2
Description of EDS Treatment Processes and Liquid Waste Streams

DESCRIPTION OF THE EXPLOSIVE DESTRUCTION SYSTEM

The Explosive Destruction System Phase 1 (EDS-1) is a trailer-mounted mobile system that is intended to destroy explosively configured chemical warfare munitions deemed unsafe to transport or store. The EDS-1 can also be used to destroy small quantities1 of recovered chemical munitions, with or without explosive components. A schematic view of the EDS-1 appears at Figure 1–1; a detailed description of the system and its operation is found in Appendix F.

The heart of the EDS-1 is a 6.5-cubic foot (189-liter) explosion containment vessel mounted on a 20-foot-long flatbed trailer. The vessel body and door are made of forged 316 stainless steel and are designed to contain detonations of up to one pound (0.45 kg) of TNT equivalent. The vessel has an inside diameter of 51 cm and is designed to handle three commonly found chemical munitions: a 75-mm artillery shell, a 4.2-inch mortar, and a Livens projectile.

The explosion containment vessel contains the explosive shock, fragments, and chemical agents during the explosive opening of the munition and also serves as a processing vessel for subsequent treatment of the chemical agent that was contained in the munition.

The Army expects to fabricate three EDS-1 units. The EDS developer, Sandia National Laboratories, is also designing and fabricating an EDS Phase 2 (EDS-2), capable of handling nonstockpile items having up to three pounds of TNT-equivalent explosives. This larger version of the EDS will be able to destroy munitions as large as 155-mm projectiles. The Army plans to construct four EDS-2 units, two for operations and two for replacement units.

EXPERIENCE WITH THE EDS-1

At this writing, there have been two campaigns in which the EDS-1 was used to destroy actual chemical munitions: one at Porton Down, United Kingdom, and one at Rocky Mountain Arsenal, Colorado.

Porton Down

The EDS-1 was shipped to a United Kingdom testing facility at Porton Down on October 11, 1999. Following setup and personnel training, pretrial activities took place in November 1999 to test the explosive accessing methods and the introduction of reagents into the EDS-1.

After completion of the preliminary testing, tests of the EDS-1 with containers and munitions filled with the chemical agents phosgene (CG), sulfur mustard (HD), and sarin (GB) took place between late November 1999 and November 9, 2000. For phosgene, 11 items were tested: four cylinders and seven 4-inch Stokes mortars. For mustard, 14 items were tested: two cylinders, seven 4.2-inch mortar rounds, and five 4.5-inch projectiles. For sarin, one steel cylinder containing 1.3 pounds of agent was tested (U.S. Army, 2000).

Following the disposal of each munition and container at Porton Down, neutralents, rinsates, and vapors were tested by the Defence Evaluation and Research Agency of the U.K. Ministry of Defence. Using gas chromatography/mass spectrometry, tests were conducted for residual GB and HD, metals, and volatile and semivolatile organic compounds. Test results and analytical procedures used were provided to the committee (U.S. Army, 2001h). The compositions of the neutralents produced from EDS disposal of two representative mustard-filled munitions are presented in columns 3 and 4 of Table 2–1; the composition of the neutralent produced from the disposal of the single GB-filled container is shown in column 3 of Table 2–2, and the neutralent compositions from a representative munition and container containing

