Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
Board on Army Science and Technology Mailing Address:
500 Fifth Street, NW
Washington, DC 20001
www.nationalacademies.org
November 5, 2009
Mr. Conrad Whyne
Director
Chemical Materials Agency
5183 Blackhawk Road
Edgewood Area
Aberdeen Proving Ground, MD 21010-5424
RE: Disposal of Legacy Nerve Agent GA and Lewisite Stocks at Deseret Chemical
Depot
Dear Mr. Whyne:
At your request, the National Research Council of the National Academies
established a study committee to assess the disposal of stocks of legacy nerve agent GA
and lewisite at Deseret Chemical Depot. (See Attachment A for the statement of task.)
Specifically, the Committee on Disposal of Legacy Nerve Agent GA and Lewisite Stocks
at Deseret Chemical Depot reviewed information provided to it on the 50 percent design
of the Area Ten Liquid Incinerator (ATLIC) facility.
The findings and recommendations in this letter report are based on the
information that the committee received in July and August 2009 from the Army and its
contractors on the 50 percent design. The information came from presentations at the first
meeting, which took place at Deseret Chemical Depot in Tooele, Utah, on July 21-22,
2009, a teleconference on July 29, 2009, and written design plans and other documents
provided to the committee upon request. The committee was not asked to consider
alternative methods of destroying the GA and lewisite.
The committee focused on differences between the process design being used for
the ATLIC facility and those used at the Tooele Chemical Agent Disposal Facility
(TOCDF), which has operated successfully for over a decade in campaigns to destroy
nerve agents GB and VX and mustard agent. This letter report provides the technical
information necessary to support the general and specific findings and recommendations
of the committee. 1 Also, the committee took into consideration that the ATLIC facility
would operate for only 3 months or so following approximately 4.5 months of
systemization. Closure of the facility was not an issue within the scope of the design
review conducted by the committee, nor were any specific details on this provided.
Certain physical properties of GA and lewisite are provided in Attachment B. Process
1
The key findings (General Findings 1 to 8 and General Recommendations 1 and 2) are presented
in this covering letter. Specific findings and recommendations are found in the remaining portion of this
report, which is the detailed analysis of the 50 percent design for the ATLIC facility.
-1-
OCR for page 2
flow diagrams for GA and lewisite and a schematic diagram for the ATLIC facility
pollution abatement system can be found in Attachment C. Abbreviations and acronyms
are listed in Attachment D.
The committee notes that the GA stored at Deseret Chemical Depot contains no
significant concentration of metal contaminants and is similar to GB. Because the
lewisite contains a large amount (37 wt percent) of arsenic, this letter report is largely
focused on destruction of the lewisite and management of the resulting waste streams.
What follows is the analysis of the committee based on the 50 percent design
information provided for the ATLIC facility. This analysis is provided to satisfy the
requirements in the committee’s statement of task as summarized in the following extract
from the statement:
Examine the process design and procedural steps to be used for treating GA
and lewisite at the Deseret Chemical Depot;
Provide an assessment of the process design which includes a new incinerator
and associated pollution abatement system to be tailored to the requirements
for treating the GA and lewisite;
Provide an assessment of the process design to determine the system's ability
to reduce arsenic and mercury emissions to within the Maximum Achievable
Control Technology (MACT) new source regulatory limits
Produce a report covering the topics listed above.
The committee’s general findings and recommendations are as follows:
General Finding 1. The committee believes that the Area Ten Liquid Incinerator and
associated pollution abatement and process control systems being installed at Deseret
Chemical Depot for processing GA and lewisite will safely and completely incinerate the
GA and lewisite. However, the incinerator and the pollution abatement system constitute
a first-of-a-kind system. Although this incinerator was available, it had never been used
to destroy agent and is being modified with new burners and injectors and new integrated
pollution abatement and integrated process control systems.
General Recommendation 1. The integrated Area Ten Liquid Incinerator and pollution
abatement system should be assembled, tested, and debugged prior to installation in Area
Ten and prior to systemization. The system should be tested with the same auxiliary fuel
that will be used in the Area Ten Liquid Incinerator facility.
General Finding 2. The committee believes the principal challenges to be addressed for
the Area Ten Liquid Incinerator facility arise largely from the arsenic content in the
lewisitethat is, from capture of the arsenic species and management of the resulting
arsenic waste streams. Since GA is chemically similar to GB and contains no
extraordinary amounts of regulated metals, its incineration is expected to be
straightforward and to present no issues that have not been successfully resolved during
GB disposal operations at the Tooele Chemical Agent Disposal Facility.
-2-
OCR for page 3
General Finding 3. Based on its review of the Army’s approach to ensure compliance
with all required environmental regulations, the committee expects the necessary permits
to construct and operate the Area Ten Liquid Incinerator facility for destruction of GA
and lewisite to be forthcoming.
General Finding 4. The method proposed to access agent from the ton containers has
been used extensively to obtain samples of other agents at Deseret Chemical Depot or to
completely drain ton containers at other sites. The committee believes the Army can
successfully remove the liquid GA or lewisite from the containers.
General Finding 5. In the information available to the committee, procedures for
decontamination, rinsing, and sampling of ton containers to meet Chemical Weapons
Convention treaty obligations as administered by the Organization for the Prohibition of
Chemical Weapons and Resource Conservation and Recovery Act regulatory
requirements for shipment off-site, were not provided in detail. However, the Army has
prior experience with similar nonthermal decontamination procedures, such as those that
were used at the Aberdeen, Maryland, site. The committee expects that procedures will
be implemented for the Area Ten Liquid Incinerator facility pursuant to the conditions
established in the state-approved Resource Conservation and Recovery Act permit.
General Finding 6. The committee expects that the pollution abatement system of the
Area Ten Liquid Incinerator facility, as described in the 50 percent design, will
effectively remove arsenic and mercury to below the EPA’s Maximum Achievable
Control Technology standards. This expectation is based on the use of redundant unit
operations for arsenic and mercury removal in the pollution abatement system and in
view of well-established principles for mercury removal. These factors counterbalance
uncertainties that exist concerning the chemical forms of the arsenic and mercury that
will be present, a lack of prior history on the effectiveness of removal for the high
concentrations of arsenic that will be present in the gas stream, and uncertainties arising
from the potential interferences caused by simultaneous arsenic and mercury removal.
General Finding 7. Until systemization and trial burns for the Area Ten Liquid
Incinerator facility are completed, available data do not allow a determination of what
form and in what waste streams the arsenic and mercury from agent destruction will be
found.
General Recommendation 2. In the trial burns of lewisite, the Army must determine the
removal efficiency of the arsenic and mercury and the distribution of these elements in
the waste streams of the pollution abatement system of the Area Ten Liquid Incinerator
facility. These analyses are critical to viable operation of the Area Ten Liquid Incinerator
facility and waste management.
General Finding 8. The Army has been working with the Utah Division of Solid and
Hazardous Waste to (1) identify all secondary waste streams generated at the Area Ten
Liquid Incinerator facility; (2) adopt a comprehensive waste analysis plan that establishes
appropriate sampling and analysis methods and waste control limits for each secondary
-3-
OCR for page 4
waste to be treated in the Area Ten Liquid Incinerator facility or to be shipped off-site for
treatment or disposal; and (3) ensure each secondary waste stream is properly managed in
accordance with the facility-specific permit waste analysis plan and all applicable
Resource Conservation and Recovery Act regulations and requirements.
