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 8
A Modified Baseline Incineration Process for Mustard Projectiles at Pueblo Chemical Depot 2 Experience at JACADS with Mustard Munitions BASELINE INCINERATION SYSTEM The baseline incineration system, for which JACADS provided the prototype, is thoroughly described in the 1994 NRC report Recommendations for the Disposal of Chemical Agents and Munitions (NRC, 1994). The baseline incineration system was designed to separate, process, prepare, and dispose of four agent and munitions waste streams: agent (liquid incinerator [LIC]), energetics (deactivation furnace system [DFS]-rotary kiln), contaminated metal components (metal parts furnace [MPF]), and packaging and other materials (dunnage furnace). Neither JACADS nor TOCDF, the two baseline facilities that have processed both agent stored in bulk (ton) containers and chemical munitions, was built and operated exactly according to the original design. JACADS, the first baseline facility to become operational, was also the first to encounter unanticipated problems, the majority of which were solved in ways not envisioned in the original design of the system. For example, although a dunnage furnace at JACADS was included in the facility design, constructed, and had a successful trial burn, it has not been used in recent years. Prior to full-scale JACADS operation, trial burns were conducted to confirm that the destruction and removal efficiency (DRE) for agents of all of the baseline system furnaces met the criteria of the Environmental Protection Agency (EPA). Because the processing of chemical agent and chemical munitions was unprecedented, and because the process failures could have catastrophic effects on personnel and the environment, Congress mandated that the Army conduct an operational verification test program to confirm that operations were safe before allowing construction of baseline facilities to begin in the continental United States (U.S. Army, 1992). Table 2–1 summarizes test conditions and results from the four 4-hour test runs conducted in August 1992 on ton containers in the MPF (U.S. Army, 1992). Table 2–2 shows the results of trial burns for metal emissions (U.S. Army, 1992). As Table 2–1 shows, the treatment of the mustard agent (HD) resulted in stack concentrations, DRE, operating temperatures, carbon monoxide concentrations, stack particulate concentrations, and hydrogen chloride emissions that were all within required limits for all four test runs. At the time the trial burn report was prepared, no limits had been established for metals. However, the report notes that the measured concentrations were very close to the detection limits in all cases (U.S. Army, 1992). It is not clear whether the low concentrations were due to low metals content in the agent stream or to the effective removal of metals during processing. Disposal campaigns for HD ton containers and three types of projectiles containing HD were completed at JACADS in mid-1999, at which time all mustard agent stored on Johnston Island had been destroyed. Table 2–3 shows the total number of HD items processed. There were no HT mustard-filled containers or munitions at JACADS. While HD and HT are very similar, and the results of their combustion could be expected to be much the same, the committee has no evidence of analysis or testing to this effect. IMPROVED PROCESSING OF HD MUNITIONS Multipurpose demilitarization machines (MDMs) are used in the baseline system to extract the press-fit burster well from the projectile body and drain agent from the munition. During the processing of a significant number of the HD projectiles at JACADS listed in Table 2–3, the agent foamed and overflowed the projectile casing when the burster casing was ruptured or removed (U.S. Army, 2000a). The MDM then had to be shut down and the mustard agent, which is corrosive, had to be decontaminated with a neutralizing solution; then the MDM had to be cleaned and repaired. Maintenance personnel, who performed this work in demilitarization protective ensemble (DPE) suits, thus generated large amounts of spent decontamination solution (SDS) and contaminated DPE suits. The additional maintenance reduced
OCR for page 9
