In 1993, the United States signed the Chemical Weapons Convention (CWC), an international treaty outlawing the production, stockpiling, and use of chemical weapons.1 The chemical weapons stockpiles at five of the U.S. chemical weapons storage sites have now been destroyed. At those sites, the munitions were robotically opened and the chemical agent was removed, collected, and incinerated.2 Chemical agent stored in bulk containers at two other chemical storage sites have been destroyed by hydrolyzing the agent with either hot water or caustic.3
The remaining two sites with chemical weapons stockpiles are the Pueblo Chemical Depot (PCD) near Pueblo, Colorado, and the Blue Grass Army Depot (BGAD) near Richmond, Kentucky. The choice of technology to destroy chemical weapons at these sites emerged out of a complex history. As discussed in Chapter 3, in the event that currently planned treatment processes underperform or fail, this history will influence the way that stakeholders approach any consideration of alternatives. The early history of the chemical weapons destruction program (described in more detail in Appendix A) was characterized by significant controversy and social distrust about the Army’s preference for incineration to dispose of the stockpile. Much of this controversy had its origins in the questions raised by residents living nearby PCD and BGAD.
A dramatic culmination in the dynamic of growing social distrust and Army insistence on incineration occurred in 1997. A number of factors led to this shift, including (1) mounting pressure on the Army from Congress and (2) in local communities in opposition to incineration to achieve chemical demilitarization (Durant, 2007; Futrell and Futrell, 2012; GAO, 1995a and 1995b). In 1996, Congress enacted laws that froze funds for construction of baseline incineration facilities at PCD and BGAD.4 These laws further directed the Army to identify and evaluate at least two alternative technologies for the chemical agent destruction. The Assembled Chemical Weapons Assessment (ACWA) program was established to demonstrate other means of destroying the chemical agent.
Congress also required a more robust effort at public involvement by ACWA, which led to the initiation of the 5-year-long dialogue on assembled chemical weapons assessment in 1997 (Goldberg, 2003; Futrell, 2003; and Keystone Policy Center, 2004), referred to from here on as the Dialogue Group. Input was sought from a broad range of interested and affected parties about technical and social criteria for comparing alternative technologies, assessment of the alternative technologies using these criteria, and identification of sites appropriate for the implementation of the alternative technologies. The Dialogue Group included representatives of local citizens; federal, state, and local regulators; the Army; and the National Research Council (NRC). The NRC also conducted reviews of candidate nonincineration technologies for treatment of the chemical stockpiles in Colorado and Kentucky.
Published accounts of the Dialogue Group highlight its importance for rallying public support, enhancing ACWA responsiveness to public concerns and preferences, building the capacity of local stakeholders to participate in highly technical discussions, and rebuilding trust (Goldberg, 2003; Futrell, 2003; and Futrell and Futrell, 2012). After the technologies had been selected, the Dialogue Group was disbanded and Citizens’ Advisory Commissions (CACs) were established in Colorado and Kentucky. Both CACs include members of the Dialogue Group. The ACWA program has resulted in the selection of alternative technologies at the two sites and is now the Program Executive Office (PEO)
1 Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and on their Destruction. The treaty entered into force in 1997.
2 These sites were located on Johnston Atoll and in Anniston, Alabama; Pine Bluff, Arkansas; Tooele, Utah, and Umatilla, Oregon.
3 These sites were located at Aberdeen Proving Ground, Maryland, and Newport, Indiana.
4 Public Laws 104-201 and 104-208.
for Assembled Chemical Weapons Alternatives, still known by the acronym ACWA.