Additional more specific findings and recommendations are provided in the detailed
analysis of this letter report, which follows.
Sincerely,
Robert A. Beaudet, Chair
Committee on Disposal of Legacy
Nerve Agent GA and Lewisite
Stocks at Deseret Chemical Depot
Attachments
A Statement of Task
B Pertinent Properties of GA and Lewisite
C Process Flow Diagrams for GA and Lewisite Processing; Schematic Diagram of
ATLIC Pollution Abatement System
D Abbreviations and Acronyms
E Committee on Disposal of Legacy Nerve Agent GA and Lewisite Stocks at
Deseret Chemical Depot
F Acknowledgement of Reviewers
-4-
OCR for page 5
Detailed Analysis of 50 Percent Design for the ATLIC Facility
Approximately 44 percent of the more than 31,000 tons of the chemical agents in
the original U.S. stockpile declared under the Chemical Weapons Convention (CWC)
treaty were stored at Deseret Chemical Depot (DCD). Included in this material were
some relatively small quantities of the nerve agent GA (also known as tabun) and the
blister agent and lung irritant lewisite. Destruction of the large quantities of nerve agents
GB and VX stored at DCD was begun in August 1996 at the Tooele Chemical Agent
Disposal Facility (TOCDF) located at DCD and completed in 2006. Some mustard agent
munitions and ton containers containing low levels of mercury have also been destroyed.
The remaining mustard agent containing high levels of mercury will be destroyed
following installation and systemization of the pollution abatement system carbon filter
system at the TOCDF, which were under way when this report was being prepared.
This letter report examines the 50 percent process design provided by the Army
and its contractor for the small destruction facility called the Area Ten Liquid Incinerator
(ATLIC) facility, which will be used to destroy the GA and lewisite stored at DCD (see
also the statement of task in Attachment A). The agents are being stored in bulk in sealed
vessels commonly known as “ton containers” (TCs). The ATLIC facility is being
designed to destroy 4 TCs of GA, 10 TCs of lewisite, and another 10 TCs that may
contain lewisite residues. The ATLIC itself is an incinerator that had originally been
constructed for another purpose but was available (albeit disassembled); its capacity is
approximately one fourth that of the liquid incinerators at the TOCDF. This incinerator is
being used to avoid interfering with the mustard agent destruction ongoing at the TOCDF
or otherwise prolonging the latter’s overall schedule. Table 1 indicates the approximate
composition and total mass of the GA and lewisite materials to be destroyed. One
environmental challenge is the approximately 37 percent arsenic content in the lewisite
that must be captured and sent to a hazardous waste treatment, storage, and disposal
facility. 2
At the time this report was prepared, DCD and the contractor considered the
design of the ATLIC facility to be 50 percent complete. The Army’s Chemical Materials
Agency (CMA) requested the National Research Council (NRC) to assemble a study
committee to assess the design and associated issues relating to the ATLIC facility
processes (see Attachment A).
PERMITTING AND REGULATORY CONTEXT
TOCDF currently operates under the conditions established by a Resource
Conservation and Recovery Act (RCRA) permit issued by the state of Utah, Division of
Solid and Hazardous Waste (UDSHW). Because the ATLIC facility is not included in the
current TOCDF RCRA permit, its construction in the Area Ten storage area of the DCD
will require a RCRA Class 3 permit modification to be filed with and approved by the
UDSHW (Utah Rule R315). RCRA regulations concerning hazardous air emissions
(Utah Rule R315-8-15) do not apply to hazardous waste incinerators that demonstrate
2
Kevin Morrissey, SAIC, “Characterization of lewisite and tabun stored at DCD: Application to
DCD-LITANS,” November 4-5, 2008. Briefing paper provided to the committee on July 24, 2009.
-5-
OCR for page 6
TABLE 1 Composition of the Legacy Chemical Agents Stored at Deseret Chemical
Depot
Total Mass
of Contents
Item Contents and Major Contaminants (wt-%)
(tons)
4 TCs containing the agent ~2 GA (tabun): (32-63%)
GA Chlorobenzene (4-15%)
Diethyldimethlyphosphoramidate (0-11%)
Tetramethylphosphorodiamide cyanide (3-5%)
Diethyl 1,2-tetramethyldiamido-diphosphate (3-9%)
Mercury (nondetect)
Arsenic (36 ppm)
10 TCs containing lewisite ~13 Lewisite 1: ClCH=CHAsCl2 (~73%)
Lewisite 2: [ClCH=CH]2AsCl (~17%)
Other arsenical compounds (3-5%)
Mercury (56-536 ppm)
10 “transparency” TCs with ~0 Now known to be empty except for traces of liquid
lewisite residuesa
a
The term “transparency” is associated with a designation from a CWC treaty perspective for the
10 TCs possibly containing lewisite residues. When declaring the total chemical stockpile, there was a
question of whether these 10 TCs contained agent. In order to be transparent in the declaration, the 10 TCs
were declared as part of the stockpile.
SOURCE: Kevin Morrissey, SAIC, “Characterization of lewisite and tabun stored at DCD: Application to
DCD-LITANS,” Adapted from a November 4-5, 2008, briefing paper provided to the committee on July
24, 2009.
compliance with the Hazardous Waste Combustor Maximum Achievable Control
Technology (MACT) requirements.
The Utah Division of Air Quality (UDAQ) has incorporated by reference the
federal Environmental Protection Agency (EPA) National Emission Standards for
Hazardous Air Pollutants (Utah Rule R307-214; 40 Code of Federal Regulations 63,
Subpart EEE), effective as of July 1, 2007. This rule stipulates emission standards based
on the performance of maximum achievable control technology. Section 112 of the Clean
Air Act required the EPA to establish emissions standards for hazardous air pollutants.
These National Emission Standards for Hazardous Pollutants are commonly referred to as
MACT standards because the EPA used the MACT concept to determine the levels of
emission control. 3 In essence, MACT standards ensure that all major sources of air toxic
(i.e., hazardous air pollutant) emissions achieve the level of control already being
3
The MACT standards reflect the “maximum degree of reduction in emissions of . . . hazardous
air pollutants” that the Administrator determines is achievable, taking into account the cost of achieving
such emission reduction and any non-air-quality health and environmental impacts and energy
requirements [Section 112(d)(2)].
-6-
OCR for page 7
achieved by the better-controlled and lower-emitting sources in each category. The EPA
found that this approach assures citizens that each major source of toxic air pollution will
be required to effectively control its emissions of air toxics. 4 For new hazardous waste
incinerators, the MACT standards limit emissions of chlorinated dioxins and furans,
carbon monoxide and hydrocarbons, toxic metals (including mercury and arsenic),
hydrogen chloride and chlorine gas, and particulate matter.
