Review of International Technologies for Destruction of Recovered Chemical Warfare Materiel (2006)

Based on its information gathering, the committee assigned Tier 2 status to those technologies that did not at this time warrant the more extensive evaluation given to Tier 1 technologies in Chapters 4 and 5. As described in Chapter 1, the Tier 2 technologies are of two basic types. The first comprises those technologies that may eventually prove to be applicable to Non-Stockpile Chemical Materiel Program (NSCMP) projects but that are still at an early stage of development for such applications. The second type includes those technologies that were tried out in operations for destroying recovered CWM but have so far not proven to be satisfactory for various reasons. Both types are presented below in alphabetical order, first the munitions processing technologies and then the agent-only processing technologies. These technologies (for the reasons given above) do not warrant the investment of U.S. Army resources to develop a treatment process for the non-stockpile program, whether the vendor of the technology is a U.S. entity or a foreign entity. The remainder of this chapter describes these technologies and provides reasons for this conclusion.

Initial development of the acid digestion process (ADP) was carried out by the Société Nationale des Poudres et Explosifs (SNPE) Propulsion in France (Gaudre et al., 2001).¹ The development effort was continued by Battelle, and the Shaw Group is now apparently involved (Soilleux, 2005). The discussion of the ADP in this section is in part based on published documents and on information obtained during visits to overseas sites. The committee asked for information from the vendor that is currently developing this technology, but because any such information would have been classified as “export confidential” under U.S. regulations, it was not possible for the committee to accept this information. It is known that the NSCMP is evaluating this technology in detail and it is possible that operational or demonstration trials will be performed. The assessment contained in this report was based on the information in the public record and, to a significant extent, on the opinions of experts in chemical demilitarization operations overseas who have additional knowledge of the technology. The committee recognizes that information that could have been provided by the vendor might have altered the conclusions that were reached. However, the report Possible Replacement Technologies for Operation Pongee After 2007 stated as follows: “More recent communications with the French prime contractor for Secoia reveal that they have revisited all candidate technologies and again rejected acid digestion on the grounds of safety and an incomplete process” (Soilleux, 2005, p. 14). Based on this information and on other sources, the committee concludes it is unlikely that acid digestion will prove to be a technology that is superior to other currently available technologies.

ADP uses 7M nitric acid to access munitions contents, destroy munition bodies, and oxidize the fills. The acid digests the steel that contains the hazardous substances, renders the fuzing systems inoperable, and decontaminates the chemical agents. The acid neutralizes/reacts with the agent by entering the agent cavity through the fuze. A significant amount of heat is generated by the reaction of the acid with the steel and must be removed. The system comprises the following parts:

• Reactor vessel where the munition is placed,

• Acid storage tank with recirculation pump,

• Acid heating and cooling systems,

• Air emissions control system,

• Gas hold-up tank, and

• Caustic and permanganate scrubbers to treat the gaseous effluents.

The munition is placed in a sealed reaction vessel and nitric acid is circulated through the reactor, with the exothermic heat of reaction removed by a heat exchanger in the recirculation loop. The reaction is continued until both the agent and energetic fills are neutralized. The reactor is then flushed and drained and the munition removed. The unit is mounted on skids, and the reactor chamber can be tailored to accommodate a wide variety of shapes and sizes of material.

The result of the treatment is a liquor that requires further treatment for recovery or disposal. The liquor includes spent nitric acid, permanganate and sodium hydroxide wastes, and cleanup rinse waters. It also contains the explosive in the munition, which is, theoretically, untreated and must be removed by filtering.

Recovered munitions containing phosgene, mustard agent, or chloropicrin are reported to have been destroyed using ADP. Lab tests are reported to have shown it can also neutralize diphosgene, hydrogen cyanide, and nerve agents. The reported energetic material and chemical warfare fills that the process can be used to treat are given in Table 6-1.

The fate of the energetics has not been firmly established. Although it is claimed that the energetics can be filtered out of the waste liquid, there is a potential issue with energetics contaminating the downstream liquid system.

Advantages of ADP over the explosive destruction system (EDS) may be greater throughput capacity, smaller volume of waste produced, and no requirement for explosive containment. Disadvantages appear to be lower destruction efficiencies (based on tests conducted by Battelle), a requirement for offgas treatment, and safety concerns.

The report Possible Replacement Technologies for Operation Pongee After 2007 (Soilleux, 2005, p. 13) states as follows:

The information surrounding the abandonment of ADP by the French and an explosive incident associated with it was confirmed by representatives of some European organizations involved with chemical demilitarization who supplied information to this committee.²

ADP appears to be theoretically suitable for processing munitions that are damaged or corroded. A munition of any size can be processed, in theory, assuming the reactor is large enough to contain it. According to published claims, no explosion is possible, the process may be considered safe, and the reaction vessel does not have to be designed to withstand an explosion. However, in view of undocumented reports of an incident in France, this claim should be examined carefully.

Several disadvantages concerning ADP can be envisaged, but it may be possible to apply safeguards. However, given the lack of detailed information, there are some potential issues:

• The use of highly concentrated nitric acid poses a significant hazard for personnel that must not be underestimated or minimized.

• The munition must be handled, which presents another hazard for personnel.

• It is not clear on what basis the reaction is considered complete. It may be that a reaction time is specified, or it may be that some objective criterion is used, such as an analysis.

• Side reactions could generate an unstable compound during the reaction with nitric acid. Such unstable

reaction by-products have caused numerous explosions elsewhere in the chemical industry. It may be difficult to eliminate this concern. Reports from a variety of sources on a least one (and possibly more) explosion incidents with the ADP were noted during information-gathering visits to European organizations involved with chemical demilitarization (Soilleux, 2005).³

• It is not clear whether this technology is suitable for all agents that require treatment or only for some.