A Modified Baseline Incineration Process for Mustard Projectiles at Pueblo Chemical Depot TABLE 2–1 Summary of 1992 Trial Burn Tests for the Treatment of HD Ton Containers in the MPF at JACADS Parameter Units Test Run 1 Test Run 2 Test Run 3 Test Run 4 Requirement Average quantity of agent per container lb 12.8 51.8 58.4 60.3 HD evaporation rate (maximum) lb/hr 54 114 111 105 ≤146.2 Concentration of HD in stack mg/m3 NDa ND ND ND <0.03 DRE for HD % >99.9996 >99.9997 >99.9997 >99.9997 ≥99.99 Operating temperature in primary chamber °F First zone 1,408 1,425 1,418 1,457 1,450±250b Second zone 1,456 1,453 1,460 1,461 1,450±250 Third zone 1,459 1,449 1,449 1,451 1,450±250 Afterburner temperature °F 2,003 2,003 2,003 1,998 2,000+250 or −100 Afterburner CO output (corrected to 7% O2) ppm 9.0 9.0 1.7 1.1 (Average PAS one-hour rolling average) 13.0 12.7 2.0 2.1 ≤100c Particulate concentration in stackd mg/dscme Corrected to 7% O2 10.92 2.68 3.13 0.89 ≤180 Corrected to 12% O2 13.35 3.29 3.79 1.03 HCl emissions lb/hr NDf 0.0497 ND ND ≤4 Stack gas flowd acfmg 11,338 11,918 11,801 11,772 dscfmh 6,002 6,333 6,150 6,121 Oxygen (Orsat)d % 14.5 14.8 14.5 14.2 Carbon dioxide (Orsat)d % 4.6 4.3 4.5 4.9 aND=none detected. Detection limits were 0.0054, 0.0066, 0.0069, 0.0066 mg/m3 for test runs 1, 2, 3, and 4, respectively. bThe permit requirement of 1,450±150°F was revised to 1,450±250°F by memorandum of understanding. c100 ppm is the hourly rolling average limit. The peak limit for five minutes is 200 ppm. Both corrected to 7% oxygen. dAverage of MMT (multimetals train) and M5AT (method 5 acid train) particulate results. eDry standard cubic meter. fND=none detected. Detection limits were 0.00326, 0.00277, 0.00328 lb/hr for test runs 1, 3, and 4, respectively. gActual cubic feet per minute. hDry standard cubic feet per minute. Source: Adapted from U.S. Army, 1992. processing rates, thus increasing overall facility costs. An innovative solution to the frothing problem was developed by freezing the agent in the projectile before opening the agent cavity. About 200 rounds were successfully processed in this manner (Tomanek, 2000a, 2000b). During the processing of munitions at JACADS, operators also noted that a high percentage of HD-filled projectiles, especially 4.2-inch mortar rounds, did not conform to the design criteria of the baseline system, which specified that agent be in a liquid state and that 95 percent of the agent fill be removed. In fact, data showed that on average only 60 percent of the agent fill was removed by the MDMs (SAIC, 1998). Even at this reduced rate, plugging of the drain system and interruptions in processing occurred, and the removal rate had to be reduced to 50 percent to avoid plugging of the drain system (U.S. Army, 2000b). This resulted in a higher-than-expected agent load to the MPF for a significant number of projectiles. Because the RCRA permit was based on agent loading rather than on the number of projectiles processed, the processing rate was severely curtailed (EPA, 1998). In the absence of experience with chemical agent disposal operations, the original permit issued by EPA for operating the MPF had set a feed rate based on the schedule established for accomplishing the JACADS disposal mission, rather than on the capacity of the MPF to destroy agent feeds to the required DRE. With EPA cooperation, a new trial burn was undertaken in 1999 to support a RCRA permit modification that would allow the processing of batches of 96 mortar projectiles, which had energetics and burster wells removed and were completely filled with agent, through the three zones of the MPF at JACADS. Table 2–4 shows the results of the 1999 trial burn, which confirmed that the MPF could process more agent than the 5 percent residual agent heel permit limit and, in fact, was capable of destroying completely filled projectiles in full compliance with RCRA requirements for agent DRE (U.S. Army, 1999b). Comparison of the 1992 and 1999 Trial Burn Results The principal differences between the 1992 and 1999 trial burns were the rates of agent loading and the types of containers processed. In 1992, agent was introduced to the MPF in a single ton container that had been punctured; in 1999, agent was introduced in a tray of 96 projectiles filled to
OCR for page 10
