1
Introduction
BACKGROUND
Chemical Agent Demilitarization in the United States
For the past two decades, the U.S. Army has been engaged in activities aimed at destroying its aging stockpile of chemical agents and munitions, which are located at eight sites in the continental United States.1 Approximately 29 percent of the original stockpile of more than 30,000 tons of nerve and blister (mustard) agents has been destroyed to date.2
As a signatory to the international treaty known as the Chemical Weapons Convention, which was ratified by the international community on April 29, 1997, the United States had 10 years to destroy its stockpile, with an allowable extension of 5 years. It recently has been acknowledged that the United States will need the additional 5 years to complete destruction operations, that is, until April 29, 2012.
At four of the disposal facility locations, the destruction process is based on incineration, which is how most of the stockpile has been destroyed to date. Two of the other sites not using incineration (that is, other than the sites at Pueblo, Colorado, and Blue Grass, Kentucky) store chemical agents only in bulk ton containers: mustard agent at the Aberdeen, Maryland, site and VX nerve agent at the Newport, Indiana, site. These stocks are to be destroyed by a neutralization process (hydrolysis), followed by secondary treatment of the products of the hydrolysis treatment. Destruction operations are under way at Aberdeen and pending at Newport. At Aberdeen, secondary treatment of hydrolysate is being performed at the DuPont industrial wastewater treatment facility in Deepwater, New Jersey.
Assembled Chemical Weapons Alternatives Program
In 1996, in response to local opposition to the use of incineration for destroying the stockpile of chemical agents and munitions, the U.S. Congress passed Public Laws 104-201 and 104-208 that (1) froze the funds for construction of destruction facilities at Pueblo Chemical Depot in Colorado and at the Blue Grass Army Depot in Kentucky, (2) required the Army to demonstrate at least two alternatives to incineration to destroy assembled chemical weapons, and (3) required the Army to coordinate with the National Research Council. This program became known as the Assembled Chemical Weapons Assessment program and has since been renamed the Assembled Chemical Weapons Alternatives (ACWA) program.
After an elaborate selection process in which the public was extensively involved, six technologies received acceptable technology grades, and the Army chose three of these for demonstration (Demo I) of their
technical viability to meet destruction objectives. Two of these technologies were found acceptable after demonstration testing, and they proceeded to engineering design studies (EDSs) for assessing their acceptability for implementation to destroy the chemical stockpile at Pueblo, which comprises nearly 800,000 projectiles and mortar rounds filled with mustard agent.
Subsequently, in 1999, the Congress passed Public Laws 106-79 and 106-52, requiring the Army to demonstrate the remaining three technologies (Demo II) and to consider all viable technology alternatives for destroying the chemical weapons at Blue Grass Army Depot in Kentucky, where both mustard agent and nerve agent munitions are stored. Two of the technologies demonstrated in Demo II and one of those in Demo I were selected to undergo EDS as candidates for destroying the weapons at Blue Grass.
A final, site-specific Environmental Impact Statement to satisfy National Environmental Policy Act (NEPA) requirements was completed in April 2002. The Department of Defense’s (DOD’s) Defense Acquisition Board issued a Decision Memorandum in July 2002 that approved neutralization followed by biotreatment for full-scale pilot testing at Pueblo and directed acceleration of the destruction of the stockpile. The Record of Decision was signed in July 2002. The Request for Proposal (RFP) to design, build, operate, and close a chemical agent destruction facility at Pueblo was issued in July 2002. Although the RFP specified that hydrolysis followed by biotreatment was to be used in the process, the selection of all other unit operations was left to the RFP respondents. The only other requirement was that the process design should include the destruction of all hazardous materials on-site. The system contract was awarded to Bechtel National, Inc., in September 2002, and work to develop a full-scale pilot plant design began in December 2002.3,4
Involvement of National Research Council in Assembled Chemical Weapons Alternatives Program
In accordance with congressional guidance, the National Research Council (NRC) formed the Committee on Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons (referred to as the ACW I Committee) in 1997 to evaluate the candidate technologies and to assess the three technologies chosen for Demo I. After producing two reports evaluating these technologies, the ACW I Committee completed its tenure in February 2000. However, the Army requested the NRC to evaluate the remaining three technologies that were to be tested during Demo II and to assess the engineering design studies of the first two technologies selected from Demo I.