• It is not clear where the waste from the reactions with the agent and explosive would go, the extent to which this waste is hazardous, and what secondary treatment might be required. This waste stream, containing concentrated nitric acid and other wastes, would be difficult to handle.

• ADP requires an offgas treatment system of some complexity.

• The explosives filtered from the liquid waste would have to be destroyed in some manner, presumably by incineration on-site.

Other issues may exist that can only be discovered by a detailed review of the process, which was not possible given its current “export confidential” classification. In short, notwithstanding its apparent simplicity, the issues pertaining to ADP include the use of concentrated nitric acid, the potential for the generation of unstable nitrate compounds, the probable difficulty with handling the secondary waste streams, and the probable requirement to filter out explosives from the waste liquid and then treat the explosives. The French purportedly abandoned ADP after at least one explosive event and after having invested considerable effort to develop it. All of these considerations make it likely that this technology should not be pursued. However, as stated earlier, if more detailed information could be obtained from the vendor without having to treat it as confidential, the committee might be able to change its recommendation.

Vitrification is the process of melting materials to produce a glass or glass-like substance. The GeoMelt Division of AMEC Earth and Environmental, Inc., offers a bulk vitrification process for waste treatment that it calls In-Container Vitrification (ICV). AMEC is an English company, and the technology received a favorable evaluation in Soilleux (2005) for possible use in the United Kingdom for the destruction of recovered CWM.

ICV is being considered for the treatment of low-radioactivity tank waste now stored at the Department of Energy’s (DOE’s) Hanford, Washington, site (CH2MHILL, 2005). At Hanford, the bulk vitrification technology might be used to supplement the operation of the main tank waste treatment plant, possibly treating up to 42 percent of the 53 million gallons of liquid and solid wastes now stored in 177 large underground tanks. The ICV technology will use vitrification technology for the treatment of less active waste. The main waste treatment plant also uses vitrification technology and can treat more active waste.

The Washington State Department of Ecology approved a RCRA research, development, and demonstration permit for the ICV pilot plant (CH2MHILL, 2005; WNO, 2005).⁴ The permit is very specific, allowing DOE to do testing for no more than 400 operating days and to treat no more than 300,000 gallons of waste from Hanford’s tank No. S-109. The building must be torn down when testing is completed. The testing will be done at full scale, i.e., using the same size container as would be used in the production facility.

Laboratory testing at various scales, up to an engineering scale that was 1/16 the size of the planned full-scale unit, has been under way since about 2003. The objective of this testing has been to identify additives that will promote the formation of a suitable final glass waste form. It is important that the waste formed exhibit minimal leaching of radioactive components and other contaminants of concern, such as RCRA heavy metals (Buelt et al., 1987; Loehr et al., 1992; Thomas and Treat, 2001).

The waste to be treated is mixed with glass-forming additives and local silica-rich earth, then dried, using a batch-mode rotary mixer/dryer and indirect steam heating (WDOE, 2005). The mixer/dryer has a capacity of 10,000 liters (2,645 gallons) at a fill fraction of 45-50 percent. During drying, moisture is removed from the offgases, and the dried gas is added to the main offgas treatment system for additional emission control.

The waste container is expected to be a steel box approximately 3 m (10 ft) high, 2.4 m (8 ft) wide, and 7.3 m (24 ft) long. Before waste is placed in it, the container will be lined with insulating board, sand, and a layer of castable refractory. The refractory will be in contact with the waste. A layer of melt-initiating graphite and soil will be placed over the refractory in the bottom of the container. The container will have one or more ports for sampling the vitrified waste after it has cooled.

A steel lid with attached electrodes will then be placed on the container. The lid is attached to the container using bolted flanges and a refractory gasket. The lid has several ports for waste addition, electrode connections, venting, sampling, and addition of postvitrification material. Some or all of the waste is placed in the container (see next paragraph), and electric

power is applied. The waste is heated to about 1300C over about 140 hours, forming a molten material. After passing through a high-efficiency particulate air (HEPA) filter, ambient air flows into the top of the container to maintain flow to the offgas treatment system, cool the vitrification offgases, and provide thermal protection to HEPA filters in the offgas treatment system. Vitrification offgases are vented under induced draft and flow to an offgas treatment system. As melting occurs, the depth of the waste decreases.

After all the waste has been melted, the melt is allowed to cool, forming the vitrified glassy material. Soil is added to the container so that it is at least 90 percent full. After the vitrified waste is sampled and tested, all connections to the container are removed. The container, still containing the electrodes, is taken away for storage or burial. During processing the volume of waste shrinks by 50 percent or more. Used personal protective equipment (PPE) and other secondary wastes can typically be recycled to the next batch of waste being treated, minimizing the waste.

Offgas treatment is extensive and involves use of sintered metal filters, quench systems, venturi scrubbers, a condenser, a mist eliminator, an offgas heater, parallel HEPA filters, a carbon filter for radioactive iodine removal, a baghouse, and a selective catalytic reduction unit; a packed tower scrubber system is used as a backup.

GeoMelt claims that its base technology has been granted a national Toxic Substances Control Act permit for the treatment of wastes containing up to 1.7 percent polychlorinated biphenyls (PCBs) (Campbell et al., 2005). Bulk vitrification was demonstrated for the treatment of waste from the Rocky Flats Environmental Technology site near Denver, Colorado.⁵ About 21,500 pounds of waste containing PCBs and low-level radioactivity were shipped from Rocky Flats to Waste Control Specialists near Andrews, Texas, and treated.

