Assessment of Explosive Destruction Technologies for Specific Munitions at the Blue Grass and Pueblo Chemical Agent Destruction Pilot Plants (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Rating of Explosive Destruction Technologies for Proposed BGCAPP and PCAPP Applications INTRODUCTION used to assess the various EDTs were process maturity, process efficacy, process throughput, process safety, the This chapter presents the committee’s evaluation of public or the regulatory acceptability in a U.S. context, the explosive destruction technologies (EDTs) under secondary waste issues, destruction verification capa- consideration. When evaluating a technology such as bility, and process flexibility. These factors are dis- an EDT, various factors must be considered. One of the cussed in detail in Chapter 2. The EDTs considered in important factors is the economics of the technology. this chapter are discussed in detail in Chapter 3, and the This includes the capital cost of the equipment, the oper- four EDTs that are candidates for destroying chemical ating costs, utility costs, waste disposal costs, and closure weapons are described in Appendix A and Chapter 3. costs. These life-cycle costs have not been assessed by The methodology for assessing the technologies was the committee in this study. An analysis of proprietary as follows: capital cost data was not part of the committee’s task, nor did the committee have sufficient resources to predict the • For each of the four requirements—BG-1, BG-2, other components of the life-cycle costs. Therefore, the BG-3, and P-1 (see Table 4-1)—the committee recommendations generated by the committee relate only members independently assessed each of three or to what might be termed “technical acceptability,” in this four EDT candidates with respect to the eight fac- case the eight evaluation factors discussed in Chapter 2. tors and assigned the technology a value between The four requirements for the EDTs that the commit- 1 and 10. A committee member’s assessment tee was asked to consider are listed in Table 4-1. Some was based on available information, committee judgment has been used when one of the requirements in discussions, and individual expertise. Table 4-1 appears to be more suited to a particular EDT • The independent ratings of the committee mem- owing, say, to the number of items to be processed or bers were averaged to obtain a number from the time available for processing, among other consid- 0-10, reflecting the committee’s collective judg- erations, but cost as such was not considered. Therefore, ment about each EDT with respect to the various certain recommendations made by the committee might factors—safety, throughput (to which Table 4-2 require modification when the life-cycle costs of the applies), efficacy, and so on. various EDTs are fully understood. • For example, an EDT assessed by the committee as excelling in a particular evaluation factor for Basis for Assessment a particular requirement was rated at 10. Indeed, it was possible for two or more technologies to Eight factors were used to evaluate the suitability of receive a score of 10 if they were both (or all) a technology for a particular requirement. The factors outstanding. Ratings of less than 10 reflected 55 

56 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES TABLE 4-1  Requirements Proposed for EDT Processing of Chemical Stockpile Items at Blue Grass Army Depot and Pueblo Chemical Depot Requirement No. Requirement Description BG-1 Requirement BG-1 is the processing of about 70,000 M55 rocket motors at Blue Grass and about 15,000 munitions that are not contaminated with agent. Current plans call for shipment of these noncontaminated rocket motors to an off-site location for processing; destruction in an EDT is being considered as an alternative. BG-2 Destruction of all 155-mm mustard agent H projectiles at Blue Grass. BG-3 Destruction of both noncontaminated M55 rocket motors and mustard agent H projectiles at Blue Grass. P-1 Destruction of all leakers and reject munitions at Pueblo. About 1,000 mustard agent-filled munitions—a mixture of 4.2-in. mortars, 105-mm projectiles, and 155-mm projectiles—would be destroyed. TABLE 4-2  Throughput Rates of Five EDTs and Their Implications for Schedule: Requirements BG-1, BG-2, and BG-3a,b Requirement TC-60 TDC D-100 DV65 Proposed DV120 SDC2000 BG-1: 70,000 rocket motors N/A 180/day 36/day 72/day 100/day 389 days 1,945 days 972 days 700 days 65 weeks 324 weeks 162 weeks 117 weeks 1.25-2.50 yrs 6.23-12.46 yr 3.11-6.22 yr 2.25-4.5 yr BG-2: 15,000 155-mm mustard 17/day N/Ac 18/day 36/day 30/day agent projectiles 882 days 834 days 417 days 500 days 147 weeks 139 weeks 70 weeks 84 weeks 2.83-5.66 yr 2.67-5.34 yr 1.34-2.68 yr 1.60-3.20 yr BG-3: 70,000 rocket motors and N/Ad N/Ad 2,779 days 1,389 days 1,200 days 15,000 mustard agent projectiles 463 weeks 232 weeks 200 weeks (combined BG-1 and BG-2) 8.90-17.8 yr 4.46-8.92 yr 3.85-7.70 yr aTen-hour operating days, 6-day work weeks, and 52 weeks of operation/year are assumed. bRanges are shown for the processing times, calculated for each EDT to meet each of the requirements. For example, the expected pro- cessing time (campaign length) for the use of the D-100 to destroy the 70,000 rocket motors at Blue Grass is from 1.25 years to 2.50 years. The lower number is based entirely on the lowest demonstrated elapsed time between detonation events, the number of munitions expected to be destroyed in each detonation event, and operation of the system at this maximum capacity. The inputs for these calculations are shown in the table. Thus, it is expected that 180 rocket motors can be destroyed per day if the D-100 system is operated at its maximum capacity. The 70,000 rocket motors could be destroyed in 389 processing days, corresponding to 65 weeks, or 1.25 years. This is the lower number of the range shown. The second number shown is the upper end of the range and results from doubling the number for the lower end of the range. This is an attempt to account for the effects of scheduled and unscheduled maintenance and other causes of delay. Note that in certain cases the Army may choose to operate in a fashion that does not relate to the throughput rate capability of the EDT. For example, rocket warheads might be processed through the main plant more slowly than the rocket motors could be destroyed in an EDT. In that case, the Army might choose to destroy the rocket motors in an EDT at the same rate as the warheads are treated in the main plant. c The D-100 is not intended for destroying chemical munitions. d Refer to the entry for Requirement BG-2 for the TC-60 TDC and the entry for Requirement BG-1 for the D-100. Both of these units in combination are necessary to meet Requirement BG-3 because the D-100 is not intended for destroying chemical munitions. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 57 shortcomings of varying degrees. If an EDT had from CH2M HILL was compared with only single serious problems but they could be corrected systems from other vendors for Requirement BG-3. while retaining all the main features of the pro- It is expected that ACWA will be able to consider the cess, it might be assigned a rating of 4, the lowest committee’s evaluations and recommendations for rating given in the course of this exercise. Requirements BG-1 (noncontaminated rocket motors • For each technology, the ratings were summed, and only) and BG-2 (mustard agent projectiles only) and the summed ratings (see Tables 4-3 through 4-6, will come to its own conclusions on the use of such which appear in association with the discussions combinations. for Requirements BG-1 through P-1, respectively) formed the basis for the committee's recommenda- Requirement BG-1: Destruction of tions. If there were small differences between the Approximately 70,000 NonContaminated selected EDT and one of the other EDTs, the com- M55 Rocket Motors at Blue Grass mittee noted this. Small differences in the summed ratings, up to about five points, were not considered Requirement BG-1 applies to the rocket motor com- to be significant by the committee. The evaluation ponent of M55 rockets but not to the warhead compo- factors were not weighted, but it would be possible nent. Therefore, Requirement BG-1 does not involve for the Army to assign its own weighting factors the processing of agent and can be considered to be a and generate revised summed ratings. conventional munitions disposal application. The com- mittee assumed that the rocket motor will be removed The Army’s Explosive Destruction System (EDS) from its associated shipping and firing tube (SFT) seg- would not be able to satisfy Requirements BG-1, BG-2, ment and, owing to its polychlorinated biphenyl content, and BG-3, so it was not considered for those require- will be disposed of in a Toxic Substances Control Act ments. After discussions with Assembled Chemical (TSCA)-approved treatment facility. The M55 rocket Weapons Alternatives (ACWA), it was decided to not motor contains 19.3 lb M28 double base (nitroglycerin evaluate the TC-60 transportable detonation chamber and nitrocellulose) cast grain propellant. The D-100 (TDC) for the destruction of noncontaminated rocket from CH2M HILL, the detonation of ammunition in a motors using either a static firing approach or a donor vacuum integrated chamber (DAVINCH) DV65 from charge approach for Requirement BG-1. The TC-60 Kobe Steel, Ltd., and the Dynasafe SDC2000 were TDC is not designed for such an application; more- evaluated for this requirement. over, CH2M HILL offers the D-100 system, which is The M28 propellant in M55 rocket motors contains designed to destroy conventional weapons and, if test- 2 percent lead stearatea significant amountand ing is successful, should be capable of being used for the detonator contains a smaller amount of lead azide static firing of the noncontaminated rocket motors. Also, (BPBGT, 2004). All of the technologies considered for as indicated in Chapter 3, a D-100 system is already this requirement face the issue of working with this installed at the Blue Grass Army Depot (BGAD). quantity of lead and disposing of the lead-containing Thus, the D-100 system was evaluated for Require- residues. For those technologies that require opera- ment BG-1, and the TC-60 TDC was evaluated for tors to insert materials into the detonation chamber Requirement BG-2. For Requirement BG-3, the com- through a large open doorall except the Dynasafe mittee evaluated the combination of the two CH2M SDC2000exposure to dust containing lead and other HILL technologies (the D-100 for the noncontaminated particulates is possible. Operators of these systems rocket motors and the TC-60 TDC for the mustard should wear respiratory protection when working near agent-filled projectiles). This was done with the concur- the open door. rence of ACWA. Further, this combination of systems Finding 4-1. When processing noncontaminated rocket motors, operators of the DAVINCH DV65 or the Personal communication between Joseph Novad, Deputy Pro- CH2M HILL D-100 could be exposed to substantial gram Manager, ACWA, and Margaret Novack, NRC, study director, amounts of dust containing lead when working near the September 23, 2008. open door of the detonation chamber.  Personal communication between Joseph Novad, Deputy Program Manager, ACWA, and Richard Ayen, committee chair, September 23, 2008. Information at www.fas.org/man/dod-101/sys/land/m55.htm. 

