ce ch lo rin ate d organic compounds which might be present in the k iln em issions. This report describes the te st program ca rrie d out Irrwthe laboratory and presents the data obtained using s p e c if ic ch lo rin ate d compounds as rep resen tative d o !lutants. H .3 BacKground Information >* Previous studies involving the incineration of chlorinated organic compounds [H .1] have indicated th a t, i f complete combustion to CO^, H^O and Cl^ or HC1 Is not r e a liz e d , then trace amounts o f v o la t ile compounds such as CCl^, CHCI^ and CH2 C I2 may be present tn the combustion ca se s. (Though it was u n lik e ly that any HC1 or C l j produced would pass through a cement k iln without reactin g to form a lk a li c h lo rid e s , It was necessary to consider these gases as p o ssib le p o llu tan ts for a n a ly sis also .) The EPA Method 5 t ra in for p a rtic u la te sampling is a lso used to c o lle c t organic compounds in the Impingers by condensation and scrubbing. The e f f ic ie n c y of c o lle c t io n for a p a r t ic u la r substance Is dependent on gas flow rate and v o l a t i l i t y of the organic compound. For compounds more / o la t ile than w ater, the c o lle a t Ion e f f ic ie n c y is probably clo se to zero. * To c o ile c t these v o la t ile compounds, s p e c if ic organic solvents have been used in the impingers, rep lacing the water normally p resen t. There i s , however, l i t t l e data a v a ila b le on the c o lle c tio n e f f ic ie n c ie s obtained. *1*1 AC* c c e z s s 209 In recent y e a r s , o r g a n i c compounds in the ambient a i r have been c o lle c te d using inert adsorbents [H.2]. These m a te rials have tended to replac e a c t iv a t e d carbon for most a p p l i c a t i o n s s in ce they are unaffected by water vapour, and reco ver y o f adsorbed s p ec ie s i s co n s id e r e d , in g e n e r a l , to be e a s i e r and more e f f i c i e n t . Recovery o f a ds orbed s p e c i e s may be accompl i she d e i t h e r by thermal d e s o r p t i o n o r s o l v e n t e x t r a c t i o n An a p p l i c a t i o n o f t h i s te ch ni qu e has been to c o l l e c t p o l y c y c l i c o r g a n i c compounds from co n b u st io n e f f l u e n t s [ H.3] B e a r i n g in mind t h a t , in the s tu d y to be c a r r i e d out at SLC o n l y trace amounts o f more v o l a t i l e o r g a n i c compounds were expected to be in the k i l n e m i s s i o n s , i t was decided to e v a l u a t e both a d s o r b e n t s and a d s o r b e n t s in an EPA Method 5 sa mp li ng t r a i n f o r c o l l e c t i o n , r e t e n t i o n and subsequent reco ver y of compounds such as CHCl^ and CH^Cl^. I f a c o l l e c t i o n method proved to be e f f e c t i v e f o r ca p tu r e o f these compounds, then i t would almost c e r t a i n l y be e f f e c t i v e f o r the c o l l e c t i o n of h i g h e r mo lecular weight s p e c i e s , such as those c h l o r i n a t e d compounds present in the waste feed m aterial. H.A Test Methodology The normal gas flow r at e thr ough a p a r t i c u l a t e sampli ng t r a i n to m a i n t a i n i s o k i n e t i c s a mp l i ng f o r most s o u r c e s i s 0 .5 to 1.0 cfm. A flow rate of 0.5 to 0.7 cfm was, therefo re, chosen for i n i t i a l experiments u s i n g the EPA Method 5 t r a i n , which was I l l u s t r a t e d in schematic form as F i g i r e A . 5 in Ap pendix A. H.A.l Retention and c o l l e c t i o n stu d ie s H.A.1.1 A b s o r b e n t s . I n i t i a l l y , t e s t s were conducted to fi n d a s u i t a b l e h i g h b o i l i n g s o l v e n t f o r use in an impinger system, in o r d e r to tr ap low m ol e c u l a r weight o r g a n o c h l o r i ne compounds, e .g . CHCl^, C Cl ^ , DCE and TCE. The f o l l o w i n g s o l v e n t s were tested fo r t h e i r s u i t a b i l i t y as an impinger s o l u t i o n f o r use in an EPA s t a c k s ampli ng t r a i n : - reagent grade toluene; - reagent grade xylene; and, - reagent grade decan^. A l l s o l v e n t s were r e d i s t i l l e d p r i o r to use in o r d e r to remove or l i mi t to u s a b l e l e v e l s i n t e r f e r i n g peaks in the GC-EC chr omat ogr aphic ACM CCE3CC 210 p ro file s which. tended to obscure'measurement of the components of interest. - T e s ts were made to see i f trace quantities of v o la tile chlorinated compounds could be retained in a solvent such as toluene with a high gas flow rate passing through the train for a period of four hours, a time anticipated for tests at SLC. A sampling tra in was, therefore, assembled in the laboratory with probe and oven temperature set above 250F to warm incoming gas. Toluene (100 ml) containing known amounts of 1,2-dIchIoroethane (EDC) and 1 ,1 , 2 -trichloroethane (TCE) (main constituents of the f i r s t waste feed in the waste chloride study at SLC) was placed In impinger A (see Figure A .5 ). 100 ml of pure toluene was placed in impinger B, impinger C was empty and impinger D contained s i l i c a g el. A ir was pulled through the sampling tra in a fte r cooling the impingers to ice* water temperature, at a flow rate of 0.5 cfm. After running for four hours, the contents of each impinger wer? noted and analyses made for EDC and TCE. At sp e c ific times during each t e s t , the parameters normally recorded during a source sampling te s t, such as impinger in le t and outlet temperatures and o r if ic e ip , were recorded. At the conclusion of each te s t, the total volume of gas sampled was determined. This type of experiment was repeated, with 100 ml of water in impinger A, 100 ml of toluene containing EDC and TCE in impinger B, impinger C empty and impinger D containing s i l i c a g e l. Both experiments were then repeated using xylene and decane as absorbents. Additional experiments were made using decane with known q u an tities-of CCl^, CHCl^ and CHCI2 In the standard impinger solution. H .4.1.Z Adsorbents. Though a large number of adsorbents are capable of adsorbing a wide v a rie ty of organic compounds, Chromosorb 102 was used for the tests described below because it was read ily a v a ila b le , and could be purchased without delay from various companies supplying chromatographic m aterials. In order to determine f a ir ly rapidly i f Chromosorb 102 would adsorb v o la tile chlorinated compounds, the following experiment was performed. 100 ml of decane containing known q uantities of CCl^, CHCl^ ir*. and TCE was placed in impinger B of the sampling tra in , and 25 gm of --- ,.rt*' - v-+1, . - Ch CCo3Cl S' " V .y 211 Chromosorb 102 in impinger D. Impinger A contained 200 ml of d i s t i l l e d water and impinger C was empty. Room a ir at 0.5 cfm was then pulled through the system fo r one hour and at. in te rv a ls of 10 minutes the decane so lu tio n was analyzed for ch lo rin ated organic concentration. A fter one hour the Chromosorb was removed from the impinger, placed in a septum sealed conta: ner and heated to 100C. Head-space samples were then analyzed for the re sp e ctiv e ch lo rin ated organic compounds. A ll three halocarbons present in the spike had been retained by the re sin . Following t h is experiment, a s e r ie s of te s ts were c a rrie d out using the 0RF te st duct system, depicted In Figure H .l. Th is system allow s dust or gaseous p o llu ta n ts to be added at a co n tro lle d rate into an a i r stream flowing through a one-foot diameter duct at flow rates of 500 to 250C cfm. Sampling ports are located at id eal and non-ideal p o sitio n s in the system, to allo w sampling of the a i r stream with an EPA type sampling t r a in or other c o lle c tio n equipment. A low flow of a ir was allowed to bubble through CHC1. in a midget impinger and bleed into the main a ir flow in the duct. By varying the c a r r ie r flow through the CHCl^. concentrations of 10 to 140 ppb of CHCI^ were obtained. In Che duct a ir stream. These concentrations were determined both from weight loss of CHC1. with time and d ire c t a n a ly sis of the a i r flowing' in the duct. j Tests were made with Chromosorb in impingers C and D but considerable carry over of the adsorbent occurred at flow rates of 0.5 - 0 .7 cfm. This was minimized in fu rth e r te sts by p lacing a fin e mesh screen on top of the Chromosorb layer in the impinger. At a CHCl^ concentration of 13 ppb, breakthrough of t h is p o llu ta n t with 25 gm of Chromosorb Pn the th ird impinger occurred w ith in 30 minutes. It was recognised that high volume flow rates through the adsorbent caused not only contaminant problems w ith the m aterial In the sampling tra in but saturated the bed f a i r l y ra p id ly . This la t t e r occurrence was proved by repeating, the experiment with the Chromosorb packed tig h tly in a c o lle c to r . A fin a l s e r ie s of te sts were, th erefo re, completed, evaluating both Chromosorb and a c tiv a te d carbon in tube c o lle c to rs at much lowe.r a ir flow ra te s of 1-^ 1itres/m in u te. The CHCl^ concentration in the a ir stream was analyzed before and a ft e r the c o lle c t o r at selected time in te rv a ls over a period of four hours. For both adsorbents c o lle c tio n of CHC1^ was more than 903: e f f ic ie n t for the duration of each t e s t . Since ACK CCb3C2 11 f|l!I 212 H> 1 DUST FEEDER 2 BLOWER 3 SAMPLING PORTS NONIDEAL LOCATION 4 SAMPLING PORTS IDEALLOCATION 5 SAMPLING PORTS 6 TO ATMOSPHERE F IG U R E H.l TEST DUCT SCHEMATIC CC83C2 213 desorption of adsorbed sp ecies from Chromosorb was r e a d ily accomplished normally and s in c e , u n like a ctiv a te d carbon, the adsorption c h a r a c te r is t ic s of Chromosorb are unaffected by'm oisture In the gas stream, Chromosorb i 32 was selected as a su ita b le adsorbent for the SLC waste ch lo rid e program, H. 5 C o llectio n System Evaluation This sectio n o u tlin e s the experimental protocol undertaken to v e rify the performance of Chromosorb 102 as a c o lle c tio n media for the . m cen tretio n and a n a ly s is (by adsorption/desorptlon) of low m olecular veight chlo rinated hydrocarbons from an a i r stream under conditions relevant to f ie ld sampling. The concentration and a n a ly s is procedure is e s s e n t ia lly an extension of the method used to concentrate organohalogen compounds present in potable water [H.A], Absorptive cap acity values were determined. Adsorptive capacity is defined as the amount of so lu te vapour retained by a given weight of sorbent. In th is study only low le v e ls of ch lo rin ate d hydrocarbons (1 - 10 ppb) were examined sin ce a n tic ip a te d le v e ls of organohalfdes In the sampling streams were expected to be lav ( I . e . ng/m^). The sampling c a rtrid g e s consisted of g la ss tubes (11 mm l .d . x various lengths) containing vario us depths of sorbent bed supported by plugs of s ila n iz e d g la ss wool. The tubes were conditioned by heating to 200C and passing a stream of through them (40 ml/min) for four hours. Known concentrations of a ir - s o lu t e vapour m ixtures were prepared. Tna so lutes examined were CHCl^, CCI^ and dfchloroethane^(DCE). The so lutes were examined both in d iv id u a lly and as m ixtures. The a ir - s o lu te vapour mixtures were prepared as fo llo w s. M ic r o litre q u a n titie s of organic compounds were in jected Into gas ja r s of known volume. The gas J a r was heated to 50C and, a ft e r co o lin g , a liq u o ts were*removed and added to Ted lar bags. The bags i n i t i a l l y were evacuated and then h a lf f i l l e d with a i r , at which stage the aliq uot from the gas j a r was in jected Into the bag. The bag was f in a lly f i l l e d to the d esired ca p acity (6, 20 and 40 l i t r e s ) . AQM CLt2C<i A schematic diagram of the component parts of the test assembly is shown below. The various parts of the assembly were connected to each other by means of Teflon tubing. D ifferin g a ir-so lu te mixture concentrations of CHC1j were used to determine the adsorptive capacity of the Chromosorb ID2. The CHC1^ concentrations used were 5 ppb and 10 ppb. A flow rate of 250 ml/min was maintained throughout the run. For these conditions the co llectio n e ffic ie n c y of Chromosorb 102 for CHCI^ was 100 percent. At various time in te rva ls the pump was stopped and the gas Jar disconnected. Aliquots (I ml) were removed from the gas j a r and analyzed for CHC1^ by gas chromatography (GC). The f i r s t detection of CHCI^ In the gas ja r was taken as the stage in the test for the determination of breakthrough volume, .e . the volume of a ir necessary to purge the adsorbed vapour through the cartrid g e. From the breakthrough volume of the a ir-so lu te mixture of known concentration and the weight of sorbent in the cartrid g e It was possible to calcu late the adsorptive capacity of Chromosorb 102 for CHCI^ under the given experimental conditions* From the data determined with respect to the adsorptive capacity of Chromosorb 102 for CHC1^ and the knowledge that a four hour sampling period at a flow rate of 2 5 0 ml/min would be required in the fie ld te s ts , ACfc C C 6 3 C 5 215 sampling ca rtrid g e s of the follow ing dimensions were constructed ; 11 mm i .d . x 11.5 cm long containing 6 g of CKromosorb 102. These ca rtrid g e s were used to determine desorption e f f ic ie n c y data. Qesorption efficiency q uantity CHCl^ from Chromosorb 102 CHC1, concentration Volume of a ir in a ir sampled sampled The same te st assembly as described p revio u sly was used to c o lle c t CHC1^ on the c a rtrid g e . The adsorbed CHCl^ was therm ally desorbed into an evacuated gas j a r . The adsorbent tube and the gas j a r were connected by Teflon tubing. The tube was heated to a fixed temperature by wrapping with heating tape co n tro lle d by a v a ria b le transform er (170C). A fter reaching maximum temperature, the stopcock of the gas j a r between the j a r and the tube was opened. Heating of the sample tube was continued fo r a fu rth e r 15 minutes, a ft e r which time the stopcock of the gas Ja r was clo sed , the gas j a r removed from the sample tube, taken outside the laboratory and allowed to f i l l up with the cle a n e st p o ssib le a i r . The contents of the gas j a r were then subjected to GC a n a ly s is . A liq uots {1 ml) were removed from the 'gas j a r using an a ir - t ig h t syring e and in jected into a gas chromatograph. Gas chrot-utography was conducted on a Varian 1200 s e r ie s chromatograph equipped w ith an electro n capture d etector. The GC parameters used are shown below. Co1umn Column temperature Injecto r temperature Oetector temperature Detector Flow rate Range and attenuation Chart speed1 - 2 m x 2 ftm SS containing Chromosorb 102 (80/100 mesh) - 180C - 215C - 215C - EC - ^ 3 1 ml/min - as required - as required The concentration of so lu te s present in the gas j a r * follow ing desorption was determined by measuring peak areas of the vapours and comparing them with those of prepared standards. The I n i t i a l concentration ACM CC63C6 216 in the Ted lar bag was known, and, from the volume of a i r drawn through the c a rt r id g e , the actual concentration contacting the ca rtrid g e was c a lc u la te d . The desorption e f f ic ie n c y was then ca lcu la te d . H. 6 R esu lts and D iscussion Experiments using toluene and xylene as absorbents showed that a consid erable loss of both solvent and ch lo rin ate d organic compounds occurred a f t e r a four hour t e s t . The solvent was found in a l l Impingers following the one in which the standard so lu tio n was placed . Estimated solvent lo sses were 40-60%. Losses of CCl^, CHCI^ and TCE ranged from 50 to 90%. With decane as the adsorbent, the solvent lo ss was le ss than 10%. However, the loss of ch lo rin a te d compounds was 60 to 100%. Oue to time r e s t r a in t s and the obvious problem of solvent or p o llu tan t lo ss a t the high gas flow r a te , t h is method of c o lle c t io n using absorbents was rejected . As discussed in Section H .4 .1 .2 , reten tio n of v o la t ile halocarbons by the adsorbent at high gas volume fjpw ra te s was poor and the break through rap id . The r e s u lts fo r the c o lle c t io n system f in a l l y selected (Section H .5) are summarized below. Adsorptive Capacity Component CHC13 ul/g (sorbent) 0.1. ^ Oesorption E ffic ie n c y Tedlar bag concentration ___________ o f CHC13___________ 1 .2 5 ppb Desorption ^Efficiency ______________ %______________ 90 96 104 5 -0 ppb a 92 no 1 0 .0 ppb 94 102 (The higher concentration samples required a second thermal desorption in order to remove a l l the so lu te from the sorbent. However, the f i r s t desorption was always b etter than 75$.) ACh CC03C7 r* 217 Two additional experiments were performed. They were: a) An a ir-s o lu te mixture of CCI^.and DCE (5.00 ppb le v e l; 40 L Tedlar bag) was run through the system using a back-up gas ja r in order to test whether these solutes were retained by the Chromosorb 102. Regular checks of the gas Jar using GC to analyze for the presence of CCl^ and/or DCE fa ile d to show any trace. Thermal desorption of the tube followed by GC an alysis gave good recoveries for CCl^ and DCE (>90%). b) An a ir - s o lu t e mixture of C H C ly CCl^ and DCE (2 .5 ppb le v e l; 40 i Tedlar bag) was prepared. This mixture was drawn through the test system which contained two additional traps prior to the Chromosorb 102 ca rtrid g e . One trap contained water (50 ml) and the other 5% NaOH solution (60 ml). These traps were included in the 0RF train, in order to remove residual Cl^ and/or HC1. Thermal desorption of the tube contents gave good recoveries of the solutes (>9 0 %). H.7 Conclusions The laboratory study undertaken showed that low molecular weight halocarbons cannot be e f f ic ie n t ly co llected by absorbents or adsorbents from a gas stream at high flow rates (n* 0 . 5 cfm). Tnere are numerous lite ra tu re references to the technique for co llectin g trace q uantities of component vapours by drawing a ir through cartridges containing sorbent media such as polymer beads, act Ivoted carbons or stationary liquid phases chemically bonded to so lid supports This report has demonstrated such a procedure for the co llectio n o* low molecular weight chlorinated hydrocarbons on Chromosorb 102. It has been shown to give q uantitative c o lle c tio n together with quantitative recovery of the trapped species using a thermal desorption technique. a The sampling tube technique appears to give better re su lts for sampling traces of low molecular weight organochlorine pollutants In a ir than impinger methods. ADM C C 1 3 C b * ' ' ' ' ' * ' * `!'* **'*''*' ,,r, M . 218 REFERENCES H .1 Marine Environmental Monitoring of Vuleanus Research Burn 1!, December 2-10, 197*1 P relim in ary rep o rt, U .S. Environmental P ro tectio n Agency, December 10, 197*t. H.2 P e l l i z a r r i , E . D . , Bunch, J . E . , and Carpenter, B. H. , Env, S c L Technology 9; 552 (1975). H.3 Jones, P. W. , Giammar, R . D . , Stru p , P . E . , and Stanford, T . B . , Env. Sc?, t Tech. 10 608 (1976). H.J* B e l l a r , T . A . , and Llchtenberg , J . J . , J . Am. Water Works A sso c .. 66 739 (197*0. H .5 P e l l i z a r r i , E . D . , EPA-600/2-75 076 Nov. 1975. 3 i i 7 .r \ ACP- C C 8 3 C S APPENDIX I GC/HS/COMPUTER DETERMINATION OF CHLORINATED HYDROCARBONS AND PlCB'S .t Chemistry Division A ir Pollution Control Directorate Environmental Protection Service Environment Canada Submitted by Dr. R.C. Lao V Ii II 221 APPENDIX 1 GC/MS/COMPUTER DETERMINATION OF CHLORINATED HYDROCARBONS AND PCB's In February 1970, 30 extracted samples were received from Dr. G. Thomas of the O ntario Research Foundation. These samples were taken from the burning of ch lo rin a te d hydrocarbon wastes in a cement k i l n. Requests were made to determine the nature and q uantity o f any heavy chlorinated hydrocarbons, in p a rtic u la r PCB's, present in the samples. Experiments were completed and the procedures are b r ie f ly described as fo llo w s: M aterial : A ll apparatus and reagents used tn the experiment were examined by gas chromatograph-flame io n iza tio n d etector (GC-FID) a n a ly s is of a pure hexane e x tra ct to ensure freedom from organochlorine residue contamination. The ch lo rin ated isomers (PCB's) and Aroclor^mixtures were obtained from Ana l abs, North Haven, Conn., U. S. A. GC-FID: A Perkin-Elm er 990 model GC-FID with a datasystem PGP-1 was used. The procedures fo r standard c a lib r a tio n s were given in a previous paper (to be published as a chapter in the book "Advances In Dynamic Mass Spectrometry11 1976). Chromatograph operating parameters were as fo llo w s: Column 12' x 1/8" O.D. Column packing Column temperature In jectio n temperature Manifold temperature C a rrie r gas stainless steel Apiezon L on chromosorb W 80/100 mesh I n i t i a l 100C programmed at 4C/mIn. to f in a l temp. 200C and hold 250C 250C Helium 40 ml/min. CC/HS/Computer: A Flnnigan 1015 D GC/MS system was used with a data system 6000 s e r ie s . I t s performance has been studied and documented tn the same paper as mentioned above. It a ls o Includes standard PCB mass spectra and computed reconstructed A roclo r Chromatograms. The instrumental data are as follow s: ACM CC6 3 1 1 222 r 'N Finntqan 1015P GC-MS Instrumental Data A Instrumental Data GLC Co Iumn 12' x " a l l g la ss Column packing 6% Apiezon L 80/100 mesh chromosorb W - Column temperature 225C In jectio n temperature 2S0C C a rrie r gas helium 10 ml/min. Sample s i z e 5 to 7 mi c ro i 1tr s per in je c tio n B Instrumental Data Ms Filament current TOO microamperes Electron energy 70 or 20 eV Operating pressure 6.68 x l O 3 N/m^ (5 x l< f6 to rr) Scanning speed A seconds Standard deviation of , spectra maximum 570 A fter the quadruple MS operating parameters were adjusted the sample was in je cte d Into the GC. The Ion abundance chromatogram of the GC e fflu e n t was acquired by scanning the mass range (40 to 400). The dialogue required fo r mass spectrometer c o n tro l, data a c q u isitio n and obtaining the p lo t are given by softw are programs. At the end of the GC run the computer p lo ts a reconstructed gas chromatogram (Ion abundance chromatogram) of to ta l Ion amplitude versus the spectrum number. Id e n tif Ica t Ion of these chromatographic peaks can be accompfished by p lo ttin g the mass spectrum of a sp e c ifie d peak or by a lim ited mass \ pl ot chromatogram which Is obtained under computer control by searching through the c o lle c te d sp ectra and id e n tify in g spectra containing ions with a s p e c i f i c m/ va lu e. ADM CCe3i* i *- ' 223 Resul t and D iscussion (All chromatograms and computer reconstructed chromatograms are kept on f i l e a t the A1r P o llu tio n Technology C entre, Department of .the Environment, "Ottawa, Canada.) Computer reconstructed gas chromatograms of A ro clo r 1242, 1254, and 1260 were made. By focusing on s p e c if ic mass/charge r a t io (m/e) peaks such as 2<)0 (tetra~ ch lo ro b ip h e n y ls), 324 (penta-chloro) or 358 (h e x a -ch lo ro ), an ion cu rren t p lo t for a p a r t ic u la r PCB Isomer versus the A roclor spectrum is obtained. Gas chromatograms on Perkin Elmer 990 GC were done fo r some blank so lu tio n s and concentrated samples ( a ll samples were concentrated from about 25 ml to 1 ml ) . No ch lo rin ated hydrocarbons nor PCB*5 were found in the samples, i f they are present the concentrations are below the s e n s i t i v i t y l i mi t of GC-F1D o f 1 ng or le s s . Hass sp ectra fo r the GC peaks of spectrum numbers 365 end 487 from the reconstructed chromatogram p lo t of sample (WBC-T3, EPA t r a in , Impinger) reveals th at only hydrocarbons are present In the sample. No evidence for ch lo rin a te d compounds was found. A l i mi t mass searching technique has fa ile d to give any In d icatio n of the presence of PCB's. The computer lib r a r y search of standard A roclo rs did not match those obtained from the sample. It was concluded, th e re fo re , that there were no heavy ch lo rin ated hydrocarbons or PCB's in the samples. I f they were present the weights were l ess than 1 ng. v \ ACh 006313 Or. Z. 1. Wallen Environmental Protection Agency Office of Toxic Substances 401 H Street 8.W. Washington, DC 204C0 Dear Dr. Wallen: Monsanto has a concern over the validity of the perchlorlnation technique used by SPA and others to Measure and/or confirm polychlorinated biphenyls in environmental materials. Xn our investigation of the perchlorination method, we have found that such chemicals as biphenyl, alkylated biphenyl, and many other substituted biphenyls, interfere with the perchlorination technique. Zt also appears that various petroleum oemponents may interfere. Zf these chemicals were present in environmental materials that were .being tested for PGB's using the perchlorination method, erroneously high PCB concentrations would be reported. A recent article in the Journal of the AOAC (Yol.59, Wo.3, 1975) points out two other limitations of the perchlorination procedure: 1. High and variable reagent blanks, which cause erroneously high findings. 2. Formation of bromononaChlorobiphenyl, which causes low recoveries. A copy of the article i^\attached. ACh 008315 ) Dr. X. B. Wallen -2- March 15, 1976 -/ Since these limitation* can laad to significant errors in dataraining trace levels of PCB's in environsiBntal samples, va suggest that EPA carefully review the validity of the perchlorlnation technique. Results that have been obtained using this technique nay not be valid. Xf any other information is needed, please let us know. Sincerely, J. Coleman Weber Manager, Product Acceptability aah cc: Dr. A. C. Trakovski Environmental Protection Agency bcc: W. C. Hammann - T3A R. E. Keller - TIB J. P. Mieure D. Wood - B2SC K. W. Easley - Washington 1920 R. G. Kaley - T2P vV AO* CC6316 i Ai DECS 1575 UNITED. STATES ENVIRONMENTAL PROTECTION AGENCY WASHINGTON. O.C. 20460 O FTIC E OP W ATER ANO H A ZA R D O U S M A TE R IA LS December 1, 1975 Hr. V.B. Papageorge Manager, Product Acceptability Specialty & Process Chemicals Monsanto Industrial Chemicals Co. 800 N. Lindbergh Boulevard St. Louis, Missouri 63166 Dear Mr. Papageorge; Pursuant to your request attached Is a copy of the paper "Environmental Levels of PCB's." Also attached, Is a contractor's report describing FCB levels In soil samples taken from the area surrounding your facility In St. Louis. If I can be of any further help, please contact me on 202-755-6956. Sincerely yours, Vincent J. DeCarlo, Chief Monitoring & Information Systems Branch Office of Toxic Substances (WH-557) v* ADM C C 3 17 DtVIRONMEHTAL LEVELS OF PCB' Doris J. Xuopp V lacont J. DuCarlo U.S. Environmental l'rotrctloa Agency Washington, D.C. 20460 ABSTRACT Sines the 1966 discovery ' in Svadsn that chlorinated biphenyls ware widely dispersed in the environment their envlronarntnl levels have bean* the subject `qf many studies with the results indice tint tliet PtB'*- ceo have adverse ecological and. toxicological effects* However, a well planned national approach ro environmental sampling has not bean attempted thus hindering oaklng a national assessment of the FCB proble*a. This paper reviews the levels that are currently Using found in. the environnent. The data from the satlonal surveys show that s great deal of effort has been expended ar.d continues to be expended but that the date ere very limited. On,examination, ell States are believed to shew some level of PCB. contamination. In Table 2 States which had significant levels in at least or.e nodlum ere H e t c d with their reported TCR levels for surface and ground water, bottom sedinents and fish. In Tabic 3, a nuitber of localized studies are listed along*with Che reported environmental data. In all these studies ' the most extensive treesuremsnss have been la fish. TABLE 1 introduction Since the 1966 discovery In Sweden that chlorinated biphenyls verc widely dispersed In the environment, thtlr levels have bean the subject of any studies both abroad and in the United Staten. Extensive laboratory and environmental a.>aaurevents Indieete that ? C5's can U.-'ve adverse ecological and toxlcologiesl rffects, art very resistant to environmental degradation end are being discljrgad by any different aeurces. This paper will focue on the currant FCB data base in an effort to assess the TCB levels in the environment on e notional beale* ENVIRONMENTAL TCtt SOURCES Before the TCB data base it examined, it would be of interest to review how these chemicals get into the envlrersent. The major sources 'contributing to tha environment:.') 1 levels of PCB's sra Industrial tf fluents, manufacturing processes, consumer end Industrial waste materials, sewage treatment facilities and accidental spills. Lacking extensive air and soil data from these sources, available water data arc shown In Table 1. The 5 major FCB spills In 1973-75 all Involved transformers. In 4 of thocc Incidents, the PCB's verc spilled on soil while In the fifth the transformer was dropped on a pier and tha contents were spilled Into the water, In the latter incident, 2B3 gallons were spilled end It wee eetlmated that only 70-90 gallons ware recovered. In ell cases the material that could be recovered wee drurced up end entombed at a cost approaching <2.3 million. 1 EVIEV or ITT. PCB DATA BASE To eesoes the eeope end extent of PCS'a In the environment, all national survey* and national date bases aintained by eh* U.S. Environmental Protection Agency (ETA) were examined. Including S7DRET, the national water quality storage and retrieval system, the National Sells Monitoring Program for Pesticide Residues and the Human Monitoring 'Program. In addition, dace from special dati bases .maintained by the United Statro'Urological 5urvcy and the open literature were else examined along with available unpublished reports. Using these date, maps were conscructed showing the concentrations of PCS'* la urban soils end the aquatic environment. ENVIRONMENTAL SOURCES OF PCB's SOURCE EFFLUENT WATER CON CENTRATIONS (twM PAPER KILLS . Wisconsin1 WASTE PAPER HILLS Wisconsin-1 INDUSTRIAL EFFLUENTS 0.1 - 10.5 18.5 California* Wisconsin2 Ohio2 , Michigan3 MUNICIPAL VASTE VATER TREATMENT PLANTS _ Michigan3 Wisconsin1 Ohio2 California2 CAPACITOR AMD TRANSFORMER FACILITIES . New York , Massachusetts3 0 . - 76 .04 - 0.25 . 0-17 .1 - 7000 0.5 .05 - .10 0.16 - 17.0 17 76 2800 42.5 TCB MANUFACTURING FACILITY East St. Louis3 SPILLS , .87 Variable 1. Kleinert, S. J . ,'Environmental Status of PCB in Wisconsin, May 8, 1975, Wisconsin Department of Nstor.il Resources. 2. PCB's and the Environmental, Report of the Intsrdoparruental Task Force on PCB's, National Technical Information Service, 1972. 3. Sifttenant of C o n e c m s of the Lake Michigan Toxle Substances Committee Related to Poly chlorinated Biphenyls, June 1975. Prepared by Kerl F. Bremer, USEFA, Chicago, 111. 4. Unpublished Data - Roy.il J. Nadeau and Robert P. Uavla, Investigation of Polychlorinated llphcnyls in tha Hudson River, Hudson FallsFt. Edward Area, August 1974. 5. Field Sampling and Analysis of Toxic Pollutants Interim Report, Battalia, Pacific Northwest Laboratories, August 1974. ACH C C 6 3 1 b WATT* On the bail of tho national aquatic environmental data collected In 1971-1972 and in 1974, a continuing,widespread acehmulatlon of P CS' In water, aedleant and fish appears to be occurring, forever, no trend analysis la possible with tho available measurements, For- example, whole water measurements have been taken throughout n o s t . of the country but thoee atatea reporting non zero readings are few In relation to the number of s t a t u showing zero concentrations. This is due both to the low eolublllty of VCB'7 and to the usual analytical procedura that U n i t s detectability to the 0.1 ppb level. Kora meaningful water concentrations were obtained In the Lake Ontario* and Orange County, California* studies where concentration* in water at the ppc (ng/1) level varc measured. At theae levels, changes In FCB concentrations could be found with distance and d e c and eould be related to other measurable parameters such ss PCS concentrations In scdlcent, flora and fauna. sediments Since PCB's are relatively Insoluble, It la not surprising that bottom deposits have shown significant concentrations. Although 30 states collected samples In the 1974 study, IS had less chan 4 ststlons reporting and showed zero readings. Of the remaining 15, with at least 4 reporting tationa, 13 shoved dateetablc concentrations. Slnee the 13 states involved were not necessarily the ease In both studies, only a very broad comparison nay be made, l.e., five states had lower concentrations, four states were higher and four did not sample again.* Ve say conclude, then, that in theae states with any monitoring rffort In the 1374 study, PCB's are present in bottom deposits end the levels aie not any lover than in the 1971-72 study. Considering, the persistence of PCB's, a significant proportion of the nation's waters are now affected and will continue to be. directions. FCB`s were detected in t0i* tc botli sampling locations. Concentrations ranged from the detection limit of 0.001 ppn co ove 20 ppm Tne distribution of all PCB's *nalyd appears higher near the plant elte and generally decreases -with distance from the site.4 Details of the sampling sites and the concentration levels measured arc shown In Figure 1. TABLE 2 SELECTED STATE DATA OH ENVIRONMENTAL LEVELS OF PCB's '. State Surface and Cround W a t e r 1, Mg/1 Bottom Sediment1 bg/kg Fish* ppm AL AR CA CO CT FL CA IL IA KD MA MI ' MH MS NB NJ NY OH 0.1 0.3 0.1-0.2 0.1-2.1 0.1 0.2 0.1-0.3 0.1 0.1-6.0 20-2,600 20-190 5-350* 5-3,200 10-1,300 10-1,200 1.53-5.48 1.69-3.88 2.16-5.36 0.10-1.25 0.52-1.18 1.21-11.3 0.35-1.41 0.56-1.31 50-170 3-800* 3-13,000* 4.00-11.7 0.44-1.09 0.47-4.58 0.10-4.00 2.68-9.50 1.73-8.07 OR 1-5-140 0.71-3.62 SOILS FA 0.2 6-700* 1.94-2.68 The Rational Soils Monitoring Progrim Is a small earpling effort studying only 5 urban areas each year. However, TCB's were detected In three of the five cities sampled In eaeh of the years for which data are available, 1971-73. Of the 22 positive readings 17 of them were below 1 ppm. Of the cities sampled in 1973, Pittsfield, Mass., was of particular interest because It has a large transformer end condenser plant using large quantities of' PCB's. Six different sampling sices vlthln one mile of the plant shew no detectable FCB* residues. In contrast, the soil surrounding another facility In Illinois using- FCS'a in the manufacture of Investment casting vsxcs was recently measured for FCB content. Samples taken in an area ono-qua near mile radius around the plant ranged i value from 0.77-5.2 ppm. The PCB's Identified were mixtures of Aroclor 1260 and daeAehlorcblpheuyl. Levels up to 1.6 ppm Aroclor 1260 were found at 1-1/6 miles from the plant. Saaplee were collected .from this facility as ypart of a study being conducted for the Office of Tbxlc Substances at sites < suspected to have PCB concent re d one. Other cites samples'la Kerch of 1975. ware In the vlclnliy of on lnvcstnrnt casting company in Michigan aqd the FCB manufacturer In Illinois. Surface sell samples were collected at each site up to a distance of approximately one mile from the plant boundary# at 1/4 mile Intervale In all PR SC TX UT VA wv . VI 0.1 0.1-3.0 i 30-200 7.9-290 N 0.1 5-80 . 