Two melts were conducted, with each melt taking 3 days and producing monolithic blocks about 80 cubic feet in size, a reduction in waste volume of more than 50 percent. The refractory and insulation from the first batch were reused for the second batch. PCBs, trichloroethylene, and perchloroethylene were reduced to below detection limits (not reported).

Information on gaseous emissions was likewise not reported. However, it is possible and perhaps likely that some of these organic compounds were thermally desorbed from the solid waste and that the extent of their destruction depended on the effectiveness of the offgas treatment system.

The technology was also demonstrated at Waste Control Specialists for the treatment of 9,575 pounds of another waste, source not indicated, that also contained PCBs and low levels of radioactivity.⁶ As received, the waste was about 25 percent oil and aqueous fluid and 75 percent sludge. Zeolite, bentonite, and soil were added to the waste, and treatment was carried out in five batches. The average batch treatment time was 53 hours. PCBs were reduced to below detection limits (not reported).

To the committee’s knowledge, the process has never been used on chemical weapons or agent. If it were to be used for this purpose, the high temperatures and long residence times would be expected to result in complete removal of any agent from the solid phase. However, some agent might be thermally desorbed from the feed material, which would mean depending on the offgas treatment system to destroy the balance of the agent. This could cause problems for public and regulatory acceptability in the United States. Also, large amounts of electricity are consumed, which might hinder public acceptance.

GeoMelt’s in situ technology has been tested on a small scale for the destruction of energetics, such as HMX, RDX, and TNT, in soil. These tests resulted in a vitrified product with no detectable explosives (Osborne, 2003; Campbell et al., 2005).

About 65-90 percent of heavy metals such as lead, cadmium, and arsenic are retained in the melt. The fugitive arsenic is captured in the offgas treatment system, “treated by standard arsenic treatment technology” (unspecified), and returned to the next melt (Osborne, 2003, p. 3).

While technically interesting, bulk vitrification has to date not been tested on either munition or agent destruction. It was therefore judged by the committee to not warrant listing as a Tier 1 technology for this report.

The firing pool is a large pool filled with an aqueous solution that can neutralize the agent in a chemical weapon (Guir, 1997). Guir (1997) says that “the firing pool technique discussed here was designed by Société Nationale des Poudres et Explosifs (SNPE) Ingénierie and based on an unpublished laboratory study at the Centre d’Etudes du Bouchet (CEB)” (Guir, 1997, p. 161).

Explosive charges are placed on the munition, which is then submerged into the center of the pool, and the explosive charges are detonated. The munition is converted into fragments and the burster is exploded. The pool is designed to withstand multiple explosions, and it is expected that one weapon can be destroyed every 15 minutes. From laboratory-scale experiments, it was concluded that a firing pool with a 12-meter diameter and a 6-meter depth, filled with approximately 500 cubic meters of aqueous decontamination

solution, could be used to destroy about 100 metric tons of mustard agent (Guir, 1997).

A concern with the process is that mustard agent that has been contained in a munition for 70 or more years is likely to have polymerized, becoming viscous and sticky. This sticky mustard agent is likely to adhere to the fragments of the munition. The fragments are expected to be large and few in number. As a consequence, it is expected that agitation of the pool would not be sufficient to cleanse the agent from the fragments, necessitating the use of a downstream fragment cleaning system. It is also expected that large amounts of agent would escape from the surface of the pool, requiring the use of some means of destroying the agent in the vapor phase, with an attendant air pollution control system.

No mention of the development of the firing pool technology past the bench scale was found in the literature. The firing pool technology was not considered further by the committee because it had not been further developed and because of the problems that were uncovered during the bench-scale studies.

Biological approaches for the destruction and detection of chemical agents are being developed for applications such as the mitigation or remediation of contaminated media, sensors to detect the presence of agent, degradation of agents using enzyme extracts, as well as a few other less mature applications. None of these approaches has been fully developed into a technology on a large scale. However, several hold promise to complement the large-scale physical/chemical techniques used for the destruction of non-stockpile chemical weapons. For the nerve agents and blister agents (vesicants), microorganisms and some of the mechanisms for degrading/neutralizing the agents have been identified. Several enzymes from bacteria and fungi have been identified that are capable of degrading a wide range of organophosphate compounds. A considerable effort has been expended on the organophosphate-degrading enzymes owing in part to their potential utility for pesticide destruction. None of these concepts was developed sufficiently to justify its inclusion in this report as a Tier 1 technology.

Naturally occurring organisms have been found that can degrade a range of the chemical agents. Degradation in soils has been documented. Bioreactors have been developed for select chemical agents, either parent compounds or various degradation products of parent compounds. None of these have been expanded beyond the pilot scale, and most exist only at the bench scale. The stability of the microbial cultures remains a challenge. Much work has been done on the isolation of key enzyme systems. Although some studies exist that demonstrate the performance of these enzymes in aqueous suspensions, a significant effort has been expended on developing immobilized systems. These have been in the form of cryoimmobilized beads, packed columns, and impregnated cotton materials. Bioremediation technologies for soils contaminated by chemical agents were discussed at the Third International Workshop “Biotechnological Approaches to Chemical Weapon Destruction,” in Saratov, Russia, in 2000. The feasibility of using naturally occurring microorganisms of genetically modified bacteria has been proven, but no field-scale operations have been reported. There is also the potential to use a bioremediation/phytoremediation approach for treating contaminated soils. Patents have been filed for two degradation processes: one uses enzymatic processes and the other a reactor system for phosphonate degradation.