58 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES Recommendation 4-1. When processing rocket motors included 25,000 155-mm M483 projectiles containing and working near the open door of the detonation cham- M42 or M46 submunition grenades (CH2M HILL, ber, operators of the DAVINCH DV65 or the CH2M 2007). The D-100 system at BGAD is available for HILL D-100 should be required by the Army to wear this requirement and its DDESB approval for 49.3 lb respiratory protection to minimize exposure to lead. TNT-equivalent total explosives is adequate. However, as of the writing of this report, the Resource Conservation and Recovery Act (RCRA) permit for the Process Maturity D-100 at BGAD was in process. Also, the committee was not aware of any previous use of the D-100 or D-100 related systems for destruction of rocket motors. As described in Chapter 3 and as discussed above, BGAD and CH2M HILL propose using a CH2M HILL DAVINCH D-100 system already installed at BGAD to destroy noncontaminated M55 rocket motors. Other D-100 DAVINCH is a mature technology that should be systems and other CH2M HILL destruction chambers able to destroy M55 rocket motors. The DV65 has an for conventional weapons from CH2M HILL’s product inside diameter of 2.6 m and a length of 6.9 m; the line have been or are in routine operation at military M55 rocket motor body is only about 1 m. The M55 bases and used for destruction of a wide variety of rocket motor contains 8.8 kg of double base propel- conventional munitions. A transportable T-10 model lant. In testing the destruction of a simulated M55 with a Department of Defense Explosive Safety Board rocketi.e., both the rocket motor and a surrogate- (DDESB) approval for up to 4.3 lb total explosives filled warheadanother 22.2 kg of donor charge was was used at the Massachusetts Military Reservation used, for a total TNT-equivalent explosive loading of to destroy over 25,000 munitions ranging from small 31 kg, well within the 65-kg containment capability of arms to 105-mm projectiles. This same system had the DV65. Thus, a scale-up of the DAVINCH would been deployed at five sites as of May 2007. At Camp not be required for application to Requirement BG-1, Navajo Army National Guard Base in Arizona, several although for increased throughput, a proposed, but not types of white phosphorus munitions, including 162 yet built, longer version of the DAVINCH, the DV120, 81-mm mortars, were destroyed. The system was used might be used. at four sites in California to destroy 28,858 munitions Although the DAVINCH technology has been per- in 15 days, although a more typical throughput is 25 mitted to operate in Japan and in Belgium and has munitions per day. The system was also employed at destroyed over 2,500 items of recovered chemical Redstone Arsenal, Alabama, for destruction of smoke- weapons materiel as of late May 2008, it has not yet producing fills, riot agent fills, and incendiary fills. A been permitted to operate in the United States. Also, as D-200 was installed at the Crane Naval Surface Warfare is the case with the other EDTs, the DAVINCH has not Center in 2002. Of more direct relevance to this require- had an opportunity to demonstrate an ability to destroy ment, two D-100 systems, the same model proposed M55 rocket motors. for use by BGAD in its static firing proposal, have been installed at the Milan Army Ammunition Plant SDC2000 in ­Tennessee. CH2M HILL reported that the systems were inactive as of the writing of this report; however, Dynasafe proposes to use the model SDC2000 for they had been used for 3 years to destroy munitions that destruction of the noncontaminated rocket motors. Dynasafe has extensive experience with this system in J. Germany and Taiwan. See Chapter 3 for descriptions Quimby, 2007, “Current status of transportable controlled d ­ etonation chambers (CDCs) offered by CH2M HILL,” Pre- sentation to the National Defense Industrial Association Global Personal communication between Brint Bixler, Vice President, D ­ emilitarization Symposium and Exhibition in Reno, Nevada, in CH2M HILL, and Richard Ayen, committee chair, August 29, May 2007. Available online at http://www.dtic.mil/ndia/2007global_ 2008. demil/2007global_demil.html. Last accessed February 18, 2009. Personal communication between Brint Bixler, Vice President, Personal communication between Brint Bixler, Vice President, CH2M HILL, and Richard Ayen, committee chair, July 23, 2008. CH2M HILL, and Margaret Novack, NRC, study director, July 10, Personal communication between Harley Heaton, Vice President 2008. for Research, UXB International, Inc., and Margaret Novack, NRC, 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 59 of the production experience in Germany and of the over approaches where fragmentation occurs. Note, testing done to obtain destruction efficiencies (DEs) for however, that the committee did not meet with or com- mustard agent and to obtain information on the feeding municate with military personnel involved in operating and discharging of the rocket bodies. The feed system these systems to obtain information on maintenance at Münster, Germany, was too small to accommodate needs. the long rocket motors, but the vendor states that the feed system can be enlarged if a new system is built DAVINCH for the Blue Grass Chemical Agent Destruction Pilot Plant (BGCAPP). In limited testing to date, the DAVINCH has The aluminum fires experienced in the simulated d ­ emonstrated the ability to destroy a simulated M55 rocket motor testing at Münster were attributed to the rocket contained in a wooden box. Additional testing use of too much aluminum. More extensive experience carried out with a variety of linear-shaped charges has in Taiwan was cited as an indication that aluminum fires demonstrated the ability to access the agent, burster, will not be an issue at BGCAPP. The only aluminum and propellant through both an overpack and the fiber- components of rocket motors are the fins; the bodies are glass SFT. steel. Also, the multiple vessel rotations used to remove If, as proposed by Kobe Steel, Ltd., a DV65 were scrap from the detonation chamber are considered by used to destroy four rocket motors per shot, then 17,500 the manufacturer to be part of the normal operation of shots (detonation events) would be required. The load- the system. ing would be fairly high: 35.2 kg energetics and 8.8 kg propellant per rocket. The associated donor charge would have to be sufficient to destroy the propellant Requirement BG-1 Ratings for Process Maturity but not so large as to exceed the 65 kg explosive con- For the process maturity criterion for Requirement tainment capacity of the DV65. The DAVINCH may BG-1, the committee assigned a rating of 8 to the well be able to process this number of shots. The inner DAVINCH DV65, a rating of 8 to the CH2M HILL vessel is replaceable; it can be rotated to distribute wear D-100, and a rating of 6 to the Dynasafe SDC2000. around its surface and is reinforced with four layers of The committee downgraded the Dynasafe SDC2000 steel around its outer perimeter. by two points because of slight uncertainties over the Although various versions of the DAVINCH have new air pollution control system and the apparently been used to destroy over 2,500 items of recovered solvable problems experienced during the simulated chemical warfare materiel, the destruction of about rocket motor testing. 70,000 M55 rocket motors at BGAD would represent an increase of more than an order of magnitude in the number of items processed and the number of shots, Process Efficacy especially since these would presumably be destroyed in a single DAVINCH vessel. Rocket motors would be D-100 destroyed in the same way as projectiles: by placing The D-100 and related systems appear to be reliable external donor charges on the rocket motors, hanging and robust, as evidenced by their extensive use to date. them in slings so they are suspended in the middle of The T-10 system has destroyed over 163,000 munitions the inner vessel, and detonating the donor charges. A since 2000; the more than 3-year campaign at Milan static firing stand would not be used.10 Army Ammunition Plant used two D-100s to destroy 25,000 155-mm projectiles containing submunition SDC2000 grenades (CH2M HILL, 2007). The static firing con- cept (see Chapter 3), which does not involve fragmen- The testing at Münster involved the feeding of 2.3 kg tation of the munition, appears to offer an advantage propellant, along with 2.3 kg aluminum and bottles of water, to the SDC2000 (UXB, 2007). The primary study director, July 17, 2008. Personal communication between Brint Bixler, Vice President, 10Personal communication between Joseph Asahina, Chief of CH2M HILL, and Richard Ayen, committee chair, August 29, Technology, Kobe Steel, Ltd., Douglas Medville, committee mem- 2008. ber, and Richard Ayen, committee chair, September 27, 2008. 

60 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES objective was to show that the water effectively con- Process Throughput trolled the temperature rise during the rocket firing, The throughput rates shown in Table 4-2 (and appli- which it did. The vendor claims that the modifications cable for Requirements BG-1, BG-2, and BG-3) repre- that would be built into a new system for BGCAPP sent peak rather than average throughput rates. Actual would accommodate the 1-m-long rocket motors and processing rates will likely be slower during routine that the aluminum fires experienced at Münster would operations because of the downtime associated with not occur during operations at BGCAPP because the both scheduled and unscheduled maintenance. This ratio of aluminum to steel would be much lower than will reduce system availability and, accordingly, the that at Münster. The M67 rocket motor contains only average processing rate. The lower end of the projected 0.28 kg aluminum, not 2.3 kg. Additional testing using campaign lengths in Table 4-2 assumes production is a more appropriate amount of aluminum is needed. The always at the peak rate. The upper end of the projected net explosive weight (NEW) limit for the SDC2000 campaign length is twice the lower end and is an system at Münster was set by permit at 2.3 kg, about attempt to take into account downtime for any reason. one-fourth the NEW of the rocket motor. For a new system constructed for BGCAPP, ­Dynasafe claims the NEW limit can be up to 10 kg depending D-100 on the choice of the inner chamber option. This is just As indicated previously, the D-100 system for sufficient to withstand the unexpected detonation of a conventional weapons, with its 49.3 lb NEW limit, is single rocket motor with its 19.3 lb (8.8 kg) of propel- expected to be able to accommodate up to six rocket lant. Finally, Dynasafe proposes to drop the rocket motors per cycle, firing the motors sequentially in a fir- motors into the hot detonation chamber and to not ing stand. A throughput rate of 18 munitions per hour is employ static firing, as proposed by CH2M HILL. The anticipated. If the system operated for 10 hours per day, committee expects that the motors will move around 18 motors per hour, 6 days per week, the 70,000 motors energetically inside the chamber when the heat of the would be destroyed in about 1.25 years. The committee chamber causes them to fire. The extent to which this thus projected a campaign length of between 1.2 years could be an operational problem is not known, but the and about 2.5 years. vendor’s analysis indicates that the problem would not be serious (UXB, 2007). The system appears to be robust and reliable. DAVINCH The DAVINCH manufacturer claims that the exist- Requirement BG-1 Ratings for Process Efficacy ing DV65 can process four M55 rocket motors per shot with a throughput rate of nine shots per 10-hour day—a None of the EDTs being considered for Requirement cycle time of slightly over 1 hour. If 36 rocket motors are BG-1 has demonstrated the destruction of whole M55 destroyed per 10-hour day, the 70,000 motors would be rocket motors, although some testing has been done destroyed in 1,945 days, or 6.23 years, assuming 6-day using the DAVINCH and the SDC2000. The static operating weeks. The committee thus projected a cam- firing concept proposed for the D-100 appears to be paign length from about 6.2 years to about 12.5 years. effective because the rockets are not allowed to move A proposed longer and higher explosive containment energetically around the chamber when they fire and capacity version of the DAVINCH, the DV120, could are not subjected to fragmentation. The committee also be used to destroy the rocket motors. This version decided that reliability was the most important sub- would be used to destroy nine rocket motors per shot factor for this evaluation factor, and the rating would and would have a throughput of eight shots per day. If be based primarily on past performance on all muni- this version were to be used, the rocket motors would tions. All have favorable reliability records. Thus, for be destroyed in 972 days, or 162 6-day operating weeks Requirement BG-1, the committee rated all three—the (3.11 years). The committee thus projected a campaign DAVINCH DV65, the Dynasafe SDC2000, and the length ranging from about 3.1 years to about 6.2 years. CH2M HILL D-100—at 9. This could be acceptable if it is less than the expected duration of operations for BGCAPP. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 61 Based on operating experience to date, the DAVINCH occur. The rocket motors and their donor charges can manufacturer states that about 30 minutes per day, an be inserted into the inner DAVINCH vessel by a robotic additional 3 hours per week, and still another 3 hours arm, with no worker exposure to energetics during that per month are required for routine maintenance activi- operation. ties. While these routine activities do not affect peak Since donor charges are needed to access the rocket throughput, unscheduled corrective maintenance activi- motors, these explosives need to be stored in the vicin- ties will probably take place over a 3- to 6-year operat- ity of the DAVINCH. This could create an additional ing period and could reduce this throughput. hazard. SDC2000 SDC2000 The throughput rate expected by the vendor for the Once the munitions are transported to the SDC2000, SDC2000 is high, 10 motors per hour, 10 hours per day, the processing is automatic and no donor explosives for a total of 100 motors per day (UXB, 2007).11 This have to be attached. Overall, few manual operations are translates to about 700 operating days to destroy all the associated with the static detonation chamber (SDC) noncontaminated rocket motors. The committee thus for Requirement BG-1. projected a campaign that would take about 2.2 years to about 4.5 years. Requirement BG-1 Ratings for Process Safety For the process safety criterion for Requirement Requirement BG-1 Ratings for Process Throughput BG-1, the committee assigned a rating of 8 to the For the process throughput factor for Requirement CH2M HILL D-100, a rating of 8 to the DAVINCH BG-1, the committee assigned a rating of 10 to the DV65, and a rating of 9 to the Dynasafe SDC2000. CH2M HILL D-100, a rating of 5 to the DAVINCH DV65, and a rating of 8 to the Dynasafe SDC2000. Public and Regulatory Acceptability in a U.S. Context Process Safety D-100 RCRA permits have been issued for the two D-100 D-100 systems at Milan Army Ammunition Plant and for Use of the D-100 with static firing would not require several other CH2M HILL detonation chambers for donor explosives to be attached to the rocket motors. conventional weapon destruction. The RCRA permit This would minimize the amount of explosives in the for the D-100 at BGAD is in progress. work area and the handling of those explosives. The rocket motors must be handled manually, so DAVINCH exposure to lead is a possibility. The motors must be loaded into firing racks. Ignition wires must be con- Although the DAVINCH technology has not so far nected to the igniters, and new igniters might need to been permitted for destroying either energetics such as be installed. the M55 rocket motors or chemical agent in the United States, from a regulatory perspective, it should be able to meet environmental regulatory requirements and DAVINCH achieve permitted status for processing and destroying Personnel protective equipment (PPE) should not be noncontaminated rocket motors at Blue Grass. needed when handling noncontaminated rocket motors at Blue Grass, except when working within the chamber SDC2000 or near the open door, where exposure to lead could The Dynasafe SDC2000 has not been issued regula- tory permits in the United States but has operated in 11Personal communication between Harley Heaton, Vice Presi- Germany, where emissions testing based on U.S. regu- dent for Research, UXB International, Inc., and Margaret Novack, NRC, study director, July 17, 2008. latory requirements has been performed. The commit- 

62 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES tee anticipates that U.S. regulatory bodies would grant SDC2000 environmental permits based on satisfactory testing The SDC2000 generates up to 150 Nm 3/hr of prior to full-scale operations. The manufacturer indi- nitrogen, water vapor, and carbon dioxide offgas. If cates it would change the design of the pollution abate- the ­ offgas system is modified for use in the United ment system for use in the United States, converting States, the manufacturer states that the acidic and basic the secondary combustion chamber to an electrically scrubbers would produce no liquid effluents but would heated oxidizer. produce up to 500 lb/day of salts as a filter cake. Since the rocket motors contain 2 percent lead stearate, the Requirement BG-1 Ratings for Public and Regulatory salts resulting from their processing could be defined as Acceptability in a U.S. Context hazardous waste owing to the lead content. The scrap metal can be released for ­unrestricted use. The CH2M HILL D-100 and smaller and larger models of the same technology have been permitted and operated in a production mode at several locations Requirement BG-1 Ratings for Secondary Waste in the United States. It should be the easiest to permit Issues and to be highly acceptable to the public. It is rated a Both the D-100 and the DAVINCH DV65 produce 10. Neither the DAVINCH nor the Dynasafe SDC2000 only minimal waste streams. Both were rated at 9. have been permitted or operated in the United States. Because of the production of salts from the scrubber, Both are rated a 7. the SDC 2000 is rated lower, at 7. Secondary Waste Issues Destruction Verification Capability D-100 Destruction of M55 rocket motors per Requirement BG-1 does not involve chemical agent, so compliance Secondary waste will consist of spent rocket motor with the provisions of the Chemical Weapons Conven- bodies and dust resulting from pulsed jet cleaning of the tion (CWC) treaty should not be an issue for any EDT particulate filter. The dust from the filter will contain used for this purpose. Therefore, a rating N/A (not lead from the lead stearate in the propellant. It could applicable) applies. possibly be defined as a RCRA hazardous waste, but the motor bodies are not expected to be so defined. Contamination of the rocket bodies with agent is not an Process Flexibility issue since the rocket bodies have never been exposed Process flexibility was considered to be not appli- to agent. This is true for any EDT used for Require- cable to Requirement BG-1, which involves a single ment BG-1. feedstock, noncontaminated rocket motors. DAVINCH Summary Assessment for Requirement BG-1 The volume of solid and liquid wastes other than metal rocket motor casing and fin fragments is expected D-100 to be low: about 0.9 kg of filter dust and 0.2 m3 of con- The D-100 has many advantages. It is designed for densate water and cooler blowdown per shot. Again, the destruction of conventional weapons, and Require- the dust from the filter will contain lead and could be ment BG-1 is just such an operation. The pollution defined as a RCRA hazardous waste. Although agent control system of the D-100 is more appropriate for contamination is not an issue, the offgases will require conventional munitions than those of the other tech- treatment. Their volume and constituents will depend nologies. The waste streams are minimal. The projected on the materials that are detonated and combusted in length of the campaign would average 1.2 to about the DAVINCH vessel—that is, rocket propellant and, 2.5 years, the shortest of all the EDTs considered. A possibly, overpacking materials. program to demonstrate that the D-100 will work as expected has been proposed, but no testing has been 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 63 TABLE 4-3  EDT Ratings Summary for Requirement BG-1, Destruction of Approximately 70,000 Noncontaminated M55 Rocket Motors at Blue Grass Evaluation Factor Public and Regulatory Secondary Destruction Process Process Process Process Acceptability in Waste Verification Process EDT Maturity Efficacy Throughput Safety a U.S. Context Issues Capability Flexibility Total D-100 8 9 10 8 10 9 N/A N/A 54 DAVINCH DV65 8 9  5 8  7 9 N/A N/A 46 SDC2000 6 9  8 9  7 7 N/A N/A 46 NOTE: The above values for each evaluation factor are the average of each committee member’s rating on a scale of 0-10. These average values were then summed to arrive at the totals given in the last column. Small differences in the summed ratings, up to about five points, were not considered to be significant by the committee. There was no weighting. done. Tests with actual rocket motors would be needed side, the throughput rate is very acceptable, and the before this technology could be selected for Require- process has many inherently safe features. Additional ment BG-1. testing would be needed before this technology could be selected for Requirement BG-1. DAVINCH Overall Ratings for Requirement BG-1 The DAVINCH technology could be used to destroy noncontaminated motors from the M55 rockets at Table 4-3 summarizes the ratings for three EDT tech- BGAD. However, the large number of rocket motors nologies for requirement BG-1. The D-100 static firing would take between 6.2 and 12.5 years, assuming system would be the most satisfactory EDT to meet that the throughput rates estimated by the technology Requirement BG-1. The fact that three such systems provider can be maintained. Although no significant have been installed and two have been permitted and technical issues have been identified that would lead operated in the United States is a major plus. One of the the committee to question the ability of the DAVINCH three units is already installed at BGAD and is available to destroy propellant, both DAVINCH and its offgas for this requirement. Larger and smaller systems from treatment system would need to be demonstrated the same CH2M HILL product line have been built before the applicable regulatory permits could be and operated. The summed rating for the D-100 unit is issued. Based on prior experience, the manufacturer 54 out of a possible 60. The Dynasafe feed chambers estimates that it will take 18-24 months for fabrica- and the discharge chute would have to be resized. The tion and installation of a DAVINCH unit at Blue DAVINCH DV65 and the Dynasafe SDC2000 are both Grass, with additional time required for the above- rated 46. Neither has been permitted or operated in the mentioned testing. Limited testing demonstrated that United States, and their throughput rate is lower than a DAVINCH system is capable of destroying a simu- that of the D-100. lated rocket motor, but tests with actual rockets would be needed before this technology could be selected for Finding 4-2. The CH2M HILL D-100 detonation Requirement BG-1. chamber for conventional munitions, using static fir- ing of the rocket motors, is best suited for Require- ment BG-1. The DAVINCH DV65 and the Dynasafe SDC2000 SDC2000 are acceptable second choices. Information is needed on many aspects of the SDC2000, including the nature of the flameless oxi- Recommendation 4-2. For Requirement BG-1, if dizer and the ability of Dynasafe to obtain DDESB testing is successful, the Army should use the CH2M approval to feed whole rocket motors. The system has HILL D-100 detonation chamber at BGAD, with static not been permitted in the United States. On the plus firing of the rocket motors. The Army should consider 

64 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES the Dynasafe SDC2000 and the DAVINCH DV65 as the well-documented subsequent destruction of mustard acceptable second choices. agent-filled 25-pounder projectiles (see Chapter 3). However, the design and operating issues encountered during the recent Schofield Barracks operations point to Requirement BG-2: Destruction of the need for the Army to continue to correct design defi- approximately 15,000 Mustard Agent ciencies, especially considering the large number (15,000) H-filled 155-mm projectiles at of munitions involved in Requirement BG-2. As noted in Blue Grass Chapter 3, these deficiencies were being corrected as this Implementation of Requirement BG-2 would allow report was being written. Also, the TC-60 TDC has little an EDT to process all the mustard agent H muni- experience with destruction of 155-mm projectiles. In the tions, including leakers, stored at BGAD in parallel 2008 campaign at Schofield Barracks, 38 phosgene-filled with the disposal processing of nerve agent VX and 155-mm projectiles were destroyed. No mustard agent- GB projectiles through BGCAPP. The Program Man- filled 155-mm projectiles have been destroyed. ager for Assembled Chemical Weapons Alternatives (PMACWA) staff indicated that this would shorten DAVINCH the overall schedule for the destruction of the BGAD stockpile by 8 months. Although the U.S. Army’s EDS DAVINCH is a mature technology for chemical technology has proven its ability to process the type agent destruction. It had destroyed over 2,500 recov- of munitions that are associated with Requirement ered chemical warfare materiel items (as of late May BG-2, its low processing rate (one 155-mm projectile in 2008) in Japan and Belgium, including lewisite, mus- 2 days) would necessitate a very long campaign: about tard agent, and agents Clark I and II in bombs and a 100 to 200 years for a single EDS. The EDS was there- variety of projectiles. Although the DAVINCH has not fore eliminated from further consideration for Require- been used to destroy mustard agent-filled M104 and ment BG-2 (and also for Requirement BG-3). Because M110 155-mm projectiles, it should be able to do so. In the CH2M HILL D-100 is designed for destruction of Japan, it destroyed over 1,600 Yellow bombs containing conventional weapons only, it, too, was eliminated. a 50/50 mix of mustard agent and lewisite. The mustard agent fill in those bombs was 9.45 kg as compared to 5.31 kg mustard agent in the 155-mm projectile. Their Process Maturity explosive weight was 2.3 kg as compared to 0.186 kg in the 155-mm projectile. In Poelkapelle, Belgium, TC-60 TDC the DAVINCH destroyed 150-mm Clark agent-filled Several versions of the TDC have been used exten- munitions that were comparable in size to the 155-mm sively for the destruction of chemical weapons. A TC-10 projectiles at BGAD (50 cm length vs. 68 cm length for system and a TC-60 system were used at ­Poelkapelle the 155-mm projectiles). The Belgian artillery rounds in Belgium to destroy 3,200 recovered chemical muni- have a NEW of 2.5 to 2.75 kg, while the 155-mm pro- tions. A TC-25 system was tested at Porton Down in jectiles have a NEW of 0.186 kg. Thus, the DAVINCH the United Kingdom in 2003. A TC-60 was extensively has demonstrated the ability to destroy munitions hav- tested at Porton Down from 2004 to 2006. As described ing a greater NEW and containing more agent than the in Chapter 3, this same system was permitted and 155-mm projectile. operated at Schofield Barracks in Hawaii to destroy Scale-up is not required since the DAVINCH DV65 71 munitions containing phosgene and chloropicrin. has an inside diameter of 2.6 m and a length of 6.9 m, CH2M HILL claims that the Army owns this TC-60 more than sufficient to destroy the 155-mm-diameter, system, and it might be made available for Require- 68-cm-long projectiles at BGAD. ment BG-2.12 Of special interest is the upgrading of DAVINCH has not been permitted in the United the system at Porton Down between 2004 and 2006 and States. Before it can obtain a RCRA operating permit, the equivalent of trial burns with agent or agent sur- rogate will have to be conducted to demonstrate the ability of the system to achieve the required degree of 12Personal communication between Brint Bixler, Vice President, agent destruction. CH2M HILL, and Jim Pastorick, committee member, September 3, 2008. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 65 SDC2000 in Chapter 3, these problems were being corrected as this report was being written. A heat exchanger failed, The Dynasafe SDC is a mature technology for indicating improper selection of materials of construc- destruction of the type of chemical weapon in Require- tion or a failure to prevent acid gases from migrating ment BG-2. As reported in Chapter 3, more than 13,000 past the lime feed system to the heat exchanger. This recovered munitions were destroyed at the Münster, same heat exchanger had also failed during testing at Germany, facility. Also described in Chapter 3 is the Porton Down. It was redesigned using new materials 3-day test series carried out at Münster to demonstrate of construction and rebuilt to allow completion of the that the Dynasafe SDC2000 system could effectively Schofield Barracks operation.14 Robustness is also a destroy mustard agent-filled munitions. Three exten- concern. It is not clear that the TC-60 could process sively monitored runs were conducted using distilled 15,000 mustard agent projectiles without very high (sulfur) mustard agent HD-filled 100-mm mortar levels of maintenance. rounds. For each run, either two or three mortars at a However, the TC-60’s approach to munitions destruc- time were fed as a single batch to the SDC approxi- tion is fundamentally sound and, as already noted, the mately three times per hour. HD sampling was con- operating problems were being corrected and the heat ducted at three sampling points in the pollution abate- exchanger that failed has been replaced with a unit with ment system. The results of the 3-day HD tests showed improved materials of construction. Use of the TC-60 is that an overall process DE of >99.999999989 percent continuing, and it can be expected that more improve- (nine nines) was achieved. The technology has not been ments will be made. The current schedule for BGCAPP demonstrated in the United States. Also, Dynasafe has provides about 5 years to make these improvements. As indicated it would not use the same version of the air shown by the testing of the TC-60 on mustard agent- pollution control system used at Münster in the United filled 25-pounder projectiles at Porton Down, the DEs States. As noted in Chapter 3, Dynasafe has provided are high enough (see Chapter 3). Mustard agent con- some information on its planned new system, but the centrations during the testing were below the limit of system has not been designed, built, or tested. detection at the entrance to the final filtration/activated carbon adsorption operation. Requirement BG-2 Ratings for Process Maturity For the process maturity evaluation factor for DAVINCH Requirement BG-2, the committee assigned a rating The DAVINCH has proved to be robust and reliable, of 8 to both the TC-60 TDC and the DAVINCH DV65 having destroyed over 2,500 items of recovered chemi- and gave a rating of 7 to the Dynasafe SDC2000. cal warfare materiel through late July 2008. The items contained a variety of fills, including mustard agent, Process Efficacy lewisite, and Clark agents. If used to destroy mustard agent-filled 155-mm TC-60 TDC projectiles at Blue Grass, the explosive loading would be well within the capability of the DAVINCH Reliability is a concern for the TC-60. As described units—that is, even if a 20- to 30-kg donor charge is in Chapter 3, operational problems were observed used, the combined explosive weight of the projectile during the Schofield Barracks operations. Problems burster (0.186 kg) and the donor charge would still were encountered with the detonator initiation system. be less than the explosive containment capabilities The lime feed system appeared to not be operating of the DV65. properly and was subsequently found to be feeding The DE appears to be more than adequate, based lime too slowly. At the end of the campaign, approxi- on results to date. The DE for agent in the DAVINCH mately 50 gallons of aqueous fluid with a pH of 1 was vessel for a 50:50 mix of mustard agent and lewisite unexpectedly found in the expansion tank.13 As noted has been 99.9999 percent. In a separate test an addi- 13Communication via teleconference between Dave Hoffman, 14Personal communication between Brint Bixler, Vice President, Chemical Materials Agency (CMA), and the committee, August 18, CH2M HILL, and Richard Ayen, committee chair, August 15, 2008. 2008. 

66 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES tional agent DE of >99.99 percent (below agent detec- paign length ranging from about 2.8 years to about tion limits) was achieved in the cold plasma oxidizer 5.7 years. that treats the offgas. In theory, the two DEs could be multiplied together to obtain an overall DE. However, DAVINCH analytical limits of detection would prevent confirma- tion of that result. The DAVINCH DV65 is capable of destroying two If the DV65 is used, 7,500 shots would be required 155-mm projectiles per shot for nine shots per 10-hr to process the 15,000 155-mm mustard agent H pro- day. At this throughput of 18 projectiles per day, it jectiles at BGAD, and this large number of shots could would take 834 days, or 139 six-day weeks (2.7 years), adversely impact the inner vessel. This number of shots to destroy the 15,000 mustard agent H projectiles at is well in excess of the cumulative number of shots con- BGAD. The committee thus projected a campaign ducted in all DAVINCH vessels to date. The larger pro- length ranging from about 2.7 years to about 5.3 years posed DV120 would absorb the impact of 3,750 shots. for the DV-65. The DAVINCH manufacturer estimates These quantities should be acceptable since as noted that the larger proposed DV120 will be able to destroy above, the inner vessel is replaceable and can be rotated four 155-mm projectiles per shot, again doing this to distribute wear around its surface. Furthermore, it nine times per 10-hour day. If this estimate is correct, can be strengthened by several layers of steel plates on the processing time would be 417 days, or 1.34 years. its outer surface, as has been done with the DV50 in use The committee thus projected a campaign length rang- in Belgium. The system is not transportable. ing from about 1.3 years to about 2.7 years for the DV-120. SDC2000 SDC2000 The Dynasafe SDC2000 has also demonstrated the ability to process the types of munitions associated with The Dynasafe SDC2000 can destroy one 155-mm Requirement BG-2. DEs are high enough, although projectile per cycle and can conduct three cycles per they have not been demonstrated with the proposed hour.16 This corresponds to 30 projectiles per 10-hour new air pollution control system. Reliability has been day. Operation in this mode would result in the destruc- good during the operations in Germany. The system is tion of the 15,000 mustard agent H projectiles at BGAD not transportable. in 500 operating days (1.6 years). The committee thus projected a campaign length ranging from about 1.6 years to about 3.2 years. Requirement BG-2 Ratings for Process Efficacy For the process efficacy evaluation factor for Require- Requirement BG-2 Ratings for Process Throughput ment BG-2, the committee gave the TC-60 TDC a rat- ing of 4 and the DAVINCH DV65 and the Dynasafe For the process throughput criterion for Require- SDC2000 ratings of 9. ment BG-2, the committee assigned ratings of 8 to the TC-60 TDC and the DAVINCH DV65. The Dynasafe SDC2000 was rated at 10. Process Throughput15 TC-60 TDC Process Safety TC-60 TDC operations at Porton Down showed TC-60 TDC that one detonation every 35 minutes is possible. A 35-minute cycle would correspond to 17 detonations The TDC has been operated extensively in both per 10-hour shift. At this rate, 882 days of operation production and testing modes, and the committee is (2.83 years) would be required to destroy the 15,000 not aware of incidents causing injuries. ECBC reported projectiles. The committee thus projected a cam- 16Personal communication between Harley Heaton, Vice Presi- 15The reader is reminded that footnote b in Table 4-2 explains the dent for Research, UXB International, Inc., and Margaret Novack, use of ranges for projected throughputs. NRC, study director, July 17, 2008. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 67 that TC-60 operations were conducted safely during Public and Regulatory Acceptability in a U.S. Context their 2004-2006 testing at Porton Down (DiBerardo et al., 2007). TDC The TDC requires individual handling of the agent- The TC-60 TDC has been permitted and operated in containing munitions, manual attachment of the explo- the United States but only for the destruction of ­phosgene sives to the munitions, and manual hanging of the and chloropicrin. For this reason, although eventually a munitions in the detonation chamber. In the event of a full RCRA operating permit would be required, opera- misfire, there would be additional risk for the operations tions could be initiated under a research, development, personnel because they would have to open the cham- and demonstration (RD&D) permit. To allow continued ber containing the munitions and their donor explosives operation of the TC-60 TDC, BGCAPP’s Title V CAA to correct the cause of the misfire. Also, this technology permit would need to be modified. When obtaining the requires the storage of explosives in the vicinity of the permits for operation of the TC-60 TDC in Hawaii, there unit, which creates an additional hazard. was no public opposition. As previously discussed and as indicated in Appendix A, permits have been obtained DAVINCH for similar systems built for destruction of conventional weapons in the United States. The TC-60 TDC uses a The DAVINCH vessels, consisting of inner and outer catalytic oxidizer but no open flame in the pollution steel chambers, have been safely operated over a 6-year abatement system. The catalytic oxidizer does not period, first in Japan and later in Belgium. Some of the appear to be a liability for the public or the regulatory munitions destroyed are heavier than those found in the authorities. Noise levels are not extreme. The system is 155-mm projectiles at BGAD and contain more agent transportable, a positive factor. and explosives. To date, there have been no incidents that the committee is aware of that have compromised either worker or public safety and there have been no DAVINCH releases of agent or offgas containing residual quanti- Although the DAVINCH technology has not been ties of agent. permitted for destroying either energetics or chemical DAVINCH operations involve munition handling, agent in the United States, from a regulatory perspective, the manual placement of donor explosives around the it should be able to satisfy environmental regulations and munition, and the placement of a detonator into the obtain permitted status for processing and destroying donor explosive. When handling the mustard agent- mustard agent-filled 155-mm projectiles at BGAD. filled 155-mm projectiles at BGAD, PPE may be The use of DAVINCH for destroying the 155-mm required, although if the projectiles are inserted into projectiles will require a RCRA operating permit. Since the DAVINCH vessel using a robotic arm, as is done it has not yet been used to destroy either energetics or in Japan, that operation should not expose workers to chemical weapons in the United States, it should be an explosives or agent. ideal candidate for beginning operations under a RCRA Since donor charges are used to destroy the munition, RD&D permit. they need to be stored in the vicinity of the DAVINCH. Public perceptions of DAVINCH may be favor- This could create an additional hazard. able in light of the high degree of agent destruction (99.9999 percent for the vessel itself and an additional SDC2000 >99.99 percent destruction of any agent remaining in the offgas in the cold plasma oxidizer) and also because The Process Safety section for the SDC2000 for all process residuals can be held, tested, and recycled Requirement BG-1 applies. through the DAVINCH vessel and cold plasma oxidizer for further treatment (if needed) prior to release. Requirement BG-2 Ratings for Process Safety Since there are no DAVINCH units operating in the United States on which to base a perception, the public For the process safety factor for Requirement BG-2, reaction to using DAVINCH for destroying the 155-mm the committee assigned a rating of 7 to the TC-60 TDC, mustard agent-filled projectiles might be favorable. The a rating of 8 to the DAVINCH DV65, and a rating of 9 reaction could also, however, be ambivalent if it is per- to the Dynasafe SDC2000. 