Under the MACT rule, mercury emissions from new incinerators are currently
limited to 8.1 μg/dscm corrected to 7 percent oxygen (40 CFR 63.1219(b)(2)); this would
be applicable to mercury emissions from the proposed ATLIC. Also under the MACT
rule, arsenic emissions are currently limited to 23 μg/dscm corrected to 7 percent oxygen
(40 CFR 63.1219(b)(4)); this, too, would be applicable to arsenic emissions from the
proposed ATLIC. As part of the MACT requirements, a comprehensive performance test
must be conducted that includes a surrogate trial burn to satisfy both MACT and RCRA
requirements. Therefore, a notice of intent to comply (NIC) will have to be filed with the
UDSHW, including a proposed comprehensive performance test plan.
TOCDF operations are also conducted under the provisions of an existing depot-
wide Clean Air Act Title V operating permit issued by the UDAQ. Although parts of the
ATLIC facility already exist, the facility has not yet been permitted for the destruction of
GA or lewisite. Therefore, the Army filed a notice of intent (NOI) with the UDAQ (Utah
Rule R307-401-4) on September 3, 2009, to modify the existing Title V operating permit
and approval order to include all ATLIC facility emissions and emission sources.
Construction cannot begin until receipt of the approval order for installation of the
ATLIC facility. An approval order will be issued if the UDAQ determines that the degree
of pollution control for emissions, including fugitive emissions and fugitive dust, is at
least the best available control technology and that the facility complies with all
applicable requirements for other air quality conditions, including the National Emission
Standards for Hazardous Air Pollutants. The Army expects to receive temporary
authorization to begin some preparatory construction and installation of equipment prior
to any operations.
In addition to compliance with any Clean Air Act approval order and permit
requirements, the generation, storage, treatment, and disposal of secondary wastes (i.e.,
wastes generated during GA and lewisite treatment) must comply with all applicable
RCRA characterization and management regulations, including compliance with any
waste control limits (WCLs) for GA and lewisite, as established in the RCRA permit
modification.
In addition, the ATLIC facility must also comply with the requirements of the
Organization for the Prohibition of Chemical Weapons (OPCW) for declaring the ton
containers to be empty of agent. 5
Finding 1. As detailed in documentation provided to the committee, the Army appears to
be complying with all required environmental regulations that pertain to the planned Area
4
64 FR 53038, September 30, 1999, as amended at 65 FR 42297, July 10, 2000; 67 FR 6986,
February 14, 2002; 70 FR 59540, October 12, 2005.
5
The OPCW, headquartered in the Hague, Netherlands is the implementing organization for the
CWC treaty, to which the United States is a signatory.
-7-
OCR for page 8
Ten Liquid Incinerator facility and should be expected to receive the necessary permits to
construct and operate the facility.
DESIGN REVIEW OF THE ATLIC FACILITY
Brief Description of the Overall Process
As noted previously, the ATLIC facility will be located in Area Ten of the agent
storage area at DCD, minimizing transport of GA and lewisite TCs from storage to the
processing area. The ATLIC facility uses incineration to destroy the agents and other
liquid secondary waste streams. It employs a two-stage incinerator that was originally
built for use at another location but never used because it was no longer needed. In the
overall process, TCs containing agent are transported from storage igloos to the ATLIC
facility using the same type of equipment and methods that were used to sample mustard
TCs stored in Area Ten for eventual disposal at TOCDF. 6 Figures 1 and 2 in Attachment
C show steps in the processing of TCs containing GA and lewisite, respectively. Figure 3
is a schematic diagram of the pollution abatement system.
TCs arriving at the ATLIC site from storage are loaded onto TC carts and moved
through a vestibule into one of two glove boxes. In the glove boxes, the TC vapor space
and liquid are sampled, and, after their composition has been verified, the liquids are
transferred to the incinerator or an incinerator feed tank in the toxic cubicle. After the
drained TCs are treated with decontamination fluids, the fluids are drained to a collection
tank for spent decontamination solution in the toxic cubicle. After decontamination, the
TCs are rinsed with process water and sampled to verify that agent concentrations meet
OPCW requirements for release from the declared stockpile. The rinse water is sent to the
same collection tank. The liquid contents of the TCs as well as associated
decontamination fluids and rinse water are fed to the ATLIC from the toxic cubicle. The
agent contents drained from the TCs are sent to an injector in the primary chamber of the
liquid incinerator, where the liquid is sprayed into the primary chamber burner flame.
Except for the acetic acid solution, which will be sent to the primary chamber, and the
nitric acid, which will be recovered by diffusion dialysis, spent decontamination liquids
and rinse water will be injected into the burner flame in the secondary chamber.
Exhaust gas from the secondary chamber of the liquid incinerator is fed to a
pollution abatement system (PAS). The PAS uses an aqueous quench to lower the
temperature from 2000°F to approximately 185°F. The cooled gas flows first through a
three-stage packed-bed scrubber and then a venturi to remove gaseous pollutants and
entrained particulates from the cooled exhaust gas of the liquid incinerator. Caustic
solution is used in both the scrubber and venturi. The gas is cooled to remove excess
moisture, then reheated and fed to a baghouse along with powdered sulfur-impregnated
activated carbon (SIC) to remove additional particulates, arsenic, and mercury. After the
baghouse, the filtered exhaust gas flows through two SIC filter assemblies (mounted in
6
Jim Clark, URS, “GA/lewisite–Area Ten Liquid Incinerator (ATLIC) project overview,”
Presentation to the committee on July 21, 2009.
-8-
OCR for page 9
parallel flow) to remove any remaining mercury and arsenic and then through an induced
draft fan, which discharges to a stack. 7
All process areas are ventilated with conditioned air that is then discharged to the
atmosphere through activated carbon filter banks to remove any agent or other toxic
materials from the ventilation air. The carbon filters, which are identical to those used in
ventilation systems at other chemical agent disposal incineration sites, including TOCDF,
have an excellent record of performance.
Process exhaust gas streams and ventilation air streams are monitored for agent at
intermediate points in the flow paths to verify that the various filters are operating
effectively. Monitors will also be used between the beds of HVAC carbon filters to verify
carbon filter effectiveness (URS, 2009).
Accessing GA and Lewisite from Ton Containers
Accessing operations for the TCs are designed to remove all drainable agent or
other liquids in each TC and to decontaminate the TCs. All agent and decontamination
fluids removed from the TCs are pumped to the appropriate ATLIC feed tank or directly
to the incinerator. The standard approach for accessing all of the TCsthat is, the 4 GA
TCs, the 10 lewisite TCs, and the 10 transparency TCs that might contain lewisite
residuesincludes the following key steps:
All accessible liquid is transferred either to the primary chamber feed tank
(lewisite and acetic acid rinse) or directly to the incinerator primary chamber
(drainable GA liquid), followed by addition of the appropriate
decontamination fluid to the TC. Decontamination fluids are agent-specific.
For GA, 100 gallons of 18 percent NaOH solution will be used and will be
drained to the spent decontamination solution tank located in the toxic
cubicle.
For lewisite, 100 gallons of 20 percent acetic acid will be used to dissolve
and remove the remaining lewisite and then pumped to the acetic acid
rinsate tank located in the toxic cubicle. From there, the acid will be
pumped to the primary combustion chamber injector.