This report pertains to the destruction of chemical weapons containing nerve agent currently stored at BGAD, and also of energetic materials contained within these munitions. The nerve agent stockpile at BGAD consists of GB (sarin) and VX, in varying munition configurations, including 115-mm M55 rockets, 155-mm projectiles, and 8-in. projectiles. There is also an inventory of chemical munitions containing mustard agent at BGAD. The mustard stockpile will be disposed of using an explosive destruction technology in a separate facility at BGAD. The energetics in the M55 rocket consist of a burster (Comp. B: 60 percent RDX and 40 percent TNT) and M28 propellant (primarily 60 percent nitrocellulose and 23.8 percent nitroglycerine). The nerve-agent-containing projectiles at BGAD do not contain energetics.5
The facility being constructed to destroy the nerve agent munitions at BGAD is called the Blue Grass Chemical Agent Destruction Pilot Plant (BGCAPP). At the writing of this report, main plant construction is over 90 percent complete. The process of preparing the plant and its employees for munitions destruction is called systemization. According to ACWA, “Systemization is the testing of all process components, subsystems, and systems, and the demonstration the plant, procedures, and personnel are ready for toxic operations.” (BPBGT, 2012, p. 20)
Some parts of the plant have been in various phases of systemization while systemization for other parts of the plant has yet to be started. At the writing of this report, main plant systemization was approximately 30 percent complete. Processing of the nerve agent munitions through the plant is scheduled to begin in the second quarter of FY2018. BGCAPP expects that it will require approximately 3 years to completely destroy the nerve agent in the BGAD stockpile and complete secondary waste processing, with operations scheduled to be completed in the second quarter of FY2021.6
The GB and VX will be destroyed by caustic (sodium hydroxide, NaOH) hydrolysis. Energetic materials contained within the chemical munitions will also be treated by caustic hydrolysis. The hydrolysis process results in a waste residual referred to as hydrolysate. BGCAPP intends to blend the agent and energetics hydrolysates and then treat the combined hydrolysate in supercritical water oxidation (SCWO) reactors. There will be a GB phase and a VX phase of operations; the GB and VX hydrolysates will be blended with energetics hydrolysate, but GB and VX hydrolysates will not be blended with each other. The nerve agent hydrolysates, blended with energetics hydrolysate, will go through SCWO separately. The SCWO effluent then will be processed through a water recovery system (WRS), the heart of which is a reverse osmosis unit. Recovered water will be recycled to the SCWO process as quench water in lieu of drawing makeup water. The BGCAPP hydrolysis treatment processes are described in Chapter 2 of this report.
While destruction of the nerve agents is conducted under the auspices of the CWC, because the hydrolysate contains Schedule 1 and 2 compounds under the CWC, the SCWO process is also subject to CWC oversight.7 The Schedule 1 and 2 compounds in the hydrolysates are identified in Chapter 2.
The project is to be undertaken in two reports: a first report, on the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP), to be delivered 9 months from start of contract, and a second report, on the Blue Grass Chemical Destruction Agent Pilot Plant (BGCAPP), to be delivered at the end of the 15-month contract period.
The National Research Council will establish an ad hoc committee to consider the following study objectives for the BGCAPP second report:
- Develop criteria for successfully treating the hydrolysate in the designed SCWO and Water Recovery Systems;
- Identify systemization data that should factor into the criteria/decision process;
- If the present treatment is not satisfactory, identify potential modifications that would allow continued onsite processing, e.g., additional buffer storage or shipment of excess hydrolysate beyond onsite capacity;
- Identify the downstream impacts to plant operations if offsite shipment is required;
- Determine if aluminum recovery from energetic hydrolysate prior to shipment offsite will be necessary;
- Determine if separate waste disposal options are required for agent and energetics hydrolysates;
- Identify any regulatory requirements for offsite hydrolysate shipment and treatment options, and evaluate transportation risks that could be expected; and
- Consider stakeholder interests and solicit stakeholder input.
5 J. Barton, BGCAPP chief scientist, Battelle, “BGCAPP Agent and Energetics Treatment Processes,” presentation to the committee on January 27, 2015.
6 R. Goetz, assistant project manager, Parsons, BGCAPP, “Project Overview,” presentation to the committee on January 27, 2015.
7 Under the Chemical Weapons Convention, Annex on Chemicals, Schedule 1 chemicals are those that were developed, produced, stockpiled, or used as a chemical weapon, or are chemicals that would pose a high risk to the object and purpose of the Convention by virtue of their high potential for use as a chemical weapon. Schedule 2 are those chemicals that pose a significant risk to the object and purpose of the Convention because they possess such lethal or incapacitating toxicity and other properties that could enable them to be used as a chemical weapon, or may be used as a precursor in one of the chemical reactions at the final stage of formation of or is used in the production of a Schedule 1 or a Schedule 2 chemical, https://www.opcw.org/chemical-weapons-convention/annexes/annex-on-chemicals/.
This committee has done its work in two distinct parts as broken down by the statement of task. The first part addressed the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP), while this report, the second part, addresses the hydrolysate treatment processes at BGCAPP. The composition of the committee changed slightly between the two parts. The committee that completed the PCAPP report is referred to as the PCAPP hydrolysate committee. The committee that completed the work in this report is referred to as the BGCAPP hydrolysate committee (or, simply, the committee).