Some materials have been developed for the protection of personnel and farmers who might be exposed to chemical agents or pesticides. The clothing that is currently available has an absorptive polyurethane layer impregnated with activated carbon, which offers protection but does not neutralize/degrade the agents. It is expected that organophosphate degrading enzymes will be incorporated into this material. Enzyme-containing materials are also being developed for self-decontaminating clothing and surfaces. Because the preparation and purification of enzymes is expensive, work is focusing on immobilization techniques that can maximize enzyme performance and longevity. A brief summary follows:

• Pseudomonas diminuta (organophosphate hydrolase) immobilized on nylon performed for weeks without leaching (Caldwell and Raushel, 1991a, b). The type of support material was a key factor in the successful degradation or organophosphates (Havens and Rase, 1991).

• Taking advantage of cloning methods, an E. coli with surface-expressed organophosphate hydrolase was immobilized on cotton used in fabrics and filters (Ritchens et al., 2000; Wang et al., 2002). Grimsley et al. (2001) demonstrated that the tremendous capacity of cotton to absorb the nerve agent-degrading enzymes made it appropriate for use as decontaminating towelettes, gauze, swabs, bandages, and wound dressings.

• A variety of bioplastics and enzyme-polymer composites for use as reactive monoliths, foams, fibers, wipes, and coatings have been developed (Kline et al., 2000; Gill and Ballesteros, 2000a, b). In this form, the enzymes were found to maintain stability under

normally denaturing conditions. Easily prepared enzyme-containing polyurethanes have a wide range of properties (Braatz, 1994). Foams and gels can be prepared, but there has been variable success with these due to long-term stability problems. However, a sol-gel product has been shown to retain high activity and to have good stability (Gill and Ballesteros, 1998, 2000c).

Sandia Laboratories, in collaboration with EnviroFoam, has commercialized a product that is effective in the decontamination of materials exposed to chemical agents as well as to some pathogenic viruses and bacteria.⁷ This product consists of several chemicals, including some enzymes, and has been tested at several sites for different types of contamination.

Several biosensors are currently being developed primarily for detecting the release of agents into the air. However, some could be adapted for detecting agent in water or soil media. A project funded by DOE’s Initiatives for Proliferation Prevention program has resulted in an analyzer capable of detecting organophosphates, carbamates, and other inhibitors of butyrylcholinesterase. The electrochemical biosensor analyzer for detection and discrimination of different neurotoxins can analyze water, soil, and food samples, and has potential for air control and medical applications.

Exposure to high temperatures was used to destroy mustard agent and mustard agent/lewisite mixtures in a bench-scale study at the DSTL, Porton Down, England (Anderson et al., 2003). The agent or agent mixture was evaporated and mixed with an artificial air containing 79 percent nitrogen by volume and 21 percent oxygen.

Whether or not the amount of added air was less than or more than the stoichiometric amount for full oxidation of the agent(s) was not stated. The agent/air mixture was passed through quartz tubes contained within three close-coupled electric furnaces. The first oven was maintained at 600C or 750C, the second at 1100C or 1200C, and the third at 800C. Product samples were collected and analyzed. All agent was reportedly destroyed. AsCl₃, 1,2,4-trithiolane, 1,3,5-trithiane, cyclic octa-atomic sulfur, and As(SC₄H₉)₃ were identified in the products. Trace quantities of dioxins were also detected. The investigators concluded the results were sufficiently encouraging to warrant “consideration of the method as a means of dealing with the problem posed by the old and abandoned chemical munitions. The process appears to be safe, reliable and robust” (Anderson et al., 2003, p. 15).

Without knowing the agent/air mixture ratio, it is not possible to determine whether the destruction process was similar to combustion or to pyrolysis. Only one paper was found, and no further work was carried out by DSTL. This technology must be considered to be at a very early stage of development and therefore presently of little interest for non-stockpile application.

An earlier study (NRC, 2001a) for the U.S. Army NSCMP concerned with the treatment of liquid neutralent wastes produced by NSCMP disposal operations examined two electrochemical processes, the silver Ag(II) process, which was developed by the firm AEA Technologies in the United Kingdom, and the cerium Ce(IV) (or “CerOx”) process. Although the latter process was developed in the United States, the two processes are similar, and both are included here for completeness. A description of these two processes from the earlier NRC report follows:

The Ag(II) process was also examined as a potential technology for the disposal of assembled chemical weapons at two of the U.S. stockpile storage sites. A report on the Assembled Chemical Weapons Assessment program demonstration phase testing conducted for this process (NRC, 2001b) listed eight findings and four recommendations, which pointed to a number of observed shortcomings with regard to the application of this technology to chemical demilitarization at that time: continuing major design changes during the testing period, migration of organic material across the electrochemical cell membranes, and generation of new energetic compounds in the course of processing.

A further disadvantage of the Ag(II) process is that large quantities of silver (a toxic heavy metal) and nitric acid (a corrosive) are required, along with the attendant potential for production of toxic emissions and effluents.

The CerOx process avoided some of the difficulties of the Ag(II) process in that cerium is much cheaper and less toxic than silver. However, the most serious disadvantage for the CerOx process, at least at the time of the report (NRC, 2001a), was that it was found to not be as mature a technology as the Ag(II) process, and it had never been tested with any neutralents. As with the Ag(II) process, it also uses large amounts of nitric acid and thus rated poorly in terms of pollution prevention criteria.

The committee could not locate any information to indicate that either the Ag(II) process or the CerOx process had been developed significantly since 2001 in terms of their applicability to processing either stockpile or non-stockpile chemical agents or munitions.