1  

The processing rate of the EDS-1 is currently about one item every two days. Thus, it may be most appropriate for treatment of a few tens of items, rather than hundreds or thousands of items.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System 2 Description of EDS Treatment Processes and Liquid Waste Streams DESCRIPTION OF THE EXPLOSIVE DESTRUCTION SYSTEM The Explosive Destruction System Phase 1 (EDS-1) is a trailer-mounted mobile system that is intended to destroy explosively configured chemical warfare munitions deemed unsafe to transport or store. The EDS-1 can also be used to destroy small quantities1 of recovered chemical munitions, with or without explosive components. A schematic view of the EDS-1 appears at Figure 1–1; a detailed description of the system and its operation is found in Appendix F. The heart of the EDS-1 is a 6.5-cubic foot (189-liter) explosion containment vessel mounted on a 20-foot-long flatbed trailer. The vessel body and door are made of forged 316 stainless steel and are designed to contain detonations of up to one pound (0.45 kg) of TNT equivalent. The vessel has an inside diameter of 51 cm and is designed to handle three commonly found chemical munitions: a 75-mm artillery shell, a 4.2-inch mortar, and a Livens projectile. The explosion containment vessel contains the explosive shock, fragments, and chemical agents during the explosive opening of the munition and also serves as a processing vessel for subsequent treatment of the chemical agent that was contained in the munition. The Army expects to fabricate three EDS-1 units. The EDS developer, Sandia National Laboratories, is also designing and fabricating an EDS Phase 2 (EDS-2), capable of handling nonstockpile items having up to three pounds of TNT-equivalent explosives. This larger version of the EDS will be able to destroy munitions as large as 155-mm projectiles. The Army plans to construct four EDS-2 units, two for operations and two for replacement units. EXPERIENCE WITH THE EDS-1 At this writing, there have been two campaigns in which the EDS-1 was used to destroy actual chemical munitions: one at Porton Down, United Kingdom, and one at Rocky Mountain Arsenal, Colorado. Porton Down The EDS-1 was shipped to a United Kingdom testing facility at Porton Down on October 11, 1999. Following setup and personnel training, pretrial activities took place in November 1999 to test the explosive accessing methods and the introduction of reagents into the EDS-1. After completion of the preliminary testing, tests of the EDS-1 with containers and munitions filled with the chemical agents phosgene (CG), sulfur mustard (HD), and sarin (GB) took place between late November 1999 and November 9, 2000. For phosgene, 11 items were tested: four cylinders and seven 4-inch Stokes mortars. For mustard, 14 items were tested: two cylinders, seven 4.2-inch mortar rounds, and five 4.5-inch projectiles. For sarin, one steel cylinder containing 1.3 pounds of agent was tested (U.S. Army, 2000). Following the disposal of each munition and container at Porton Down, neutralents, rinsates, and vapors were tested by the Defence Evaluation and Research Agency of the U.K. Ministry of Defence. Using gas chromatography/mass spectrometry, tests were conducted for residual GB and HD, metals, and volatile and semivolatile organic compounds. Test results and analytical procedures used were provided to the committee (U.S. Army, 2001h). The compositions of the neutralents produced from EDS disposal of two representative mustard-filled munitions are presented in columns 3 and 4 of Table 2–1; the composition of the neutralent produced from the disposal of the single GB-filled container is shown in column 3 of Table 2–2, and the neutralent compositions from a representative munition and container containing 1   The processing rate of the EDS-1 is currently about one item every two days. Thus, it may be most appropriate for treatment of a few tens of items, rather than hundreds or thousands of items.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System TABLE 2–1 Composition of Mustard (HD) Neutralent Derived from Treatment with 90 Percent MEA Compound/Element MMD at DPG, HD Sample (NRC, 2001a), Bench-Scale Test EDS at PD, HD Projectile (U.S. Army, 2000a), Munition 92703 EDS at PD, HD Mortar (U.S. Army, 2000a), Munition 92704 Sulfur mustarda <50 μg/L <200 μg/Lb <200 μg/Lb 1,4-Dithiane 80–1,600 mg/Lb 113 mg/Lb 170 mg/Lb Explosives N/A N/D N/D Semivolatile organics detected 1,4-Dichlorobenzene N/R <0.6 mg/L 4.9 mg/L Pentachlorophenol N/R <0.8 mg/L 4.9 mg/L Pyridine N/R <0.7 mg/L 2.2 mg/L Volatile organics detected Benzene N/R 6.1 mg/L 11.4 mg/L Chlorobenzene N/R 178.4 mg/L 152.4 mg/L Chloroform 0.14–0.2 mg/L <1 mg/L <1 mg/L Vinyl chloride 5.8–6.9 mg/L <1 mg/L <1 mg/L Metals detected Lead N/R 9,600 μg/L 2,300 μg/L Silver N/R 15 μg/L 3.6 μg/L Selenium 3.0–3.6 ppm <0.15ppmc <0.15ppmc Arsenic 0.14–0.23 ppm <0.05 ppmc <0.05 ppmc Calcium N/R 12 mg/L 11 mg/L Mercury N/R 2.1 μg/L 0.6 μg/L Chromium 0.53–0.62 ppm 0.34 ppmc 0.32 ppmc Barium N/R 3.0 mg/L 4.4 mg/L NOTE: N/A, not applicable; N/R, not reported; and N/D, not detected. Explosives detection limits at Porton Down: 50 mg/L for HMX, RDX, TNT, tetryl; 65 mg/L for nitrocellulose; 30 mg/L for nitroglycerin. Semivolatile organics (SVOCs) analyzed at Porton Down: 1,4-dichlorobenzene, 2,4- and 2,6-dinitrotoluene, hexachlorobutadiene, nitrobenzene, o-cresol, pentachlorophenol, pyridine, 1,2,4-trichlorobenzene, 2,4,5- and 2,4,6-trichlorophenol. Detection limit on all SVOCs except pyridine was 0.6 mg/L; that for pyridine was 0.7 mg/L. Volatile organics (VOCs) analyzed at Porton Down: benzene, carbon tetrachloride, chlorobenzene, chloroform, 1,4-dichlorobenzene, 1,2-dichloroethane, 1,1-dichloroethylene, 1,1,1,2- and 1,1,2,2-tetrachloroethane, trichloroethylene, 2-butanone (methyl ethyl ketone (MEK)), and vinyl chloride. Detection limit was 1 mg/L on all VOCs. aThe higher detection limits for sulfur mustard in the EDS tests at Porton Down reflect the GC/MS analytical procedure used by the Defence Evaluation and Research Agency in the United Kingdom, which, along with contracted commercial laboratories, performed the neutralent analyses. bFor ease of comparison, the original units of μg/ml and % are converted to mg/L or μa/L. cFor ease of comparison, the original units of μg/L and mg/L have been converted to ppm (mg/kg). phosgene are shown in columns 3 and 4 of Table 2–3. These analytical results are based on samples of each neutralent. Rocky Mountain Arsenal After the completion of testing at Porton Down, the EDS-1 was shipped to the Edgewood Area of the Aberdeen Proving Ground, Maryland, on November 24, 2000, to prepare it for operational testing. Before this could take place, however, the EDS-1 was sent to Rocky Mountain Arsenal (RMA) near Denver, Colorado, on December 17, 2000, to support the emergency destruction of six M139 bomblets, each of which contained about 1.3 pounds of the nerve agent GB. These bomblets had been recovered from a scrap pile in a former parking lot and, based on their fuze design, were considered to be unsafe for transport to a storage area. Following a preoperational survey, the EDS destroyed the bomblets without incident between January 28 and February 9, 2001. The EDS was then decontaminated and returned to Aberdeen Proving Ground on March 3, 2001. Following the destruction of each bomblet at RMA, neutralent and rinsate samples were tested by the on-site Environmental Analytical Laboratory for residual GB, metals, GB decomposition products, residual explosives, and volatile and semivolatile organic compounds. Test results and analytical procedures were provided to the committee (Foster Wheeler, 2001). The composition of the neutralent for a representative munition (bomblet 3) is shown in column 4 of Table 2–2. These data are based on a single analytical sample of the neutralent. For all six bomblets, the ranges of both neutralent and water rinse components are presented in Appendix C, Table C-1.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System TABLE 2–2 Composition of Sarin (GB) Neutralent Derived from Treatment with 45 Percent MEA in Water Compound/Element MMD at DPG, GB Sample (NRC, 2001a), Bench-Scale Test EDS at PD, GB Cylinder (U.S. Army, 2000), Sample GB-DOT-01 EDS at RMA, Bomblet 3 (U.S. Army, 2001e), Munition MEAB31 Monoethanolamine (MEA) 33.9–40.3% N/R 44.9% Water 49.0–49.4% N/R 55.1% GB degradation products 0.7–8.5% N/R 4,300 ppm on 4-hr sample MEA IMPA salt (O-isopropyl methylphosphonate)   Diisopropyl methylphosphonate (DIMP) 0.03–0.36% DIMP detected 22,100 μg/L Sarin (GB) <25 ppb <100 ppba <103 ppb pH N/R 12.3 12 Explosives N/A N/D <1 mg/L Volatile organics detected Benzene 6.5–6.8 mg/L 3.3 mg/L 2.64 mg/L Chlorobenzene N/R 1.1 mg/L N/R Dichloromethane N/R N/R 97.1 μg/L Toluene N/R N/R 810 μg/L Metals detected Lead 550–1,300 ppbb 96 ppba 237 ppba Silver N/R <2.0 μg/L <174 μg/L Selenium N/R <150 μg/L N/R Arsenic 660–760 ppbb <50 ppba <200 ppba Calcium N/R 1.6 mg/L N/R Mercury N/R 0.4 μg/L <0.5 μg/L Chromium 410–1,080 ppbb 450 ppba 703 ppba Barium N/D-750 ppbb 3,000 ppba <44.7 ppba Aluminum 76–81.5 ppm N/R 9.35 ppmc Cadmium N/R N/R <6.81 μg/L Zinc N/R N/R 30,000 μg/L Copper N/R N/R 18,200 μg/L NOTE: The RMA results are on waste drum 1 (MEAB31) or, for the IMPA analysis, on the 4-hr sample from bomblet 3 (2001–02–02). Analyses for bomblet 3 wastes were chosen as being the most comprehensive. N/A, not applicable; N/R, not reported; and N/D, not detected. Explosives detection limits at Porton Down: 50 mg/L for HMX, RDX, TNT, tetryl; 65 mg/L for nitrocellulose; 30 mg/L for nitroglycerin. Semivolatile organics (SVOCs) analyzed at Porton Down: 1,4-dichlorobenzene, 2,4- and 2,6-dinitrotoluene, hexachlorobutadiene, nitrobenzene, o-cresol, pentachlorophenol, pyridine, 1,2,4-trichlorobenzene, 2,4,5- and 2,4,6-trichlorophenol. Detection limit on all SVOCs except pyridine was 0.6 mg/L; that for pyridine was 0.7 mg/L. Volatile organics (VOCs) analyzed at Porton Down: benzene, carbon tetrachloride, chlorobenzene, chloroform, 1,4-dichlorobenzene, 1,2-dichloroethane, 1,1-dichloroethylene, 1,1,1,2-and 1,1,2,2-tetrachloroethane, trichloroethylene, 2-butanone (methyl ethyl ketone (MEK)), and vinyl chloride. Detection limit was 1 mg/L on all VOCs. aFor ease of comparison, the original units of g/ml, μg/L, and mg/L have been converted to ppb (μg/kg). bAs contained in the Utah MMD permit, available at <http://www.eq.state.ut.us/eqshw/cds/MMDPermit.htm>. RCRA toxicity-characteristic component concentration is less than TCLP regulatory limit. cFor ease of comparison, the original units of μg/L have been converted to ppm.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System TABLE 2–3 Composition of Phosgene (CG) Neutralent Derived from Aqueous Caustic Treatment Compound/Element MMD at DPG CG Sample (NRC, 2001a), Bench-Scale Test EDS at PD CG Mortar (U.S. Army, 2000), Munition 1051905 EDS at PD CG Cylinder (U.S. Army, 2000), Munition CG-MOCK-2 Sodium hydroxide content or pH 8–9% pH 13–14 pH 13–14 Explosives N/A N/D N/D Semivolatile organics N/R N/D N/D Volatile organics N/R N/D N/D Metals detected Lead N/R 2,100 μg/L 280 μg/L Silver N/R 19 μg/L 13 μg/L Selenium N/R <150 μg/L 2,200 μg/L Arsenic N/R 0.42 mg/L <150 mg/L Calcium N/R 7.2 mg/L 23 mg/L Mercury N/R 1.5 μg/L 0.9 μg/L Chromium N/R <0.05 mg/L 5.3 mg/L Barium N/R 11.5 mg/L 8.5 mg/L NOTE: N/A, not applicable; N/R, not reported; and N/D, not detected. Explosives detection limits at Porton Down: 50 mg/L for HMX, RDX, TNT, tetryl; 65 mg/L for nitrocellulose; 30 mg/L for nitroglycerin. Semivolatile organics (SVOCs) analyzed at Porton Down: 1,4-dichlorobenzene, 2,4- and 2,6-dinitrotoluene, hexachlorobutadiene, nitrobenzene, o-cresol, pentachlorophenol, pyridine, 1,2,4-trichlorobenzene, 2,4,5- and 2,4,6-trichlorophenol. Detection limit on all SVOCs except pyridine was 0.6 mg/L; that for