Document dQmdXDYOJqoEvGQrzDYX1vkr9

Chlorinated Solvents: Will the Alternatives be Safer? Katy Wolf, Ph.D. Institute for Research and Technical Assistance Los Angeles, California Azita Yazdani Pollution Prevention International Los Angeles, California Pamela Yates University of California, Los Angeles Los Angeles, California Over the next decade, use ofchlorinated solvents, a widely employed class of chemicals, will decline significantly because of increasingly stringent environmental regulations. These solvents pose certain health and envi ronmental problems and they have been heavily scrutinized. The alterna tives to the solvents are being adopted without controls. In some cases, these substances will pose other health and environmental problems that are.likely to be as serious; in other cases, the alternatives have not been examined for their health and environmental effects at all. This case study demonstrates that regulations on chlorinated solvents and their potential alternatives are inconsistent with one another and conflicting. on the new path users should take. At this juncture, there is an opportunity to develop a systematic and coherent approach to regulating the solvents and phasing in safe alternatives. This approach will require a good under standing ofthe processes where chlori nated solvents are used and a knowl edge of the alternatives that are likely to be adopted. It will also require a program to evaluate the health and environmental consequences of em ploying the alternatives so they can be used wisely. Chlorinated solvents have been widely used in a variety fapplications, partic ularly over the last few decades. They have extremely attractive properties and have become ubiquitous in the industrial base. Domestic demand for the solvents amounts tojust underone million metric tons (mt) annually. Chlorinated solvents pose a range of health and environmental problems. The study ofhistorical chlorinated sol vent regulation and use reveals inter. estingtrends. Regulations have pushed users from one chlorinated solvent to another and from chlorinated solvents to other chemical classes which cause health or environmental problems ofa different kind. This case study illus trates that regulations are frequently conflicting and inconsistent, and that there has been no systematic approach to better management ofsolvents. Two of the five chlorinated solvents considered here will be banned over the next 15 years because they contrib ute to stratospheric ozone depletion. The other three will be increasingly restricted because they are suspected carcinogens. Alternatives to these sol vents--including new and existing August 1991 Volume 41, No. 8 chemicals, processes and products-- are being marketed and others will be identified in the future. Five generic classes ofalternatives will be employed in place of the chlorinated solvents. These include flammable solvents which pose a workplace danger and are photochemically reactive; combustible solvents which pose similar problems, hydrochlofofluorocarbons which de plete the ozone layer and contribute to global warming; hydrofluorocarbons and fluorocarbons which contribute to global warming; and aqueous form ulations which require increased en ergy and water use and cause sewer pollution. In the case of aqueous for mulations virtually none of the new substances or additives have been ade quately tested for chronic toxicity. The hydrochlorofluorocarbons are undergo ing a very complete series of toxicity tests under the sponsorship ofa consor tium of worldwide chemical producers. It is unlikely that many of the other alternatives will ever be tested. Existing and future regulations and policies will push users away from chlorinated solvents. These policies, like those of the past, give no guidance Characteristics of Chlorinated Solvents There are five mqjor chlorinated sol vents used extensively in commerce today. Table I lists these solvents, their chemical formulas and their abbrevia tions used below. The chlorinated solvents are widely used in a variety of industries for a range of purposes. They show excel lent solvency for many contaminants. Chlorinated solvents are compatible with most substrate materials and are nonflammable. They were introduced into the market after World War II and have found widespread use, partic ularly in the last two decades. Over the next 15 years, use of all five of the solvents is expected to decline signifi cantly because ofexisting and impend ing regulations. Health and Environmental Features The health and environmental char acteristics of the five chlorinated sol vents are summarized in Table II. The Copyright 1991--Air & Watte ManageoieiiiAovocUtion 1055 SL 037306 Chlorinated Solvents: Will the Alternatives b Safer? Katy Wolf, Ph.D. Institute for Research and Technical Assistance Los Angeles, California 'O' JG l Azita Yazdani Pollution Prevention International Los Angeles, California ' Pamela Yates University of California, Los Angeles Los Angeles, California Over the next decade, use ofchlorinated solvents, a widely employed class of chemicals, will decline significantly because of increasingly stringent environmental regulations. These solvents pose certain health and envi ronmental problems and they have been heavily scrutinized. The alterna tives to the solvents are being adopted without controls. In some cases, these substances will pose other health and environmental problems that are.likely to be as serious; in other cases, the alternatives have not been examined for their health and environmental effects atall This case study - demonstrates that regulations on chlorinated solvents and their potential alternatives are inconsistent with one another and conflicting. on the new path users should taka At this juncture, there is an opportunity to develop a systematic and coherent approach to regulating the solvents and phasing in safe alternatives. This approach will require a good under standing ofthe processes where chlori nated solvents are used and a knowl edge of the alternatives that are likely to he adopted. It will also require a program to evaluate the health and environmental consequences of em