Plasma arc is a very high temperature process that has proved most effective for liquid waste streams. It could be used to destroy neat agent or treat secondary waste streams resulting from agent destruction, including detonation offgases. It could also be used for destroying metal parts, dunnage, and energetics.

Plasma arc technology utilizes the electrical discharge of a gas to produce a field of intense radiant energy and high-temperature ions and electrons that cause target chemical compounds to dissociate within a containment chamber. Large volumes of high- temperature vapor are generated that require a treatment system composed of a series of gas scrubbers, HEPA filters, and monitors to ensure that the system meets regulatory emission limits.

Variations of the plasma arc process are numerous and involve different plasma gases and reactor designs that provide either an oxidizing or a reducing environment. One system, developed by MGC Plasma AG in Switzerland (the MGC/PLASMOX process), has achieved destruction efficiencies greater than 99.99999 percent (seven nines) when processing adamsite, Clark I and II, phosgene, lewisite, yperite and a mixture of yperite, and lewisite. PLASMOX employs closely coupled, staged reaction zones (characterized as controlled pyrolysis) to completely destroy organic compounds. The Army has also investigated the PLASMOX process for destruction of neutralent waste streams as part of its technology test program (NRC, 2002).

MGC/PLASMOX developed a portable unit, Model RIF 2, that was put into operation in 1994 and has since built additional units. The RIF 2 is skid-mounted and designed to be moved by four standard tractor-trailers. The unit has been used in Europe and is permitted under both Swiss and German environmental laws and regulations. It was used successfully to destroy chemical agents for the Swiss Army at its chemical materiel laboratory in Spiez, Switzerland. The PLASMOX tests run by the Germans and Swiss indicate that the system will destroy chemical agent safely and rapidly (Burns and Roe, 2001).

As part of a technology test program for non-stockpile CWM, the NSCMP hired Stone & Webster to conduct tests of the MGC/PLASMOX plasma arc process on simulated H and GB neutralents with MEA. MGC conducted these tests from January 8 through January 19, 2001, under a subcontract to Burns and Roe Enterprises at the MGC/PLASMOX facility in Switzerland.

The NSCMP had proposed that plasma arc technology be used primarily for the destruction of neutralent waste streams, e.g., it was a candidate for the direct destruction of the binary CWM components DF and QL, stored at Pine Bluff Arsenal. MGC/PLASMOX tests indicated a through-

put rate for neutralent processing of approximately 13 liters per hour; process availability was reported to be 50 percent (NRC, 2002). However, there has been no recorded destruction of non-stockpile materiel by plasma arc technology in the United States.

MGC/PLASMOX technology was used successfully to destroy approximately 20 tons of toxic chemicals, including chemical munitions, in Albania. A portable unit was sent from Switzerland to Tirana, Albania, in time to begin the destruction of the toxic chemicals in July 2001. During the destruction process, unexpected major problems were encountered that affected the operation of the exhaust pipe, the air cooler, and the quenching process. In addition, the centrifuge in the plasma furnace was damaged by the high oxygen content in the destroyed pyrotechnic devices. Fortunately, these problems were solved without having a significant impact on the schedule. This project was completed in September 2001, and the plant was returned to Switzerland (Huber and Werner, 2002).

The Army has identified approximately a dozen vendors of plasma arc technology in the United States, although none is currently permitted to treat hazardous waste or non-stockpile CWM (NRC, 2002).

Stone & Webster recommended that the MCG/PLASMOX system receive further testing on typical NSCMP liquid and solid waste streams, with particular attention paid to the deposition of solid materials in the system. Its report concluded that further improvements would have to be made to ensure that the system would comply with all EPA and state requirements (Stone & Webster, 2001).

A number of regulatory issues were raised by the Army’s test results for the MCG/PLASMOX technology that must be resolved before it could be permitted in the United States. These include improvements to the gas scrubber system, more complete knowledge of the fate of key components of the non-stockpile CWM (e.g., phosphorus), and better characterization of the solid, liquid, and gaseous secondary waste streams.

The key public concern about plasma arc processes for the destruction of non-stockpile CWM in the United States centers on whether plasma arc offers a true alternative to incineration. Depending on the type of plasma gas used and the configuration of secondary oxidation zones, quench, and scrubber processes, plasma arc systems may produce gas volumes and reaction products that are quite similar to or quite different from those associated with incinerators. When oxygen is used as the plasma gas and/or if the plasma arc process has additional treatment chambers, the process may be practically indistinguishable from incineration. On the other hand, vendors often highlight the fact that plasma arc processes that do not use oxygen as the primary plasma gas differ from incineration, although even in these systems, oxidation generally takes place at a subsequent stage of the process. However, the levels of dioxins, furans, and other hazardous pollutants are likely to be below regulatory limits when the plasma arc system is optimally designed and controlled.

A noteworthy indication of the variance in public attitudes toward plasma arc is that in one case, after careful consideration, the Assembled Chemical Weapons Assessment program Dialogue Group accepted plasma arc as a valid alternative to incineration. However, in contrast with this, a spokesperson for the Non-Stockpile Chemical Weapons Citizens Coalition characterized plasma arc as a synonym for incineration and expressed concern that NSCMP was prematurely embracing the technology. As with incineration, the degree of public concern about plasma arc may vary with specific implementation and specific location (NRC, 2002).

Consequently, although the committee believes that this high-heat technology could successfully destroy the various chemical agents, its problem with public acceptance has kept it from being a viable option. Therefore, the committee determined that further support for the technology by the NSCMP would be a waste of resources and assigned this technology to Tier 2.