pyridine was 0.7 mg/L. Volatile organics (VOCs) analyzed at Porton Down: benzene, carbon tetrachloride, chlorobenzene, chloroform, 1,4-dichlorobenzene, 1,2-dichloroethane, 1,1-dichloroethylene, 1,1,1,2- and 1,1,2,2-tetrachloroethane, trichloroethylene, 2-butanone (methyl ethyl ketone (MEK)), and vinyl chloride. Detection limit was 1 mg/L on all VOCs. OPERATIONAL PROCESSES OF THE EDS-1 Predetonation Procedure Before the munition is placed in the EDS containment vessel, it is surrounded by a fragment suppression system (FSS) (see Appendix F) to protect the walls of the EDS containment vessel from high-velocity fragments of the detonation. The FSS is also used to mount and properly locate the explosive charges used to access the chemical fill and detonate the fuze and burster of the munition. A linear shaped charge cuts open the munition and exposes its contents for chemical treatment, and conical shaped charges detonate the internal energetic compounds. The shaped charges are composed of RDX or Composition A, an RDX/polyethylene composite. Following the preparation of the sealing surface of the EDS-1 vessel door and installation of a new o-ring, the door is closed and a leak test is conducted. If no leaks are detected, detonation is initiated. Although the detonation produces high temperature and pressure transients, the reactions with the neutralizing reagents take place at 25° to 60°C and pressures slightly above ambient. The explosion of the RDX linear shaped charge and the Composition A-3 conical shaped charge was estimated by Army experts to be 99 percent complete (U.S. Army, 2001h, enclosure 1). In only one of the RMA tests was any intact RDX found in a solid residue. In all the RMA tests, the liquid effluent contained less than 1,000 ppm of RDX or other explosives. Although the explosion may destroy some of the chemical agent contained in the munition, the agent destruction products appear to arise primarily from the subsequent neutralization process. The concentration of IMPA, the primary GB hydrolysis product, in the neutralent corresponds to between 70 and 90 percent of the agent originally present (U.S. Army, 2001c). Other hydrolysis products were also present, suggesting that, at least with GB, most of the agent survives the blast and is destroyed by the neutralizing reagent. Postdetonation Treatments and Liquid Waste Streams After the contents of the munition have been released into the EDS vessel, a neutralizing reagent is introduced to destroy the chemical agent and any remaining energetics. The reagents used in the EDS-1 to neutralize three of the most common chemical munition fills—sulfur mustard, phosgene, and sarin—are shown in Table 2–4.2 2   A complete list of proposed reagents for various chemical agents is provided in U.S. Army, 2001e, table 3–1.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System TABLE 2–4 EDS Treatment Solutions and Liquid Waste Handling Chemical Fill Solution and Waste Disposition Chemical Treatment (Source of Neutralent)a Second Treatment (Source of Rinsate)b Rinses (Source of Rinsate)c End-of-Run Cleaning Solutiond End-of-Campaign Cleaning Solutione Mustard (HD) Composition 90% MEA/H2O (60°C) Water (100°C) Water (ambient) Water/detergent (ambient) H2O/vinegar Collection Drum 1 Drum 2 Drum2 Drum 3 Drum 3 Phosgene (CG) Composition 20–22% NaOH/H2O (ambient) Not needed Water (ambient) H2O/detergent H2O/vinegar Collection Drum 1   Drum 2 Drum 3 Drum 3 Sarin (GB)c Composition 45% MEA/H2O (25°C) Water (100°C) Water (ambient) H2O/detergent H2O/vinegar Collection Drum 1 Drum 2 Drum 2 Drum 3 Drum 3 Lewisite (L)f Composition 20% NaOH (ambient) NA Water (ambient) H2O/detergent H2O/vinegar Collection Drum 1 Drum 2 Drum 2 Drum 3 Drum 3 NOTE: There are no known nonstockpile munitions containing VX. Were such munitions to be found, the reagent is expected to be a combination of MEA and aqueous NaOH. aTemperatures shown in parentheses are vessel temperatures. bInitial tests at Porton Down included a second chemical treatment with reagent, but this was discontinued after munition 1061602. cInitial tests at RMA included rinses with MEA/water solution, but this was discontinued after bomblet 2. dDetergent is used periodically to remove lubricant from metal door seal. eUnited Kingdom only: denatured alcohol, potassium hydroxide, bleach, and water. RMA: vinegar (acetic acid) and water are used for line rinsing periodically. Water, bleach, and detergent are used for hardware cleaning. fLewisite has not yet been treated in the EDS at this writing. SOURCE: Michael L.Duggan, Group Leader, Mobile Systems Acquisition, “Explosive Destruction System (EDS),” presentation to the committee, June 15, 1999; U.S. Army 2001d; U.S. Army 2001e; U.S. Army 2001f. Based on current and expected operation of the EDS, four types of liquid wastes are generated: neutralent, consisting of the initial treatment of agent with active reagent (e.g., MEA) and any subsequent chamber washes with chemical reagent (if used) rinsates, consisting of additional agent treatment with water and chamber washes with water after opening the EDS cleaning solutions, consisting of washes (water/detergent) that are made between processing of each munition, and final washes (e.g., water/vinegar) made after completing a munitions campaign final washes, using, for example, water and acetic acid, carried out after completing a munitions campaign The expected source and handling scheme for these wastes is presented in Table 2–4. The term “treatment” is used to describe steps involving addition of reagent or water to the EDS and agitation for a certain period prior to opening the chamber. Note that water treatment and rinse water wastes are combined (rinsate). To date, the Army has chosen to segregate the four types of waste, in order to keep open the option of a separate treatment strategy for the waste streams that are primarily water.3 For treatment of the residues from an HD-containing munition, a monoethanolamine (MEA) solution (90 percent MEA, 10 percent water) is heated to 60°C. The EDS chamber is at 50°C at the time of the addition. After the MEA is added, the mixture is agitated by rocking (EDS-1) the containment vessel (the EDS-2 design envisages rotating the containment vessel),4 and the temperature is maintained at 3   The treatment neutralent is transferred to a 55-gallon drum. The subsequent water rinsates are transferred to a second drum, and cleanup wash water is discharged to a third drum (see also Figure 1–1). 