A Modified Baseline Incineration Process for Mustard Projectiles at Pueblo Chemical Depot TABLE 2–2 Metal Emissions in 1992 Trial Burn Tests at JACADS on HD Ton Containers in the MPF Test Run 1 2 3 4 Sample volume (dscm) 1.25 1.39 1.34 1.33 Metal Concentrationaμg/dscmb Arsenic 0.8 18.4 NDc 1.6 Selenium 5.7 4.2 ND ND Chromium 1.5 2.3 2.6 2.6 Lead ND 16.0 6.3 17.0 Barium ND ND ND 8.2 Tin ND ND 12.0 ND Phosphorus 97.0 43.0 81.0 84.0 Zinc 21.0 27.0 21.0 22.0 Boron 205.0 81.0 161.0 132.0 Manganese 0.2 8.6 409.0 2.1 Copper ND 5.1 3.8 ND Mercury ND 6.4 4.7 2.0 aValues blank corrected using the field blank results. bMicrograms per dry standard cubic meter. cND=none detected; detection limits listed below. Run 1: 4.0 μg/dscm for lead and copper 8.0 μg/dscm for barium and tin 1.6 μg/dscm for mercury Run 2: 7.2 μg/dscm for barium and tin Run 3: 3.7 μg/dscm for arsenic and selenium 7.5 μg/dscm for barium Run 4: 3.8 μg/dscm for selenium and copper 7.5 μg/dscm for tin Source: Adapted from U.S. Army, 1992. TABLE 2–3 Number of HD Items Destroyed at JACADS Mustard (HD) Munition/Container Quantity 155-mm projectiles 5,779 105-mm projectiles 45,154 4.2-inch mortar projectiles 43,660 Ton containers 68 Source: U.S. Army, 1999a. capacity, with the agent cavity of each projectile punctured and open to the atmosphere inside the MPF. This increased the weight of HD processed from 60 lb to more than 500 lb. There were also differences in MPF operating temperatures and other parameters, such as the surface area for agent evaporation. In 1992, EPA required an operating temperature of 1,450°F±250°F. In four test runs, the actual measured temperatures ranged from 1,408°F to 1,461°F (U.S. Army, 1992). In 1999, EPA required a primary chamber “set point” temperature of 1,600°F with an allowable range of 1,425°F to 1,750°F. The multirun trial burn average temperature was 1,596°F (U.S. Army, 1999b). Table 2–5 is a comparison of limits from the 1998 JACADS RCRA permit and 1992 and 1999 trial burn data. Table 2–6 is a comparison of selected emissions, including those that exceeded permit limits. The 1992 trial burn results did not violate existing air quality standards; however, standards for metals had not yet been established. Both cadmium and mercury emissions in the 1999 trial burn exceeded 1998 standards. The JACADS project manager ascribed the source of mercury in the emissions to spillage of mercury from manometers during the filling of the projectiles with agent, which probably was not a problem in the ton containers used in the 1992 trial burn (personal communication from Gary W.McCloskey, PMCD JACADS Site Project Manager, June 22, 2000). The source of the cadmium emissions is silver solder. Extensive sampling of the stack gases for organic compounds was also conducted. No emissions of dioxins, furans, polychlorinated biphenyls or any of the 138 other semivolatile organic compounds were detected. All emissions were well within RCRA permit levels (U.S. Army, 1999b). The DRE of agent was 99.9999 percent, well above the RCRA 99.99 percent target. Processing of 4.2-Inch HD Mortar Shells Through the MPF Despite the emissions of cadmium and mercury that exceeded current standards during the 1999 trial burn, EPA extended the JACADS operating permit for the MPF to allow incineration of punched but undrained (energetics and burster well removed) 4.2-inch mortar shells in trays of 96 rounds. The target temperature setting for Zone 1 of the MPF had to be lowered from 1,600°F to 1,475°F because of the larger quantity of agent being oxidized at peak loading (Webster, 2000). Additional modifications to process parameters were made to adjust for the disposal of SDS in the LIC afterburner and for the effects of processing secondary wastes. Observations Based on HD Operations at JACADS The modified baseline process for Pueblo that the committee examined was developed from successful operations of the baseline system at JACADS and lessons learned during the processing there of mustard-filled munitions. Solutions developed and tested at JACADS for problems encountered during the disposal of mustard-filled munitions were the basis for process modifications and are considered to be applicable to the PCD stockpile. The committee’s principal observations from the JACADS experience are as follows:
OCR for page 11