The complete statement of task, above, required assessment of hydrolysate treatment processes at both PCAPP and BGCAPP. Construction of the PCAPP plant was anticipated to have been completed by the time the BGCAPP hydrolysate committee was to begin its deliberations and was scheduled to begin agent operations in September 2015.8 It was therefore determined that it would be most beneficial for the NRC to first address PCAPP. Hence, the report prepared by the PCAPP hydrolysate committee was nearing completion and publication as the effort for BGCAPP was beginning.
The first BGCAPP hydrolysate committee meeting took place in January 2015 and the PCAPP report was released in March 2015. While there are a number of important differences between PCAPP and BGCAPP, including the agents being destroyed and the treatment methods, the main thrusts of both reports were the same:
- Establish criteria for successfully treating the hydrolysate,
- Identify systemization data that may factor into the criteria and the decision process,
- Suggest potential modifications that would allow continued onsite processing of the hydrolysate should the hydrolysate treatment systems develop issues, and
- Evaluate the potential for offsite transport of hydrolysate should the hydrolysate treatment processes underperform or fail.
This report follows the underlying structure and concepts used in the PCAPP report. The committee determined that having the opportunity to consider important concerns expressed by PCAPP stakeholders regarding the PCAPP report would in fact be beneficial in writing the BGCAPP report. Therefore, the committee determined, with input from the NRC, that feedback from the PCAPP report would be considered in writing the BGCAPP report.
As indicated in the NRC’s 2013 report on SCWO, SCWO is considered first-of-a-kind (FOAK) technology (NRC, 2013). SCWO reactors at BGCAPP will subject blended agent and energetics hydrolysate to very high temperatures and pressures, resulting in the destruction of organic materials within the hydrolysate to carbon dioxide, water, and salts. The total organic carbon content of the water that will be released from the SCWO reactors is anticipated to be less than 10 parts per million.9
The chemical agent will be treated to 99.9999 percent destruction in agent concentration before entering the SCWO process.10 The hydrolysis process breaks up large complex molecules into smaller ones, destroying the nerve agents and eliminating their acute toxicity. Thus, the hydrolysate should no longer contain GB or VX. The level of destruction efficiency was first established by the Kentucky legislature and is a regulatory requirement. Details of how the regulatory process impacts decisions at BGCAPP are provided in Chapter 4.
Very little, if any, agent will remain in the hydrolysate. However, because of the high salt content of the hydrolysate, including a number of different salt species, the blended agent and energetics hydrolysate feed presents a challenge to SCWO. Considering the nature of the hydrolysate and issues with corrosion and salt buildup, extensive FOAK testing was conducted prior to the units being shipped to BGCAPP. This FOAK testing was reviewed by an NRC SCWO committee in 2013, and while that committee had many findings and recommendations for BGCAPP, it nevertheless concluded that “SCWO is a mature technology with a strong scientific and engineering base underlying it”11 (NRC, 2013). For example, SCWO solutions have been designed for the treatment of diverse waste materials, water purification, the recovery of precious metals from catalytic materials, and the production heat and power. The NRC 2013 concluded that while extensive testing of these systems has been conducted in a laboratory setting, the process has never been operated at full scale on the chemical feeds it will process at BGCAPP. This committee agrees with that conclusion.
In addition, ACWA had commissioned an earlier NRC report, in 2012, to evaluate the BGCAPP WRS (NRC, 2012). That committee identified a number of issues in its report, but, overall, it had no overarching concerns that the WRS process would not work on BGCAPP SCWO effluent if its recommendations are followed.12 WRS issues are discussed in Chapters 6 and 7.
8 The NRC committee learned at a meeting of the NRC’s Committee on Chemical Demilitarization held on February 18, 2015, in Washington D.C., that initiation of agent destruction at PCAPP has been delayed to the beginning of January 2016.
9 G. Lucier, deputy chief scientist, BGCAPP, D. Linkenheld, SCWO startup supervisor, and L. Austin, waste manager, BGCAPP, “SCWO Process: Cradle to Grave,” presentation to committee on January 28, 2015.
10 401 KAR 34:350.