4   Rotation of the EDS-2 containment vessel should improve mixing of the released chemical agent with the neutralizing reagents and thus could reduce the munition processing time, compared with the EDS-1.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System 60°C for 4 hours. Samples of the liquid in the vessel are withdrawn periodically until the HD concentration is less than 50 ppm, the current detection limit in HD neutralent solutions that may contain compounds that interfere with the detection of HD. When the treatment is complete, the vessel liquid contents are drained from the vessel and transferred to a storage drum. The vessel is then treated with 100°C water to decompose small amounts of mustard occluded in solid residues as well as to remove residual MEA and reaction products. After repeated rinsing with water, the vessel is flushed with helium or other inert gas to remove residual HD vapor that may be present in the headspace. The helium and any agent vapor are passed through a carbon filter to capture the agent. The helium is then released to the atmosphere. Before the vessel is opened, a gas sample is bled off into a test bag and tested. If there is a positive reading for agent, additional treatment will take place. Several detectors are available that can be set to detect various agent types; the analysis takes about 6 minutes. The vessel is then opened to remove any solid residues. The neutralent, the 100°C posttreatment water/rinse water, and the solids are stored in separate waste drums. Similar procedures are used to neutralize other chemical agents, but the treatment cycle and the neutralizing reagent are specific to each type of chemical agent. For treatment of the GB-containing bomblets at RMA, the EDS vessel was maintained at 25°C and oscillated during reaction with a 45 percent MEA solution in water. A neutralent sample was collected after 2 hours and was analyzed for agent concentration. Additional samples were taken every 2 hours until the GB concentration dropped below the detection limit of 1 mg/L (1 ppm). When this end point was reached, the neutralent was transferred to a container for liquid waste. The vessel was then rinsed with 65 L water at 40°C to 50°C (U.S. Army, 2001d). Appendix C describes the neutralent, rinsate, and cleaning solution waste streams that resulted from destruction of the GB bomblets in the EDS. The reagents used to treat the postdetonation materials contained within the EDS chamber are essentially identical to those used in the MMD (NRC, 2001a). Like those used in the MMD, the EDS reagents produce neutralents and organic-contaminated water washes that require posttreatment before ultimate disposal. The aqueous EDS reagents, listed in Table 2–4, differ substantially from the nonaqueous reagents used in the RRS system to destroy CAIS items (see NRC, 2001a). RRS reagents and procedures are not appropriate for the munitions to be processed in the EDS and are not discussed in this report. EDS LIQUID WASTE VOLUME AND COMPOSITION DATA Useful data on the composition and quantity of EDS waste streams have come from the tests at Porton Down (PD) (U.S. Army, 2000) and the emergency destruction campaign at TABLE 2–5 Summary of Liquid Wastes from EDS Tests to Date   Liquid Wastes Test Site/Agent Munitions Destroyed Average Volume/Test (gal)a Agent Content (μg/ml)b PD/CG 4 cylinders, 7 mortar rounds 53 <1c PD/HD 2 cylinders, 12 mortar rounds or projectiles 61 <0.2 PD/GB 1 cylinder, 1.3 lb GB 67 <0.1 RMA/GB 6 GB-filled bomblets, 1.3 lb GB in each 74 <0.1 aThe volume reported is an average and includes neutralent, a postreaction wash, and rinse liquid. The postreaction wash was either dilute reagent or water; subsequent rinses used water. The liquids were collected and stored separately. The volume reported does not include the cleaning solution to clean the EDS after test completion. bThe concentration of CWM in the neutralent after 4 hours reaction was below the analytical detection limit. The detection limits of <1 ppm (1 μg/ ml~1 ppm for dilute aqueous solutions) for CG and <0.2 ppm for HD were safely lower than the treatment goal of <50 ppm for these two agents (U.S. Army, 2000a). Likewise, the detection limit of <0.1 ppm for GB was adequately lower than the treatment goal of <1 ppm (U.S. Army, 2001a). cThe liquid waste was analyzed for phosgene in only one test. The vapor phase often showed low levels of phosgene; original units of mg/L changed to μg/ml. SOURCE: Compiled by the NRC from Army sources. RMA (U.S. Army, 2001d). Table 2–5 summarizes the amount of liquid waste produced in test experiments and the effectiveness of the neutralization as judged by the amount of residual agent in the liquid effluents produced. More detail on the EDS-1 waste compositions is provided in Tables 2–1 to 2–3, along with the available data on the compositions of MMD neutralents.5 MMD and EDS neutralent component listings are not comparable because different sets of constituents were analyzed, the analysis detection limits were different, and the magnitude of operations/tests differed. As noted in Tables 2–1, 2–2, and 2–3, the analytical data are for neutralents from specific munition items. The ranges of analyses for neutralents from the six GB bomblets treated at RMA are shown in Table C-1. However, the data 5   The MMD neutralent analysis data come from bench-scale tests conducted at Dugway Proving Ground (DPG), Utah, rather than from actual operation of the MMD.