Thus, a second NRC committee, the Committee on Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons: Phase II (ACW II Committee) was formed, with largely the same membership as that of the ACW I Committee. The ACW II Committee produced three full-length reports, one evaluating engineering design studies for technologies considered for implementation at the Pueblo site, a report on the three technologies tested in Demo II, and a report evaluating the second set of engineering design studies for technologies considered for implementation at Blue Grass. Because the committee had to terminate data gathering before all of the tests were completed, two letter reports were also issued to update the committee’s analyses and findings for the full-length reports. The ACW II Committee was dissolved in March 2003.5
In August 2003, the Army requested the NRC to form another committee to assist in evaluating the designs for the pilot plant facilities for the sites at Pueblo, Colorado, and Blue Grass, Kentucky. The present committee—the Committee to Assess Designs for Pueblo and Blue Grass Chemical Agent Destruction Pilot Plants (referred to as the ACWA Design Committee)—was established in October 2003.
STATEMENT OF TASK
The ACWA Design Committee was requested to review and evaluate both the Pueblo Chemical Agent
Destruction Pilot Plant (PCAPP) and the Blue Grass Chemical Agent Destruction Pilot Plant (BGCAPP) design plans. However, this report only covers the committee’s interim evaluation of the PCAPP design. The committee’s statement of task is as follows:
The Program Manager for Assembled Chemical Weapons Alternatives (PMACWA) has awarded contracts for the design, construction, systemization, pilot testing, operation and closure activities aimed at destroying the assembled chemical weapons stockpiles at Pueblo Chemical Depot and Blue Grass Army Depot. Chemical neutralization-based technologies form the basis for destroying the agent and energetics associated with both stockpiles, along with new or adapted processes for preparing weapons for disposal and treating secondary waste streams generated during the primary neutralization processing step. These facilities differ from previously constructed baseline incineration facilities and from those constructed for bulk chemical agent disposal. To assist the PMACWA, the NRC will initially examine planning documentation and designs for the Pueblo and Blue Grass facilities and provide comments and recommendations. Separate reports will address the specific issues for each facility.
The NRC will:
-
Assess planning documentation for design and construction of the Pueblo and Blue Grass facilities.
-
Assess process and facility designs of the Pueblo and Blue Grass Chemical Agent Destruction Pilot Plants.
-
Consider design issues raised by permitting considerations and public acceptability (e.g., design aspects of facility closure).
-
Produce reports within three months following the date the initial and intermediate designs are provided to the NRC for Pueblo and Blue Grass.
SCOPE OF THE REPORT
Although this committee first met in November 2003, new Army security regulations significantly delayed the delivery of detailed information concerning the initial design for the Pueblo facility. Because the design development was continuing at a rapid pace, the Army requested an interim report based on the information that was available to the committee at the time this report was prepared.6 Thus, this report focuses on significant issues that have come to the attention of the committee in order that they may be addressed as soon as possible by the Army and its contractors.
The full initial design plans for the Pueblo facility were only made available to the committee in May 2004. Other sources of information included test plans and various test reports and trade studies that had been completed and made available. The committee also attended the intermediate design review at Bechtel headquarters in San Francisco on May 19–21, 2004. Although the committee members could attend all sessions and were given presentations during the intermediate design reviews, they were not allowed to have either paper or electronic copies of the presentation slides nor of the intermediate design package, since these items had not yet undergone operations security (OPSEC) review. Thus, neither the presentation slides from this review nor the complete intermediate design package was available to the committee when this report was prepared.
This report attempts to provide a general overview of the technical risk assessment program and of certain potential difficulties with the unit processes that are associated with the PCAPP design. The limitations in available information described above necessarily mean that the committee’s work has not yet been completed. The omission of a particular topic should not be taken to suggest either approval or disapproval by the committee. A subsequent report on the PCAPP design is planned when the intermediate and final design plans become available.
ORGANIZATION OF THE REPORT
Chapter 1 provides concise background information on the ACWA program, an introduction to the committee’s task, the rationale for preparing an interim report, and the PCAPP contractor’s design approach. Prior NRC reports on the ACWA program are noted for the reader who may wish to have more detailed information.
Chapter 2 describes the overall process and the unit operations being designed for PCAPP. These descriptions are based on the initial design plans and on the intermediate design presentations given at the contractor’s design review meeting on May 19–21, 2004, in San Francisco. Certain unit operations are also evaluated in Chapter 2.
Chapter 3 discusses in detail areas of concern identified by the committee. It does not include any subsequent design modifications that may have been made since the May 19–21, 2004, presentations by the contractor. Detailed concerns as well as findings and recommendations on specific unit processes and related issues are also included in Chapter 3.
Chapter 4 reviews briefly the public involvement program for PCAPP. General findings and recommen-
dations on the PCAPP design activities and related issues are given in Chapter 5.
DESCRIPTION OF PUEBLO CHEMICAL DEPOT STOCKPILE
The projectiles and mortar rounds stored at Pueblo Chemical Depot contain either HD, which is distilled mustard agent (bis (2-chloroethyl) sulfide), or HT, which contains nominally 60 weight percent HD and 40 weight percent residual that contains T (bis[2-(2-chloroethylthio)ethyl] ether) and related homologues. HT has the advantage of a lower freezing point than that of HD (U.S. Army, 2004a). The inventory of munitions at Pueblo is given in Table 1-1. The physical properties of field-grade liquid HD and liquid HT are given in Table 1-2. However, the composition of the agent in these munitions has changed in the four or five decades since the munitions were manufactured and placed in storage. A typical composition of the HD and the HT found in the old mortars is given in Table 1-3. Diagrams and technical specifications of the munitions to be destroyed are provided in Appendix A.
CONTRACTOR DESIGN STRATEGY FOR PUEBLO CHEMICAL AGENT DESTRUCTION PILOT PLANT
The Army procurement request for the Pueblo Chemical Agent Destruction Pilot Plant called for bidders to propose an integrated approach to the full scope of necessary activities, from design through construction, operations, and eventual decommissioning or closure. In response, the selected contractor has used a design-build strategy that is described in the Design-Build Plan for the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP) project (U.S. Army, 2003a). This approach can save both time and money for plants using well-developed processes. Under this approach, construction is begun before certain unit operations are designed and tested.
The selected contractor team, called the Bechtel Pueblo team, includes Bechtel National, Inc., as the systems contractor, together with several teaming subcontractors: Battelle Memorial Institute, Parsons Infra-structure and Technology, and Washington Demilitarization Company. In turn, the latter two companies have subcontractors—General Atomics and General Physics, respectively—complementing them. Appendix B indicates the responsibilities of each of these participants.
As required under the contract, the activities of the contractor team include all design, procurement, facilities construction, fabrication and testing of process components, installation, systemization, operations, and eventual closure. Included in these activities are safety analysis, licensing and environmental permitting, technical risk assessment, the use of lessons learned from previous chemical weapons disposal pro-
TABLE 1-1 Chemical Weapons Stockpile of HD- or HT-Filled Munitions at Pueblo Chemical Depot
Munition Type |
Model No. |
Chemical Fill |
Energetics |
Configuration |
105-mm cartridge |
M60 |
1.4 kg HD |
Burster: 0.12 kg tetrytol Fuze: M51A5 Propellant: M1 |
Unreconfigured: semifixed, complete projectile: includes fuze, burster. Propellant loaded in cartridge. Cartridges packed two per wooden box. |
105-mm cartridge |
M60 |
1.4 kg HD |
0.12 kg tetrytol |
Reconfigured: includes burster and nose plug, but no propellant fuze. Repacked on pallets. |
155-mm projectile |
M110 |
5.3 kg HD |
0.19 kg tetrytol |
Includes lifting plug and burster but no fuze. On pallets. |
155-mm projectile |
M104 |
5.3 kg HD |
0.19 kg tetrytol |
Includes lifting plug and burster but no fuze. On pallets. |
4.2-inch mortar |
M2A1 |
2.7 kg HD |
0.064 kg tetryl Propellant: M8 |
Includes propellant and ignition cartridge in a box. |
4.2-inch mortar |
M2 |
2.6 kg HT |
0.064 kg tetryl Propellant: M8 |
Includes propellant and ignition cartridge in a box. |
NOTES: The terms “unreconfigured” and “reconfigured” are defined in the column labeled “Configuration.” The M1 propellant present in 105-mm cartridges that have not been reconfigured is present in M67 propelling charges, that is, granular propellant contained in bags as specified in MIL-DTL-60318C. SOURCE: Adapted from U.S. Army, 2004b. |
TABLE 1-2 Physical Properties of Mustard Agents at Pueblo Chemical Depot
Agent Characteristic |
HD |
HTa |
Chemical name |
Bis(2-chloroethyl) sulfide or 2,2-dichlorodiethyl sulfide |
Same as HD, with 20 to 40 wt % agent T, bis[2-(2-chloroethylthio) ethyl] ether |
Chemical formula |
C4H8Cl2S |
Not applicable |
Molecular weight |
159.07 |
188.96 (based on 60/40 wt %) |
Vapor density (relative to air) |
5.5 (calculated) |
6.5 (calculated based on 60/40 wt %) |
Boiling point, °C |
218 (extrapolated) |
No constant boiling point |
Decomposition temperature, °C |
180 |
165 to 180 |
Freezing point, °C |
14.45 |
1.3 (measured as melting point) |
Vapor pressure, torr (mm Hg) at 25°C |
0.106 |
7.7 × 10−2 (calculated based on Raoult’s law equation) |
Volatility, mg/m3 at 25°C |
9.06 × 102 (calculated from vapor pressure) |
7.83 × 102 (calculated from vapor pressure) |
Diffusion coefficient for vapor in air (cm2/sec) |
0.060 at 20°C (68°F) |
0.05 at 25°C (77°F) |
Flash point, °C |
105 |
Flash point range: 109 to 115 |
Surface tension, dynes/cm |
43.2 at 20°C (68°F) |
42.01 at 25°C (77°F) |
Viscosity, cSt |
3.52 at 20°C (68°F) |
6.05 at 20°C (68°F) |
Liquid density, g/cm3 |
1.2685 at 25°C |
1.263 at 20°C |
Solubility, g/100 g of distilled water |
0.092 at 22°C (72°F); soluble in acetone, carbon tetrachloride, chloroform, tetrachloroethane, ethyl benzoate, ether |
Slightly soluble in water; soluble in most organic solvents |
Heat of vaporization, Btu/lb (J/g) |
190 (82) |
Not available |
Heat of combustion, Btu/lb (J/g) |
8,100 (3,482) |
Not available |
aHT is prepared by a chemical process that synthesizes the HT directly in such a way that it contains both the HD and T constituents without further formulation. SOURCES: Abercrombie, 2003; adapted from U.S. Army, 1988. |
grams, and interaction with the local community concerning all of these activities.
There are many challenges involved in controlling cost and schedule for the PCAPP project. These challenges stem from the PCAPP being a first-of-a-kind facility that will require the integration of many sub-system process components that have yet to be designed, built, or tested at a scale consistent with the anticipated throughput of the facility. These technical and scheduling issues appear to be understood by the contractor team, which has developed specific steps, plans, and procedures in its preliminary design documents in order to obtain the necessary data.
Ultimately, the as-built drawings of the facility itself, the installed equipment, the piping, the electrical equipment and wiring, and the control instrumentation are mandatory components for conducting operations and maintenance. To acquire and manage these design components, the PCAPP contractor team has organized the design packages into 11 elements that are mostly centered on the various processes and the buildings that encompass the six major unit operations. These design
TABLE 1-3 Compositions of Liquid HD and Liquid HT Agent Drained from 4.2-inch Mortars at Pueblo (Excluding the Composition of Any Solids in the Munitions)
Chemical Name/Abbreviations |
Chemical Structure |
Area % by GC/MS-Cl |
|
HD |
HT |
||
Bis(2-chloroethyl) sulfide, HD |
(ClCH2CH2)2S |
95.7 |
62.6 |
1,2-Dichloroethane |
ClCH2CH2Cl |
0.7 |
0.3 |
1,2-Bis(2-chloroethylthio) ethane |
ClCH2CH2SCH2CH2SCH2CH2Cl |
2.1 |
3.6 |
1,4-Dithiane |
S(CH2CH2)2S (six-member ring) |
0.7 |
2.0 |
1,4-Oxathiane or 1,4-thioxane |
O(CH2CH2)2S (six-member ring) |
0.04 |
0.7 |
Bis(2-chloroethyl) disulfide, HS2 |
(ClCH2CH2)2S2 |
0.02 |
0.04 |
Bis(3-chloropropyl) sulfide, BCPRS |
(ClCH2CH2CH2)2S |
0.08 |
|
2-Chloropropyl 3-chloropropyl sulfide |
(CH3CHClCH2)S(CH2CH2CH2Cl) |
0.07 |
|
2-Chloroethyl 4-chlorobutyl sulfide, |
CCBS (ClCH2CH2)S(CH2CH2CH2CH2Cl) |
0.3 |
|
2-Chloroethyl 3-chloropropyl sulfide, |
CECPRS (ClCH2CH2)S(CH2CH2CH2Cl) |
0.07 |
|
Bis[2-(2-chloroethylthio) ethyl] ether |
(ClCH2CH2SCH2CH2)2O |
|
23.5 |
2-(2-Chloroethylthio) ethyl 2-chloroethyl ether |
ClCH2CH2SCH2CH2OCH2CH2Cl |
|
5.1 |
Bis(2-chloroethyl) ether |
(ClCH2CH2)2O |
|
0.2 |
Polythioethers (4 compounds tentatively identified) |
|
|
2.0 |
Total |
|
99.8 |
100.0 |
NOTE: HD mortars contained an average of 80 weight percent liquid and 20 weight percent solid. SOURCE: Chemical Compositions of Liquid HT, Solid HT, Liquid H, and Solid H. Briefing by Yu-Chu Yang, ACWA Chief Scientist, to the Mustard Working Group Meeting, September 23, 2003. |
packages are complemented by a full three-dimensional computer model for PCAPP that includes all civil structures, equipment, piping and electrical infrastructure, control instrumentation, and interfaces, and provides considerable detail on the installation process and operational parameters.7
An important element of any design-build activity is configuration control. This effort entails keeping track of changes necessitated as the design and process details change and as equipment is manufactured and installed.
Finding 1-1. The three-dimensional computer model for the Pueblo Chemical Agent Destruction Pilot Plant is a key element in the formal configuration control program that has been developed by the Bechtel Pueblo team. All design and procurement documentation is maintained in its most current form and related via the three-dimensional model.
Recommendation 1-1. It is essential that the Bechtel Pueblo team, the contractor team for the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP), follow through on procedures to ensure that field changes are incorporated back into the three-dimensional computer model for PCAPP so that as-built drawings reflect the actual installation.