Neutralization (Chemical Hydrolysis)
GENERAL CHARACTERISTICS AND EXPERIENCE
One potentially attractive approach to the destruction of chemical agents is chemical hydrolysis (i.e., reaction with water to form products of reduced toxicity). Because some of the reaction products are acids, a base such as lime or sodium hydroxide is generally used to neutralize the acidic products, hence, the generic description of the process as neutralization. The base may also catalyze the hydrolysis to permit it to be carried out more rapidly and with more efficient use of process equipment. The hydrolysis of the nerve agent GB (sarin) has been studied extensively and is described by the reaction:
Neutralization is potentially applicable to the whole family of phosphate-based nerve agents and to blister agents such as the sulfur ''mustards'' and arsenic-based Lewisite, although the processes for agents other than GB have had much less study. Despite several known problems in hydrolysis technology, neutralization has attractive features that commend it for study as an alternative to the liquid incinerator of the baseline system. Hydrolysis is usually carried out under mild conditions, generally atmospheric pressure and less than 100ºC, in fairly conventional chemical process equipment. After neutralization, the products are nearly neutral solutions containing inorganic salts and organic compounds of greatly reduced toxicity. The product solutions can be treated further by a variety of processes to produce "mineralized" products suitable for disposal like normal chemical wastes. Because process conditions are mild, the volatile components of the reaction mixtures should be contained easily.
The U.S. Army's experience with neutralization of the nerve agent GB has confirmed some of the potential virtues as well as the practical problems associated with this technology. In an extensive field program carried out at the Rocky Mountain Arsenal from 1973 to 1976, 4,188 tons of GB were hydrolyzed successfully (Flamm et al., 1987). The nerve agent was treated with a large excess of aqueous sodium hydroxide to produce a water solution of inorganic salts and organic compounds. The solutions were evaporated, and the solid residues were deposited in a hazardous waste landfill. With hindsight and better analytical capabilities, it appears that the amount of solid waste could have been substantially reduced by working with a much smaller ratio of alkali to GB. In addition to the U.S. Army experience with hydrolysis of GB, various neutralization processes have been used to destroy multi-ton quantifies of the agent in Great Britain, Canada, the Soviet Union, and most recently, Iraq. Overall chemical hydrolysis has much to recommend it for destruction of bulk quantities of agent GB.
In contrast to the large-scale use of neutralization to destroy GB, the experience with hydrolysis of nerve agent VX and the various mustard agents is largely confined to laboratory- and pilot-scale studies. A major problem is that these agents, in contrast to GB, are much less soluble in water and react slowly under neutral or alkaline conditions. Several approaches to deal with this problem have been explored (Yang et al., 1992). For agent VX, aqueous solubility is greatly increased by working in acidic solution because the acid protonates the amine function of the agent to form a soluble salt:
Although both acid and base catalyze the hydrolysis of the P—S and P—O bonds, these reactions are slow and incomplete at ambient temperature. At elevated temperature (75-90ºC), alkaline hydrolysis detoxifies VX in less than one hour (NRC, 1993a).
In a related process, Russian workers have utilized an acid-catalyzed transesterification of VX with ethylene glycol to give glycol esters of methylphosphonic acid that had low toxicity and could be burned safely.
The alkaline hydrolysis of VX can be accelerated by the addition of a stoichiometric quantity of hydrogen peroxide to the conventional sodium hydroxide solution (Yang et al., 1993). The peroxide has a twofold role in the process. With base, it generates a powerful nucleophile, the hydroperoxide anion, that catalyzes specific P—S bond cleavage to form products of lower toxicity than are formed by base alone. It also oxidizes the aminothiol hydrolysis product to a sulfonic acid of low toxicity, which is not readily converted back to VX. The peroxide-accelerated reaction appears promising as a means to detoxify VX. Use of hydrogen peroxide is convenient because the reagent is widely available commercially and because the only reagent-derived product is water.
