Evaluation of the Army's Draft Assessment Criteria to Aid in the Selection of Alternative Technologies for Chemical Demilitarization (1995)

The primary guiding principle for previous Stockpile Committee recommendations has been minimization of the total cumulative risk ¹ (NRC, 1994b). Minimization of total risk continues to be the primary guiding principle. A decision on whether to proceed to a pilot demonstration of an alternative technology requires a direct comparison between the proposed alternative technology and the baseline system. The relative advantages and disadvantages of piloting an alternative technology should be clarified by comparison.

The comparison process being used by the Army is a multistep process. The first step is determining the primary factors and subfactors that form the basis for comparison. The second step is establishing criteria for each factor, against which baseline system performance and alternative technology process development results can be compared. While factors identify subjects that are central for the comparison, the Army's criteria are formulated as questions designed to solicit numerical, pass-fail, or yes-no responses that form the basis for the ultimate decision.

This chapter discusses the factors that the NRC Stockpile Committee considers central to the comparison between the alternative technologies and the baseline system. The factors included in this list were developed based on review of the Army's April 26, 1995, draft document, Assessment Criteria to Aid in the Selection of Alternative Technologies for Chemical Demilitarization and on input solicited from the communities and interested parties at the stockpile locations. In November 1994, letters were sent to Citizens Advisory Commissions, local regulatory authorities, governors, and other interested parties at all the chemical stockpile locations. The letters solicited information on critical factors to be considered by the Army when evaluating the alternative technologies. Copies of the solicitation letter and its distribution list are included as Appendix B of this report. Approximately 40 responses were received. The requirements of Public Law 102-484 ( Appendix A ) and the Chemical Weapons Convention (United Nations, 1993) were also considered when developing the critical factors.

Using the guiding principle of minimization of total risk, coupled with the requirements of Public Law 102-484, the Chemical Weapons Convention treaty, and input from host communities surrounding the stockpile sites, the Stockpile Committee determined four primary factors, each of which corresponds to a question that must be answered when evaluating an alternative technology. These factors and their corresponding questions are:

1. Process Efficacy. Does the alternative agent-destruction process, when integrated with other necessary destruction system components, effectively meet agent-destruction requirements?

2. Process Safety. Is the alternative technology at least as safe as the baseline system? The criterion of “at least as safe” adopted by the Stockpile Committee is minimization of total risk to the public and to the environment (NRC, 1994b).

3. Schedule. What are the impacts of implementation of an alternative technology on the schedule for stockpile destruction?

4. Cost. Is the alternative process cost effective?

Answering each of these questions requires consideration of both the system requirements for stockpile destruction technology and the uncertainties associated with implementation of either the baseline system or an alternative technology. Several of the questions are interrelated. For example, schedule requirements directly impact programmatic costs. In addition, given the uncertainties associated with various program components, answers to these questions may be based on the personal perspectives of the individual(s) carrying out an evaluation and, therefore, may be subjective. Results from the alternatives RDT&E program and other program components, as well as information from external sources (e.g., permitting agencies, etc.), will be used to answer these questions.

Each of the four primary factors has several subfactors that may be interdependent. A negative judgment for a specific subfactor need not be considered a negative outcome for its respective primary factor. The relative significance of each factor and subfactor is determined based on the judgment of the ultimate decision-making organization. The subfactors and their interdependencies are discussed below.

Determination of process efficacy requires considering not only whether the process can destroy the agent of interest, but also whether the process can be controlled, function reliably, and meet applicable regulatory and treaty requirements. The subfactors for process efficacy are:

• detoxification of agent;

• achieving treaty requirements;

• achieving environmental and other regulatory requirements;

• management of process residuals;

• process stability, reliability, and robustness;

• process monitoring;

• natural resource requirements;

• scale-up requirements; and

• applicability to other wastes.

Destruction of the agent of interest (e.g., mustard or VX) requires that the detoxification reaction proceed to residual concentrations below a defined limit and that the products be reduced in toxicity to the extent necessary for management of the process residuals ² through aqueous discharges to waste-water treatment facilities, solid waste disposal at hazardous-waste landfills, or release as atmospheric emissions. Maximum allowable residual concentrations and toxicity limits must be defined for aqueous, solid, slurry, and gaseous streams. The agent-destruction process selected also must facilitate decontamination and management of agent bulk storage containers, as well as nonprocess wastes as part of the overall disposal system. These requirements are specified by the environmental permits required for the agent-destruction facility and the associated waste-management facilities.

The requirements of the 1993 Chemical Weapons Convention necessitate destruction of the primary agent and further reaction or destruction so that none of the end products is a “scheduled” precursor compound, which can readily be reconverted to the primary agent. Acceptable residual concentrations of agent specified under the treaty negotiations are likely to be less stringent than applicable environmental permit requirements. The process must also fulfill applicable verification requirements of the convention.

The agent-destruction process that is implemented must comply with applicable state and federal regulatory requirements. Key components of regulatory requirements include specifications for acceptable process discharges (i.e., aqueous, air, and solid or hazardous waste) and waste-management practices. Other regulatory compliance issues will include workplace requirements (e.g., those set by the Occupational Safety and Health Administration) and management of non-process wastes (e.g., decontamination fluids, personal protection suits). The Army should compile a listing of

applicable regulatory requirements, and it should explain the differences in requirements for an alternative technology compared to the baseline system.

Disposal of process residuals is a critical aspect of any destruction process. Alternative technologies will have different process residuals from the baseline system. Differences include the physical state, composition, and quantity of the residues, but all process residuals, including any remaining agent and reactants, must ultimately be disposed of. For the baseline system the major process feeds are agent, air, fuel, and scrubber solution (water and caustic). The major process residuals are atmospheric emissions (i.e., combustion gases and water vapor), slag or ash, dried salts from spent scrubber solutions, and spent activated carbon (from air cleaning). For an alternative neutralization-based technology, the primary process feeds will be agent, water, and caustic. Use of an aerobic biodegradation process after neutralization will also require air, biomass, and, potentially, nutrients. The major process residuals are carbon dioxide, water, salts, organic-reaction products, and spent activated carbon (from air cleaning). Use of a biodegradation process will also add biomass as a process effluent.

One major question for management of aqueous residuals from an alternative technology will be whether to generate an aqueous discharge to a waste-water treatment facility or to evaporate water and discharge it as an atmospheric emission as in the baseline system. Some form of water release will be required even with aggressive water recycling. The extent of water recycling will be important from the viewpoint of cost effectiveness. A second major question for management of process residuals is at what point wastes can be transferred to private sector facilities for subsequent management. This question requires consideration of appropriate waste-management options (e.g., aqueous discharge, solidification/stabilization, land filling, thermal destruction) for individual waste streams and available private sector capacity. Composition and management practices for each of the residual streams from both alternative and baseline systems should be compared. An appropriate basis for this comparison is the overall process mass balances and component mass balances for major chemical species (e.g., water, nitrogen, sulfur, chlorine, and carbon species).

The agent feed to the destruction process will be variable as a result of the differences in agent purity, impurity composition, and aging processes that result from varied production and storage operations. It has been reported that some agents, particularly mustard, become gelled in storage containers and, consequently, may be difficult to remove from the containers and to feed to the destruction process (NRC, 1994b).

The selected process must function effectively and reliably in spite of changes in the process feed. Operating conditions that result in inherent process instabilities (e.g., poor performance, process failure, or catastrophic failure) must also be examined. These include extreme operating conditions (e.g., high pressures or temperatures), phase changes (e.g., formation of multiphase process streams), and corrosive process reactants, residuals, or operating conditions. Development of process-control strategies and process flexibility must be defined so that the process can be controlled effectively even if an upset such as power failure or loss of agitation occurs. The selected process must also provide for the decontamination and management of storage containers and other contaminated metal parts.