After the acetic acid rinse, each lewisite TC will be treated with 100 gallons of
7.0 M nitric acid. The used nitric acid will be drained to the nitric acid rinsate
tank in the toxic cubicle, where it will be analyzed and then sent to a diffusion
dialysis unit for acid recovery. 8
Transparency TCs, if necessary, may be processed as lewisite TCs based on
the results from solid samples taken with a borescope to ascertain the presence
of agent. (This testing had not been completed by DCD and the contractor
when this report was being prepared.)
7
Note that what are termed “carbon filters” in Army vernacular are more accurately described as
carbon adsorption beds.
8
Based on the NOI, either acetic or nitric acid may be used as the decontamination solution (URS,
2009). However, a presentation to the committee by Jim Clark, URS, “GA/lewisite–Area Ten Liquid
Incinerator (ATLIC) facility design review,” on July 21, 2009, indicated that both acetic acid and nitric acid
would be used.
-9-
OCR for page 10
For nitric acid rinses, the used acid solution will be sent to a diffusion dialysis
unit to recover any remaining nitric acid.
After each of the decontamination fluids has been used, the TCs will be rinsed
three times, each time using 100 gallons of water and rotating the TCs for a
prescribed cycle. The rinse water will be sampled to confirm that the agent
concentration is below the WCL established in the state-approved RCRA
permit. Rinsate will be sent to the spent decontamination solution tank in the
toxic cubicle and then injected into the secondary chamber of the ATLIC.
Alternatively, rinsate below the WCL could be disposed of off-site at a
qualified TSDF (TOCDF, 2009).
Finding 2. The direct transfer of ton container liquid contents and decontamination fluid
and rinse water to the Area Ten Liquid Incinerator primary chamber may not ensure that
the flow and composition of the liquid to the primary chamber injector nozzle is uniform
to thereby provide optimum incineration conditions. However, the Army is addressing
this problem by providing a high fuel:feed ratio and by designing the injector to
accommodate a range of fluid properties.
Finding 3. The procedures and chemicals used for decontaminating or removing residual
agent from the ton containers appear to be adequate to ensure that the decontamination
fluids and rinsates do not contain agent above the waste control limits. However, it is
unclear whether occluded agent could be present on the internal surface of the ton
container walls. Also, while no liquid had been found in the transparency ton containers
at the time this report was prepared, a definitive determination of agent contents had not
been completed.
Recommendation 1. The Army should establish a procedure for verifying that after
decontamination the ton containers meet any requirements established in the approved
waste analysis plan necessary to allow the containers to be cut and transported off-site in
accordance with regulatory requirements.
Diffusion dialysis is a commercialized ion exchange membrane technology that is
used in various electroplating processessuch as printed circuit board manufactureto
recover acids that have become contaminated with metals (Steffani, 1995). Diffusion
dialysis separates acid from its metal contaminants by using an acid concentration
gradient between two solution compartments, one filled with contaminated acid and the
other with deionized water, separated by an anion exchange membrane. Acid diffuses
across the membrane into the deionized water. Metal ions are blocked by their charge and
the selectivity of the membrane. Diffusion dialysis does not employ an electrical potential
or pressure gradient across the membrane. Rather, the transport of acid is driven by the
difference in acid concentration in the two compartments separated by the membrane.
Nitric acid recovered from the dialysis unit will be reused after adjusting the
concentration of acid to 7.0 M by adding fresh concentrated nitric acid. Waste solution
from the nitric acid diffusion dialysis unit, which will contain heavy metals, will be sent
to the incinerator secondary chamber or disposed of as hazardous waste.
-10-
OCR for page 11
Finding 4. Diffusion dialysis is a commercial method of acid recovery and appears to be
a valid method of recovering a portion of the nitric acid for reuse. But the application of
this technology to nitric acid that contains both mercury and high concentrations of
arsenic, as well as other contaminants from the lewisite ton containers, is untested.
Alternatively, it might be possible to send the waste nitric acid that will be contaminated
with arsenic, mercury, and other metals off-site for disposal to an appropriate treatment,
storage, and disposal facility under conditions established in the state-approved Resource
Conservation and Recovery Act permit, including any waste control limit for agent.
Recommendation 2. The Army should consider shipping waste nitric acid off-site for
disposal without dialysis, provided it meets the agent waste control limit for off-site
shipment established in the Resource Conservation and Recovery Act permit.
The ATLIC design uses the same common tanks to collect different liquids used
for treating the TCs before feeding them to the primary or secondary incineration
chamber. Thus, the flows of liquids from the TCs to these tanks must be carefully
managed to avoid unexpected reactionsfor example, reactions between acids and
NaOH or between organic materials and 7.0 M nitric acid. These concerns have already
been identified by the contractor and placed in the hazard tracking log, which is discussed
in a later section of this report. 9
Finding 5. Using the same tank to alternatively collect both acidic and basic (low- and
high-pH) liquids or other solutions containing reactive chemicals during operations could
pose a serious hazard.
Recommendation 3. If the Area Ten Liquid Incinerator facility design continues to call
for the use of common tanks to collect different liquids from ton container treatment over
the course of operations, each tank used in this manner should be flushed several times
until the original liquid is no longer present in the tank before it is used to store a new
liquid. Alternatively, different tanks could be installed for each of the different liquids
collected from treating the ton containers.
Incinerator Design
The design of the existing small scale liquid incinerator is similar to that of the
two liquid incinerators (LICs) that have been used at the TOCDF and other facilities to
successfully destroy GB and VX and that are being used to destroy mustard agent
contaminated with mercury at the TOCDF. The ATLIC has approximately one-fourth the
capacity of the LICs at the TOCDF, based on the lewisite flow rate. It was started up
during acceptance testing on fuel oil but never used to destroy agent. Design
modifications that distinguish it from the TOCDF LICs resulted from the need to
transport the small-scale incinerator to perform the task for which it was originally
designed but never used. These modifications include horizontal instead of vertical
primary and secondary combustion chambers and a water quench tower followed by three
short interconnected scrubber towers with a common sump for scrubber liquid. Thus,
9
New hazard tracking log provided to the committee by CMA staff, July 20, 2009.
-11-
OCR for page 12
exhaust gas flows in series through the three scrubber towers while the scrubber liquid
flows through these towers in parallel. This is in contrast to the single, tall scrubber tower
used in the PAS units for the TOCDF LICs. Although this existing incinerator was
available, it was never used to destroy agent and is being modified with new burners, new
injectors, and new, integrated PAS and process control systems. Current plans call for
testing to be performed off-site.
Finding 6. The Area Ten Liquid Incinerator and its pollution abatement system constitute
a first-of-a-kind system that has not been integrated or tested in its final configuration. It
is being modified with new burners and injectors. The incinerator unit has been stored in
the open air at DCD for a number of years.
Recommendation 4. Prior to systemization, the integrated Area Ten Liquid Incinerator
and pollution abatement system should be assembled, tested, and debugged off-site with
the same auxiliary fuel to be used in Area Ten. Its design specifications must be verified.
This preliminary testing prior to systemization will decrease the overall schedule and
minimize unexpected problems.
