|
|
|
Originally published in
Environmental Law Reporter, April 1993, vol. 23, pp. 10216-27. Burning Mad:The Controversy Over Treatment of Hazardous Waste in Incinerators, Boilers, and Industrial Furnacesby David B. Kopel
This
Article sorts through the competing, and sometimes contradictory, claims about
the thermal destruction of hazardous waste. The Article deals with the burning
of hazardous waste in incinerators, boilers, and industrial furnaces. The
Article focuses only on hazardous waste as defined by the federal Resource
Conservation and Recovery Act (RCRA), 2 and not on the burning of
municipal solid waste or hazardous wastes mixed with radioactive wastes (mixed
waste). After providing a background to the development of the present
controversy, the author offers an overview of how thermal destruction devices
operate, and why these devices pose special regulatory difficulties. The Article
then details the special approaches that the U.S. Environmental Protection
Agency (EPA) and the states have taken toward dealing with incinerators and
BIFs, with particular attention to the most controversial of all BIFs: cement
kilns that burn hazardous waste. Following a summary of permitting issues unique
to incinerators and BIFs, the Article concludes with an analysis of the policy
issues that the burning controversy raises. When EPA finally promulgated core RCRA regulations around 1980, 4 the costs of land disposal began to climb sharply. Hazardous waste had to be sent to an approved treatment, storage, or disposal (TSD) facility, and such facilities were required to maintain increasingly expensive systems to prevent release of their hazardous wastes into the environment. In 1984, Congress enacted the Hazardous and Solid Waste Amendments (HSWA), 5 which amended RCRA to impose a near-total ban on land disposal of hazardous waste. Pursuant to the HSWA, EPA in the late 1980s and early 1990s promulgated the land disposal restrictions (land ban) 6 to bar land disposal--except under very restrictive conditions--of untreated hazardous waste that poses a potential threat of groundwater contamination. Because land disposal is extremely expensive, other disposal options became increasingly attractive. The Federal Water Pollution Control Act 7 and the Clean Air Act, 8 both of which predate RCRA, had already foreclosed the possibility of disposing of large quantities of hazardous waste in the water or air. Hence, disposal of waste through burning became the most economical, and in some cases, the only option for a large class of hazardous wastes. The concept of disposal through burning has generally been supported by environmental regulators. EPA encourages regulated burning as a treatment option, and considers incineration to be the best demonstrated available technology (BDAT) for most wastes. EPA's land ban regulations actually mandate that certain wastes may be treated only by incineration. 9 The
environmental community, which at first viewed waste-to-energy facilities as
environmentally positive, has become increasingly skeptical of burning hazardous
waste. Currently, almost any government regulation or permit allowing the
burning of hazardous waste is met with vociferous opposition from at least one
environmental or public interest group. Interestingly, the legislative history of RCRA indicates that Congress did not consider incineration to be "disposal"; accordingly, EPA regulates incineration under its authority to regulate "treatment" of hazardous waste. 13 Scientifically speaking, incineration really is treatment rather than disposal. Organic wastes fed into an incinerator are not destroyed, but rather are thermodynamically converted, through oxidation, to simpler forms. The mass at the start of the process is the same as the mass at the end. What changes is that larger molecules are rapidly oxidized and broken down into smaller molecules--mostly water and carbon dioxide. The heat resulting from incineration is the energy released which is no longer used to hold the larger molecules together. Almost
all states have incinerator regulations, similar to the federal regulations,
14 in their own hazardous waste programs. In April 1990, EPA proposed
additional incinerator regulations, the main element of which was to place
additional controls on particular types of incinerator emissions. 15
EPA has not announced plans to issue the draft 1990 incinerator regulations in
final form. Currently, EPA applies the proposed 1990 regulations on a
case-by-case basis during the permitting process. 16 Some states are
moving to adopt the 1990 EPA proposals in their own incinerator regulations. The BIF rule incorporates many of the existing rules applicable to incinerators, as well as EPA's proposed but unfinalized 1990 incinerator rules. Thus, formally speaking, the current BIF regulations are actually stricter than the current incinerator regulations. 22 States that want to have an EPA-authorized program, so that primary enforcement regarding hazardous waste boilers and industrial furnaces is ceded to the state, must adopt BIF regulations by July 1, 1993, if no statutory change is needed for the regulations, or by July 1, 1994, if statutory authorization is needed. 23 Before
the effective date of EPA's BIF rule, about 925 boilers in the United States
were burning hazardous waste fuels. 24 About 600 of them are exempt
from most of the BIF rule because they qualify as "small quantity burners."