A Modified Baseline Incineration Process for Mustard Projectiles at Pueblo Chemical Depot TABLE 2–4 Results of the 1999 Trial Burn of Mustard-containing Projectiles at JACADS Emissions Parameter Permit Section Module Va Permit Limit MPF Results (four-run average) Compliance Limit Met Agent DRE V.B.1 99.99% >99.999999% Yes Participate matter V.B.3 180 mg/dscmb 3.8 mg/dscmb Yes Hydrogen chloride (HC1) V.B.2 1.8 kg/hr <0.003 kg/hr Yes Carbon monoxide (CO) V.F.2.c 100 ppmb 14.1 ppmb Yes Agent HD concentrationc V.F.2.j 0.03 mg/m3 <0.000150mg/m3c Yes Trace Metals EPA Permit Limitsd (g/sec) MPF Results (g/sec) Compliance Limit Met Antimony (Sb) 5.87 E-05 <4.68 E-07 Yes Arsenic (As) 8.52 E-05 1.64E-06 Yes Barium (Ba) 6. 14 E-05 <3.37 E-05 Yes Beryllium (Be) 2.38 E-05 <3.33 E-07 Yes Cadmium (Cd) 2.98 E-05 <4.02 E-05 Noe Chromium (total) 2.87 E-05 <1.54E-06 Yes Lead (Pb) 7.93 E-05 9.54 E-06 Yes Mercury (Hg) 4.29 E-05 <2.25 E-04 Noe Silver (Ag) 1.91 E-05 <4.52 E-06 Yes Thallium (T1) 1.91 E-05 <1.48 E-07 Yes aJACADS permit (July 1998), module V. bMilligrams per dry standard cubic meter (dscm) corrected to 7 percent O2 or parts per million by weight. cDetermined from analysis of special DRE Depot Area Air Monitoring System (DAAMS) sorbent tubes at Station 18 Common Stack. dJACADS RCRA permit Table 5–12, “Maximum Allowable Stack Emissions Limits.” eThe analytical data in Table 2–4 are the averaged results from four different trial burns conducted in 1999. For both mercury and cadmium, in three of the four analyses, the levels were below the detection limit. However, in one case for each metal, the measured levels were above the detection limit and in excess of the MACT standard, which is probably the reason the EPA showed noncompliance. Source: Adapted from U.S. Army, 1999b. TABLE 2–5 Comparison of Limits from the JACADS RCRA Permit and Results of 1992 and 1999 Trial Burns Emission Parameter 1998 Permit Limit 1992 Trial Burn 1999 Permit Level Agent DRE 99.99% 99.9997% >99.999999% Participate matter, corrected to 7% O2 180 mg/dscm 4.41 mg/dscm 3.8 mg/dscm Hydrogen chloride 1.8 kg/hr 0.03 kg/hr <0.003 kg/hr Carbon monoxide 100 ppm 7.45 ppm 14.1 ppm Agent HD concentration 0.03 mg/m3 None detected <0.000150 mg/m3 Source: U.S. Army, 1992, 1999b. Although JACADS encountered numerous operational problems during processing, many of which were unanticipated in the baseline system design, the facility maintained a good safety record for agent processing and handling. Munitions filled with HD drained more slowly and with more difficulty than expected, and the unexpected frothing of agent created serious maintenance and production problems. Freezing the munitions before opening the agent cavity to the atmosphere minimized the frothing. Although freezing minimized problems associated with accessing the agent cavities of mustard rounds, it added a processing step. Only a small number of rounds was processed that way at JACADS, and no frozen rounds were introduced to the MPF for final processing. The MPF can effectively process much larger residual agent heels than the 5 percent limit stipulated in the original EPA operating permit. Safe, effective processing of munitions was demonstrated, even with 100 percent of the original HD agent fill. A separate fur
OCR for page 12
A Modified Baseline Incineration Process for Mustard Projectiles at Pueblo Chemical Depot TABLE 2–6 Comparison of Selected Emissions (including those exceeding permit limits) for JACADS Trial Burns Trace Metal 1998 Permit Limit (g/sec) 1992 Trial Burn (μg/dscm) 1992 Trial Burn (g/sec) 1999 Trial Burn (g/sec) Arsenic 8.52 E-05 18.4 7.29 E-05 1.64 E-06 Cadmiuma 2.98 E-05 No record No record <4.02 E-05 Chromium 2.87 E-05 2.25 8.91 E-06 <1.54E-06 Lead 7.93 E-05 9.83 3.89 E-05 9.54 E-06 Mercurya 4.29 E-05 3.28 1.30 E-05 <2.25 E-04 NOTE: 1992 trial burn data were reported in micrograms per dry standard cubic meter (dscm) of stack gas. 1998 standards and 1999 trial burn data were reported in grams per second. 1999 stack gas volume was reported, after computation, at 3.96 dscm per second. 1992 stack gas volume, after computation, was reported to be 2.90 dscm per second. To facilitate comparison, 1992 micrograms per dscm were multiplied by 1999 stack gas volume per second times E-06. This converts the 1992 unit contaminant levels to total grams per second for the 1999 stack gas volume. aExceeds 1998 limits. The analytical data are the averaged results from four different trial burns conducted in 1999. For both mercury and cadmium, in three of the four analyses, the levels were below the detection limit. However, in one case for each metal, the measured levels were above the detection limits and in excess of the MACT standard. Source: Adapted from U.S. Army, 1999b. concurrently with facility design. Delay in closure planning adds time and expense to the life-cycle cost estimate. nace for processing liquid agent is not necessary for destroying projectiles filled with mustard agent. Only one projectile out of 94,000 was found to be leaking HD from the agent