11 Two of the 2013 SCWO committee members are also members of this BGCAPP hydrolysate committee.
12 Two of the 2012 WRS committee members are also members of this BGCAPP hydrolysate committee.
The committee received 2 days of briefings from ACWA and BGCAPP staff, from public stakeholders, and from Kentucky regulators during the January 2015 meeting at BGCAPP. The committee has since reviewed extensive documentation on the BGCAPP process, including the earlier 2013 and 2012 NRC reports, and additional documentation provided by ACWA and BGCAPP. As discussed in Chapters 6 and 7 of this report, however, there are a number of technical concerns with operation of the SCWO and the WRS. BGCAPP plans to conduct extensive preoperational testing concurrent with facility systemization that will evaluate these concerns. This testing is expected to result in facility changes (e.g., equipment and operating procedures) that will increase the likelihood of success. Technological underperformance or failure of the SCWO and/or WRS could result in BGCAPP not meeting its overall performance criteria, as set forth in Chapter 6, including not only efficient and effective operations but also regulatory and treaty compliance, public expectations, and an established destruction schedule.
Finding 1-1. It is expected that extensive preoperational testing of the SCWO and the WRS will be performed concurrently with systemization to help reduce the uncertainty in expected technological performance. However, considering the first-of-a-kind nature of these technologies and the need for them to properly function in tandem, there is a possibility that technological issues with these systems may prevent BGCAPP from meeting all of its overall performance criteria.
As mentioned above, there are a variety of reasons that the SCWO or WRS systems might not perform satisfactorily. Chapter 7 provides a detailed discussion of the technical factors that could lead to underperformance of these treatment systems. For example, corrosion and salt clogging could lead to frequent repairs that could shut down hydrolysate treatment in the SCWO units. Occurrences like these could also lead to other issues, such as storage of the hydrolysate for unanticipated times, and idle periods for BGCAPP workers. More importantly, however, if these or similar problems occur, the actual munitions disassembly and hydrolysate production may have to be interrupted until the downstream processes are able to catch up. In the event that one or more of the hydrolysate treatment systems are shut down, and as hydrolysate storage nears capacity, destruction of the primary stockpile at BGAD may need to be halted unless there is an alternative means for treating the hydrolysate.
A delay in the disposal of munitions and agent at BGCAPP is extremely undesirable. The nerve agent munitions to be treated at the BGCAPP have been stored at the BGAD for over 50 years, representing a steady-state or even increasing risk profile for the community (due, for example, to the potential for the munitions to spring new leaks); final destruction of these munitions at BGAD will eliminate this risk. Delays in the destruction process will halt this risk reduction and protract the risk that the community faces. If the SCWO or WRS processes do not perform as expected, it would be necessary for any decisions that might delay BGCAPP destruction operations to consider potential impacts beyond the plant, such as the risk associated with storing aging chemical weapons for longer periods of time, public expectations, regulatory and treaty issues, and employee reassignment, furloughs, or layoffs. The committee firmly believes that destruction of the stockpile at BGAD must continue, because it is destruction of the munitions and the agents that will eliminate the primary risk to the community. Hence, the committee believes that it is necessary to establish a backup plan—an alternative to the onsite hydrolysate treatment processes.
Recommendation 1-1. Considering that the SCWO and the WRS may not perform satisfactorily and that this underperformance could result in delays in the destruction of chemical agent at BGCAPP, increasing the primary risk to the community associated with continued munitions storage, BGCAPP should establish a backup plan as an alternative to the onsite hydrolysate treatment processes.
One alternative for treatment of the hydrolysate that might be more quickly implementable than other alternatives would be to ship the hydrolysate and/or SCWO effluent offsite to an existing, prequalified, and fully permitted treatment, storage, and disposal facility. The decision to implement offsite shipment would impact the downstream plant units that were constructed to treat the hydrolysate generated from the agent and energetics hydrolysis processes. Workers associated with the discontinued SCWO or WRS processes might need to be reassigned and retrained, placed on furlough, or laid off, resulting in the loss of specially trained personnel. This is not the desire of ACWA program staff, the BGCAPP contractors, or the local stakeholders.
The choices faced with possible underperformance of the SCWO or WRS are highly complex. All alternatives will entail interconnected considerations about technological capabilities and limitations, in conjunction with ACWA commitments, contractual requirements, public stakeholder opinion, regulatory permitting requirements, schedule delays, and cost.
To study this situation at BGCAPP, ACWA requested that the NRC form an ad hoc committee, the Committee on Review Criteria for Successful Treatment of Hydrolysate at the Pueblo and Blue Grass Chemical Agent Destruction Pilot Plants, to assess the SCWO and WRS processes, develop criteria by which to judge success, consider onsite mitigation strategies should the SCWO or WRS systems underperform, and explore the potential for offsite transport of the hydrolysate.