10 0.10-7.3 0.10- .22 0.15-2.14 0.31_1.20 1.24-14.8 All data ln this ieolumn taken from Hans J. Cruap-Wiesner, liftman K. Fclrz and Marvin L. Yates, A Study of the Distribution of Poly chlorinated Biphenyls in the Aquatic Environ ment, Jour. Research ,U.S. Ccol. Survey 1_, 603 (1972, unless otherwise noted. 2. USCS Sediment dara, 1974. 3. A H data In this column'taken from Croswcll Henderson, Anthony logli and Wendell L. Johnson, Organoehlorlne Insecticide Residues In -Fish Fall 1969 National Peetlelda Monitoring Program, Pesticide Monitoring Journal Y , 1(1971). O 2 ACM CC31S TABLE 3 SELECTED PCS STUDIES lakes U b O ntario^ 2 Lake I r l* 3. * Lake Superior* lake Huron* U k a Michigan*. $ Cayuga Lake10 U k a St. Clair**. 1IVERS Fieh (ppo) 0.14 0.2 9.17 7.B 0.3 3.6 0.16 0.1 3.4 tr 11.0 163.3 9.7 * 3.1 Water (uc/1) 0.033 - 0.097 Sediment (ve/kn) A i - 243 / 0,1 - 0.2 3.72 - 636 Plankton teP"> 3.4 - U . B Mississippi River** 0.9 " 5.5 Taaaa Brazos River w Utah Green River*? Nev York Hudson 13 Michigan Saginaw River*1* Kalamazoo Hiver?*** Portage Creek? Wisconsin Milwaukee * 7 330 6.9 - 165.3 <0.01 164.56 MARINE EKVISOSaKT Atlantic Ocan*/ 1* ay of Fundy*^ Gulf of Mcxieo/Cnrib- bean Sea?0* 21 Southern California?? 0.038 0.07 0.032 . 0.04 0.190 1.54 0.059 6.6 A Fred Lee and William C. Doyle, Chlorinated Hydrocarbons in the Lake Ontario Eeoystcn, EVA-660/3-75-002, . June 1975, U.S. E.F.A., Corvallis, Oregon. 2 Klaus L. Kaiser, Nircx, An t'nrccognircd Contznl- nant of Fishes from Lake Ontario, Science 18, 323 (1974). 3. Richard L. Carr, Charles E. Fir.stewcidtr and Michael J. Sehibl, Pesticides Monitorlug Jour nal 6:23 (1972). 4. John R. >1. Xelso and Rieharo Frank, Orpanochlorina Residues, Mercury, Ccppr.r and Cadmium in Yellow Perch, White Bass and Smallmouth Bass, Long Point Bay, Lake Erie, Trans. Aaer. Fish. Soc., 103:577 (1974). , 3. Unpublished data. Gilman D. Vcith and C. E. Class, PCB's and DDT In Fish from Eastern Lake Superior, . U.S. E.P.A., Duluth, Minnesota. 6 Unpublished Date - John L. Hesse, Bureau of Water Menagemenc, Michigan Departcent of Natural Re sources, June 1973. 7. Unpublished data - State of Michigan Water Re- sources Cosslsslon, Polychlorinated Biphenyl Sur vey of the Kalamazoo River and Portage Creek in the Vicinity of the City of Kalamazoo, 1972. S. Gilman D. Vcith end Fred G. Lee, Chlorohlpheuyls In the Milwaukee River, Water Research, 1971. .9. Unpublished data - Schncht 1974, EPA 600/3-74-002. 10 Carl A. Bachc, James W. Scrum, William D. Youngs and Donald J. Link, Polychlorinated Biphenyl. Residues: Accumulation in Cavuga Lake Trout . with Age, Science 177:1191 (1972). 11 Lauren G. Johnson and Robert L. Harris, Chlorin- stsil Insecticide Residues in the Eggs of Some Freshwater Fish, Bulletin of Environmental Con tamination end Toxicology 11:503 (1974). \ 12. Jean.A. Schulzo, Douglas B.,tlanlgold end Freemen L. Andrews, Pesticides In 'Selected Western Streams 1968-1971, rcstlcldus Monitoring Jour nal 7:73 (1973). 0.1 - 0.3 0.1 1.0 - 2800 6.6 6,700,000 0.1 0.1 0.1 - 0.21 - 0.45 - 0.48 100 - 560 10 - 475,400 .02 - 2.17 0.0009 - 0.0036 . - 0.0015 - 0..Q19 0.157 - 1.055 * 13. Unpublished data - Royal J. Nadeau end Robert P. Davie, Investigation of Polychlorinated Bi phenyls in the Hudson River (Hudsoa Falls - Ft. Edvard Area, August 1974. 14. Unpublished dar - John !.. Hesse, Monitoring for Polychlorinated Biphenyls in the Aquatic Envi ronment, Kay 1973. 13. Unpublished data - Michigan Water Resources Com mission, Evaluation of cha Aquatic Environnent, of the Kalamazoo River Watershed, Kay 1972. 16. Michigan Department of Agriculture, Bureau of Consumer Protection, 1973 Great Lakes Environ mental Contaminants Study. 17. T. F. Bidltccn and C. E. Olnsy, Chlorinated Hyrdocarbons in the Sargasso Sea Atmosphere and Sur face Water, Science 163:516, 1973. IB George R. llarvey, Helen P. Miklas, Vaughan T. Boven end William C. Steinhauor, Observations on the Distribution of Chlorinated Hydrocarbons in Atlantic Ocean Organisms, Journal of Marine Re s e a r c h 32:103 (1973). 19. V. Zltko, 0. Hitlinger and P.M.K. Choi, Contamin ation of the Bay of Fundy - Culf of Maine Area with Polychlorinated Biphenyls, Polychlorinated Ter phenyls, Chlorinated Dibenzodioxlr.s and Dibenzofurans, Environmental Health Perspectives, 1:47 (1972). 20. C. S. Glam, A. R. Hanks, R. !.. Richardson, W. M. Sackctt and M. K. Wong, DDT, DDE, and Pclychloclnarcd Biphenyls in Biota from the Culf of Mex ico end Caribbean Sea - 1971. Pesticides Moni toring Journal 6:139 (1972). 21. C. 5. Claa, M. K. Wong, A. R. Hanks, . K. Sackecc end R. L. Richardson, Chlorinated Hydrocarbons In Plankton from the Culf of Mexico and Northern California, Bulletin of Environmental Contam ination end Toxicology 9:376 (1973). 22. The Ecology of the Southern California Bight: Implications for Water Quality Management, Southern California Coastal Water Research Pro ject, 1500 East Imperial Highway, El Segundo, 1973. 3 ACK CCtidC Figure 1 Concentracin of FCC'* ln Soil with Diacanca (PP) Aroclor 1260 Invasecene Casting Corpany Investeent Casting Wax Manufacturer Aroclor 1260' FCB Manufacturar doeadiiorebiphenyl Investment Casting Wax Manufacturer 4 AC* CCfii PCI measurement! In air lent with the related transport studies era practically nonaxlstant. Karvey and Stalnhautr* have reported lavals ranging from 3.9 -* 3.3 ng/m3. Air measurements by Bldlenan and Olnay for Abode Ialand ranged froa 2.1-9.4 ng/n3 and iron 0.21 - 0.65 ng/n3 at Bermuda.8 Sq o w _oelt voter data froa Wisconsin ranged froa aero to' 0T24 ppb. The suggestion has been Bade that atmospheric fallout may be the scat significant aourec of PCB discharge to the vatara of the atate of Wisconsin. 7 ns Flah have been studied on a nationwide baela by the Departnent of the Interior since 1967. In the 1969 study, PCD levels were Identified In flah froa 35 states. 8 Data from the nationwide sampling programs, 1970-1973, are currently being prepared but were not available for this raport. Consequently 1 nationwide flah data are not available ony core currently than 1969, with tha exception of the Isolated studies listed In Table 3. The 1969 national study showed PCB levels generally within the Fl\\ 5 pp** guideline. In the Great Lakes area, PCB levels in flah fron Lake Michigan were so high, 7.6-10.9 ppn, that FDA seized shipments of coho saloon In Kay 1975.* Although transport mechanisms are not veil known and would vary through different ecosystems. It Is Interesting to consider tha Lake Ontario Ecosystem Study *urjvi:ud I11 Tab!* 6. From this study on estimate of bioaagnifleatlon is possible giving a sediment to fish ratio of 1:120. The FDA guideline of 5 ppn in the edible portion of fish corresponds to a sediment concentration of 61 ppb - a figure exceeded In all 13 acates reporting through uses in 1976. HUMAN'S With FCIt'i as widespread across various media es they are, It la expected that levels would be Identified In humans as veil. . Tha Human Monitoring Program, In 1972, found that 3035 out of 4102 aamplea from 31 Stater, showed levels ranging fro* lea* than 1 ppn to more than 3 ppm. In 1973, 966 out of 1277 samples from 23 Statra again showed levels In the same range. In both years approximately 75Z of tha, adipose tissue analysed contained eome polychlorinated biphenyl a. Unfortunately wc do not know age, occupation or residential histories of the cosea involved so It la difficult to trace the levels back to the potential sourcts. CONCLUSIONS Although thcro are many national and state groups collecting environmental PCI data, limitations in the current data base prevent ona from making a uniform national assessment ofv^ PCB environmental levels. It appears that if a. well .planned national approach - sampling vare attempted, the data base could be Improved In a very short period of time. LAXE ONTARIO ECOSYSTEM* FISH, lfg/g Alcwlfa Smelt Slimy SculpIn .14 - 4.36 1.40 - 3.49 1.58 - 9.17 HATER, ng/1 38 ' - 97 SEDIMENT, ng/g 43 - 245 VET PLANKTON, pg/g 3.4 - 11.8 X. Halle, C. L., Velth, C. D., Lee ' ---- * * --'l ij v e v . h v v ( i o a* tha Lake Ontario Ecosystem (IFiCL) June, 1975. INFERENCES 1. Haile^ C.L., Velth, G.D., Lee, G.F., Boyle, V.C., Chlorinated Hydrocarbons In the Lako Ontario' Ecoayetea (1FYGL), June 1975. 2. McDermott, D.J., and Hansen, T.C., Inputs of D D T , 1 PCB and Trace Metals from Harbors, Coastal Water Research Project Annual. Report, 1975. 3. U.S. Ceological Survey. 6. Environmental Polychlorinated Biphenyl Contamination near Sites of Manufacture and Use, Environmental Science and Engineering, Inc., 1975. 5. Harvey, C.R. and Stelnliauer, W.G., Atmospheric Transport of Folyehlorinstcd Biphenyls to the North Atlantic, Atmospheric Environment:, Vol. 8, 1974. 6. Bldlcman, T.F., and Olney, C.F., Chlorinated Hydro carbons In the Sargasso Sea Atmosphere and Surfeea Water. Science, Vol. 1S3, February 1974. 7. Klelnert, S.J., Environmental Status of PCB'fi in Wisconsin, >1ny 8, 1975, Wisconsin Department of Natural Resources. 8. Henderson, C., Inglls, A., Johnson, V.L., Organoehlorlnc Insecticide Residues in Fish - Fall 1969 National Pesticide Monitoring Progrun, Pesticides Monitoring Journal, Vol. 5, No. 1, June 1971. 9* lremer, Ibirl E., Draft-copy of "Statement of Con cerns of tha Lake Michigan Toxic Substances Com mittee related to PCD*a," June 1975. 5 ACH CC8322 Data from V ic in it y of Monsanto Co., Sauget, IL. A ro clo rs have been produced by Monsanto Co. at the Sauget, I l l i n o i s s it e since p rio r to 1957. PCB mixtures ranging from 16 percent chlorin e to 68 percent chlorin e have been produced over the years. Production of the higher percent ch lo rin e m ixtures has been v o lu n t a r ily reduced since 1970 and lower su b stitu te d isomers have been introduced. Polychlorinated terphenyls and chlorinated benzenes have a lso been produced at t h is fa c ilit y . It is reported (Papageorge, 1975) that terphenyl production was suspended in 1971. A ty p ica l chromatogram o f a s o il .sample taken in the v ic in it y o f the \ i* Monsanto f a c i l i t y i s shown in Figure 8 along with the chromatogram of *\ several other PCBs run under the same instrument cond itio ns. This p a rtic u la r s o il sample contains 11 ppm A ro c lo r 1242, 9.3 ppm Aroclor 1260 and 1.0 ppm decachlorobiphenyl. Lig h te r chlorin ated A roclor mixtures are l i k e l y present as w e ll. The instrument c o n d itio n s employed, however, do not give adequate separation of these m ixtures to permit accurate q u a n tific a tio n , The presence o f decachlorobiphenyl at the concentrations observed is in te re stin g . Decachlorobiphenyl has been reported as a compo. nent o f only one A ro c lo r m ixture, A ro c lo r 1268 (Hutzinger, Safe and Z itko , 1974). Whether i t may'be a component o f any o f the terphenyl mixtures or biphenyl/terphenyl m ixtures or a by-product i s not known. ACH C L c . 3 The measured s o il concentration o f A ro clo r 1242, A ro c lo r 1260 and deca d ilo robf phenyl in the v ic in it y of the Monsanto f a c i l i t y are depicted In Fig u re s ?, TO and 11, re sp e ctive ly. Figure 12 shows the long-term Average wind d ire c tio n in the area. The d is t r ib u t io n o f a ll PCBs analyzed appears to be higher near the plant s it e and generally decreasing with distance .from the s it e . Furthermore, there i s evidence that generally higher concentrations are present in the s o i l s located to the southeast. T h is corresponds to the predominant wind d ire c tio n and may suggest an airborne tran sport of the PCBs from the f a c i l i t y . One 15 cm deep core sample was taken 1/2 mile southwest of the Monsanto s it e . A n a ly sis of the bottom 11 cm of t h is s o il yield ed PCB concentrations b a s ic a lly equivalent to the top 4 cm. A 24-hour composite sample from the in flu e n t and the e fflu e n t o f the Sauget v illa g e sewage treatment plant taken beginning 0800 on February 17, 1975, was provided by p lant personnel. The Monsanto Co. j o in t ly owns t h is f a c i li t y . A n a ly sis o f these samples were as fo llo w s: *t . A ro c lo r 1242 A roclor 1260 Decachlorobiphenyl. influent 6.8 pg/1 3.7 pg/1 ^ <0.01 pg/1 effluent 4.5 yg/1 4.7 ug/1 <0.01 pg/1. The treatment plant e ffe c ts a reduction of 34 percent A roclor 1242 and 46 percent A roclor 1260. No decachlorobiphenyl was detected. ADM C08325 AC* CC36 i r AC* GO0327 4 1 mi. Figure 11. Concentration o f decachlorobiphenyl In s o il as a function of distance from Monsanto Mfg. Co. Concentrations are expressed 1n ppm; AC* CCfciiit 'F ig u re 12. Ten Year Averaged Hind Rose fo r St. Lo u is, M isso u ri, January through March, 1951-1960. V Scale : l \ . 55 Calm Winds 2.05 CM C 0 6 3 2 S bcc C-. .R. Wllmore - NEMA Tyler Nourse - EIA bbcc D. Hosmer A . Leisy D. Wood February 18, 1976 Mr. John P. Lehman Director Hazardous Waste Management Division U.S. Environmental Protection Agency Washington# D.C. 20460 Dear Mr. Lehman: Enclosed are our comments and suggestions to your draft "Recom mended Procedures for Disposal of PCB Containing Wastes", m We have made several changes on pages 1 2 5, 7 and have added Appendix B - Disposal Services and ohanged your Appendix B to C with a minor correction. I would appreciate receiving a copy of your revised guidelines. 1 If we can be of any help# please let me know. My phone number is (314) 694-2027. Also enclosed for your information Is a Monsanto summary on FCBs Sincerely# CB oc K. W, Easley J. Coleman Weber Manager# Product Acceptability CC633C Recommended Procedures for the Disposal of FCB-Contairir-.Wastes (Industrial Facilities) On December 22, 1975, the Administrator of the Environmental Protection Agency announced an action plan of regulatory and administrative actions to help ensure thatpolychlorinated biphenyls currently in service do not enter the environment. One portion of that program includes investigation and recommendation of appropriate disposal procedures for industrial, users of this family of chemicals. The Agency is also evaluating the consumer waste disposal problem and may issue further guidance on that subject at a later time. General Background Polychlorinated biphenyls (PC3s) are chlorinated aromatic organic compounds which give rise to concern because of their toxicity coupled with their pervasiveness and persistence in the environment and their tendency to accumulate in food chains. It is this bioaccumulative characteristic which manifests Itself by possible adverse effects on animals, including man, that % causes the present and continuing concern over the disposition of PCBs to the environment. FC3s have been used both in "closed" systems, or sealed as dielectric fluids in transformers and capacitors, and in "open" applications, where the PCBs are used in and consumed with the product. Uses of PCBs in "open" applications included impregnation of \ cotton and asbestos for braided insulation of electrical wiring; "* a plasticizer in wire and cable coatings; in plasticizers of AC* GC6331 2 vinyl chloride polymer films; in high-pressure hydraulic fluids; specialized lubricants and gasket sealers; heat trans fer agents; and machine tool cutting oils. Former open appli cations also include: formulation into some epoxy paints; protective coatings for wood, metal, and concrete; adhesives; and in carbonless reproducing paper. Today PC3s are supplied by Monsanto only to the electrical power distribution industry where they are marketed as dielectric fluids in "closed" or sealed systems such as transformers and capacitors. The sole producer of PCBs in the United States is the Monsanto Company, under the trade name of Aroclor. PC3s have also been marketed and used by manufacturers In their products under various trade names. Appendix A .provides a listing of the trade names used by domestic and foreign manufacturers. Cumu lative sales in the U.S. since the introduction of PCBs are estimated to be 1.5 billion pounds. In 1970 Monsanto volun tarily restricted sales only to the electrical power distribu tion industry where they are marketed as dielectric fluids in "closed" or sealed systems such as transformers and capacitors. t As a result, current production is down more than 50% from the peak years before 1970, but still amounts to about 30 mil lion pounds per year. Monsanto has introduced Aroclor 1015 which is reported'to have lower toxicity and higher biodegra dation rates. Most capacitors AC** c c t 3 d 3 (using PCBs) produced in the past two years use Aroclor 1016. Almost all of the PCBs ever produced still present a seifious threat to the environment because of their toxicity, bioaccumulative characteristic in the food chain and slow degradation. An estimated 750 million pounds of PC3s produced a*e still in use; 300 million pounds are estimated to be present in landfills. It is estimated that less than half the PCBs ever produced have entered the environment; of this amount, only about 10% (50 million pounds) are estimated to have degraded or been incinerated. The recommendations of this document are directed only at the PCBs presently in service (or which will enter service as a result of future production). Purpose Pursuant to Section 204 (b)(1) of the Solid Waste Disposal Act, as amended, the Administrator may issue advisory guidance and recommendations related to the disposal of waste materials. These recommended procedures do not have standard-setting or regulatory status, but-.represent the best information available on preferable disposal optionsfor PCB-contaminated waste materials. In addition, the Agency plans to conduct studies to determine other options for treatment apd disposal of PCB-containing wastes (including incineration of capacitors containing PCB liquids) during calendar 1976. The Agency is also aware that the American National Standards Institute is in the process of revising its guideline for disposal of askarels containing PCBs ACM CCbli 3 S' A (ANSI C107-. 1-1974). As significant new information becomes available from these and other sources, this Federal Register issuance will be revised. The purpose of this issuance is to provide guidance for the disposal of PCB-containing wastes. This guidance is addressed primarily to industrial users of PCBs, particularly those manufacturing and/or using capacitors and transformers containing PCBs, although the recommended procedures are appropriate to all PCB-containing wastes. Technical Background Polychlorinated biphenyls are, a class of synthetic compounds which have no known counterpart in the natural environment. PCBs are manufactured by the chlorination of biphenyl with anhydrous chlorine using iron filings or ferric chloride as a catalyst. The three most important physical properties of PCBs are low vapor pressures, low water solubility, and high dielectric constants.. They are miscible with most organic solvents. The chemical properties that make PCBs desirable industrial materials are their excellent thermal stability, their strong resistance to both acidic and basic hydrolysis, and their , general inertness. They are quite resistant to oxidation. In the process of replacing hydrogen atoms with chlorine atoms, a large number of substitution combinations arise. A C * CC633<t \ Theoretically, there are 210 possible chlorinated biphenyl compounds but only --bout 100 are likely to occur in corr-ercis I products. In commercial preparation, PC3 compounds contain from 12 to 68 percent chlorine;' the typical product is a mix ture of sovera1 compounds. Monsanto, the sole U.S. producer of PCBs, has adopted a four-digit designation for its Aroclors; the last two digits represent the approximate chlorine content by percentage weight. The first two digits indicate the type of material: biphenyl, triphenyl, or mixture of the two. The higher chlorinated biphenyls are non-flammable and have extremely m low volatilities. Unfortunately, some of the characteristics (stability, non-degradability) which make PCBs so valuable in industrial applications also make them highly persistent In the environment. The chemical stability, low volatility, high dielectric constant, and compatibility with other chlorinated hydrocarbons have resulted in many and varied industrial applications for the PCBs. The solubility of PCBs in nonpolar solvents ex plains why they are readily absorbed into fatty tissue and into the liver. Their resistance to oxidation or other types of chemical degradation explains their persistence and accumulation in animal tissue. The latter effect Is enhanced both by their insolubility in water and solubility In organic solvents. Their chemical inertness and resistance to \ metabolism account for their low acute toxicity. But as they AtK CC0325 \ 6 slowly build .up in a living system,-their concentration approach^ toxicity levels, at which point chronic effects make themselves evident- The low solubility of ?C2s in water probably limits the rate at which they are dispersed by water systems, but does not seem to limit their.range of dispersion. FCSs discharged into a river or lake will accumulate in the sediment in relatively high concentration. Plants and animals can concentrate PCBs above their level in water alone. Shrimp and oysters exposed to 10 ppb of Aroclor 1254 have been known to bioaccumulate from 130- to-;3300-'iold'.' PCBs are not intended to get into the environment, but they do because -their unique chemical properties prevent them from being destroyed by usual waste disposal methods. Thus, they inadvertently escape and become widely dispersed. Environmental transport models for PCBs have not been developed. Conclusions regarding their behavior, particularly their distribution and transport in the environment, are based largely upon the results of DDT research. Recent data indicates m that aerial transport is the most efficient route, although dumping and river runoff will contribute more effectively to local contamination. Disposition of PCBs The persistence o'* PCBs necessitates extremely careful attention to final disposition of PCBs and PC3-contaminated A t * CCfc i 3 1 7 materials. Initial attention within the manufacturing or PCB-user facility should be given to reducing the use of FCEs. Manufacturers or users may either use substitute (less contami nating) materials or more efficient methods which produce leec waste. However, no matter how many substitutes are develcred and no matter how efficient the operation may become, there may still remain some applications for which PCBs are the cur rent optimum material and there still will be waste materials. Spills, damaged goods, and housekeeping materials will require disposal. Recommended options for the disposal of PCB-containing wastes (in priority order) are: Reclamation or recycling Incineration - Controlled land disposal Recycling - Reclamation or recycling of PCBs is usually feas ible only with relatively uncontaminated liquids. Monsanto maintains a .toll-free telephone number (800/325-3350) for the public for advice on scrap PCBs. If liquids cannot be reused, these liquids needs to be disposed of by high temperature incin eration. Several incineration services are available such as Rollins Environmental Services, Inc., Chem-Trol Pollution Services, etc. Addresses are given in Appendix B. Monsanto also has a high temperature incinerator. A technical descrip tion of the Monsanto incinerator is provided in Appendix C. Recycling of transformers is accomplished by General Electric Co. and- several independent companies. The tank and inside copper of the transformer Is reclaimed; the PC3 liquide, ack CCo 8 if they can't be- reused, can be drained and incinerated at , 0 facilities such as General Electric's Pittsfield, Massachusetts plant. It is quite possible to leach out the PC3s from solid wastes with a suitable solvent, e.g., acetone, and then dispose of the decontaminated waste. The solution containing PCBs may be separated by distillation into PCB, which may be reclaimed, and the solvent which may be used again. This technique (of leaching and reclamation) is especially useful after accidental spills of liquid PCBs. The'spills should be absorbed on dry sand, ash, sawdust, or commercial absorbents and then processed as above. Incineration - Present knowledge indicates that the proper incineration of waste PCBs must involve a suitable balance among temperature in the incinerator, dwell time in the firing .. chamber, plus oxygen availability. Also, a suitable scrubber must be provided on the exhaust stack to remove the hydrochloric acid mist that is formed. % Preferred requirements ares (1) 2-second dwell time at 1100C(2000F) -and 3% excess oxygen in the stack; or (2) 1 1/2 second dwell time at 1500C(2700F) and 2% excess oxygen in the stack gas.* These alternates are recommended. Open hearth incinerators are not considered suitable; the relatively * These values are fr&m ANSI C107.1-1974. Guidelines for * Handling and Disposal at Capacitor and Transformer-Grade Askarel Containing Polychlorinated Biphenyls, American National Star.card Institute. ACM' CC33 9 low operating temperature of this equipment would only volatilize the PCB and pollute the atmosphere. The above recommendations are suitable for liquids. Incineration of solid wastes containing PCBs has not been demonstrated. However, such destruction does appear feasible. For example, a rotary kiln furnace, with an afterburner and scrubbing system could' probably safely -incinerate solid wastes containing PCBs. Likewise, other smaller, high temperature incinerators with sufficient residence tines may be satisfactory Land Disposal - Wastes containing PCBs should not be disposed of with other mixed wastes in a sanitary landfill. Characteristics of transport of PCBs through the soil are not definitively established. The interaction with other decomposing wastes is not well understood. The ubiquity and persistence of PCBs indicate that their disposal should be carefully controlled until additional data are developed. While these data are being gathered, PCBs (when disposed to the land) should only be placed in a secure chemical waste landfill. In general terms, a chemical waste landfill provides complete long-term protection for the quality of surface and subsurface waters from hazardous waste deposited therein, and against hazards to public health and the environment. Such sites should be located or engineered to avoid direct hydraulic continuity with surface and subsurface waters.' Generated leachates should be contained, and subsurface flow into the disposal area"eliminated. Monitoring wells should be ACh CCb 23 '' 10 established and a sampling and analysis program conducted. Additional characteristics of a chemical waste landfill are described in EPA publication, Landfill Disposal of Hazardous Wastes: A Review of Literature and Known Approaches (EPA/53/ SW-165)**. Encapsulation of wastes in cement prior to burial in a sanitary landfill has been used for small quantities of solids or sludges contaminated with PCBs. Only those specific - sites which have been state-approved for PCB wastes should be used. The EPA publication, Hazardous Waste Management Facilities, (EPA/530/SW-146) provides some information to potential disposers; this data when used in consultation with State solid waste management officials can assist in planning a disposal action. ** Documents referenced here may be obtained from Solid Waste Information, U.S. Environmental Protection Agency, Cincinnati, Ohio 45268 AG* GCfcjHC Appendix A Listing of Names Used for PCBs by Manufacturers Name Aroclor Asbesto! Askarel Chlorextol Diaclor Dykanol Elemex Hyvol Inerteen No-Flamol Pyronol Saf-T-Kuhl Manufacturer Monsanto American Corporation Allis Chalmers Sangamo Electric Cornell Dubilier McGraw Edison Aerovox Westinghouse Electric Wagner Electric General *Electric Xuhlman Electric Clophen Fenclor Kennechlor Phenoclor PK Pyrolene Santothenn Bayer (Germany) Caffaro (Haly) Kenneclor (Japan) Prodelic (France) Caffaro (Italy) Prodelec (France) Mitsubishi (Japan) * generic name used for insulating liquids in capacitors and transfomers; may contain PCBs, \ CCc Appendicej* (These Appendixes ire n;ii 2 p in of American National Standj iJ Guidelinei for lljnJ!:o- and [> ; C jfj;i:o i-a n d rrjnilurm cr-CrjJe Aikarclt Cor'^innu Poi>wnloriftJied biphenyls. C Iu ;~ [.1974 uicluJcJ foi infoiinauon purposes only.) i Appendix 3 Disposal Sem ces In addition to the supervised dry landfill sites that gases are passed through a packed column scrubber to may be used for the disposal of aslcard-coniair.ing scrap, remove HC1. the followm3 additional Known fjcil.nes ana services have been established, and others may be available.' Nuclear Engineering Company Eastern Division Chem-Trol Pollution Services, Inc P.O. Box 200 1550 Balmer Road P.O. Box 146 Morehead, Ky 40351 Phone; 606 7S4-6611 Model City, N.Y. 14107 Nudcax Engineering Company Phone: 716 754-8231 Disposal Division This organization has facilities and services capable of handling: Sheffield, 01.61361 Phone; 815 454-2624 (1) Liquids. Askarels alone or mixed with solvents This organization provides containerization, trans or oils. Disposal by hirii-temperature incineration. portation. and diiposal services of all itauids and s.-hds 0 (2) Solids (software;. Adisre1-soaked compounds, (inducing hardware). Cispcsai is in controlled charmed rags cartons, absorbing earths, etc. Disposal by incinera and scientific landfill area. Licensed by Atomic Er.mry tion or scientific landfill. Commission for radioactive waste disposal. The organi (3) Solids (hardware). Capacitors, transformer tanks, zation also has two West Coast locations, in the states cores, askarel-soaked metals. Disposal by scientific land* of California and Washington. fill. Has solvent extraction capability. Rollins Environmental. Services, Inc Monsanto Company 800 North Lindber^t Boulevard St. Louis, Mo 63166 P.O. Box 2349 Wilmington, Del 19899 Phone: 302 658-845 U Phonei 314 694-3352 This organization has facilities and services capable This organization has facilities and services capable of handling askarel liquids alone or mixed with other oils or solvents by high-temperature incineration. Liquid is pumped through a gun with atomizing steam into.incinerator. Temperatures are maintained at 2000aF2500*F with auxiliary natural gas. Exit gases are quenched to 1SO**F by contact with water. Gas is then passed through a high-energy ventun scrubber for re moval of particulates. Before exhausting to air (110eF), of handling: (1) Liquids. Askarel alone or mixed with solvents or oils. Disposal is by high-temperature incineration. (2) Solids (software). Askarel-soaked compounds, rags, cartons, absorbing earths, etc. Disposal :s ov incin eration at combustion temperatures up to 2500'F. In cineration gases are scrubbed, and entrained soucs are removed before exhausting to air. Rollins Environmental Services maintains disposal facilities in the following areas: 'T h e tilting given herein 11 representative of lame o f the lourett (hat provide ihn icrvice, and 11 hot preturned to be complete. Any other organizations tint wuh to tie lined ihould notify the Standard! Institute so that they may be included in the neat edition of this standard. Fhiladelphia/Camden: Rollins Environmental Services, Inc Route 322 Lagan Township Bridgeport, N.J. 08014 19 A CP CC 6 3*14 AP1T.NDIX Baton Rouge: 4. - Rollms Environmental S e rie s, (nc Scenic Highway fit West Chatham Lane Scotlandvdle _ * East Baton Rouge Parish, La 70807 Houston: Rollins Envifoirw.iut Services, Inc Tidal Road &. Higiiwjy 134 Deer Park, Tex 77536 Appendix Vf Description of- Monsanto Incineration Facility . _ . ...Sauget, Illinois . . . Background The Krummrich Plant is one of Monsanto's larcra chemical1manufacturing plants. The product line includes sulfuric acid, benzene, chlorine, polychlorinated biphenyls, several rubber compounds, and various chemical intermediates. The plant is located immediately south of East St. Louis, ill. near the Mississippi River. Since June 1971, Monsanto m has operated a liquid injection incinerator to dispose of in-house liquid wastes and contaminated PC3s from customers. Waste Characteristics The large majority of the wastes burned are;PC3 derivatives. The sources of these wastes are process still bottoms i sy-rited-batche s , and contaminated % transformer oils. The heating value of the materials is about 9000 BTU/lb. Other in-house high BTU liquid wastes are also incinerated. Phosphorous compounds cannot be burned due to the formation of particulates (P20g) wiiich foul the injection system. The system is not equipped to handle suspended solids. \\ Four * l > CCt'j Air Pollution Control f fumes exit the oxidizer and enter a water quench*column. The main purpose of the quench is to reduce the temperature of the hot fumes. Particulates ' are removed next in a high energy venturi. Finally, the emissions are cleaned in a packed bed (polypropylene packing) at the base of the stack. The 40 ft. stack is equipped with a demister. Scrubber liquors are drained to the sewer without treatment. v\ faLP. CG63 5 ! * REFERENCES ' 1. Polychlorinated biphenyls and the environment, Inter departmental Task Force on PCB's. Depts. of Agriculture, Interioi HEW, Commerce, and EPA, May 1972. 2. Carnes, R . , Doerger, J . , Spearks, H.L. Polychlorinated biphenyls in solid waste and solid-waste-related materials. Cincinnati. 1973. .3. American National Standards Institute (ANSI). Guidelines for handling and disposal of capacitor-and transformer-grace askarels containing polychlorinated biphenyls. New York, January 1974. 4. Peaksall, D.B PCB's and their environmental effects. Critical Reviews in Environmental Control. CRC. 1975. 5. Broadhurst, M. Use and replaceability of polychlorinated biphenyls. Environmental Health Perspectives. Oct. 1972. 6. Versar, Inc. Assessment of wastewater management, treatment technology, and associated costs for abatement of PCBs * concentrations in industrial effluents. Jan. 1976. (Draft report) : 7. Bremer, C. State of concerns of the Lake Michigan Toxic Substances Committee related to PCB. Chicago. Internal EPA f\ \ Report. 8. Aroclor.Polychlorinated polyphenyls (biphenyls). Technical Bulletin O-FF/IR. St. Louis: Monsanto Industrial Chemicals Company, Nov. 1971. 9. PCB*s-environmental impact. Environmental Research. 1972. 10. PCB's their use and control. Organization for Economic Cooperation and Development. Paris. 1974. 11. Sewage sludge incineration. Section IV. Effect of incineration on metals, pesticides, and polychlorinated biphenyls. EPA Task Force. March 1972. 12. PCB Conference. National Swedish Environment Protection Board. 1970.' pg. 83-86. 