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System in the tables permit concluding that EDS and MMD neutralents are similar and should be treatable by the same posttreatment technologies. This having been said, the difference in analytical methods means that this conclusion must be stated in a qualitative rather than a quantitative way. In the destruction of the six bomblets at RMA, a total of 196 gallons of neutralent were produced along with 248 gallons of rinse water (U.S. Army, 2001f). The compositions of the GB neutralents, the water rinses, and the liquids resulting from cleaning the EDS transfer lines between operations are tabulated in Appendix C. It appears that the transfer lines were cleaned with aqueous acetic acid between campaigns and with aqueous detergent between treatment of individual rounds. The used cleaning solutions were combined for disposal.6 As noted in Table C-1 of Appendix C, surprisingly high levels of chloroform are noted in the used cleaning solutions from cleaning the EDS after a munition has been processed. The concentrations are sufficiently high to raise a question about the suitability of this aqueous waste stream for processing in a FOTW facility. (This waste stream from the RMA operation was sent to a commercial TSDF for disposal.) That the concentration of chloroform in the cleaning solution is higher than in the neutralent or rinsate suggests that it arises from the cleaning process rather than from the detonation or the neutralization processes. The Army supposes that it comes from the commercial lubricant PermaSlik, which is used to lubricate and seal the door seals on the EDS.6 The lubricant may be washed off the door seal during the detergent rinse of the apparatus after treatment of each munition. This explanation seems likely because the lubricant was formulated from a mixture of organic solvents including trichloroethane. Newer formulations substitute ethyl acetate for the trichloroethane and may not present this problem. Comparison of EDS and MMD Liquid Waste Compositions The MMD and the EDS neutralents are similar. The phosgene neutralents, which are alkaline, aqueous solutions of inorganic salts (NaOH, NaCl, Na2CO), are nonflammable, single-phase liquids containing small quantities of suspended solids. The MMD and EDS neutralents derived from HD and GB are also single-phase liquids over the range of 5:1 to 200:1 MEA reagent:CW agent (Lucille Forrest and James Horton, personal communication, October 1, 2001). However, most of the information on physical properties comes from studies on laboratory simulations of the neutralization process (U.S. Army, 1997). More data from actual EDS test neutralents would be valuable. Differences may arise from the contrasting types of munitions processed in the two systems and the divergent ways they are processed. The MMD system does not process items containing explosives, while the EDS can handle munitions containing bursters and/or fuzes. The EDS system introduces the explosives RDX and Composition A to open the munition and to detonate any explosives contained therein. The EDS waste analyses did not show residual explosives except for RMA bomblet 1. The solid residue from this device contained 18.3 ppm of RDX, which presumably resulted from incomplete detonation of the shaped charges used to open the bomblet. The absence of RDX (detection limit ca. 1 ppm) in the neutralent from other bomblets probably reflects the ability of MEA to destroy RDX. The Army has stated as follows: “Both MEA and aqueous alkali have been shown to efficiently destroy TNT and RDX” (U.S. Army, 2000). Apparently no analyses were performed to detect RDX-MEA reaction products, which may be complex organic amines. The explosives present in the munition7 could also produce a variety of compounds in the EDS-1 liquid waste streams that may need to be removed or destroyed by a subsequent treatment technology.8 If the munition fuze and burster charges are successfully detonated by the shaped charges, there may be little residual explosive remaining after the blast. If, on the other hand, the fuze and burster charges in the munition are not successfully detonated by the shaped charges, as may occur especially in older munitions,9 the energetics are likely to be hydrolyzed to a considerable extent by the MEA-water or sodium hydroxide-water reagent (NRC, 1999a, Appendix E).10 Even if there is little residual explosive remaining in the neutralent, hydrolysis products from the energetics will still be present in the liquid waste streams. Another difference between the MMD and EDS wastes may be the presence of higher levels of metal ions and metal oxides of Fe, Co, Ni, Pb, Al, Zn, Sn, and Cu in some EDS neutralents. The fragmentation of the munition bodies in the EDS, as well as the hours of rocking back and forth during the neutralization process, may expose more metal surface to the MEA reagent, which is a fairly good extractant for these metal ions. In addition, heavy metal constituents of the munition fuze (e.g., mercury from mercury fulminate) will 6   Lucille Forrest, Office of the Project Manager, Non-Stockpile Chemical Materiel, “Re: Cleaning Solutions for EDS,” memorandum to the committee, May 10, 2001. 7   Fuzes were made from a variety of primary explosives, such as mercury fulminate. The most common explosives used in burster charges are tetryl, TNT, and Composition B (a mixture of TNT and RDX). 8   Various compounds that may be present can be found in Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons, Appendix E (NRC, 1999a). 9   In the Porton Down tests, the fuze or burster in the munition failed to detonate in several cases. 10   The committee did not receive information on procedures that might be followed if detonation of the shaped charges results in cracking of the munitions casing but fails to expose the chemical agent fully to the neutralizing solution.