Note that the statement of task, above, calls on the committee to, among other things, “develop criteria for successfully treating the hydrolysate in the designed SCWO and Water Recovery Systems.” Chapter 6 lays out two categories of criteria:
- Performance requirements. Conditions that must be met under regulatory permits and under CWC treaty obligations, and
- Performance goals. Primarily oriented toward process performance and schedule and should be met to enable satisfactory system performance.
Performance requirements are crucial in achieving successful operation of the BGCAPP. If these requirements cannot be met, and if the time it takes to destroy the actual munitions increases as a result of the degraded performance, risk reduction goals associated with destruction of the stockpile will not be achieved in a timely manner, and consideration of the offsite hydrolysate and/or SCWO effluent option becomes more likely. Performance goals, in comparison, represent goals for the treatment process that should be met to achieve satisfactory performance, but nonattainment of one or more of these goals will typically not result, at least immediately, in consideration of the offsite option.
The committee did not develop new success criteria for the SCWO and WRS, as called for in the statement of task. While criteria are presented in Chapter 6, they were not “developed” by the committee. Rather, these criteria and goals were garnered from a review of regulatory requirements and BGCAPP documentation. The committee believes the criteria presented in Chapter 6 are reasonable for this stage of the project because they are based on prior SCWO testing with both simulants and actual hydrolysates and on SCWO performance modelling. The committee feels strongly that any new criteria should be developed based on preoperational testing, which has yet to be initiated. The committee did not want to lock BGCAPP into new criteria developed without the benefit of preoperational testing and systemization data, and was especially concerned that new criteria it developed could be incorporated into permit documentation. Hence, while Chapter 6 establishes performance requirements and goals, no new criteria were developed by the committee.
In the course of its interactions with public stakeholders, the committee received an inquiry from the Kentucky CAC about alternative technologies and approaches that could be employed onsite should the SCWO or WRS systems underperform or fail altogether. Specifically, as indicated earlier, BGAD intends to use an explosive destruction technology—that is, the Static Detonation Chamber (SDC)—to destroy its entire mustard munition stockpile. The committee was asked following its January 2015 meeting if this technology could be used to dispose of the nerve agent and energetics hydrolysates.13
The statement of task does not call for the committee to evaluate technologies other than SCWO and the WRS. Therefore, the committee did not conduct an evaluation of other possible onsite technologies for the treatment or disposal of hydrolysate.
It is, nevertheless, possible to make some general predictions in connection with adopting the SDC technology to replace SCWO for the onsite treatment of hydrolysate. The SDC is a double-walled detonation chamber designed to destroy munitions containing chemical agent. The heating of the explosives in the munitions and/or the pressure generated from the heated liquid agent contents eventually cause the munitions to rupture and the agent to be destroyed in the >1,000°F operating temperature in the chamber.
Thus, the SDC is designed to treat an entire munition that contains agent rather than a bulk liquid material such as agent-derived hydrolysate. There are a number of other technologies that were considered for treatment of the hydrolysate in the course of selecting SCWO for BGCAPP. However, as indicated previously, the technologies selected for BGCAPP, specifically hydrolysis followed by SCWO, arose out of a complex history involving many stakeholders; many different technologies were evaluated during this process, culminating in the selection of hydrolysis and SCWO. Further, any effort to adopt another technology would necessitate a formal evaluation effort to select the most appropriate technology. Considering, as stated previously, that continued storage of the nerve agent munitions represents an ongoing risk to the community, processes to select an alternative technology, develop designs and equipment, construct and test new facilities, and obtain permits and other regulatory approvals, would significantly prolong this risk. In addition, further delays would contravene CWC treaty obligations.
In the event that the SCWO and WRS underperform or fail altogether, the most likely option that would help reduce the risk to the community in a timely manner and in compliance with regulatory requirements is offsite transport of the hydrolysate. Future consideration of the full range of options, if contemplated by ACWA, CAC, and the Kentucky Chemical Destruction Community Advisory Board, would weigh these issues very carefully. While a comprehensive and detailed assessment of adopting another technology to replace SCWO for the onsite treatment of hydrolysate is beyond the scope of charge to the committee, the discussions of the decision framework (Chapter 6), public involvement (Chapter 3), and
13 E-mail from Douglas Hindman, chair, Kentucky Citizens’ Advisory Commission, to committee on February 3, 2015.
permitting (Chapter 4) would apply to any alternatives that might be considered.