ploying the alternatives so they can be used wisely. Chlorinated solvents have been widely used in a variety ofapplications, partic ularly over the last few decades. They have extremely attractive properties and have become ubiquitous in the industrial base. Domestic demand for the solvents amounts tojust under one million metric tons (mt) annually. Chlorinated solvents pose a range of health and environmental problems. The study ofhistorical chlorinated sol vent regulation and use reveals inter estingtrends. Regulations have pushed users from one chlorinated solvent to another and from chlorinated solvents to other chemical classes which cause health or environmental problems ofa different kind. This case study illus trates that regulations are frequently conflicting and inconsistent, and that there has been no systematic approach to better management ofsolvents. Two of the five chlorinated solvents considered here will be banned over the next 15 years because they contrib ute to stratospheric ozone depletion. The other three will be increasingly restricted because they are suspected carcinogens. Alternatives to these sol vents--including new and existing chemicals, processes and products-- are bring marketed and others will be identified in the future. Five generic classes ofalternatives will be employed in place of the chlorinated solvents. These include flammable solvents which pose a workplace danger and are photochemically reactive; combustible solvents which pose similar problems, hydrochlorofluorocarbons which de plete the ozone layer and contribute to global wanning; hydrofluorocarbons and fluorocarbons which contribute to global warming; and aqueous form ulations which require increased en ergy and water use and cause sewer pollution. In the case of aqueous for mulations virtually none of the new substances or additives have been ade quately tested for chronic toxicity. The hydrochlorofluorocarbons are undergo ing a very complete series of toxicity tests under the sponsorship ofa consor tium ofworldwide chemical producers. It is unlikely that many of the other alternatives will ever be tested. Existing and future regulations and policies will push users away from chlorinated solvents. These policies, like those of the past, give no guidances Characteristics of Chlorinated Solvents There axe five mtyor chlorinated sol vents used extensively in commerce today. Table I lists these solvents, their chemical formulas and their abbrevia tions used below. The chlorinated solvents are widely used in a variety of industries for a range of purposes. They, show excel lent solvency for many contaminants. Chlorinated solvents are compatible with most substrate materials and are nonflammable. They were introduced into the market after World War II and have found widespread use, partic ularly in the last two decades. Over the next 15 years, use of all five of the solvents is expected to decline signifi cantly because ofexisting and impend ing regulations. Health and Environmental Features The health and environmental char acteristics of the five chlorinated sol vents are summarized in Table II. The Copyright 1991--Air ManagementAwodalioo August 1991 Volume 41. No. 8 SL 037307 105! Table I. Major chlorinated solvents. Solvent Chemical formula Trichloroethylene Perchloroethylene Methylene Chloride 1.1.1-Trichloroethane 1.1.2-Trichloro-1,2,2- trifluroethane CHCI2CH2C1 CC12CC12 CH2C12 CCI3CH3 CC12FCC1F2 Abbreviation TCE PERC METH TCA CPC-113 Other names Tetrachioroethylene Dichloromethane Methyl Chloroform Fluorocarbon-113 Freon-113 first column lists the Permissible Expo sure Level (PEL) in the workplace set by the Occupational Safety and Health Administration (OSHA). It is the time weighted average exposure allowed for a worker in parts per million (ppm) for an eight-hour work day, 40 hour work week. In general, the lower the PEL, the more acutely toxic is the chemical. PERC and TCE have low PELs, whereas CFC-113 has a PEL of 1,000 ppm, the highest value assigned to any chemical. OSHA is expected to lower the PEL of METH from the current level to 25 ppm within the next few months. The second column of Table II indi cates whether or not the chemical causes cancer in laboratory animals. Althoughthe results remain controver sial, at one time or another, TCE, PERC and METH have given positive carcinogenicity results. EPA considers TCE to be a probable human carcino gen based on several positive mouse inhalation studies and one marginally positive rat inhalation study.1 There have been positive mice studies on PERC and one rat study demonstrated elevated levels of cell leukemia and kidney tumors.2 The International Agency for Research on Cancer (IARC) classifies PERC in Group 2B, "possi bly carcinogenic to humans."3 A Na tional Toxicology Program (NTP) study on METH indicated that it caused cancer in both mice and rats.4 The IARC places METH, like PERC, in Catagory 2B.6 The third column of Table II indi cates whether the chemical is regu lated as a precursor to photochemical smog (ozone) in the lower atmosphere or troposphere. All chemicals are con sidered to contribute to ozone forma tion unless specifically exempted. METH, TCA and CFC-113 are ex empted under the Clean Air Act (CAA) and most local air districts, whereas TCE and PERC are not.6 TCE and PERC are regulated under most state implementation plans (SIPs) devel oped to attain the national ambient air quality standard (NAAQS) for ozone. There is evidence that PERC does not actually contribute to smog formation but it is not likely to be exempted because it is a suspected carcinogen. The fourth column of Table II indi cates whether or not the chemical causes depletion of the stratospheric ozone layer. CFC-113 and to a smaller extent, TCA, contribute to ozonedeple tion. Under the provisions of the Lon don Amendments to the Montreal Pro tocol CFC-113 will be banned worldwide by the year 2000; TCA will be banned somewhat later--in 2005. The new Clean Air Act Amendments call for a ban of the CFCs in 2000 and ofTCA in 2002.9 The fifth column of Table II indi cates whether the chemical is consid ered a hazardous air pollutant under Section 112 of the Clean Air