Process stability, reliability, and robustness can be enhanced by using technology proven in commercial service. Standard equipment and materials used in the chemical and petroleum processing industries are developed with reliability as a goal. Adaptation of processes from these industries can build on experience with proven process stability and robustness (i.e., ease of control in spite of fluctuations in feed stocks and reaction rates).

Implementation of an alternative technology will require development of techniques to monitor concentrations of agent and products from agent neutralization in liquid and slurry or solid process streams. Sampling procedures, response times, and required detection limits will have to be defined. Because a neutralization-based process is unlikely to produce large volumes of hot, gaseous effluents, the monitoring requirements may be quite different from those for the baseline system.

Natural resource consumption (e.g., energy and water) should be considered as one aspect of technology selection at locations where specific natural resources, such as water, are limited.

Implementation of an alternative technology will require process demonstration using near-full-scale equipment prior to full implementation. Equipment required for process demonstration may be different for mustard than for VX. Selection of the demonstration location should consider availability of agent and permit acceptance. Two primary options for process demonstration are to test both the mustard and VX processes at the Chemical Agent Munitions Disposal System at Tooele Army Depot, Utah or, alternatively, to test a mustard process at Aberdeen and a VX process at Newport.

Use of a destruction technology that is broadly applicable to common industrial wastes is a concern to local host communities because of the potential for importing additional wastes once stockpile destruction is completed. Thus, selection of a technology that would result in the construction of a versatile waste-destruction facility may undermine the Army's credibility with respect to decommissioning the destruction facility at the conclusion of the stockpile destruction.

Process safety encompasses the perspectives of worker safety, community health risks and environmental risks, and storage risks. Evaluation of process safety should include assessment of in-plant risks and hazards, as well as risks to the neighboring community and environmental resources. Risks associated with both agent release and release of nonagent process residuals should be evaluated and provided in a side-by-side comparison of the alternative technology to the baseline system. The following paragraphs discuss the subfactors in this category.

Plant safety and health risks should focus on the nature and magnitude of hazards within the processing facility. This evaluation should include preliminary assessment of the following:

• risk of catastrophic failure and agent release;

• risk of exposing plant workers to agent;

• risk of plant worker exposure to other hazardous chemicals; and conditions.

In the Chemical Stockpile Disposal Program, site-specific, accident quantitative risk assessments are used to evaluate these risks. The time available before the 1996 Defense Acquisition Board decision is inadequate for complete risk assessments, but preliminary risk comparisons between processes should be possible.

This subfactor should focus on the nature and magnitude of risks to the communities and natural resources adjacent to the processing facility. Risks from acute exposure to agent or accidental release of process residuals, as well as latent health effects from low-level exposure to process emissions and discharges, should be considered. Potential for impacts on natural resources should include site-specific considerations. For example, many individuals living near Aberdeen Proving Ground, Maryland, expressed concern about negative impacts on the Chesapeake Bay, with consequential adverse effects on fishing and on recreational activities. Evaluation of exposure risks from agent destruction should be based on processing throughout the projected facility operating period and consideration of the limited scale and finite time of stockpile destruction operations at each site. The risk evaluation should include comparison of:

• risk of agent release and exposure;

• risk from latent health effects of exposure to nonagent releases (carcinogenic and noncarcinogenic endpoints; this information is included in comparisons of site-specific health effects);

• risks from process residuals management;

• risk of exposure to other hazardous materials, including possible releases during transportation;

• total environmental burdens (process mass and energy balances);

• potential impact on natural resources (e.g., agriculture, water bodies), especially from aqueous discharges, atmospheric emissions, or solid-waste management; and

• special considerations that may affect emergency preparedness or emergency response.