68 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES ceived that what happens in the cold plasma oxidizer is ≤1 vapor screening level (VSL) prior to removal from a form of incineration. the detonation chamber. As explained in Appendix A, this thermal decontamination is accomplished by purg- ing the detonation chamber with hot air. SDC2000 The Dynasafe SDC2000 has not been permitted DAVINCH in the United States. It has operated in Germany, and emissions testing based on U.S. regulations has been When processing 155-mm mustard agent H projec- performed. The committee anticipates that regula- tiles, several waste streams will be produced. These tory bodies would grant a permit based on satisfac- include fragments of metal from the munition ­bodies, tory testing prior to full-scale operations. The vendor burster wells, and other metal parts; dust; small quan- indicates it would be prepared to change the design of tities of liquids from spray decontamination of the the pollution abatement system for use in the United DAVINCH vessel; condensate water from the cold States, converting the secondary combustion chamber plasma oxidizer; activated carbon from the filters; to a flameless oxidizer and making other changes (see and the treated offgases from the process. The metal Chapter 3). The committee cannot anticipate either a parts will have been explosively treated in the vessel favorable or an adverse public reaction to this system to a point where they can be released or recycled. This since it has not yet been designed, built, or tested. treatment includes exposure of the metal munition frag- ments to detonations of conventional explosives (see Appendix A). Following treatment in the cold plasma Requirement BG-2 Ratings for Public and Regulatory oxidizer, the process offgas enters a retention tank for Acceptability in a U.S. Context testing. If the quantity of agent in the offgas exceeds the The TC-60 TDC has been permitted and operated in VSL, it is recycled through the DAVINCH vessel and the United States, and no public or regulatory opposi- the cold plasma oxidizer for further treatment. tion was encountered in the process. However, it does The volumes of each waste stream resulting from not have the hold-test-release feature of the DAVINCH. the processing of 155-mm projectiles are not known, The TDC is rated a 9. but absent a large volume of liquid wastes, they are The DAVINCH has not been permitted but has a expected to be small. (Note: The total weight of all sub- hold-test-retreatment capability, which is a positive. It stances recovered from destruction of a similar muni- is rated a 7. tion, the 150-mm Clark projectile in Belgium, is 42 kg The Dynasafe SDC has not received an operating per one munition shot.) The original total munition permit in the United States, and the design of its pol- weight was 40 kg. The TC-60 TDC would be expected lution abatement system will be different from that of to produce about the same relative amount of recovered the system used at Münster; it has not, however, been substances. The Dynasafe SDC2000, with the proposed specified in full or built and tested. The Dynasafe SDC new air pollution control system, would produce more was assigned a rating of 7. recovered substances because of filter cake produc- tion. The EDS would also produce more recovered substances because of hydrolysate production. Secondary Waste Issues TDC SDC2000 The TC-60 TDC produces relatively small amounts Offgases from the SDC2000 are up to 150 Nm3/hr of secondary waste. These are described in Chapter 3. of nitrogen, water vapor, and carbon dioxide. The It is likely that the spent lime and spent pea gravel will acidic and basic scrubbers would produce no liquid not be a RCRA hazardous waste for mercury since the effluents but would produce up to 500 lb per day of Levinstein mustard agent in the projectiles at BGAD salts as a filter cake, which would need to be tested contains low levels of mercury. Overall, the second- for contaminants of regulatory concern. The scrap ary waste, including the spent activated carbon, is not metal resulting from the munition bodies is suitable expected to contain compounds of regulatory concern. for ­unrestricted release; however, this waste is a listed The scrap metal is thermally decontaminated to a waste in ­Kentucky and can therefore be sent only to a 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 69 hazardous waste treatment, storage, and disposal facil- agent fill in these projectiles would take place in the ity (TSDF) or to a recycler that is allowed to receive DAVINCH ­vessel, where in tests to date, a 99.9999 per- this waste, such as the Rock Island smelter. cent destruction and removal efficiency (DRE) has been demonstrated. Additional agent destruction in the vessel offgases takes place in the cold plasma oxidizer, where Requirement BG-2 Ratings for Secondary Waste Issues a DE of >99.99 percent of any remaining mustard agent The DAVINCH produces a relatively small amount has also been demonstrated. The process provides for the of secondary waste and is designed as a hold-test- retention and testing of postoxidizer offgases and, if need release system, which ensures that offgases are free of be, recycling them through the system for any additional agent or other compounds of regulatory concern. It is agent destruction needed before release. rated a 9. The TC-60 TDC generates a small volume of waste that is not, however, expected to contain con- SDC2000 taminants of regulatory concern. The discharged scrap metal is ≤1 VSL. It does not have a gaseous emission The Dynasafe SDC2000 at Münster is not designed hold-test-release feature. The TC-60 TDC is rated an for nor does it operate in a hold-test-release mode. As 8. The amount of secondary waste produced by the indicated in the Dynasafe description in Appendix A proposed SDC2000 system for BGCAPP is not well and as more recently confirmed by the vendor, such known because of the proposed changes to the air pol- a system has been designed but has not been built lution control system, but it promises to be modest. or tested.18 Operation in the hold-test-release mode However, it will not be as low as the secondary waste reduces the throughput rate by an unspecified amount. from the TC-60 TDC or the DAVINCH. The discharged The throughput rates presented in this report do not scrap metal is acceptable for unrestricted use. The apply to operation in the hold-test-release mode. Dynasafe SDC2000 was rated a 7. As reported in Chapter 3, testing was carried out on the Dynasafe SDC2000 system at Münster, ­Germany, to determine the DE for mustard agent HD. Three Destruction Verification Capability HD destruction tests were conducted using 100-mm mortar rounds. The results of the 3-day HD tests TC-60 TDC showed the absence of any detectable agent near the As reported in Chapter 3, well-monitored testing was end of the process and an overall DE for the sys- carried out at Porton Down during March 2006. During tem of >99.999999989 percent (nine nines). DEs this testing, 101 mustard agent-containing 25-pounder for the detonation chamber only were measured at projectiles were destroyed. No agent was detected in 99.99481 percent and 99.99508 percent. the offgas from the pollution abatement system, and the overall DE was determined to be >99.9999 percent.17 Requirement BG-2 Ratings for Destruction Verification Because the munition bodies are shattered into small Capability pieces by the detonation, they cannot be reused. The TC-60 TDC presently does not have the ability For evaluating the verifiability of the destruction for to hold and test the gases it generates before they are Requirement BG-2, the committee assigned a rating of released to the atmosphere. 9 to the TC-60 TDC, a rating of 10 to the DAVINCH DV65, and a rating of 9 to the Dynasafe SDC2000. DAVINCH Process Flexibility The DAVINCH technology would destroy the 155-mm projectile bodies by fragmenting them through Process flexibility was considered to be not appli- use of donor explosives. As a result, the munition cable to Requirement BG-2, which involves a single b ­ odies could not be refilled. Destruction of the mustard feedstock, 155-mm mustard agent H-filled projectiles. 17Brint Bixler, Vice President, CH2M HILL, “Destruction of 18Personal communication between Harley Heaton, Vice Presi- chemical weapons using CH2M HILL’s transportable detonation dent for Research, UXB International, Inc., and Margaret Novack, chamber,” presentation to the committee, May 8, 2008. NRC, study director, July 17, 2008. 

70 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES Summary Assessment for Requirement BG-2 inner vessel and (2) periodic rotation of the inner vessel to distribute the impacts of high-velocity metal parts TC-60 TDC on the vessel walls. If need be, the inner vessel can be replaced per the DAVINCH design. The TC-60 TDC could execute Requirement BG-2 The DAVINCH technology has not been permitted in a reasonable time as a consequence of its higher in the United States, hence RCRA and other permits throughput. The high probability of being able to obtain and appropriate performance testing will have to be an operating permit is also an advantage. Its current obtained for any application. In the United States, pub- lack of an ability to hold-test-release is a disadvantage, lic awareness of the DAVINCH is low and reactions to as is the need to store explosives near the unit during its characteristicsthermal treatment in both the vessel operations. The need for additional upgrading of the and in the plasma oxidizerare unknown. unit operations downstream of the detonation cham- ber to improve robustness and reliability had a slight adverse impact on its rating for maturity. SDC2000 The Dynasafe SDC2000 also could execute Require- DAVINCH ment BG-2 in a reasonable time based on its through- put. In addition, it has a proven record of destroying a The DAVINCH technology is capable of destroying similar number of munitions in Germany without any the approximately 15,000 mustard agent-filled 155-mm reported problems. It has a distinct safety advantage projectiles at BGAD. It has destroyed more than 2,500 for Requirement BG-2 as it minimizes handling by the comparable bombs and projectiles, some containing operating staff and does not necessitate storing addi- mustard agent, in previous and ongoing applications tional explosives in proximity to the unit during opera- overseas. The large number of projectiles, combined tions. It has not been permitted in the United States, and with the relatively low processing rate of the DAVINCH this introduces uncertainty. The vendor has indicated technology (estimated destruction rate of 18 per day in that a new pollution control system would be used. This an existing unit and 36 per day in a proposed longer system has been outlined but not described in detail. It version), imply a campaign length range of about 2.7 to has not yet been designed, built, or tested. about 5.3 years for the DV65 and about 1.3 to about 2.7 years for the proposed DV120, assuming that the throughput rates claimed by the technology developer Overall Ratings for Requirement BG-2 can be maintained over these time periods. The large The TC-60 TDC received a summed rating of 53 number of detonations involved could adversely affect out of a possible 70. The DAVINCH DV65 and the the DAVINCH inner vessel, although as noted above D ­ ynasafe SDC2000 received summed ratings of 59 and for Requirement BG-1, this is mitigated by (1) four 58, respectively (see Table 4-4). layers of steel placed around the outer perimeter of the TABLE 4-4  EDT Ratings Summary for Requirement BG-2, Destruction of 15,000 Mustard Agent H-Filled 155-mm Projectiles at Blue Grass Evaluation Factor Public and Regulatory Secondary Destruction Process Process Process Process Acceptability in Waste Verification Process EDT Maturity Efficacy Throughput Safety a U.S. Context Issues Capability Flexibility Total TC-60 TDC 8 4  8 7 9 8  9 N/A 53 DAVINCH DV65 8 9  8 8 7 9 10 N/A 59 SDC2000 7 9 10 9 7 7  9 N/A 58 NOTE: The above values for each evaluation factor are the average of each committee member’s rating on a scale of 0-10. These average values were then summed to arrive at the totals given in the last column. Small differences in the summed ratings, up to about five points, were not considered to be significant by the committee. There was no weighting. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 71 Finding 4-3. The DAVINCH DV65 and the Dynasafe Requirement BG-3 Ratings for Process Maturity SDC2000 are rated approximately equally and slightly For the process maturity criterion for Requirement higher than the TC-60 TDC for Requirement BG-2. BG-3, the committee assigned a rating of 6 to the D-100 and TC-60 TDC combination, a rating of 8 to Recommendation 4-3. The Army should give prefer- the DAVINCH DV65, and a rating of 7 to the Dynasafe ence to the use of the DAVINCH DV65 or the ­Dynasafe SDC2000. SDC2000 for Requirement BG-2, the destruction of 15,000 mustard-filled projectiles at BGCAPP. The TC-60 TDC is rated lower but would also be Process Efficacy acceptable. D-100 and TC-60 TDC Combination Requirement BG-3: Destruction of See the discussion in the section “Process Efficacy” Approximately 70,000 NonContaminated for the D-100 for Requirement BG-1 and the discussion M55 Rocket Motors and approximately in the section of the same name for the TC-60 TDC for 15,000 Mustard Agent H-filled 155-mm Requirement BG-2. Projectiles at Blue Grass As was the case for Requirement BG-1, the EDS DAVINCH is not evaluated for Requirement BG-3 because it is The discussions provided for Requirements BG-1 not able to destroy rocket motors. The D-100 is not and BG-2 apply. able to process mustard agent-filled projectiles and is likewise not evaluated. As noted in the introduction to this chapter, however, the combination of two CH2M SDC2000 HILL technologies, the D-100 and the TC-60 TDC, is See the discussions in the “Process Efficacy” sec- evaluated for Requirement BG-3. The D-100 is used tions for Requirements BG-1 and BG-2. for destruction of the 70,000 noncontaminated M55 rocket motors, and the TC-60 TDC is used for the destruction of the 15,000 mustard agent-filled 155-mm Requirement BG-3 Ratings for Process Efficacy projectiles. To evaluate process efficacy for Requirement BG-3, the committee rated the D-100 and TC-60 TDC com- Process Maturity bination at 7 and both the DAVINCH DV65 and the Dynasafe SDC2000 at 9. (Table 4-2, footnote b, explains D-100 and TC-60 TDC Combination the use of ranges for projected throughputs.) See the discussion in the section “Process Maturity” for the D-100 for Requirement BG-1 and the discussion Process Throughput in that same section for the TC-60 TDC for Require- ment BG-2. D-100 and TC-60 TDC Combination The committee projects the D-100 would require DAVINCH about 1.2 to about 2.5 years to destroy the 70,000 noncontaminated M55 rocket motors, and the TC-60 The discussions provided for Requirements BG-1 TDC would require about 2.8 to about 5.7 years to and BG-2 apply. destroy the 15,000 mustard agent-filled projectiles. However, parallel operation is possible. Thus, depend- SDC2000 ing on whether the campaigns are done sequentially or in parallel, the projected campaign length range would See the discussions for the SDC2000 in the “Process be either 2.8 to 5.7 years for parallel operation or 4.1 to Maturity” sections for Requirements BG-1 and BG-2. 8.2 years for sequential operation. 