13. Hazardous waste management facilities in the united states. Leshendok, T. Environmental Protection Agency SW-146, revised, 1976. v A O CC6346 / ' 2 14. Scurie A. t al...Incineration in hazardous waste managt int. Environmental Protection Publication SW-141, 1974 . 15. Lindsey, F. and Fields, T. Landfill disposal of hazardous wastes: review of literature and known approaches. Environmental Protection Publication SW-165, 1975. 16. Farb, D. information about hazardous waste management facilities. Environmental Protection Publication.SW-145, 1975. 17 Tucker, E., et al. Migration'of PCB's in soil induced by percolating water. Monsanto Co. Bulletin of Environments 1 Contamination and Toxicology, Vol. 13. 1975 18. ASTM. Standard method of test for rapid gas chromatographic estimation of higher boiling homologues of chlorinated ( biphenyls for capacitor askarels. specification, D 1810-%3' (1973)- , 19. HEW. Registry of toxic effects of chemical substances. 1975 Edition NIOSH June 1975. 20. Mitre Corp. Environmental cycling of PCB's. 1975 (unpublished report). \ \ ACK GCB3<.7 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY WSHINGTON. D C. 20460 October 22, 1975 Mr. D. B. Hosmer U t ilit ie s and Environmental Protection Director Monsanto In d u strial Chemicals Co. 800 N. Lindbergh Boulevard St.' Louis, M issouri 63166 Dear Mr. Hosmer: iI t 1s a great pleasure fo r me to in vite you to attend the "National _ Conference on Polychlorinated Biohenvis sponsored by the Environmental Protection Agency in cooperation with the U. S. Department of A g ric u l ture;, the Council on Environmental Q uality, U. S. Department of Health, Education and Welfare, and the U. S. Department of the In te rio r. Your attendance and p articip ation w ill help ensure that the conferees benefit from the presence of recognized author it ie s in th is fie ld . Please reserve the dates ofQlovember 19-21 , j )?75 for th is meeting to be held at The Pick-Congress H&tel, CMUayu, I l l i n o i s . Enclosed' for your convenience is a hotel reservation card. I would suggest that you make reservations as soon as possible. - The program w ill la s t fo r three days. The objectives of the con ference are to: ... bring together the la te st data and best available expertise to help c la r if y the problems associated with the manufacture, use and disposal of PCBs. ... help assess the effectiveness of steps taken to reduce the problems associated with PCBs. ... provide a platform fo r interested parties to present previous y neglected data concerning PCBs. ... help c la r if y the f e a s ib ilit y and complications of steps to reduce the problems associated with PCBs. VI s _ r ^- - J- L A t* C C 3 H_e The Adm inistrative Coordinator for the conference is Mr. Franklin A. Ayer, Research Triangle In stitu te , Post Office Box 12194, Research Triangle Park, North Carolina 27709; telephone (919) 549-8311, ex tension 640 or 632. You are encouraged to write or telephone him i f you have questions concerning th is meeting. In addition to the enclosed hotel reservation card, you w ill find a conference re g istra tio n card which you should f i l l out and return to Mr. Ayer as soon as possible. I am looking forward to a stim ulating conference and hope that you w ill be able to attend. Sincerely, Enclosures: Hotel Reservation taro Conference Registration Card PRELIMINARY AGENDA. NATIONAL CONFERENCE ON POLYCHLORINATED BIPHENYLS FCBs in Che Envlronaenc - 1975 The Plck-Congres9 Chicago, Illinois / November 19*21 1975 Sponsored by The Environmental Protection Agency in cooperation with the Depart ment of Agriculture, Council on Environmental Quality, Department of Health, Education and Welfare, and the Department of the Interior. Program - Outline - General Chairman - John L. Buckley, Ph.D. Environmental Protection Agency Wednesday^ November 19. 1975 10:00 KEYNOTE ADDRESS - Honorable Russell E. Train, Administrator, United States Environmental Protection Agency 11:00 HEALTH EFFECTS AND HUMAN EXPOSURE - David P. Rail, M.D., Ph.D., Director, National Institute of Environmental Health Sciences and Chairman, Department of Health, Education and Welfare Consulttee to Coordinate Toxicology What exposures to FCBs have vhat effects on man? Review of toxicological, clinical and epidemiological data obtained since 1972. 12:00 Lunch 1:00-3:00 HEALTH EFFECTS AND HUMAN EXPOSURE (CONTINUED) 3:30-5:30 USES, SOURCES, AND IDENTIFICATION - David Garrett, P.E., Chief, Special Projects Branch, Office of Toxic Substances, Environmental Protection Agency What are FCBs used for now, how much is used, where do they come from, how do they reach the environment? Review of present manufacture, import, use, distribution and disposal, and review of chemical analytical procedures. Thursday. November 20, 1975 8:30-12:00 ENVIRONMENTAL FATE AND OCCURRENCE - Ian C. T. Nlsbet, Ph.D., Research Massachusetts Audubon Society, How much occurs where, how does It move in the environ ment, end vhat happens to It? Report on current levels in various parts of the environment -- water, soil, air, fish, wildlife, human foods end man. Review vhat Is known of transformations and degradation - mechanisms and rates. To the extent possible, synthesize in mass-balance terms, with particular attention to ex posure routes snd levels for man and other living organisms. AC* CC835C ( 12:00-2:00 LUNCHEON -_ Speaker (To Be Announced) 2:00-3*30 ECOLOGICAL EFFECTS AND EXPOSURES - Donald I. Mount, Ph.D., Director, National Uater Quality Laboratory, Environmental Protection Agency Vhat is happening/ln freshwater, marine, estumarlne and terrestrial environments because of PCBs? Review of laboratory and field studies completed since 1972. 4:00-5:30 ECONOMICS AND SUBSTITUTES - Warren Muir, Ph.D., Council of Environmental Quality ^ What are PCBs used for now, what substitutes are available, and what are the economic, health and environmental aspects of proposed substitutes? 5:30 Dinner 7:00-9:00 GENERAL SESSION - Christopher M. Timm, Director, Surveillance and Analysis Division, Region V, Environmental Protection Agency Brief statements by Interested groups, or individuals expressing opinions on the PCB situation, and elaboration of discussion points curtailed by lack of time during sessions. Statements from the floor from persons registering re quests prior to the opening of the General Session. Friday. November 21. 1975 8:30-10:15 APPROACHES TO CONTROL (Chairman to be Announced) What can be done to reduce amounts of PCBs reaching the environment and to limit human exposure? Panel discussion by Federal and State Agency spokesmen, and representatives of industry. 10:30-11:45 SUMMARY SESSION - Richard Carpenter, Executive Secretary, Commission on Natural Resources, National Academy of Sciences. Concise sumoarles of the Information presented and significant points raised during discussions. 11:45-12:30 CONFERENCE HIGHLIGHTS - John L. Buckley, Ph.D. . Integration and synthesis of the most significant findings and recommendations that emerged during the conference. Richard Carpenter. CONCLUDING REMARKS - Conference Chairman ACh CC8351 October 7, 1971 Dr. Ronald G. Webb Research Chemist Environmental Protection Agency Southeast Water Laboratory Athens, Georgia 30601 Dear D r . Webb: Thank you for sending me the chromatograms and isomer samples in which I expressed an interest. Upon my arrival in St. Louis, I found that the Unilever article which we discussed had been published (J. Chromatography, 60, 15-32, 1971) For this reason, I did not send you a copy of the unedited preprint. I have, however, enclosed a copy of Dr. Berg's paper which discusses his novel approach to PC3 quantitation. We currently do not employ Dr. Berg's procedure in our laboratories, but plan to evaluate the use of the perchlorination technique for quantitation. I enjoyed listening to your presentation on the "Identities of PCB Isomers" and being able to meet and talk with you after^ wards. Sincerely, ,^E. S. Tucker Research Group Leader Analytical Chemistry Group w. B. Papageorge v AO* CC6352  ci ENVIRONMENTAL PROTECTION AGENCY SOUTHEAST WATER LABORATORY Athens, Georgia 30601 / September' 28, 1971 Dr. E. Scott Tucker Monsanto Company 1700 South Second St. St. Louis, Missouri 63177 Dear Dr. Tucker: As I promised you at the PCB Analysis Meeting in Washington, September 17, X am sending under separate cover samples of 2,5,21,51-tetrachlorobiphenyl and 2 , 4,5,2 * ,S '-tetrchlorobiphenyl. I am also including some electron capture gas chromatograms of the compounds. You mentioned that you have a preprint of Telling, Sissons, and Welti's paper on identification of PCB isomers by MS and NMR. I would like to have a copy of that paper. Sincerely, r> Ronald G. Webb Research Chemist : fri-- xtj*.* lu ; ;* r- i. fa/.'-- -I^ ACH 006353 "Sr -c 0CT18 W q UNITED STATES ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460 f October 12, 1978 OFFICE OF ENFORCEMENT Mr. V. B. Papageorge' Monsanto Company 800 North Lindbergh Blvd. St. Louis, MO 63166 Dear Mr. Papageorge: This office is In the process of developing a disposal control program for Polychlorinated Biphenyls (PCBs). It would be very helpful to our effort if ve were able to discuss past uses of PCBs, and the nature of the former PCB-uslng community with members of your organisation. Our principal concern is to identify feasible commercial routes for the disposition of PCBs that are available to persons who are disinclined to pay the cost of proper storage and disposal. Your knowledge and experience would be helpful In this analysis. If you, or other persons In the Monsanto organization could provide some time to discuss PCBs with my staff, I would be grate ful. My staff can travel to St. Louis or elsewhere at your con venience. t Thank you for your consideration of this request. 1 look forward to hearing from you. Sincerely \ ACM CC835A M O N SAN TO INDUSTRIAL CHEM ICALS CO. BOO N. Lindbsrgh Boulevard S t. Louis. M issouri 63166 Phons: 014) 694-1000 November 2 0 , 1976 Dr. Jerry M. Johnson Acting Chief, Criteria Documentation, Section 1 NIOSH, DCDSD, CDB, RM 142 4676 Columbia Parkway Cincinnati, Ohio 45226 Dear-Dr. Johnson: Attached are answers to the questions on the draft criteria document concerning occupational exposure to polychlorinated biphenyls. We hope our comments are helpful and appreciate the opportunity to participate in the preparation of a responsible criteria document. Sincerely, WBP/ken attachment W. B. Papageorge Mgr., Product Acceptability u n it ol Me.- s a n io C o m u iriy ACK CC8355 .r'l l. ' A p o l y c h l o r i n a t e d '-b i p h e n y l e x t e r n a l ' r e v i e w q u e s t i o n s 1. The document inadequately documents industrial exposure related effects of PCB. It also inadequately defines clinical tests which could be used to monitor for potential adverse effects of PCB exposure among the industrial population. The document dwells at great length upon the effects'observed in a population of Japanese citizens who were accidentally exposed to an acute dose of PCB in a cooking oil. No relation is established between this accidental acute oral exposure to a relatively high concentration of PCB and industrial type exposures. No relationship is established defining relative body burdens resulting from acute oral exposure versus other routes of exposure, dermal absorption and respiratory, that are more likely to occur in an industrial environment. Another study given seemingly unwarranted consideration, involved an unpublished report of an epidemiological study conducted among a group of. research workers with some unknown previous exposure to PCB. No sound basis has been established to support the contention that PCBs are human carcinogens. It is suggested that efforts be extended to accurately assess the clinical health of U.S. workers exposed to PCB. 2'. In studies with rodents and subhuman primates the most sensitive target organ was usually found to be the liver. In these studies the dietary level which did not produce an effect on the liver was between 1 and 10 ppm. Based upon these studies the most probable toxic effect of excessive PCB exposure would be chronic hepatic injury 3. From the manner in'which the document is put together it is impossible to tell if the recommended standard is based upon a logical scientific extrapolation of the information cited. Specifically, no attempt is made to relate the body burdens which would result from an industrial inhalation or dermal exposure to those obtained frpm animal experi mentation or the usual data following oral administration. Chapter V consists of a series of data relating the toxic effects observed after many different types or routes of exposure. It then states that on the basis of all this data the environmental limit should be 0.05 mg/M3 of PCBs in air. There is no logical support for the stated limit as based on the industrial usage this document is intended to relate. There has been no evidence shown to indicate that, as inferred on lines 4071-4072, "teratogenicity, fetotoxicitv ar.c carcinogenicity has occurred to man and to laboratory animals when they have been subjected to air-borne PCB levels below the current standard of 1.0 mg/M3 (42% chlorine)." 4a. To be supportable a standard must be based on the application of a safety factor to experimental animal data and/or upon the utilization of previous work history exposure concentration and clinical worker monitoring data. Which of these was used in establishing this standaz' is not clear. Specifically, the limit of 0.05 mg/M3 has not been supported by any data relating industrial worker exposure to clinical effects. In fact, the arrival of 0.05 mg/M3 as a "safe" level has not been explained at all. 1 b. The document on line 3929 states that the current PCB standard is based on an 8-hour dermal exposure. This is not entirely c o r r e c t . AC* CCtlfct 2. The current standard does include a "skin notation" which refers .to the potential contribution to the overall exposure by cutaneous, route or particularly by direct skin contact. An environmental limit for air-borne material cannot totally prevent direct skin contact with material and the resulting dermal exposure. This.concern has been aptly addressed by the ACGIH .TLV .committee on line 3925 by their reference to the fact that'the current standard may not guarantee protection, against dermal effects. The recommended environmental limit cannot guarantee such also since workers can still put unprotected hands in the material if they wish. c. Inadequate information exists to properly evaluate the need for and type of protection for the pregnant worker. d. Until better scientific evidence is brought forth supporting a necessary change in the current OSHA standard on PCBs, no justificatic exists for modifying the existing environment limit. 5. We recommend that the current users of PCBs, capacitor and transformei manufacturers, be consulted regarding feasibility of compliance. 6. To exclude biological monitoring in a standard for a material known to accumulate in fatty tissues would not appear advisable. .Biologic testing is an acceptable monitoring device if it is shown that a reasonable correlation exists between body burden, detrimental health effects, and degree of worker exposure. Biological monitoring may well give an indication of dermal exposure and degree of skin absorption that might not be evident through the use of air monitorinc 7. We assume that if the composition of PCBs is unknown, the sample chror togram does not match the chromatogram of any commercial PCB standard mixture. In this case, measurement by electron capture gas chroma tography is inadequate for two reasons. First, interferences may be present which may account for the differences in the chromatograms. Secondly, as pointed out in Table I of the draft, electron capture detector response varies by orders of magnitude for different PCB isomers. Hence, the analyst would not know what response factor to apply, even using the weight factor technique of Webb & McCall (see attached comments for line 5354). Therefore, when the PCB composition is unknown, a more specific and accurate technique must be used. We recommend using GC/MS in the selected ion monitoring mode as the detector device. The parent ion and a characteristic fragment ion are monitored for each PCB homolog. This technique is much less subject to interferences and is more accurate because MS response factors are nearly constant. GC/MS detection can be used with samples collected on Florisil if sensitivit is adequate. If sensitivity is inadequate, samples should be collects on a porous polymer sorbent such as Tenax GC. .Collected components are then thermally desorbeddirectly into the GC/MS for analysis. 3. In those cases where the sample chromatogram matches a standard PC3 chromatogram, the sampling procedure is adequate. However, as mentioned in 7., we recommend sampling with Tenax GC as a more general method. Collected components can be thermally or solvent desorbed from Tenax GC. For further details, see pp 147-160 of the transcript of the Second NIOSH Solid Sorbents Roundtable in December, 1973. CC6357 3. 9. No, several deficiencies should be corrected. See attached comments relative to lines 5252, 5253, 5258, 5273, 5291, 5292, 5303, 5310, and 5320 of the draft. 10a. We recommend collection on. Tenax.GC adsorbent, thermal desorption and measurement by GC/MS. See 7.,^tf., and attached notes. b. See attached comments, especially the notes relevant to lines 5354, 5381, 5396, 5410, 5460, 5533, and 5590 of the draft. 11a. Annual comprehensive medical examinations are considered appropriate. b. The medical requirements listed in the document under Section 2 are patterned after the criteria document on Tetrachloroethylene. It would seem that the medical tests for PCB exposured workers could be more specific for that particular material. Pulmanary function tests should be included as a check on capability of wearing respir atory protective devices. Contrary to lines 202-203, medical surveillance should be made available to all workers occupationally "exposured" to PCBs who contact the material and yet not be "exposed" on the basis of the NIOSH definition given on lines 176-177. 12. There is no persuasive evidence supporting the "cancer suspect agent" labelling all PCBs, therefore such labelling is inappropriate. Warning signs should be posted in those areas in which workers are exposed above some specified limit or where potential skin contact is possible. To be functional a warning sign must inform the reader of specific effects of a material. PCBs have relatively mild irritant properties and has not been associated with acutely toxic effects. However, tissue accumulation and target organ effects may occur following prolonged exposures above the established permissible limits. A meaningful sign would warn of potential tissue PCB accumulation and indicate avoidance of skin and eye contact. 13. Considering that the concentration limits as listed in Table 1-1 represent the protection factors as shown below, the limits are considered to be appropriate and consistent with the respirator decision logic used in the Standards Completion Project. Along with the protection factor limits are listed various additional respirator types which should provide adequate protection when used in areas of respective air contaminant concentrations. Concentrations of Polychlorinated Biphenyls 0.5 mg/M^ or Less (lOx^ Respirator Type (Modifications to Table) 3. Chemical cartridge respirator with replaceable cartridges approved for organic vapors and mists and half-mask facepiece. 4. Gas mask, chin-stvle or front or back mounted organic vapor and mist cannister. ACf* C C 6 3 5 6 4. Concentrations of Polychlorinated Biphenyls cont'd. 2.5 mg/M3 or Less (50x) *t j Respirator Type (Modifications to Table) cont'd.______________ 3. A gas mask with a full facepie and front or back mounted vapc cannister with high efficiency filter. 5.0 mg/M3 or Less (lOOx) (No modification) Greater than 5.0 mg/M3 (lOOx) (No modification) 14. Line 292: Detectable level is mentioned. Since the detectable limit of PCBs has not been addressed, this level is meaningless. It would make more sense to reduce air-borne concentrations of PCBs to below the (TWA) environmental limit or permissible exposure limit. Lines 379 through 382 should be deleted since it seems unnecessary to have escape type devices available in all areas of "potential" exposure where the exposures might be brief at low or moderate levels. Escape type devices are generally made available in defined high risk areas containing potentially hazardous operations involving extremely toxic substances. The document does not provide data which supports PCBs as being considered extremely toxic. Earlier comments addressed the modifications in Table 1-1 on respirator types vs. concentration limits. 15. Line 405: Do hazards- imply only health related hazards or environment hazards also? In general, Section 5 provides adequate avenues for communicating work hazards to workers. 16. The work practices and control procedures are considered adequate for the protection of workers. 17. The recommended record retention period appears to be appropriate. 18. Additional research is needed in the following areas: a. Comparative data on the PCB body burden resulting from oral, dermal and inhalation exposures. b. Comparative metabolism studies of an homologous series in humans, non-human primates and rodents. These studies should include both a characterization of the metabolites formed and kinetic studies to determine the rates of mobilization and excretion. c. Clinical evaluation including liver function tests on industrial populations with previously documented PCB exposure. d. Epidemiological studies on major populations of PCB-exposed worker 19. Attached is a report compiled by Monsanto and distributed in November, 1975 to representatives of concerned governmental agencies, (NIOSH, OSHA, EPA, FDA, NCI, NIH) relating to histopathological evaluation of livers from rats treated with PCBs. ACK CC8 35 5 COMMENTS ON SAMPLING ftND ANALYTICAL METHODS Minor conments are on the o r ig in a l NIOSH d ra ft. y* General corments: In using trade nams, the document should be consistent. Any . . shorthand such as dropping re g istra tio n , notation should be documented at the e a r lie s t opportunity and then a standard format adopted thereafter e.g. ^roc-]or R Flo risil R Environmental Sampling and A n a ly tic a l Methods 3714 ___ evaluated under lab oratory conditions using aerosol generation tech n iq u e s. F ie ld stud ies were a ls o performed in a PCB environment containing aerosol and vapor components. The statement as i t stands leaves open the question of aerosol or vapor stu d ie s. The document should be completed by in c lu sio n of th is informa tio n in a manner s im ila r to the above amended statement, and expressing the true nature o f the e xercise. Both aerosol and vapor concentrations must be included. 3724 -3726 E a rlie r in 'th e text, 2 solvents are mentioned along with the use of sin g le and m ultiple impingers. Which solvent was- used should be c la r if ie d (toluene or ethylene g ly c o l) and the number of impingers. The conclusions reached imply a study using an aerosol o f PCBs or an equilibrated vapor environment. Is th is true? 3737 to use when personal sampling fo r PCBs. This circumvents other techniques such as area monitoring for calculating effective exposures. 3738 NIOSH has not tested a ll a va ila b le s o lid sorbents (e.g. porous polymers). Hence, t h is lin e should be reworded to re fle c t only those methods tested. 3752 Electron capture gas chromatography i s the most widely used method, but not n e c e ssa rily the "p re fe rre d " method. For example, the wide variancein response factors referred to in Table I is a d istin c t disadvantage. 3801 -3803 In th is day, i t ' is not p o ssib le to elim inate a method based on the lack or expense of a computer. In many small companies, the OSHA requirements w ill ra p id ly force the purchase o f such a unit fo r many a p p lica tio n s. 3809 -3810 One might prefer NBS as a re p o sito ry fo r "standard" samples. 3824 P e rch lorin ation should d e fin ite ly not be used to quantitate PCBs. We have confirmed that not only biphenyl, but a lso sub stituted biphenyls, cause large p o sitiv e e rro rs. At le a s t one of the replacement flu id s fo r PCBs in capacitors' is a su b stitu te d biphenyl (Chemical and Engineering News, p. 25, Nov. 15, 1976). Hence t h is interference is already present in the manufacturing f a c i li t y and precludes use o f p erchlorination. See attached copy o f le tte r from J. Coleman Weber to Dr. I. E. Wallen. AW CC36C -2- 3834 -3835 3843 -3845 See consents on alternate stand ard ization procedure, lin e 5354. See above comments on p e rch lo r*in at//io n . APPENDIX I 5248 5252 -5253 5258 -5260 5266 -5268 5273 5291 5292 -5293 5303 5304 5308 5310 5320 Samples collected must be representative o f the personal exposure of individual workers. Records should include: Pump model & S e ria l no. Sampling Tube type and no. i. e . sampling equipment d e ta ils C a lib ra tio n data should a lso be recorded. The aim of sampling i s to permit determination of the personal exposure level p rim a rily . T his leads to a decision on compliance/ non-compliance situ a tio n s. I t cannot lead to the "lowest feasible le v e l" without other major input beyond the scope o f sampling and a n a ly sis. A irflo w through the pump sh a ll be co n tro lla b le w ithin 5% of the desired rate during the e n tire sample period. " A more frequent requirement fo r pump c a lib ra tio n i s required. At le a st monthly and preferably in the method fo r a d a ily setup caTibration. SpotTchecked i s an open statement and means nothing. For d ig it a l readout pumps sp e c ify p a rity check between expected stroke count versus actual. For pump with rotameters sp e c ify before and a fte r readings. We would prefer a c a lib ra tio n a fte r flow s t a b iliz a t io n and again before pump is turned o ff , fo r each sample. The sorbent tube should not be v e rtic a l. P o in tin g down - loses g la ss wool and F l o r i s i l Poin ting up - dust from overhead, hard hat, etc. enters. Suggest a close to horizontal location , attached to the c o lla r of the worker. A ir being sampled should pass d ire c tly in to the open in le t of the sorbent tube This precludes f ilt e r s . i The recommended sample volume fo r th is method i s 50 l i t e r s (U .5 .`s p e llin g ) \ The sorbent tubes should be la b e lle d . At th is point i t should re ite ra te checking label and I.D . Number. A bulk a ir sample is o f no value unless the re s tra in ts required to obtain a good and recoverable sample are f u ll y documented. fiL I* C G 6 3 fc1 A ) 5324 5346 5354 f -3- Throughout Appendix I I nomenclature fo r PCB and PCBs is confused. In general, PCBs should be used as a noun d e scrib in g a PCB mixture, w hile PCB i s an adjective. This reads more smoothly i f w ritten , "T his would correspond to a detection lim it of 40 ng of* 5 ml o f desorbent". i*i This alternate standardization procedure is of highly questionable u t ilit y . As pointed out in . lin e 5358, i t has not been evaluated in NIOSH la b o ra to rie s. As co rre c tly seated in the d ra ft, the reason an alternate procedure is needed are two-fold. F ir s t , most o f the GC peaks in a chromatogram o f a PCB product contain more than one component. Secondly, electron capture detector response d if f e r s s ig n if ic a n t ly fo r d iffe re n t PCB isomers (See Table I o f the d ra ft). These two fa cto rs ind icate that the absolute quantity o f PCB represented by a given peak containing unresolved components va rie s depending on the re la tiv e amounts o f the components. The re la tiv e composition o f a given peak can vary fo r the Yollow ing reasons: 1. PCBs with a given degree o f c h lo rin a tio n (i.e . 42%) from d iffe re n t manufacturers contain different isomer ratios. 2. Component r a tio s vary fo r products with d iffe re n t degrees of c h lo rin a tio n . For example, from. Table X II - 6, the ra tio of d ic h lo ro - to t r i chiorobiphenyls fo r the peak with RRT 28 in A ro clo r 1221, i s 85:15. From Table X II - 8, the same peak in A ro c lo r 1242 has a ra tio o f 25:75. S im ila r ly , fo r peak RRT 70 in A ro c l r 1242 and 1254, the te tra ch lo ro -to pentachlorobiphenyl ra tio s are 90:10 and 25:75. 3. Vapor pressure d ifferen ces a lt e r the composition of vaporized PCBs re la tiv e to the liq u id . (See attached figu re showing chromatograms of liq u id and vapors from A ro clo r 1016). 4. In te rfe rin g components may be present which d is t o r t the sample chromatogram. This is l i k e ly to be a major problem in capacitor and transform er manufacturing f a c i l i t i e s because several of the PCB replacement products ( i.e . phthalate e sters and ch lorobutyldiphenyl ether) have s im ila r GC retention times to PCBs and give an electron capture response. The weight fa c to r technique o f Webb and McCall does not adequately account for these d iffic u lt ie s . For those situ a tio n s where the sample chromatogram c lo se ly matches the chromatogram of a reference PCB mixture, the Standard A n a ly sis described in 5554 f f should be followed. However, i f the chromatograms do not match, we believe th only sa tis fa c to ry a n a ly tic a l method which w ill y ie ld accurate re su lts is gas . chromatography/mass spectrometry (GC/MS) using selected ion m onitoring (SIM ). The detection lim it o f t h is technique for A ro clo r 1016 i s about one nanogram of PCB injected into the ^ instrument. Considering the environmental lim it of 0.05 mg/m recommended in t h is document, i t might be p o ssib le to reach th is lower detection lim it fo r the total procedure by sampling as outlined in the d ra ft, evaporating the hexane desorbent and analyzir. the concentrate by GC/MS. AG* CC6-362 ia 5381 5396 5410 5460 5515 5533 5590 6118 -4- A lte rn a tiv e ly , thermal desorption from a su ita b le s o lid sorbent d ir e c tly in to the GC/MS can be used. This well documented technique analyses the e n tire collected sample at once, rather than o n ly an a liq u o t at a time. Response factors vary only s l i g h t ly fo r d iffe re n t PCB isomers and most interferences are eliminated. / The buTk sample must not/be shipped in the same container as the sample tubes. The volume of a ir sampled can be measured to w ithin at le a st 5%. Delete next sentence. Delete 100% and conclude sentence with recovered with a relative standard deviation of 4.4% fo r 27 spiked samples". Delete next sentence. Type o f electron capture detector i s not important. r in s in g with p e sticid e grade acetone and hexane should follow the water rinses. Argon/methane i s a lso a su ita b le GC c a r r ie r gas. The flow rate fo r e ith e r c a rr ie r gas should be only about 30 ml/min fo r a 2 nri id column. Actual retention times should be used in c a lc u la tio n s i f a va ila b le . I f not a v a ila b le , then the measurements described here can be used. The vapor pressures in Table X I I I - 5 appear to be too high. What i s the source o f these data, and has th e ir accuracy been v e rifie d ? ! ACH C C 6 3 t 3 I I March 13, 1976 Dr. I. E. Wallen Environmental Protection Agency Office of Toxic Substances 401 l\ Street S.W. Washington, DC 20460 Dear Dr. Wallen: Monsanto has a concern over the validity of the perchlori nation technique used by UFA and others to measure and/or confirm polychlorinated biphenyls in environmental materials. In our investigation of the perchlorlnation method, we have found that such chemicals as biphenyl, alkylated biphenyl, and many other substituted biphenyls, Interfere with the perchlorination technique. It also appears that various petroleum components may interfere. If these chemicals were present in environmental materials that were being tested for PCD's using the perchlorination method, erroneously high PCD concentrations would Do reported. A recent article in the Journal of the AOAC (Vol.SG, Ko.3, 1973} points out two other limitations of the perchlorination procedure: 1. High and variable reagent blanks, which cause erroneously high findings. 2. Formation of. bromononachlorobiphcnyl, which causes low recoveries. v A copy of the article is attached. ADK CC 36*1 D r* Z* E . W a lle n 2- K a r c h 1 5 , 1 9 7 6 Since these limitations 'can lead to significant errors in determining trace levclc'of rcu'c in environmental samples, we suggest that EPA carefully review tho validity of the perchlorinaticn technique. Fesults that have been obtained Using this technique may not be valid. If any other information is needed, please let us know. Sincerely, J7 Coleman Weber Manager, Product Acceptability mah cc: Dr. A. C. Trakovski Environmental Protection Agency bcc: W. C. Hammann R. E. Keller J . P .*Mieure D. Wood - B2SC K. Vi. Easley - Washington 1920 R. G. Kalcy - T2F V\ Alt' CC365 L 990900 *av Aroclor 1016 Vapors 1 Figure 7 V Monsanto SEC7 W i MONSANTO INDUSTRIAL CHEM ICALS C O . 8 0 0 N. ndbor$h Boutovard S t. Louia. Mraaouri 63166 Phon: (3143 6 9 4 -10 0 0 JC O A L T r CHCM!C*1 QWtSiQH December 6, 1976 y Hr. Thomas Kopp Environmental Protection Agency Office of Toxic Substances 401 H Street, S.W. Washington, D.C. 20460 Dear Mr. Kopp: Enclosed are the first'thru third quarter production and sales figures for Monsanto's Aroclor products which you requested. Sincerely, mg Enclosure bcc: K. W. Easley C. Paton * R. A. Stohr^ D . Wood \ a unit Of Montcnto Company ACM C063fc7 o US PRODUCTION SALES1 DOMESTIC EXPORT TOTAL DOMESTIC SALES BY PCB GRADE Aroclor 1221 Aroclor 1242 Aroclor 1254 Aroclor 1016 Total b. o .** n moLv Cftr MONSANTO INDUSTRIAL CHEMICALS COMPANY PCB MANUFACTURE AND SALES (THOUSANDS OF POUNDS) lo . 