Storage risks have been cited as one motivation for prompt destruction of the stockpile. Risks associated with storage also have been the focus of contentious debate at several of the stockpile locations. A comparison of storage risks associated with an alternative technology with those of the baseline system is integrally related to the schedule for each option (see schedule factor, below) and to any options that may be exercised for storage risk reduction at individual sites. Evaluation of storage risks should include comparison of the range of risk based on the schedule uncertainties for both the baseline system and alternative technology options. Storage risks should consider both acute effects (human health and environmental) associated with a high-level release and latent effects to human health from low-level releases.

Impact of technology selection on the program schedule is difficult to project. Planned schedules have repeatedly been subject to delays resulting from both internal program issues and external events. However, it is also possible that development of significantly simpler destruction technologies may allow schedule acceleration. The time of treaty ratification also will influence the disposal schedule. Public Law 102-484 requires that agent destruction be completed by December 31, 2004. This date was specified by the law as one that would comply with treaty obligations, based on anticipated rapid ratification by the United States of the Chemical Weapons Convention treaty. The treaty requires a schedule for completion of agent destruction that is based on the date the treaty is put into effect either by ratification by the United States or by a sufficient number of other countries. Currently, the treaty has not been put into effect by either route; therefore, the destruction schedule required to fulfill this obligation remains indeterminate. The treaty date for destruction of agent also does not necessarily require completion of all waste-disposal actions or facility removal. It is reasonable to consider that Congress may modify Public Law 102-484 to allow extension of the destruction schedule within treaty requirements. The schedule is a significant factor, but it is not a “make or break” determinant.

Estimated schedules with time ranges for the baseline system and for the alternative technologies at each potential site are needed. Key steps that may result in schedule delays or offer the potential to accelerate the schedule should be identified. Schedules for completion of stockpile destruction using the baseline system and using alternative technologies should be compared on the basis of currently projected schedules and should include both optimistic scheduling and reasonable worst-case delays. Potential schedule delays due to technology development for alternatives, and to permitting delays possible in communities opposing implementation of the chosen system should be taken into account. The results of the schedule estimates will directly affect site-specific storage risk estimates and cost estimates.

Five factors should be considered in evaluating process costs:

• extended stockpile storage interval;

• process development;

• process implementation and operation;

• plant disassembly; and

• discounting.

The cost associated with maintaining the stockpile until destruction is completed should include consideration of delays that may occur owing to the technology selected. For example, implementation of the baseline system may result in prolonged storage and

associated costs because of permitting delays in a hostile community. Implementation of an alternative technology may result in increased storage intervals because of development requirements. Comparison of costs for the baseline system and alternative technologies should be site-specific and should include estimates based on currently projected schedules and reasonable worst-case delays for each technology.

The costs associated with process development for the baseline system and alternative technologies should be compared. This comparison should include future research, development, and demonstration costs for both processes, all of which would be specific to the stockpile location.

The comparison of the baseline system and alternative technologies should include costs associated with full-scale facility construction and operation for complete destruction of the stockpile at the storage location.

The costs of complete decommissioning and disassembly of the destruction facility for the baseline system and alternative technologies should be compared. The comparison should include restoration of the plant site and stockpile storage location for the intended future use. The nature and extent of facility decontamination and dismantlement required will significantly impact decommissioning costs. Since these actions are not treaty requirements, the schedule for facility removal should not be based solely on treaty deadlines.

It is a standard business practice to discount cost streams to their present value when comparing alternative investment opportunities. The cost of government activities, such as destruction of the chemical weapons stockpile, should be discounted at a rate appropriate to account for the opportunity cost of using tax dollars for the proposed activity rather than for other purposes. Discounting is particularly important when the time pattern of costs for one alternative action differs from the time pattern of costs for another (which might be the baseline system in this case). For comparison purposes, all costs, including those for any continued storage, process development, process implementation and operation, and plant disassembly, should be discounted for both the baseline system and any alternative technology.