The committee reviewed key materials of construction at the 50 percent design
stage and deemed them adequate for the planned operating period of about 3 months of
actual agent operation after approximately 4.5 months of systemization. The steel shells
of the primary and secondary chambers are protected by the same refractory brick linings
as used at TOCDF. The quench tower uses AL6XN alloy for the first 4 feet and Type 316
stainless steel for the remainder, protected by six nozzles spraying caustic on the vertical
walls at the gas inlet.
Confidence in the performance of the ATLIC can be derived from the similarity
of its operating conditions to those in the TOCDF LICs, including the following:
Temperatures in the primary and secondary chambers of both the ATLIC and
the TOCDF LICs are maintained at 2700°F and 2000°F, respectively.
The design of both the ATLIC and the TOCDF LICs calls for 30 percent
excess air, corresponding to an oxygen concentration of 5-6 percent at the exit
of the secondary combustion chamber.
Differences between the incinerator components of the ATLIC and those of
TOCDF LICs are these:
The planned feed rate for GA is 200 lb/hr in the ATLIC versus a baseline feed
rate for GB of 1,000 lb/hr in the TOCDF LICs. The primary burner rating for
the ATLIC is 3 MMBtu/hr compared to 14 MMBtu/hr for each of the TOCDF
LICs. The agent in the ATLIC provides a lower fraction of the total heat
release in the ATLIC than does the agent in the TOCDF LICs. This provides
for more robust operation but results in more natural gas or fuel oil
consumption when the feed of agent or spent decontamination solution is cut
off but the incinerator temperature must be maintained.
-12-
OCR for page 13
There are minor differences in configuration. The primary and secondary
combustion chambers in the ATLIC are horizontal, while those in the TOCDF
LICs are vertical. The ATLIC primary combustion chamber is tied directly
into the secondary chamber, which is also horizontal. This contrasts with the
TOCDF LICs, where the vertical primary combustion chamber is connected to
the vertical secondary chamber through a crossover duct. Agent (or spent
decontamination solution) is injected through a separate gun angled at the
combustion zone created by the incinerator fuel in the ATLIC rather than
directly through the fuel burners as in the TOCDF LICs. The horizontal
secondary combustion chamber may experience some drop-out and
accumulation of noncombustibles on the bottom at the end of the chamber
under the 22-inch induced-draft exhaust opening. Owing to the short operating
period, this accumulation is not expected to cause operating problems;
however, noncombustible material removal and disposal should be provided
for.
The sum of the residence times in the primary and secondary combustion
chambers of the ATLIC is reported to be 5 seconds compared with 2 seconds
in the TOCDF LICs. 10 The residence time for the TOCDF LICs does not,
however, include the time in the cross-over duct, which is maintained at
1800°F.
Finding 7. Liquid incinerators used at chemical demilitarization facilities have provisions
for removal of noncombustible material (slag). Noncombustible materials (e.g., slag) can
be expected to accumulate at the bottom of the horizontal secondary chamber of the Area
Ten Liquid Incinerator.
Recommendation 5. Although the operating time for the Area Ten Liquid Incinerator is
of relatively short duration, the Army should consider the need for removal of
noncombustible materials from the secondary chamber.
The increase in residence time for the ATLIC compared with that for the TOCDF
LICs should lead to an even higher destruction efficiency for the agents to be processed
in the former. The lower flow rates and smaller dimensions in the ATLIC will lead to
lower Reynolds numbers and lower turbulence levels than in the TOCDF LICs, which
might lead to reduced micromixing in the ATLIC. The change in the injection of agent
will lead to differences in macromixing of the agent as well as the combustion patterns in
the ATLIC relative to those in the TOCDF LICs. These changes will impact the details of
the combustion patterns, but, given the reported increased total residence time, the Army
expects that the destruction efficiencies will meet the RCRA requirements. 11 Trial burns
with agent and spent decontamination solution liquid surrogates will be conducted to
10
Jim Clark, URS, “GA/lewisite–Area Ten Liquid Incinerator (ATLIC) project overview,”
Presentation to the committee on July 21, 2009.
11
A measure of the margin of safety in the ATLIC design is the requirement provided by the EPA
Office of Enforcement and Compliance Assurance (OECA) for the destruction of PCBs, compounds
considerably more refractory than GA and lewisite. One set of conditions specified by EPA is a residence
time of 2 seconds, a temperature of 1200°C (2192°F), and an excess oxygen concentration of 3 percent,
conditions exceeded by a safe margin in the ATLIC (OECA, 2004).
-13-
OCR for page 14
ensure that these changes do not impair the performance of the ATLIC. The ATLIC, with
its longer residence time, will meet the destruction and removal efficiency (DRE)
standards for the organic compounds in the GA and lewisite containers.
Finding 8. The committee believes that the temperatures and times in the primary and
secondary combustion chambers of the Area Ten Liquid Incinerator will ensure
destruction efficiencies that meet regulatory requirements for the organic content of the
GA and lewisite ton containers.
The process for permitting a new incinerator under RCRA and MACT regulations
requires trial burns to determine the range of operating conditions that will achieve the
desired DRE. In place of the agent, the trial burn may involve injection of a surrogate
compound that is as difficult or more difficult to destroy than the agent (NRC, 2007). The
established measure of the relative destruction efficiency of hazardous chemicals is the
incinerability index, which ranks compounds in order of their ease of destruction
(Thurnau, 1989). 12 The ranked compounds are divided into seven classes, with Class 1
representing the most refractory compounds. Chlorobenzene, present at up to 15.3
percent in GA, is in Class 1. DCD estimates that GA and lewisite are in Class 5. GB also
was ranked in Class 5 by the University of Dayton Research Institute (UDRI), the
developers of the incinerability index (Taylor and Dellinger, 1990).
The substances selected for the ATLIC trial burns are (1) a mixture of 20 percent
chlorobenzene and 80 percent of a Class 5 (or lower) organic with 100 ppm lead and
sufficient arsenic to determine removal efficiency and (2) lewisite.13 The committee
believes these are judicious choices. Chlorobenzene is one of the more refractory
compounds in Class 1 and is also found at up to 15.3 percent in the four GA TCs (Taylor
and Dellinger, 1990). Inasmuch as the lewisite at DCD contains 37 percent arsenic, there
are few arsenicals with such a high arsenic content that could be used as a surrogate, so
that the use of lewisite for the second trial burn is a logical choice. The lewisite TCs also
contain 56 to 536 ppm mercury, with a mean of 154 ppm, and an average of 34.6 ppm
lead, so that no metal spiking is required.
The potential success of the ATLIC trial burns can be anticipated from the trial
burns conducted on TOCDF LIC2 (NRC, 2007). These were successfully conducted with
surrogates containing trichlorobenzene (Class 1) 14 and the agents GB (Class 5), VX
(Class 5), and mustard agent H and HD (Class 4).