20 10645 6261 \ 6955 1383 8338 5991 960 6951 YTD 16906 12946 2343 15289 3Q 5691 5202 1062 6264 YTD \ -4ft -- 22597 18148 3405 21553 4 4 15 19 19 1376 732 2108 862 2970 476 396 872 377 1249 ?0?9 ifiia 9947 2262 13910 6955 5991 12946 5202 18148 YTD r P T \ W zj UNITED STATI ENVIRONMENTAL PROTECTION AGENCY WASHINGTON. D.C. 20460 _ n ^ .OCT 1 } 1975 .' OFFICE OF ENFORCEMENT Monsanto Industrial Chemicals Company 800 N. Lindbergh,Boulevard St. Louis, Missouri 63166 Gentlemen: Q C T 2 2 V':*5 Recent governmental sampling data indicate the presence of Polychlorinated Biphenyls (PCBs) and comparable chemical substances in the air, in water bodies, and in fish in several areas of the country. In order to determine the nature and extent of the possible adverse effects resulting from the presence of PCB compounds in the environment, the Environmental Protection Agency (EPA), in cooperation with other federal and State agencies, Is attempting to determine the sources and amounts of PCBs entering the environment. It is important that this effort be carried out without delay. It is our understanding that your company handles PCB compounds or mixtures or comparable chemical substances in its operations. I am therefore requesting, pursuant to the authority provided by Section 308 of the Federal Water Pollution Control Act, as amended, 33 U.S.C. 1318, and Section 114 of the Clean Air Act, as amended, 42 U.S.C* 1857c-9, that your company furnish EPA with information pertaining to your use and handling of PCBs and comparable chemical substances. In addition to a general description, which should Include Information as to sources, quantities, uses, and ultimate disposition, you should respond in detail to the enclosed questions. If any question is not applicable to your company or operations, please so indicate by responding "not applicable." The information requested herein must be provided notwithstanding Its possible characterization as confidential Information or trade secrets. Should you so request, however, any information (other than effluent or emission data) which the Administrator of this Agency determines to constitute methods or processes entitled to protection as trade secrets will be maintained as confidential, pursuant to procedures specified in 40 CFR Part 2. \ PLA IN TIFF'S a* bVi [> C C c 3 9 2 Within 14 days of receipt of this letter, your company must provide all information concerning /our current status and activities and covering the twelve month period immediately preceding receipt of this letter. Within 30 days following receipt of this letter, your company must provide all.information for all of the prior years indicated. The information required herein should be sent directly to the address indicated below. If you have any questions'"you may call the person ' I n 2 i c a ted below or Mr. Blake A. Biles of our office k*202) .75r'->73! . We appreciate yuur prompt cooperation in this matter. Enclosure Regional Contact: -- * Mr. Earl J. Stepucubh Director, Enforcement Division Environmental Protection Agency 1735 Baltimore Kansas City, Missouri 64108 Telephone: (816) 374-2576 \ AC* CC637C POLYCHLORINATED BIPHENYL (PGB) COMPOUNDS OR MIXTURES Within 14 days of receipt of'this letter, your company must provide all information concerning your current status and activities and covering the twelve month period immediately preceding receipt of this letter. Within 30 days following receipt of this letter, your company must provide all Information for all of the prior years indicated. FOR PURPOSES OF THIS LETTER, THE PHRASE "PCB COMPOUND OR MIXTURE" INCLUDES ALL CHEMICAL SUBSTANCES KNOWN OR BELIEVED BY YOU TO BE PCB COMPOUNDS OR MIXTURES OR OF A SIMILAR CHEMICAL NATURE, IRRESPECTIVE OF TRADE NAME, AND SPECIFICALLY INCLUDING * CHLORINATED TERFHEKYLS. 1. For each PCB compound or mixture produced or imparted bv your company, for each company facility, during each year of ^ 7 1 ^ 1972. 1973._1974._and the first two_quarters of 1975.L- a. The total amount of each PCB compound or mixture produced or imported. b. The name end address of each of your company's facilities which handle' PCB compounds or mixtures* (Including production S H f V . Ki facilities and wholesale and retail outlets), and the + amount of each PCB compound or mixture distributed through each facility. C . The name and address of each customer of each PCB compound or mixture, the amount of each PCB compound or mixture obtained by each customer from each facility, and each customer's delivery polnt(s) for the receipt of each PCB compound or mixture. .2 2. For each PCB compound or mixture Incorporated by your company Into its products, for each company facility, during each year of 1971, 1972, 1973, 1974, and the first two quarters of 1975: a A description of each product. b* For each product the total amount Incorporated of each PCB compound or mixture. CC6311 alp 2 2. (continued) /,, c\ For each product the name*and address of each source from which your company obtained each PCB compound or mixture, and the amount ^of each PCB compound or mixture obtained from each source. d. The name and address of each of your company's facilities which handle such products (Including production facilities and wholesale and retail outlets), and the amount of each product distributed through each facility. e. The name and address of each customer of each product, and i the amount of each product obtained by each customer from ' each facility. For consumer products list only the total production of each product at each facility and the total number of customers. Do not provide the name and address of each customer of consumer products. 3. For each PCB compound or mixture used by your company in its operations other than for incorporation into its products, for each company facility, during each year of 1971, 1972, 1973, 1974, and the first two quarters of 1975: a. A description of each use. b. For each use the total amount of each PCB compound or mixture. c. For each use the name and address of each source from which your company obtained each PCB compound or mixture, and the amount of each PCB compound or mixture obtained from each source. 4. For each PCB compound or mixture reclaimed by your company, for each company facility, during each year of 1971, 1972, 1973, 1974, and the first tiro quarters of 1975: a . A description of each method of reclamation. b. For each method of reclamation, the total amount of each PCB compound or mixture reclaimed. i e. The total amount of each PCB compound or mixture reclaimed. AC* CC37i t * 3 d. The name and address of each source from which your company obtained FCB compound's^ or mixtures, and the _ amount of each FCB compound or mixture obtained from each source. e. For each method of reclamation, the location of every reclamation site, the name and address of each party Involved *in the reclamation of each FCB compound or mixture, and the amount of each FCB compound or mixture reclaimed by each party. 5. For each FCB compound or mixture disposed of by your company (with or without the Involvement of other parties), for each company facility, during each year of 1971, 1972, 1973, 1974, and the first two quarters of 1975: a. A description of each method of disposal. b. For each method of disposal the total amount of each FCB compound or mixture disposed. c. The total amount of each FCB compound or mixture disposed. d. The name and address of each source from which your company obtained FCB compounds or mixtures, and the amount of each FCB compound or mixture*obtained from each source. A e. For each method of disposal, the location of every disposal site, the name and address of each party Involved in the disposal of each FCB compound or mixture, and the amount of each FCB compound or mixture disposed of by each party. 6. The composition by chemical name and percent by weight of each PCB compound or mixture jjjg^uce^, imported, sold, reclaimed, used, and/or disposed of by your company since J a n u a r ^ ^ ^ ^ T ^ . 7. The results of any and all sampling and analysis performed by your company, its agents or contractors since January 1, 1971, concerning the .following: a. Concentrations of PCB compounds or mixtures In the effluent of any discharges by the company (Into vatara of the United* States or Publicly Owned Treatment Works) or In the emissions of the company into'the air. 4CP CC6373 4 b. The flow and/or composition of any discharges or emissions by the company (into waters of the United States or Publicly Owned Treatment Works, or in^o the air) which contain PCB compounds or mixtures. c. Concentrations of PCB compounds or mixtures in receiving waters (both upstream and downstream) of any discharges by the company which contain PCB compounds or mixtures and concentrations of PCB compounds or mixtures in the air in the area of the company. 8. The methods by which PCB compounds or mixtures are transported to, by, and/or from your company, including: a. A description of each method of transportation, and the form in which PCB compounds or mixtures are trans ported by each method. Where different methods of trans portation are used at different facilities specify which transportation method Is used at each facility. b. The names and addresses of all known transporters of PCB |1 compounds or mixtures. 9. All occasions (including spills) of which you sre aware on which PCB compounds or mixtures have or may have been Introduced into .the environment. In particular, describe such occasions Insofar as they involved PCB compounds or mixtures in their liquid state, as incorporated into closed systems, or as Incorporated into open systems. For each occasion, indicate name and address of party involved; dates, time, and location of the discharge or spill; and the amounts involved. 10. A description of sny adverse health or environmental effects which you know or believe to have resulted from the introduction of PCB compounds or mixtures Into the environment. Indicate any specific occasions including dates, times, locations, amounts and parties Involved for which such effects are known. 11* Any and all other information which you possess concerning: a . The production, importation, reclamation, use, distribution, and disposal of PCB compounds or mixtures. v AC* CC6374 b. The discharge of FCB compounds or mixtures into the environment. / In addition to the above, EPA would appreciate receiving any other information which you possess concerning the distribution and discharge of PCB compounds or-.mixtures by sources other than your company, Including your company's customers and sources. \\ ACM GC6375 ? August 1975 Mr. Stanley V. Legro Assistant Administrator for Enforcement Environmental Protection Agency Room 1111 - Vest Tower EO-329 Ath and M Streets, N.W. Washington, D, C. 2OA60 Dear Mr. Legro: In accordance with your request of yesterday and In the national Interest, X violate our and Industry's usual practice by handln? you herewith a copy of Monsanto's present customer list with respect to its sale of PCBs. As I indicated In our two telephone conversations, Monsanto, about three years ago, voluntarily eut back Its production of PCBs and terminated Its tales of such produots to all exoept those who 1) could satisfy us that they are using PCBs In closed systems and 2 ) either executed Indemnification agreements In favor of Monsanto or obtained liability Insurance satisfactory to Monsanto. The last page of the list, which sets forth four company names, has to do with thoss customers whose Insurance has expired ar.d, therefore, are not now able to buy PCBs from Monsanto, but who presumably will be able to satisfy Monsanto's Insurance require ments and rssume purchases sometime In the future. Because we do not believe it appropriate to reveal details con cerning customers of products which we make and sell, we are particularly sensitive to the transmission of this customer list. Ve, therefore, request that It be handled with care and not used with the media or In any other fashion that might Identify it with Monsanto or cause the ccnpany any type of embarrassment. v \ Sincerely, t l* I^ IN T IF F ^ Edwin J. Putxell, Jr A L * CCfc7t bec: H. S. Bergen - B2S P, Cunningham - W. B. Papageorge - March 3 1972 Mr* Pan! DsPsloo, Jr. Regional Adminlatxator 0*3. Snvironmental Protection A g n e y Region SC 100 California I treat Saa Francisco, California 94111 Dear Mr. DePaloot Z have been ashed to respond to your letter addressed to Hr* B. J. Bock, asking for oertaia information regarding sales of po 1yd) lorlasted biphenyls by Monsanto and Central Solvents and Chemicals in the Los Angeles County area. Z know that you have dis cussed this by telephone with Nr. V. B. Papageorge and Z have revisesd the situation with him. Zt has always been Monsanto's policy not to divulge the names of our customers for the products we make, nor the quantities of products these customers purchase. Me consider this to be proprietary informatlon and therefore hold it confidential. Consequently we do not feel free to comply with your request fox data on customer names and quantities purdiaeed. On the other hand, last Koveufcer we did release to governmental Agencies, Including the EPA, Monsanto's total production and sales figures on P 9 ' s for the years 1960 through 1911. These statistics have been released publicly end earlier this year we discussed them la detail with members of the scientific P L A IN T IF F 'S ^ EXHIBIT-;^ Cf* C G 8 3 7 7 Mr. Paul DeFalco, Jr. March 3, 1972 Pa? 2 . . / ccnffounity and with representative of the Washington multi-agency PCB task fore of which EPA is a member. Por your ready reference I era enclosing a copy of this infox&ation. We certainly share your concern and desire to control the discharge of PCB'a into the environment. It is because of this concern that we implemented a program of self-imposed curtailment of sales of these products so that now we are only offering continuing supplies for closed system dielectric applications where the potential danger of environmental contamination is carefully controlled. Furthermore, as another step in safeguarding the environment we have informed all customers of PCB materials of the potential dangers should these materials escape and have strongly urged them to take every precaution to prevent loeses through spills, usage, leakage; disposal, vaporisation or otherwise. I regret we cannot comply with your request for customer information but if we can be of assistance to you in other ways we will be glad to visit you or your designees to discuss the PCB situation further. Sincerely, W. R. Corey, Director Functional Product Groups / jf Enclosure cc: Mr. Charles H. Soomer/Tionsantc v AC* GC637B Monsanto FOR RELEASE IMMEDIATELY 1971 MONSANTO INDUSTRIAL CHEMICALS COMPANY E . V . John P U B L IC R E L A T IO N S 800 N. Lina borg."i 3c_ e*a'b St. Louis. Missouri 63*30 (314) 694-2891 (314) 567-3628 (Horte) MONSANTO RELEASES PC3 PRODUCTION FIGURES TO DEPARTMENT OF COMMERCE WASHINGTON, D.C., Nov. 30 -- Monsanto Company announced today it has released figures to the United States Department of Commerce showing the company's U.S. production and sales of polychlorinated biphenyl (PCB) from 1960 through estimated 1971. Closed-system electrical uses ranged from 11,000 tons per year to 20,000 tons per year during the period. The report also shows Monsanto voluntarily reduced its nonelectrical sales of.PCB from the high of 16,000 tons in 1970 to 4,400 tons in 1971. Substantial further reduction will be accomplished in 1972, the company said. C. P. Cunningham, managing director of Monsanto Industrial Chemicals.Co., an operating unit of Monsanto, said, "As part of our control program, the data were originally prepared for the Federal PCB Joint T$sk Force and other concerned scientists \ for use on a confidential basis so scientists can properly evaluate the figures and relate them as to impact on the environment." -more- ADK CC637S -- 2 MONSANTO: RELEASES P.CB PRODUCTION FIGURES xxx environment V II Monsanto has now cho&en to release the figures publicly "because scientists have continued to express a need for the data to make a proper estimate of how much PCB might be in the environment. We also believe the figures will inform the public about Monsanto's positive actions to limit the sale of PCB to confined applications," the company executive said. "Now that the figures are public information," ~~ Cunningham added, "we are confident both the scientific community and the public press will Interpret the data properly. These are production and sales figures only. It should be clearly understood that all this material is not in the I* environment. "For example," Cunningham said, "over 60 per cent of all sales in the past 12 years have been for closed-system electrical uses. It would be a completely false and irresponsible conclusion to say all PCB produced is in the environment. Much of the material Is still contained in sealed electrical equipment where it acts as a fire-resistant safety IIIII fluid, protecting human lives and property from the danger of fire and explosion." v\ '-more- A C Cb3eL * I -- 3 MONSANTO: RELEASES PCB PRODUCTION FIGURES xxx explosion." / Cunningham concluded,' "Monsanto continues its program to control the uses and prevent the release of PCB into the environment. To-date, we have cut off all sales.to paints, plasticizers, sealants, adhesives and all other open-system uses. We are receiving used PCB from customers for regeneration or safe disposal. And we are engaged in an all-out research effort to develop less persistent forms of PCB or suitable products which retain the fire safety features of PCB. We call' on other producers, worldwide, to join in this effort." -0O0 NOTE TO EDITORS: PCB production and sales figures attached. PRODUCTION AND SALES POLYCHLORINATED BI,PHENYLS (PCB) MONSANTO INDUSTRIAL. CHEMICALS COMPANY (SHORT TONS) YEAR U.<S. PRODUCTION DOMESTIC SALES Electrical* All** Insulating Other Liquids Uses Uses U.S. EXPORT SALES All Uses 1971 (Est) 1970 1969 1968 1967 1966 1965 1964 1963 1962 1961 1960 * 20236 42527 38194 41427 37655 32925 3024Q 25417 22367 20986 20333 20786 14417 20268 18564 20568 20387 18897 16203 13769 11448 11683 11108 12444 4401 16263 ISO34 11990 10846 10642 9695 8666 7618 7339 7661 5163 4938 6826 `5312 5616 4062 3426 2117 2048 1824 1809 2076 1827 * Closed Systems - Transformers and Capacitors ** Including Heat-transfer Systems, Hydraulics/Lubrlcants, Plascicizers PCB MANUFACTURE AND SALES MONSANTO INDUSTRIAL CHEMICALS COMPANY 1957 Thru 1964 (Thousands of Pounds) 1957 1958 1959 1960 1961 1962 1963 1 1964 U.S.PRODUCTION DOMESTIC SALES (LBS.) U.S. EXPORT SALES (1) 32299 (2) DOMESTIC SALES BY CATEGORY Heat Transfer Hydrau lj.es/Lubricants Mise, "industrial Transformer Capacitor Plasticizer Applications Petroleum Additives --. 1612 704 12955 17028 (1) - DOMESTIC SALES BY PCB GRADE Aroclor 1221 Aroclor 1232 . Aroclor 1242 Aroclor 1248 Aroclor 1254 Aroclor 1260 Aroclor 1262 Aroclor 1268 Aroclor 1016 23 196 18222 1779 4461 7587 31 -- (1) 26061 (2) 1549 755 5719 14099 3939 - 16 113 10444 2559 6691 5962 184 72 -- (1) 31310 (2) 37919 35214 (2) - 2685 1569 5984 16499 4573 - - . 2523 1559 7921 16967 6244 - 254 240 13598 3384 6754 6619 359 102 -- 103 155 16196 2827 6088 7330 326 189 -- 36515 37538 (2) 38353 38043 (2) 44734 38132 .\ 3647 50833 44869 4096 - 157 4110 3915 2114 1681 6281 7984 15935 15382 9098 . 8924 -- 94 241 19827 4023 6294 6540 361 158 -- 140 224 20654 3463 6325 6595 432 210 _ 582 929 3945 4374 1528 1692 7290 7997 15606 * 19540 9181 10337 - N_ 361 13 18510 5013 5911 7626 414 284 -- 596 13 23571 5238 6280 8535 446 190 (1) Production figures and Plasticizer Applications figures unavailable during year indicated. (2) U.S. Export Sales figures unavailable during year indicated. -t J- PCB MANUFACTURE AND SALES MONSANTO INDUSTRIAL CHEMICALS COMPANY 1965 Thru 1971 + 1972 Estimate (Thousands of Pounds) 1965 1966 1967 1968 1969 1970 Eat . 1971 1 1972 U.S.PRODUCTION DOMESTIC SALES (LBS.) U.S, EXPORT SALES 604B0 51796 4234 DOMESTIC SALES BY CATEGORY Heat Transfer Hydrau^lics/Lubricants Mise:" Industrial Transformer Capacitor Plasticizer Applications Petroleum Additives 1237 4616 1841 8657 23749 11696 - 65849 59078 6852 1766 4258 1779 8910 28884 13481 - 75309 62466 8124 l 2262 4643 1426 11071 29703 13361 - 82854 65116 11231 2529 5765 1283 11585 29550 14404 - 76387 67194 10624 3050 8039 1079 12105 25022 16460 1439 85054 34994 73061 34301 13651 \ 3950 7403 1627 13828 26700 19537 -- 3060 0 1552 0 1155 . .0 11134 (25000 to 30000 14141 (total dielectr 3259 ' 0 - N0 DOMESTIC SALES BY PCB GRADE Aroclor 1221 Aroclor 1232 Aroclor 1242 Aroclor 1248 Aroclor 1254 Aroclor 1260 Aroclor 1262 Aroclor 1268 Aroclor 1016 369 7 31533 5565 7737 5831 550 196 528 16 39557 5015 7035 5875 768 284 442 25 43055 4704 6696 6417 840 287 136 90 44853 4894 8091 5252 720 200 507 273 45491 5650 9022 4439 712 300 1476 260 48580 4073 12421 4890 1023 330 2215 171 21901 213 4661 1725 1 - 3334 300 300 4000 0 6000 600 0 0 17000 i ) AROCLOR'APPLICATIONS CURRENT USES Electrical Capacitors: Aroclor 1016*, the major product used; Aroclor 1221; Aroclor 1254; very limited usage. Electrical Transformers: Aroclor 1242; Aroclor 1254; Aroclor 1260. * Similar to Aroclor 1242 with 5 chlorine and higher isomers removed. FORMER USES Electrical Capacitors: Aroclor* 1242; Aroclor 1254. Electrical Transformers: Aroclor 1242; Aroclor 1254'; Aroclor 1260. Heat Transfer Systems: Aroclor 1242; Aroclor 1248; Aroclor 1254. Vacuum Pumps: Aroclor 1248; Aroclor 1254. Gas Transmission Turbines: Aroclor 1232; Aroclor 1242. Hydraulic Fluids: Aroclor 1232; Aroclor 1242; Aroclor 1248; Aroclor 1254; Aroclor 1260. Plasticizers in Synthetic Resins: Aroclor 1248; Aroclor 1254; Aroclor 1260; Aroclor 1262; Aroclor 1268. In Adhesives: Aroclor 1221; Aroclor 1232; Aroclor 1242; Aroclor 1248; Aroclor 1254. In Surface Coatings: Aroclor 1221; Aroclor 1248; Aroclor 1254; Aroclor 1260; Aroclor 1262. ` A C ? CC 3 6 AROCLOR APPLICATIONS FORMER USES Continued . Plasticizer in Rubbers: Aroclor 1221? Aroclor 1232; Aroclor 1242; Aroclor 1248; Aroclor 1254; Aroclor 1260. Wax Extender: Aroclor 1242; Aroclor 1254; Aroclor 1268. Dedusting Agents: Aroclor 1254; Aroclor 1260. Pesticide Extenders: Aroclor 1254. Lubricants, Cutting Oils: Aroclor 1254. Carbonless Reproducing Paper: Aroclor 1242. Inks: Aroclor 1254. \ 0 CC638t * UNITED STATES environmental protection agency REGION IX lOO C A LIF O R N IA S T R E E T SAN FR A N C ISC O . C A LIFO R N IA 94111 Mr. .J. Bock President and Chief Executive Officer Monsanto Company 800 IU Lindbergh St. Louis, Missouri 63166 Dear Mr. Bock: v Information available to this office indicates that quan tities of polychlorinated biphenyls are being discharged to the Pacific Ocean via the Los Anceles County Sanitation District's sev.-erage system. In order to take scraps to brine about a reduction in this discharge, it is necessary that the points of entry of this material into the sewerage system be determined. To assist us in making that determination, it is requested that you provide us with the following information:" 1. A list of all shipments of polychlorinated biphenyls by Monsanto to Los Angeles County, California within the past eighteen (13) months with the name and address of each addressee as well as the quantity delivered. 2. A similar listing of all sales of polychisrir.a biphenyls made by the Central Solvents and Chemicals Company, Los Angeles, California. We would hope to hear from you in this regard within the next two weeks. Your cooperation in responding to this request will be appreciated. Sincerely, July 1, 1977 Ms. Kaye Jacobs Attorney United States Environmental Protection Agency Region V 230 S. Dearborn Street Fourteenth Floor Chicago* Illinois 6060U Rex Letter Request for Information* Dale S* Bryson to John R* Eck, June 6* 1977 Dear Ms* Jacobs: - In the above referenced letter the U* S. EPA asked for certain information concerning operations at Monsanto's William G. Krucmrlch Plant* located at Sauget, Illinois* This letter was subsequently discussed and clarified during our telephone conversation June 21* 1977 The Information contained herein is provided to you in response to said letter and discussion* Beginning July 1, 1977 for a period of six months we will monitor the Incinerator discharge and the combined discharge* taking a 24-hour vertically Integrated composite sample once a week at each location. A substantial rehabilitation project is underway concerning the plant's sewer system* A description of this project is as follows* The main operation area of the plant occupies an area of approximately 100 acres* All effluent discharges from the plant operating departments and facilities are discharged into a complex network of sewer systems* This sewer system drains approximately 140 acres* A map of the plant premises and sewer system is attached* ALK CC6<t87 M. Kaye Jacob July 1, 1977 - Page Two The plant sewer effluent are collected Into small service sewers, transferred to larger laterals, and then finally" discharged into three large trunk lines which are part of the Village of Sauget sewer system. Wastes . from the entire Village of Sauget flow to the Village of Sauget Treatment Plant, and af'.er treatment are dis charged to the Mississippi River, Both domestic and Industrial sewage are handled by this system. The attached map divides the plant into five main areas of drainage. Areas 3 and 4 are the main operating areas of the plant. An examination of the sewer system indicated that the sewers servicing area 3 were deterio rated and in need of replacement or repair. Accordingly a major project was initiated to perform the necessary repairs or replacement of this system. No major leaks, failures or deterioration were discovered in the sewers servicing area 1, 2, 4 and 5. The rehabilitation project replaces tho existing trunk sewer for section 3 by extending two existing sewers northward along "g " and "E" streets plus the installation of a new sewer along "D" street. The existing usable ewer lines will be tied into the new system. Sewer lines being replaced will either be excavated and removed, or filled with grout and abandoned in place. The old sewers which are being replaced are shown in light blue on the attached drawing, with the new sewers shown in red. New ewer piping will be one of these three types: a. Ductile iron soil pipe, schedule 40 b. Reinforced concrete culvert c. Vitrified clay pipe with a reinforced concrete encasement. All Joints will be sealed with furan resin and poured sulfur. New manhole will be acid proof brick lined, with an average box depth of 11 feet below grade. The project will include installation of approximately 2,473 feet of vitrified clay pipe, sizes 6' diameter through 36" diameter; 230 feet of ductile iron sewer ACM CC6^3ti Ms* Kaye Jacobs July 1, 1977 * Page Three pipe, 4" through 12" diameter; and 600 feet of rein forced concrete culvert. Additionally, repairs will 'be made to other existing sewer lines as necessary :to upgrade the entire system to "like new" condition. `Nineteen new manholes will be constructed, and several existing manholes will be repaired. Total cost of the project is estimated to be $2,450,000. Work commenced in January, 1977, and is expected to be completed by December, 1977. This project does not directly affect the sewer system servicing the Monsanto PCB production facility, located In Area 4 on the attached nap. Effluent from this facility flows south on the map across Area 4, and then west in a 24" trunk line leaving the plant and flowing into the Village of Sauget sewer system. As far as we are aware, these sewers have no major leaks or failure points and are adequate for present and future use. The sewers within the PCB manufacturing area will be disconnected and sealed off as part of the PCB shutdown program. The above referenced letter requested information concerning the total amount of PCBs currently being stored on site, plans for distribution of any PCBs not distributed by October 31, 1977, and disposition of PCB contaminated equipment. As I pointed out to you during our discussion, the EPA Office of Toxic Substances will be promulgating regulations in the near future on the handling and disposal of PCBs. We will, of course, comply with those regulations. At this time the amount of waste PCB awaiting incineration at the plant varies from around 1,000,000 to 1 ,500,000 pounds. It is contained in storage tanks at the incineration location. The amount of PCB product inventory on hand averages around 1,500,000 pounds. It is contained in storage tanks in the production department and in tank cars. Some drum inventory of product is maintained in protected indoor warehouses. We plan to incinerate any PCBs which have not been distributed on sor before October 31, 1977. As mentioned during our discussion, we plan to deal in an orderly manner with equipment which has been used in the pro duction and handling of PCB. Such equipment which can be adequately cleaned will be reused internally for industrial type chemicals. Nonreuaable equipment, including all remaining ACH CC B 39 Ms. Kaye Jacobs July 1, 1977 Page Four * ` processing equipment and piping, will be drained thoroughly to remove PCBs. Liquids will be incinerated. All equipment will be"sealed and buried in an approved landfill. After all equipment has been removed, the production site will b# cleared to approximately one foot below grade level. Sewers will be disconnected and sealed off. All contaminated rubble will be removed to an approved landfill for proper disposal. All washings used in the decontamination will be treated to remove PCB and/or will be incinerated. In summary, we plan to proceed in an orderly and environmentally sound manner in closing down our PCB operations. We hope that the information provided above satisfactorily responds to your questions. Please direct any further inquiries on this subject to me at the above address. * Yours very truly, it Enclosure bcc: R. E. Bilger J. R. Eck R. W. Flint R. L. Harness A. E. Leisy J. W. Molloy M. A. Pierle Phocion S. Park Environmental Counsel AL i* CCt*1 c bcc : 0 Wr B. PaDage' orge W. R. R i c h a r d R. H. Munch 0. Wood H. S. Ber gen May 12. 1975 Dr. Leonard J. Guarrla USEPA W aterside M a l l , East Tower 401 M S t r e e t , S.W. Washington, D.C. 20460 Dear Dr. Guarrla: We w a n t t o t h a n k y o u f o r a l l o w i n g M r . P a p a g e o r g e and I t o p r e s e n t o u t c o m m e n t s on t h e M a r c h 13 v e r s i o n o f t h e PC3 c r i t e r i a document fo r the Toxic P o llu ta n t E fflu e n t Stan d a r d s . We e n j o y e d o u r d i s c u s s i o n s w i t h y o u and M r . L e w i s and look forw ard to a c o n t i n u i n g exchange of p e r t i n e n t 1nforraat1on. On my r e t u r n t o S t . L o u i s I l e a r n e d t h a t two o f o u r p a p e r s on b i o d e g r a d a t i o n of p o l y c h l o r i n a t e d b ip h e n y ls have r e c e n t ly been accepted f o r p u b l i c a t i o n . For your In fo rm a tio n th e t i t l e s o f t h e s e a r t i c l e s and J o u r n a l s 1n w hi ch th ey w i l l appear are as f o l l o w s : 1. ' D e g r a d a t i o n o f P o l y c h l o r 1 n a t e d B i p h e n y l s by M i c r o O r g a n i s m s " by B a x t e r , et a l , S c ie n c e of the T otal Env1ronment. 2. ' A c t i v a t e d S l u d g e B i o d e g r a d a t i o n of P o l y c h l o r i n a t e d B i p h e n y l s ' , by T u c k e r , et a l . B u l l e t i n o f E n v l r o n mental C on tam in atio n and T o x lc o lo q y , Volume 14, IT s 'ue I T .------------- ------------------------- The f i r s t r e f e r e n c e 1s t h e same a s t h e d r a f t we l e f t w i t h you d u r i n g o u r v i s i t . The second 1s a c o m p i l a t i o n of data p r e v io u s ly su p p lie d to Mr. Lewis d e sc rib in g our semicontinuous activated sludge studies. We w i l l c o n t i n u e t o k e e p y o u I n f o r m e d o f a n y new d e v e l o p m e n t s from our l a b o r a t o r ie s re gardin g the environmental s i g n i f i cance of PCBs. S i n c e r e 1y y o u r s , -Or. James P. M l e u r e Research Group Leader db PLA IN T IFF'S EXHIBIT ACM C C 8 ^ ^ b c c M. A. P i e r l e p. S. Park A. E. L e i s v - 1740 C. F. B uckley - 1740 June 14, 1976 Mr. B lake B ile s E nforcem ent D iv is io n (EN-342) E nvironm ental P r o te c tio n Agency 401 M S t r e e t , S.W. W ashington, D.C. 20460 D ear Mr. B i l e s : T h is i s in re sp o n se to Mr. S ta n le y L e g ro 's l e t t e r o f May 2 9 , 1976 c o n c e r n i n g t h e E?A i n s p e c t i o n r e p o r t o f o u r W. G. K rum m rich p l a n t . We h a v e e x a m in e d t h e d a t a e n c l o s e d w i t h Mr. L e g r o ' s l e t t e r and fin d an e r r o r in the c a lc u la tio n s used to determ ine th e a i r lo s s e s from sto ra g e tank em issio n s. The v a p o r p r e s s u r e o f A ro c lo r 1242 i s 0.35mm a t 100C, n o t 3mm a t 100C a s u s e d i n t h e r e p o r t . T h i s means th a t th e c o r r e c t s to ra g e tank em issio n s should be 70.6 K g/year, not 597.4 K g/year. These c o rre c tio n s apply to p ag es number 509 and 548. The o th e r in fo rm a tio n p r e s e n t a p p ea rs to be c o r r e c t . S incerely, J . Coleman Weber Manager Product A cceptability u " ! O' A C H C C 6 2 6 fc **'*'* UNITED STATES ENVIRONMENTAL PROTECTION AGENCY W ASHINGTON. D C. 20460 2 9 MAY OFFICE OF ENFORCEMENT M r. W. B. P a p a g e o r g e JUN 4 S 7 Manager, Product Acceptability P rocess Chemicals Division Monsanto Industrial Chem ical Co. 800 N. L indbergh B o u lev ard St. L o u is, M is s o u ri 63166 Dear Mr. Papageorge: An Environm ental Protection Agency (EPA) inspection team recen tly visited a facility operated by your company to obtain in fo r m ation concerning the handling of PCB m a te r ia ls . The in sp ectio n w as conducted under a u th o rity of sec tio n 308 of the F e d e r a l W ate r P o llu tio n C o n tro l A ct (F W P C A ). a s a m e n d e d , 33 U. S .C . 1318, and s e c tio n 114 of the C l e a r A ir A c t, as a m e n d e d , 42 U . S . C . 185TC-9. E nclosed is a copy cf the plant v isit re p o rt p re p a red by the in s p e c tion te a m which r e c o r d s its findings with r e g a r d to the handling of PCB m aterials at your facility. It is possible that EPA will receive public requests for re le ase of the inform ation contained in the enclosed re p o rt. Such re q u e sts w ill be handled by E PA in a c c o rd a n c e with p ro v isio n s of the F r e e d o m of Inform ation A ct (FOLA), 5 U .S .C . 552, and E PA reg u latio n s is s u e d th e re u n d e r, 40 CFR P a r t 2. EPA is re q u ire d to m ake this in sp ectio n r e p o r t available in re s p o n se to FOLA re q u e s ts u n le ss the A d m in is tra to r of this Agency d e te rm in e s that the re p o rt contains in fo rm atio n c h a r a c terized as confidential com m ercial inform ation or methods and p ro cesses entitled to protection as trade s e c re ts . Pursuant to the FWPCA and the Clean A ir Act. effluent and em ission data m ay not be considered confidential inform ation or trad e s e c re ts subject to exem p tion f r o m m a n d a to ry d is c lo s u re r e q u ir e m e n ts of the FOLA. \ ACf CCJfcS PLAINTIFF'S EXHIBIT 2 In o rd e r to facilitate the A gency's tim ely resp o n se to any public in q u irie s, while giving due consideration to your com pany's right to req u est confidentiality, please provide the following information: 1. N o tic e of a n y f a c tu a l i n a c c u r a c i e s th a t you find in the rep o rt which you believe should be corrected. Although the rep o rt itself cannot be altered or m odified in responding to FOIA re q u e sts, we would appreciate your comments concerning any facts which you believe to have been reco rd ed inaccurately. 2. Any inform ation, except effluent and em ission data, contained in the report which you believe to be com m ercial confidential inform ation or tra d e s e c re ts entitled to exem ption fro m the m a n d ato ry d is c lo su re re q u ire m e n ts of the FOIA. Your com m ents should be addressed to Mr. Blake B iles, Enforcem ent Division (EX-342), Environm ental Protection Agency, 401 M S tr e e t, S. W. , W a sh in g to n , D. C. 20460, and should r e a c h this a d d r e s s not l a t e r th a n 14 d ay s a f t e r y o u r r e c e i p t of th is l e t t e r . F a i l u r e by your f ir m to su b m it w ithin that p erio d of tim e a w ritte n re q u e s t that inform ation be characterized as confidential or privileged will be treated by EPA as a w aiver by your company of any claim s fo r confidentiality regarding inform ation contained in the PCB investigation report. Thank you for your cooperation in responding to our inquiries concerning PCBs. Enclosures for Enforcement AC* CC63S UNITED STATES ENVIRONMENTAL PROTECTION AGENCY I SUBJECT: FROM: PCB Inspection, Monsanto Krunnrich Plant, Sau:at,"Illin o is. i M u date: March 9, 1 Chief, Air Surveillance Branch, Surveillance and Analysis Division l TO: 'Karl Brerr.er, Member, Lake Michigan Toxic Substances Committee I A preliminary survey was made of the above installation to identify and sample point sources of potential PCB emissions to the atmosc/.=.*s. The sampling and analysis were carried out at this source to furrisinformation to arrive at a further decision that may require an incersivtesting program to quantify the PCB emissions. The sample analysis revealed the following: SAMPLING LOCATION NO. OF SAMPLES PCB concentrt::*: - Steam Ejector - Ground Level, Site #1 2 * 3A - 27.5 ' 3B - 33.9 Steam Ejector - Top Level, Site r2 2 3A - N.A. 3B - N.A. Calculated Storage Tank Emissions Virgin Product Product Loading and Waste Storage 597.4 Kg/year 0.02 Kg/year Monsanto has performed a stack test at the Sauget location quantify* the PCB emissions to the atmosphere. Stack-sampling and other r e o c rts describing process and sampling methodology used at this site are attache * 3A - Right Side of sampling train - 3B - Left Side CPA f+rm 1320-4 (R#*. 4-72) A Cf*. CC63S1 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY SUBJECT: Emissions of PC8 from the Incinerator at Monsanto - d a t e : February 1 1,-I Sauget, Illinois. f r o m : R. Edwin Zylstra, Technical Advisor, Special Projects Section Tn- Chief, Air Surveillance Branch On January 27-28, 1976 sampling was conducted at the Krummrich Plant of Monsanto in Saucet, Illinois to determine emissions to the air. The samples were taken rear the exits of the steam je t ejectors of the non-fractionating distillation process. The exits of the fractionating distillation steam ejector were not sampled because of inaccessibility. The incinerator used to destroy waste PCB's was not sampled because Monsanto had tested the stack and the results were made available to us. ! I have reviewed the test procedures and the data from three stac! tests and I feel that the test results accurately indicate the quantity of PCB being emitted into the atmosphere from tne incinerator. The !1a ^ 3 i* procedure for stack testing the incinerator for PCB and particulate is shown in ['psf'11y 'Vf !-- -- A A." The metned is O r I EPA Method z : T i'C " ' half is identical to Method 5 ar.d consists of a sample nozzle, a hss-:' probe, and a heated glass'fiber f ilt e r . The back half is as folic., s: First impinger empty; 2. Second impinger with 150 ml. of N, M'-dim.et'-. formamide; 3. Third impinger with 75 ml. KaOH solution and 75 ml. Na solution; 4. Fourth impinger with 200 g silica gel. Sampling is performed the same as Method 5. To determine the PCB emissions the sa: is handled as follow: T. Probe, nozzle and f ilt e r are rinsed with hexane. The rinse is mixed with any condensate from the first impinger and then analyze for PCB by means of electron capture gas chromotograph. Z. The NN dimethy formamide in the second impinger is analyzed for PCB by means of electron capture gas chromotograph. The table shows the results of stack tests for PCB on the incine^: performed by Monsanto. Emission ranged from .COCCI to .CCC243 lbs/nr. Computer printouts of data collected and the results of the last three te: performed were supplied to us- for evaluation. The calculations vers cm*: and no errors werefound. Monsanto*^ and our computer printouts are in Appendix B <r CPJU/i^-l 370-6 (R... AO CCciSi  \ iA \\ 'i 4 . -Vv 1 * .V - . 1*. :vT .*r `Staci: tests on >:o::sA:rro rcn iiiciKinATOn The fo lio -./in g t e s t s w ere r.ade u s i n g a rr.ot: r : __ Method 5 and b o th fiov; and c o n c e n t r a t i o n s verc CC` n PCB cone ... Date in nc/n^ Flow n 3/.n: , * * 7/25/73 I"' 8/7/73 ..8/16/73 V 8/24/73 *: ' - 10/2G/73 V , *11/7/73 : 11/20/73 *V 11/28/73 . V 8/22/74 9/12/74 . 5/23/75 io/:/75 0.0063 0.0167 0.0176 0.0165 `0.0077 . 0.0056 0.0119 _ 0.0013 *0.0149 . 0.0289 0.0300 0.0ig2 * V * %" *,> * t . 63 94 106 106 48 58 63 66 71 51 44 52 1 . % . > . ; .l V vv . h/ 2. 3 7 2. 3 '1 a ui '/ *;// /.'/ /. * 5 /.? 2.i T ALf* C C t I S 2 . APPENDIX A MONSANTO PROCEDURE FOR STACK TESTING FOR PC3s ACH CC63S^ y O' INCINERATOR STAC;: TES? OA O SCO'E: This procedure determines the amount of PCB's lates present in the incinerator stack gas. 1 . R. A .C. portable sta ck gas sampler a. sample box b. unbilical cord 25 feet ( 2 each) c. d. nozzles 0 .250 or 0.375 clamps ( 1 0 each. ) 23/15 Fisher 5-rQoo_? e. Z>1 combine cion probe and pi cos tube is-.a f. g. glass conn ector s (4 eaefc) 500 ml Ore cnburg-Snlth Impinger it ni each) Fisher Sci entif ic Company, St. Loui s, -sJ cat, no. 9 -257, or ecu ivalent. Tn e *tu impinger i o mod ifled `by repiici T*"*--o *u" standard tip extending to with a 1 /2 inch 1 0 on.e-half inch from gla bot S3 tU i.w_..,1 0 Impinger flask. 2 . Gla ss orifice to regulate air flow* thro .'-.Ui r tub e. ( 2 each) Can be made ..y Vw. oAr*.,5 r. whcrr"- 3. Thermometer 0-220? _ * 4. Thermocouple and potentiomete: -5. - 6. 7. *8. Nomograph Clip board Air Pump (Dyna pump) Stopwatch (2 each) : 9. Tape ruler * . 10. 3 way plug * . 1 1 . Stopcock grease * . f . 12. 50 ft. of 5/8" rope V ?. . Adjustable wrenches ( 2 each) PARTICULATE TEST ADD: ` 1. Cyclone and Erlcnmcyer flask* * 2. Clamps 2 each 28/15 ACh CCb3S5  - i* Glass fiber filter holder' 4. Glass fiber size 7 . 0 C'A (Piehcr cat. i * Dry in oven and leave in ccciclator. i i * 1 . ' . % AC* CCfciSt V a. Sodium Sulfite, Fisher Scientific Company, St. Loui Cat. No. S-*r^0 in unstar.dardized solution. Weight dinsulve in v/atcr and make up to 1 ltcr. b. N,N-Dinethyl Formamide, Fisher Scientific CO., St. Lou Missouri, Cat. No. D-119. c. .Sodium Hydroxide 50 solution, Fisher Scientific Com pa St. Louis, Missouri, Cat. No. SO-S-25^, in unstandardi solution. Add 52.4 ml to 500 ml distilled water and n 'to a lcter. d. Hexane; nono^rade (Mal Under odt Chemical Works, St.- Lc Missouri, Cat. No. 4159) I e. Silica Gel Grade 933 Darison Code 933-03-03-226 Lav^sc Chemical; Baltimore, Maryland, 21203. I f.' Crushed ice. \' . At^ CC3S7 I SAKPLIi.'O ?nOC"3'jnS: 1. Measure stack diameter and determine minimum number cf sam-.li.points needed. :. 1I 2. Measure stack temperature ; X i ;. 4 .3 . Determine percent moisture. This may be found from previous tests. i> * : Place sample box on duorail and assemble imoir.gers. Leaving the first impinger empty. Pill the second impinger with 1 3 3 ml of DM? solution. Add 75 ml in NaCK plus 75 ml in ::2213j ir the third impinger. The last impinger v/ill contain 233 g silica gel in a modified impinger. (Weigh silica gel for uati determinatio.n later). *; - ' & : ... 5. ' Place the cyclone and glass fiber filter in the heat chamber and connect to impingers. >. Connect vacuum line of umbilical cord to inlet of meter box and outlet of las.t impimger. Plug up the inlet to the filter holder and pull 15 in. Kg vacuum to check fer leaks. Leakage rate should not exceed C.C2 C.P.M. "7. Connect glass probe- to cyclone. : ` .:8. Complete all connections on umbilical cord. i ' 19. Purge air through pitot tube using tubing which is attached to umbilical cord and connect to air pump. ; ' 10. Pack ice around impinger and add water. ! 1I 4 I. ai. ! ' 112. 1 1 i ' U4. 1 as. Make a traverse*of stack on one side to determine velocity ari 'stack pressure. Ee -sure you have balanced draft gage before you take your readings. Heat probe and chamber in box to a temperature of 10 higher than the stack temperature. Record temperature at gas meter, (inlet and outlet.) Use Nomograph to determine nozzle size and isokinetic sampling Tate. Record gas meter readings. .; m '6. As 'soon as heat chamber in sampling box is up to temperature *t start'your test. * / *% ' ACM C C 6 3 S 6  SAMPLE TRAIN OPERATION 1. . For each run, record the data required on the example she shown in Figure 1-2. 2. Determine the time required at each sampling point. A . minimum time of one hour is required for this test. 3. To begin sampling, position the nozzle at the first trave point with the tip pointing directly into the gas stream. Immediately start the pump and adjust the flow to isckine conditions. Sample for at least 5 minutes at each traver point; sampling time must be the same for each point. 4. Record temperature at gas meter inlet and outlet, about c minute before it is time to change traverse points.. Also .** be sure to read temperature'of last impinger. i 5. Add more ice during test to keep the temperature of gases leaving the last impinger as low as.possible .and preferat : at 70 F or less. : - 6 . Turn off pumps at the conclusion of sample run and reccmd final reading. 7^. Remove probe -from stack and disconnect from impingers. T piobe and impingers and prepare them for lab analysis. B.. Measure the increase in volume of each impinger plus the increase in weight of the silica gel to determine percent -St Rinse probe and cyclone with hexane and pass rinsing thro glass fiber filter. Mix this extract with the water (con from the first impinger. Carefully remove the filter pad plus rubber washer and dr oven then place in desicator before weighing for particul ~n r Collect all washings and give to lab for analysis. Upon receipt of analysis from b b , record results in grar - * on sheet shov/n in Figure 1-1 and complete for computer ru LjISCUSSIOjj: ^ . ACh CC63SS ^Ehc collection system in this test may also be used to deter*, other components of the stack gas. The ones of most interest *be HC1, H0 PO4 , and Cl2. To do this have the lab save all aqu .layers from washing used in preparing the solutions for PI 13 Take the aqueous layer from the first impinger (water cordon; and run for IiCl and JI3 PO4 analysis. Analyze the dimethyl ;cr Urorn the second impinger for CI2 and 1I3 ?0/. and the Na-S03 ir.-.p for Cl2. VH FT IS T 'i ACh CCc<iC 1 I ;/I *i:'.lo r of eels of data (\-') ' 2 I j .''ur.bor c^onnonciits to be analyzed #. * 3 !-/.iy r y T T T n fi- m 7 | 7 m ~ i ( H i D F i n t s n 0-12) U,:tc "(13-20) Time i7 i7 r r i7 i^ rn 7 / ix f fj i - n ~ B i r n n n ^ r i (21-l|.l|) Stack Sampled m un c m 1Z on es 10 2 J l j J in 1 ! | / | . I '! m G ('l5-ij0)Saa ree OpMo. (119-50) (5 i~56)StackDlair.(5 7-6 l)Uar. (62-6 6 )Nzzle iiii ii rrm (6 7 -7 2 )sisc'x: liuti * Im s iV . fll Process Vit lPloltfl.vl/)l!;rFl-flRl iwbltiUI I I 1 I I I I 1 l I I I I l I I I l j l (6-iJ0) Units of production per hpur * .* E'TJ/ n7| RTiTTTZTj ; \u\a \;>\ \a\Ao \ irMUJ \ilh\cW-\\i\i-\ I I i I I I i I I I I I I II- I I i I I ' (1 --5C) Sampling dene by: c !11 v-.v::\r!\o loi lA'-l/rl k-i bU/i\o\ I/.-IL-lsigisi l/k ilrli Ivk-l/il/flS (l-;n3) Sampling .Method - Indicate modifications rrm Correction Factors K l:> 0 \L F5 0 o Jl__ 3 Decree: CFntlcrade or -- I FV AAVhrenheit M Q1 ( 5 1 - 'J f0 O r i f i c e (55-58) P ito t (59-60) (Al-A1*) i'eter -. j2s 1s lnJVn 1eFrac t ons (Tdentifv Comnonrnt 7> If present bv M.V/. and nane or forni 1a) 7 i l Q e j i i i o ] f c - r n y b i v i f o i, i / i v i % \ m f r i 1 i i l n n i n . b i | n o i - it>i f f i ? tv i g i i ~ n (1-5) 02 (6-10) Ma (11-15) C02 (16-20) CO (21-2^)1120 (26-30) Z (31-3*OiWZ (35-,`3) rare of Z Moisture determination by absorption I v i a i ; Ie - F - (T / I - r G n ^ RU n i `on " Ar sdc/aCril itvci . a1n9dE T T A - Iwei^hln^. v G T . ] v 1z G 3 E E ^ pLeak test on Snr.pler D "! (l-fi)S&mplcVol. (7-11)Temp. (12-15)Press (16-23) Tare v/t.. (2^-31) Cross wt. C32-36)Ci' re.. A n a ly s is fo r Components t> a..;1n 1 V L M- ir.i. . ! 'j 2 0 H ( \ O !) c 1 ---- !i M/.Ui A i r Vi!? O 71/ O Jp _ lo -- l-IDKar.e of Cor.p. (12-15) (16-25)Tare \v r '.V t/. r \ .. f l i t * 36,.r" H,w' . 0 E b1 0 09 T|i>0 W O1O 41 0 ') / ;(?$ ,s L 0 # 0 J l f l .1 V ty r: 7 T zr i i. 1 (26-35)Cfrone Wt.(c) (3-*l7 )`How dittrcilned ` * . VJ V* ^* APPENDIX B > STACK TEST COMPUTES PRINTOUTS *\ "i ACM CC<tC2 v ' . U i . - , TO I N D U S T R I A L CHc J C A L S CO. i *-' '<UKRlCJi*PLANr SAUGtT.IL 62201. 10 NO. 63 1?I A-r :.iT `.LPT IS 197A ST AST TIKE 1:00 PM SAKPLFp- INCINERATOR SOURCE OPFrta n:)- Sl.NPLl'!G DON Y NEIL SULLIVAN ANO ED nrUKAXN 'J ` ,i I Si-Is-LJUG. METHOD-ERA METHOD ti5 DMF Am O NA2S03 IKPiNOEkS d et er m in a ti on of emission rate of I A sPaRTICULATE . ; - . :P.C.B. CiriCL . O:n3P04 ` P'%CSSS 'WEIGHT Ra T= 772.6G POUNDS PR HOUR TZST RESULT: . i /ACTUAL CFM AT /STD CFM DRY STD CFM / 1 / STACK CONDITIONS/ AT 70 DEG.F AND 29 .92 IN. HG/ STACK ^OLU ME 1903.1 .1803.8 1711.6 .1 CLST.MIST* OR GAS CONCENTRATIONi IN wET 6 DRY STACK GAS AT STD .CO;o it i . * COMPONENT A: a: B: B: V - Cs C: GRAINS l b s ; p e k LBS. PER PER MILLION 1000 CUBIC FT CUBIC FT L B S GAS 0 070U iG .1I(S3 C. 1331 0*0740 10.6611 0.1374 fl . '" o . o o o ' 0.0018 0.0000 0.0000 0.0019 0*0000 V- .0307 .0*0323 4.316 4.6176 0.0576 0.0595 MGS. PER CUBIC METER ]6?.0'*H6 170.7756 0.0289 0.0304 70.1866 73.9665 parts PER MILLION 134.9245 142-1910 ** i-.'ET ; IN u*Y G 0-0028 IN V.'ET G U *0029 IN DRY G 46-4308 tn V.ET G 43.9313 IN DRY G. D: '0.0827 11.6131 0.1554 165.2287 46.6235 IN v.'ET g D: 0.0&71 12.4493 0.1604 199.4194 49.1344 IN DriY G t W S S J O N RATE:* * l b s . per CCr^POiJflNT HOUR Mi 1 0*348 B: o . o o c ? Cs 0.4742 Vi 1.278S. GRAMS PER SEC. 0.1379 0.00-J 00597 0*1611 LBS. PER UNIT OF PRODUCTION 0*0014 0.0000 0*0006 0.0017 V AC* CC64C3 , .MONSANTO INDUSTRIAL CH.->1ICALS CO. 5'' ^ W.fi. KSUMMRICH PLANT S A U G S T . I L 6 2 2 0 1 . I D NO. 63 121 AAC DATE SSPT 1? 1974 START T I M E i : 00 PM STACK SAMPLED- INCINERATOR SOURCE OPERATION n o ! *y. I. . ; ' F I E L D DA<TA:it . '.. . . O R I F IC E CONSTANT K = 0 .5 0 6 PITOT TUbE FACTO* FS= 0.83 TEMPS. RECORDED IN F A ' DEGREES : ' METER CORRECTION FACTOR= 1.02 i - . ... ! ... .... ;; GAS A N A L Y S I S : Z=N0NE MOLECULAR WT OF Z= * ' i. MOLECULAR WT. OF STACK GAS = 2 9 . AO , : DRY MOLECULAR WT. OF GAS = 3 0 . 0 2 .0 02 N2 C02 CO . H20 Z VOL TW PW * "i*0.10ft 0.-902 0*0^2 0.0 0.051i 0*0" AO.630 100.20 29.6^ A ; pa rt icu la te . _ determ ined hy by weight ........ p.c.a. ` determ ined by ' ''"'b y 'height. 0 . I 8 8 A GRAMS 0 . 0 0 0 0 . GRAMS i -C' -h CL DETERMINED 5Y . BY WEIGHT 0 . 0 9 1 6 GRAYS r' H3P04 DETERMINED RY BY WEIGHT 0 . 2 2 0 0 GRA.'-o t r - ! ''' . .LEAK TEST ON ENU IPNENT = 0 . 9 0 1 CFM AT 15 .0 0 IN .HG -BAROMETERS' 2 9 . 6 4 i n . hg ; STACK D I A h = -19. 000 -IN-T V NOZ Z U f DI Mm -- 0 * J . . _ VEL STACK STACK T IM E VOLAT CONd iMETER TEMP METER P RE SS FLO* i HEAD TEMP P R E S S A TEND END TEMP START FIN CPI: t ' -ZONE HP TS H20 ZONE ZONE I N OUT HG KG H2*x .aoC o .1 * * : 0 0.0 95.0 0.6 o.o 935.30 D 90.0 0.0 0.0 G. ; l 0 . 0 6 0 9 5 . 0 0 . 8 6 . 0 9 3 9 . 7 1 64 0 9 8 . 0 9 0 . 0 0 . 0 ' 0 . 0 1.1 2 o.o e o 9 5 . 0 0 . 9 16.0 9 4 4 . 7A 64 .0 196.0 91 .0 0.0 0.0 1 . ; .3 0 .0 9 S 9 5 .0 0 .8 2A .0 9 5 0 . 2 2 6 5 .0 .111.0 9 2 .0 0.0 0.0 1 . 1 i A 0.0*95 9 5 . 0 0 . 8 3 2 . 0 9 5 5 . 7 1 6 5 . 0 1 IT . 0 9 4 . 0 0.0 0 . 0 l . i ' ' S 0.065 9 5 . 0.8 ' 40.0 960.50 63.0 196.0 *95.0 0 . 0 1o.c i ! 6 0 . 0 8 0 9 5 . 0 C. 8 A S . O 9 6 5 . 2 8 64 C l U ' . O 9 8 . 0 0.0 G 0 l . S ! . 7 0 . 0 9 0 9 5 . 0 0 . 8 5 6 . 0 9 7 0 . 6 5 6 5 . u 1 1 3 . 0 9 7 . 0 0.0 C. 0 1 . v . ' ' 8 0 . 0 9 0 9 5 . 0 0 . 8 6 2 . 0 9 7 5 . 9 3 6 8 . 0 1 1 4 . 0 9 7 . 0.0 0 . 0 1 . V ack cct^c^ MONSANTO INDUSTRIAL CHEMICALS CO. a/ J W.G. KftUMK.'MCH PLAiJT SAUiit7.IL 62201 . 10 NO. 63 121 AAC DATE SEPT 12 197A ST APT TIME 1:U0 PM STACK SAMPLED- INCINERATOR ` SOURCE OPERATION NO CALCULATION ANi) EVALUATION: ' j THE ISOKINETIC'CONSTANT HC/HP FOR THIS EQUIPMENT IS 23.626a' 631 VLC; SAMPLE/STACK/AVfi. P S S / AVG. TEMP/SAMPLE VOL./SAMPLE RATE /P.C. ZONE TIME .VEL. Mt ter STACK m e t e r s t a c k ME TER s ta ck ACTUAL ISC.n IN dev 1 8.0' 13.8 ?y.7 29.7 552.0 555.0 4.4 4.6 0.60 .G.64 -1>.34 2 8*0 16.0 29.8 29.7 566.3 555.0 5.0 S 4 0.63 0.74 3 8.0 17.4 29. * 29.7 560.0 555 0 5.5 5.9 0.73 0*80 --8.67 4 8.0 17.4 29. A 29.7 562.0 555.0 5.5 5.8 .. 0.73 _ 0.8:0 -rf.c3 .5 8.0 14.4 2.9.7 29.7 561.5 ,555.0 4. 5.1 0.64 0.66 -3.8A 6 8.0 16.0 29.8 29.7 562.3 555.0 4.6 5.1 064 0*74-13.60 7 6.0 16.9 29.3 29.7 564.5 555.0 5.4 - 5 *7 0.71 0.78 -6 S 1* S 6.0 16.9 29.6 29.7 565.3 555.0 5.3 5.6 0.93 0.76 19.39 TOTAL 0 62.0 16.1 .29.8 .29.7 560.5 555.0 .40.6 .43.3 0.70 .0.74 -5.73 T AV AV AV AV AV . 1 T AV AV AV i - TOTAL SAMPLE VOL* CORRECTED FROM STACK C ON D IT IO NS TO STD CONOI TIO.\S = 4 1 . 0 6 CUR1C FT. OF WET GAS TOTAL SAMPLE VOL. CORRECTED TO DRY S I D CONDITIONS = 3ft.96 CU6IC FT. % I* oce'.c MONSANTO INDUSTRIAL CHEMICALS CO. W.O. KRtJMMRICH PLANT SAUGET.IL 62201. II) NO. 63 1?1 AAC DATE OCT 1 19/5 ' START TIME 1:30 PM .| . STACK SAMPLED- ' INCINERATOR ..... ..SOURCE OPERATION NJ SAMPLING DONE BY NEIL SULLIVAN.AND EO HEUMANN - i, * 831 -'.. SAMPLING METHOD-EPA .METHOD 5 . DMf AND NA2S03 ,,IMPINGERS__________ I ' DETERMINATION Or EMISSION RATE OF L___________________ :___________A PARTICULATE___________!____ !_______ __ B:P.C.B. ,* CtCL2 . PROCESS WEIGHT RATE=__ 0.90 _THOUSAND POUNDS PER H O U R ________ TEST RESULT:. .T . ; /ACTUAL CFM AT /STD CFM . DRY STD CFM /. /.STACK CONDITIONS/ AT 70 DEG.F AND 29.92 IN. HG/ __' STACK VOLUME 3749.6 3257.1 2353.4 DUST* MI ST *OR GAS CONCENTRATION IN WET.6 , DRY. STACK _GAS_At_SjD_ CONDITIONS GRAINS LBS. PER LBS. PER MGS. PER PARTS _ . PER __ . .MILLION___ .1000 _ .,,.CUBIC___ . PER____ - _____ COMPONENT CUBIC FT CUBIC FT LBS. GAS METER MILLION * 0.0725 10.3620 0.1494 165.9835 138.2010 IN r.ET GAS -0.3.004. _ 1 4 .3 4 0 3 . .0.1836 .22').7183 .191.?679 IN DRY GAS B* - 0.0000 ' 0.0011 . 0.0000 0.01B2 0.0018 IN WET GAS ___ - j p * ___ _ 0.000.0 ......0.0016..... 0.0000 .._ ..*0.0252.. ... 0.0024 IN DRY GAS.. C. .... c : . .. 0.0079 .0.0110 1.1335 1*5687 0.0163 0.0201 10.1570 25.1290 6.1749 IN WET CAS .8.5460 IN DRY GAS EMISSION RATES ..... LBS. PER COMPONENT HOUR _____ A l ________ . 2 . 0 2 S 0 8 s 0.0002 cs 0.2215 " * * -* GRAMS'r T b sT p ER UNIT ~ PER SEC.' OF PRODUCTION . . . 0 . 2 5 5 1 __________ . . . 2 . 2 4 5 0 ________________ 0.0000 0.0002 0.0279 0.24S6 !1 AC* CC6<tC6 h. - 4 W J U U < W.G. KR'JMMR I CM PLANT S A U G E T . I L 6 P P 0 1 . ID NO. A3 12"l" A.\C ... DATE OCT 1 1975 START T IM E i : 3 0 PM STACK S AI M P L E D - . I N C I N E R A T O R . . . '. ' ***** *Mt SOURCE 0 3ERATIO,`j f;3. ! _ f i e l d d a t a : ...____________ ___________ 1______________ _ 0 O R I F I C E CONSTANT K=0.50f> PITOT TUBE FACTOR FS= 0.8 3 T EM P S . RECORDED .IN FA D E G R E E S ____________ _________________ METER CORRECTION FACTOR= 1.02 Oi .. .GAS A N A L Y S I S :.Z=N0ME. . ___ ______MOLECULAR. WT 0.F..Z5___ 0 . 0 f MOLECULAR WT. OF s t a c k GAS = 2 6 . 8 3 Q DRY MOLECULAR WT. OF GAS = 3 0 . 2 2 .. . O __ Q 3 3 a0 o 3 & ) 02 - N2 C02 CO H20 Z VOL TW PW . . .. ___,, 0 . 0 6 R 0 . 8 1 8 0 . 1 1 A 0 . 0 0 . 2 7 7 0 . 0 3 0 . 2 0 0 .. 70.89 29902**^ I I i PARTICULATE DETERMINED _. P . C . B < _________ _ D ETE RM IN ED C L 2\* * * DETERMINED BY BY. . _ BY BY WEIGHT#/ = BY W T K . iT I ON . .. = _ BY 'TITRATION = 0 . 2 0 0 2 GRAMS 0 . 0 0 0 0 GRAMS 0 . 0 2 1 9 GRAMS ' L E A K TEST ON EQUIPMENT = 0 . 0 0 5 CFM AT_ .15 .0 0 IN . hg . BAROMETER'= 2 9 . 9 0 IN.HG ; STACK D IA M = 19. Q0U IN . : NOZZLE 01 Mn1J --- ^ V" ---- 1 - VEL STACK STACK TI ME VOLAT COND METER TEMP METER f :;E 5 5 r HEAD T E mp P R E S S AT END END TEMP START FIN C" r 2 0 N E HP TS H20 ZONE ZONE I N OUT HG HG H23 . 0 0 * 0 1 4 9 . 0 - 0 . 4 0 . 0 4 2 8 . 5 0 D . 6 4 . 0 . . 6 4 . 0 . . .0.0. .0. 0 C.O Co o -4 ui o -- n 'O 1 0,200 149.0 -0 .4 0.0 431.84 51.0 60.0 64.0 0.0 0.0 0.3 = 2 0 . 3 0 0 1 4 9 . 0 - 0 . 4 1 6 . 0 43:5. 2 5 1 . 0 7 4 . 0 6 4 . 0 0 . 0 0 . 0 0.?i . 3 0 . 3 ^ 0 1 4 9 . 0 - 0 . 4 . 2 4 . 0 _ 4 3 9 . 9 7 .. 5 3 . 0 .80.0 . 6 5 . 0 . ,, 0 . 0 0 . 0 11 11-- -4 0 .3 3 0 14 9.0 - 0 . 4 3 2 . 0 4 4 4 . 1 1 5 5 . 0 80 .0 6 6 .0 0.0 0.0 1 .07 5 o;i7o 149.0 -0 .4 40.0 447.29 53.0 74.0 66.0 0.0 0.0 0.5 = h 0 . 2 3 0 1 4 9 . 0 - 0 . 4 4 8 . 0 4 5 0 . 3 4 . .54.0 8 0 . 0 . 6 6 . 0 ... 0 . 0 0 . 0 0.7*. 1 7 0.300 149.0 -0 .4 56.0 454.84 56.0 83.0 68.0 0.0 0.0 0.05 .B 0.270 149.0 -0 .4 64.0 458.70 58.0 82.0 68.0 0.0 . .. 0.0 0.57 * c & c r & W.G. KR'JNMPCH PLANT S A U G E T . I L 6 2 2 0 1 . - ID NO. 63 1P1 A.-.C o . (>AT o c r 1 1975 START TIME 1 :30 PM o I: o o STACK S A M P L E D - I N C I N E R A T O R . ______________ SOURCE o p e r a t i o n n o . .. CAL CUL AT ION AO' EVA LU A TI ON : THE I S O K I N E T I C CONSTANT.HO/UP FOR TrIIS EQUIPMENT I S X3 . 0 114 THE TAHLE BELOW SHOWStFOR EACH ZONE THE METER VOL. an q t *'F AVG AND P R E S S , I N ABSOLUTE U N I T S . FROM T H I S I S CALCULATED f n E O'j : STACK CONDITIONS a ND D I V I D E D MY TIME TO G IV E THE F>nw I nT th ^ L. h* *iu : , .COMPARED.WITH THE I S O K I N E T I C FLOW PERCENTAGE DEVIATION IS RECORDED, RATE CALCULATED. FROM ... THF STArw V ..... _S A M P LE /S T A C K /A V G . P R E S S / AVG. .t e m p /s a m p l e VOL./SAMPLE RATE /P.i o ZONE TIME VEL. METER STACK METER I SDK I i1 8.0 27.6 29.9 29.9 525.0 609.0 3.3 5.5 0.69 1.27 !___ 2 ___ . 8 0 . 3 3 . 8 .. 3 0 . 0 . . .2 9 . 9 5 2 7 . 5 . 609.0,_ . 4 . 0 6 . 5 0.31 . 1 . 5 6 - 4 7 , o 3 8 . 0 3 6 . 0 3 0 . 0 2 9 . 9 5 3 0 . 3 6 0 9 . 0 4 . 2 6 . 7 0.R4 1 . 6 6 - 4 9 . 4- 8 ; 0 3 5 . 5 3 0 . 0 2 9 . 9 5 3 2 . a 6 0 9 . 0 4 . 1 6 . 7 0 . 8 4 1 . 6 3 -43. . . 5..,.. 8 . 0 2 5 . 5 2 9 . 9 2 9 . 9 5 3 1 . 5 6 0 9 . 0 ... 3 . 2 5 . 2 0 . 6 4 1.1 7 - 4 5 . o r ; 6 7 8 . 0 2 9 . 6 3 0 . 0 2 9 . 9 5 3 1 . 5 6 0 9 . 0 3 . 6 5 . 6 0 . 7 2 1 . 3 6 -4?. 3 . 0 3 3 . 8 3 0 . 0 2 9 . 9 5 3 4 . 3 6 0 9 . 0 4 . 0 6 . 5 0 . 8 1 1 . 5 6 -43 L... 6__ _.e.o._ .32.1... 3 0 . 0 ,,. 2 9 . 9 5 3 5 . 3 . 6 0 9 . 0 ___3 . 9 ,,0.78 * 1.43 -47. TOTAL 0 64.0 31.7 30.0 29.9 531.1 609.0 30.2 0.77 1.46-47.: i 0 T AV AV AV AV AV * T AV AV i ___ TOTAL SAMPLE VOL. CORRECTED FROM STACK C O N D IT IO N S TO STD CONO I TI C-. = 4 2 . 5 9 C UB IC F T . - O F WET GAS , TOTAL SAM PLE _VO L. _ CORRECTED TO..DRY. S T D _ C O N D IT IO N S _ = _ 3 0 . 7 8 CUBIC,: ! .O - O LI _ r o: f-- I t O' O ': a?" o r 5:i . ACP- CC*iCc iiONiA.'; TO z c p .->*icai s co. u * vi. y f >w ' ' t. ANT 5 MJGET *11, 0 2 2 0 1 . 10 NO. A3 121 AAC DATE 23 lOVs). START TT-r. 10:.i0 AM STACKNvv^j__x.u^/iNC lMiCHAfoR sTACK SOURCE OPERATIC.* SAILING UDNL BY NEIL SULLIVAN AND ED H SAMPLING METHOO-EPH METHOD 5. 6 -Dm F AND NA2S03 Im PIn GERS DETERM INATION OF EMISSION RATE OF a : P ARTICULATE *8 :P.C.tl. PROCESS v. I g .-;t r a t e s JEST RESULT o o CiHCL 0 :CL2 THOUSAND PCjNOS PER HOUR /a c t u a l CF AT /STO CFM DRY STO .CF / STACK CONDITIONS/ AT 70 DEG F a n O 29 .92 IN. .-.3/ STACK VOLUM - 171 1 . 6 /- 1649. 0 1569*5 DUST i IST jOR GAS COn CENTRAT ICiV .I\: *ET L RY STACK GAS AT STS 3RA INS : PER COMPONENT CU-ilC FT AS 0.U522 A: 0.01-A9 L1***G.A7L(3.*-**^'`M- LMS. PER 05. PER PA--*T3 1000 C:-jIC PER Co '.v r l V.46".? LBS. GAS G.CV62 iME0TEt:R -a '*ILL ION 99.49-.S : 7-A343 ' 0.09 1 125.5560 I04 54?.j 1 B: o .noo j 0.0 0)6 u .00:c 0.(j?s 7 '0.0025 : 6: 0,0000 0.0017 0.cc00 0.0270 0.0026 I C: .G027 P.3737 0.0049 6.0656 4 *0a 0 C: 0.002 0.3979 n.ooso 6.3731 4.2160 1 D: 0.0021 0.2970 0.0036 4 75*0 16 iA1 I : 0.0022 0.3121 C0039 4.9991 1.7001 I e m i s s i o n k ;*T: LBS. PH - GRAMS COMPONENT HOUR PER SEC. A: 0.7331 0.0930 B: 0.0002 o.nnon C: 0.0375 0.0047 0 : 0.0294 0.0037 LRS. PER UNIT - OF PRODUCTION 0.73RI 0.0002 0.0375 0.0294 ACM CGe4C9 riOUSAN'TO INGWS T* IAl_ OiE'-UCAl S CO. ' CC J* *W.G. KKUMm PICiL LANT SAUGET IL 62201. -ID NO. ft3 1?1 A>\C HATE .SAY 23 1975 ' ST AWT TIME 10:30 AM . STACK SAMPLl O- INCINERATO*. STACK SOURCE 0=cPA7IG.\ ,\G. 3; . FIELD DATA: O^irrCc CONSTANT k=o.50ft PTTOT TUdE FACTOR Fs= 0.83 TEMPS. ^i'v-0DEC IN FA D.EGPEFS METE* CORRECT ION FACTORS 1.02 GAS ANALYSIS: Z=NONr KOLECU LAP IVT OF Z= o.0 MOLECULAR WT. OF STACK GAS = 30.00 DRY MOLECULAR WT. OF GAS = *30.60... . * .. 02 N2 CO? CO H20 -2 VOL T.v .P*. jO.O2d 0.824 0.143 0.0 0.048 0.0 45.63C 114.4.' 29. i- *- PARTICULAT E P.C.H. HCL CL2 DETFP:**.In ED a Y 0ETFPi*INE0 "Y DETr Pm In EO F,Y d e t f p m INFO GY- KY VE Cri KY v.:E IGHT KY WEIGH? KY WE IGnT ^ 0.15*.' G-.A = 0.0 f - \l~.n4 ~ r 0.0:77 GP- 2 0*ij0s j G-.-- ' A LEAK TEST ON EC'.jIaMr*:T---1ft/ft*ft*/* Cf N 11 15.CO haRJMETER= 29.*64 IN.r,3 STACK DIAM = 14.000 IN *** IN .; NOi?.iv--.rZ .1'V'-"'V- -- Vr twV3 VEL STACK STACK TIME VOLAT CONO ME TEP TES? METER PRESS Fj_; " HAD TEMP P^ESS- ATFNU ENO TEMP ST AP T r IN GrC* ZONE NP TS H20 ZONE 7ONE IN OUT HG Ho nc . 0 O.n 86.0 l 0.0ll*0 86.0 0.8 0.8 0.0 926.60 01 86.fi eh6.0 8.0 932.60 0.0 1 C6 f) 94.0 0.0 n .o b.j c .o C 1. *> 2 0.O90 86.0 3 Q.OftO 86.0 0.8 1ft.0 93d.71 0.8 ?4.0 944.48 0.0 122.0 1 04.0 0.0 132.0 IG4.0 0. O.o '*c C .0 21i,.-*.T- * O.OftO 66.0 0.8 32.0 950.04 0.0 134.0 1 05.0 0.0 0.0 1 .-5 5 0.950 86.0 6 0.070 86.0 0.8 40.0 955.74 0.8 48.0 960.50 0.0 128.0 110.0 0.0 134.0 112. 0 0.0 0.0 0.u -0.0 1 *u 1 .*7^ 7 .070 86.0 0.070 86.0 0.8 56.0 9ft7.85 0.8 64.0 972.23 0.0 134.0 114.0 0.0 134.0 113.0 0.0 0.0 C 0 1 V c*u 1 .71v- ACM CC641G MONSANTO IN DU ST RI A L CHEMICALS CO, i. J W.G. KRUm m RICH PLANT SAUGftT.lL 62201. ID NO. S3 121 AAC DATE MAY 23-1975 START TI`ME 10**30 AM - STACK SAMPLED- * I\'CTNRATOR s t a c k CALCULATION AND EVALUATION: SOURCE o p e r a t i o n .-NO IS T ' THE ISOKINETIC CONSTANT HO/h P FOP This ECUIP:-*.,~\T IS 24.7933 THE TAriLE GELOW SHOWS .FOR Ei.Cri `/ONE. THE METER VOL AND TnE AV AND PRESS. IN ABSOLUTE UNITS. FROM THIS IS CAL C*i*LATFU STACK CONDITION'S AND DIVIDEO m Y TIME TO GIVE THE FLOW Th e 'c O j IN ji-.E 'V ft. COMPARED WITH THE ISOKINETIC. FLOW PATE CALCULATED FROM THE ST PERCENTAGE DEVIATION IS RECORDED. rn ri SANPl E/STACK/AVr,. PRESS/ AVG. TEMP/SA.-u l E VOL./Sample Pa " ZONE TIME VEIL. ME7rR STACK METER ST ACN METER STACK A TUAi. I3CKI* 3 , -1 -. H.O -15.7 * 29.8 *29.7 553.0 546.0 --6 .0" 6 4 1 0.8n 0.72 ;:.22 2 H.O 16.6 29. a ?9.7 S66 .5 546.0 6. 1 6.3 0.79 G.77 ;.'i 3 8.0 13.b 29.7 ?9.7 575.5 5A.fi 5.8 5.9 0.7*1 0.63 17** 4 8.0 13.6 29.7 29.7 579.5 546. 5.6 5,6 * 0.70 C.t>2 . *: 5 8.0 12.4 29.7 ?9.7 5h 0. 5A 5.0 5.7 5. B 0.7? 'j .oi 6 8.0 14.7 29.8 29.7 581 .G 546.0 4.6 4.8 0.60 G*6`"-- ... "j 7 8.0 14.7 29. d 29.7 523.5 546.0 7.3 7.4 0.92 .6B 3*; fi 8.0 14.7 29.8 29.7 555.0 546.0 4.4 4.4 0.55 0.6 - 1 I. TOTAL -- 64.0 14.5 29.8 ?9.7 575.5 546.0 45.6 46.5 . 0.73 T AV AV .AV * AV AV T T AV AV TOTAL SAMPLE VOL. CORRECTED FROM STACK CONDITIONS TO STD CDn DITI:n S= -44.83 CUK1C FT. OF WET GAS TOTAL SAMPLE1 VOL. CORRECTED TO ORY STD CONDITIONS = . 42.67 CUBIC FT. i ACM CC6411 ii. 3. ;.m . r.:". .*..tM i '. : A ir. j I i ! . . '.::::;I* r ii.O STAC!. Tl. ::*c w :`.'IT.* :*;: fn ; '11 .V J ""T ;i ::..C7i.v* co: m *:v r i s v a u : in :: r r i f i t b I-"!.. **'*'' y : r : i\ " * " v*? * s-u.: iu v i: n - ti::: r . - - i . . 2, .*7 A 1, Amount n i u a t c r c o lle c te - : C` : i i \ t I i L c r - ) .Volume o w a te r va-in- c - 'lie e t c ! UZ,Z' 2.'';'-;?''. 2 . DTy - - r . r .e tc r v o lu r c n t u c t e r cou M ic ::- ; (cuV Ie rr-.-l) l. ir n ` 'c : r i c 'fc r . s u r c ( i.;c \-zz o." r - r v c . r -- ) 27.'-. Aver.'X" * o r i 2c e m *c ;;r.::c *!r,-- * ( ir .e .ie r- c i v c t t i ; 1. f rti Average r a te r t e r r e r : : re " 133 *7 3 . Cas n o t e r v o lu r.e n t S" ? ( i r y c u b ic f e e t ) 22.22 4 . P ercentage ci n e is tu r c in -ns scree;*. 5 .1 " ;i 5 . P e rc e n t e o rtp o s ic io a e f -.ss n tr e .v .: on................. 1*>. K i t r o r c a .................. C \ V '7 ! Carbon J io :;i..e .. . 1.2 '*" Carbon ;!or.p::i i c . . . "'r *: c. P e rc e n ta g e o cucees a i r " 1 3 3.25 :: 7. Hr;* : .n in c u ls r :: b i ~ h t c i ;;as s t r e r *. (1 j . / l b -- r la * 27. " 1:et n o ic c - la r - wC i3 .;t a: j i s s t r c . t r ( l b . . i n - r a l a ) " 2 : . . ` . 0. M in t tube coci ic ie r .t C i - casio-airs?) .53 funbcr o f p it o t tuba rea:!i". A verage oT s r a : re r o o ts o i v s la c it* -* pre s sura*-, -* . r * : : Avcra;*,c fp s r.trc s :; te -p a ra tu r ( * " ) 5 i*" Gas s tre a m v e in c i t y ( : : / hcc) l ^ . l ' r - P ito t tube rc a u in 's : .06 .OS .0? 30 .33 .3"' .72 ^1 9. Stal: dien.? in n s , e irc u l.-.r w ith lia - c t e r ( fe te ) - 1.23 Stack arca (r-tiare fe e t) - 1.0C- V o lu m e tric flo u a t S7" (cub ic J e e t/n in ) - 1717 10. Nnzr.lc d im e te r use.! fo r s m ilin g (in c h e s ) .273 Sam plin'; tin e (rtin u e rs ) * 2 .* 3 *V e ir.h t o f p a rtic u la te rete;* c o l le e t e l ( r iil li^ r ; ; - ) .33 11. P a rtic u la te con cen tre tion (lb ./S C " ) ''l P a r tic u la te c o n c e n tra tio n ( " ra ia /'iC !') . 0 "'7 1 2 P a r t ic u la r s c o n c e n tra tio n ro rre e * .n l to 5*1 2 cuecen a i r i'.b . '7* P a r t ic u la t e c o n c e n tr a tio n c o r r e c t e ! to S'* * cu e cst a i r (-.*.! -_n. / 1 2 . Is o b in e tic rate (.2) - 12.15 13 . Itan s c n is s io n r a t e ( l b . / h r ) .''T Date__ j f ' L h S l*n a tu 1*7 r.ija. l ACfc CCfc^tU i-..-.; v a i ?. s i,` :,. \ : u .:.a::c .: svac< r.iiv i:? ! VI I' si- v.: c n - a::v__ "T '-.y-n __________________________ stack I'acstIr i c . \ T "'` i .............................................. CITI UlTl'Kit HACK O'* _____ * *p. i 1. A ro u n t o f v o t e r c o lin e t o ! ( : : i 11 i 1 J tr-v c ) Volur.e o i ( in te r va p o r c o llu d o - ,; ( 3 i:* l it, ,,?o 2. I ! *7'; 2. Dry -a s (le c e r v o lr .r c i t i-u te r c m ; i t * ' "a s {c-;'--'c iu c t ) a A ". *3 D . ir o r x t r ic p r e s s ili'i! (in c h e s oC m e rcu ry) a 27. *'i A ver .ire o r i f le a o rc a s e re re*; ( io.a`-:..s o : '. >s r) 1..V.9 Average r.c te r tc-.-.p&raturrt 11A *F 3. Gas p e te r volum e .at 3" ? (d ry c u b ie f r e t ) 41.67 Percentage o f r.oir.eurc in -ss stream s. P e rc e n t c o m p o s itio n o f gas e t r a cm: 4.T1 V ! fls m e r.. . . . . . . . . . 2.3"" n i t r o g e n .................. Carbon .!o . . . 14. : gg Carbon ^o n o n id e .. iy,f 1 G. P e rc e n ta g e o f excess a i r 1 4 .7 7 " 7. D ry n o lc c v .ln r v c ig .-.t o f pas c e rc a r ( l ' i . / 1 :-- a ir.) l i s t n o la c u ln r i:t_ ig '.iC o f gar. a t r o c i ( l a . / l b - - e l t ) a P it o t cube c o e f f ic ie n t (1; u n s im lc s s ) * .33 Murijfc.r o f p ie c e tub e r s a s in r s C Average o : squ are ro o ts o f v a ' d i ;* p r . ir a v e s A verage gaa n tre v -. tc -m a ra c u re * :' ) 77*7 Gas s c re e .; v e l o c it y ( f t / s c c ) l i . 5 3 " 1 T icoc tube rea.!in*a*. .05 .09 .*5 . 0G .95 .07 I 't ? .97 r.i 9. S ta c h d irc c n a io n s , c i r c u l a r v i t h d r . - c t e r ( f e e t ) - 1. *9 S tack .area (squ are fe e t) 1.9G V o lu m e tric flo w a t S"? (c u b ic C e e t/fin ) 13^3 HC. llo c r.le d ia m e te r used f o r sa m p lin g (in c h e s ) .375 Sam pling tin e (m in utes) 64. W eight o f p a r tic u la te m a tte r c o ilc c ta J ( r ili ig r a o ) - 21. P a r c lc u la c a c o n c e n tra tio n ( I b . / 7 C r ) " . r.a vclcu l.a te c o n c e n tra tio n (grain/GC**) . n -'V 'ig P a r t ic u la t e c o n c e n tra tio n c o r re c tc J to 59 e::cc"!-. a i r ( lb .'" " ' * .............. P a r t ic u la t e c o n c e n tr a tio n c o v rc c r.i: ' co 3*1 ?. cueces a i r ( r . r i i .""7 .22 Isolane tie rate (*) " 1^6. M 13. Knse cilinaio raen (lb./hr) .00 Dote F s j . / , 97G Signature i * ft A C * CC 6 ^ 13 c o *.i:'a :;v o-nrr s t a g ;; i;j:::rrin c a t in:; "cm incinerator ron I j)ATU or Ul*:: Or.c. l ,1075 1. Amount or water collectcr*. (nillilir.crs) 244.Til Volume of wiater vapor collected (ST") * 11.52451 2. Dry as neter volume at neter- 00:1.'.:.tier.a (cubic feet) - 3". 10 Barometric pressure (inches of mercury) = 70.00 Average orifice pressure drop (inches of water) = .360 Average neter.temperature " 71 F t 3 . Gas neter volume at S7P (dry cubic feet) * 30.13 4. Percentac of no is turn, in stream 27.75 7. 5. Percent composition of as stream: OiTy^cn. Nitrogen........ Carbon Dioxide... Carbon Ilonaxida.. 6.C07 S1.2nn 11.405 .00:: .6 Percentage of excess air 45.36:; 7. b./lh-- ole) s 30, C*1 b. /ll.-:-olc) * 26.73 "\ 8. ) a .33 Similar of pitot tube readings 8 Average of square roots of velocity presaurns = .5133 Average as stream temperature (T) - 140*7 Gas stream velocity (t/scc) 31.323 Pitot tube rc-adinGs: .20 .30 34 .33 .17 .23 .27 ".n 9 . Stack dimensions, circular wita dianetcr (fent) 1.53 Steel: area (square feet) " 1.96 Voluretrie flow at ST? (cubic fcct/nin) a 2352 IT), llosnle dianetcr used for n .moling (inches) .37' S x m l i n G tine (minutes) 64.00 'e ig h t o f particulate ratter collected (nillirran) .'ll 31, 12. Particulate concentration (lh./SGV) " .00nr'000 Particulate concentratioa (y,rnin/SCr) * .000011 Particulate concentration corm c L c d to 50 7, excess air Particulate concentration corrected to 50 7. excess air Isokinetic rate C O * 51.23 (lb./CT)'(Graiii/C?) 13. llasr. emission rate (lb./hr) .00 ACH CCE'il't Date_ ,/fyc Cl S7 \ y MONSANTO IM O U S T H IA L . C H E M IC A L S CO. f W.G. KKlJMMrllCH PLANT SAUGET* I L 6 2 2 0 1 . ID NO. 63 121 AAC DATE HAY 2 3 - 1 9 7 5 START T IM E 1 0 : 3 0 AM - STACK SAMPLED- INCINERATOR STACK . s o u r c e o p e h a t i o n . n o . 03i: CALCULATION AND e v a l u a t i o n : THE IS O K IN ET IC -C O N S T A N T HO/HP FOR T H I S EQUIPMENT I S 2 4 . 7 9 3 8 THE TAHLE QELOW SHOwS * F O R EACH ZONE.THE METER VOL. AND J H E a |v G. HE R TE:-:j AND P R E S S . IN ABSOLUTE U N I T S . FROM TH IS I S CALCULATED T h E EOUIVALi T VOL. STACK C O N u I T I O N S ANn D I V I D E D RY T I M E TO G IV E THE FLOW IN THE NOZZ! . Tr! ; S COMPARED WITH THE I S O K I ' N E T I C FLOW RATE CALCULATED FROM TH STACK V LOOT TY. PERCENTAGE DEVIATION IS RECORDED. SAMPLE/STACK/AVfi. PRESS/ AVG. : * ZONE TIME VEL. METER STACK METER *-- -1----- - H . 0 - - 1 5 . 7 * 2 9 . ft - 2 9 . 7 5 5 3 . 0 1. 2 ti.O 1 6 . 6 29.. 8 ? 9 . 7 5 6 6 . 5 3 a.o 13.6 29.7 29.7 575.5 ; . ...4 . 8.0 13.6 29.7 29.7 579.5 5 8.0 12.4 29.7 29.7 580.0 i -6 .0 14.7 29.8 29.7 581.0 - 7 - 8.0 - 14.7 -29 .8 29.7 583.5 ft 8 . 0 1 4 . 7 2 9 . 8 2 9 . 7 5 8 5 . 0 * total - -- 64.0 -14.5- 2 9 .6 ` ;?9.7 575.5 TEMP/SAM^LE VO L ./S AM PL E R ATE /R STACK METER STACK a c t u a l I SOKIN DEV 5 4 6 . 0 --" 6 . O' . 6 * 4 ... 0 . 8 0 * 0 . 7 2 1 0 . 2 3 546.0 6.1 6.3 0.79 0.77 3.33 546.0 5.8 5.9 * 0.71 0.63 17.49 5 4 6.0 . *5.6 5 .6 * 0 .7 0 0.b3 12. 546.0 5.7 5.B 0.72 0.57 26.00 546.0 4.8 4.6 0.60 0.6S-11.C6 546.0 7.3 : 7.4 * 0.9? - 0.6B 36.74 5 4 6 . 0 4 . 4 4 . 4 0 . 5 5 0 . 6 8 - 118 . 7 2 546.0 43 . 46.3 i .0 `vJ 0 6 T 9.04 , T AV AV .AV AV AV . T T . AV AV AV *, TOTAL SAMPLE VOL. CORRECTED FROM STACK COND ITIO NS TO STD C O N D I T I O N S * 4 4 . A 3 C UB IC FT. OF WET *GAS ' TOTAL SAMPLE VOL. CORRECTED TO DRY STD CO ND ITIO NS = . 42.67 CUBIC FT. ACM C C6 415 . -, MONSANTO IND USTR IA L CHEMICALS;.CO. * dJ' W.G. KRUMMR1CH PLANT- -SAUGET1 1L 62201. ID NO.- 63 121 AAC DATE OCT 1 1975 START* TIME 1:30 PM STACK SAMPLED- INCINERATOR ---....... . SOURCE OPERATION NO.' 831 SAMPLING DONE BY NEIL, SULLIVAN AND ED HEUMANN SAMPLING METHOD-EPA METHOD 5 . DHF AND NA2S03.IMPINGERS__ DETERMINATION OF EMISSION RATE OF ' * _____________ :________:________:___ A :PARTI CULAT.E___ !____ . BiP.C.B. I C:CL2 PROCESS WEIGHT.RAIE=__ 0.90. .THOUSAND POUNDS_PER HOUR TEST RESULT: . > /ACTUAL CFM AT /STD CFM DRY STD CFM / -- ;__ _ .... . / STACK CONDITIONS/ AT .70 DEG F AND 29 .92 IN. .h g /._ STACK VOLUME 3749.6 3257. 1 2353.4 OUSTiMISTt OR GAS CONCENTRATION IN WET .L ORY.STACK.GAS .AT.STD CONDITION GRAINS . LBS. PER LBS. PER MGS. PER PARTS ...PER ... . MILLION. ...1000 . CUBIC ,,.'PER COMPONENT CUBIC FT CUBIC FT LBS. GAS METER MILLION a: 0.0725 10.3620 0.1494 165.9835 13S.2010 IN WET GAS ___ A?_____ ...0.1 004 14.340B.. . 0.1336 229.71 A3 191.?679 IN DRY GA3 B?: ... _.b:: . 0.0000 0.0011 - 0.0000 0.0182 0.0018 IN WET GAS .... o.oooo . .. 0.0016. . 0.0000._.,, 0.0252,,. 0.0 024 IN DRY GA3 C C3 0.0079 1.1335 0.0163 18.1570 6.1749 IN WET GAS o.ono .__ ,1.5687. . _00201 ... 25.1290 . 8.5460 IN DRY GAS EMISSION KATE! LBS. PER COMPONENT' HOUR _ 2.0250 EF-: 0.0002 C3: 0.2215 GRAMS LBS. PER UNIT PER SEC. OF PRODUCTION 0.2551.. . . 2.2450___ 0.0000 0.0002 0.0279 0.2456 i V \pj W.G. KR'JMMRICH PLANT SAUGET.IL 62201. I NO. 63_ 121 h ' r DATE OCT 1 1975 START TIME 1:30 PH STACK SAMPLED-. JNCINEKATOR.................... , SOURCE 0=ERATlCw fO 0 FIELD DATA! .. o ' ORIFICE CONSTANT K=0.50f> PITOT TUBE FACTOR FS= 0.B3 TEMPS. RECORDED IN FA DECREES - -.______________________ HETE CORRECTION FACTOR 1.02 ' OAS ANALYSIS :.Z=NONE ____MOLECULAR. WT_OE_Z=___ 0.0 ^ MOLECULAR- WT. OF STACK GAS = 2 6 . 8 3 - - __ O | ORY MOLECULAR WT. OF GAS =' 30.22 S31 O ............ , 02 2 C02 CO H20 1 VOL TW PW WG . 0.068 0.818 0.114 0.0 0.277 0.0 30.200 .70.89 29.90244,6 t/I U) i/> C ! PARTICULATE DETERMINED 8Y BY WEIGHT C ( * = i_._P.C.B<_______ DETERMINED BY _ .BY TgT.'hVM-ON CL2 DETERMINED BY BY TITRATION = 0.2002 GRAM 0.0000 GRAM 0.0219 GRAM O !' _ LEAK TEST ON EQUIPMENTS. 0 0 5 CFM AT.. IS. 00.. IN.HG____ ____ ____ r 8AR0METER= 29.90 IN.HG5 STACK DI AM= 19.000 I N .: NOZZLE DIAM = 0.375 Ss VEL.... s t a c k STACK TIME VOLAT <CONIO 1METER. TEMP 1METER PRESS r* ~ *r MEAD TEm P .PRESS AT END END TEMP START fin CR l FI_ ONE HP TS H20 ZONE' ZONE IN OUT HG HG H20 T 0 0.0 149.0 -0.4 0.0 428.50 D 64.0 64.0 0.0 0.0 0 D *' * 1 0*200 149.0 -0.4 8.0 431.4 51.0 68.0 64.0 0.0 0.0 C .=*53 0 2 0.300 149.0 -0.4 16.0 435.82 51.0 74.0 64.0 0.0 0.0 0.7 =3 i___ 3 0.340 149.0 -0.4 24.0 439.97 53.0 80.0 65.0 ... o.o 0.0 i.i: 3 4 0.330 149.0 -0.4 32.0 444.11 55.0 80.0 66.0 0.0 0.0 i ij 0r _ 75 o a ? o 149.0 -0.4 40.0 447.29 53.0 74.0 66.0 0.0 6 0.230 149.0 .-0.4 48.0 450.84 . .54.0 . 80.0 .66.0,__ 0.0 0.300 149.0 -0.4 56.0 454.84. 56.0 83.0 68.0 0-0 0.0 0.5 =9 0.0 0.793 0.0 0.550 3 l' 8 0.270 149.0 -0.4 64.0. 458.70 58.0 82.0 68.0 0.0 0.0 0.870 - .1 ft r it* ( -/ \ ACP 0 C b U 7 - W W J I^ k,l I^ I 1 i V - M ^ U U ' .*'. -W.G.*KRUMMRICU p l a n t SAUGET 1L 62201, I D NO. 63 121 A/.C 3 ' DATE OCT 1 1975 - - START .TIME 1 : 3 0 PM ,,/ /u /- : STACK S A M P L E D - IN C IN E R A T O R .......... ..... ...... . SOURCE OPERATION NO. `g 3 " CALCULATION AND EVALUATION: 3 J . . THE I S O K I N E T I C CONSTANT HO/HP FOR T H I S EQUIPMENT I S 13 .01 14 THE TAriLE BELOW SHOWS,FOR EACH ZONE, THE METER VOL. AND THE AVG. !!-:: : AND P R E S S . IN ABSOLUTE U N I T S . FROM T H I S I S CALCULATED The Ec-u i v a l e v 3 STACK CONDI-TIONS .a n d D I V I D E D BY T IM E TO G IV E THE FLOW In the NOZZl :, . COMPARED WITH THE I S O K I N E T I C FLOW RATE CALCULATED FROM, the STACK VE. PERCENTAGE DEVIATION IS RECORDED. -- - **SAMPLE/STACK/AVG. PRESS/ AVG. TEMP/SAMPLE VOL. /SAMPLE RATE . / p .C. : ZONE T IM E VE L . METER STACK METER s t a c k METER STACK a c t u a l I S O K I N ' l 'EV ;1 e.o 27.6 29.9 29.9 525.0 609.0 3.3 5.5 0.69 1.27-42,07 ;___ 2 __ _ 8 ; 0 _ 3 3 . 8 . ,, 3 0 . 0 2 9 . 9 5 2 7 . 5 6 0 9 . 0 . 4 , 0 . .... 6 . 5 . . 0 . 8 1 1 . 5 6 - 4 7 . 7 4 , 3 6.0 36.0 30.0 29.9 530.8 609.0 4.2 6.7 0.84 1.66-49.10 '4 e:o 35.5 30.0 29.9 532.8 609.0 4.1 6.7 0.84 1.63-436 6 r ...5 .. . 8.0.. 2 5 . 5 2 9 . 9 2 9 . 9 5 3 1 . 5 6 0 9 . 0 . 3 . 2 , 5 . 2 . 0 . 6 4 . 1 . 1 7 - 4 5 . CO r6 8.0 * 29.6 30.0 29.9 S31.5 609.0 3.6 5.8 0.72 1.36-47,19 o ;7 8.0 33.0 30.0 29.9 534.3 609.0 . 4.0 6.5 0.81 1.56-42.14 1 . . 8 .. . 8 . 0 ...32.1 3 0 . 0 . 2 9 . 9 5 3 5 . 3 60 9. 0. __ 3.9., __ 6 . 2 . . ,, 0 . 7 8 1 . 4 3 - 4 7 . 3 5 ', total . - 1 o 31.7- 30.0 29.9 531.1 609.0 30. 49.0 0.77 1.46-47.55 >< j1 T AV AV AV AV T T AV AV AV lOTAL SAMPLE VOL. CORRECTED FROM STACK C O N D IT IO N S TO STO C O N D I T I O N r . 4 2 . 5 9 CU BI C FT. OF WET GAS TOTAL SAMPLE VOL. CORRECTED TO DRY STD C O N D IT IO N S = _ J 3 0 . 7 8 CUBIC F 3. 3V I. _/ *L 37~ 3* v 3*i ACM CC416 */ II F a c ilit y : Monsanto Company W.G. Krummrich P la n t Route 3 '. Sau ge t, IL 62201 Date o f Inspection: January 27, 1976 P a rtic ip a n ts: Monsanto Paul H eisler Clarence Buckley U.S. Environmental Protection Agency Edwin Z y lstra John Connell Charles M ille r Background: Monsanto i s the so le m anufacturer o f PC3 in the United S ta te s. P r f ig u r e s f o r the f i r s t n in e months o f .1375 are l is t e d below: Aroclor* 'I 1016 ! .1242 1254 Thousands o f Pounds ' 10350 5120 * ' 6980 S t a r t in g d u rin g the l a t t e r p a rt o f 1972,' the o n ly use fo r PCS compounds i s in the manufacture o f tra n sfo rm e rs and c a p a c ito rs. Process; The c h lo r in a t o r i s charged with, biphenyl and f e r r i c c h lo rid e (catalyst) . and then heated- Vaporized c h lo r in e i s fed in to the c h lo rin a to r. The conta ct tim e v a r ie s from 12 to 35 hours, depending on the type o f Arcelor to he produced, which determ ines the degree o f c h lo r in a t io n . The vapor : from the c h lo r in a t o r (HC1 c o n ta in in g PCS) goes to the scrubber where it i s washed w it h liq u id A ro c lo r. The gaseous HC1 i s se n t to the purification .sed tton o f the p la n t. The crude A ro c lo r goes to the Blower tank where * t t t s blown w ith dry* a i r f o r se ve ra l hours. The a ir i s scrubbed with - - - water and vented' to the atmosphere through, a.d em ister, the crude product tS'sent to a sto ra g e tank, whara a few tenths o f 1# o f a lk a l i i s added to react u it h any remaining hydrogen c h lo rid e or f e r r ic chloride. r ACM CGt*r19 I -2- The method o f d i s t i l l a t i o n o f the raw A ro c lo r v a r ie s `depending upon the product to be produced- Fo r types 1254, 1242, and 1221, the process is the same. Each i s d i s t i l l e d in a vacuum s t i l l , the condensate being the f in is h e d product and the bottoms being the fa n ta rs which are sent to in cin e ra tio n . For A ro clo r 1016, the crude a ro clo r is d is t ille d in a vacuum d i s t i l l a t i o n tower. The steam frcm the steam j e t e je cto rs is p a r t ia lly condensed; the condensate being discharged into the plant d isc h a rg e sump and the vapor exhausted to atmosphere. The overhead frcm the d i s t i l l a t i o n tcv.'er i s the product 1016 and i s se n t to sto rage . The bottoms are se n t through another c h lo r in a t io n and d i s t i l l a t i o n cycle. The overhead from t h is s t i l l i s the f in is h e d product and the bottoms are th Montars which are se n t to in c in e r a t io n . The three p o s s ib le places wher PCB can escape in to .th e atmosphere are: vapor from j e t e je c to rs, exhaus from the scru b b er; and su rfa c e area evaporation (in ge n e ra l). ri- fl) ti> A l l a ro c lo r s are sto re d a t 150F w ith a la y e r o f n itro g e n on top o f the liq u id . . In c in e ra tio n : There are three points in the process that go to the in c in e ra to r: far.cars, bottoms from se p arato r sump, c o lle c t io n from a ll d rip pans. In a d d itio n waste m aterial from the u se rs o f PCB i s se n t to Monsanto to be destroyed.. The type and con ce n tration o f these waste are unknown but i t i s estimated th a t they con tain about 90% PCB. The amount o f contaminated waste disposed o f f o r the f i r s t nine months o f 1975 i s lis t e d below: i /v Thousands o f Fcur. Customer returns, records e x ists Monsantoi records e x ists Total receipts Destroyed by in c in e ra tio n 1109 322 1431 2808 T h e 'liq u id waste stream i s steam atomized and fed in t o the f i r e box. The feed i s in c in e ra te d at a tem perature above2200F w ith natural cos used fo r combustion w ith 52 excess oxygen and a re te n tio n time o f 2-3 seconds, .he gases go through ^ quench pot, the exhaust of which passes through a venturi scru b b e r and then through a packed tower which i s ir r ig a t e d by the.weak m u ria tic acid o r ig in a tin g from the quench pot. The exhausts are vented to atmosphere through a dem ister and are monitored. ; ACf* CCfc^mC * 3 ^ \ Loading and Unloading: The m a jo rity o f the PCB components and the products made u sin g PCB are l i q u i d and are tran sp o rte d in b u lk o r in ste e l drums. Bulk shipments are made in r a ilr o a d tank car and tank tru c k s. Waste m aterial shipped in from o th e r lo c a t io n s in 55 g a llo n drums o r tank tru c k s are unloaded in to a concrete p it ; the m aterial in t h i s p it i s p e r io d ic a lly pumped to one o f fo u r 20,000 g a llo n in c in e r a t o r waste feed tanks. The waste m aterial that i s shipped in by r a il tru ck is unloaded in to a 500,000 gallon storage tank and i s pumped in to the feed ta n k s when req u ire d . In the truck o r r a i l c a r ' loading area, drainage i s directed in to a small concrete p it which is p e r io d ic a lly pumped in to the b a s in s located in the m anufacturing area. R e lie f va lve lin e s and atmosphere ve n ts are routed through catch tanks cr are redirected to underground se t t lin g basins. *Ci* CC6421 ** _l mr i j i ji lift if .tut ..U ri.'Od CIA >tt (rJ>*'< UNITED STATES G O V E R N M E N T Memorandum I .. TO : Gerald F. Regan, C hief, A ir Surveillance Branch j y d a t e : March 5, 3 FROM : Technical Advisor, Special Projects Section Air S u rv e illa n ce Branch subject: 1 I Calculation of PCB A ir Emissions from Storage and Dispensing Operations at the Krurmrich Plant of Monsanto t Potential a i r emissions from the storage and dispensing of PCB were calculated based on information supplied by Monsanto. The results of these calculations are: Storage tanks for v irg in product Product loading Storage and handling waste aroclors 597.4 kg/year 0.'0l84 kg/year 0.00066 kg/year Attachment (C alculatio ns) R. Edwin Z y ls tr a I j i V 'tliWl aom coe^tZ2 Buy U.S. Savings Bonds Rrfularly on the Payroll Savings Plau C laren ce Buckley of Monsanto estim ated that storag e and loading o p eratio ns at the p lan t would d is p la c e the follo w in g volume of headspace gases: ft 1. Storage tanks 635,000 ft3 2 . Loading 460t0C0 ft3 Calculation of emissions: From F ig . 2 .2 from the Draft Mitre report the vapor pressure of Aroclor 1242 is 3 ntn Hg at 100C and 10~4 at 21C. The a ro c lo rs ar stored at 100C. It may be assumed that the terr.p. in the tanks bei loaded would be approximately 70F ( 2 l C ) . . From D olton! s law o f p a r t ia l p re ssu re : I Bx Px where Bx proportion by volume of a gas component P mix. * Px p a r t ia l p re ssu re o f gas component and P mix. * ab solute p ressu re of gas m ixture. Bx 3 0.0035 o r 3500 ppm at 1CCC 760 1 To change ppm to mg/M^ use the follow ing form ula: mg/M^ * ppm x mol, w t. molecular volume m ol. w t. 261 m olecular volume * 30.62 lit e rs / m o l. at 100C , mg/H3 3900 x 261 - 33242 30.62 Em issions from sto rag e ta n k s: Em issions kg/year - V o l. o f gas d isp laced /ye ar x concentration o f PCB Kg/yr - 635000 f t3 x .0283 m3 x 33242 x , ftl 1 kg 100 mg kg/year 597.4 Em issions ffbm loading o p e ra tio n s: Bx - Px *,, io"^ 1.3 x 10"7 or .13 ppm P mix. 7b mg/h3 .13 x 261 1.41 mg/H3 24.04 ACM 0CS*23 I 1 -2.I jj Em issions: I K g/yr. * 460,000-x ..0283 x 1.41 x 1j 1 kg I0& mg Kg/yr. - 0.0184 Em issions from handling of waste a r o c lo r s : T o ta l w aste m a te ria l received into p lan t as reported by Monsanto fo r 9 mos. of 1975 " 1,431,000 lb s . Assume t h a t a l l t h i s m a te ria l is received and stored at 70F and th a t i t d is p la c e d an equal volume o f tank headspace gas a t 7 0 F. Volume o f produce 1.1.31 .000 l b s . X .. 1 a l . 11.5 lbs. x 8 g a l. 1 X i a * " hJ " " * 71 h3 The co n cen tratio n would be the same as in the leading operation th a t is 1.41 mg/M3. Em issions : K g/yr. - 471 m3 x 1.4 mg/M3 x ' 1 'kg 1& mg Kg/yr. * .00066 1 j .. \ AC* CC 5\ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY SUDjCCT: FROM: TO: PCB M onitoring - Monsanto, Krummrich Plant, Sauget, Illinois. . t' DATE: F f 9 `1 * Charles M il 1er/Ed Zylstra/John Connell, Technical A d viso rs, SPS _i Gerald F. Regan, C hief, A ir S u rve illa n ce Branch On January 27, 1976, the SPS f i e l d team ( M i ll e r , Z y ls t r a , Connell) performed ambient a i r m o n ito rin g f o r P C B 's a t the Monsanto Krum.mricn P lant, Sauget, I l l i n o i s . Although Monsanto p re se n tly manufactures P C S's a t t h is f a c i l i t y , p ro d u ctio n o f P C B 's i s to be g ra d u a lly phased-out. I A ft e r a b r i e f d is c u s s io n and o r ie n t a t io n period w ith Monsanto o f f i c i a l s , ASB and ILDO personnel toured the PCB m anufacturing area in c lu d in g the in c in e ra t o r f a c i l i t y . The "p la n t " area where P C 3 's are manufactured i s re t . an enclosed stru c tu re per se. I t i s an open stru c tu re constructed c f ste e l g r a t in g ( f l o o r s , s t a i r s ) , s a fe t y r a i l i n g s , a maze o f p ip in g , c o lle c t io n and reaction v e sse ls, v a lv e s, etc. a ll o f which are completely exposed to the elements. Mr. C la ir e Buckley o f Monsanto was a ssigned as e sc o rt fe r ASB personnel. O nly two lo c a t io n s were deemed a p p ro p ria te as sam pling s i t e s fo r P C B 's. Other p o s s ib le s i t e s were in a c c e s s ib le o r had re c e n tly been tested by Monsanto - ie . - the recen t in c in e r a t o r t e s t r e s u lt s were fu rn ish e d by Monsanto. These two sam pling lo c a t io n s , the c le a n in g o f glassv.'are, eq u ip ment used, sam pling procedure, samples obtained, etc. i s described bslcw: .Site Location r l fSEG) - Steam E je cto r - Ground L e v e l. T h is steam ejector pipe terminated j u s t above a grate which covered a recessed concrete trough in the flo o r at ground le ve l. For co lle c tio n -o f "the sam ples, the SPS f i e l d team used the g la s s b u b b le rs, Y-connector w ith treducer c o u p lin g , U-tube, cold tra p and vacuum pump. The H-frame base, emd 2x4 m anifold su p p o rt and g la s s m anifold were not used. Sam pling was performed fo r a one-hour p e riod , 1046 - 1146 hours. . S i t e Location #2 (SET) - Steam E je c to r - Top le v e l or 2nd Level Above G r r r i . T h e se twin steam e je c to r p ip e s (somewhat o f a candy-cane c o n fig u ra tio n ) lerm tnated out beyond the 2nd le v e l s a fe ty r a i l i n g . Sample c o lle c t io n was perform ed u sin g the same equipment as a t S it e Location #1 and included the !H-frame base, one se c tio n o f 2x4 m anifold su p p o rt, and one s e c tio n o f g l a s s m anifold. Sam pling w^s performed f o r a one-hour p e rio d , 1205 - 13C5 ` `h o u r s . nUb'.D304 (RV..72) ACP C C f i^ S T h e 'fo llo w in g n a rr a t iv e d e fin e s the process fo r cleaning of the sam pling glassw are , the actual p re p aratio n f o r sam pling, and handling o f the samples collected in the bubbler tra in . The ASB, SPS f i e l d team cleaned a l l PCB m onitoring glassw are v ia the follow ing procedure: 1. Thoroughly wash a l l gla ssw are w ith a s o lu t io n o f PCB fre e detergent The co ncentration o f the detergent i s determined by the manufacture in stru c tio n s. I 2. .Rinse thoroughly w ith tap water. I 3. * jRinse thoroughly w ith d is t ille d water. 4 . R in se th o ro u gh ly w ith acetone ( re a g e n t 'grade o r b e tte r). 5. R in se w ith hexane, r in s e each bubbler three times with' 50 ml. o f hexane f o r each r in s e and d is c a rd . 6. Wrap a l l sm all g la s s p a rts in aluminum f o i l and ensure proper s e a li . crim ping o f the f o i l . Seal a ll b all jo in t s of the bubblers with aluminum f o i l (a lte r n a te ly , app rop riate b a ll j o in t f it t m c s - sealo on one end and th o ro u g h ly cleaned - may be clamped in place in ste a d o f u sin g aluminum f o i l ; the purpose o f the above i s to prevent co n ta m in a tio n -o f the now PCB fre e g la ssw a re ). Upon determ ination o f a sam pling s i t e lo c a t io n , the sam pling equipm s u p p o rtiv e framework i s se t-u p and a ll glassw are bubbler box, pump, etc. are p o sitio n e d . The sam pling procedure i s as fo llo w s: 1* Remove s e a ls ( f o i l o r ground j o i n t s and clamps) from the bubblers. 2.- Place 300 ml. hexane in to bubbler #1, shake, pour into bubbler #2, Shake, pour in to bubbler 3, shake, pour in to a sample bottle a p p ro p ria te ly designated as a reagent blank. This procedure is performed f o r both s id e s CA.and B) o f the sam pling tr a in . \ AC* CG6<ti6 -3 'I 3.*. Pour hexane, 400 m l., 200 m l., and 300 ml. in to bubblers 1, ?2, and i3 , re sp e c tiv e ly . 4. Connect the e n tire sampling system - U-tube, manifold, reducer c o u p lin g , Y-tube, U-tubes between each bubbler and the copper tee (containing hypodermic needles - one fo r each sid e ), the quick d isc o n n e c ts between the tee and the cold trap and between the cold tra p and the vacuum pump. Ensure th a t a ll b a ll j o in t s are t ig h t ly (clamped. Do not use any lu b ric ;.n t/ se a lin g compound on the ground /gla ss ball jo in ts. 5. ( F i l l the plywood box (c o n ta in in g 4 o f the s i x b ub b lers) a n d `the (cold trap with crushed dry ice. * 6. I S t a r t the vacuum pump; note th e time and vacuum gage read in g; sample ,, fo r one-hour. ... 7. At the conclusion o f the one-hour sam pling period, not*vacuum cage re a d in g, t i g h t l y double the rubber hose betv/een the c c p rsr tee a rt .the quick-disconnect thus stopping a ir flow, pull both disccnr.eccs apart, and re le a se the p re v io u s ly clamped o r kinked rubber hose; .shut o f f the vacuum pump. ` 8. Disconnect the Y-tube from the tra in . 9;. D isconnect the b ubb ler U-tubes and the rubber hose - g la s s connection from l a s t bubbler to the copper tee on one s id e o f the tr a in . - IQ*. fo u r the absorbing so lu tio n s from one sid e o f the tra in (3-bubblers) *tnto one sample b o ttle a p p ro p ria te ly id e n t ifie d . A ll sample hotele . ' screvt caps must have T e flo n l i n e r s . A bsorb ing s o lu t io n s and r in s e s , .'from e a c h .sid e o f the t r a i n must be in sep arate sample b o ttle s - 1 B o ttle per s id e o f travn., 1 1 * In tro d u ce 'a p p ro xim ate ly 50 ml. o f hexane in to bubbler t! , shake, p :u r tn to b ubb ler #2, shake, pour in to bubbler 3 , shake, and pour in to th e 'a p p ro p ria te sample b o ttle from Step 10 above. Repeat th is another 50 ml. p o rtio n o f hexane. Repeat f o r the second or other s id e o f the'sampling tra in . ACM CC642 7 -.4 12. In tro d u c e 300 ml hexane in t o bubbler #1, shake, pour in to bubbler #2, shake, pour in to bubbler 3, shake, -and pour in to another a p p ro p ria te ly id e n tifie d sample bottle. T his i s also a reagent blank. Repeat fo r the other sid e of the tra in . j 13 .' The bubb lers can now be p ro p e rly f i l l e d w ith the designated amounts Of hexane (400 m l., 200 m l., 300 m l.) fo r the next sampling period. I f a d d itio n a l sam pling i s not required, the equipment (glassw are) must be wrapped/sealed as defined in Step 6 o f the cle an in g procedure cc:`. Timm Kallgren Torrez Zylstra Connal 1 ACP The ta b le below d e fin e s the v a r io u s samples c o lle c te d at tr.e PCB m anufacturing f a c i l i t y : DATE _: LOCATION 1/27/76 1/27/76 1/27/76 Reagent Blank, Pre (SEG) 1 R1i eagent B lan k, Pre (SEG) Sample (SEG) 1/27/76 Sample (SEG) 1/27/76 1/27/75 Reagent Blank, Pre (SET) 1 I Reagent Blank, Pre (SET) 1/27/76 Sample (SET) .1/27/76 Sample (SET) 1/27/76 . Reagent Blank, A fte r (SEG) Sample 1/27/76 Reagent Blank, A fte r (SEG) Sample 1/27/76 Reagent Blank, A fte r (SET) Sample 1/27/76 Reagent Blank, A fte r (SET) Sample ASB DES I Gt:ATI Cf1 3A-T27-1 * 3B-T27-2 ' 3A-T27-3 / 3B-T27-4 ' 3A-T27-5 ` 3B-T27-6 ` 3A-T27-7 ' ' 3B-T27-8 ' 3A-T27-9 3B-T27-10 3A-T27-11 ' 3B-T27-12 L-3. VO. 76-13555 76-13557 76-13553 76-13559 76-13550 .76-13551 76-13562 76-13553 76-1555; 76-13555 76-13555. 76-13557 r/ps ACH C C 'iC s'/#? /Pit/ i ACH C C e O l AGP CCc**3i 21 of 23 ACM CCt>636 r i Monsanto-Saugct, Illinois Sampling Location Site '1 Steam Ejector-Ground Level (SEC) Site 2 Steam Ejector-fop Lcvcl-(SET) Sample No. Grit iral Or i[ 1re 1*1owra i.e in-*/lionr 1'.' i;.rnt ill;iik Values lli'furn & After 1 K.n li Sample nv,/:;aiiiplc Average Reagent Blank Values ug/s.miplc Sample Values ug/sample DifferenceSample Minus Ave. Blank Value Final Value ug/m3 76-13658 3A-0.71602 76-1 365* I'.07 7.0 / 16.8 6./ / 3.3 11.9 6.0 31.5 25.0 196 21.6 27.5 33.9 76-13662 3A-0.7I402 N.A. / N.A. 76-13663 3B-0.6j602 12.1 / 8.4 10.3 72.5 N.A. S- i i VJ V. I -J /[ _ _ / . ' ..................... ''Monsanto Interpretation of the data chart Column 1- Sampling Location - corresp o n d s to the preceding n arrativ e defining sam pling sites. j1 Column 2- Sam ple N u m b ers - assig n e d by ASB and fo r use by the ana lytical laboratory. Column 3- C ritic a l O rifice F lo w ra te (m 3/h o u r) - calib rated flow rate of the c ritical orifice utilized on each side of the sam pling traim i- i Column 4- Reagent Blank V alues - B efore and A fter Each Sample (ug/ sam ple_ - defines the analytical resu lts determ ined for each reagent blank. "* Column 5- A verage Reagent Blank V alues - sim ply the average of the two values in the preceding column. C olum n 6- Sam ple V alues (u g /sa m p le ) - a n a ly tic a l r e s u lts of the* sa m p le s subm itted. i Column 7- D ifference - Sample Value Minus A verage Blank Value - defines a value representing only the sam ple. Column 8- Final Value (ug/m 3) - defines the final or actual value of F C B 's obtained at the p a rtic u la r sam pling location; determ ined by divid in g the tru e .sam pling a ir flo w rate in to the c o rre c te d value (Column 7). .\ AC*. C'Cti<t'35 Environmental Polychlorinated Biphenyl Contamination near Site s of Manufacture and Use I prepared by Environmental Science and Engineering, Inc. P.0. Box 1354, G ain e sville , Florida 32604 for OFFICE OF TOXIC SUBSTANCES ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460 Contract 68-01-2978 \ July 1975 ^ P D U N T IF P S ^ ACM 006*1*3 INTRODUCTION Polychlorinated biphenyls (PCBs) have a basic chemical structure as follows: t ' Cl Cl From one to ten chlorine atoms may be attached to the structure re- , s u itin g in 210 p ossib le isomers. PCB's have certain physical properties; including high heat capacity, chemical s t a b ilit y , and excellent d ia le c tric properties which make them high ly desirable for a number of in d u stria l uses. Concern has been expressed in recent years over the widespread r d isp e rsio n 'o f PCB's in the environment. These compounds are suspected to have s ig n ific a n t adverse effects on organisms at the top of the food web, including man. Biom agnification of th is contaminant in natural predator-prey food chains has been documented by numerous studies. To complicate the problem, the rate o f degradation of PCB's in the environ ment, p a rtic u la rly the higher substituted isomers, is postulated to be extremely slow. This situ a tio n allows for the possibly sig n ific a n t accumulation of these p e rsiste n t chlorinated organic compounds in the environment. A thorough discussion of the problem and of the environ.mental data that are available has been presented by S e lik o ff (1972): and the U.S. Government Interdepartmental Tasks Force (1972). v AOK CCt**1* In response to the r is in g concern over PCB's in the environment, the' I sin gle domestic manufacturer of .these m aterials (Monsanto Co.) has | ' v o lu n ta rily taken two actions to reduce the environmental hazard. The trade name for these domestically-manufactured compounds is Aroclor. Beginning in 1970, the production of Aroclor was v o lu n ta rily reduced by the manufacturer and supply was discontinued to those users who could not control release to the environment. Production of the more high ly chlorine substituted Aroclor mixtures has since been curtailed i or discontinued and a low er-substituted, refined mixture (Aroclor 1016) has been introduced. Some researchers have concluded that the lower] substituted PCB isomers do not p e rsist as long in the environment as the higher substituted isomers, and that they may be amenable to b io lo gic a l degradation. Aroclors are s t i l l employed in e le ctrica l capacitors", transformers, vacuum pumps and gas-transm ission turbines (Hutzinger.i i Safe and Zitko, 1974). There are a number of foreign producers of PCB. The major foreign products are liste d in Table 1. Some domestic in d u strie s, whose supply of Aroclor has been curtailed by the producer, have turned to foreign supplied PCB's. The extent of the use of these foreign products in the U.S: i s not know at the present time. The Environmental Protection Agency, Office of Toxic Substances has , undertaken an envirpnmental monitoring program to assess the magnitude and potential hazard of PCB contamination of the environment and to assess the effectiveness of voluntary controls on use of these compounds. The f i r s t phase of th is mor^toring program has been completed. The in it ia l e ffo rt was directed toward the evaluation of PCB contamination A C K C C 6 4^.5 Table 1. THE WORLD'S MAJOR PRODUCERS OF PCB* Producer Monsanto Bayer Prodelec Country U.S.A. and Great B ritain Germany France Kanegafuchi M itsu b ish iMonsanto Caffaro Sovol Chemko Japan Japan Italy U.S.S.R. Czechoslovakia Tradename of PC3 Aroclor Clophen Phenoclor and Pyralene' Kanechlor Santotherm l Fenclor after Hutzinger, Safe and Zitko (1974), \ ACM CC644fc of s o ils in the v ic in it y of potential in d u stria l sources and of wastl- jI waters from these sources. ! So il samples have been taken in the v ic in it y of Yates Manufacturing Co. in urban Chicago, I l l i n o i s ; Valcast, Inc. in Troy, Michigan, a suburb of D etroit; and the Monsanto Co., Krumnrich Plant in Sauget, I l l i n o i s , near East St. Louis. P rio r to 1972, the Yates Manufacturing Co. used Aroclors 5460, 6090 and 5442 in the production of wax mold material for the investment casting (lo s t wax) process. These mixtures contain both PCB and polychlorinated terphenyl (PCT). Subsequent to 1972, th is company has been purchasing a PCB of foreign o rig in . Fenclor DK, a 1 product of Caffero Corporation in Milan, It a ly , has been su b stitu te d 1 for the unavailable Aroclors. Valcast, Inc. is an investment casting f a c ilit y . I t has been reported that Valcast purchases the necessary' wax compounding m aterials from Yates Manufacturing Co. The Krumnrich Plant of Monsanto Co. is the only PCB production f a c ilit y in the U.S: A ll domestic Aroclors are produced at th is location. ACM CC6<t47 EXPERIMENTAL Sampling Methods Surface s o il samples were collected at each s ite from the plant i t 1 boundary to a distance of approximately one mile at h mile in te rvals j in each direction subject to the physical constrain ts encountered. At each sample point, fiv e sub-samples were collected within a radius of 10 m to a depth of 2.5 cm over a 100 cm2 area. This mode of sampling, which was designed to c o lle ct a homogeneous surface s o il sample representative of the chosen sample point, is illu stra te d in Figure 1. Vegetative matter was removed with forceps and the fiv e , i sub-samples were composited and thoroughly mixed. Every attempt was 1 made to select sample points where surface vegetation was sparse and ' where i t appeared u n lik e ly wind scour would be a major so il transport factor. Samples were placed in pre-cleaned, wide-mouth glass jars with aluminum f o il- lin e d closures. i Water samples were collected from a drainage ditch adjacent to the i Valcast plant, of Valcast effluent cooling water and from the sanitary sewer system downstream of Valcast. Wastewater samples were taken from the in flu e n t and e ffluent of the Sauget, I l l i n o i s sewage treatment plant. The Monsanto f a c i li t y uses th is treatment plant for wastewater treatment. These la tte r water samples were 24-hour composites, of the treatment plant influen t and effluent; the former were grab samples. A ll samples were taken in 1 1. glass containers, composited and stored in 4 1. pre-cleaned glass-stoppered reagent bottles. vi A D M CCfc44fc V \ Figure 1. Diagram showing surface s o il sampling method. Not to scale. Five sub-samples are composited to y ie ld a single sample. CC c * 5 Extraction and Concentration The s o il sample is thoroughly mixed and sectioned into quarters. Two of the quarters are rejected and the remaining quarters are again mixed and quartered. This is continued un til a representative sample size of about 50 g is obtained. This sample fraction is allowed to a ir dry at room temperature. PCBs are extracted from the s o il using a soxhlet apparatus and 75 ml of 1:1 nanograde hexane/acetone for a 10 g so il sample. The extract is concentrated to 5 ml on a Kuderna-Danish apparatus and dried with sodium sulfate. This concentrate is evapo rated to dryness under dry N2 at room temperature then redissolved in 1 ml of nanograde hexane. This procedure b a sic a lly follows that described by the EPA (1974). Water samples are t r ip le extracted with 15 percent nanograde ehtyl ether in hexane, dried with sodium su lfa te , concentrated on a Kuderna-Danish apparatus, evaporated to dryness under N? and redissolved in 1 ml nanograde hexane. This method is sim ila r to that recormended by the EPA (1973.). Sample Cleanup Both the s o il and water extracts are cleaned-up by the use of a s ilic a gel column. The column is eluted with nanograde pentane followed by nanograde benzene. The PCBs present in the extract appear in the pentane fraction ; pesticid es appear in the benzene fraction. The pen tane fraction i s further treated with concentrated H2 SO4 to eliminate unstable organic compounds. The s ilic a gel separation procedure is described by Snyder and Rennert (1971). The s u lfu ric acid cleanup was adapted from Murphy (1972). ACM CC6<5C Analytical Methods All analyses are by gas chromatography. A dual-column Varian 2760 instrument is employed with the following conditions: a) detection and quantifications glass column, 6' length, 1/8 ID 1.53/1.95% 0V-17/QF-1 liquid base Chrom W-HP, 80/100 mesh support N- 38 psig inlet pressure, 68 ml/min detector temperature 225 C column temperature 200 C injector temperature 215 C b) confirmation stainless steel column, 5' length, 1/8 ID 1.530V-101 1iquid phase Chrom G-HP, 100/120 mesh support No 30 psig inlet pressure, 43 ml/min detector temperature 225C column temperature 205 C injector temperature 210 C Detection is by electron capture using a tritium detector. Mass spectrographic confirmation has also been obtained on selected samples. The PCBs are quantified in a two-step process. First, the elution pattern is compared to those of standard mixtures and the closest pat tern match is considered to be the PCB present. The relative peak heights in a mixture are a good indication of a match with a standard, as is the presence or absence of later eluting peaks. Good matches with standard Aroclor mixtures have been obtained. The presence of decachlorobiphenyl is easy to ascertain since it is a single peak. Once the elution pattern is subjectively matched to a standard PCB pattern, quantification is conducted by electronic integration of peak areas using a Hewlett-Packard Integrator Model 3380A. The total area of all accepted matching peaks in the elution pattern of the sample is compared ALM C C 4 5 1 with the total area of the same peaks in the standard mixture and a sample concentration i s calculated. The A roclors used for preparation- of standards were obtained from the Monsanto Company and the EPA Southeast Environmental Research Laboratory. Fenclor DK was provided by EPA Region V personnel. Sample Replication and Recovery Studies A number of replicate analyses were conducted to determine the repro^ d u c ib ilit y of both the analytical method and the sampling method. The re su lts of these re p lic a tio n s are tabulated in Tables 2 and 3. A ll re p lic a tio n s conducted to date are of so il samples. Table 2 l i s t s the re su lts of the a n aly sis of two fraction s of a sin g le so il sample. Table 3 l i s t s the re su lts of the an alysis of replicate samples and, hence, represents both sampling and analytical v a r ia b ility . The re su lts o f several PCB recovery studies are tabulated in Table 4. A C M CCe<52 Table 2. ANALYTICAL REPLICATES OF SOIL SAMPLES Reolicate Y-l " V-1 V-2 V-3 V-4 V-5 V- 6 M-l H- 2 M-3 M-4 Aroclor 1260 (ppm) 0.86 0.51 mean 0.69 <0 . 0 1 0.11 mean 0.06 0.07 <0 . 0 1 0.01 mean 0.04 mean 0.13 0.11 0.12 0.13 0,15 mean 0.14 <0 . 0 1 <0 . 0 1 <O T <0 . 0 1 0.03 . 0.04 mean 0 . 0 2 0.06 0.13mean 0 . 1 0 0.21 0.30 mean 0.26 1.8 0.95 mean 1 ,4_ * Decachlorobiphenyl 0.65 0.50 mean 0.58 0.084 0.16 mean 0 . 1 2 0.066 0.096 mean 0.081 0.95 0.34 mean 0.90 0.49 0,67 mean 0.58 ACM 0 0 8 * 1 5 3 Table 3 . ' SAMPLE REPLICATES OF SOIL SAMPLES Replicaie Y-l M-l M-2 M-3 Aroclor 1260 (ppm) 0.53 0.40 mean 0.47 9.8 9.3 mean 9.6 0.12 0.46 mean 0.29 0.23 <0 . 0 1 oTTI Decachlorobiphenyl (ppm) 1.0 1.4 mean 1 . 2 1.3 V.O mean 1 . 2 0.015 0.069 mean 0.042 0.64 0.58 mean O'. 61 ACH CG645<t Table 4; RESULTS OF RECOVERY STUDIES FOR PCBs PCB Type Aroclor 1242 Aroclor 1260 Aroclor 1260 Percent Recovery 105 85 122 RESULTS A,ND DISCUSSION PCBs were detected in s o ils at a ll three sampling locations. The elution patterns obtained most closely matched Aroclor 1242, Aroclor 1260 and decachlorobiphenyl. S o il concentrations ranged from the detection lim it of 0 . 0 0 1 ppm to over 20 ppm. Data from V ic in ity of Yates Manufacturing Co., Chicago, IL. Figure 2 illu s t ra t e s : a) a chromatogram of Fenclor DK, the foreign supplied PCB presently used by Yates Mfg. Co. in the production of investment casting wax compounds; b) a chromatogram of a standard Aroclor 1260 and; c) a typical chromatogram of a so il extract taken in the v ic in it y of the site . Figure 2a demonstrates that Fenclor DK is nearly pure decachlorob'iphenyl. This compound is unusually pure for a technical grade material. One minor peak elutes ju st before the predominate decachlorobiphenyl peak. The retention time of the minor contaminant peak is such that i t may be expected to be an octachlorobiphenyl since it exhib its the same retention time as the la te st eluting peak in a standard Aroclor 1260 mixture. I t is apparent by comparison of peak retention times with the sample chromatogram, Figure 2c, that decachlorobiphenyl is present in the soil* sampled. The measured concentration of decachlorobiphenyl in th is p articu lar sample was 1.0 ppm. The presence of decachlorobiphenyl in th is sample has been confirmed by mass spectrometry. Figure 3 illu s t r a t e s a portion of the mass spectrum of the sample and a standard decachlorobiphenyl obtained from a different source. The sample elution pattern ajso indicates the presence of Aroclor 1260 as can be seen by comparison with the Aroclor 1260 elution pattern. AGh C C c O (a) Fenclor DK r i ( J Figure 2. Typical chromatograms of standard PCBs and of a so il sample taken in the v ic in it y of Yates Mfg. Co., Chicago, Il l i n o i s . Cf* CC8*t57 Relative Intensity t* n X r> om s- \_r o> ritjuri* 3. (a) P o r t i o n o f ili s s spec t run o f sample taken in v i c i n i t y of Yates Mhj. Co. (1)) Parent rtolecule nas s c l u s t e r of d e c d i h l n r o h i p h n n y l . The concentration of Aroclor 1260 in this sample is 0.53 ppm. There is no indication of degradation of the Aroclor 1260 since virtually all peaks in the sample chromatogram are present in the same relative propor tion as in the standard. There are no significant, undefined peaks in the chromatogram following the acid cleanup procedure. Certain of the sample chromatograms indicate the presence of lower weight PCBs, however the quantities present tend to be far less than either Aroclor 1260 or decachlorobiphenyl at this site. Figure 4 depicts the quantity of Aroclor 1260 measured in samples taken in the vicinity of Yates Mfg. Co. There exists no apparent spatial pattern of Aroclor 1260 distribution over the study site. There is lit t le to indicate the source of the Aroclor 1260 contamination is Yates Mfg.Co. Aroclor 1260 has not, in fact, been used by Yates. Prior to 1972 Aroclors 5442, 5460, and 6090 were purchased by this company. Subsequent to 1972, Fenclor DK and Aroclor 5460 have been in use. Aroclors 5442 and 5460 are terphenyl mixtures with 42 percent and 60 percent chlorine by weight, respectively. Aroclor 6090 is 90 percent terphenyl and 10 per cent Aroclor 1221. Aroclor 1221 does not appear in the soils at this site. Figure 5 illustrates the distribution of decachlorobiphenyl in soils near the Yates facility. The distribution has some elements of a pattern of higher levels closer to the Yates facility with decreasing concentration with distance. Such a pattern may be expected if Yates were the source of the decachlorobiphenyl (or Fenclor DK) contamination. ACM CC6<55 I Figure 4. Concentration of Aroclor 1260 in soil as a function of distance from Yates Mfg. X;o. Concentrations are expressed in ppm. c- ACfc C C e fcC Figure 5. Concentration of decachlorobiphenyl in soil as a function of distance from Yates Mfg. 0. Concentrations are expressed in ppm. C K CC6<tfcl Data from Vicinity of Valcast Corp., Troy, HI. Valeast Corp. is an investment casting facility located in a small indus trial park in suburban Detroit, MI. PCBs are a constituent of the wax mold compound used to fashion intricate shapes which are to be cast. Figure 6 is a typical chromatogram of a soil sample extract taken fron this location. Aroclor 1260 is present in many of the soil samples. Decachlorobiphenyl is absent in the soils. Figure 7 shows the concen trations of Aroclor 1260 observed in the vicinity of Valcast Corp. No distribution pattern is evident. The one observed concentration of 18 ppm Aroclor 1242 appears to be anomalously high, however, replicate analysis yielded values of 16 ppm and 19 ppm, respectively. Aroclor 1242 was not detected in any other soil samples from this area. A small drainage ditch passes adjacent to the north boundary of the Valcast facility. This ditch serves to remove stormwater runoff in the vicinity and receives discharged cooling water from Valcast and other local small industries. Analysis of the Valcast cooling water at the point of discharge and of water in the drainage ditch failed to detect PCB levels greater than the detection limit of 0.1 yg/1. Two bottom sediment samples taken from this drainage ditch, however, had concentra tions as follows: Aroclor 1242 2.3 ppm 9.4 ppm Aroclor 1260 6.7 ppm 3.9 ppm Decachlorobiphenyl 0.09 ppm 0.11 ppm Three grab samples of water and bottom sludge were taken from the sanitary sewer system aboutM/4 mile down stream of the Valcast facility. ACM C C8 e2- J- (a) 38 57 75 min. - h -+ H-----------:---- 9*7900 wnv W M l U A a A -,. .... --J riyurti G. Typical chromatograms of soil samples taken ir the vjcinity of Valcast Corp., lroy, Micliiyan--(d) *0.01 ppm, (b) 0.014 ppm Aroclor 1260 I I N V Figure 7 . Concentration of Aroclcr 1260 in soil as a U n c tio n r f distance from Val cast "fg. Cc. Concentrations ars expressed ir ncr. ACM CC Table 5. ANALYTICAL RESULTS OF SAMPLES TAKEN FROM THE SANITARY SEWER - SYSTEM OF TROY, MICHIGAN. AT INTERSECTION OF ELLIOT AND EXECUTIVE DRIVE. Date Time Type Sample 23 Feb 75 1530 water 22 Feb 75 1815 water 23 Feb 75 m s water 22 Feb 75 bottom sludge Aroclor 1260 0.1 wg/1 0.1 wg/1 7.0 wg/1 0.11 ppm Deeach1,oreb i oha nv1 <0.01 wg/1 <0.01 wg/1 4.1 wg/1 0.034 ppm iCM CCSMt This sewer system serves numerous small in d u strie s in the area. Results of the an alysis of these samples are tabulated in Table 5. It appears that some Aroclor 1260 and decachlorobiphenyl are finding their way into the sewer system. In su ffic ie n t information is available to identify the source of these PCBs. Data from V ic in ity of Monsanto Co., Sauqet, IL. Aroclors have been produced by Monsanto Co. at the Sauget, Il l i n o i s s ite since p rio r to 1957. PCB mixtures ranging from 16 percent chlorine to 68 percent chlorine have been produced over the years. Production of the higher percent chlorine mixtures has been v o lu n ta rily reduced since 1970 and lower substituted isomers have been introduced. Polychlorinated terphenyls and chlorinated benzenes have also been produced at this f a c ilit y . I t is reported (Papageorge, 1975) that terphenyl production was suspended in 1971. A typical chromatogram of a s o il sample taken in the v ic in it y of the Monsanto f a c ilit y is shown in Figure 8 along with the chromatogram of several other PCBs run under the same instrument conditions. This p articu lar so il sample contains 11 ppm Aroclor 1242, 9.3 ppm Aroclor 1260 and 1.0 ppm decachlorobiphenyl. Lighter chlorinated Aroclor mixtures are lik e ly present as w ell. The instrument conditions employed, however, do not give adequate separation of these mixtures to permit accurate quan tification. The .presence of decachlorobiphenyl at the concentrations observed is interesting. Decachlorobiphenyl has been reported as a compo nent of only one Aroclor mixture, Aroclor 1268 (Hutzinger, Safe and Zitko, 1974). Whether it may be a tomponent of any of the terphenyl mixtures or biphenyl/terphenyl mixtures or a by-product i s not known. A C M C C 6 4 1 fc (a-) A roclor 1242 ti 'J v " ' J u (b) Decachlorobiphenyl J Figure 8 . Typical chromatograms of standard Aroclors, decachlorobiphenyl and a soil sample taken in the v ic in it y of Monsanto Co., Sauget, Il l i n o i s . alk i The measured so il concentration of Aroclor 1242, Aroclor 1260 and decachlorobiphenyl in the v ic in it y of the Monsanto f a c ilit y are depicted in Figures 9, 10 and 11, respectively. Figure 12 shows the long-term average wind direction in the area. The d istrib u tio n of a ll PCBs analyzed appears to be higher near the plant s ite and generally decreasing with distance from the s ite . Furthermore, there is evidence that g en erally higher concentrations are present in the s o ils located to the southeast. This corresponds to the predominant wind direction and may suggest an airborne transport of the PCBs from the f a c ilit y . One 15 cm deep core sample was taken 1/2 mile southwest of the Monsanto site . A n alysis of the bottom 11 cm of th is s o il yielded PCB concentrations b a sic a lly equivalent to the top 4 cm. A 24-hour composite sample from the influent and the effluent of the Sauget v illa g e sewage treatment plant taken beginning 0B00 on February 17, 1975, was provided by plant personnel. The Monsanto Co. jo in tly owns this f a c ilit y . A nalysis of these samples were as follow s: Aroclor 1242 Aroclor 1260 Oecachlorobiphenyl influent 6 . 8 pg/1 8.7 pg/1 <0 . 0 1 pg/1 effluent 4.5 pg/1 4.7 pg/1 <0 . 0 1 pg/1 The treatment plant effects a reduction of 34 percent Aroclor 1242 and 46 percent Aroclor 1260. No decachlorobiphenyl was detected. \ l Al> CCHCt '" 'S I Figure 9. Concentration of Aroclor 1242 in so il as a function of distance I from Monsanto Mfg. Co. Concentrations are expressed in ppm. i C P CCd<tfc5 IV. I \ Figure 10. Concentration of Aroclor 1260 in so il as a function of distance from Monsanto Mfg. Co. Concentrations are expressed in ppm. v A C * CCfcWO i I Figure 11. Concentration of decachlorobiphenyl in s o il as a function of distance from Monsanto Mfg. Co. Concentrations are expressed in ppm. ACM CC6*t 7i ? N Figure 12. Ten Year Averaged Wind Rose for St. Louis, M issouri, January through March, 1951-1960. \ Scale: 1" = 5% Calm Winds = 2 .OS i ALP CCc Hit REFERENCES Environmental Protection Agency (1973). Meth'od fo r Polychlorinated Biphenyls (PCB's) in In d u strial Effluents. NERC, Cincinnati, OH. Environmental Protection Agency (1974). A n alysis of Pesticide Residues in Human and Environmental Samples. Pesticides and Toxic Substances Effects Laboratory, Research Triangle Park, N.C. Hutzinger, 0.; Safe, S .; and Zitko, V. (1974). The Chemistry of PCB's. CRC Press, Cleveland, OH. 269 p. Murphy, P.G. (1972). S u lfu ric acid for the cleanup of animal tissu e s for analysis of acid-stable chlorinated hydrocarbon residues. JAOAC 55:6; 1360-1362. Papageorge, W.B. (1975). Personal coimunication. Manager, Product A cceptability, Specialty and Process Chemicals, Monsanto In d u strial Chemicals Co., St. Louis, MO. S e lik o ff, I. J . (1972). ed. PCBs - Environmental Impact. Environ. Res. .5:3; 249-362. U.S. Government Interdepartmental Task Force on PCBs (1972). PCBs and the Environment. NTIS document C0M-72-10419. \ ACK CCb CONCLUSIONS 1 . The d istrib u tio n of Fenclor DK suggests that th is compound may be transported to s o ils in the v ic in it y of Yates Mfg. Co. by way of airborne emissions. 2. A roclor 1260*is present in urban Chicago s o ils . There is no evidence Yates Mfg. Co. is the source of th is s o il contamination. 3. Detectable, but lower le v e ls, of Aroclor 1260 are present in s o ils in Troy, Michigan. There is no evidence Valcast Corp. is the source of the s o il contamination. No decachlorobiphenyl (or Fenclor DK) was detected in s o ils in the area. 4. S ig n ific a n t q uan tities of Aroclor 1242 (18 ppm) were observed in a sin gle s o il sample taken in a resid ential area near Valcast Corp. The importance of t h is sin g le data point should not be over-emphasized. However, Aroclor 1242, Aroclor 1260 and decachlorobiphenyl were also detected in drainage ditch sediments at much higher concentrations than the observed PCB le ve ls fo r s o ils in the area. 5. The drainage ditch sediments near Valcast Corp. contained detectable le v e ls of decachlorobiphenyl. It s source is unknown. A number of small in d u strie s are located 1n the immediate v ic in it y , any of which could be the source as well as interm ittent discharge, leakage or runoff from the Valcast site or fa c ility . 6 . One of three grab samples taken over a period of tine from the sa n i tary sewer system near Valcast Corp. contained detectable levels of Aroclor 1260 (7.0 pg/1 and decachlorobiphenyl (4.1 pg/1). The other two samples were below detection lim its of 0 . 0 1 pg/1 and 0 . 0 0 1 pg/1 , respectively. Sewer sludge contained both Aroclor 1260 and decachlorobiphenyl. These AC* CC647*t m aterials are apparently discharged to the sewer system on an in te r mittent b asis. The source is unknown. 7. Decachl.orobiphenyl is re la tiv e ly abundant compared to the other PCBs in s o il s in the v ic in it y of the Monsanto f a c ilit y . Decachlorobiphenyl is not reported to be a major component of any of the Aroclor mixtures. It may be a by-product associated with Aroclor production which is eliminated or greatly reduced in the commercial preparations by refin e ment at the Monsanto f a c i li t y and may, therefore, find i t s way into the environment at th is location. 8 . - Aroclors 1242 and 1260 were also present in s o i l s near the Monsanto plant. There appears to be a tendency for higher PCB levels close to the plant with decreasing concentration with distance. There is some evidence that higher s o il concentrations e x ist along the northwest to southeast axis which corresponds to the dominate wind direction at th is location. This suggests airborne transport of the PCBs from the plant with subsequent deposition, of at le ast the heavier isomers of PCB, on nearby s o ils . 9. PCBs are present in Sauget, I l l i n o i s wastewaters. The sewage treatment plant removes le ss than 50 percent of the Aroclor 1242 and Aroclor 1260. Decachlorobiphenyl is not present in these wastewaters. 10. Aroclor 1260 was found to be present in the s o ils of a ll three site s sampled. This product is currently used in e le ctrica l transformer flu id s and has been used in the past in hydraulic flu id s , p la stic iz e rs and dedusting agents (Hutz\nger, Safe and Zitko, 1974). Aroclor 1242 ACM CC8<t75 was found in comparable concentration near the Monsanto f a c ilit y and in drainage ditch sediments near Valcast. This product has consistently been the most heavily produced* by Monsanto of a ll the PCB mixtures. 11. Decachlorobiphenyl may be more widely spread than would be a n ti cipated by projection from Monsanto production figu re s for Aroclors. \\ A t * CCfc^Tt J su g g est io n s for CONTINUED STUDY OF PCBs 1. Conduct total PCB a n a ly sis employing the perchlorination technique o f selected samples on hand.- Determine what fraction o f the total PCB pre-sent has been quantified as Aroclors 1242 and 1260 and decachlorobiphenyl. 2. Conduct polychlorinated terphenyl (PCI) an alysis o f selected samples on hand. 3. Assess PCB and PCT le v e ls in various urban and non-urban s o ils and stream sediments. 4. Conduct a n a ly sis o f airborne and precipitation-borne PCB and PCT in the v ic in it y of known users. 5. Evaluate PCB and PCT le v e ls in and near dumps and la n d f ills in the a ir , water, s o il and groundwaters. 6. Investigate the potential fo r chlorination of biphenyl at normal wastewater and d rin kin g water treatment le ve ls. 7. Investigate the p o s s ib ilit y of chlorinated dlbenzofuran, dioxin and naphthalene contamination of various foreign and domestic PCB products. 8. Evaluate the rate o f degradation of PCBs in the environment. vv A C H CCfi^7 7 ,,nU iU/Ky UNITED STATES ENVIRONM ENTAL PROTECTION AGENCY WASHINGTON. D C 2C4Q r September 23, 1975 Dr. Robert E. Keller Manager, Applied Sciences Monsanto Industrial Chemicals, Inc. 800 North Lindbergh Blvd. St. Louis, Mo. 63166 Dear Dr. Keller: In reference to our Las Vegas conversation, the following are attached: 1. "Environmental Levels of PCB's" 2. PCB data from the Sauget, Illinois area. If possible, I would be interested in having a copy of the PCE list of references you had' compiled. Call if I can be of further help. My phone number is 202-755-6956. Sincerely yours, Vincent J. DeCarlo, Chief, Monitoring and Information Systems Branch Office of Toxic Substances \ ACM February 25, 1972 Dr. Scocc Tucker Monsanto Company Organic Chemicals Division 1700 Souch Second Street Sc. Louis, Missouri 63177 Dear Dr. Tucker: On May 22 and 23, 1972, chere will be an informal inter agency work session on PCS's in the environment, co be held ac che Holiday Inn ac Duluch, Minnesota. Our objecclves are co assess che presenc scace of knowledge and efforc, and co identify additional study need priorities. You (or in some cases, a member of your -agency) are invited to participate. A copy of che proposed agenda and time schedule is enclosed. Please let us know by April 1, 1972, if you will attend. Additional information will be sent later regarding reservation times and places. Very truly yours End Donald I. Mount, Ph.D Director \ ACM CC647S i INTERAGENCY MEETING ON PC3's IN THE ENVIRONMENT Monday Morning. May 22, 1972 WELCOME Donald I- Mount, ?h.D. arrangements AND ANNOUNCEMENTS STATE OF IKE ART Presentation of present efforts by attendees. THE OUTLOOK EPA ENFORCEMENT Monday Afternoon THREE CONCURRENT SESSIONS ON NEEDS This session, after its discussion regarding regulation, will split with attendees going to the session of their choice, either B or C. A. Regulation - Head and Srini Use of isomer measurements quantification of PCB's Residue tolerances, etc. Oil-sediment regulations Discharge permit usefulness Etc., etc. 2. Analytical - Donaldson and Veith Separation of isomer and aroclors Clean-up Precision accuracy Positive identification Labeled compounds Monitoring needs Sensitivity Type sample Preservation v ACM CC6WS I C. Biological Effects - Additional toxicity studies Aquatic Terrestrial Residue studies and half-life Food chain studies Interaction with pesticides Duke and Schoetcger INTEREST GROUP Monday Evening Informal Meetings (about 3) Tuesday Morning, May 23, 1972 PRESENTATION BY EACH OF THREE WORK GROUPS Head, Donaldson and Duke MEETING SUMMARY AND IDENTIFICATION OF NEEDS AND PRIORITIES V # Monsanto January 14, 1976 Hr. Robert D. Kleopfer USEPA Region V II Laboratory 25 Funston Road Kansas C ity , Kansas 65115 Dear Hr. Kleopfer: Per our phone conversation o f 1-14-76, I am sending you a copy o f ASTM Method D-3304, which was developed from analytical methodology provided by Monsanto. I am also enclosing a copy o f my presentation at the National Conference on Polychlorinated Biphenyls in Chicago on November 19, 1575. .This paper describes the p rop erties and composition of PCB'mixtures. The text I recommended as a reference fo r further information is lis te d as reference 1 in the paper. I hope th is information i s useful to you. Sincerely, ss Enclosures(2) bcc: R. E. K e lle r J. C. Ueber D. Wood \ J. P. Mieure, Ph.D. Research Group Leader aunn ofMontante Company CM CCS4G1 fi^TiniAL row *HAT IftICPiTMM UNITED STATES GOVERNMENT Memorandum t o : see Belov ENVIRONMENirti* PROTECTION AGENCY National Water Quality Laboratory 6201-Congdon Boulevard Duluth, Minnesota 55804 DA7Z: M ay 2, 1972 FRO : Director,. NWQL, Duluth, Minnesota SUBJECT: Interagency Meeting on FCB's In the Environment The progress for the "Interagency Meeting on PCB's In the Environment" scheduled for May 22 and 23, 1972 to be held In Duluth, Minnesota has now been finalized. The program remains essentially the same as the one which was announced earlier and a copy is enclosed. Starting time will be 9:00 a.m., May 22, 1972 In the Conference Room at the Duluth Holiday Inn. For the presentation of present efforts by attendees It Is requested that one individual present the efforts of his group or laboratory. Presentations should be limited to ten minutes or less and be strictly informal. We would expect presentations from the following groups: 1. Fish Pesticide Laboratory 2. Southeast Water Laboratory 3. Perrine Primate Laboratory 4. Chamblee Toxicology Laboratory 5. Monsanto 6. Justice Department 7. Great Lakes Fisheries Laboratory 8. National Marine Water Quality Laboratory 9. Analytical Quality Control Laboratory 10. Gulf Breeze Laboratory 11. National Water Quality Laboratory 12. Food and Drug Administration 13. Patuxant Wildlife Research Canter, and 14. Department of Fisheries and Wildlife, Michigan State University. If other individuals desire to make a presentation of important findings or facts, please contact John I. Teasley, National Water Quality Laboratory, Duluth, Minnesota 55804. For Monday evening a group dinner is belhg planned preceded by a cocktail hour and followed by the interest group discussion. This will be an excellent time.to meet other attendees to discuss mutual problems. V B u y U S . S a v in g s B o n d i R e g u la rly on the P a y r o ll S a v in g s P la n A C ? C C 64 6 Interagency Meeting on FCB's In the Environment 2 Those who do not have accommodations In Duluth as yet can obtain them directly at the Duluth Holiday Inn (please specify the National Water Quality Laboratory FCB Meeting), or If you desire, contact John Teasley vho will make the necessary arrangements for you. Likewise, If possible, a schedule of your arrival time In Duluth would be helpful. Attachment: Program (2 pgs) y Donald I. Mount, Ph.D. TO: Mr. William T. Donaldson ) Dr. Ronald Webb ) Dr. Leo Azarraga ) Mr. George Baughman ) Southeast Water Laboratory, Athens, Georgia Dr. Thomas Duke ) Dr. Del Nimmo ) Mr. Albert Wilson ) Mr. John Couch ) Gulf Breeze Laboratory, Gulf Breeze, Florida Dr. Richard Schoettger ) Fish Pesticide Laboratory, Dr. David Stalling ) Route #1, Columbia, Missouri 65201 Mr. August Curley, Chamblee Toxicology Laboratory, Chamblee, Georgia 30341 Dr. Henry F. Enos, Perrine Primate Laboratory, Perrine, Florida 33157 w m i a r n Papageorge ) Mons Dr. Scott Tucker ) St. LoulSi Missouri 63166 Dr. Gilman Veith, University of Wisconsin, Madison, Wisconsin 53706 Dr. Elizabeth Anderson, Environmental Protection Agency, Washington, D.C. 20460 Dr. John Buckley, Environmental Protection Agency, Washington, D.C. 20460 Dr. Jerry Burke, Food and Drug Administration, Washington, D.C. Dr. Stan Hegre ) National Marine Water Quality Laboratory Dr. Peter Rogerson ) West Kingston, R.l. Dr. Howard Johnson, Michigan State University, East Lansing, MI 48823 Mr. William Potter, Food and Drug Administration, Minneapolis, Minnesota Mr. Wayne Willford. Great Lakes Fisheries Laboratory, Ann Arbor, MI 48107 Dr. John Hesse, Michigan Department of Natural Resources, Lansing, MI 48895 Mr. Jim Lichtenberg ) nr. Tom s a n a r ) Quality Control Laboratory, Cincinnati, OH \ ACM CG6463 INTERAGENCY MEETING ON PCB'a IN THE ENVIRONMENT Monday Morning, May 22. 1972 WELCOME Donald I. Mount, Fh.D. ARRANGEMENTS AND ANNOUNCEMENTS STATE OF THE ART s Presentation of present efforts by attendees. THE OUTLOOK EPA ENFORCEMENT Monday Afternoon THREE CONCURRENT SESSIONS ON NEEDS This session after Its discussion regarding regulation will split with attendees going to the session of their choice either B or C. A. Regulation - Walpole and Srlnl Use of Isomer .measurements quantification of PCB'a Residue tolerances, etc. 011-sediment regulations Discharge permit usefulness Etc. etc. B. Analytical - Donaldson and Velth Separation of Isomer and aroelora Clean-up Precision accuracy Positive Identification Labeled compounds Monitoring needs Sensitivity Type sample Preservation * \ ACK c c a '. c C' Biological Effects - Additional toxicity studies Aquatic. Terrestrial .. Residue studies and half-life Food chain studies Interaction with pesticides Duke and Schoettger INTEREST GROUP Monday Evening Informal Meetings (about 3) Tuesday Morning. May 23, 1972 PRESENTATION BY EACH OF THREE WORK GROUPS Walpole Donaldson and Duke MEETING SUMMARY AND IDENTIFICATION OF NEEDS AND PRIORITIES \ ACM c e s a t o UNITED STATES ENVIRONMENTAL PROTECTION AGENCY DEC 29 1975 WASHINGTON. O.C. 20M0 ' December 22 1975 OFFICE OF THE ADMINISTRATOR Dc*r Hr. Hanley * For *c least five years, the Federal Government and the manufacturing Industry have been evare that polychlorinated biphenyls (PCBa) pose a threat to human health and the environment. Last month, at EPA's National Conference on Polychlorinated Biphenyls In Chicago, leading experts clearly documented the fact that PCBs are pervasive In the environment, are causing disruption and economic harm to commercial fishing in many areas and are creating a serious potential health hazard to the public. While certain voluntary measures have been Initiated In the past, the widespread and persistent occurrence of PCBa underscores the failure of existing practices to adequately control the escape of PCBs Into the environment. Today I announced a broad EPA program of regulatory and nonregulatory actions to reduce the levels of PCBs In the environment. statement, (which Is enclosed), stated that we must, as a society, accept and work toward a goal of totally eliminating the production. Importation and use of PCBs as rapidly as possible. Toward that end, I am asking you, ss a principal member of the PCB Industry, to cosmic yourself to a deliberate and expeditious search for alternatives. Z recognize that this effort will not be easy. A careful examination and balancing of the environmental consequences of alternatives will be required. The economic Impacts of the changeover will have to be considered. Z also recognize that such an effort cannot be accomplished overnight, nevertheless, I have confidence that American Industry has the capacity to solve this difficult task, and I solicit your full cooperation and participation in this effort. So that we can begin this undertaking as soon as possible, I am inviting you and other representatives of your industry to meet with me on January 14 to discuss this undertaking and work towards the development of plans on how it csn be accomplished as smoothly and rapidly as possible. X have also asked representatives of the principal users of PCBeontainlng transformers and large capacitors to meet with me on January 22 to discuss ways that we can effectively manage and control PCBa In current use. 1 am enclosing a copy of my letter to these . companies. At our meeting on January 14 we might also discuss some of the matters outlined in that letter. ACfc G 0 6 4 6 7 I am looking forward to M a t i n g with you at EPA la Room 11121 1921 Jefferson Davis Highway, Crystal Mall #2, Arlington, Virginia, on January 14, from two to flva In the afternoon. If you have any questions concerning the arrangements for this meeting, you may contact Dr~. X. E. Wallen, (202) 755-6956. Sincerely yours Mr. John W. Hanley President Monsanto Chemical Company 800 H. Lindbergh Boulevard St. Louis, Missouri 63166 Enclosures Administrator \ AOh CC4tt Monsanto MONSANTO INDUSTRIAL CHEMICALS CO . 8 0 0 N. lind ii0r-n Boulevard S t. Louji.'M issouri 63168 Phon: (3UJ 63-1-to0 0 .S P c o A L rr r.^s : Dear Customer: Monsanto will stop accepting polychlorinated biphenyl ("PCB'M waste returns effective August 31s 1977- We new have substantial volumes of PC3 waste on hand which must be incinerated along with our production unit's residue before the incinerator is decontaminated and shutdown later this year. For this reason we are not able to accept any significant volumes between r.ow and August 31st. The American National Standards Institute's publication AMSI C1G7. -1974 lists the following firms as having incineration facilities capable of handling PCB's. Chem-Trol Pollution Services, Inc. P.0. Box 200 1550 Salr.er Read Model City, Mew 'fork 14107 Phone: 716-754-5231 Rollins Environmental Services, Inc. P.0. Box 2349 Wilmington, Delaware 19899 Phone: 302-658-3541 'It is, of course, your responsibility to determine the adequacy of. the incineration facility for the disposal of PCS waste and its compliance with applicable laws and regulations. Sincerely, PLAINTIFF'S' "janies A. Alley Industry Specialist Dielectrics tme i unit of M ontarne Company AC* CCc^tSC M ONSANTO IN D U STR IA L C H EM IC A LS 8 0 0 N. L io a C ';* ' a s - ' t v f C St. low s. M.ssown 53*55 Phon#; 3 U , 6 94 -iO C O CO. June 30, 1977 Mr. George F. Wirth, Chief Special Chemicals Branch Office of Toxic Substances (WH-557) U.S. Environmental Protection Agency Washington, D.C. 20460 Dear Mr. Wirth: ,* Enclosed is a letter we are sending to our customers notifying them that we will stop accepting polychlorinated biphenyl ("PCB") waste returns effective August 31, 1977. Our incinerator was built to dispose of our PCB production unit's residue and to assist our customers to dispose of their waste. This incinerator was designed to handle only "straight" PCBs or those contaminated with an organic solvent such as trichlorobenzene. As stated in our letter, we now have sub stantial volume of FC3 waste on hand which must be disposed of, along with our production unit's residue and whatever residue remains from the shut down and decontamination of our PCB production facility. This action is in line with Monsanto's announcement to shut down our PCB manufacturing unit and exit the business by October 31, 1977. If you have any questions, please let me know. Sincerely JCW:rd Enclosure cc: J. A. Alley J. Coleman Weber Manager Product Acceptability Specialty Chemicals Division PLAINTIFF'S EXHIBIT # u n it of Momi "io C o m o i 'i AC* CCo `fcS Monsanto August 4, 1978 Office of Toxic Substances (TF-794) Environmental Protection Agency 401 "M" Street, S.W. Washington, D. C. 20460 ATTENTION: Ms. Joni T. Repasch Dear Ms. Repasch: On June 7, 1978, the EPA published a proposed rule in 43 Federal Register beginning on page 24,802 covering Manufactur ing, Processing, Distribution in Commerce, and Use Bans of Polychlorinated Biphenyls (PCB's). This letter sets forth Monsanto Company's comments on the proposed rule. Please include these comments, with the attachment, in the" official record of this proceeding. Our comments are divided into two parts. The first part provides general comments which focus on the EPA's proposed maximum concentration of 50 ppm and the justifiable need for a heat transfer system use authorization. The second part sets forth specific language and recommendations related to these matters, and to certain other matters in specific sections of the proposed rule. GENERAL COMMENTS The proposed immediate, mandatory removal from service of heat transfer systems containing fluids whose PCB concentra tion equals 50 ppm or greater is not justified in view of the lack of any significant risk to health or the environment. To require industry immediately to remove from service and -drain all these systems or to replace the systems entirely to meet a 50 ppm concentration level is an arbitrary mandate and will generate a substantial, immediate risk of injury `H. .and environmental harm from spillage and result in an abrupt, /^massive buildup of PCB contaminated fluid without available *'"disposal facilities or time to develop orderly plans to -9- carry out the disposal". Furthermore, the EPA cites r.o evidence in the preamble of. any environmental or health hazard to support the proposed reduction m the regulated concentration of PC3's m a PCS mixture under Section 761.2(vi from 500 ppm to 50 ppm, r.or "does the May, 1978 EPA Support Document/Draft Voluntary Environmental Impact Statement upon which the proposed rules are based support this reduction. Therefore, Monsanto strongly recommends the EPA withdraw this proposed reduction. Yet even with the withdrawal of the 50 ppm concentration, it is imperative that an authorization be granted for continued use of heat transfer systems containing residual PCB's. The evidence conclusively establishes that the heat transfer system use activity meets the requirements for granting an authorization to allow continued operation, and Monsanto requests the EPA to grant such an authorization. On page 42 of the EPA Support Document, the EPA states that before it grants an authorization for continuation of a non-totally enclosed use activity the following two requirements must be met: (1) the activity authorized must not pose an unreason able risk of injury to health or the environment, and (2) the ban would'cause major and extensive economic disruptions. Regarding the first requirement, an analysis of potential PCB exposure from various sized PCB contaminated heat transfer systems is illustrative of the insignificant risk of injury to health or the environment posed by authorizing continued use of these systems. The highest risk area for leakage in these systems is in the pump seals, but inspection maintenance for pump seals is also emphasized because of user awareness of this risk. A pump seal failure would be rapidly detected by inspectors or automatic monitoring devices and loss of fluid would be unlikely to exceed from 1% to 5% of the total fluid volume in the system. For example, a 500 gallon system containing 1% volume of PCBs has a total PCB content equal to 60 pounds, and a leak of 5% of the total fluid (which is unlikely) would release only 3 pounds of PCB's. Similarly, a 10,000 gallon system containing a PCB contamination of 500 ppm has a total PCB content of 60 pounds, and a leak of 5% of the total fluid (again unlikely) would release only 3 pounds of PCB's. A leakage of 3 pounds of PCB's is below the EPA's designated /harmful quantity for PCB spills into navigable waters (10. pounds) under 40 CFR 118. issued pursuant to the Federal A i ^ CCfc*iS> c -3- Water Pollution Control Act and, in Monsanto's opinion, dees not create a significant risk to health or the environment. Furthermore, the above examples are worst case situations because they assume no spill prevention program (for example, diking) or effluent control program m effect. If such programs"were in effect it would be highly unlikely for any amount of PCB's approaching 3 pounds to escape into the environment. The proposed immediate, mandatory removal of these systems from service also meets the standards for the second require ment for an authorization by resulting in substantial, widespread economic disruption. Since the EPA states in the preamble and the Support Document that it lacks information necessary to evaluate the impact of the ban on the heat transfer system use activity, the following information is presented which is generally consistent with the specific information requests in 43 FR 24810 relative to hydraulic die-casting systems, and which clearly and conclusively demonstrates this economic impact: 1. Number of systems contaminated. It is estimated that when the sale of FCB heat transfer fluids was discontinued in 1972 there were 450 heat transfer systems using this fluid. This estimate is based on (a) the current level of new systems introduction, (b) the ten year period (1962-72) during which PCB heat transfer systems were specified, and. (c) the total volume of 20 million pounds of PCBs sold for heat transfer systems from 1962 through 1971 as reported by the 1972 Interdepartmental Task Force on PCBs COM-72-10419 pages 6 and 7. 2. The average liquid volume of the systems. It is estimated that about one-half of all PCB contaminated heat transfer systems have less than a 500 gallon liquid volume. Systems greater than 500 gallon volumes are estimated to average about 2,000 gallons. These estimates are based upon a review of systems filled in 1977 and 1978 and upon the average system size of Monsanto units converted in the 1970-72 period - a typical cross section of the industry pattern. 3. The range of liquid volume. Heat transfer systems of greater than lo/ooo gallons are unusual. Small experi mental pilot plant units may contain 50 gallons. 4. Fluid top-off. Heat transfer fluids are designed to be thermally stable within the recommended temperature range of operation and to operate at low pressure. Accordingly, routine top-off is minimal. Average top- fiCM CCe** - o f f may be h i g h e r b e c a u s e o f f l u i d r e m o v a l from c a r t s o f th e sy stem when m a in te n a n c e i s p e rfo rm ed on o t h e r components of the system such as pumps, valv es and heaters or in the event a system malfunction causes a th e rm a l d e g r a d a t i o n o f `t h e f l u i d . M o n san to 's e x p e r ie n c e i n d i c a t e s t h a t some s y s t e m s may r e q u i r e no t o p - o f f f o r t h r e e t o f i v e y e a r s ; o t h e r s y s t e m s may r e q u i r e a t o p o f f o f 50% m one y e a r i f m a j o r m a i n t e n a n c e p r o g r a m s r e q u ir in g drainage are undertaken. However, systems are not routinely drained. 5. E f f o r t s t o r e d u c e PC3 c o n t a m i n a t i o n . We e s t i m a t e t h a t g r e a t e r t h a n 90% o f t h e p l a n t s o p e r a t i n g w i t h PCS h e a t t r a n s f e r system s c o n v e r te d to a l t e r n a t i v e non-PCB f l u i d s m 1 9 7 0 - 7 2 . Some o f t h o s e c o n v e r t i n g d r a i n e d and r e f i l l e d t h e s y s t e m w i t h non-PCB f l u i d ; some d r a i n e d , f l u s h e d and r e f i l l e d ; some d r a i n e d , c h e m i c a l l y c l e a n e d and r e f i l l e d ; and o th e r dism antled and d isco n tin u ed use o f l i q u i d s y s t e m s . We e s t i m a t e t h a t t h e s e e f f o r t s have g e n e r a l l y r e s u l t e d i n c u r r e n t s y s t e m PCB l e v e l s m t h e i n d u s t r y r a n g i n g from a r o u n d 100 ppm t o s e v e r a l t h o u s a n d ppm. R e c o v e r y from h e a t t r a n s f e r s y s t e m s wo uld g e n e r a l l y be e x p e c t e d t o be g r e a t e r t h a n 90%. Note t h a t f o r t h e purposes of th e se comments the term "recovery" is defined as the percentage of t o t a l system f lu i d volume e x tra c te d from the system by d ra in in g . 6. S e r v i c i n g f l u i d s t o remove PCB1s from s y s t e m s . F o l l o w i n g is a review ana a n aly sis of the p o te n tia l success of t h r e e m e th o d s t h e EPA h a s c o n s i d e r e d f o r PCB r e m o v a l : (a) Drain and r e f i l l . I f small systems (500 g allons or le s s ) are d ra in e d and r e f i l l e d w ith a non-PC3 f l u i d , a 98-99% re c o v ery o f d is s o lv e d PC3' s can be expected. This serv ice method w ill not, however, remove PCB a b s o r b e d by t h e h e a t t r a n s f e r s y s t e m w a l l s . In t h e c a s e o f sy stem s l a r g e r t h a n 500 g a llo n s a 90-99% re c o v ery can be expected depending upon the com plexity of the pip in g la y o u t and heat exchanger design. (b) D is tilla tio n is of questionable e ffic ie n cy in s e p a r a t i n g and removing PCB' s from h e a t t r a n s f e r f lu i d due to the s i m i l a r i t y in b o ilin g ranges of PCB's and non PCB h e a t t r a n s f e r f l u i d s . The b o ilin g ranges for these re sp e c tiv e flu id s are as fo llo w s: PCB T h e r m i n o l FR 1 617F - 690F T h e r m m o l FR 2 644F - 707F AC* CCfcHSs Ncr.-?C3 Therm m o l 5 5 T h erm m cl 66 635 3F 643 3F 34 0F 663 3F (c) Carbon tre a tm e n t experim ents to remove ?C3's are cu rren tly being conducted to in v estig ate the c a rb o n t r e a t m e n t removal method, b u t no d a t a are presently available. 7 . P r e s e n t PC3 l e v e l s m s y s t e m s - We e s t i m a t e t h e r a n g e o f PC5 c o n c e n t r a t i o n l e v e l s i n h e a t t r a n s f e r s y s t e m s to be from a b o u t 130 ppm t o a b o u t 3 0 , 0 0 0 ppm. a. C o st o f new sy ste m s - The c o s t o f a new system c o u ld r e s u lt m a large d o lla r expenditure compared to t.te much low er c o s t o f r e p l a c i n g f l u i d . R e p l a c i n g a n e a t t r a n s f e r s y s t e m may w e l l mean t o t a l l y r e h a b i l i t a t i n g a plant because these systems are in te g ra l p arts of the u s e rs' p la n ts , and replacem ent of heat t r a n s f e r systems is impossible without replacement of reactors, coolers and h eat exchangers. Such replacem ent would u su ally require a m ultim illion dollar expenditure. Further, the c o sts of l o s t p ro d u c tio n would be e x tr a o r d in a r ily high in terms of l o s t wages, l o s t s a le s and product shortages which could impact a wide area of other manufacturing industries using finished products during the length of time required to replace each system. Thus the immediate, mandatory removal o f th ese systems from s e rv ic e would r e s u l t m s u b s t a n t i a l , widespread economic d is ru p tio n m in d u s trie s using heat tra n s fe r systems and the products produced from p ro cesses using such systems. 9. C o s t o f d r a i n i n g and r e f i l l i n g s y s t e m s - We e s t i m a t e the c o s t of d r a in in g and r e f i l l i n g systems would be from S10 m i l l i o n t o $30 m i l l i o n , d e p e n d i n g upon a u t h o r i z e d PCS l e v e l , and b a s e d upon t h e model o f an a v e r a g e 2 , 0 0 0 g a llo n system and th e e s ti m a te d 450 systems in the i n d u s t r y . Fo r e x a m p l e , PCB r e d u c t i o n t o a c o n c e n t r a t i o n l e v e l o f 500 ppm i n a 2 , 0 0 0 g a l l o n s y s t e m , w i t h a c u r r e n t 1% PCB volume ( 1 0 , 0 0 0 ppm) and a s s u m i n g a 90% flu id recovery, would r e s u lt in a marginal removal cost o f a p p r o x im a te ly S220 p e r pound. I t i s e s t i m a t e d t h a t t o r e d u c e from a 500 ppm l e v e l t o a 50 ppm l e v e l would req u ire a marginal removal c o st of about S30,000 per p o u n d . S t a r t i n g w i t h a PCB c o n t e n t m t h e 2 , 0 0 0 g a l l o n sy ste m o f 240 pounds, d r a i n i n g t h e system (90% r e c o v e r y ) would leave 200 g a llo n s o f f l u i d c o n ta i n in g 2 g a llo n s o f PCB's o r 24 pounds. T h is q u a n t i t y s h o u ld , when th e system is r e f ille d with 2,000 gallons of flushing f l u i d , r e s u l t i n a PCB c o n c e n t r a t i o n o f 1500 ppm. A A L * CCe<S5 ' 'S' 1 r second draining and refilling wi ushi ,g r . u A should initially drop ?C3's*to 2 und s or -o out :r.:s amount w o u i g shortly increase as ?C3's are released from the surface.walls into the fluid, proper concentration has be erf obtained to assur concentration remains below' BCG ppm, the flushi would be drained and the system refilled with n heat transfer fluid. a The direct cost estimates of flushing and draining, excluding costs associated with lest production and resulting disruption of supply to consumers, are as follows : (a) Flushing fluid - 2x2,000 gallons at 52.00 per gallon (b ) New fill of non-PCB heat transfer fluid - 2,000 gallons at S8.00 per gallon S15.0C- (c) Disposal of initial drained fluid and 2 flushing fluid changes - 6,000 gallons at S2.26 per gallon 515,000 (d) Labor and overhead - 4 men 10 to 15 days 512,000 Total 552,000 This estimate assumes that only two flushings are required and that draining and flushing proportionately reduces PCB concentration level without a major breakdown of the system involving gasket replacement or rewelding which could easily double or triple costs. Based upon this estimate, the cost of removal of 235 pounds would give a marginal removal cost of S220 per pound to achieve a residual level below 500 ppm. To achieve a level of 50 ppm it is likely that at least two further drain and flush procedures and chemical cleaning would be required. At S50,000 per procedure, the final 1.6 pound removal would carry a marginal removal cost of at least S30,000 per pound. The above information relative to environmental and health injury and economic impact is clear evidence which meets the two requirements necessary to support the grant of a use authorization for heat transfer systems containing residual PCB's. Accordingly, Monsanto recommends the proposed rules be revised to authorize for a period of 5 years the continued use and servicing of all 500 gallon capacity or less heat transfer systems used and serviced in a manner other than a totally enclosed manner to the extent these systems contain 1% or less by volume of PCB's, provided users attain this i% A C * CCc<t5fc ?C3 volume w i t h i n cr.e y e a r a f t e r t h e e f f e c t i v e d a t e o f tr.e r u l e . For s y s t e m s w i t r . g r e a t e r t h a n SCO g a l l o n c a p a c i t y . Monsanto recommends t h e EPA a u t h o r i z e f o r 5 y e a r s c o n t i n u e d use ana s e rv ic in g of these systems in a manner o th er than a t o t a l l y e n c l o s e d manner t o t h e ex t e n t t h e y c o n t a i n a ?C3 c o n c e n t r a t i o n o f 500 ppm o r l e s s , p r o v i d e d u s e r s a t t a i n t h i s 5C0 ppm PCS c o n c e n t r a t i o n w i t h i n 2 y e a r s a f t e r t h e e f f e c t i v e date of tr.e ru le , These t r a n s i t i o n time p erio d s would p e r m i t an o r d e r l y r e d u c t i o n o f PC3 l e v e l s d u r i n g whic h d is p o sa l of flu id s by in c in e r a t io n could be planned, m e r e r y a v o i d i n g s p i l l a g e and b u i l d u p o f PCS c o n t a m i n a t e d f l u i d i n v e n t o r i e s a w a i t i r.g d i s p o s a l . P l e a s e n o te t h a t p ro p o se d l a n g u a g e f o r t h i s use a u t h o r i z a t i o n i s s e t f o r t h as new S e c tio n 7 6 1 .3 1 (h ) on pages 10-11 o f th e s e comments. S pecific Suggested Changes m the Proposed Rules S e c . 7 6 1 . 2 (w) "?C3 m i x t u r e . " In a d d i t i o n t o t h e p r o p o s e d a r b i t r a r y 50 ppm c o n c e n t r a t i o n , t h e d e f i n i t i o n i n c l u d e s any m a t e r i a l , no m a t t e r how d i l u t e i n PC3 c o n c e n t r a t i o n , i f t h e m a t e r i a l i s c o n t a m i n a t e d by a s u b s t a n c e c o n t a i n i n g a PC3 c o n c e n t r a t i o n o f 50 ppm o r g r e a t e r . Fo r e x a m p l e , i f a g a l l o n o f h e a t e x c h a n g e f l u i d c o n t a i n i n g 500 ppm PC3 were d i l u t e d to 1,000 g a llo n s w ith a non-PCB f l u s h in g f l u i d ( f i n a l c o n c e n t r a t i o n 0 . 5 ppm PCB), t h e r e s u l t i n g f l u s h i n g f l u i d when d r a i n e d would s t i l l be a r b i t r a r i l y c o n s i d e r e d a PCB m i x t u r e . We recommend t h e d e f i n i t i o n be c h a n g e d t o read: (w) "FC3 M i x t u r e " means any c o m b i n a t i o n o f c h e m i c a l s u b s t a n c e s w h i c h c o n t a i n s 500 ppm (0 .C 5 0 p e r c e n t on a d r y w e i g h t b a s i s ) o r g r e a t e r o f a PC3 c h e m i c a l s u b s t a n c e . This d e fin itio n includes, but is not lim ited to, d ie le c tric f lu id and contam inated s o lv e n ts, o i l s , waste o i l s , h eat tran sfer fluids, other chemicals, rags, so il, paints, d e b ris , sludge, s l u r r i e s , dredge s p o ils , and m a te ria ls contaminated as a r e s u l t of s p i l l s . S e c . 751. 2( b b ) " M a n u f a c t u r e ' f o r Commercial P u r p o s e s . ' " We recommend fo r c l a r i f i c a t i o n th e a d d i t i o n o f the follow ing subsection to the d e fin itio n , which is c o n sis te n t with the d i s c u s s i o n by t h e EPA s e t f o r t h m column 1. 43 FR 24805: ( 3 ) As t h e d e s i r e d p r o d u c t . I n a d v e r t e n t m a n u f a c t u r e as a b y - p r o d u c t m t h e m a n u f a c t u r e o f a n o t h e r c h e m i c a l is not "manufactured for commercial purposes." Sec. 7 6 1 . 2 ( cc ) "PC3 S e a l a n t , C o a t i n g , o r D ust C o n t r o l Agent." This term is d efin ed based on the s t a t e of the A O C C c *i^ 7 -3- analytical art. In complex mixtures such as waste oils, analysis for PCB is not sensitive because cf background interference*. We recommend that a concentration level be set related to the toxicological properties of PCB1s and the demonstrated injury risk associated with human and environ mental exposure. Such a concentration would, among other things, avoid a moving target definition depending upon the sophistication of the analytical equipment used. Sec. 761.2(dd) "process 'for Commercial Purposes.1" We recommend the addition of the following subsection to the definition. This is consistent with the discussion m column 1, 43 FR 24805 and our proposed modification of section 761.2(bb). (3) for means other than PCB removal. Processing which removes PCB from the material to be used for commercial purposes does not constitute "processing for commercial purposes." Sec. 761.2(ff) "Significant Exposure." This term is defined based on the state of the analytical art. We believe that significant exposure should be, as discussed above relative to Section 761.2(cc), a specific concentration related to the toxicological properties of PCB's and to injury risk associated with human or environmental exposure. We recommend, consistent with the 1977 American Conference of Governmental and Industrial Hygenists publication of Threshold Limit Values for Chemical Substances in the Workplace, that this section be amended to read as follows: (ff) "Significant Human Exposure" means any exposure of humans to PCB chemical substances or PCB mixtures m excess of 0.5 mg/cu. meter TWA for a 40 hour week by an applicable analytical method. Sec. 761.1(gg) "Small Quantities for Research and Development." The definition of this term is unjustified and arbitrary. No data are presented in the preamble by the EPA that use in research has posed any measurable risk of injury to health or harm to the environment. Monsanto has no evidence of any adverse effect of PCB's in laboratory use. We suggest as an alternative replacing section 761.2(ff) with the following * wording taken in part from the TOSCA Inventory Reporting 0 Regulations, 42 FR 64576, section 710.2(y): * ' (gg) "Small Quantities for Research and Development" means any quantity of PCB chemical substance or PCB mixture manufactured or processed only for purposes of scientific experimentation or analysis of chemical ACM CCe^Sb -9- research en, or analysis cf, PC3's, including research or analysis for the development of a product that tl> are no greater than reasonably necessary for such purposes, and (2) are used by, or directly under the supervision of, a technical!'/ qualified individual ts i. It would also seem appropriate to include as r.ev section 751.2(kk) the definition of 'Technically *ua ified Indir based upon section 710.2(aa) of the TCSCA In entory Repo Regulations. v al" --a Sec. 761.2(hh) "Totally Enclosed Manner." This term is defined with reference to section 761.2(ff) and, consistent with our recommended changes above to that section, we suggest section 761.2(hh) be changed to read: (hh) "Totally Enclosed Manner" means any manner that will ensure any exposure of human beings or the environ ment to PC3 chemical substances will be insignificant. Sec. 761.2(jj) "Applicable Analytical Method." We propose the following new definition of this term: (]j) "Applicable Analytical Method" means a.method (other than the perchlorination procedure) which distin guishes PCB Chemical Substance as defined in section 761.2(t) from all other materials. The basis for exclusion of the perchlorination procedure is y discussed more fully in the attached letter dated March 15, 1976 from J. Coleman Weber of Monsanto to Dr. I. E. Wallen of the EPA. Section 761(kk) "Technically Qualified Individual." We proposed a new definition for this term with the following wording taken from Section 710.2(aa) of the TCSCA Inventory Reporting Regulations, 42 FR 64576: (aa) "Technically Qualified Individual" means a person (1) who because of his education, training, or experience, or a combination of these factors, is capable of appreciating the health and environmental risks associated with the chemical substance which is used under his supervision, (2) who is responsible for enforcing appropriated methods of conducting scientific - experimentation, analysis, or chemical research in order to minimize such risks, and (3) who is responsible for the safety assessments and clearances related to the procurement, storage, use, and disposal of the chemical substance as may be appropriate or*required ACM CC84S9 'H'- within the scope of conducting the research and develop J ment 'activity. The responsibilities in clause (3) of this paragraph may be delegated to another mdiviudal, or other individuals, as long .as each meets the criteria m clause (I) cf this paragraph. Section "61.31(f) Hydraulic die casting systems-use. The same considerations that support a use authorization for hydraulic die casting systems apply to other industrial hydraulic systems,,-as well. Therefore, we recommend that all references in this section to the words "hydraulic diecasting system" be changed to "mdustri al hydraulic system." Section 761.31(h) Heat transfer systems - use. We recommend this new use authorization as follows: (h) Heat Transfer Sytems - Use. Heat transfer systems containing PCB mixtures may be used and serviced in a manner other than a totally enclosed manner until five years after the effective date of this rule subject to the following conditions: (1) One year after the effective date of this rule no heat transfer systems of 500 gallon capacity or less may contain fluid whose PCB mixture volume is greater than 1%. (2) Two years after the effective date of this rule no heat transfer systems of greater than 500 gallon capacity may contain fluid whose concentration of PCB mixture exceeds 500 ppm. (3) Ninety days after the effective date of this rule each person who owns a heat transfer system containing residual PCB's above the authorized levels shall report to EPA, and retain records of, the number of such systems he owns, the volume capacity of each such system, and the PCB volume or concentration, as appropriate, of the fluid contained in such systems. This report shall be sent to the Pesticides and Toxic Substances Enforcement Division (EN-342), Environmental Protection AGency, 401 M Street S.W., Washington, D. C. 20460. Each person who owns such a system shall also keep a current record of the dates of each draining or refilling and the measured PCB concentration or volume, as appropriate, of the fluid in the refilled systems on those dates for each system. If any such system is ADM CCtbCC j ib aj M *1 sold, the tr a n s a c tio n and the p a r tie s th e re to s h a l l he r e p o r t e d t o EPA by t h e S e l l e r . At i t s d i s c r e t i o n , EPA n a y r e q u i r e m e s u m i s s i o n of a copy of a person's cu rren t record. - (4) Each r e p o r t s u b m i t t e d t o EPA u n d e r p a r a (3) of th is sectio n sh all contain the c c a t i o n fo u n d m 7 6 1 . 3 1 ( fc ) (2 ). ( 5 ) Each p e r s o n who owns a h e a t t r a n s f e r s y s t e m c o n ta in in g r e s id u a l PCB's above the authorized le v e ls s h a ll develop and implement a plan for t h e c o n t r o l o f PCB e x p o s u r e s and c o n t a m i n a t i o n m acco rd an ce w ith Annex VI I . Your fav o ra b le c o n s id e r a tio n of the above comments is respectfully requested. Sincerely, W. R. C ore y D i r e c t o r , TCSCA A d m i n i s t r a t i o n ACP CCcoCI D r. I . Z. W allen Zr,v i r c r.me r. t a 1 ? r o - a c " i. c r O f f i c e c f T o x i c OUmS wiT. 401 11 S t r e e t S. M, Washi n c t e n , DC 2 0 4 60 Ac a r. c y Dear Dr. Waller.: .`i o r . s a n t o h a s a c o n c e r n e v e r t h e v a l i d i t y o f t h e p e r c h 1 c r n a t i o n t e c h n i q u e u s e d by EPA a n d o t t e r s t o m e a s u r e a r . c / o confirm. p o l y c h l o r i n a t e d b ip h e n y ls ir. e n v iro n m e n ta l r a t e r In our in v e s tig a tio n of the' p e rc h lc rir.a tio n method, we have fcunc th a t such chem icals as biphenyl, a ik y la ted biphenyl, and many e t h e r s u b s t i t u t e d b i p h e n y l s , i n t e r f e r e w i t h t h e perchlorination teennique. I t also appears th a t various p e t r o l e u m c o m p o n e n t s may i n t e r f e r e . I f t h e s e c h e m i c a l s w ere p re se n t ir. en v iro n m en tal m a te r ia ls t h a t v/ere being te s t e d for PC3's using the p e rc h lo rin a tio n , method, e rro n eo u sly high PCB c o n c e n t r a t i o n s w o u l d b e r e p o r t e d . A r e c e n t a r t i c l e i n t h e J o u r n a l o f t h e ACAC ( V c l . 5 3 , No . 3 , 1975) p o in ts out two o th e r li m i t a t i o n s of the p e rc n lo rir.a tic r. procedure: 1. High and v a r ia b le reag en t b lan k s, which cause erroneously high findings. 2. Form.atioa of bror.or.cnachlorobinheny i , which cau ses lew re c o v e rie s . A copy of the a r t i c l e is a tta c h e d . PLAINTIFF'S EXHIBIT ACP CCfcbC 'h j - l T *^ Dr. X . I . Wallen 2- - March 13, 1375 it M Since these lir.'.irations can-lead to significant er in caosrnininc trare levels of ?C3 's in er.vircr_ren sa-ples, *.ce susresz than I?A* carefully rsviev; the val; of the oerchlorinaticn technique. Pesults that have 3 cbrained using this technique .ray r.ct be valid. If any other ihforration is needed:, please let us hr.c* S inter e l*r, J. Colersn Weber .'Ir.nacer, Product A r.an cc: Dr. A. C. TraJcowsJci Environmental Protection Agency \ ACM CQ8dC3 ItGTlCS: Tills material Z2.7 be protected by copyright 466 joltin'u. or the ao\ c fV:.. 55. v 3 2 Limitation on the L>e of Ant i mony Pent a chl or i de for Perchlorination of Polychlorinated Biphenyls WILLIAM J. TROTTER and SUSAN J. V. YOUNG nision of Chemistry a*i Phyr.cs, Food and Dr.g Administration, 7,'jjhtngion, DC Two contaminants re present in comm er cially available antimony pentachloride (ShCH) used 10 p e r c h lo r in a ie p o l ch lo rtn aied biphenyls ( P C B s ) to d e c a c h lo r o b ip h e n y l ( D C B ) . DCB ia found in the SbClj perchlorination reaction blank in which no PCBs were added. Bromon onachlorobipheny I (BN CB) is found after use of SbClj to perchlorinate PCBs. Levels of DCB found in the SbClj reaction blanks from vari ous distributors ranged from 8 to 972 n j DCB/ ml SbClj. The relationship of the formation of BNCB to am ounts of various PCB Aroclors perch lo rin ated is exam ined. Polychlorinated bipken>l (PCBs) residues are extracted, cleaned up, and detected by methods similar to those used for org.inochlonne pesti cides. PCB residues are quantitatively deter mined by comparing the cas-liquid chromato graphic (GLC) response of the multicomponent residue and commercial PCBs (Arcelor?) or a mixture of Aroclors producing a GLC response pattern similar to that of the residue (1). This approach is limited because the mui:'.component PCB residue may not have the same propor tional composition as the Aroclor or Aroclors used as the quantitation reference Residues car. be composed of mixtures of ch!erob:pheny| com ponents from more than 1 Arcelor Metabolic and other environmental factors complicate the description of the PCB residue composition. There has been considerable work to de\e!op methods to convert the multicomponent PCBs to a single derivative on which to base the resi due determination. Procedures have been re ported to catalyttraHy rtechlorinate PCBs with hydrogen over palladium or platinum to bi phenyl, cyclohevylbenier.e, and bieyclohexyl (2, 3). A principal disadvantage with that pro cedure is that the hydrocarbon product is determined with a GLC flame ionization detec tor, re-ulting in low sensitivity. Attempt? have been made tq;convrn PCBs to the fully chlor inated decachlorobiphenyl (DCB) (3-6). Ar mour (0) reported optimum conditions for perchlormatir.g PCBs with antimony pentachloride iibC lj). The method provides a cuai.tat confirmatory procedure for PCS de'erm.r v... The GLC electron capture de'ec'or reir.or.-e enhanced becir.;e total PCBs are r_a::.:es*ed _s single peak for DC3. In measuring the s.r.o peak for DCB the analyst is cot faced analytical judgments sura as baseline co:*-r method of integration, or discrimination tefve PCBs ar.d non-PCB components. However, t necessary to be aware that the various A.*::., give rise to different equivalents of DCB and that the nonchlonr.aied bip'cenvl false - as a fungicide) is perchlorinated by SbClj DCB. Nonetheless, using the perchlonra: der.va'ization can reinforce the residue vv determined by measuring a raulticomporPCB residue. During attempts to apply the perchlor nat. derivatization in determine? low residue !*v of PCB and make use of the increa.-ea c:ectr capture response to DCB, 2 contamipacvs v: indicated which led to erratic recoveries of DC Experim ental Rta/inti and Apparatus fa) Antimony pewarhfonde.--Hocker Chem-' Niagara Fails. NY 14302 (recer. ed 1a flass hrwith Uad-lincd cap); Mnti.esoa Coleman 1 3 (MCB). Not wood. OH 43212 freazent 'Bd:A (.Allied Chemical), Morr.itowa. NJ C7` (reaceat grade. 99"c): Research O'giaic-lr.prz; Chemical (ROC-RIC). Bellevi'Je. NJ C- ' (9999<"r); and J. T Baker Chemical. Phillipst'NJ GSS65 (Baker Analysed Reagent). tb) G n ch'pmalog-'i? \ --Searle-Aaal'tie (I Plaines. IL 6G06S) Model 5360 ntn 5' x 4 r.n elas? column contaiaioz 1G: OV-lCl cn S3mesb Chromosorb JV (HP). Operatic? coni:*. * column flow, 60 ml oitrczea. r.in ; coiumn. 2C2* dc'ector. 202*C; injector, 22j*C. pm-cup de? electron capture detector with iitaamm *!! I defector voltace (constant dc) adjusted ro c*. one-half full scale recorder deflection nr 0 7 DCB w|i*n full scale 'flection ij 1 X '.0 * 'w (c> }fn>* sprrtrnmett*--Van.m MAT .`5 ?. 72. Spnczfleld. NJ 070iD CH5-DF mu# etcr (MS) rouplcd to Vartan Aerocruph 27iD Mi 7R0TTLR * YOt'NC PUtCHLORIVATTON* CF ?C3 467 ehroxsiocnph via i j- g liii 'f e n usinc 'AVsca- 3:f~ U 3 epintor. CLC opera: *c 1:z z j : i' x t c_n id tikis column * c a ! i . i . i r 3"e OV-1 c : $O-!j0 mesh C ' - . - 'iOiOfh TV ' H ? > ; column dew. 60 x l hi: .m r . . a : 2;3'C MS open.a? 'fe d ,:.e ta t.T \i:*n or*v>?D m-; en_ss.cn current. 300 ,,a; n u l t . r L e r - : 2c*. 3 2 kv. 1 ReeuJt nd Dicuion A peak identical :o rhat of DC3 was found in the reaction biar.k for :hc \ m r : r p*-ch!ornation procedure (0) tr*h the described GLC cperatmr conditions The .iej:ifica\on of DC3 was con fixed by GLC-M5 of a hexane v'rict of a hydrolyzed sample cf S'.G , which had not been subjected to the perohJcnraticn procedure. Various quantities <02-2 0 mi) of ;bC!i ;Vni the 5 commercial sources w*r? examined to de termine the presence of DC3 5LCL alone was carried through the percLlonnation reaction (6) except that no CHClj wj;- present with 5hG , in the reaction vessel. DC3 was determined by GLC. Table 1 lists the amounts of DC3 found. After perchlcnnatmg ?CBs with 5bCl;, a secondary peak -.nth a CLC retention time rela tive to DCB of 1.31 was observed similar to that reported by Huckins tt al. ID . This later eluting peak is seen in Fig. 1. the chromatogram from he 02 ml 5bCI, (Hooker Chemical) perchlorination of 0.50 ui Aroclor 12-1. This peak was found when SbClj from each supplier wa* used. The peak was determined by GLC-M5 to be due to bromononachlorobiphenyl (3N C B ). BN'CB was assumed to be a competing product w-.rh DCB arising from a small amount of 5bCl.Br in SbCI,. so parameters relating to possible limitation.' of the perchlormaiioa procedure were studied. Vanous quantities i0 5--10 ue) of Aroclcrs 1221. 1242, 1234, and 12-50 in CHCI, were perchlorinated. Recoveries of ECB and estimates of the relative amounts of BN'CB formed are given in Table 2. Calculation of the relative Tbl 1. O C I (m / m i) 0 r m t 4 f r t m v a rla u ! SbCI, SbCI,. ml Supplier a 2 1.0 2 0 Ay. Sootier C h e m ic a l <*C8 A oc m e - T. 8 > tr ir u 47 4] is 42 M U MO 1 W 911 972 12 1J 12 12 9 7 71 FIG. 1-- electron cotur# GLC eurva from rh* 0.2 m l SbCI, iH oo m r Cbam.cai) p trcM on nal'O n of 0.50 _g Aroclor 1221; 0 31 ng #og.aint Aroclor 1221 injected. Reek 1 represent* 0 > 1 ng DCS. Paak 2 represent 0.2 ng SN C S. amounts of BNC3 product formed was based on comparison of the electron capture GLC peak height of BXCB with that of a DCB reference. The amount of DCB determined in the reac tion blank was directly proportional to the 'amount of SbCI, used (Table 1). This indicates SbClj was the source of the DCB and that contamination from other possible sources dur ing the pcrchJormaiion was negligible. The pro cedure for perchlorinaring PCBj specifies the use of 0 2 ml SbClj. SbCI, producing $-972 ng SbClj/'ml m the reaction blank would add OS65 ppb. based on a 3 g sample. DCB produced in the reaction blank was as sumed to come from PCB contamination of SbClj. In an effort to locate the ortgm of this contamination, SbCI, bottle closures ere i n s t i gated. GLC analysis of hexane, in which the plastic caps were soaked for 4 days, did not reveal PCBs. Hooker Chemical, the sole do mestic source of 5bG ,. supplied SbG, in glass bottles with lead-lined caps. This bulk supplier of SbCI, indicated that the production of chlo rine in carbon anode half-ceils with linseed oil or other organic binders forms certain organic compounds: however, the destructive oxidative environment in ihe electrolytic cells would make the production of PCB unlikely .1 a reult of this pathway On^hc other hand, antimony metal is commonly obtained as a metallurgical by-prod uct by carbon redumon of its oxide: therefore, ACH C C 6 5 C 5 468 journal cf the ao\c (Vcl.5?.N?3.*. T b l* 2. D C S i " d B N C H r a m p * r e * io rln a t lo n oI v a rio u s A ro c lo rs >* 0.2 ml S S C I'* A ro d o r Am t, Mi CCB reed, % B.N C3' reed, % C C B - 3*<C3* to n s irti r#C. C 1260 1254 1212 1221 1260 1254 1242 1221 1260 1254 1242 1221 10 13 13 10 4 4 4 4 0.6 0.5 06 0 .5 16 A4 12 87 11 10 71 70 19 ;i 72 60 0 0 4 16 0 0 1 11 2 6 10 11 16 24 92 21 11 20 26 22 91 24 22 79 * Hoohor C hem ical S a C 'j. ` Q u a n tify ca icu lo to d oy e o m o o n so n o * l* c tr o n e a > t u r * G lC ro s o o n s o to B N C B * . ro sp o n so to D C B r * f* r - nc* standard. it li conceivable that PCBs could be associated with the antimony metal employed in the SbCl, process. No heat tranter systems containing PCBs are used in either the chlorine or SbCl; production facilities, and SbCl* does not come into contact with plastics in the manufacturing operation or in shipping container? (Hooker Chemical and Plaitics Ccrp , 1074. private com munication). Two parameter? (various quantities and vari ous Aroclors) were itudied in relationship to the production of BNCB as a competing prod uct of DCB during the pcrchlorin,.'ion of PC3s. BNCB was calculated by comparison of the elec tron capture GLC response to BNCB vs. the response to DCB. Several factors are considered: ( /) In this reaction hromination :* kir.eticaliy favored over chlorination. With perchlormation of lower amounts of PCBs the re'at. ? BNCB to DCB .i grea'er bccau.-e : -s c** ating anent is the limiting quant.ty :* mated SbCl,. (J) Brormnation occurs 'o a !_ degree for a given quantity of 'he *.- c* nated PCBs such as Aroclors l l - l a*d rather than for 1254 and 1260. Ti..i .ikely to a greater number of reactive sites t . i steric hindrance, (i) In the rare? of r 71chlormatcd (0 5-10 >ig) in the above :tu,,y. likely that with lower amounts of PCBs or less chlorinated Aroclors the decrease .n I recovery is principally due to the incre.u BNCB formed. One of the major advantages of perch!:* tion in determining minute quantities cf ; is the inherent increase in elective GLC : tor response. Contaminated SbCl,, as riesc* here, would preclude its use iq many of * cases. Rctxxcncu (1) Offizinl Method* of Anal'jsu (10751 l?*h AOAC, Washington, DC. sees. 29 CCI-I9 (2) Asai. R., Cunther. R.. Westlake. '.V. I Y. (1971) J. Agr. Food Cficn 19 Z'.o-Z'jz (3) Berg. O. W.. Dioeidy, P. L . P.ees. G (1972) Bull. Environ. Conlam. Tor.cz 333-547 (4) Hutzinger. O. W., Safe. S . Ziiko. V. ( Ini J Environ. Anal. Chem 2. 95-IG6 (5) Hutzinger. 0 . W , Jamieson. D . Safe. Zitko. V. (1973) JAOAC 56. 9S2-556 (6) Armour. J. A. (1973) JAOAC 56 757-?"' (7) Huckuu. J. N.. Swanson. J. E.. SvaZm L. (1974) J.40.4C 57. 415-417 R-n*U A u ftul Jt 1.: T h u paper S i 3re**ted 1 lh SS, "i A .---al '! the \O A C . Oet. 11-17. IS*4, at W a j.n .s ro n . DC AC* CCfcSCt J# Monsanto c o a m c * ~c i. a w 2 e * 0 * y jc * ~ July 27, 1973 c Substances (TS-79*0 Protection Agency S. W. C. 2QU60 Attention: Ms. Jonl T. Repasch Dear Ms. Repasch: This notification is sent to you in accordance with instruct ions contained in the June 7, 1978 issue of the Federal Register in connection with proposed rules for the Manufacturing, Processing, Distribution and Commerce, and Use Bans of Poly chlorinated Biphenyls. The instructions require that requests to participate in the hearing, to be conducted August 21, 1978, are to be sent to the above address by July 31, 1978. Monsanto Company does not at this time plan to present testimony at the hearing. We do, however, desire to participate in the hearing in every other respect, including the questioning of witnesses which do testify. We are, therefore, hereby giving notice to this effect. Thank you for including this notification in the record of this proceeding. Yours very truly Phocion S. Park Environmental Counsel \ CC85C7 March 16, 1978 Mr. Ban Leolich Office of Toxic Substances (TS-788) Environnantal Protaction Agencv 401 M Streat S.W. Washington, D.C. 20460 Daar Mr. Lemlichj Confirming our talaphona convareation of March 14, 1978, you approval our proposed nodificationa to tha Ml and Mg label* for PCB'a. Tha M|_ change involved tha "A toxic environmental..." statement where "with" in tha second line dropped to tha third line, "Regulations" to tha fourth line and "contact" to tha fifth line. Tha M* change concerned tha "for proper d i s p o s a l . s t a t e m e n t where "information" was moved to the first line and "U.S. Environmental" to the second line with all copy centered right and left. These modifications afford maxima, overall type aise utilisation with adequate word spacing in the area de fined for these labels by the regulations published in the Federal Register dated February 17, 1978. Based upon your approval, we are proceeding to have labels p r o d u c * d l n ordar to .chi.ve the riieet p o o i M compliance. vahvc tr u ly i Kenneth E. Smith. L a b e lin g S u p e rv iso r \ \