cavity into the burster well at the time the energetics were to be removed. Eighty others were found to be leaking externally. The JACADS experience suggests that leakage will not be a major problem during disposal operations at PCD. Emissions of cadmium and mercury in excess of standards were measured during the 1999 JACADS trial burn. Except for charging SDS to the LIC afterburner during operations, secondary wastes were not treated as they were produced. Substantial inefficiencies that were created from having to handle massive quantities of secondary wastes at the end of JACADS disposal operations included protracted storage and monitoring; multiagent mixing in storage areas, thus complicating monitoring; the possibility of further contamination of igloos and surrounding areas; and increases in the risk of worker exposure and transportation-related accidents. Additional secondary waste was generated as a result of contaminating storage containers, such as the drums used for holding DPE suits. Closure is an expensive, time-consuming, and complex process requiring thorough planning beginning FINDINGS AND RECOMMENDATIONS Finding 2–1. Trial burn results from the 1999 JACADS tests confirmed the required destruction and removal efficiency of 99.9999 percent when 4.2-inch mortar projectiles filled with mustard agent were incinerated through the metal parts furnace at a feed rate of 96 rounds per batch. Subsequently, almost 95,000 mustard projectiles were successfully destroyed in the JACADS MPF by mid-1999. Recommendation 2–1. Based on the successful JACADS campaigns, the Army should evaluate a process design for Pueblo in which the munitions filled with mustard are processed through an MPF. Finding 2–2. The 1999 JACADS trial burn did not include introduction of frozen projectiles into the metal parts furnace for final processing. Mustard in the frozen projectiles thawed before entering the MPF. Recommendation 2–2. The Army should determine whether freezing projectiles before opening the mustard agent cavity to the atmosphere is necessary to mitigate frothing. If so, the Army should determine, by testing, whether frozen projectiles can be processed successfully through a metal parts furnace, or as an alternative, if it is feasible to allow the agent to thaw before the projectiles are fed to a metal parts furnace. Finding 2–3. HT mustard-filled munitions were not processed at JACADS. HT and HD consist of similar chemicals and will most probably result in much the same products of combustion. Nevertheless, there is no evidence that testing and analysis of HT combustion have taken place. Recommendation 2–3. Regarding HT, the Army should verify that the combustion of HT will produce results akin to the combustion of HD. These results should be considered in the development of the modified baseline process. Finding 2–4. The 1999 JACADS trial burn results indicated that mercury and cadmium were emitted at unacceptable levels during the disposal of some mustard agent. Recommendation 2–4. The Army should prove, through testing, an acceptable technique for capturing emissions of heavy metals—particularly cadmium and mercury—from the metal parts furnace when processing mustard-filled projectiles. An acceptable disposal plan for accumulated heavy metals must be included in the modified baseline process or any other process.
OCR for page 13
A Modified Baseline Incineration Process for Mustard Projectiles at Pueblo Chemical Depot Finding 2–5. Secondary wastes, including dunnage and demilitarization protective ensemble suits, were reported to be successfully processed through the metal parts furnace (MPF) at JACADS, but only limited data on rates, operating conditions, and other parameters for handling these wastes in the MPF have been presented. The best way to handle spent carbon appears to be through the use of a micronizer system. Recommendation 2–5. The Army should determine whether adequate data are available from JACADS to support the efficacy of processing secondary wastes in the metal parts furnace. If not, the Army should determine the additional tests required to confirm a disposal process. A plan based on these results should also be developed for handling and disposing of all secondary wastes from processing the Pueblo stockpile, including demilitarization protective ensemble suits and hoses, spent carbon filter materials, scrubber brine solutions, plant cleaning wastes, and dunnage.
Representative terms from entire chapter: