| Munitions Management Device Version One
Comments on RCRA Part X Research and Demonstration Permit
prepared by Clearwater Revival Company
on behalf of Non-Stockpile Chemical Weapons Citizens Coalition
January 1999
Project Understanding
The Research, Development, and Demonstration draft permit for the MMD-1 would allow treatment of 1,675 pounds of chemical warfare agent in Building 3445 at Dugway Proving Ground, Toelle County, Utah. Treatment in the MMD-1 is intended to render the chemical warfare agent "...ineffective for use in weapons" (page 4B-1 line 16).
The items intended for treatment include 301 pounds of chemical warfare agent currently stored at Dugway Proving Ground, 352 pounds of chemical agent that will be transported from the Deseret Chemical Depot, 921 pounds of phosgene that will be purchased from a commercial vendor for the purpose of this test. The source of the additional 101 pounds of chemical warfare agent is not specified in the permit, and the permit would allow Dugway to receive chemical warfare agent from Deseret and other Army bases at the Munitions Storage Magazine that is part of the permit application.
The MMD-1 is not a detoxification process. Approximately, 11 pound of acutely hazardous waste will be produced for each pound of chemical agent treated. The draft RCRA permit requires this liquid waste and all other wastes to be incinerated. In addition to commercial hazardous waste incinerators, the draft permit would require that commercial laboratories and hazardous waste transporters handle the residues of this chemical warfare agent treatment.
Comment 1.0 - Environmental Performance Standards
A three paragraph evaluation of the risks to human health and the environment from the permitted activity is contained in Attachment 1, Section 9.12. This explanation begins: "The risks posed to human health by operating the MMD-1 system will be minimal." This statement is based in part on the fact the "...releases occurring outside of engineering controls will be cleaned up and managed in accordance with the contingency plan." (p. 9-25)
These reassuring statements contradict the permit concerning MMD-1 safety. The proposed permit states any release of chemical agent outside of engineering controls "...is considered a potential endangerment to human health and the environment" (Condition III.C.12). The permit also states that even treated agent should be considered acutely toxic (Condition III. D.1)
It is therefore reasonable to conclude that the information contained in Section 9.12 does not meet the requirements of R315-8-16 [40 CFR 264.601 (a)(8)-(9), (b)(10)-(11), and (c)(6)-(7)]. Specifically, Section 9.12 should be revised to accurately represent the "...potential for health risks caused by human exposure to waste constituents." Noticeably absent from the 500 plus pages of information supporting the draft permit is a discussion of the short and long-term symptoms of chemical agent exposure. An estimate of the distance to a toxic end point during a worst case release scenario from the MMD-1 should be prepared and included in Section 9.12.
Comment 2.0 - Air, Liquid, and Solid Waste Treatment Standards
2.1 Air Treatment Standard Clarification - Air sampling is to be used to determine if treatment standards are reached for air, carbon breakthrough, equipment decontamination, and solid waste decontamination. It is the intent of the draft permit to establish this standard as 0.2 times the TWA values shown in Table 10-1 (condition III.C.11). Discrepancies exist between this treatment standard and permit documents which often show the standard as 1.0 TWA.
2.2 Air Treatment Standard for Mustard Gas (HD) - The air treatment standard for mustard gas is based on the limits of analytical detection and may not be health protective. A footnote appears throughout the report to indicate that no safe level of mustard exposure is known. Section 9.12 of the permit should be revised to address the health effects that mustard gas may cause at any level of exposure. The Closure Plan should be revised to reflect the fact that "clean closure" of MMD-1 is not possible because no safe level of worker exposure to mustard gas is known at this time.
2.3 Liquid Treatment Standard Clarification - The liquid treatment standard of 50 mg/L is based on the limits of analytical methods to detect chemical warfare agents in a solution of monoethanolamine (MEA). The Army will be required to document their attempts to reach the 1.0 mg/L, which the State of Utah has identified as a desirable level of treatment (condition III.C.5). One obvious way to reduce analytical detection limits is to reduce the amount of MEA used to treat agent in the MMD-1. Do lower ratios of MEA to chemical agent increase the ability to verify treatment effectiveness through lower detection limits?
2.4 Treatment Standard for EA 2192 - Measurable levels of a nerve toxic compound, sodium S-(2-diisoprpoylaminoethyl)methylphosphonothioate (EA2192, sodium salt) have been found when VX reacts in the presence of water (p. 4B-37). A treatment standard should be proposed for EA2192 and a reliable and accurate method of analysis developed and included in the Waste Analysis Plan.
Comment 3.0 - Closure Performance Standard
3.1 By-products of Chlorine Bleach Rinse - A five percent solution of chlorine bleach is to be used to decontaminate the equipment between each chemical warfare agent campaign, and at closure (p. 11-7). The by -products of a reaction between chlorine bleach and mustard, sarin, phosgene and GB are not included in the permit information. This information should be incorporated into the Waste Analysis Plan, Health and Safety Plan, Environmental Monitoring Plan, and plans for off-site waste management.