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System remain in the vessel and may well dissolve in the neutralizing reagents. Unfortunately, as indicated in Tables 2–1 to 2–2, an adequate comparative analysis for these metals, which are likely materials of construction in various munitions, is not currently available. In general, the EDS and MMD neutralents are similar apart from the concentrations of metal ions and explosives-derived organic compounds. The latter are likely to be destroyed by posttreatments that destroy MEA, because they are less stable chemically. The metal components may require stabilization or solidification before final disposal, especially if toxic metals such as lead or mercury are involved. The toxicities of the MMD and EDS-1 neutralents are discussed and compared in Appendix D. POTENTIAL FOR ENCOUNTERING UNUSUAL CHEMICAL SPECIES IN THE EDS Munitions Containing Arsenic and Other Toxic Metals The preceding discussion of munitions to be treated in the EDS covers those most likely to occur in Non-Stockpile Chemical Warfare Materiel (NSCWM) “finds.” However, some munitions containing arsenic-derived chemical agents may also be encountered. Past experience indicates that there may be projectiles filled with lewisite, an organoarsenic blister agent. Some old German traktör rockets stored at Pine Bluff Arsenal appear to contain Winterlost, a mustard formulation containing phenylarsenic dichloride (PD) and diphenylarsenic chloride (DA) as freezing-point depressants (Martens, 1998). There may also be items containing adamsite (DM), an arsenic-containing vomiting agent intended for riot control. It is uncertain whether the EDS will be used to destroy munitions containing arsenic. If so, the neutralizing reagent will need to be tailored to the anticipated chemical agent because their neutralization chemistry is quite diverse. The Army has proposed neutralizing agents and treatment conditions for the agents reasonably expected to be treated in the EDS (U.S. Army, 2001e). Alkaline neutralization like that used for the destruction of phosgene is planned for the destruction of lewisite in the EDS (U.S. Army, 2001e). The agent from an opened munition is to be treated with 22 percent aqueous sodium hydroxide at ambient temperature for 1 hour. Hydrolysis with hot aqueous sodium hydroxide solution forms the basis for neutralization of bulk lewisite in a new process to be implemented at Gorny in the Saratov region of Russia (Petrov et al., 1998; N.N.Kovalyev, Monterey-Moscow Study Group on Russian Chemical Disarmament, personal communication to G.W.Parshall, member of the committee, February 25, 1997). The alkaline treatment destroys the lewisite, forming acetylene gas, sodium chloride, and sodium arsenite. The arsenite may be oxidized to arsenate to facilitate disposal. In a related process, to be implemented at the Chemical Agent Munitions Disposal System (CAMDS) in Utah (Utah, 1998), the lewisite is first oxidized with aqueous hydrogen peroxide, then hydrolyzed with sodium hydroxide solution. The resulting sodium arsenate solution will be solidified with cement for disposal in a hazardous waste landfill. The solidification process is being studied under an NSCMP test (Edward Doyle, Office of the Project Manager, Non-Stockpile Chemical Materiel, communication to the committee, March 16, 2001). As practiced in the United States and some European countries, the sodium arsenate could also be converted to ferric arsenate, a less soluble salt that is resistant to leaching by groundwater. The separation of arsenate can be accomplished by precipitation with ferric chloride to form ferric arsenate, which is insoluble in either cold or hot water (Hodgman, 1963). The CAMDS process is based on the process used to destroy Canada’s stockpile of lewisite in Project Swiftsure (McAndless et al., 1992). Based on the Draft Final Standard Operating Procedure for the EDS (U.S. Army, 2001e), the mixture of sulfur mustard and phenylarsenic chlorides contained in Winterlost is likely to be treated according to the same protocol as HD alone, i.e., 90 percent MEA, 10 percent water, at 60°C for 4 hours. While MEA will break the As-Cl bonds associated with the vesicant activity of DA and PD, it is unclear whether the treatment will break the phenyl-As bonds. Further posttreatment may be needed to prepare the neutralent for disposal. Much the same situation arises with the proposed neutralization of DA and PD in the absence of HD. The Army plans to treat the agent and munition residues in the EDS with denatured 95 percent ethanol (or possibly acetone for DA) at ambient temperature for 1 hour. While this treatment will dissolve the organoarsenic compounds and partially hydrolyze the As-Cl bonds, it will produce a neutralent requiring significant further treatment. Adamsite and the phenylarsenic chlorides are customarily hydrolyzed with acid rather than alkali (Wertejuk et al., 1998). Treating these organoarsenic compounds with hot concentrated hydrochloric acid destroys the compounds, but this reagent may be too corrosive for use in the current EDS system. Alkaline reagents like aqueous sodium hydroxide hydrolyze the As-Cl chemical bonds in these compounds, thus reducing their toxicity, but do not necessarily destroy the organoarsenic structures. For the small number of munitions containing these agents, it may be most appropriate to treat the detonation products with aqueous sodium hydroxide reagent and then destroy the neutralent in a treatment unit or facility capable of substantially reducing the toxicity, mobility, and volume of the arsenic wastes. At present, the only commercial facilities available are commercial incineration facilities that are permitted to treat wastes containing arsenic. The requirement that the initial arsenic-containing products be oxidized to arsenate salts gives an advantage to neutralent treatment processes that have an oxidative character. This class includes chemical oxidation, electrochemi-