To facilitate the interpretation of trial burn results, it would be helpful to conduct
equilibrium calculations for the agents to be destroyed under the expected incineration
conditions. For the design of the PAS, it would have been prudent to know the amounts
12
The regulations are built around DRE = 100 × (Input mass – Output mass in exhaust gas)/Input
mass. This allows for a combination of destruction such as by incineration (a measure of which is given by
the incinerability index) and removal. For the primary and secondary combustion chambers, the only term
of importance is the destruction efficiency. After the PAS, removal is included, so DRE is the proper
measure. Thus, destruction efficiency is discussed for the primary and secondary combustors and DRE for
the trial burns, which includes both the combustors and the PAS.
13
Drew Papadakis, URS, “ATLIC environmental permitting,” Presentation to the committee, July
22, 2009.
14
Chlorobenzene is ranked 19th and the 1,2,4- and 1,3,5- isomers of trichlorobenzene are ranked
22nd and 23rd of the 320 compounds in the incinerability index (Thurnau, 1989).
-14-
OCR for page 15
and form of the elemental and inorganic constituents in the combustion products that
enter the PAS. In the absence of high chlorine concentrations, experiments show that
As2O3 is the principal product of arsenic combustion (Hirsch et al., 2000; Wasson et al.,
2005; and Hara and Maeda, 2007). However, in the presence of chlorine, AsCl3 may be
formed. Equilibrium calculations by Wu and Biswas (1993) indicate that in methane-air
flames with a large excess of air, the hydrogen competes with the arsenic for the chlorine.
For the proposed ATLIC incinerator operating conditions (100 percent excess air, 7 × 10-
7
arsenic/methane molar ratio), AsCl3 began to form in appreciable amounts only for
Cl/As ratios exceeding 10 at 1521°F and 100 at 2240°F. These ratios are much higher
than the ratios anticipated for the incineration of lewisite.
The Wu and Biswas (1993) calculations indicate that at a Cl/As ratio of 3, as
occurs in lewisite, the arsenic will be emitted from the secondary combustion chamber
primarily as As2O3. The conditions in the ATLIC, however, differ from those used by Wu
and Biswas in one important respect: The ATLIC has an auxiliary fuel with a lower
hydrogen content, a factor that may favor some AsCl3 formation, meaning that some
AsCl3 formation in the ATLIC secondary combustion chamber cannot be ruled out.
Mercury leaving the primary combustion chamber at 2700°F will be present
primarily in the elemental form. However, as the temperature drops to 2000°F in the
secondary combustion chamber, the high Cl/Hg ratio in the gas stream from the
combustion chambers during lewisite incineration means that the mercury will be present
primarily as HgCl2, according to equilibrium calculations and experiments (Fransden et
al., 1994; Wu and Biswas, 1993; and Widmer et al., 1998).
Finding 9. Based on thermodynamic modeling calculations, the arsenic leaving the
secondary combustion chamber for the case of lewisite incineration should be primarily
in the form of As2O3.
Finding 10. Extrapolation of experimental measurements and thermodynamic analysis
suggest that HgCl2 is likely to be the dominant form of mercury at the exit of the
secondary combustion chamber of the Area Ten Liquid Incinerator.
Pollution Abatement System
Based on the chemical composition of GA and lewisite, the anticipated products
of combustion that will require treatment by postcombustion pollution abatement
processes will be compounds containing arsenic (As), mercury (Hg), or phosphorus (P),
and hydrogen chloride (HCl). Although oxides of nitrogen (NOx) will also be present, as
a pollutant, NOx from GA and lewisite destruction will fall under the existing, sitewide
RCRA permit. The PAS is intended to remove these combustion products from the gas
stream. The exact chemical forms of the arsenic and mercury compounds, their relative
proportions, and the transformations they may undergo throughout the PAS have not
been well defined for this first-of-a-kind system.
To control emissions of arsenic, mercury, phosphorus, and hydrogen chloride, the
PAS consists of the following components and processes (in order of gas flow):
-15-
OCR for page 16
Rapid aqueous gas quench, with the gas temperature decreasing from ~2000F
to 185F,
Packed tower wet scrubber,
Venturi scrubber,
Entrainment separator (mist eliminator),
Gas reheater, increasing the gas temperature from ~ 70F to 180F,
Injection of powdered sulfur-impregnated activated carbon (SIC),
Pulsed jet fabric filter (baghouse),
Parallel filter assemblies each consisting of a prefilter, a HEPA filter, two
granular (pelletized) SIC filter beds, and a second HEPA filter, and
A variable-speed induced draft fan and stack.
The removal of compounds of arsenic and mercury can occur in one or more of
the PAS components. The use of the PAS to control emissions of these compounds and
also hydrogen chloride and phosphorus is discussed separately below.
Arsenic
The total residence times and temperatures that are maintained in the ATLIC are
expected to yield combustion products whose composition reflects thermodynamic
equilibrium. Although DCD has not conducted equilibrium calculations, those reported in
the literature provide some guidance. Extrapolation of the equilibrium calculations by
Fransden et al. (1994) determined that arsenic is present as AsO under standard oxidizing
conditions at 2000°F. At lower temperatures, varying amounts of As2O3 and As2O5 will
exist in equilibrium with the AsO. As discussed earlier (see “Incinerator Design”), similar
equilibrium calculations for arsenic in the presence of chlorine indicate that the 3:1
chlorine:arsenic ratio of lewisite will cause the arsenic leaving the secondary chamber of
the ATLIC to be primarily in the form of AsO (Wu and Biswas, 1993). Even if the
chlorine:arsenic ratio were high enough to produce significant concentrations of AsCl3
leaving the secondary chamber of the ATLIC, laboratory experiments suggest that
subsequent hydrolysis in the PAS would convert AsCl3 to As2O3 (Hara and Maeda,
2007). While As2O3 is not water soluble, laboratory experiments conducted on cacodylic
acid suggest that as much as 60 percent of the As2O3 particulates could be captured in the
wet scrubber (Hara and Maeda, 2007).
Finding 11. After quenching, the arsenic removed by the Area Ten Liquid Incinerator
pollution abatement system is expected to be predominantly in the form of arsenic oxides.
Any remaining AsCl3 will be removed in the quench towers because it is water soluble.
DCD expects to capture arsenic through multiple pathways in the current PAS
design: nucleation of As2O3 through (1) a rapid quench, (2) wet scrubbing of As2O3 in the
packed tower wet scrubber, (3) induced growth of the particulate As2O3 in the wet
scrubber, (4) subsequent removal of the larger size fraction of particles in the venturi, and
(5) filtration of the smaller size fraction of the particles and adsorption of the remaining
solid As2O3 by the fabric in the baghouse. The DCD estimates a 95 percent removal
efficiency of arsenic across the venturi scrubber. Although DCD expects some arsenic
-16-
OCR for page 17
removal to also occur in the packed tower wet scrubber and in the baghouse, no removal
efficiency models or measurements were provided at the 50 percent design stage. There is
little experience with the formation of As2O3 aerosols from the combustion of compounds
with arsenic concentrations as high as those in lewisite. As the As2O3 nucleates and
agglomerates, it will form an aerosol with a mean particle size that increases with the
concentration of As2O3 in the gas phase and with the residence time of the combustion
products from the point of condensation of the aerosols. The most relevant study is that of
Hara and Maeda (2007), who observed As2O3 particles of about 0.5 microns produced
from the combustion in an electrically heated laboratory furnace of a solution of 0.1 to
0.2 weight percent cacodylic acid [(CH3)2AsO2H], 54 percent arsenic by weight.