25 Of the 325 non-exempt boilers, about 200 are expected to stop burning
hazardous waste because of increased cost attributable to the BIF rule. Thus,
about 125 boilers will be operating under the BIF rule. 26
Additionally, some classes of industrial furnaces, particularly cement kilns,
have been granted waivers from part of the BIF rule. More importantly, no matter how much a thermal device is studied, it is impossible to know with certainty what is going on inside. The average temperature inside a thermal destruction device is generally at least 1,400oF, and temperatures sometimes exceed 3,000oF. Accordingly, it is too hot for a person to stand inside and watch what goes on. Likewise, standard monitoring devices have difficulty operating precisely. Since the inside of the device cannot be monitored effectively, regulators must rely on information gleaned from air emissions and other outputs, such as ash, to attempt to discern what happens inside the devices. But relying on emissions is also difficult. Emissions monitoring is imprecise and costly. State-of-the-art commercial emissions monitors cannot continuously measure releases of the most toxic emissions, such as heavy metals and dioxins. Such releases are sampled only occasionally, and lab analysis is quite expensive. Continuous monitoring of toxic releases may become feasible in the middle-term future. The U.S. Department of Energy predicts Fourier transform infrared spectroscopy (FTIR) will be increasingly used commercially within the next several years. 32 This technology is capable of detecting emissions at the parts per billion level. In the more distant future, Cornell's Solid Waste Combustion Institute group foresees emissions monitors that are accurate at parts per trillion. In practical terms, while very sophisticated monitoring may be feasible in the laboratory, continuous emissions monitoring for small quantities of toxic chemicals is only beginning to be made to work in the demanding environment of an emissions stack. As a result of the monitoring problems, thermal destruction device operators simply do not know what kind of air pollution they are creating with the degree of certainty demanded by the public. At first impression, it appears that what goes out of a thermal destruction device is similar to what goes in. If the device is burning spent toluene, the most dangerous emission from the stack would be uncombusted spent toluene. But, the products that result from the burning of hazardous waste can be considerably more dangerous than the original wastes. Since no burning device operates at 100 percent efficiency, some of the items end up only partially burned. These are called products of incomplete combustion (PICs). PICs are created when fragments of partially burned materials stabilize or recombine to form new chemicals. A simple type of PIC is carbon monoxide (CO); rather than being fully oxidized and acquiring two oxygen atoms to become carbon dioxide (CO2), the carbon atom acquires only a single oxygen atom to become CO. Among the most dangerous PICs are dioxins, furans, polychlorinated biphenyls (PCBs), and other complex organochlorines, with dioxins and other halocarbons being the most dangerous. 33 One type of PIC is made from fragments of original input constituents which result from partial oxidation or simple substitution reactions. For example, the burning of PCBs produces hexachlorobenzene as a by-product. Another type of PIC is a reaction product, resulting from recombination. As recombinations, these PICs usually have a high molecular weight. Examples include naphthalene, fluoranthane, and pyrene. A third type of PIC is a simple fragment which is a universal by-product of combustion of organic compounds. As a simple fragment, this type of PIC is usually of lower molecular weight. Examples includes chloroform, carbon tetrachloride, trichloroethylene (TCE), tetrachloroethylene, benzene, phenol, toluene, and chlorobenzene. Due to the primitive state of emissions monitoring, many facilities may not know exactly what PICs they are emitting. Only zero to 60 percent of unburned hydrocarbon (HC) emissions at any particular facility have been identified. 34 Some of these unidentified HC emissions may be toxic PICs. Dioxins and other chlorinated PICs are often formed in cooler areas--such as in the air emissions stack, where the gas temperature may have cooled to about 450 degrees, or in the ash--rather than during the actual burning. 35 Hence, high temperatures in the burning process are not a complete protection against PICs. In fact, high temperatures and lots of oxygen--conditions associated with a high level of combustion efficiency--sometimes result in their own set of PICs. Significant deviations from operating parameters are called "upsets," and PICs form at higher rates during upsets. Upsets are not at all uncommon, although without continuous monitoring it is difficult accurately to gauge their frequency. Some upsets may be just a minor deviation from numerical standards. For example, if the boiler is supposed to operate between 1,600 and 1,700 degrees, the temperature might drop to 1,580 degrees for a minute. Other upsets may be much more spectacular--and dangerous. A gas buildup in a furnace can spew pollutants 240 feet in the air, blow the furnace door off, and melt light sockets 30 feet away from the furnace. Upsets are best avoided through redundancy, so that if one component fails, there are backups. In