Act. In 1985, EPA published an intent to so list TCE, PERC and METH but never did so.W Under the Clean Air Act Amendments, TCE, PERC, METH and TCA are designated as toxicaircontam inants.9 In California, the state Air Resources Board (ARB) has classified METH as a toxic air contaminant and is expected to so classify PERC in the near future. Solvent End Use Table III summarizes the nuyor end uses of the chlorinated solvents. TCE is used primarily in cleaning applications--vapor degreasing and cold cleaning. A small amount is used as an intermediate--a chain termina tor in the production of polyvinyl chlo ride. Smaller quantities are also used Table II. Health and environmental characteristics ofthe chlorinated solvents. Solvent Permissible exposure level (ppm) Animal carcinogen Smog contributor Ozone deoleter Toxic air contaminant TCE PERC METH TCA CFC-113 50 25 500* 350 1000 suspect suspect suspect -- -- regulated regulated exempt exempt exempt no no no yes yes listed listed listed listed -- * OSHA is expected to lower the PEL to 25 ppm or below shortly. Sources: References 1-9. in the electronics and textile indus tries. PERC's mqjor use is as a dry clean ing agent. A significant amount is used as an intermediate in the production of CFC-113 and CFC-114. The chemical is also used in cold cleaning and vapor degreasing operations. Small amounts are used in maskant formulations in aerospace coating operations, in elec tronics, in aerosol applications and in the textile processing industry. The primary application ofMETH is paint stripping. It is used in a variety ofother applications as well: as a blow ing agent in the production of flexible elabstock foam; as a vapor pressure depressant and flammability suppres sant in aerosols; in cold cleaning and vapor degreasing; in electronics for stripping photoresist; in pharmaceuti cal operations in pill coating and as a chemical reaction or processing me dium; as a cleanup solvent in the adhe sives industry; as an extraction solvent in the food industry; in pesticide formu lations; and in the textiles industry. TCA, like METH, is ubiquitous. It is used primarily in vapor degreasing and cold cleaning. It is used in aerosol, adhesive and coatings formulations. TCA is employed as an intermediate in the production of a hydrochlorofluoro carbon, HCFC-142b. It is used as a photoresist developer and for printed circuit board defluxing. A small amount of TCA is used in textile processing and in pesticide formulations. A signif icant new use of TCA--one not shown in Table III--is in flexible foam. The primary use of CFC-113 is in the electronics industry for printed circuit board defluxing, for cleaning semiconductors and for various preci sion cleaning operations. It is also used in cold cleaning and vapor degreasing. It is employed as an intermediate in the production of fluoropolymers. A small amount is used for dry cleaning specialty items. In other uses, it func tions as a refrigerant and a foam blow ing agent. The Regulatory Regime There are three major sets of regula tions that have affected the pattern and level of chlorinated solvent use in the past. These same regulations will influence future use of the chemicals as well. One of these--lower atmo spheric smog regulations--has had a significant impact on chlorinated sol vent use over the last fifteen years. The other two--OSHA workplace ex posure levels and ozone depletion regu lations--will influence the pattern of chemical use over the next fifteen years. These regulations have gener ally been imposed independently by different government agencies or by different offices of EPA without regard 105G SL 037308 .1 Air Waste Manaoe A to the consequences and with no coher ent systems approach to developing an overall toxic chemicals policy. As dem onstrated here, the regulations have tended to push users from one target set of substances to another ^et that eventually proves dangerous in an other way or from one target medium to a less regulated medium. This frag mented approach has not led to better protection of human health and the environment. Air Regulations--Photochemical Smog During the 1970s, there was sub stantial emphasis on promulgatingand implementing regulations that would reduce the formation of lower atmo spheric ozone, more commonly known as smog. EPA regulates photochemically-reactive substances or so-called volatile organic compounds (VOCs) un der Section 111 of the Clean Air Act. The VOCs are defined as any volatile compound of carbon except methane, carbon monoxide, carbon dioxide, car bonic acid, metallic carbides or carbon ates, ammonium carbonate and vari ous exempt compounds. "Exempt" compounds, or those not traditionally regulated, include the fully halogenated chlorofluorocarbons (CFCs), TCAandMETH. The South Coast Air Quality Man agement District (SCAQMD) in South ern California is one of the most strin gent in the nation. Over the last several years, SCAQMD's mission has been to reduce the levels of lower atmospheric ozone or smog which is particularly troublesome in the natural meteorolog ical basin surrounding Los Angeles. SCAQMD is the most progressive air district in the nation and the District's VOC rules are frequently adopted later by EPA or by other air districts. Under the SCAQMD rules, when substances are "exempt" their emis sion fees are lower than those of "nonexempt" substances. In June 1989, for firms emitting between five and 25 tons of VOCs, the emission fee was $289 per ton annually. For those emitting more than 25 tons, the fee was $327 per ton. In contrast, the fees for METH, TCA and CFC-113 (which were exempt substances) were lower, at '$52 per ton for emissions between five and 25 tons and $58 per ton for emissions greater than 25 tons.16 In earlier years, the fee structure was similar. CFC-113, and to a smaller extent TCA, are the solvents of choice in electronics. Because of the stringent smog regulations, the same two sol vents--and particularly TCA--have been adopted in place of the photochemically reactive solvents in a range of applications. This in part explains a Table in. U.S. chlorinated solvent use--1988 (thousand metric tons). Application TCE PERC METH TCA CFC-113 Total Vapor degreasing Dry cleaning Intermediate Cold cleaning Electronics Aerosols Paint stripping Adhesives Coatings Flexible foam Pharmaceuticals Textiles Food processing Pesticides Other Total demand* Production 47.1 -- 7.0 14.1 3.2 -- -- -- -- -- -- 1.0 -- -- -- 72 82 18.1 120.0 80.0 6.7 1.3 3.0 -- -- 7.0 -- -- 2.0 -- -- 20.0 258 226 5.8 -- -- 17.2 16.9 20.0 50.0 5.0 -- 23.2 14.4 --- 4.2 1.0 49.3 207 229 106.0 -- 22.5 48.0 17.0 40.8 -- 26.0 17.2 -- -- 7.0 -- 3.0 10.5 298 328 17.7 2.0 5.3 4.2 40.2 0.6 -- -- -- -- -- 0.5 -- -- 7.5 78 78 194.7 122.0 114.6 90.2 78.6 64.4 50.0 31.0 24.2 23.2 14.4 10.5 4.2 4.0 87.3 913 943 * Total demand is assumed to be equal to production minus exports plus imports. Note: The aerosols category includes 9 thousand mt of TCA and 2 thousand mt of METH used in aerosol pesticides. Sources: References 10-15. 20 percent growth in TCA production since 1978. As the figures in Table III indicate, TCA is the most heavily used solvent by far in vapor degreasing and cold cleaning applications. In many adhe sive formulations TCA has increas ingly replaced photochemically reac tive solvents.13'17,18 In coatings, TCA has also replaced these solvents on a large scale. In 1982, the Chemical Marketing Reporter indicated that one percent of TCA demand went toward coating applications (CMR, 1982). In 1989, the same publication stated that five percent of TCA demand went to coatings.19 In the South Coast district, because of the pressure to move away from photochemically reactive sol vents, many adhesive and coating formulators reformulated their products to include TCA so that users in the basin could comply with air regula tions. Military specifications--particu larly for coatings--have been rewrit ten to require TCA-based formulations. This was necessary because of the large concentration of aerospace firms in the southern California area. Over the last decade or so, there has been substantial substitution away from VOCs--the substances that con tribute to smog. To meet the require ments of new smog regulations and to minimize fees, users have increasingly adopted ozone depleting substances. Air Regulations--Stratospheric Ozone Depletion In 1974, the theory of ozone deple tion was first proposed. Certain sub stances, called chlorofluorocarbons or CFCs, are extremely stable in the atmo sphere. Unlike photochemically reac tive substances, which break down readily in the lower atmosphere or troposphere, the CFCs have long atmo spheric lifetimes that in some cases are on the order of 100 years. Eventually, these chemicals make their way to the upper atmosphere or stratosphere. Once there, ultraviolet light impinges upon them, breaking the chemical bond between the carbon atom and the chlo rine atom. The chlorine atom partici pates in a catalytic reaction with ozone and the net effect is depletion of the ozone layer that protects the earth from harmful ultraviolet radiation. In 1976, propellant use of CFC-ll and CFC-12, the two major CFCs em ployed in aerosol applications, amounted to 138 thousand mt.20 In 1978, the first CFC regulatory steps were taken by the Food and Drug Administration (FDA) and U.S. EPA. These agencies acted to ban the use of CFCs in nonessential aerosol applica tions. At the time, CFC use as aerosol propellants represented about half the total U.S. production level of the two CFCs.21 They were favored in certain personal care products--particularly hair sprays, deodorants and antiperspirants--and many spray paints. Hydro carbon propellants were used in many other aerosol products. Alternatives to CFC propellants in cluded hydrocarbons like isobutane and propane, compressed gas propellants like carbon dioxide and substitute pack aging like pump sprays. Hydrocarbon propellants, which are significantly cheaper than CFCs, became the most widely used alternatives. In conjunc tion with the hydrocarbon propellants, formulators began increasingly adopt ing METH which has properties simi lar to CFC-ll. METH functions to suppress the flammability ofthe hydro carbon propellant; it acts to depress the vapor pressure; and it also serves to solubilize the resins and propellants into an alcohol base. The demise of CFCs contributed to growth in METH use in aerosol applications. In 1971, aerosol applications accounted for 10 August 1991 Volume 41, No. 8 SL 037309 1057 percent ofMETH use.22 By 1983, aero sol use--at 24 percent--was the single largest use of the chemical.23 In December of 1985, the FDA pro posed a ban on METH in cosmetic products including those in aerosol packages. This action was based on the results of the NTP study mentioned earlier. The CPSC also considered ban ning consumer products containing METH. InDecember of 1987, the Com mission voted not to restrict METH, but instead to work with industry to develop warning label standards and a consumer education program. Because of the threat of a ban and the practice of labelling, METH use in aerosols declined between 1984 and 1988.5 In many cases, formulators moved to TCA to avoid the labelling requirements of the METH based products. Between 1986 and 1989, the percentage of TCA devoted to aerosol uses increased from seven to 10 percent13'24 and use of TCA in such applications approxi mately doubled. It is worth noting here that in Cali fornia, the ARB and SCAQMD have expressed an intent to regulate aero sols. ARB has passed regulations that will severly limit the content of photochemically reactive substances in aero sol antiperspirants and deodorants over the next few years. They are now targeting other types of aerosol prod ucts. The rule specifies that ozone depleting substances cannot be used in place of VOCs. As discussed later, vir tually all possible substitutes either deplete stratospheric ozone or contrib ute to photochemical smog so the use of aerosol products must necessarily decline substantially in the future. In the aerosol case, there has been significant substitution over the last decade. Formulators moved from CFCs which deplete stratospheric ozone to hydrocarbons which are flammable and contribute to smog and METH, a sus pected carcinogen. They then moved from METH to TCA which depletes the ozone layer, in a sense coming full circle. Now, formulators are beingpres sured to move away from hydrocar bons. Workplace Regulations In 1989, 03HA revised the worker exposure levels for some 400 sub stances. In this process, the PEL of PERC was lowered from 100 to 25 ppm. For the first few years--until December 1992--workers will be al lowed to use personal protective equip ment to satisfy the new requirement. After that, engineering controls will be necessary. As mentioned earlier, since the ma jor use of PERC is in dry cleaning, dry cleaners will be strongly affected by the new regulation. Nationwide, about half the dry cleaners in the country have older transfer machines.26 In these machines, the wash and extrac tion steps take place in one unit and the clothing is transferred to another machine for drying. Dining the trans fer, emissions of PERC occur in the workplace. The other half of the dry cleaners nationwide have the newer dry-to-dry machines where the wash, extraction and drying steps all occur in one unit.25 Dry cleaners with transfer machines are not likely to be able to meet the new OSHA level; most dry cleaners with dry-to-dry machines probably can. New dry-to-dry machines with con trols are expensive--ranging from about $30,000 to $180,000 for a facil ity, and many dry cleaners will be unable to afford the capital outlay. There is likely to be significant consoli dation in the industry as the smaller transfer machine dry cleaners go out of business. PERC use will decline significantly in the future. OSHA is still evaluating METH use and has not yet announced the new workplace exposure level for the chem ical. It is likely that the new PEL will be set over the next few months at 25 ~ppm. This is substantially lower than the present level of 500 ppm shown in Table II. The major use of METH, as indicated in Table III, is in paint strip ping. In the consumer paint removal sector, which accounts for about twofifths ofMETH use in that application, the lower PEL will require significant substitution in contract stripping appli cations and alternative chemical strip pers will have to be employed. Characteristics of Chlorinated Solvent Alternatives The last two sections demonstrated that chlorinated solvents are under intense regulatory scrutiny and are likely to be even more heavily regu lated in the future. CFC-113 and TCA will be banned because they deplete stratospheric ozone. TCE has been and will increasingly be regulated be cause ofits contribution to lower atmo spheric smog and because it is a sus pected carcinogen. The new low workplace exposure level for PERC together with the fact that it is a suspected carcinogen will lead to a much reduced use of the chemical in the future. METH, also a suspected carcinogen, will be assigned a much lower workplace exposure level in the future. Its use is expected to decline significantly. Because of the certain substantial decrease in chlorinated solvent use in the future, alternatives are being inves tigated and marketed vigorously. The methods that will be employed to re duce or eliminate the use of the chlori nated solvents include substitution of other chemicals, substitution of other processes, substitution of other prod ucts and adoption of waste and vapor recycling techniques. In situations where elimination of the chlorinated solvent is necessary because of a ban, only substitution of alternative chemi cals, processes and products is appro priate. Substitute Chemicals All available or emerging alternative chemicals can be classified into generic categories. Table IV lists these catego ries and provides several specific exam ples of each type. The characteristics ofthese alternative chemicals are com pared along a number of dimensions with one another and with the chlori nated solvents. The first column of Table IV shows the ozone depletion potential (ODP) of the chlorinated solvents and their po tential alternatives. The ODP is the potential for ozone depletion of one kilogram of a chemical relative to the potential of one kilogram of CFC-11 which has a defined ozone depletion potential of 1.0. Two factors determine the ODP of a chemical. The longer the atmospheric lifetime, the higher the ODP. The higher the chlorine content, the higher the ODP. CFC-113 has a long atmospheric lifetime because it is fully halogenated. Its ODP is high--at 0.8--TCA has a lower ODP of about 0.1. The chemical is not fully haloge nated and its lifetime is of the order of 6.5 years. The HCFCs are not fully halogenated so they have relatively low ODPs. The ODP of HCFC-225 has not yet been estimated but it is likely to be in the range of those of HCFC123 and HCFC-141b. The HFCs, the FCs, the flammable solvents and the combustible solvents contain no chlo rine, so they do not contribute to ozone depletion. The London Amendments to the Montreal Protocol call for a ban of the fully halogenated CFCs in 2000 and a ban of TCA in 2005. The HCFCs, even though their contribution to ozone depletion is small, are considered only "interim" substitutes. The London Amendments call for a ban of the HCFCs between 2020 and 2040. The Clean Air Act Amendments ban CFCs in 2000 and TCA somewhat earlier than the London Amendments--in 2002.9 The second column of Table IV indi cates whether or not the chemical con tributes to smog. The flammable and combustible solvents are not exempt under the Clean Air Act and are pre sumed to contribute to photochemical smog. TCE and PERC are also not exempt. The CFCs have traditionally been exempted because of their long 1058 SL 037310 J Air Waste Mananp Ae<;nr- atmospheric lifetimes. The HCFCs have recently been exempted by EPA8 and most local air districts. Although their lifetimes are much shorter than those of the fully halogenated CFCs, they are still long enough to prevent them from contributing to smog. Smog data on pentafluoropropanol are not yet available. Flammable solvents are generally low molecular weight hydrocarbons, alcohols and ketones. They have high vapor pressures and evaporate readily. Their atmospheric lifetimes are gener ally short so they readily form ozone precursors. The combustible solvents have short atmospheric lifetimes and also contribute to smog. However, their vapor pressures are generally lower than those of the flammable solvents and TCE, so their rate of evaporation is much lower. D-limonene, a terpene, is one of the most photochemically reactive substances.29 Because of its relatively low vapor pressure, signifi cant evaporation may not occur in some applications; its net contribution to smog may be lower than that of a flammable solvent with a higher vapor pressure that is used in a similar man ner. A disadvantage of the flammable and combustible solvents is that they are regulated by EPA and local air districts because of their contribution to smog. Fees for their emissions will be much higher than for exempt sol vents and in some areas--notably Southern California--air districts may refuse to grant new permits for their use altogether. In other locations, stringent emission controls may be required if the solvents are used. It is worth noting that nearly all possible chemical solvents either con tribute to photochemical smog or de plete the ozone layer. Exceptions are METH, PERC and the HFCs. Both PERC and METH have atmospheric lifetimes in the mid-range--not short enough to contribute to smog nor long enough to cause ozone depletion. PERC, nevertheless, is regulated as a smog contributor. HFCs do not de plete the ozone layer because they do not contain chlorine and they probably do not contribute to smog formation either. The third column of Table IV shows the Global Warming Potential (GWP) of the chemical. The GWP of a chemi cal is the potential of one kilogram of the chemical to cause global warming relative to the potential of one kilo gram of CFC-11 to cause global warm ing. CFC-11 has a defined GWP of 1.0. The GWP is higher for chemicals that have long atmospheric lifetimes and strong absorptions in the infrared spec tral region. Halogens, including chlo rine and fluorine, absorb strongly. The flammable and combustible solvents do not have GWPs because they do not contain halogens. The CFCs--because of their long atmospheric lifetimes-- have high GWPs. HCFCs and HFCs, because of their shorter atmospheric lifetimes, have lower GWPs. The FCs are fully halogenated and thus have long atmospheric lifetimes. They there fore would be expected to contribute significantly to global warming. Al though there are currently no regula tions on global warming gases, it is likely there will be in the future. The fourth column of Table 4 indi cates whether the chemical has a flash point or is combustible. The flam mable solvents have flash points below 100 " F. The combustible solvents have flash points above 100 F. The chlori nated solvents, the CFCs and the HCFCs do not have flash points. How ever, TCA and HCFC-141b will bum under certain conditions. The flash point of pentachloropropanol, if it has one, is not available. Worker safety is tied to the flash point of the chemical. In the 1960s and 1970s, there was a movement away from the flammable solvents to the chlorinated solvents and CFCs because they were perceived to be safer for workers. The fifth column of Table IV speci fies whether the chemical has been tested for chronic toxicity. It is worth noting here that most chemicals used for many years in commerce have not been adequately tested. Nearly all new chemicals and existing chemicals desig nated for new uses are allowed to be marketed without a requirement for testing. Although isopropyl alcohol has been used for many years, EPA has only recently required manufacturers to test it.27 Mineral spirits have similarly been widely used and there is no require ment for testing them at this time. The combustible solvents are rela tively new to the market. A major ingredient of many terpene cleaning formulations--d-limonene--has been tested and found to give positive carci nogenicity results in male rats.28 EPA has recently issued a proposed test rule on NMP.26 Neither DBE nor alkyl acetates have been tested for chronic toxicity. Chlorinated solvents and CFCs have been tested extensively for chronic toxicity; their characteristics were presented in Table II. All three of the HCFCs listed in Table IV are cur rently in testing. The battery oftests is being sponsored by a consortium of worldwide CFC producers. The results for HCFC-123 and HCFC-141b should be available in the 1992-1993 time frame. The testing of HCFC-225 has just been initiated and the results will Table IV. Characteristics of generic solvent categories. Generic solvent category ODP* Tested for Photochemical chronic reactivity GWP8 Flash point* toxicity Flammable solvents Isopropyl alcohol Mineral Spirits Combustible solvents'1 Terpenes DBE NMP Alkyl Acetates Chlorinated solvents TCE PERC METH TCA Chlorofluorocarbons (CFCs) CFC-11 CFC-113 Hydrochlorofluorocarbons (HCFCs) HCFC-123 HCFC-141b HCFC-225 Hydrofluorocarbons (HFCs) and Fluoro carbons (FCs) Pentafluoropropanol ____ -- -- 0.1 1.0 0.8 0.02 0.1-0.18 NA Yes Yes Yes Yesf No No No No -- ____ 0.02 1.0 1.4 0.02 0.09 NA NA NA F Rule issued No C Limited' No Rule issued No Yes ____ Yes ____ Yes ____ Yes -- Yes ____ Yes In testing -- In testing -- In testing NA No * ODP or ozone depletion potential is defined in the text. b GWP or global wanning potential is defined in the text. c F refers to flammable; 0 refers to combustible. d DBE is dibasic esters; NMP is N-Methyl-2-PyrroIidone. ' One of the terpenes, d-limonene, has been tested. f Although PERC is not photochemically reactive, it is not exempt under the Clean Air Act. NA means not available. Sources: References 26-28. August 1991 Volume 41, No. 8 SL 037311 1059 not be available until later. Pentafluoropropanol is not currently being tested. Substitute Processes/Products In addition to chemical substitutes, other processes and products can sub stitute for chlorinated solvents. The major process substitute for the chlori nated solvents is water. Referring to Table III, aqueous based processes can be used to some extent in place of vapor degreasing, cold cleaning, elec tronics, aerosols, adhesives, coatings, textiles and pesticides. The increased use ofwater may carry with it certain disadvantages. First, water cleaners generally have addi tives and some of these additives are either known to pose health problems or they have not yet been tested for chronic toxicity.31 Second, the use of water frequently requires higher en ergy use which may exacerbate global warming. In cleaning, coating and ad hesive applications, in particular, wa ter requires more energy for drying, and in some cleaning applications it requires more energy for heating. Third, in cleaning applications where water is used, metals and organics will enter the sewer. Fourth, increased use of aqueous based systems will lead to a higher demand for water, an increas ingly scarce commodity. Alternative products can substitute for chlorinated solvents as well. In aerosol applications, for instance, pump sprays provide an alternative to aero sol products. In some instances, how ever, like furniture protectors, pump sprays cannot dispense the product in an efficient manner. In paint strip ping, abrasive techniques can be used in place of chlorinated solvents. There remains a question of whether these techniques cause substrate damage or corrosion.32 Summary of Trends In the 1960s and 1970s in particu lar, chlorinated solvents were widely substituted for flammable solvents be cause they better protected workers from flammability. As smog regula tions were promulgated in the late 1960s and early 1970s, users moved from the photochemically reactive flammable solvents and TCE to the other "exempt" chlorinated solvents. In the 1980s, there was increased scru tiny of the chlorinated solvents. TCE, FERC and METH were considered undesirable because of their suspect carcinogenicity, and TCA and CFC113 were being examined for their contribution to stratospheric ozone de pletion. At this stage, as Table IV illustrates, there are really only three choices. 1060 First, flammable solvents are gener ally not a goodchoice because ofworker danger, the expense of rendering the facility explosion-proof and obtaining insurance and the difficulty of getting permits to use photochemically reac tive substances. Furthermore, many of the flammable solvents remain un tested for chronic toxicity. Second, users can adopt the combus tible solvents which are simply nonchlorinated hydrocarbons that are of higher molecular weight than the flam mable solvents. Potentially, there is an infinite number of these combinations of carbon, hydrogen, oxygen or nitro gen and generally these materials are comparatively unscrutinized for their effects on human health and the envi ronment. For instance, there are hun dreds of terpenes. In some cases, the combustible solvents become flam mable upon use and require explosion prooffacilities and the associated insur ance. The fact that they are "biode gradable" simply means they do not contain chlorine. After they have been used to clean, they will contain hazard ous components which generally are not "biodegradable." They do not dry readily because of their comparatively low vapor pressure. Generally they must be used with a water rinse to completely flush the solvent from the part. Drying will thus require higher energy. After the combustible solvents have been used for cleaning, they con tain the residues from the cleaned material, like metal chips, salts and organics such as greases and oils. These contaminants can render the solvent hazardous. Third, just as the combustible sol vents are more complex flammable solvents, the HCFCs, HFCs and FCs are simply variants of the CFCs. The HCFCs and HFCs contain hydrogen, making them less stable in the atmo sphere and less likely to contribute to ozone depletion. The HCFCs are in testing, and their effects on human health and the environment are likely to be better understood than any other set of chemicals that has been mar keted. The HCFCs will eventually be banned because of their contribution to ozone depletion; in the meantime, at least, their health and environmental characteristics will be anticipated and reasonably well understood before they have achieved widespread use. The HFCs and particularly the FCs be cause of their long atmospheric life times may eventually be restricted for their contribution to global warming. In the years to come, there will be a movement from the chlorinated sol vents to combustible solvents, to HCFCs, HFCs and FCs and to aqueous based systems. For the most part, the health and environmental conse quences of the combustible solvents SL 037312 are unknown and most of the new solvents are not likely to be tested for chronic toxidfy. In contrast, the health and environmental effects of the HCFCs will be relatively well under stood before they achieve widespread use. The effects of an increased use of aqueous based processes remain largely unanticipated and the sewer remains the relatively unregulated medium. The increased energy and water re quirements have not been studied; the health and environmental effects of many additives are unknown; and the consequences of increased sewer load ing have not been adequately exam ined. Policy Implications The information presented in this paper demonstrates that past policies have pushed users from one substance to another and from one medium to another without regard for the conse quences. There has never been a coher ent policy for hazardous substances use in the solvents arena. Governmen tal regulations generally provide a clear incentive not to use certain substances; rarely do they give guidance as to what users should adopt instead. We are presently at a crossroads and we have the opportunity to develop more sensible and thoughtful policies. The London Amendments to the Mon treal Protocol call for a ban on many ozone depleting substances used widely in commerce. These include TCA and CFC-113, two of the chlorinated sol vents considered here. Other stringent regulations have been and will be placed on TCE, PERC and METH, the three other chlorinated solvents ad dressed here. As a consequence ofthese regulations, there will be a massive restructuring of the solvent-using in dustries in the years to come. From this work, two specific recom mendations emerge. First, a knowl edge of the industrial processes where solvents are used is critical