The committee believes that the multiple redundancies built into the ATLIC PAS
to capture arsenic (as well as mercury) will be sufficient as designed to reduce
concentrations of arsenic in the combustion gases to below MACT emission limits. At the
same time, however, a number of uncertainties exist that are associated with specific PAS
components and their role in removing arsenic. At the 50 percent design stage, some of
these components had been specified in relatively sparse technical detail, limiting the
degree to which their operating conditions and performance could be independently
assessed by the committee.
Specifically, the committee has not been able to satisfactorily evaluate the
proposed operation of the following PAS components:
Venturi scrubber, specifically the pressure drop of the gas.
Entrainment separator, specifically the suitability of the component design.
and materials of construction for application to a potentially high loading of
particulate arsenic.
Reheater, specifically its effectiveness in drying fine, moisture-laden arsenic
particles prior to the baghouse.
Injection of powdered SIC, specifically the potential for interferences to occur
during the simultaneous removal of arsenic compounds and mercury or
mercury compounds onto the powdered SIC.
The committee believes that an important implication of these uncertainties is that
DCD cannot yet state with reasonable certainty what the removal efficiency of arsenic
will be in each PAS component. Nonetheless, the committee believes the bulk of the
arsenic will be collected in the common sump for the quench tower, packed tower
scrubber, and venturi scrubber.
Finding 12. At the 50 percent stage, the design status of the various components of the
Area Ten Liquid Incinerator pollution abatement system ranges from design concepts to
actual hardware that is ready for installation. While the design as a whole is 50 percent
complete, the committee did not have access to detailed specifications for several of the
more conceptual components and thus could not conduct an in-depth assessment of their
operation.
-17-
OCR for page 18
Mercury
The mercury entering the PAS is primarily in the form of HgCl2 and elemental
mercury. DCD plans to capture mercury at three different locations in the PAS: HgCl2
will be absorbed in the wet scrubbers, and elemental mercury and any remaining HgCl2
will be adsorbed onto the powdered SIC injected into the gas stream upstream of the
baghouse, or onto the granular SIC held in the filter beds. DCD has estimated mercury
removal efficiencies for these steps to be 50 percent for the wet scrubber and 95 percent
each for the injected SIC in the baghouse and the granular SIC in the filter beds.
DCD has stated that the amount of granular SIC in the final filter beds of the PAS
has been estimated based on the total mercury load expected to be processed during the
campaign, the reported mercury adsorption capacity of the SIC within a flow of simulated
sulfur mustard combustion products, and a safety factor. 15 DCD asserts that the granular
SIC filter bed will be capable of achieving the required overall mercury removal
efficiency for the PAS even without contributions from the other PAS components. The
committee believes that the sizing of the granular SIC filter beds will, under normal
operating conditions, be sufficient to reduce concentrations of mercury in the combustion
gases to below permitted levels.
While absorbing HgCl2 in the packed tower wet scrubber and adsorbing elemental
mercury on the SIC filter beds are both proven ways to control mercury emissions, the
committee has concerns about the simultaneous capture of mercury and arsenic
compounds on the powdered SIC in the baghouse. As noted above, excess moisture could
be introduced into the dust cake on the fabric filters if fine, moisture-laden arsenic
particles are not sufficiently dried during the gas reheating process. Such moisture could
make it difficult to dislodge the dust cake during periodic bag cleaning. As noted earlier,
for the high concentrations of As2O3 expected, only the larger particle size fraction will
be removed by the venturi scrubber. The committee believes that simultaneous collection
of both arsenic and mercury species may be difficult to achieve without interferences.
The baghouse is an appropriate choice to use as a polishing particulate control device to
preserve the operational life of the downstream HEPA filter; however, it is not clear that
removal of arsenic and mercury compounds can be achieved independently and in
parallel on the powdered SIC collected in the baghouse. Without the powdered SIC
injection, the baghouse could remove the remaining As2O3 particulates while letting
elemental mercury pass thorough to be adsorbed by the final granular activated carbon
filters.
Finding 13. The committee believes the Area Ten Liquid Incinerator pollution abatement
system, as designed, will meet EPA’s Maximum Achievable Control Technology
emissions limits for arsenic and mercury because it includes redundant operations for
their removal.
Recommendation 6. As part of the systemization of the Area Ten Liquid Incinerator
facility, the Army should determine what fraction of mercury is captured by the wet
15
Personal communication between Drew Papadakis, URS, and Herek Clack, committee member,
July 22, 2009.
-18-
OCR for page 19
scrubbers and, based on the results, should make appropriate adjustments for how the
baghouse is to be used.
Hydrogen Chloride and Phosphorus
The ATLIC facility design incorporates a packed tower wet scrubber into the PAS
to remove HCl and H3PO4 formed from the phosphorus and chlorine in the GA and
lewisite agents, respectively. The scrubber liquor will be maintained at a pH of 7. Based
on the 50 percent design status, no issues have been identified surrounding the removal of
the phosphorus and chlorine compounds.
MANAGEMENT OF SECONDARY WASTE STREAMS
Each secondary waste stream must be characterized and managed in accordance
with its hazardous characteristics and applicable regulatory requirements. This practice is
similar to that for secondary waste streams at the TOCDF (URS, 2009). The proper
management, characterization, and ultimate disposal options for all secondary waste
streams generated during the ATLIC operations will be established in the RCRA permit
waste analysis plan (WAP), which must be approved prior to start-up of the ATLIC
facility. The WAP was not available to the committee, but existing documentation
indicates some secondary wastes (e.g., rinsates) will be treated by injection into the
ATLIC, some (e.g., scrubber blowdown brine and decontaminated/rinsed TCs) will be
shipped off-site to a properly permitted treatment, storage, and disposal facility, and some
(e.g., lightly agent-contaminated personal protective equipment and solid wastes) may be
treated in the Area Ten Autoclave System or the existing metal parts furnace in the
TOCDF (TOCDF, 2009; URS, 2009). Since neither the WAP nor the trial burn data were
available for review, the committee could not comment on the feasibility or efficacy of
the management of all secondary waste streams. However, certain secondary wastes of
concern unique to the ATLIC system are discussed below.
Secondary wastes from the ATLIC process comprise arsenic- and mercury-
containing compounds captured in the various components of the PAS. It is noteworthy
that the 37 percent arsenic content of the lewisite will lead to approximately 9,600 lb of
arsenic waste (calculated as the element) from all 10 of the lewisite TCs. The unique
waste streams of the ATLIC PAS containing arsenic are the following:
Spent liquor from the common sump of the quench tower, packed-bed
scrubbers, and the venturi scrubber;
Particulate matter from the hopper that receives material dislodged from the
baghouse filters;
Spent SIC filters and HEPA filters used in the final stage of the PAS;
Waste from the nitric acid diffusion dialysis process; and
Rinsate from the TCs and spent decontamination solution.