3.2 Closure Standard - The proposed "clean closure" strategy appears to be no different then the standard used when the equipment will be maintained by operators utilizing fully encapsulating suits and self-contained breathing apparatus. This standard based on 20 percent of the allowable occupational exposure is not a "clean closure" condition. Significant hazards would remain following closure and post-closure management requirements are appropriate for the process trailer.
Post-closure care requirements should be incorporated into the permit. These conditions should include secure storage and health and safety requirements. There is concern that maintenance without the use of self-contained breathing apparatus and encapsulating suit may expose individuals to a life threatening situation. During post-closure MMD-1 maintenance should also be restricted to secondary containment buildings with air pollution control systems similar to Building 3445.
3.3 Building 3445 Decontamination - Building 3445 has been specifically excluded from the Subpart X Permit, with the exception of the carbon adsorption system, and the west chamber 90-day storage area. The closure plan provides no details for closure of the Building 3445 carbon adsorption system or Building 3445 except to say they will be decontaminated according to standard operating procedures (p. 11-10). Building 3445 is an integral part of the MMD-1 test plan and should be incorporated into the permit. The standard operating procedures for decontamination of Building 3445 should be incorporated into the permit closure plan.
Comment 4.0 - Management of Derived-Wastes
4.1 - Incineration Mandate - The draft permit contains a provision mandating incineration of all wastes produced by the MMD-1. There is concern that the incineration mandate is not consistent with RCRA regulations which provide waste management options for spent carbon from closed vent systems. Section 40 CFR 264.1033(n) requires owners and operators who use carbon adsorption systems to manage carbon removed from the control device in one of three manners. The carbon can be regenerated or reactivated in a thermal unit, the carbon can be incinerated in a hazardous waste incinerator, or the carbon can be burned in a boiler or industrial furnace. The regulations indicate that it is the discretion of the generator, and not the regulator, to select from among the waste management options.
4.2 - Incineration Mandate Modification - A January 15, 1999 letter from the Secretary of the Army to the Non-Stockpile Citizen Coalition indicates that the army will seek a Class II modification from the State of Utah for the Rapid Response System (RRS) to remove the incineration mandate from that permit. This would allow the Army to identify an environmentally responsible form of disposal for residual wastes from the RRS. The Army further indicated during a telephone conference call with the Non-Stockpile Chemical Weapons Citizens Coalition on January 26, 1999, that it will seek a similar modification for the MMD-1 draft permit.
4.3 - Alternatives to Incineration - Alternative treatment methods for liquid waste generated by the MMD-1 should be evaluated. One alternative is the safe storage of liquid wastes produced by the MMD-1 at Dugway Proving Ground until a non-incineration based treatment technology is developed to complete detoxification. A mandate for on-site storage should be included in the permit. This mandate should require annual evaluations of developing technologies for detoxification of MMD-1 stored wastes until such time as a technology is identified and permitted for stored waste.
4.4 - Storage as a form of Treatment - The kinetics for alkaline hydrolysis reactions are favorable at low temperatures but they do not allow for complete reaction of agent in a four hour period. Even at 10:1 ratios of MEA to agent the first-order kinetics do not achieve the desired treatment goal of 1 mg/L of agent.
For example , Table 4B-3, Reaction of HD with MEA/Water, Reaction Profile, the HD concentration is 50 ppm at 117 minutes, and 17 ppm at 240 minutes. In one case, the HD was found at 0.06 and 0.32 mg/L after 24 hour storage in MEA/water solution.
For example, in studies of GB reaction with MEA/water, 12 to 94 mg/L of GB remained after three hours (p. 4B-20). According to Table 4B-7, the neutralent from GB/MEA/Water reactions contained GB concentration of 0.352 mg/L after 23 days and
<0.025 mg/L after 46 days. Storage should be considered as a form of treatment as short-term storage of liquid wastes will increase the effectiveness of the MMD-1. 4.5 - Storage of MMD-1 Wastes at unpermitted Facilities
- Three 90-day storage areas are not included within the scope of the draft permit despite the fact that these storage areas are incidental to treatment. A fourth storage area, the munitions storage magazine, is included within the permit. When hazardous waste is removed from a permitted facility such as the MMD-1 to a non-permitted facility such as a 90-day storage area the "cradle to grave" intent of RCRA is not achieved.
4.6 - Processed Munitions Casings - If chemical agent remains in a processed munitions body above air monitoring levels, the munitions body will be stored in neutralent solution in Igloo G at Dugway (Condition III.C.6.a). The permit should provide details on the long-term management plans for this waste.
Comment 5.0 Process Design Issues
5.1 Over-pressurization of the process trailer. No evaluation of the potential for over-pressurization of the process trailer and subsequent containment failure is provided in the permit support information to determine compliance with 40 CFR 264.601(c)(2). If the largest pressure vessel fails, over-pressurization of the process trailer's secondary containment may occur. What margin of safety was used to design the secondary containment volume to prevent over-pressurization during a worst-case release scenario?
5.2 Size of Waste Gas Knock-out Drum. Table 5-8 lists the waste gas knock-out drum volume as 170 gallons. The capacity of this vessel is important because it represents the capacity of the primary system used to treat gases generated during chemical warfare demilitarization in the MMD-1. The dimensions of the Waste Gas Knock-out drum is described as a 1.667-foot diameter by 4.625-foot length cylinder (p. 5-66). These dimensions provide a capacity of 75 gallons. Please clarify this information since this is the largest pressure vessel, and would provide a basis for design of secondary air containment systems.
5.3 Size of Gas Reactor Chiller. The permit indicates that the waste gas chiller was sized to remove the adiabatic heat of compression from the 7.5 horse-power vacuum pump (Table 5-8). The waste gas reactor chiller is sized at 5,500 BTU/hour which represents a heat transfer equivalent to 2.15 horse-power. This apparent imbalance between the heat added by the vacuum pump, and removed by the gas chiller indicates that heat will accumulate in the system until a pressure relief valve is activated.
5.4 Size of Gas Reactor The exhaustion of carbon in the waste gas reactor is to be indicated by a fast increase in temperatures above 200 degrees Fahrenheit. There is concern that no method exists to remove this heat from the gas reactor loop except through a pressure relief valve.
5.5 Adequacy of Air Treatment Unit The use of activated carbon to remove low levels of air contaminants is highly effective when contaminants are present at constant concentrations. These predictable conditions allow carbon breakthrough times to be accurately estimated because the mass loading and carbon temperature are constant. In units such as the MMD-1 where emissions to the carbon units will be highly variable carbon adsorption is more problematic. Short-term discharges of high concentrations of organics could cause a temperature rise in the carbon adsorber which could render it ineffective. Long-term discharges of "agent-free" air may lead to desorption of chemical agent at low concentrations into the air. The air treatment unit should be modified to account for a sudden accidental discharge of agent into a carbon treatment system.
5.6 Ammonia Monitoring Ammonia is to be removed from the process trailer waste gas by an adsorption bed immediately prior to the whelterized carbon. Ammonia would render this carbon ineffective by dissolving the copper and other metals used as a coating to increase phosgene adsorption. Breakthrough monitoring of the ammonia bed should be included in the permit requirements to ensure that the removal of chemical agent by the whelterized carbon remains effective.
Comment 6.0 - Phosgene Treatment
6.1 Land Disposal Restrictions Incineration Mandate - Land disposal restrictions set a technology based treatment standard for phosgene wastes, requiring that these wastes be incinerated. The MMD-1 therefore requires a variance to this land disposal restriction treatment requirement to use alkaline hydrolysis as an alternative treatment method.
6.2 Phosgene Compatibility with Process Systems - Phosgene is the most difficult treatment campaign for the MMD-1. The high volatility, high heat of reaction and significant air treatment requirements extend the limits of the MMD-1. During treatment of phosgene, the process will be required to be stopped so the gas reactor can be replenished with activated carbon. What is the MMD-1 design capacity for a phosgene batch? Is it 22.6 liters as listed in Table 8-5 (condition III.C.5), or 17.90 gallons as listed in Table 4-1 (condition III.B.1). What is the margin of safety (available capacity) when treating a 71.1 pound and 213.1 pound batch of phosgene?
6.3 Discretionary Treatment Most of the phosgene to be treated during the MMD-1 test is to be purchased from a chemical manufacturer. Unlike the other chemical agents for which there is a need and desire for their ultimate elimination, only a small amount of phosgene is present in recovered non-stockpile materials. Phosgene represents a small fraction of existing non-stockpile inventory, and there is a low potential that phosgene will be discovered at other non-stockpile sites in the future. It is impractical to spend 60 percent of the MMD-1 test effort on this chemical warfare agent. The State of Utah should exercise discretion in limiting the size and scope of tests with phosgene gas.
Comment 7.0 - Waste Analysis Plan
7.1 Analysis Methods for Chemical Agent Detection Chemical analysis of liquid wastes is the primary method used to determine if the MMD-1 is being effectively operated. At the time of the permit application, no reliable analytical methods for chemical agent detection in MEA have been developed. A principle complication is the reaction of quality control spikes with MEA in the sample matrix. Permit conditions require the MMD-1 laboratory demonstrate the reliability of the analytical method before operations begin (p. 10-3). A similar requirement for the State of Utah certified analytical laboratory that will be used for waste characterization tests should be included in the permit. This requirement would insure test are replicable between MMD-1 lab equipment and the outside lab equipment.
7.2 Remote Sampling In addition to challenges in developing a reliable method of analysis, the remote sample collection and sample conditioning system are also prone to reliability problems. The MMD-1 laboratory should be relocated on Figure 1-1 to a location that would minimize sample line lengths to increase sample system reliability.
|