The committee notes also that the safety of hydrolysate treatment operations is a concern. The committee, however, was not tasked to look into safety. It was apparent, from ACWA and BGCAPP presentations during the January 2015 briefings in Kentucky, as well as the committee tour of the BGCAPP facility, that safety was considered to be of very high importance. The committee believes that safety must remain the highest priority through all operations, including, if necessary, offsite transport of hydrolysate.
Finally, the costs of modifying the SCWO and the WRS at BGCAPP to improve their performance, should this be needed, must be taken into consideration by BGCAPP. If additional funding is needed to modify onsite operations or to allow offsite transportation of hydrolysate, significant delays to the BGCAPP operating schedule and increases in its life-cycle costs may result. Hence, these costs must be carefully considered. While the committee recognizes this situation, it was not tasked to conduct cost analyses of various mitigation activities. Consequently, although cost goals are recognized in this report, they are not explicitly evaluated.
Chapter 2 of this report provides information on the composition of the hydrolysates that will be produced by BGCAPP and describes in more detail the processes planned for use in treating them. Chapter 3 discusses the history of public stakeholder involvement regarding the potential for offsite shipment and treatment of hydrolysate and current concerns related to this. Chapter 4 reviews regulatory considerations at the federal, state, and local levels for onsite process modifications and possible offsite shipment and treatment of hydrolysate, and addresses the requirements of the CWC in this regard. Chapter 5 discusses Department of Transportation regulations for the shipment of hazardous material, presents historical data that quantify the risk of hazardous material shipment, and identifies risks associated with the offsite shipment of hydrolysate, including historical experience with the shipment of agent hydrolysates. Chapter 6 establishes criteria for successfully treating the BGCAPP hydrolysates and identifies systemization data that should factor into the criteria and decisions regarding onsite mitigation approaches and possible offsite transport and disposal of the hydrolysate. Chapter 7 discusses underperformance and failure risks and contingency options as well as the downstream impacts of a decision to ship hydrolysate and/or SCWO effluent offsite. Chapter 7 also addresses issues with the potential for aluminum to be present in the hydrolysate due to dissolution of aluminum M55 rocket parts and also whether the combination of energetics and agent hydrolysate might impact a decision to send combined hydrolysate offsite.
There are also four appendixes. Appendix A provides a chronology of events at BGAD to the present, with focus on public involvement. Appendix B contains material documenting committee efforts to encourage public involvement in this study and material received from the CAC/CDCAB. Appendix C provides biographical sketches of committee members. Appendix D identifies the committee meetings and locations.
BPBGT (Bechtel Parsons Blue Grass Team). 2012. Systemization Implementation Plan (SIP), Rev. 0, August 30. Richmond, Ky.: Blue Grass Chemical Agent Destruction Pilot Plant Program Office.
Durant, R.F. 2007. The Greening of the U.S. Military. Washington, D.C.: Georgetown University Press.
Futrell, R., and D. Futrell. 2012. Expertise and alliances: How Kentuckians transformed the US Chemical Weapons Disposal Program. S. McSpirit, L. Faltraco, and C. Bailey, eds., Chapter 9 in Confronting Ecological Crisis in Appalachia and the South, pp. 171-193. Lexington, Ky.: University Press of Kentucky.
Futrell, R. 2003. Technical adversarialism and participatory collaboration in the U.S. chemical weapons disposal program. Science, Technology, & Human Values 28(4): 451-482.
GAO (General Accounting Office). 1995a. Chemical Stockpile Disposal Program Review. NSIAD-95-66R. Washington, D.C.: General Accounting Office.
GAO. 1995b. Chemical Weapons Disposal: Issues Related to DOD’s Management. T/NSIAD-95-18. Washington, D.C.: General Accounting Office.
Goldberg, M. 2003. Strengthening the link between project planning and environmental impact assessment: The assembled chemical weapons assessment dialogue process. Environmental Practice 5(4): 313-320.
Keystone Policy Center. 2004. ACWA Dialogue Close-out Report (draft), https://www.keystone.org/.
NRC (National Research Council). 2012. Letter Report: The Blue Grass Chemical Agent Destruction Pilot Plant’s Water Recovery System. Washington, D.C.: The National Academies Press.
NRC. 2013. Assessment of Supercritical Water Oxidation System Testing for the Blue Grass Chemical Agent Destruction Pilot Plant. Washington, D.C.: The National Academies Press.