72 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES DAVINCH Process Safety Processing 70,000 rocket motors and 15,000 155-mm D-100 and TC-60 TDC Combination projectiles in the DV65 at the throughput rates provided by the manufacturer (36/day and 18/day, respectively) See the discussion in the section “Process Safety” for would take 2,779 days or 463 weeks (8.9 years) the D-100 for Requirement BG-1 and the discussion in assuming a 6-day operating week. The committee thus the section of that same name for the TC-60 TDC for projects a campaign length range of about 8.9 to about Requirement BG-2. 17.8 years. Using the larger proposed DV120, the time required would be 1,389 days or 232 weeks (4.5 years), DAVINCH again for a 6-day operating week and assuming that the manufacturer’s stated throughput rates of 72 rocket See the discussions in the section “Process Safety” motors and 36 155-mm projectiles per day can be for the DAVINCH for Requirements BG-1 and BG-2. achieved and sustained. The committee thus projects a campaign length range of 4.5 to 8.9 years. SDC2000 The discussions in the sections “Process Safety” SDC2000 for Requirements BG-1 and BG-2 for the SDC2000 The Dynasafe SDC2000 could destroy the 70,000 apply. noncontaminated rocket motors in 700 days at a rate of 100 per 10-hour day and the 15,000 mustard agent- Requirement BG-3 Ratings for Process Safety filled projectiles in 500 days at a rate of three per hour in a 10-hour day. A total of 1,200 operating days For the process safety criterion for Requirement (200 6-day weeks, 3.85 years) would be required. BG-3, the committee assigned a rating of 7 to the D-100 and TC-60 TDC combination, a rating of 8 to the DAVINCH DV65, and a rating of 9 to the Dynasafe Requirement BG-3 Ratings for Process Throughput SDC2000. Throughput rates for each technology and require- ment are given in Table 4-2, as are the schedule implica- Public and Regulatory Acceptability in a U.S. Context tions for these rates. These are best-case times and do not reflect downtimes for scheduled and unscheduled D-100 and TC-60 TDC Combination maintenance, facility downtime, ramp-ups, and change- overs. They are only intended to illustrate the relative See the discussion in the section “Public and Regu- times required for destroying the noncontaminated latory Acceptability” for the D-100 for Requirement rocket motors and/or 155-mm mustard agent-filled BG-1 and the discussion in the section of the same projectiles at BGAD using each technology. name for the TC-60 TDC for Requirement BG-2. Requirement BG-3 is the requirement that benefits the most from employing a technology with a high DAVINCH throughput. The projected campaign length range for the D-100 and TC-60 combination would be about The earlier discussions for Requirements BG-1 and 2.8 to about 5.7 years if the campaigns are done in BG-2 apply to the regulatory aspects of destroying parallel or about 4.1 to about 8.2 years if the campaigns M55 rocket motors and mustard agent projectiles in are done sequentially. This combination was assigned a the DAVINCH. rating of 8. The projected campaign length range for the DV65 is about 8.9 to about 17.8 years, and the projected SDC2000 campaign length range for the proposed DV120 is about 4.5 to about 8.9 years. The DAVINCH technology was The discussions for Requirements BG-1 and therefore rated a 5. The projected campaign length BG-2 apply. When processing noncontaminated rocket range for the SDC2000 is about 3.9 to about 7.7 years, motors, the scrubber salts might be hazardous owing to and it was rated a 9. the presence of lead. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 73 Requirement BG-3 Ratings for Public and Regulatory D-100 and TC-60 TDC Combination Acceptability in a U.S. Context See the section “Destruction Verification Capability” The D-100 and TC-60 TDC combination was given for Requirement BG-2 for the TC-60 TDC. a rating of 9 because both systems have been through the permitting process in the United States—the D-100 DAVINCH at Milan Army Ammunition Plant and the TC-60 at Schofield Barracks. The TC-60 has had some operating See the section “Destruction Verification Capability” experience in the United States, and the D-100 has had for Requirement BG-2 for DAVINCH DV65. considerable experience. The DAVINCH was assigned a lower rating, 7, since it has not yet received an oper- SDC2000 ating permit in the United States. The SDC2000 was also rated at 7. See the section “Destruction Verification Capability” for Requirement BG-2 for the SDC2000. Secondary Waste Issues Requirement BG-3 Ratings for Destruction Verification D-100 and TC-60 TDC Combination Capability See the discussion in the “Secondary Waste Issues” For the destruction verification evaluation factor for section for the D-100 for Requirement BG-1 and the Requirement BG-3, the committee assigned a rating of discussion in the “Secondary Waste Issues” section for 8 to the D-100 and TC-60 TDC combination, a rating the TC-60 TDC for Requirement BG-2. of 10 to the DAVINCH DV65, and a rating of 9 to the SDC2000. DAVINCH Process Flexibility The same waste streams noted in the previous dis- cussions on use of the DAVINCH for Requirements D-100 and TC-60 TDC Combination BG-1 and BG-2 apply. The D-100 and TC-60 TDC combination provides sufficient flexibility to destroy both rocket motors and SDC2000 projectiles. The discussions for Requirements BG-1 and BG-2 apply. DAVINCH The DAVINCH technology possesses sufficient flex- Requirement BG-3 Ratings for Secondary Waste ibility to destroy both rocket motors and projectiles. Issues For the secondary waste issues evaluation factor for SDC2000 Requirement BG-3, the committee assigned a rating of 8 to the D-100 and TC-60 TDC combination, a rating One issue previously discussed is that the NEW of 9 to the DAVINCH DV65, and a rating of 7 to the limit for the SDC2000 system at Münster is limited by SDC2000. permit to 2.3 kg, which is one-fourth the NEW of the rocket motor. For a new system constructed just for BGCAPP, Dynasafe claims the NEW limit can be up Destruction Verification Capability to 10 kg depending on choice of inner chamber. This This criterion is not applicable to the destruction is still relatively close to the NEW of a rocket motor, of noncontaminated M55 rocket motors (Requirement which is 8.8 kg propellant. BG-1) but is applicable to mustard agent-filled projec- The Dynasafe SDC2000 has the ability to destroy tiles (Requirement BG-2). 155-mm mustard agent H projectiles. Again, with only the one agent-filled feedstock, process flexibility is 

74 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES not a significant issue. Leaking munitions will have destroying both bombs and projectiles in applications been previously identified and placed in overpacks. in two countries, and is capable of achieving high Depending on DDESB approvals and other factors, DREs, reducing agent concentrations to below limits overpacked munitions may or may not be processed in of detection. It has the ability to hold and test waste the SDC2000. streams and, if necessary, to reprocess gaseous waste streams by recycling them through the vessel and offgas treatment system. Requirement BG-3 Ratings for Process Flexibility The D-100 and TC-60 TDC combination, the SDC2000 SDC2000, and the DAVINCH DV65 are all sufficiently flexible and are rated 9. The Dynasafe SDC2000 is also a strong candidate for Requirement BG-3. This EDT could execute Require- ment BG-3 in a reasonable length of time based on its Summary Assessment for Requirement BG-3 throughput. It has destroyed large numbers of muni- tions in Germany without any reported problems. It has D-100 and TC-60 TDC Combination a distinct safety advantage with respect to Requirement The D-100 and TC-60 TDC combination is a strong BG-3 because it minimizes handling by the operating candidate for Requirement BG-3. The projected cam- staff and does not require the storing of additional paign length for the combination would be about donor explosives near the unit during operations. It has 2.8 to about 5.7 years if the campaigns are carried not been permitted in the United States, however, and out in ­parallel or about 4.1 to about 8.2 years if done this introduces uncertainty. The vendor has indicated sequentially. Again, the Army must continue to upgrade that it will use a new pollution control system, which the unit operations of the TC-60 TDC. The D-100 and has been outlined but not designed, built, or tested. similar systems have carried out campaigns of the same magnitude as would be encountered for Requirement Overall Ratings for Requirement BG-3 BG-3. The summed rating for the D-100 and TC-60 com- bination is 62, the summed rating for the DAVINCH DAVINCH DV65 is 65, and the summed rating for the SDC2000 The DAVINCH technology would be a strong candi- is 66. The EDS is not suitable for Requirement BG-3. date for Requirement BG-3, especially if the proposed Thus, the D-100 and TC-60 TDC combination, the DV120 vessel were used to increase throughput. The DAVINCH DV65, and the SDC2000 are all rated about technology is robust, has operating experience in the same and are all viable candidates (Table 4-5). TABLE 4-5  EDT Ratings Summary for Requirement BG-3, Destruction of Approximately 70,000 Noncontaminated M55 Rocket Motors and 15,000 Mustard Agent H-Filled 155-mm Projectiles at Blue Grass Evaluation Factor Public and Regulatory Secondary Destruction Process Process Process Process Acceptability in Waste Verification Process EDT Maturity Efficacy Throughput Safety a U.S. Context Issues Capability Flexibility Total D-100 and TC-60 6 7 8 7 9 8  8 9 62 TDC combination DAVINCH DV65 8 9 5 8 7 9 10 9 65 SDC2000 7 9 9 9 7 7  9 9 66 NOTE: The above values for each evaluation factor are the average of each committee member’s rating on a scale of 0-10. These average values were then summed to arrive at the totals given in the last column. Small differences in the summed ratings, up to about five points, were not considered to be significant by the committee. There was no weighting. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 75 Finding 4-4. The CH2M HILL D-100 and TC-60 TDC Process Maturity combination, the DAVINCH DV65, and the Dynasafe SDC2000 technologies are rated approximately the TC-60 TDC same and are all acceptable candidates for Requirement See the TC-60 “Process Maturity” section for BG-3, although the time needed for use of a single Requirement BG-2. DV65 operating 60 hours per week might be considered excessively long by the Army. All will require testing or further testing before a final selection can be made. DAVINCH This is a mature technology for chemical agent Recommendation 4-4. If the results of testing on destruction that demonstrated the ability to destroy rocket motor destruction are favorable for all of the artillery projectiles in Belgium that were similar to explosive destruction technologies suitable to this task, the 105-mm and 155-mm projectiles at Pueblo and to the Army could use either the CH2M HILL D-100 and destroy recovered bombs in Japan that had a 50 percent TC-60 TDC combination, the DAVINCH DV65, or mustard agent fill, the same agent stored at PCD. In the Dynasafe SDC2000 technology for Requirement both Japan and Belgium, donor explosives were suc- BG-3. The campaign length for use of a single DV65 cessfully used to shatter thick-walled steel munitions operating at 60 hours per week might be considered (the largest Belgian shell was 28 mm thick). The largest excessively long by the Army. munition destroyed in Japan, the 50-kg Yellow bomb, had a diameter of 200 mm and a length of 1 m. The Requirement P-1: Destruction of all largest item destroyed in Belgium, the 21-cm shell, had leakers and reject munitions at Pueblo a diameter of 210 mm and a length of 0.8 m. Both of Comprising approximately 1,000 rounds these exceed the size of the 155-mm M104 and M110 of Mustard Agent HD/HT-filled MUNITIONS projectiles at PCD. The manufacturer has done testing (mixture of 4.2-in. mortars and 105- and on the disposal of munitions in overpacks and muni- 155-mm projectiles) tions encased in concrete during a campaign in Belgium concurrent with the preparation of this report. Table 1-3 in Chapter 1 lists overpacked munitions DAVINCH has not been RCRA-permitted in the currently stored at Pueblo Chemical Depot (PCD). United States. It would have to undergo the equiva- Most are 105-mm and 155-mm HD- or HT-filled pro- lent of trial burns with agent or an agent surrogate. jectiles.19 This list is expected to grow to about 1,000 This could delay implementation of the DAVINCH at munitions as destruction of munitions proceeds in the PCD. main processing unit at the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP) facility. According to the PMACWA, processing these units in an EDT will SDC2000 significantly shorten the schedule and reduce risk to the See the SDC2000 discussion in the section “Process operating staff by minimizing the requirement for inter- Maturity” for Requirement BG-2. SDC2000 is a mature mediate storage with multiple handling requirements. technology for destruction of this type of chemical The EDT can also be used to destroy contaminated weapon. As indicated in Chapter 3, over 13,000 recov- energetics and bursters, as indicated in Figure 1-2. ered munitions were destroyed at the Münster facility. The processing of these approximately 1,000 rounds Also described in Chapter 3 is the 3-day test series can be spread out over a long period of time (1 year at carried out at Münster, Germany, to demonstrate that least). Thus, a significant characteristic of Requirement the Dynasafe system could effectively destroy mustard P-1 is that it does not require as high a throughput rate agent-filled munitions. The technology has not been as Requirements BG-1, BG-2, and BG-3. In addition, demonstrated in the United States. Also, the manufac- the explosive load of these munitions is relatively turer has indicated it will modify the air pollution con- small. trol system used at Münster for use in the United States. The modified system must therefore be designed, built, and tested. The manufacturer claims that munitions in 19HT is distilled mustard mixed with bis[2-(2chloroethylthio) overpacks can be fed directly into the system. ethyl] ether. 

76 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES EDS-2 to dispose of the leakers and rejects in at least some of the overpacks used. The DV60 system (nearly identi- The EDS Phase 2 (EDS-2) is a mature technology for cal to the DV65) has been reliable and robust to date, chemical agent destruction and has been demonstrated having been used to destroy more than 1,600 bombs in the United States for all the weapons types that in Japan that were filled with a 50:50 mustard:lewisite would be encountered at PCD. It has performed very agent mix. satisfactorily in an assignment similar to Requirement P-1—namely, the destruction of over 1,200 old chemi- cal munitions at Pine Bluff Arsenal. In the course of SDC2000 doing so, an efficient procedure for operating paired The Dynasafe SDC2000 has demonstrated the EDS-2s was developed. In the current evaluation, it is ability to process the types of munitions associated assumed that a pair of EDS-2s would be operated at with Requirement P-1. However, it has not been PCD to provide an adequate throughput for Require- demonstrated in the United States. DEs are high, with ment P-1. Because the EDS-2 donor charge is used final agent concentrations below limits of detection, solely to cut open the munition and detonate the burster although they have not been demonstrated with the charge, the quantity of explosive is relatively small. new air pollution control system proposed for use in the The EDS-2 is routinely operated in a hold-test-release United States. Reliability was good during the opera- mode. tions in Germany. The throughput rate of the EDS-2 is low, especially for large munitions like the 155-mm projectile, and the process produces more liquid waste than the vendor- EDS-2 supplied EDTs, but neither problem seems serious in The EDS-2 has proven it is able to process the types the context of the task to be done at PCD. of munitions that are associated with Requirement P-1. Agent is destroyed to below acceptable levels, typically Requirement P-1 Ratings for Process Maturity 1 VSL. The system is transportable, robust, and very reliable. For the process maturity evaluation factor for Requirement P-1, the committee assigned a rating of 8 to both the TC-60 TDC and the DAVINCH DV65, a Requirement P-1 Ratings for Process Efficacy rating of 7 to the Dynasafe SDC2000, and a rating of For the process efficacy factor for Requirement 10 to the EDS-2. P-1, the committee assigned a rating of 4 to the TC-60 TDC, ratings of 9 to both the DAVINCH DV65 and the Process Efficacy D ­ ynasafe SDC2000, and a rating of 10 to the EDS-2. TC-60 TDC Process Throughput See the “Process Efficacy” section for the TC-60 TDC under Requirement BG-2. TC-60 TDC The TC-60 TDC has demonstrated throughput of one DAVINCH munition per 35-minute cycle in operations at Porton Down in the United Kingdom. At this rate and assuming The DAVINCH technology should be able to pro- that one munition is destroyed per cycle, 17 munitions cess the roughly 1,000 leaking and reject munitions would be destroyed per 10-hour day, or 102 muni- at PCD. It has, to date, safely destroyed over 2,500 tions per 6-day operating week. The 1,000 munitions chemical bombs and projectiles in applications in Japan at Pueblo would be destroyed in about 10 weeks. The and Belgium, some of which have been larger and campaign is projected to last about 10 to 20 weeks. contained more explosives than the largest projectile Even if the throughput is decreased significantly when to be destroyed at PCD (2.75 kg TNT-equivalent in munitions in overpacks are processed, the rates should the 150-mm shell vs. 0.19 kg TNT-equivalent in the still be more than adequate. 155-mm projectile). The explosive capacity is adequate 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 77 DAVINCH EDS-2 The time required for processing leaking projectiles The EDS-2 has a relatively low throughput of one and mortar rounds in a DAVINCH vessel at Pueblo will 155-mm projectile every 2 days but can destroy six depend on the number of each type of munition, the 4.2-in. mortars in the same period. It has been demon- DAVINCH unit used, and, possibly, the configuration strated that two 105-mm projectiles can be destroyed of the munitions—for example, the type of overpack per detonation, but it is likely that six 105s can be used. The DAVINCH manufacturer claims that the done at once.21 In ongoing operations at the Pine Bluff DV65 can process six 4.2-in. mortar rounds per shot Arsenal, three EDSs—one EDS Phase 1 (EDS-1) and and six 105-mm projectiles per shot, both for nine shots two EDS-2s, only two of them operated at a time— per 10-hour day. For the larger 155-mm projectile, two destroyed 1,065 munitions in less than 3 years. If, as items would be processed per shot, again at a rate of in the throughput calculation for the DAVINCH, it is nine shots per day. assumed that the 1,000 munitions at PCD are equally If the inventory of leaking munitions at Pueblo con- divided among mortars, 105-mm projectiles, and sists of about 1,000 items (about 500 known leakers 155-mm projectiles, destroying the 155-mm projectiles and a similar number of yet-to-be-found leakers and would take 333 operating cycles, or 666 days. The reject munitions) and if it is assumed that they exist mortars would require 56 cycles, or 112 days. If the in equal proportions (one-third 4.2-in. mortar rounds, 105-mm projectiles can be done six at a time (not yet one-third 105-mm projectiles, and one-third 155-mm verified), these items would require 56 cycles, or 112 projectiles), then at the processing rates claimed, it operating days. Overall, it would require 890 operating would take 6.1 days to destroy the mortar rounds, days to destroy all the munitions under consideration another 6.1 days to destroy the 105-mm projectiles, at PCD. The projected campaign will take from about and 19 days to destroy the 155-mm projectiles. The 2.9 to about 5.7 years. A pair of EDS-2s operating as at total time would be 31 days, or about 5 6-day operat- Pine Bluff could complete the mission in 445 operating ing weeks. The projected campaign length range is thus days, for a projected campaign length of about 1.4 years from about 5 weeks to about 10 weeks. to about 2.9 years. The EDS can theoretically dispose of munitions in some of the overpacks used. This requires that larger SDC2000 shaped charges be used to cut through both the over- Assume as for the DAVINCH that about 333 each pack and the munition. This process is complicated by of 155-mm projectiles, 105-mm projectiles, and 4.2-in. the need to accurately aim the shaped charges at the mortar rounds are to be destroyed. The throughput rates munition, which cannot be seen inside the overpack, given in Table 4-7 in Appendix A are 40 munitions per and by the possibility that the munitions in the over- 10-hour day for 155-mm projectiles and 120 muni- packs will not open properly. This is likely to adversely tions per day for both 105-mm projectiles and 4.2-in. affect throughput and safety and adds complexity and mortars. Thus, the total operating time is about 15 days, uncertainty to the process of disposing of overpacked and the projected campaign length range is about 5 to munitions in the EDS. Removing the overpacked muni- 10 weeks. These are very short times, and they argue tions from their overpacks before processing might be for using a smaller system than the SDC2000. It would preferable. be especially beneficial if a transportable version of the Dynasafe technology could be used, although the Requirement P-1 Ratings for Process Throughput committee is not aware that any such system exists. The throughput rate for overpacked munitions of the The throughput of the EDS-2 is marginal for Require- size that will be processed at PCD is anticipated to be ment P-1. If only one EDS-2 is used, the campaign is about one munition per hour.20 projected to last from about 2.8 to about 5.7 years. Alternatively, the Army could choose to use more 20Personal communication between Harley Heaton, Vice Presi- 21Personal communication between Allan Caplan, System Devel- dent for Research, UXB International, Inc., and Margaret Novack, opment Group Leader, Non-Stockpile Chemical Materiel Project, NRC, study director, July 17, 2008. and Margaret Novack, NRC, study director, August 19, 2008. 