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System cal oxidation, wet air oxidation, supercritical water oxidation, and plasma processes, which incorporate a secondary oxidizer (see Chapter 3). Reductive processes such as solvated electron technology and gas-phase chemical reduction can be adapted to produce arsenate via posttreatment oxidation, but at substantial additional cost and complexity. Biochemical processes are generally unsuitable for treating wastes containing arsenic, lead, and mercury because these elements deactivate the enzymes needed to degrade organic wastes. POTENTIAL FOR ENCOUNTERING OTHER UNUSUAL CHEMICAL COMPOUNDS To date, the Army has tested protocols for treating munitions containing only sulfur mustard, phosgene, and sarin in the EDS, but treatment regimes for all the anticipated arsenic-containing agents have been established based on known chemistry (U.S. Army, 2001e). Chemistry to stabilize and solidify arsenic-containing treatment wastes is being tested in a PMNSCM-funded R&D program (Edward Doyle, Office of the Project Manager, Non-Stockpile Chemical Materiel, communication to the committee, March 16, 2001). Given that many agents11 and toxic industrial chemicals12 have been used to fill chemical weapons over the years, it is possible that other fills, including the nerve agent VX, will be encountered. However, there appears to be little information with which to estimate the probability of encountering these other fills. Apart from the variety of possible agent fills, there are several other reasons why unusual compounds might be present in recovered munitions. Special formulations of agents and industrial chemicals were sometimes used to achieve certain effects. For instance, tin tetrachloride was encountered in phosgene rounds treated in the Porton Down tests of the EDS-1. This chemical was added to facilitate the penetration of gas masks and to produce a smoke that aided in spotting where rounds had landed. Chlorobenzene, possibly used as a solvent or stabilizer, was found in the mustard rounds processed at Porton Down (Table 2–1). Chlorinated rubber was used as a thickener in some mustard formulations. In addition, unusual compounds or sludges may result from chemical reactions such as corrosion and polymerization that may occur among the components over a period of decades. DATA GAPS AND UNCERTAINTIES During the Porton Down and RMA experiences with the EDS-1, the Army was still refining its treatment procedures. For example, the number and composition of rinses following the initial neutralization step varied from munition to munition, even among munitions of the same type, as at RMA. This led not only to different volumes of neutralents and rinsates generated for each munition but also to different concentrations of organics and metals in the neutralents and rinsates. These variations could affect both the choice of treatment technology and the regulatory status of the liquid wastes (see Chapter 4). As more experience is gained with treatment of a variety of munitions and agents in the EDS-1, these variations should grow smaller. Beyond the variations introduced before EDS treatment protocols become fully standardized, a number of factors will inevitably cause the composition of the neutralents and rinsates to vary from munition to munition. These include the exact composition of the chemical agents inside the munition, the age of the munition, and factors relating to the composition and status of the energetics in the munition. If, as discussed above, the fuze and/or burster in the munition are not successfully detonated by the conical shaped charges, hydrolysis products of energetics may be present in substantial concentrations in the neutralent. The neutralent chemical analysis data provided to the committee contained information on residual energetics (in general, they were not detected) but not on hydrolysis products of energetics. A potential concern relates to the possibility that residual chemical agent may inhere in the microscopic cracks and crevices of suspended solids that are inaccessible to the neutralizing reagent. The presence of such occluded agent residues was observed in early efforts to develop chemical decontamination methods for large metal parts in the U.S. Chemical Stockpile Destruction Program (Richard Magee, former chairman of the NRC Stockpile Committee, personal communication, September 7, 2001). The EDS neutralent is known to contain suspended solids, as evidenced by the fact that it is cloudy. To date, the Army has not done any tests to determine the quantity of suspended solids or whether any residual chemical agent might inhere in them, reasoning that these solids remain in contact with the neutralizing solution until the waste drums are processed through the incinerator and that any agent that might escape from a microscopic crack would immediately be neutralized. Nevertheless, if such residual agent is present in high enough concentrations in suspended solids, it is conceivable that exposures could occur to workers handling the waste drums or in the event of a spill. This potential concern could be easily resolved by a test involving filtration of the neutralent followed by heating the filtered solids and testing the off-gas for the presence of agent. The foregoing discussion shows that the composition of the neutralents and rinsates derived from the processing of 11   These include tabun (GA), sarin (GB), Levinstein mustard (H/HS), distilled mustard (HD), HT, nitrogen mustard (HN), lewisite (L), VX, and adamsite (DM). 12   These include hydrogen cyanide, bromoacetone, bromobenzyl cyanide, phosgene, cyanogen chloride, chlorine, chloroacetophenone, chloropicrin, and diphenylchloroarsine (DA).

OCR for page 12
Evaluation of Alternative Technologies for Disposal of Liquid Wastes from the Explosive Destruction System nonstockpile items in the EDS is both highly variable and unpredictable. In the future, it will be important to analyze the neutralents for all chemical species that could be of health, safety, environmental, or regulatory concern. In addition, any nonincineration posttreatment technologies used by the Army to dispose of these liquid wastes must be able to destroy a wide variety of chemical species; i.e., they must be robust. This criterion, among others, is considered by the committee in its evaluation of the alternative secondary waste treatment technologies discussed in the next chapter.