There may also be some slag from the secondary chamber of the ATLIC. Each of the
secondary wastes could be contaminated with mercury and/or arsenic. The chemical
compositions must be determined before the wastes can be shipped off-site for further
-19-
OCR for page 20
treatment and disposal at commercial facilities. Rinsate, which would be considered a
secondary waste, will be fed to the secondary combustion chamber of the ATLIC as part
of the water needed for scrubber operation or disposed of off-site.
The common sump will contain chloride and phosphate salts, dissolved HgCl2,
suspended As2O3 particulates, and dissolved arsenite salts. According to the 50 percent
design, a small fraction of the scrubber liquor will be continuously drained as a waste
stream (i.e., blowdown). This stream will generate approximately 355,000 gallons of
brine, which will be characterized and shipped off-site as a hazardous liquid waste.
Powdered SIC containing mercury and arsenic compounds is removed from the
baghouse hopper periodically and falls into a containment bin. The bin is removed
periodically and the powdered SIC is characterized and sent off-site. The final SIC beds
in the PAS will capture arsenic particulates, mercury, and trace organic vapors not
captured earlier. The spent filters will be characterized and sent off-site for disposal.
Other secondary wastes common to chemical agent disposal facility operations
generally include these:
HVAC carbon filters, HEPA filters, and prefilters;
Personnel protective equipment, including gloves, suits, aprons, booties, snd
the like; and
Decontaminated TCs.
After being drained, TCs will be decontaminated with the appropriate solution,
rinsed, and returned to processing for cutting and inspection. The decontaminated TCs
are then to be shipped off-site for disposal in a landfill (URS, 2009).
Finding 14. Available information on the management of secondary waste streams from
the Area Ten Liquid Incinerator facility seems to indicate a reasonable approach is being
taken, but a more complete review must await the development of further information
such as the waste analysis plan required by RCRA regulations.
RISK ASSESSMENTS
The ATLIC facility project team has established a system safety engineering
management plan (SSEMP) in accordance with the requirements of Mil-STD-882D and a
hazard analysis and tracking process that complies with the TOCDF PRP-SA-057 hazard
evaluation procedure. In the implementation of SSEMP, a hazard analysis that considers
design, operation, and maintenance procedures is performed at various stages of design
(conceptual through 100 percent). Hazards are identified and given a risk ranking using
the Mil-STD- 882D. The risk ranking is based on the severity of possible consequences
and the likelihood of occurrence for each hazard or identified potential failure. Using the
risk rank code, hazards are tracked in a log and high-risk hazards are assigned to design
and operations personnel for action to resolve high-risk hazards by eliminating them or
reducing the likelihood or consequence of their occurrence.
The hazard tracking log is maintained throughout the development of the design,
and all high-risk hazards must be satisfactorily resolved before design completion and
start-up. Highlights from the hazard tracking log at 50 percent design were presented at
-20-
OCR for page 21
the committee meeting in Tooele, Utah, on July 21-23, 2009. The log identified areas of
concern, including both “open” and “closed” (resolved) findings. The open findings are
being addressed as the design progresses.
After operations begin, a job hazard analysis procedure based on experience from
TOCDF will be used. This procedure will be used for assessing all changes in design and
operation so that a high level of safety is maintained throughout the life cycle for the GA
and lewisite disposal project. No changes can be implemented without satisfactory
resolution of the findings of the job hazard analysis.
All hazard analyses during design and, subsequently, operation are led by project
safety engineers independent of the design and operations teams. In addition, the
implementation of the hazard analysis and risk assessment is periodically reviewed by a
team from the Centers for Disease Control under an agreement established with the
Army.
The risk assessment and safety analysis approach being applied to the ATLIC
facility project is identical in format to the approach applied by the Army’s CMA for all
facilities in the Chemical Stockpile Disposal Program (CSDP). That such an approach
provides adequate assurance and transparency in the management of risk and safety is
demonstrated by the outstanding safety record of operating CSDP facilities.
Finding 15. A review of the hazard tracking log provided to the committee for the 50
percent design of the Area Ten Liquid Incinerator facility indicates that the risk
assessment and safety analysis techniques are being properly applied. Back-up assurance
of the continuation of this practice is provided by the planned oversight by a Centers for
Disease Control team throughout the design and testing of the Area Ten Liquid
Incinerator facility.
REFERENCES
Fransden, F., K. Dam-Johansen, and P. Rasmussen. 1994. Progress in Energy and
Combustion Science 20(2): 115-138.
Friedlander, S. 2000. Smoke, Dust and Haze: Fundamentals of Aerosol Dynamics. Cary,
N.C.: Oxford University Press.
Hara, H. and A. Maeda. 2007. “Arsenic behavior in pollution abatement system:
laboratory scale test.” Presented to the 10th International Chemical Weapons
Demilitarisation Conference in Brussels, Belgium, May. Available online at
http://www.dstl.gov.uk/conferences/cwd/2007/pres/hidemitsu-hara-pres.pdf. Last
accessed on August 26, 2009.
Hirsch, M., R. Stirling, F. Huggins, and J. Helble. 2000. “Speciation of combustion-
derived particulate phase arsenic.” Environmental Engineering Science 17(6): 315-
327.
NRC (National Research Council). 2007. Review of Chemical Agent Secondary Waste
Disposal and Regulatory Requirement. Washington, D.C.: The National Academies
Press.
-21-
OCR for page 22
OECA (Office of Enforcement and Compliance Assurance). 2004. Polychlorinated
Biphenyl Inspection Manual. Washington, D.C.: Environmental Protection Agency.
Steffani, C. 1995. Waste Acid Recycling via Diffusion Dialysis. Livermore, Calif.:
Lawrence Livermore National Laboratory.
Taylor, P., B. Dellinger, and C. Lee. 1990. “Development of a thermal stability-based
index of hazardous organic compound incinerability.” Environmental Science &
Technology 24(3): 316-328.
Thurnau, R. 1989. Incinerability Index: A Measure of Incinerator Performance.
EPA/600/D-89/081. Washington, D.C.: Environmental Protection Agency.
TOCDF (Tooele Chemical Agent Disposal Facility). 2009. FONSI (Finding of No
Significant Impact) Proposed Installation and Operation of a Small-Scale
Incineration Unit for the Destruction of Chemical Agents Tabun and Lewisite at the
Deseret Chemical Depot in Utah. Edgewood, Md.: Chemical Materials Agency.
URS. 2009. Notice of Intent (NOI) to Construct and Install the Area Ten Liquid
Incinerator (ATLIC), DAQE-AN0113390047-08. Stockton, Utah: EG&G Defense
Materials, Inc.
Wasson, S., W. Linak, B. Gullett, C. King, F. Huggins, Y. Chen, N. Shah, and G.
Huffman. 2005. “Emissions of chromium, copper, arsenic, and PCDDs/Fs from open
burning of CCA-treated wood.” Environmental Science & Technology 39(22): 8865-
8876.
Widmer, N., J. Cole, W. Seeker, and J. Gaspar. 1998. “Practical limitation of mercury
speciation in simulated municipal waste incinerator flue gas.” Combustion Science
and Technology 134(1): 315-326.
Wu, C., and P. Biswas. 1993. “An equilibrium analysis to determine the speciation of
metals in an incinerator.” Combustion and Flame 93(1-2): 31-40.
-22-