A second approach to dealing with water-insoluble agents such as VX and mustard is to carry out the hydrolysis in a polar organic solvent or in a mixture of water and an organic solvent. With VX, the alkaline hydrolysis can be carried out in a polar solvent to give mixed hydrolysis products. The products retaining the P—S bond are toxic. With mustard agents, hydrolysis in the presence of an organic solvent should follow the simple equation:
If the reaction proceeded simply as written, it might be attractive to consider isolation of the thiodiethanol (TDE) product for sale through normal commercial channels. In practical terms, however, this strategy seems unlikely because (a) TDE is easily reconverted to mustard agent by treatment with concentrated hydrochloric (muriatic) acid and, hence, is regulated as a chemical agent precursor under the terms of the Chemical Weapons Convention; (b) it would be difficult and expensive to purify by-product TDE to the purity required for normal commercial uses such as textile finishes; and (c) the hydrolysis under mild conditions does not follow the simple equation shown above.
This last point (c) derives from the fact that military-grade mustard agents are not simple compounds, but complex mixtures of chemicals, many of which are polymeric. Apart from deliberate addition of polymeric gelling agents for military purposes, much of the mustard agent in the U.S. Army's
stockpile is gelled through a normal autopolymerization reaction that occurs on prolonged storage. The initial step in the polymerization is ideally represented as
In practice, the polymerization is much more complex and leads to gelatinous masses that are difficult to drain from munitions and storage containers. Hydrolysis of these complex species gives complex mixtures of products.
An important concern in the neutralization approach to the destruction of both nerve and blister agents is that the agent must be destroyed "irreversibly" to meet the requirements of the 1993 Chemical Weapons Convention. The document specifies destruction by "a process by which the chemicals are converted in an essentially irreversible way to a form unsuitable for the production of chemical weapons." This requirement is not always met by hydrolysis processes alone. As noted above, hydrolysis of mustard to TDE is not irreversible because treatment of the product with concentrated hydrochloric acid regenerates the mustard agent. Similarly, hydrolysis of phosphate-based nerve agents may give compounds that can be reconverted to the agents. Destruction of the C—P bond in the methylphosphonate agents GB and VX is needed to meet the requirement for irreversible destruction because methylphosphonic acid is a "scheduled" precursor.
The limitations on the effectiveness of hydrolysis in destroying nerve and mustard agents suggest that neutralization may not be an adequate destruction system in itself. It may, however, constitute part of an effective system when combined with other techniques such as chemical, biological, or supercritical water oxidation. Potentially useful systems are sketched below. These integrated systems must be compared with the baseline system to establish their effectiveness for application at various stockpile locations in the continental United States.
LIMITATIONS ON THE USE OF NEUTRALIZATION
Because neutralization may not do the whole task of munitions destruction, it may be used as part of an integrated system containing several discrete processes. The general system constraints have been discussed earlier. For chemical hydrolysis specifically, the constraints and requirements may be classified as either chemical or physical.
Hydrolysis only reduces the toxicity of liquid streams containing chemical agents. Both to ensure worker safety and to meet the "irreversibility" treaty requirement, hydrolysis may have to be followed by some other process (generally oxidation) that totally destroys the toxic agent and yields products that cannot be reconverted to agent.
To meet U.S. environmental requirements, the products of agent destruction must generally be "mineralized" (i.e., converted to carbon dioxide, water, and inorganic salts, which may be stored safely in a landfill).
Three major waste streams must be dealt with effectively:
Gases: Although the gaseous waste streams from a hydrolysis process are likely to be small, they must be handled and treated like those arising from the baseline system. The treatment must ensure that there is no agent release and that there is no release of other volatile organic compounds that would impair air quality.
Liquids: Hydrolysis, whether carried out with aqueous reagents or with solutions in organic solvents, involves handling large volumes of liquids. All the liquid streams must be treated by processes that totally destroy the agent and eventually produce mineralized products. Supercritical water oxidation or wet air oxidation combined with biotreatment appears to be a viable option for mineralization.
Solids: A neutralization system must ultimately produce oxidized solids that can be contained safely in a hazardous waste landfill. Salts such as fluoride, phosphate, and sulfate should be in insoluble forms (such as calcium salts) to reduce the likelihood of leaching into groundwater.
Access to Agent
The U.S. arsenals contain chemical agents stored both in bulk containers and in munitions such as projectiles, bombs, rockets, and land mines. In order to employ neutralization as a destruction process, it will be necessary to transfer the agents from their present containers to appropriate chemical reactors. Access to the agent is easiest when the agent is a pure liquid stored in an easily drained bulk (ton) container, although agent drained from munitions can also be treated by hydrolysis. In draining munitions for agent destruction by hydrolysis, one encounters the same problems as those of the
baseline system. As a result, a munitions disassembly process is needed as a "front end" to a neutralization-based agent destruction system.
Other Munitions Components
Chemical hydrolysis is difficult to apply to the agent retained in the metal parts of munitions that have been drained after disassembly. Thorough mixing of the hydrolysis reagent with the liquid agent is necessary to ensure complete neutralization, but it is difficult to get good mixing with films of agent adhering to metal surfaces in the confines of projectiles, bombs, rockets, and other munitions. It is extremely difficult to get complete hydrolysis of agent retained in minute cracks in the metal. Neutralization alone is also ineffective in destroying "energetics" such as explosives and propellants that are contained in the munitions. Because of these physical limitations on low-temperature, liquid-phase destruction processes, it is likely that a complete system for munitions destruction must employ a thermal process as a "back end" to the system. The thermal process might be either incineration as in the baseline system or high-temperature baking of the metal parts of a weapon.
Potentially Useful Applications of Neutralization
The chemical and physical constraints listed above help define situations in which a neutralization-based system might be preferred as a way to destroy agents in the U.S. stockpile. Neutralization systems look most attractive when neat liquid agent is easily available. The best situations for its use are arsenals where the agent is stored only in ton containers. These containers are drained easily (apart from gelled agent) and present relatively little metal surface to be decontaminated after draining. If no other chemical munitions are present at such sites, the capital investment for draining, neutralization, and decontamination facilities may possibly be lower than for the baseline system even though development costs will be higher. The two arsenals that seem to meet this criterion are the Newport Army Ammunition Plant, Indiana, which stores only ton containers of VX, and the Aberdeen Proving Ground, Maryland, which stores only ton containers of mustard agents.
Since chemical hydrolysis of agents contained in munitions such as rockets and projectiles requires access and decontamination processes analogous to the liquid incinerator in the baseline system, neutralization-based systems offer little advantage at sites that store several types of munitions, especially if a variety of chemical agents are contained in the munitions. This disadvantageous situation seems to exist at all U.S. arsenals other than Newport and Aberdeen. Chemical hydrolysis could be substituted for
incineration of agent drained from munitions in the baseline system, but the substitution would probably not result in cost reduction or in improved worker safety.
Neutralization could be used at Newport and Aberdeen in either of two contexts: (1) it could be used as part of a complete on-site agent destruction and waste treatment facility; or (2) it could be used for preliminary detoxification of agent on-site, with the hydrolysis products shipped elsewhere for incineration, biological treatment, or oxidation by other means. The "stand-alone" facility, option (1), at each site would involve the full set of unit operations described in the following section.
Option (2) is economically attractive because it would require construction of only the facilities needed for neutralization. The hydrolysis products could be sent to other arsenals for final disposal. For maximum advantage, the drained metal ton containers would be decontaminated by treatment with a hydrolysis reagent to render them safe for shipment to another site having the thermal treatment facilities necessary to prepare the metal for final disposal. This approach would eliminate the need for an incinerator and might reduce the capital cost for disposal facilities.
POSSIBLE SYSTEM DESIGNS FOR NEUTRALIZATION
As discussed above, a neutralization process is but one component of the complex integrated system required for safe destruction of chemical agents. A complete system must contain the following elements:
a drain station in which containers of liquid agent are opened and emptied in a contained-atmosphere environment, just as in the baseline system;
reactors for the hydrolysis and oxidation of liquid agent;
thermal treatment to decontaminate metal parts such as ton containers or munitions such as bombs (the treatment might be done either by furnace heating as in the baseline system or by baking in an externally heated oven); and
cleanup treatments for each waste stream:
gas streams include vapors of the agent from the contained-atmosphere rooms used for munitions handling, gases evolved from the chemical reactors, and combustion gases from the thermal treatments;
liquids, primarily the aqueous effluent from the hydrolysis and oxidation reactors; and
solids, largely salts from evaporation of the liquids, but also ash from the thermal treatment.
One possible scheme for integration of these elements into a complete system for destruction of chemical agents is sketched in Figure E-1. The postulated system would be better adapted to handling bulk agent than complete munitions. Some of the unit operations (draining and gas cleanup) are similar to those in the baseline system and are well developed. Others are significantly different and would require extensive research and development. Two of these significantly different aspects (agent hydrolysis and treatment of liquid hydrolysate) are discussed in more detail below.
Liquid Agent Destruction
Most neutralization-based schemes for agent destruction involve hydrolysis and oxidation in sequence. For the nerve agent GB, the hydrolysis-neutralization step is largely developed. The oxidation to convert organic products from GB hydrolysis (mostly isopropyl alcohol and methylphosphonic acid) to mineralized products for disposal would require extensive research and development. The processes might be much the same as those required for VX. If the neutralization of GB were chosen for development, it could be useful as a front end for wet air or supercritical water oxidation. The alkaline hydrolysis of GB would give an aqueous solution that would have the reduced toxicity and low corrosivity desirable in a feed for supercritical water oxidation or wet air oxidation. Oxidation of the hydrolysate under the severe conditions of supercritical water should give mineralized products suitable for disposal after precipitation of calcium Salts and evaporation of the water for recycle.
Implementation of hydrolysis processes for VX nerve agent may require extensive research and development because VX is largely insoluble in alkaline solution and because uncatalyzed hydrolysis is slow. As noted earlier, the peroxide-accelerated alkaline hydrolysis is a promising approach. Oxidation of the hydrolysate under the severe conditions of supercritical water oxidation should give mineralized products suitable for disposal after precipitation of calcium salts and evaporation of water for recycle.
An interesting but undeveloped alternative is to combine hydrolysis and oxidation in a single step (Cooper and McGuire, 1993). A major virtue to combining the two processes is that even though the hydrolysis of the P—S bond in VX is slow, it is greatly accelerated if the sulfur is oxidized first:
The P—SOR product is readily hydrolyzed and oxidized further to give phosphonic and sulfonic acid derivatives. The desired hydrolysis-oxidation can be accomplished with chemical reagents such as oxone (KHSO5) and peroxydisulfate (). These reactions are being studied at the Army's Edgewood Research, Development, and Engineering Center. If, in fact, the combined hydrolysis-oxidation process can be done without generation of excessive quantities of reagent-derived waste (dilute sulfuric acid), it may be an attractive approach to destruction of bulk VX. Preliminary results indicate that the reagents will convert all of the organic products to carbon dioxide or carbonate salts for easy disposal.
The hydrolysis-oxidation approach to destruction of mustard agents is more difficult than with nerve agents. The mustards are heterogeneous mixtures of chemical compounds and have low solubility in water, either acidic or alkaline solution. The solubility problem can be partially overcome by vigorous agitation, elevated temperature, polar organic solvents, or phase-transfer agents, but it is a technical obstacle that will increase cost as well as adding to development time and effort.
One of the most promising approaches to neutralization of mustard involves hydrolysis with calcium hydroxide (slaked lime) or sodium hydroxide. Both reagents convert mustard rapidly to thiodiethanol and cyclic thioethers (Reichert, 1975; Yang et al., 1993). The products emerge from the reaction as an alkaline solution that may be useful for supercritical water oxidation because it should have low toxicity and reduced corrosivity. It should also be suitable for demilitarization by wet air oxidation or by biological treatment. The oxidation step must be closely linked to hydrolysis because, as noted earlier, TDE is a chemical agent precursor regulated by treaty. Its handling and disposal must be monitored and documented carefully.
An approach to combining hydrolysis and oxidation is to use strongly acidic oxidizing agents such as oxone or peroxydisulfate. These reagents oxidize mustard to carbon dioxide, water, sulfate, and chlorine, but the completeness of oxidation needs to be demonstrated. Oxone and peroxydisulfate generate very large quantities of dilute sulfuric acid as a by-product. To avoid a large sulfate waste stream, it has been proposed that the sulfate by-product be recycled to an electrochemical cell for regeneration of oxone or peroxydisulfate (Cooper and McGuire, 1983). Both the chemistry and the engineering feasibility of this approach need to be demonstrated.
Waste Stream Treatments
The waste streams from a neutralization-based process differ from those generated by the baseline system. The greatest difference is the nature of the liquid generated by the hydrolysis and oxidation steps. The gas and solid waste
streams, however, also differ significantly from those in the baseline system. A particular virtue of neutralization is that most of the products of agent destruction are liquids or solids that can be stored easily until analysis and certification are completed.
The gases and vapors generated by a system like that in Figure E-1 resemble those from the baseline system, but there are two significant differences. One major difference is that the low temperatures in the neutralization and oxidation processes should not generate dioxins and furans. The main potential for formation of these materials in the proposed system is in the furnace or oven used to decontaminate metal parts. The small amount of products of incomplete combustion formed in this operation should be handled by the charcoal filter system proposed for the ''enhanced'' baseline system. In principle, there is little opportunity for formation of chlorinated dioxins or furans in the destruction of VX because there is no significant amount of chlorine in the agent or the proposed reagents. Mustard agents contain chlorine, but they have been burned at the Johnston Atoll Chemical Agent Disposal System (JACADS) without exceeding accepted limits for products of incomplete combustion production.
A second difference between the baseline system and the scheme of Figure E-1 arises if an externally heated oven is used instead of an internally fired furnace to treat contaminated metal parts such as ton containers. The oven baking treatment may drive vapors of unreacted agent into the gas phase along with the products of thermal decomposition and combustion.
The liquid streams exiting from the hydrolysis and oxidation reactors will vary greatly depending on the agent being treated and the reagents used to destroy it. Perhaps the simplest situation is presented by the effluent from supercritical water oxidation of hydrolyzed agents. The effluent should be a slightly alkaline water solution of inorganic salts such as carbonate, phosphate, sulfate, and chloride (also fluoride if GB is treated this way). All these materials except chloride can be precipitated as relatively insoluble calcium salts that can be filtered, calcined, and deposited in a hazardous waste landfill (after proper analysis and certification). The aqueous filtrate containing chloride and traces of other salts can be treated by conventional processes as in the baseline system.
Wet air oxidation (WAO) of the aqueous hydrolysis product, followed by biological oxidation, should produce a result like that obtained with supercritical water oxidation. The combination of wet air oxidation and biotreatment is a common commercial practice.
The aqueous hydrolysis products could also be oxidized chemically. Mild alkaline oxidants such as peroxide or hypochlorite may produce liquid wastes containing significant amounts of partially oxidized organic products. These compounds should have low toxicity, but will require extensive further treatment before they can be released into the environment. Use of chlorine
or hypochlorites as oxidizing agents will also lead to high chloride concentrations in the effluents. The liquid effluents from any particular process must be characterized thoroughly in order to develop appropriate posttreatment processes.
Treatment of the agent hydrolysis products with acidic, sulfur-based reagents (oxone, peroxydisulfate) would produce voluminous waste streams because many pounds of oxidant are required to oxidize 1 pound of agent. The oxidizing agent is converted to dilute sulfuric acid, which will be contaminated with phosphate or chloride from oxidation of the agent. These waste streams could be treated with lime to precipitate calcium sulfate, but the sheer volume of this product would be a severe burden on landfills unless the sulfuric acid can be recycled.
One critical aspect to be considered with any neutralization-based process is that large volumes of liquid need to be pumped, piped, filtered, and evaporated. Although these are standard unit operations in the chemical industry, each operation introduces a risk of leaks and spills of liquids, some of which may contain the agent or other toxic compounds. Careful engineering is required to reduce the risks. Considerable forethought is also needed because all of the effluents are eventually released to the environment as carbon dioxide, water, and the solids that go to landfills. The volume of water to be released can be reduced by recycle to earlier steps in the process, but some salt-beating streams need to be "purged" from the system. The volume of these streams can be reduced by evaporation, but the process must be controlled to prevent generation of mists or aerosols that may escape containment.
The solid wastes arise mainly by precipitation of insoluble salts from the oxidation effluent and by evaporation of the salt-beating filtrates remaining after precipitation. The insoluble salts can be deposited in hazardous waste landfills, but characterization and certification of the solids must take place before transportation. The soluble salts, primarily sodium and calcium chlorides, present greater difficulties because of their potential for leaching from a landfill.
HEALTH, SAFETY, AND ENVIRONMENTAL CONCERNS RELATED TO NEUTRALIZATION PROCESSES
Many concerns about the destruction of chemical agents are common to all destruction processes. The greatest risks probably arise from the on-site transportation and handling of munitions or storage containers, the disassembly and draining operations, and the handling of liquid agent before destruction. These sources of risk are shared by the baseline system and nearly all the alternatives. The differences lie in the specific destruction
processes and the character of the waste streams that result from them. For neutralization, the unique concerns center on the liquid streams and the solids derived from them.
For neutralization, air monitoring analogous to that recommended for the baseline system must be done to protect workers. Air monitoring also needs to be part of a complete system because the thermal treatment of metal parts has the potential for agent release as well as the emission of non-agent volatile organic compounds (VOCs). Non-agent VOCs from the hydrolysis and oxidation steps are expected to present minimal risks to the workers and the community.
The unique risks in neutralization center on aqueous solutions or suspensions containing incompletely neutralized agent. Spills or leaks should not pose a threat to the neighboring community if properly contained, but they would present serious problems for workers involved in cleanup. Since both VX and mustard act on or through the skin, contact with the liquids must be avoided.
A closely related source of risk is accidental discharge of solutions containing incompletely neutralized agent to the waste treatment systems for salt precipitation, filtration, and evaporation. If such discharges occur, worker exposure and equipment contamination could be severe problems. Careful monitoring of the hydrolysis and oxidation processes will be needed, and may require the development of specialized equipment to analyze aqueous solutions containing agent and its decomposition products. Some of the known hydrolysis products are toxic, and others may reform toxic substances as solution compositions change during the waste treatment processes. Research on analytical procedures and specific solution chemistry will be necessary.
The potential environmental impacts unique to neutralization-based systems also arise from the nature of liquid discharges, whether accidental or as part of routine operations. For example, it is difficult to predict the environmental effects that would occur if dilute aqueous solutions of methylphosphonic acid from hydrolysis of nerve agents were to be released as a result of leaks, spills, or failure of wastewater treatment processes. Acute toxicity to a variety of plants and animals is easily measured, but longer-term effects on ecosystems may be difficult to forecast. A closely related problem is leaching of soluble compounds (e.g., methylphosphonic acid) absorbed on solids sent to landfills for disposal. Given the substantial volume of solids generated by a neutralization process, this potential problem should be considered carefully. Again it points to the need for careful and continuous monitoring of effluents from hydrolysis and oxidation operations.
RESEARCH AND DEVELOPMENT NEEDS
Many of the research and development needs for neutralization-based systems to destroy chemical agents have been mentioned in the preceding discussion. Because these needs will be specific to the system chosen, it may be desirable to prioritize the research activities on the basis of the systems most promising for development. The criteria for selection may include factors such as readiness of a process for development and the potential capital cost savings relative to the baseline system.
Based on the "access" limitations discussed earlier, the most promising sites for use of neutralization-based processes are those that store solely VX and mustard in bulk containers. For VX, the most attractive options appear to be base-catalyzed hydrolysis followed by oxidation or, alternatively, a combined hydrolysis-oxidation in which the two processes occur simultaneously. For mustard agents, the options seem similar—base hydrolysis followed by oxidation or a combined hydrolysis-oxidation process.
Research and Development Needs Specific to Nerve Agent VX
Understand the chemical mechanism for hydrolysis of VX and establish the potential to accelerate the process by catalysis.
Select and develop a process for oxidation of the hydrolysis products of VX:
Evaluate the NaOH or NaOH-peroxide hydrolysis products as feeds for incineration.
Evaluate the product of glycol-phosphoric acid transesterification as a feed for supercritical water oxidation. The glycol ester arising from this process should give a product with low toxicity, and even though the catalyst is acidic, the ester should have low corrosivity.
As candidates for non-supercritical water oxidation, clean chemical oxidants such as ozone and hydrogen peroxide should be tested because they do not add to the waste burden from the agent destruction process.
Analytical techniques must be developed for water solutions of VX and its hydrolysis products.
Research and Development Needs Specific to Mustard Agents
Understand the process mechanisms and solubility effects critical to the hydrolysis of mustard agents by NaOH. It will be necessary to deal with partially polymerized sulfonium salts in munitions-grade mustard agent.
Evaluate the mustard hydrolysis products as feeds for biological oxidation and for incineration.
Evaluate dean oxidants (ozone, peroxide) for posttreatment of the base hydrolysis products. Do they mineralize the products? If not, are the final products suitable for biotreatment?
Develop analytical procedures for aqueous solutions containing mustard, TDE, and the oxidation products of TDE. Analyze and characterize the final oxidation products.
Generic Development Needs
Apply the preferred destruction chemistries to munitions-grade agents, which may be gelled or highly contaminated with decomposition and corrosion products. These adaptations may involve difficult engineering challenges.
Scale up the preferred processes from the laboratory to pilot scale to production scale.
Develop process control instrumentation and protocols specific to the chosen process.
Evaluate the use of hydrolysis reagents as a means to decontaminate metal to the 3X level for shipment.
Develop a system for evaporation of aqueous wastes that minimizes aerosols, mists, and corrosion.
Develop an environmentally acceptable treatment for chloride-containing waste.
Evaluate the toxicity and ecological impact of all process effluents that will (or might be) released to the environment.
Development of neutralization systems to the stage at which they are ready for full-scale use will take longer than development of an enhanced baseline system, even for neutralization processes with which we have considerable experience, such as hydrolysis of GB nerve agent. If GB hydrolysis were to be developed, all the topics listed above under generic
development needs would require study and testing. Even with concurrent studies of the various needs, the time required to implement the process on a large scale might be three years longer than for implementation of the baseline system at a given site. For processes such as oxidation with oxone or bleach, which have a sound chemical footing but little or no development beyond the laboratory, the development period might be at least a year longer. Speculative or unproven processes, such as peroxydisulfate oxidation with sulfuric acid recycle, would take significantly longer under the best of circumstances.
The extended development times required for alternatives to the baseline system will increase costs and prolong the risks associated with continued storage of chemical agents and munitions. In a few situations, however, the potential reductions in capital costs for disposal (relative to the baseline system) may justify the necessary development costs. A significant factor in proceeding with the development of a particular disposal system may be the availability of pilot facilities in which to evaluate the alternative technology.
SUMMARY OF FINDINGS ON NEUTRALIZATION PROCESSES
Neutralization processes have potential advantages over the liquid incinerator of the baseline system for the destruction of chemical agents because they operate at low temperatures that are less apt to generate troublesome by-products such as dioxins. Neutralization may also incur lower capital costs for destruction of bulk agents, but this advantage is lost in dealing with munitions that also require extensive thermal treatment for decontamination. As a consequence, neutralization-based processes seem most promising for use at sites such as Newport and Aberdeen, where only bulk liquid agents are stored.
Neutralization must be integrated into a total chemical agent and munitions disposal system in much the same way that incineration is integrated into the baseline system. Many aspects of a neutralization-based system will be similar to those of the baseline system (e.g., transport and draining of munitions or storage containers, decontamination of metal parts, and treatment of gaseous effluents).
Neutralization processes will produce smaller amounts of gaseous effluents than the baseline system, but they will involve handling large volumes of toxic liquids. Treatment of the liquid products may generate larger quantities of solids for disposal than would result from the baseline system.
Extensive research and development will be required to provide neutralization-based systems suitable for implementation even in the most favorable situations. Although there are useful leads for the neutralization of
chemical agents (especially nerve agents), development must include both laboratory and pilot studies. Demonstration studies must also include the development of new analytical and process control techniques, and the establishment of waste treatment procedures adapted to each specific neutralization process and site.
The time required to develop a neutralization-based process for use at any specific site may be three to five years longer than for the baseline system if a complete on-site system must be developed. The time requirement would be reduced if the hydrolysis products were to be disposed of by incineration at an existing incinerator. The research and development for a stand-alone system will add significantly to the cost, although the capital investment may be reduced in favorable situations. It also prolongs the risks associated with continued storage of chemical agents and munitions at a site.