The Scottish-based environmental technology firm Albagaia has developed two systems, one a laboratory-scale portable system and the other a trailer-mounted transportable system. These systems utilize photocatalytic technology for the destruction of chemical weapons agent. Photocatalytic technology is a heterogeneous electron transfer process (either oxidation or reduction) wherein a semiconductor such as titanium dioxide (TiO₂) is activated by light energy equal to or greater than the optical band gap of the semiconductor material, catalyzing the oxidation/reduction reaction. The organic agent is mineralized to carbon dioxide, water, and mineral acids or salts through the oxidation process. The photocatalytic process operates at near ambient temperature (±5°C) and pressure. In order for the photocatalytic process to be effective for the destruction of chemical munitions, the agent must be accessed and drained prior to treatment. The systems developed by Albagaia are not capable of accessing the agent nor are they effective for munitions hardware or energetics destruction.

The destruction of chemical contaminants by TiO₂ photocatalysis is well established. Laboratory, pilot, and field studies have demonstrated TiO₂-catalyzed photodegradation of a wide range of organic chemicals. Organic chemi-

cals tested include aldehydes, alkanes, alkenes, amines, aromatics, carboxylic acids, dioxins, dyes, fuel constituents, halogenated hydrocarbons, herbicides, ketones, mercaptans, nitroglycerines, pesticides, polychlorinated biphenyls, solvents, and surfactants (Legrini et al., 1993). Typical concentrations of organic constituents that have been treated are on the order of 5 to 500 mg/L. The photocatalytic extraction of heavy metal (lead and mercury) contaminants from water has also been reported (Tennakone and Wijayantha, 1998).

Research efforts focused on materials relevant for the destruction of chemical weapons has been limited. An international collaboration between researchers in France, Russia, and the United States demonstrated complete mineralization of the mustard agent simulant in both air and water (Vorontsov et al., 2002).

The removal of arsenic via photocatalytic oxidation of arsenite to arsenate, followed by adsorption to the TiO₂ surface, has also been demonstrated (Bissen et al., 2001; Dutta et al., 2004).

Albagaia appears to be the first to develop commercial photocatalytic technology for the direct destruction of chemical weapons agent. It has developed two systems based on TiO₂ photocatalysis, the portable chemical agent destruction system (P-CADS) and the transportable chemical agent destruction system (T-CADS). These systems use a slurry form of TiO₂ powder in an aqueous batch system with an ultraviolet light source.

In the United States, a similar process was demonstrated on a rinsate generated during EDS operation. The Photo-Cat photocatalytic oxidation process was developed by Purifics Environmental Technologies of London, Ontario, Canada, for water treatment. The process employs ultraviolet light, hydrogen peroxide, and a titanium dioxide catalyst to destroy organic compounds. In 2001, the NSCMP tested the process at the Aberdeen Proving Ground (Burnham et al., 2002). The test solution was rinsate generated during the Rocky Mountain Arsenal bomblet destruction operation, in which the EDS was used to destroy four M139 bomblets with GB (sarin) fills. The test unit was rated at 7.2 kW, capable of treating 138 gallons of rinsate per day. Stable operation of the Photo-Cat system resulted in total organic carbon being reduced from 1,610 mg/L to 37.8 mg/L in 89 hours, a reduction of approximately 98 percent. Isopropyl methylphosphonic acid, a Schedule 2 compound, was reduced to below its detection limit of 3.94 mg/L from a starting concentration of 20.1 mg/L.

The technology was tested at Porton Down in 2003 with agent. Available results indicate a 99 percent destruction of HD within 6 hours using a system containing 6.42 m² of TiO₂ and continuous ultraviolet (UV) energy of 24 watts. The same system yielded 99.987 percent destruction of HD after 24 hours. To date, this is the only known photocatalytic system tested for chemical agent destruction.

While no known commercial applications of the photocatalytic technology are in use for chemical agent destruction, photocatalytic systems are in use for other applications. Systems have been permitted for use in the treatment of water and patented in the United States for the treatment of indoor air (Goswami, 1998).

The technology would need to be coupled with a munitions access and drainage system. To be considered for implementation, higher degrees of destruction would need to be demonstrated. The Albagaia systems are small enough to be easily transported. In addition they would be appropriately suited for situations where small amounts of agent are available at a given time, for a destroy-as-you-go system.

The photocatalytic process operates at near-ambient temperature and pressure and is inherently safe. The UV lamps pose the greatest hazard to worker safety, and these lamps are fully contained. In summary, photocatalytic technology uses TiO₂ and UV light for the mineralization of chemical agent. It is not capable of destroying munitions or energetics, and therefore must be coupled with another technology for complete destruction of munitions. The process operates at ambient conditions and does not require the use of reagents, and therefore is not likely to pose a hazard to workers or the environment during its operation. The products are not toxic and with the exception of arsenic would not require additional processing. Prior to adoption of this technology for agent destruction, additional testing for confirmation of efficacy for specific agents is required. While this technology is of interest for agent destruction, the low reaction rates involved and limited use to date precluded it from being considered as a Tier 1 technology.

Plasmazon was developed in Germany and uses what are termed “activated ozone structures” for the destruction of chemical warfare materiel (Ehmer and Sieke, 1998). These activated ozone structures are characterized by unpaired or outer shell electrons and can be in one of three states: (1) basic (triplet or singlet) state [O₃(³B₂), O₃(¹B₂)]; (2) excited state [O₃(¹A₁,v)]; or (3) ionized state (O₃⁺,O₃). These species are generated using plasma ozone generation and withdrawing radicals from the discharge gap as quickly as possible after the dissociation of molecular ozone to atomic oxygen (reaction 1 below) and the generation of the ozone from reaction 2. The activated ozone structures have a half-life ranging from 70 msec to 70 sec, resulting in a higher level of radical generation than with the production and subsequent dissociation of ground state ozone.

(1)

(2)

Testing of the system for the destruction of Clark I has been reported (Sieke et al., 1998). The agent was diluted with

50 ml of acetone and the activated ozone passed through at a rate of 300 ml/min. After 1.5 hours of operation, a 98 percent reduction was reported. Multiple passes increased the reduction, with the fifth and final pass yielding a reported 99.999999 percent reduction from the initial concentration.

While the above results are encouraging, the committee determined that the Plasmazon technology had not yet developed sufficiently to justify including it in this report as a Tier 1 technology.

Finding 6-1. Based on currently available information, the Tier 2 technologies described in this chapter are not likely to meet foreseeable requirements of the NSCMP.

Recommendation 6-1. In the absence of significant developmental progress or unforeseen circumstances that would warrant reconsideration, the U.S. Army should not expend further resources on the evaluation of the following technologies for NSCMP applications:

• Acid digestion,

• Bulk vitrification,

• Firing pool,

• Biological approaches,

• DSTL electric furnace,

• Electrochemical oxidation,

• Photocatalysis,

• Plasma arc, and

• Plasmazon.

Gaudre, M., M. Ferrari, and J.-M. Tauzia. 2001. Digestion process for securisation of recovered old ammunitions. Proceedings of the 2001 Chemical Weapons Demil Conference (CD). Porton Down, England: Defence Science and Technology Laboratory.

Soilleux, R. 2005. Possible Replacement Technologies for Operation Pongee After 2007. Porton Down, England: Defence Science and Technology Laboratory.

Buelt, J.L., C.L. Timmerman, K.H. Oma, V.F. Fitzpatrick, and J.G. Carter. 1987. In Situ Vitrification of Transuranic Waste: An Updated Systems Evaluation and Application Assessment. PNL-4800 Supplement 1, March. Richland, Wash.: Department of Energy.

Campbell, B., L. Thompson, and D. Fraser. 2005. Destruction of organic and explosive compounds using the GeoMelt^® process. Available online at <http://www.dstl.gov.uk/news_events/conferences/cwd/2005/proceedings45.pdf>. Last accessed February 24, 2006.

CH2MHILL. 2005. Status and Direction of the Bulk Vitrification Program for the Supplemental Treatment of Low Activity Tank Waste at Hanford, RPP-24372-FP, Revision 0, January 12. Richland, Wash.: Department of Energy.

Loehr, C.A., J.R. Weidener, and S.O. Bates. 1992. Product Evaluation of In Situ Vitrification Engineering Test 4. EGG-WTD-9868, September. Idaho Falls, Idaho: Department of Energy.

Osborne, D. 2003. GeoMelt vitrification technologies. Proceedings of the 2003 Chemical Weapons Demil Conference (CD). Porton Down, England: Defence Science and Technology Laboratory.

Soilleux, R. 2005. Possible Replacement Technologies for Operation Pongee After 2007. Porton Down, England: Defence Science and Technology Laboratory.

Thomas, T.N., and R.L. Treat. 2001. Evaluation of In Situ Vitrification for Operable Unit 7-13/14. INEEL/EXT-2001-00279. Idaho Falls, Idaho: Department of Energy.

WDOE (Washington State Department of Ecology). 2005. DOE/ORP-2003-23, Rev. 1a. Available online at <http://www.ecy.wa.gov/programs/nwp/dbvs/sub2/enc5.pdf>. Last accessed February 28, 2006.

WNO (U.S. Water News Online). 2005. Hanford breaks ground on pilot project to glassify waste, February 2005. Available online at <http://www.uswaternews.com/archives/arcquality/5hanfbrea2.html>. Last accessed February 24, 2006.

Guir, F. 1997. The technical challenge of dismantling and destroying old and abandoned chemical weapons. Pp. 156-165 in The Challenge of Old Chemical Munitions and Toxic Armament Wastes. T. Stock and K. Lohs, eds. Oxford, England: Oxford University Press.

Braatz, J.A. 1994. Biocompatible polyurethane-based hydrogels. Journal of Biomaterials Applications 9(1): 71-96.

Caldwell, S.R., and F.M. Raushel. 1991a. Detoxification of organophosphate pesticides using an immobilized phosphotriesterase from Pseudomonas diminuta. Biotechnology and Bioengineering 37(2): 103-109.

Caldwell, S.R., and F.M. Raushel. 1991b. Detoxification of organophosphate pesticides using a nylon based immobilized phosphotriesterase from Pseudomonas diminuta. Applied Biochemistry and Biotechnology 31(1): 59-73.

Gill, I., and A. Ballesteros. 1998. Encapsulation of biologicals with silicate, siloxane, and hybrid sol-gel polymers: An efficient and generic approach. Journal of the American Chemical Society 120(34): 8587-8598.

Gill, I., and A. Ballesteros. 2000a. Bioencapsulation within synthetic polymers (part 1): Sol-gel encapsulated biologicals. Trends in Biotechnology 18(7): 282-296.

Gill, I., and A. Ballesteros. 2000b. Bioencapsulation within synthetic polymers (part 2): Non-sol-gel protein-polymer composites. Trends in Biotechnology 18(11): 469-479.

Gill, I., and A. Ballesteros. 2000c. Degradation of organophosphorous nerve agents by enzyme-polymer nanocomposites: Efficient biocatalytic materials for personal protection and large-scale detoxification. Biotechnology and Bioengineereing 70(4): 400-410.

Grimsley, J.K., W.P. Singh, J.R. Wild, and A. Giletto. 2001. A novel, enzyme-based method for the wound-surface removal and decontamination of organophosphorous nerve agents. Pp. 35-49 in Bioactive Fibers and Polymers, ACS Symposium Series No. 742. J.V. Edwards and T.L. Vigo, eds. New York, N.Y.: Oxford University Press.

Havens, P.L., and H.F. Rase. 1991. Detoxification of organophosphate pesticide solutions: Immobilized enzyme systems. Pp. 261-281 in Emerging Technologies in Hazardous Waste Management II, ACS Symposium Series No. 468. D.W. Tedder and F.G. Pohland, eds. New York, N.Y.: Oxford University Press.

Kline, B.J., G. Drevon, and A.J. Russell. 2000. Enzymes in polymers and polymers from enzymes. Pp. 397-431 in Enzymes in Action, Green Solutions for Chemical Problems (NATO Science Series 1: Disarmament Technologies, Volume 33). B. Zwanenburg, M. Mikolajczyk,

and P. Keilbanski, eds. Dordrecht, The Netherlands: Kluwer Academic Publishers.

Richins, R.D., A. Mulchandani, and W. Chen. 2000. Expression, immobilization, and enzymatic characterization of cellulose-binding domain-organophosphorus hydrolase fusion enzymes. Biotechnology and Bioengineering 69(6): 591-596.

Wang, A.A., A. Mulchandani, and W. Chen. 2002. Specific adhesion to cellulose and hydrolysis of organophosphate nerve agents by a genetically engineered Escherichia coli strain with a surface-expressed cellulose-binding domain and organophosphorus hydrolase. Applied and Environmental Microbiology 68(4): 1684-1689.

Anderson, D.W., M.S. Sandford, and S. Kirkpatrick. 2003. Decomposition of sulfur mustard and sulfur mustard/lewisite mixture in an electric cylinder furnace. Proceedings of the 2003 Chemical Weapons Demil Conference (CD). Porton Down, England: Defence Science and Technology Laboratory.

NRC (National Research Council). 2001a. Disposal of Neutralent Wastes. Washington, D.C.: National Academy Press.

NRC. 2001b. Evaluation of Demonstration Test Results for Alternative Technologies for Demilitarization of Assembled Chemical Weapons: A Supplemental Review for Demonstration II. Washington, D.C.: National Academy Press.

Burns and Roe Enterprises. 2001. PLASMOX Technology for the Destruction of Chemical Warfare Materiel, March. Aberdeen Proving Ground, Md.: Program Manager for Chemical Demilitarization.

Huber, U., and D.B. Werner. 2002. VETOXA: Destruction of Toxic Chemicals in Albania, Report Number LS 2004-05, March 15. Spiez, Switzerland: Spiez Laboratory.

NRC (National Research Council). 2002. Systems and Technologies for the Treatment of Non-Stockpile Chemical Warfare Materiel. Washington, D.C.: The National Academies Press.

Stone & Webster. 2001. Evaluation of the Limited Engineering-Scale-Testing of the PLASMOX^® Technology to Treat Chemical Warfare Materiel, July. Aberdeen Proving Ground, Md.: Program Manager for Chemical Demilitarization.

Bissen, M., M.M. Vieillard-Baron, A.J. Schindelin, and F.H. Frimmel. 2001. TiO₂-catalyzed photooxidation of arsenite to arsenate in aqueous samples. Chemosphere 44(4): 751-757.

Burnham, S.L., C.A. Freedlund, and J.M. Cardito. 2002. Evaluation of the Phase II Engineering Scale Testing of the Photo-Cat^® Process to Treat NSCMP Rinsate Solutions, Revision 1, February. Aberdeen Proving Ground, Md.: Program Manager for Non-Stockpile Chemical Materiel.

Dutta, P.K., A.K. Ray, V.K. Sharma, and F.J. Millero. 2004. Adsorption of arsenate and arsenite on titanium dioxide suspensions. Journal of Colloid and Interface Science 278(2004): 270-275.

Goswami, D.Y. 1998. Goswami, D.Y., inventor. Universal Air Technology, assignee. November 10, 1998. Photocatalytic system for indoor air quality. U.S. Patent 5,835,840.

Legrini, O., E. Oliveros, and A.M. Braun. 1993. Photochemical processes for water-treatment. Chemical Reviews 93(2): 671-698.

Tennakone, K., and K.G.U. Wijayantha. 1998. Heavy-metal extraction from aqueous medium with an immobilized TiO₂ photocatalyst and a solid sacrificial agent. Journal of Photochemistry and Photobiology A: Chemistry 113(1): 89-92.

Vorontsov, A.V., A.A. Panchenko, E.N. Savinov, C. Lion, and P.G. Smirniotis. 2002. Photocatalytic degradation of 2-phenethyl-2-chloroethyl sulfide in liquid and gas phases. Environmental Science and Technology 36(23): 5261-5269.

Ehmer, K., and R.W. Sieke. 1998. R.W. Sieke, inventor. May 14, 1998. Devices and methods for producing and using ozone. European patent WO9819961.

Sieke, R.W., G. Lippke, A. Krippendorf, R. Haas, and S. Lüdtke. 1998. Destruction of diphenylarsine chloride (CLARK I) with activated ozone. Environmental Science and Pollution Research 5(4): 199-201.