78 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES than one EDS, significantly reducing the length of the Requirement P-1 Ratings for Process Safety campaign. The EDS-2 is rated 10. The TC-60 TDC, For the process safety factor for Requirement the DAVINCH DV65, and the Dynasafe SDC2000 all P-1, the committee rated the TC-60 TDC at 7, the have more than adequate throughput capacity for this DAVINCH DV65 at 8, the Dynasafe SDC2000 at 9, requirement and are also rated 10. and the EDS at 7. Process Safety Public and Regulatory Acceptability in a U.S. Context TC-60 TDC See the TC-60 section “Process Safety” for Require- TC-60 TDC ment BG-2. See the TC-60 section “Public and Regulatory Acceptability” for Requirement BG-2. DAVINCH The same factors affecting process safety that were DAVINCH discussed for Requirement BG-2 (processing mustard The same factors involved in evaluating the public agent-filled 155-mm projectiles at BGAD) apply to and regulatory acceptability of the DAVINCH for the processing of the three munition types at PCD processing 155-mm mustard agent-filled projectiles at since the same DAVINCH operations are involved. Blue Grass should also apply to Pueblo. The large explosive containment capacity allows the DAVINCH system to dispose of munitions in some of the overpacks used. SDC2000 See the section “Public and Regulatory Acceptabil- SDC2000 ity” for Requirement BG-2 for the SDC2000. See the section “Process Safety” for Requirement BG-2 for the SDC2000. The manufacturer claims that EDS-2 the munitions in overpacks can be fed directly into the The EDS systems (EDS-1 and EDS-2) have been system and that the high temperature in the chamber permitted for use at several locations in the United will cause the overpacks and munitions to be breached States. They have not experienced significant public and the agent released. opposition even for their use in urban locations. In addi- tion, the EDS has already received regulatory approval EDS-2 for operation in Colorado for destroying GB-filled bomblets. Its routine use in a hold-test-release mode The EDS-2 necessitates individual handling of the and the absence of an oxidizing offgas treatment opera- munitions and manual attachment of shaped charges tion have contributed to EDS acceptance. The DDESB to the munitions. The likelihood of a misfire is greatly has approved it on a systemwide basis. reduced by redundant firing circuits. This technology requires the storage of modest quantities of explosives in the vicinity of the unit, which creates an additional Requirement P-1 Ratings for Public and Regulatory hazard. The destruction of a munition in an overpack Acceptability in a U.S. Context through the use of larger shaped charges has been The TC-60 TDC has been permitted in the United demonstrated but is not done routinely at Pine Bluff States and encountered no public or regulatory opposi- and is not feasible for some overpacked munitions at tion. However, because it does not have as much expe- PCD. The added complexity and uncertainty associated rience in the United States as the EDS, it is rated at 9. with destroying munitions in overpacks might create a The DAVINCH DV65 was assigned a still lower rating hazard if the munitions cannot be fully opened. Thus, of 7 because it has not been permitted in the United removal of the overpacked munitions from their over- States. The Dynasafe SDC2000 would have a pollution packs before processing might be preferable. 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 79 abatement system that is not completely described and EDS-2 has not been built or tested. It is rated a 7. The EDS is The EDS-2 produces between 8 and 10 gallons of rated a 10 because it has been granted several operating liquid secondary waste per detonation. This puts it at a permits in the United States and has had no significant disadvantage in comparison with the vendor-supplied public opposition to its use. technologies. However, for Requirement P-1, where there are only a small number of munitions and a much Secondary Waste Issues larger volume of liquid secondary waste is produced in the main processing units of PCAPP, this disadvan- The mercury concentrations in the HD and HT mus- tage seems minimal. The concentration of agent in the tard agent contained in the munitions at PCAPP are liquid waste is measured to ensure it is low enough to expected to be significantly higher than concentrations be released. The ability to control mercury emissions in the H mustard agent contained in the munitions at (see discussion on mercury concentrations above) has BGCAPP.22 Thus, wastes generated at an EDT instal- been demonstrated. lation at PCAPP by any of the candidate technologies The solid wastes are primarily scrap metal from should be tested for mercury to determine if concentra- destruction of the munition bodies, bursters, and fuzes. tions are above levels of regulatory concern. The release level for this material is ≤1VSL. If prob- lems arise with residual mustard contamination in the TC-60 TDC scrap metal, the metal could be decontaminated by thermal treatment in the main plant. See the “Secondary Waste Issues” section discussion of the TC-60 TDC for Requirement BG-2. Also, note the discussion on mercury concentrations above. Requirement P-1 Ratings for Secondary Waste Issues For the secondary waste issues criterion for Require- DAVINCH ment P-1, the committee assigned a rating of 8 to the TC-60 TDC, a rating of 9 to the DAVINCH DV65, a See the section “Secondary Waste Issues” for rating of 7 to the Dynasafe SDC2000, and a rating of Requirement BG-2 for the DAVINCH DV65. Also, 6 to the EDS-2. mercury concentrations are discussed in the paragraph before last. Although the actual volume and constituents of the Destruction Verification Capability waste streams generated can be estimated, this has yet to be done since they will depend on the nature of the TC-60 TDC overpacks used and the internal constituents of the See the TC-60 “Destruction Verification Capability” munitions. section for Requirement BG-2. SDC2000 DAVINCH Note the preceding discussion of mercury concen- See the “Destruction Verification Capability” section trations. The scrap metal resulting from the munition for Requirement BG-2 for the DAVINCH DV65. b ­ odies is suitable for unrestricted release; however, it is a listed waste in Colorado and can therefore be sent only to a hazardous waste TSDF or to a recycler SDC2000 allowed to receive it, such as the Rock Island smelter. See the “Destruction Verification Capability” section for Requirement BG-2 for the SDC2000. EDS-2 22Personal communication between Richard Ward, Chief Sci- The EDS-2 has the ability to hold, test, and verify entist, PMCSE, CMA, and Richard Ayen, committee chair, at the that agent destruction has been completed to the extent CMA Committee meeting on September 19, 2008. 

80 ASSESSMENT OF EXPLOSIVE DESTRUCTION TECHNOLOGIES required before the secondary liquid waste is released manufacturer claims that the munitions in overpacks from the unit and passed to storage. Munition bodies can be fed directly into the system and that the high are destroyed. temperature in the chamber will cause the overpacks and munitions to be breached and the agent released. Requirement P-1 Ratings for Destruction Verification Capability EDS-2 The committee assigned ratings of 10 to the EDS and The EDS-2 has been demonstrated to destroy all the the DAVINCH for this factor because of their ability to types of munitions specified in requirement P-1. It has a hold and test effluents prior to release. The Dynasafe low throughput of one 155-mm projectile every 2 days SDC2000 and the TC-60 TDC received lower ratings but can destroy six 4.2-in. mortars in the same period. of 9. However, for Requirement P-1, this is not a significant concern. It can destroy some munitions in overpacks, but the 12 × 56-in. single round containers are too large Process Flexibility to fit in the EDS-2 chamber. Only 31 105-mm projec- tiles are singly overpacked in these large single round TC-60 TDC containers. They could be unpacked for destruction, as The TC-60 TDC is highly flexible in the size and is done at Pine Bluff. number of munitions that it can process. It is expected to be able to destroy the munitions in their overpacks Requirement P-1 Ratings for Process Flexibility if destruction in overpacks is allowed by the applicable regulatory permits and DDESB. All four technologies have adequate flexibility for Requirement P-1 and were rated at 10. DAVINCH Summary Assessment for Requirement P-1 The DAVINCH technology is flexible in that the vessel size can be adjusted to accommodate the explo- See Table 4-6 for a summation of the overall ratings sion containment requirements for the three kinds of for Requirement P-1. munitions to be destroyed at Pueblo, the quantity of donor explosives needed can be adjusted to ensure TC-60 TDC that the agent is accessed, and a variety of agents can be destroyed in the vessel (although this last capabil- The TC-60 TDC could execute Requirement P-1 and ity is not necessary for Requirement P-1). The large is expected to be able to destroy munitions in overpacks explosive containment capacity allows the DAVINCH if allowed by permits and DDESB approval. system to dispose of munitions in some of the over- packs used. DAVINCH The impact on processing operations of handling, placing explosive charges around, and destroying over- The DAVINCH technology should be capable of packed munitions needs to be demonstrated. Leaking destroying the roughly 1,000 leaking and reject muni- munitions may require handlers to wear a higher level tions (projectiles and mortar rounds) at PCD since of PPE, observe more stringent safety precautions, and it has destroyed a greater number of similar items allow more time per shot for placing donor and shaped elsewhere and has demonstrated the ability to destroy charges, especially if the munitions are contained mustard agent. The time required to accomplish this within overpacks. As a result, throughput rates could be should be well within the time available. The technol- lower than those estimated by the manufacturer. ogy has not been permitted in the United States. The public is not very aware of the DAVINCH technology, but it is nonetheless likely to be accepting of it. SDC2000 The Dynasafe SDC2000 has great flexibility in the size and number of munitions that it can process. The 

RATING OF EXPLOSIVE DESTRUCTION TECHNOLOGIES 81 TABLE 4-6  EDT Ratings Summary for Requirement P-1, Destruction of All Leakers and Reject Munitions at Pueblo Comprising Approximately 1,000 Rounds of Mustard Agent HD/HT-Filled Munitions (Mixture of 4.2-in. Mortars and 105- and 155-mm Projectiles) Evaluation Factor Public and Regulatory Secondary Destruction Process Process Process Process Acceptability in Waste Verification Process EDT Maturity Efficacy Throughput Safety a U.S. Context Issues Capability Flexibility Total TC-60 TDC  8  4 10 7  9 8  9 10 65 DAVINCH DV65  8  9 10 8  7 9 10 10 71 SDC2000  7  9 10 9  7 7  9 10 68 EDSa 10 10 10 7 10 6 10 10 73 NOTE: The above values for each evaluation factor are the average of each committee member’s rating on a scale of 0-10. These average values were then summed to arrive at the totals given in the last column. Small differences in the summed ratings, up to about five points, were not considered to be significant by the committee. There was no weighting. aThese ratings are based on the use of two EDS-2 units. SDC2000 Finding 4-5. The EDS-2 is well suited for Requirement P-1. It has an advantage over the other three systems The Dynasafe SDC2000 could execute Requirement with respect to “maturity.” Its hold-test-release feature P-1 in the required time. It has not been permitted in the is an advantage. The DAVINCH DV65 is a close second United States. The pollution abatement system for an choice. The Dynasafe SDC2000 and the TC-60 TDC installation in the United States has not been designed, are also acceptable choices. built, or tested, another disadvantage. Recommendation 4-5. For Requirement P-1, the Army EDS-2 should use one or more EDS-2 units or the DAVINCH DV65 technology. The Dynasafe SDC2000 and the The EDS-2 is well suited for Requirement P-1. The TC-60 TDC are also acceptable choices. committee notes that three EDS-2s will soon be avail- able for Requirement P-1. Two are completing their assignment at Pine Bluff Arsenal and a third is under REFERENCES construction. The EDS has an advantage over the other BPBGT (Bechtel Parsons Blue Grass Team). 2004. Blue Grass Chemical three systems with respect to maturity, and its hold-test- Agent-Destruction Pilot Plant Resource Conservation and Recovery Act release feature is a further advantage. Research, Development and Demonstration Permit Application, Volume I, Revision 2. Richmond, Ky.: Bechtel Parsons Blue Grass Team. CH2M HILL (CH2M HILL Demilitarization, Inc.). 2007. Explosives Safety Submission, Munitions and Explosives of Concern, Commercially Summary Finding and Recommendation for D ­ eveloped Transportable Controlled Detonation Chambers–Models Requirement P-1 T-25, T-30, and T-60. Chantilly, Va.: CH2M HILL Demilitarization, Inc. The EDS-2 has the highest summed rating, 73 out DiBerardo, R., T.A. Blades, and N. McFarlane. 2007. Demonstration/­ of a possible 80. The DAVINCH DV65 is second with Validation of the TC-60 Controlled Detonation Chamber at Porton Down, U.K., Final Demonstration Test Report, ECBC-SP-021, June. a rating of 71. The Dynasafe SDC2000 has a rating of Aberdeen Proving Ground, Md.: Edgewood Chemical and Biological 68 and the TC-60 TDC is rated at 65. Center. UXB (UXB International, Inc.). 2007. Static Detonation Chamber Testing Using a Dynasafe SDC2000. Richmond, Ky.: Blue Grass Chemical Agent Destruction Pilot Plant.