sum, it is impossible to tell exactly what is going on inside a thermal
destruction device. It is impossible to tell what quantity and type of PICs are
being formed, even though PICs may be the most dangerous pollutant emitted by
the burning process. The lack of direct knowledge leads regulators to grab hold
of a variety of surrogate indicators, in the hope that when enough surrogates
are controlled, the burning process itself and its by-products will be brought
under reasonable control. For example, technology currently exists which makes
it easy to monitor CO output on a minute-by-minute basis. Yet, monitoring CO
output is only an imperfect indicator of what is going on during the burning
process. In addition, regulators attempt to assure that the key conditions for
efficient combustion--time, temperature, and turbidity--exist. The particular
regulations affecting incinerators and BIFs, especially the monitoring of
surrogate indicators, is the topic of the next section. 36 The 99.99 percent figure is the required destruction and removal efficiency (DRE). 37 If a facility accepts, for example, 10,000 pounds of spent toluene, 9,999 pounds must be destroyed during combustion--or more precisely, changed into simpler molecular forms, captured by the air pollution control devices, or incorporated into solid residues such as ash. Only one pound of the 10,000 pounds of waste input may be emitted into the environment through the facility's stack. 38 For most wastes, the minimum DRE is 99.99 percent--"four nines." 39 Thermal devices which accept PCB or dioxin-bearing wastes must achieve 99.9999 percent--"six nines." 40 Unfortunately, current emissions equipment does not allow continuous monitoring of the actual DRE during operations. Instead, facilities applying for a permit must conduct a test burn demonstrating that under all potential burning conditions at the facility, the DRE is at least 99.99 percent. Even the test burn is problematic, since it is impossible under current technology to monitor DRE for every component. To solve the problem, the test burn must monitor the destruction of several indicator chemicals that are unusually difficult to burn. 41 The indicator chemicals are called principal organic hazardous constituents (POHCs). Carbon tetrachloride or chlorobenzene are commonly chosen, since they have high "thermal stability"--that is, they are hard to combust. Unfortunately, one POHC's DRE may not necessarily be a good indicator for the burning of other wastes. In addition, some scientists contend, the DRE estimates from the trial burn may be wildly overoptimistic. 42 Boilers are known to produce a "hysteresis effect," whereby the slow movement of POHCs and other emissions into the stack may continue for hours after the test burn is ended. 43 In the worst case, the hysteresis effect could cause underestimation of stack pollutant output by up to two orders of magnitude, although other scientists would say the effect is far smaller. Scientists also argue that the 99.99 percent DRE is based on making best-case assumptions in data interpretation, which may substantially overstate actual DRE. 44 Even if the errors were few and the DRE were still fairly high, at 99.00 percent, the result would be 100 times more emissions than assumed by the 99.99 percent DRE. Another DRE problem is that, according to EPA-funded research, the proper DRE cannot be achieved for wastes that occur in low concentrations in the feedstock. The "six nines" cannot be met if the substance to be destroyed amounts to less than 10,000 parts per million. The "four nines" cannot be met if the substance amounts to less than 1,000 parts per million. 45 The problem for low-concentration wastes is most relevant for dioxins, furans, PCBs, sludges, contaminated solids, and wood-treatment waste.
Finally, even if a high DRE is achieved--the input wastes are almost entirely
absent from smokestack output--new, more dangerous products may be formed during
combustion. These PICs may be substantially more menacing than the original
waste inputs. Critics charge that CO monitoring can reveal gross upsets, but does not reveal the production of PICs under optimal conditions. EPA counters that low CO levels have been demonstrated to indicate that PIC formation is low. The only degree to which CO is a poor surrogate, EPA believes, is that CO may set off false alarms; high CO levels are not always associated with high levels of PICs. 46 Some environmentalists complain that CO continuous monitoring standards for BIFs are based on an hourly rolling average, 47 which understates small upsets and prevents automatic waste feed cutoffs from being triggered. 48 The automatic waste feed cutoff, which is mandatory at all BIFs and incinerators, requires that equipment be configured to automatically shut off waste input and burning whenever particular operating conditions are not met. 49 The
actual levels of permissible CO output are set up on a two-tier system. The
two-tiered CO standard is intended to be flexible, because EPA aimed to avoid
major economic impacts on the regulated community. Under tier I, a facility must
monitor CO in flue gasses (stack gasses) and must not exceed a 100 parts per
million by volume (ppm/v) limit on emissions. 50 Under tier II, the
facility can exceed the 100 ppm/v on CO if the HC concentration in stack gas
does not exceed 20 ppm/v.51
The CO standards include a special, controversial exemption for certain BIFs.
52
The most important metals of concern for thermal devices are arsenic, beryllium, cadmium, and chromium--which are all carcinogens, 53 and antimony, barium, lead, mercury, thallium, and silver--which are noncarcinogens. Metals are not destroyed or broken down into simpler molecules by combustion. Burning may, however, change the form of metal, as from an elemental form to a metallic oxide or an organometallic complex. 54 Or the metal could be changed from a solid state to a vaporous or fine-particle state. Some of the metallic oxides created by incineration are more toxic than the base metal that entered the incinerator. Similarly, the solid metals that enter the thermal device, if changed into metallic vapors, are lighter, more mobile, and hence more easily inhaled or ingested. Metals which are emitted from thermal devices are usually emitted as a particulate, sometimes as metallic vapors. 55 While metals emissions are not feasibly monitored on a continuous basis, the regulations require periodic emissions testing. 56 Only the BIF regulations have specific metals standards. The draft April 1990 incinerator regulations have similar standards which, although they are not yet law, are usually put into effect by incinerator permit writers pursuant to the omnibus permitting authority. The BIF metals standards use a three-tiered control system. The principle of the three-tiered system is that the more evidence a facility can provide demonstrating the safety of its emissions, the larger the allowable level. Under tier I, no emissions testing or site-specific dispersion modeling is needed. The assumption is made that 100 percent of input metals are emitted into the air. Then, using appendix I of the rule, the permit writer cross-references the metals input, which is also the assumed output, with the facility's stack height, nearby terrain, and nearby land use to arrive at a metals feed rate limit. 57 Under tier II, the facility performs site-specific emissions testing, to demonstrate how many metals are captured before they exit through the stack. The facility determines how much of the metals go into ash; how much into the product--if there is a product, as at a cement kiln; and how much is captured by the air pollution control equipment. Then, knowing how many metals are emitted, the facility looks at the appendix I table to factor in stack height, terrain, and land use. Appendix I then supplies the allowable emission rate. If a facility has more than one stack, permissible rates are based on worst-case stack. 58 Tier III requires a facility to do all the testing required for tier II, and to also conduct site-specific dispersion modeling and emissions testing to produce a risk assessment. For carcinogenic metals, the legally acceptable level is a lifetime cancer risk of less than 1/100,000 risk for the maximum exposed individual. 59 For the noncarcinogenic metals, there are specific reference air concentrations that cannot be exceeded. 60 The flexible three-tier system allows a facility to use different tiers for different metals. Under tiers I and II, the emissions rates are based on the worst-case stack; under tier III, a facility can instead look at its stacks in aggregate. 61 Some
people have criticized the metals rules for dealing only with air emissions, and
thus placing no limits on metals in ash or residues. But for all incinerators
and BIFs other than cement kilns, ashes and residues have to be treated as
hazardous waste anyway, since they are derived from hazardous waste. 62
Accordingly, the ashes and residues are subject to the normal panoply of strict
hazardous waste controls. The
legal PM standard allows particulate emissions of up to .08 grains per dry
standard cubic foot (gr/dscf). 64 EPA had proposed a .03 gr/dscf
standard for incinerators, but was overruled by the White House Office of
Management and Budget. EPA headquarters ordered its regional office to implement
the .03 standard anyway, by forcing facilities to agree to it as part of the
permitting process.
65 Several states have adopted their own, stricter PM standards. Stringent
PM standards are relatively easy for new facilities to meet, but may be very
difficult at facilities that need to be retrofitted. Unlike
incinerators, BIFs are not required to employ technology-based emissions
controls that achieve a 99 percent reduction in hydrogen chloride (Hcl)
emissions in stack gas. 69 The rationales for the lesser standards on
BIFs are that such reductions are not technically feasible at reasonable
expense, that BIFs produce little chlorine anyway, and there is no showing that
a 99 percent Hcl reduction is necessary to protect health. 70 During the interim status period, a facility must operate in conformity with most of the same legal standards that would apply to a facility granted a permit. During interim status, the only emission standard not applicable is the 99.99 percent DRE. 71 Under the BIF rule approach to interim status, actual monitoring of compliance with the interim status standards is left largely to the facility. For example, a BIF must use engineering judgment to certify that emissions will likely not exceed emissions standards. By August 1991, BIFs wanting interim status were required to submit a certificate of precompliance, attesting that emissions of metals, Hcl, cl2, CO, and PM were likely within legal limits. 72 The facilities were also required to publish in a local newspaper of general circulation information including a facility description and the type and quantity of waste burned. 73 By August 1992, every facility was required to submit a certificate of compliance, detailing the results of a test burn demonstrating compliance with emissions standards. 74 When the BIF interim status rules were announced, they provoked intense controversy, because they allowed BIFs to continue in operation for several years before regulatory agencies certified their safe operations. In practice, the long window provided by interim status is less significant than had originally been feared. Many states put their BIF permit applications on a fast review track. Faced with having to obtain and comply with a permit, rather than merely self-certifying that interim status standards are met, many industrial furnaces, particularly cement kilns, have stopped burning hazardous waste. In addition, EPA has denied interim status to several kilns, particularly those that began burning hazardous waste just before BIF regulations went into effect.
Small Quantity Burners To qualify for the small quantity burner exemption, the facility must make hazardous waste 1 percent or less of the total fuel burned. 77 The hazardous waste burned must have a heating value equivalent to low-grade coal (at least 5,000 btu/lb). 78 The heating value standard is designed to ensure that the waste is really burned for fuel, and not simply thrown in a boiler to avoid the expense of disposal. In addition, a high heating value is thought to promote combustion that will be efficient enough to achieve a high DRE of toxic organic constituents. Unlike permitted facilities that emit metals or hydrogen chloride under tier III of BIF/incinerator regulations, small quantity burners are not allowed increased emissions based on favorable terrain. Instead, the small quantity burner regulations assume that the burner operates in the worst possible terrain--that which will promote air and water-borne movement of emissions toward vulnerable human populations. EPA reasons that because small quantity burners do not need permits, there would be no government oversight of operators' classification of terrain type. 79 And, since there will be no trial burn to verify that "the four nines" (99.99 percent) are met, the regulations assume that the small quantity burner will achieve only a 99 percent DRE. The actual amount that may be burned per month depends on the small quantity burning unit's stack height. 80 The emission limits are based on avoiding lifetime cancer or mortality of risks more than 1 x 10-5 for an individual living at ground level point of maximum exposure. Small quantity burners may not burn dioxins at all. A
small quantity burner must provide a one-time notice to EPA, 81 and
must keep records demonstrating compliance with all applicable standards.
82 The first step in the permitting process for existing facilities is the pre-trial burn period. The period begins with introduction of hazardous waste into the unit. During the pre-trial burn period--the "shakedown period," the facility may burn hazardous waste for up to 720 hours to get the unit up to operational readiness for the trial burn. 86 New facilities do not have a pre-trial burn period, but commence directly with the trial burn phase. In the trial burn, the facility must demonstrate that it can meet the 99.99 percent DRE and all emissions standards, under all potential operating variables. 87 A trial burn may cost $100,000 or more. Before a trial burn can begin at an existing facility, the facility must obtain a draft permit specifying trial burn conditions. The problem with a trial burn, of course, is that the facility's staff is probably operating with a high degree of care, but may not always so operate in the future. During the post-trial burn period, regulators and the facility analyze the trial burn results. 88 Lastly, the successful facility is granted a final permit, and begins operations in the final permit period. 89 If a proposed new facility, rather than an existing facility, applies for a permit, construction cannot begin until a final permit has been issued. The final permit will specify trial burn conditions, and if the trial burn is successful, the facility will be allowed to commence operations. Permits will set allowable operating conditions based on reliably measured control parameters. The parameters will include, inter alia: operating temperature; permissible waste inputs (operators must conduct waste analysis throughout the permit period to insure that only authorized wastes are burned); waste feed rates; combustion chamber pressure; emissions standards; and conditions for activation of the automatic waste feed cutoff. 90 As long as the facility complies with its permit conditions, it is conclusively presumed that the facility complies with the legal emissions standards. 91 Ed
Kleppinger, a critic of cement kilns, suggests that any thermal destruction
device should have the following requirements in its permit: mandatory use of
the BDAT; round-the-clock inspectors; on-line stack gas readouts available to
the community; stipulated penalties for violating operating parameters;
automatic shutdown if the automatic alarms trip more than once a month, with the
shutdown continuing until the problem is fixed; permits limited to no more than
five years; permit renewal only upon proof of achievement of current BDAT; and
community involvement paid for by the facility.
92 Some states have adopted some of the Kleppinger proposals in their own
hazardous waste programs. In the interior of the kiln, the maximum temperature may reach 3,500oF, although 2,700oF would be more standard. At these very hot temperatures, the input materials change--specifically, the limestone calcines--and fuse into something called "clinker." The clinker--the kiln's product--leaves the kiln at the lower end, is cooled, ground with gypsum into a fine powder, and transported to ready-mix concrete makers. Clinker requires metals for compressive strength; hence, kilns find that incorporation of metals from hazardous waste into the clinker may improve its quality. 93 After the clinker has been turned into cement, the cement is mixed with aggregate, water, and other materials to make concrete. Cement is basically the glue which makes concrete. 94 The amount of hazardous waste burning in kilns is rapidly increasing. Cement kilns currently burn 2 billion pounds per year of waste. 95 Of the 116 cement kilns in the United States, 40 burn hazardous waste. The number has increased by nearly 50 percent in the last few years. The amount of hazardous waste burned annually in cement kilns is the equivalent of 168 million gallons of oil, or 1 million tons of coal. 96 Opines one cement manufacturer, "It's possible that in the not-too-distant future, cement will just be a by-product of waste burning." 97 Powerful economic incentives promote burning. Cement prices fell 40 percent in the 1980s, thanks in part to competition from South Korea and other nations. 98 Burning hazardous waste lets kilns be paid for accepting fuel, rather than having to pay for coal or oil. When kilns first began burning hazardous waste, they usually burned only "nurse" fuels--high British thermal units (Btu), relatively clean, liquid wastes. Now, nurse fuels are more and more burned on-site by their generator in a captive boiler or furnace. Cement kilns today burn large amounts of low Btu solids or sludges, which have been mixed with high Btu liquids. The hazardous waste fuel can include filter cartridges from dry cleaner plants or cleaning rags from automotive repair shops. 99 While kilns operate at very high temperatures, extreme heat is not necessarily a virtue from an environmental viewpoint. Super-high temperatures are not necessary for proper combustion. Indeed, high temperatures can produce rapid combustion, exhaust available oxygen, and create pyrolysis conditions, which promote PIC formation. In addition, kilns are hot at the front end. Incinerators are hot at the other end, and are therefore better protected against upsets. Although kilns do burn waste for longer periods than many incinerators do, the length of burning is not always important. The molecular reaction to be accomplished by burning takes place in microseconds. Good combustion also requires turbulence and oxygen--both of which are often deficient in kilns. 100 Cement kilns have pockets of oxygen-rich and oxygen-poor areas, caused by cyclones of combustion gasses. The oxygen-poor areas are where pyrolysis occurs. 101 Perhaps the most important reason to be concerned about cement kilns burning hazardous waste is that they were not originally designed to do so. Almost all hazardous waste incinerators are less than five years old and are highly engineered; many kilns are much older, and date from an era when environmental concerns barely existed. 102 Commercial incinerators represent a capital investment of over $100 million. The employees will be aware that a single misstep could cost the facility its hazardous waste permit, and them their jobs. Cement kiln employees, on the other hand, are primarily in the business of producing cement, and may not be as conscious of environmental standards. In addition, cement kilns are run with a much smaller staff (three to five on low shifts) than incinerators (10 to 15 on lowest shift). 103 Hugh Kaufman, an engineer with EPA's Office of Solid Waste, charges that BIF regulations were specially tailored to the financial benefit of the cement kiln hazardous waste incineration industry. EPA, he writes,
An example of the policy Kaufman complains about is BIF rule handling of CO standards. Some BIFs, such as cement kilns, cannot meet the 20 ppm/v CO standard under any circumstances; many cement kilns are not particularly efficient devices, and if they burn natural but dirty fuels such as coal, they may easily exceed the CO limit. In such case, the BIF rule allows the permit writer to establish alternate CO limits. To qualify for the alternate limit, a facility must conduct a risk assessment to show that cancer risk by inhalation for selected PICs is less than 1/100,000. 105 While there is room for debate about whether the BIF rule does in fact treat cement kilns too leniently, there is no doubt that EPA made a deliberate policy decision to write a rule that would allow cement kilns to continue burning hazardous waste. EPA apparently believes that, since land disposal has essentially been closed off, it would be irresponsible for EPA to foreclose one of the few remaining disposal options.
Although there are many reasons to take overly optimistic claims about cement
kilns with a grain of salt, it may well be that cement kilns are not nearly so
dangerous as their critics assert. The Texas Air Control Board recently
completed an 18-month study of the environment around Midlothian, Texas, the
site of two waste-burning cement kilns. 106 From 1,000 samples taken,
only one exceeded state health standards, and that standard involved nuisance
odors rather than health effects; significantly, the Texas health standards are
a degree of magnitude more stringent than federal standards. 107 Arguably, if hazardous waste-derived cement were used at a construction site, and the hazardous constituents in the cement became a health problem, every waste generator that shipped hazardous waste to the cement kiln would be a potentially responsible party under 107 of the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). 108 Such liability, however, would more likely arise in the context of a contribution suit or other private cause of action, for EPA has never taken enforcement action against a generator that sent waste to a properly permitted BIF or incinerator. Critics of "toxic cement" assert that the hazardous constituents in the cement may eventually leach out. The cement kiln industry responds by promising that hazardous waste in cement is like lead in crystal decanters, "chemically bonded in a highly-stable crystalline matrix." 109 Unfortunately, crystal decanters have now been found to leach low levels of lead into wine. While there is potential for some leaching, Portland Cement Association analysis shows that hazardous waste cement has leaching characteristics similar to standard cement. 110 Both types of cement easily pass the toxicity characteristic leaching procedure test. 111 So does cement kiln dust.112 In fact, EPA sometimes orders the use of cement stabilization as CERCLA treatment. Opponents of cement kilns express concern that, since cement is a product, and not a waste, cement produced from kilns burning hazardous waste is subject to none of the hazardous waste regulations. 113 Cement kiln opponents favor laws like one that was recently proposed in the Texas legislature: waste-derived cement would be required to carry a warning that the product was made with hazardous or toxic waste, that the waste may leach out, that buyers use the cement at their own risk, and that they may be partly liable for resulting environmental contamination. And, some businesses and local governments are now refusing to buy cement from kilns that use hazardous waste. 114 Given that the scientific data show waste-derived cement to be no different from ordinary cement, it seems questionable whether warning labels should be mandated.
Cement Kiln Dust Under the Bevill Amendment, 120 cement kiln dust may not be regulated as a hazardous waste unless EPA conducts a study showing it to merit such treatment. When the Bevill Amendment was enacted, cement kilns were not commonly burning hazardous waste. Accordingly, the Bevill Amendment should not apply to cement kiln dust derived from hazardous waste, some kiln critics argue. EPA, however, interprets the Bevill Amendment more broadly, and allows cement kilns burning hazardous waste to, under some conditions, qualify for the Bevill exclusion. Kiln operators that wish dust to qualify for the Bevill exclusion, must show, on a facility-specific basis, that concentrations of toxics in the dust are not significantly higher than in normal cement kiln dust--dust from the kiln when it just burns coal or oil. Similar rules apply to any effort to have BIF residues classified as nonhazardous. 121 If the
facility can actually demonstrate that its waste-derived cement kiln dust is
chemically similar to normal kiln dust, it would make little sense to apply
stricter regulations to the waste-derived dust. Whether a waste is regulated as
hazardous ought to depend primarily on the hazards posed by its particular
chemical characteristics, rather than on its ancestry. If this maximally erroneous idiot suffers at least a 1/100,000 extra lifetime cancer risk, EPA pronounces the BIF/incinerator emissions excessive, and orders them reduced. Are EPA's scientists as misguided as the self-destructive MEI? Not at all. While there are numerous health studies of particular hazardous wastes, there has been virtually no research regarding synergistic, multichemical effects. 123 When two toxins have a synergistic effect, they may interact to cause far more harm than the sum of their dangers. For example, cigarette smoking causes one quanta of risk (risk A), and occupational exposure to radon causes another quanta of risk (risk B). The cancer risk of smokers that have occupational exposure to radon far exceeds A+B. The synergistic effect of smoking and radon causes a danger that is much greater than the sum of its parts. One reason that few synergistic studies have been done is that they are so expensive. The National Toxicology Program estimates that a 13-week, subchronic, single-species study of a mixture of 25 chemicals would cost $33 million. 124 Although captive boilers may burn a single waste stream--such as solvents generated onsite, the pattern for the majority of incinerators and BIFs is to burn a wide and complex variety of hazardous wastes. Accordingly, the potential for synergistic effects could be high. And, since synergistic effects are essentially unknown, EPA attempts to compensate by taking a conservative approach to estimating single-waste risks. Besides basing risk assessments on the hapless MEI, EPA makes a number of other conservative assumptions. EPA assumes that carcinogens have a linear slope effect from high to low dose. 125 In other words, if a 100-gram dose causes a 50 percent risk of cancer, a one-gram dose is assumed to cause a .5 percent risk of cancer. Assuming a slope effect is extremely conservative, since much research shows that substances that are toxic at a high dose may be harmless below a threshold dose. One reason may be that the human body is able to repair cells damaged by carcinogens, as long as the number of cells damaged remains low. At high doses, cells may be damaged faster than the body can repair or remove them, and the cancer may thus take hold and begin to spread. 126 Moreover, many carcinogens have been so classified because they are associated with cancer at the maximum tolerated dose (MTD) in animal studies. The maximum tolerated dose is the largest dose which the laboratory animal can ingest without causing significant adverse effects other than cancer.127 It is entirely possible that substances which are carcinogens at the MTD level may be harmless at lesser doses. 128 EPA also assumes that indoor air contains the same amount of pollutants as outdoor air. For noncarcinogenic health risks, it is assumed that the MEI already receives, through his other activities, a background exposure amounting to 75 percent of the reference dose. For example, if 10 grams of a chemical are harmful, the individual is assumed to have already ingested 7 1/2 grams from other sources. In addition, possible and probable carcinogens are evaluated as if they were known carcinogens. Besides addressing the problem of unknown synergistic effects, EPA's strict, conservative human health standards also serve as a proxy for environmental protection. Since few standards exist for exposure of animals to pollution, except for the Clean Water Act aquatic life standards, 129 very strict protection of human health is used as a proxy for protection of animal health. Because animals usually have lower body weight than humans, and because they usually consume their food from a fairly small geographic region, it makes sense to assume that nothing less than strict human health protection could adequately protect animal health. 130 The conservative EPA assumptions make sense not only because health effects of various chemical combinations are poorly understood, but also because it is difficult to determine with certainty exactly what an incinerator or BIF is actually emitting. Taking into account the extremely conservative assumptions made by EPA in setting BIF and incinerator standards, the health risks posed by a properly operated thermal destruction device appear relatively small. The risks of emitting small quantities of waste into the air look particularly attractive compared to the disposal option that thermal devices have replaced: putting huge quantities of waste into landfills, which could eventually leak and contaminate groundwater. Many scientists would suggest that a BIF or incinerator operating under a strict permit is no more dangerous than a facility commonly considered innocuous, such as a coal-burning cement kiln. And, while it is true that thermal destruction devices emit PICs, so do many other things, such as automobiles, fireplaces, and coal-burning utilities. What makes PICs from incinerators or BIFs particularly hazardous, argue Greenpeace and other anti-incineration groups, is that they burn halogenated wastes--fossil fuels contain few halogens--and are hence more likely to produce particularly toxic PICs, such as chlorinated dioxins, furans, and PCBs. Even so, it may be that a tiny amount of incinerator dioxin is still much less of a health threat than a great deal of benzene, to which Americans are constantly exposed in large doses from normal industrial processes, or vast quantities of unburned hydrocarbons--which result from auto emissions. Moreover, most thermal device PICs are methane or TCE, or some other PIC common to industrial emissions. Because combustion has broken complex molecules into simpler forms, most PICs tend to be simple organic compounds. For the more dangerous PICs, such as dioxins, emission levels are very low. Even if hazardous waste did not exist, cement kilns and other BIFs would continue to operate. They would presumably return to burning fossil fuels rather than hazardous waste. Yet coal, while "natural," is far from environmentally benign. Coal can contain naturally occurring radioactive material, hazardous constituents such as benzene, and polynuclear aromatic hydrocarbons--semi-volatile organic compounds similar to some of the carcinogens in cigarette smoke. Thus, a properly permitted cement kiln burning hazardous waste--which will be subject to strict government oversight--may well be safer than a cement kiln burning coal, which is subject to very little regulation. A hazardous waste incinerator, even one that occasionally makes minor deviations from its permit standards, is a much healthier neighbor than a petrochemical refinery. In sum, while it is understandable that most people want incinerators and BIFs to be carefully regulated, the facilities pose a very small health risk if forced to operate within strict permit limits, particularly when compared with the risks posed by other, more common, industrial processes. Some thermal device critics argue, however, that the size of the risk is irrelevant. Their theory is that it is immoral to subject anyone to an involuntary risk, no matter how slight. Thus, that hazardous waste generation is related to the creation of jobs does not justify the infliction of BIF/incineration risks, even if risks are less than 10-9, on the persons in the distant community where disposal takes place. While the argument about involuntary risk has an appealing moral simplicity, it is wrong. Everyone inflicts involuntary risks on people all the time. As one scholar observes, under a rule forbidding any imposition of involuntary risk, to fly an airplane would require the permission of all those living below the route, and to light a fire might require consent of the entire community. 131 The core of the attack on incinerators and BIFs is, at its heart, not an attack on the safety of the facilities themselves. Just as the spotted owl controversy is really about preserving old growth forests rather than about owls, 132 and just as the disposable diaper controversy is really about America's "throwaway" society rather than about landfill space, the core of the attack on incinerators and BIFs is the perception that they discourage source reduction, even though thermal treatment is expensive. Further, some environmentalists argue that society should eliminate all generation of hazardous waste. Any process, such as thermal treatment, that facilitates waste disposal is seen as impeding the proper goal of abolishing all hazardous waste generation. The validity of the critique of incinerators and BIFs, therefore, depends on the assumption that source reduction is a goal in itself. While EPA and most of the environmental community have made this assumption to varying degrees, critics argue that mandated source reduction is economically inefficient at best, and environmentally dangerous at worst. 133 The
pros and cons of source reduction are outside the scope of this Article. But,
unless it is determined that choking off hazardous waste treatment options is a
desirable goal, it is reasonable to conclude that strictly regulated and
supervised incinerators, boilers, and industrial furnaces have a legitimate role
to play in managing America's hazardous waste. Formerly, an assistant attorney general in the Hazardous Waste Unit of Colorado Attorney General's office, Mr. Kopel is research director of the Independence Institute, a free-market think tank in Golden, Colorado. He graduated with high honors in history from Brown University, and magna cum laude from the University of Michigan Law School. The views expressed in this Article are the author's alone, and are not intended to represent the views of the state of Colorado. Notes
1. 40 C.F.R. 266.100-.112 & apps. I-X. Some scientists argue that the human health risk of dioxin has been vastly overstated. For a brief treatment of the issue written for a generalist audience, see Dixie Lee Ray & Lou Guzzo, Trashing the Planet 88-91 (1990). It should be noted that EPA does not appear to agree that dioxin is less dangerous than previously thought. On January 17, 1992, EPA's staff made a presentation about recent studies on dioxins to Administrator William Reilly. Transparencies 17 and 18 stated:
See Waste
Wars: The Army Opens a New Front, Rachel's Hazardous Waste News,
Apr. 29, 1992, at 2. Of course, while incinerators generally have a greater incentive to stay strictly within the terms of their permit, they do not always succeed, as the problems at Chemical Waste Management's now-suspended Chicago incinerator demonstrate. See Illinois v. CWM Chem. Servs., Inc., No. 91CH4768 (Ill. Cir. Ct. July 7, 1992) (consent decree entered); Illinois v. CWM Chem. Servs., Inc., No. 88CH5048 (Ill. Cir. Ct. Sept. 24, 1990) (consent decree entered on separate issue) (some of the problems involved four disconnections of a CO monitor, one exceedence of the PCB feed rate, relabeling of storage dates of drums containing site-generated wastes, and an explosion inside the rotary kiln caused by the incineration of an explosive chemical; Chemical Waste Management attributes the problems to errors of individual employees acting without authorization, or to other problems beyond Chemical Waste Management's reasonable control, and Chemical Waste Management believes that its Illinois facilities have an overall excellent environmental record); see also Hazardous Waste Incinerators: A Technology Out of Control?, Rachel's Hazardous Waste News, Apr. 15, 1992; Julia Flynn, The Ugly Mess at Waste Management, Bus. Wk., Apr. 13, 1992, at 76-77; Concerns Mount Over Operating Methods of Plants That Incinerate Toxic Waste, Wall St. J., Mar. 20, 1992, at B1, B5. For discussion of numerous problems at an Army incinerator operated during 1990-91, see Pat Costner, Chemical Weapons Demilitarization and Disposal: The Army's Experience at Johnston Atoll Chemical Disposal System, Greenpeace, Apr. 11, 1992. 104. Letter from Hugh
Kaufman, an engineer with EPA's Office of Solid Waste, to William Reilly,
Administrator, EPA (Dec. 7, 1990).
110.
Portland Cement Ass'n, An Analysis of Selected Trace Metals in Cement and Kiln
Dust (1992). |