to deter mine which chemicals and processes may be used in the future and what controls are optimal for each new sol vent or process that may be adopted. The combustible solvents are being marketed heavily in the electronics industry as alternatives to CFC-113 and TCA in printed circuit board defluxing. Only those familiar with the industry and the industrial processes would be aware of this. Aqueous clean ing is also a potential alternative and when it is employed, lead is present in the wastewater and treatment of the effluent will generally be required be fore sewer release. Second, it is clear that a testing program for the flammable solvents, the combustible solvents and the addi tives in aqueous cleaners needs to be established. The health and environ mental effects of the alternatives also require significant investigation. Those substances that are found to pose se vere chronic toxicity problems or threats to the environment can be removed from commerce or prevented from entering the market. A fundamental truth is that it is not possible to conduct industrial opera tions without using or generating haz ardous substances. Even if the sub stances used in solvent operations are not themselves toxic, the contami nants used to remove them frequently are. In effect, there is no free lunch. It is possible to improve the current situ ation, but only by coming to terms with and acknowledging that there will be tradeoffs in adopting the chlori nated solvent alternatives. With this admission, a systematic approach that phases in reasonably safe and scruti nized alternatives can be developed. Without it, policies will continue to repeat the pattern of the past. References 1. U.S. EPA, "Assessment of Trichloro ethylene as a Potentially Toxic Air Pollutant," Federal Register 50:52422, December 23,1985. 2. U.S. EPA "Assessment of Perchloro ethylene as a Potentially Toxic Air Pollutant," Federal Register 50:52879, December 26,1985. 3. Halogenated Solvents Industry Alli ance, "Perchloroethylene, White Paper," Washington, DC, 1989. 4. Menhear, J. H. "NTP Technical Re port on the Toxicology and Carcinogensis Studies of Dichloromethane (Meth ylene Chloride)," National Toxicology Program, 1985. 5. Halogenated Solvents Industry Alli ance, "Methylene Chloride, White Paper," Washington, DC, February 1989. 6. U.S. EPA "Air Quality: Proposed Revi sion to Agency Policy ConcerningOzone SIPs ana Solvent Reactivities," Fed eral Register 48: 49097, October 24, 1983. 7. U.S. EPA "Regulatory Investigation Initiation for Methylene Chloride, Ad vance Notice," Federal Register 50: 42037, October 17,1985. 8. CFC Alliance Bulletin, "Montreal Pro tocol Calls for CFC Phaseout by 2000," Rosahm, VA July 1990. 9. "Conference Report on S.1630, Clean Air Act Amendments of 1990," Con gressional Record, H13101, October 26, 1990. 10. Wolf, K.; Canun, F. "Policies for Chlo rinated Solvent Waste--an Explor- atonr Application of a Model of Chemi cal Life Cycles and Interactions," The Rand Corporation, R-3506-JMO/RC, 1987. 11. United States International Trade Commission, "Synthetic Organic Chemicals, U.S. Production and Sales, 1986 to 1988," Washington, DC, 19861988. 12. "Chemical Profile, Trichloroethylene," Chemical Marketing Reporter, Janu ary 23, 1989. 13. "Chemical Profile. 1,1,1-Trichloro- ethane," Chemical MarketingReporter, January 30,1989. 14. "Chemical Profile, Perchloroethylene," Chemical Marketing Reporter, Febru ary 6,1989. 15. ``Chemical Profile, Methylene Chlo ride," Chemical Marketing Reporter, February 20,1989. 16. South Coast Air Quality Management District, "Rule 301.2. Fee Schedules," El Monte, CA 1988. 17. "Chemical Profile. 1,1,1-Trichlo- roethane," Chemical Marketing Re porter, January 17,1977. 18. "Chemical Profile. 1,1,1-Trichloroethane," Chemical Marketing Re porter, January 22,1974. 19. "Chemical Profile. 1,1,1-Trichloroethane," Chemical Marketing Re porter, September 22,1982. 20. Hammitt, J. K.; Wolf, K. A; Camm, F.; Mooz, W. E.; Quinn, T. H.; Bamezai, A "Product Uses and Market Trends for Potential Ozone Depleting Substances 1985-2000," The Rand Corporation, R-3386-EPA 1986. 21. Palmer, A EL; Mooz, W. E.; Quinn, T. Q.; Wolf, K. A "Economic Implications of Regulating Chlorofluorocarbon Emissions from Nonaerosol Appli cations," The Rand Corporation, R-2524-EPA 1980. 22. "Chemical Profile, Methylene Chlo ride," Chemical Marketing Reporter, February 8,1971. 23. "Chemical Profile, Methylene Chlo ride," Chemical Marketing Reporter, February 28,1983. 24. "Chemical Profile, Methylene Chlo ride," Chemical Marketing Reporter, February 10,1986. 25. International Fabricare Institute, "Fo cus on Dry Cleaning, IFI Equipment and Plant Operations Survey/' Vol ume 13, number 1 (1989). 26. National Toxicology Program, "Techni cal Report on the Toxicology and Car cinogenesis Studies of D-Lunonene in F3441N Rats and B6C3C1 Mice," Draft report, undated. 27. U.S. EPA "Isopropanol; Final Test Rule," Federal Register, 54: 43252, October 23,1989. 28. World Meteorological Organization, "Scientific Assessment ofStratospheric Ozone: 1989," Vol. 1,1990. 29. Damall, K. R.; Lloyd, A. C.; Winer, A M.; Pitts, Jr., J. N. "Reactivity Scale for Atmospheric Hydrocarbons Based on Reaction with Hydroxyl Radical," Environ. Sci. Technol. 10: 692 (1976). 30. U.S. EPA "N-Methylpyrrolidone; Pro posed Test Rule," Federal Register 55: 11398, March 28,1990. 31. U.S. Environmental Protection Agency, Office of Toxic Substances, Personal Communication with K. Wolf, 1990 (U.S. EPA 1990). 32. Wolf, K. "Source Reduction of Chlori nated Solvents in Paint Stripping," paper presented at Fifth Aerospace Hazardous Waste Minimization Conference," Los Angeles, CA May 22-24,1990. K. Wolf is with the Institute for Research and Technical Assistance, 1429 South Bundy Drive, Los Ange les, CA 90025. Azita Yazdani is with Pollution Prevention International, P.O. Box 48161, Los Angeles, CA 90048. P. Yates is with the Univer sity of California at Los Angeles, School of Public Health, Los Ange les, CA 90024-1772. This manu script was submitted for peer review on September 4, 1990. The revised manuscript was received on January 25,1991. August 1991 Volume41, No. 8 SL 037313 1061