National Academies Press: OpenBook
« Previous: 7 Lockheed Martin Integrated Demilitarization System
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 119
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 120
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 121
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 122
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 123
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 124
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 125
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 126
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 127
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 128
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 129
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 130
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 131
Suggested Citation:"8 Parsons-Allied Signal Technology Package." National Research Council. 1999. Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons. Washington, DC: The National Academies Press. doi: 10.17226/9660.
×
Page 132

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.

8 Parsons-AlliedSignal Technology Package INTRODUCTION AND OVERVIEW The Parsons-AlliedSignal team has submitted a pro- posal under the acronym WHEAT (water hydrolysis of explosives and agent technology). This technology package is comprised of five basic technologies: · The Army's baseline disassembly process, with modifications including waterjet cutting for rock- ets, is used to separate agent, energetics, and metal parts. · Hydrolysis is the primary process for detoxifying the agent and deactivating energetics. · Biological processing, supplemented by ultravio- let/hydrogen peroxide treatment, is used to con- vert the hydrolysis products to materials accept- able for discharge to the environment. · Metal parts and dunnage are decontaminated to 5X by heating in high-temperature steam. · Gas discharges from the plant go through a cata- lytic oxidation unit for treatment. TABLE 8-1 Summary of the Parsons-AlliedSignal Technology Package Table 8-1 lists how these technologies are used to perform the six major demilitarization operations listed in Chapter 2. A process flow diagram for the Parsons-AlliedSignal package is presented in Fig- ure 8-1, and a detailed description of the package is given in the next section. The technology provider addressed the processing of rockets, projectiles, and mortars but did not consider the processing of land mines. DESCRIPTION OF THE TECHNOLOGY PACKAGE Disassembly of Munitions and the Removal of Agent/Energetics The baseline disassembly process (see Appendix C) is used to a large extent in this technology package. However, some modifications to the baseline process are proposed as described below. Major Demilitarization Operation Approach(es) Disassembly of munitions Army baseline disassembly, augmented with water jet cutting Treatment of chemical agent Base hydrolysis; biotreatment of hydrolysate Treatment of energetics Waterjet wash-out; base hydrolysis; biotreatment of hydrolysate Treatment of metal parts Heat in steam to SX conditions in metal parts treater; catalytic oxidation of gas Treatment of dunnage Treatment in metal parts treater;catalytic oxidation of gas Disposal of waste Solids. Dry salts and biotreatment product to appropriately permitted landfill; calcined grit to landfill Liquids. None Gases. Discharge to atmosphere after catalytic oxidation, caustic scrubbing (and possibly carbon filtration) 119

120 o '. I ~ ~n~° CD i0 Q Q ~ (1) In, ~ Q _ ~ .O Q o (n - X US . . O i0 ~ G5 L (n . o Q o ·-O ~ . _ i0 ·X O o _ a) ~ ._ O Qua 0-0~ ~ O ~ X ~ Q' (n 'A O Q) OIL X 0= .~ U) o . _ .O . _ . ¢ 3 ~ ' ~ o ~3- it L a: o ._ 3 it ~ ~

PARSONS-ALLIEDSIGNAL TECHNOLOGY PACKAGE Projectile Disassemb/y The fuze, booster (if present), and burster are re- moved using the baseline process. The burster tubes and supplementary charges are then placed in a special fixture, and high-pressure hot water is used to dissolve and/or physically remove energetics from the tubing. The wash-out solution is fed to the energetics hydroly- sis reactor as a batch. Boosters are washed out in the same manner as bursters. The fuzes are sent to a rotary metal parts treater, which will be described shortly. The agent is drained from the projectile as in the baseline process. The projectile body is then sent to the metal parts treater, which is described later in this chapter. Mortar Disassemb/y The fuze and burster are removed by the baseline process; subsequent fuze and burster processing is the same as for the projectile. The agent is drained from the mortar as in the baseline process. Rocket Disassemb/y Rockets are disassembled with the rockets still in- side their shipping and firing tube. Chemical agent is removed using the baseline punch and drain approach and is pumped to an agent storage tank prior to hy- drolysis. Waterjet cutting with abrasive (garnet) is then used to sever the fuze from the rocket, and the fuze is sent to the rotary metal parts treater. The rocket (with the fuze removed) advances to a wash-out station, where high-pressure (4,500 psi), hot water is used to remove the burster energetics (Par- sons-AlliedSignals, 1998~. Abrasive waterjet cutting is subsequently used to separate the warhead from the motor section, and a high-pressure (15,000 psi) hot- water jet washes the propellant from the motor section. The burster energetics, propellant, and washout solu- tion are fed to the energetics hydrolysis reactor. The grit is separated from the cutting water; the cutting water is sent to the agent hydrolysis reactor; and the grit is sent to the rotary metal parts treater. Following the removal of energetics, the warhead and motor sec- tions are inspected and sent to the metal parts treater. 121 Treatment of Chemical Agents and Energetics The chemical agents and the aqueous dispersion of energetic materials derived from wash-out of the mu- nitions are considered together in this section because they follow the same processing sequence. Hydro/ysis The hydrolysis reaction conditions for agents are the same as those outlined in Appendix D. Reaction times will of this be specified to ensure very high conversion, (e.g., 99.9999 percent destruction efficiency). Sched- ule 2 compounds formed from the hydrolysis of each agent will require further treatment by bioreaction. Energetic material is fed to the hydrolysis reactor as an aqueous slurry, having been reduced to a fairly small particle size (e.g., less than a quarter inch). (An explo- sives shredder may be used to reduce particles to this size.) The reaction rate is expected to be controlled by diffusion to the solid surface of the particle and is, therefore, dependent on particle size. Reaction condi- tions are the same as those outlined in Appendix E. All hydrolysis products are transferred to storage tanks before they are fed to the bioreactors. The hy- drolysis reactors are operated as batch reactors; bio- reactors are operated continuously. Tanks are sized to accommodate the change between the batch and con- tinuous operating modes. Both VX and GB produce hydrolysates that contain a small amount of an organic phase. The reactors and storage tanks are constantly stirred to prevent the or- ganic phases from separating from the bulk aqueous phase. GB hydrolysate is more dilute than VX hydroly- sate (i.e., 8 percent of the reaction products are from GB compared to 30 percent from VX in their respec- tive hydrolysates). The dilution prevents the formation of a solid precipitate from GB hydrolysis, which is probably sodium fluoride and some iron salt (see Appendix D). Bio/ogica/ Treatment Aerobic bioreactors oxidize the hydrolysates (from chemical agents and energetics) to the following products:

122 . ALTERNATIVE TECHNOLOGIES FOR DEMILITARIZATION OF ASSEMBLED CHEMICAL WEAPONS carbon dioxide, water, and biomass (solid prod- ucts of the biological cell mass produced in the reactions) (the technology provider estimates that about 80 percent of the carbon in the process feed is oxidized to CO2; the balance is in the organic biomass (sludge), as well as a small amount of organic matter remaining in solution) other products, such as fluoride, sulfate, nitrite, ni- trate, phosphate salts in solution, and ammonia some low molecular weight, partially oxidized species (e.g., acetic acid), as well as some organic compounds that color the aqueous solution (color bodies) The biological reaction is relatively slow. The liquid residence times in bioreactors (the so-called hydraulic residence time) are typically five to 15 days, although the technology provider believes that five days will be sufficient. The average residence time of biomass can be as long as two months. The bioreactor design will include an AlliedSignal development called an immobilized cell bioreactor, which holds the biomass that develops in the reactor on a porous screen (sponge). In other bioreactor designs, the biomass floats freely in the liquid. Advantages claimed by the technology provider for the immobi- lized cell bioreactor are (1) more rapid reaction because the biomass that accumulates on the screen is more concentrated than in the free floating alternative; and (2) lower production of biomass overall usually an advantage because disposing of the biomass is a prob- lem. The lower biomass production may not be an ad- vantage in nerve agent disposal, however, because the biomaterials use phosphorus primarily to produce bio- mass, rather than for metabolism, and a bioreactor that produces more biomass will more effectively eliminate the phosphorus from the nerve agent hydrolysate. Since the original proposal was made, the technol- ogy provider has suggested using a combination of an immobilized cell bioreactor and a "conventional" bioreactor. The combined bioreactor would consist of a long box, with its long side horizontal. The reactor would be aerated to supply the oxygen for the bio- reactions. The liquid in the reactor would be stirred by the air, which would have the undesirable consequence of keeping the entire process at the lowest reactant con- centration (i.e., the exit concentration). To avoid this, the reactor would be compartmentalized, with liquid flowing horizontally from stage to stage within the re- actor. The reactor would have two to four stages. The first stages would use the immobilized cell bioreactor design for rapid reaction. The last stage would have free-floating biomass to promote the removal of phos- phorus. Bioreactors used only for mustard and energet- ics (that contain no phosphorus) are expected to use the immobilized-cell technology for all stages. The combination of a very large volume of feed (low concentration of organics and salt in water), and a hy- draulic residence time of five days, dictates the reactor volume. The basic reactor module will be a 40,000- gallon "box" the largest size that can be transported by highway. Much larger reactor volumes will be needed, however, and this requirement will be met by adding more 40,000-gallon tanks. The technology provider has suggested a basic mod- ule of four 40,000-gallon tanks grouped around a cen- tral "facilities" corridor (pumps, blowers, metering equipment and controls, etc.~. The number of modules will be determined in the final plant design by the re- quired rate of munitions destruction, as well as on the particular munitions. The technology provider has sug- gested a configuration with three modules for the Pueblo, Colorado, site and four modules for the Rich- mond, Kentucky, site. Thus, there would be 12 or 16 40,000-gallon reactors at these sites (total bioreactor volumes of 480,000 and 640,000 gallons). The reactors are operated with very dilute solutions. For example, experimental work has been done with feed concentrations of less than 0.7 wt. percent; the tentative plant design (for VX) calls for 1.2 wt percent. This means the agent and energetics feed would repre- sent 1.2 wt. percent of the material fed to the bioreactor. The technology provider has suggested that salt con- tents of up to 4 percent could be tolerated. Because the hydrolysis products, which are the feed streams, are typically in the range of 5 to 10 wt. percent, the hy- drolysate feed stream must be diluted by a large factor, as much as 10-fold, before entering the bioreactor. Effective bioprocessing requires pH control. The product from the hydrolysis reactors has a pH of 10 to 12, which is generally too high for the bioorganisms to tolerate. The pH is adjusted by adding acid to lower the pHto 8.5. The microbial population responsible for the

PARSONS-ALLIEDSIGNAL TECHNOLOGY PACKAGE bioreactions imposes additional requirements on the feed to the reactor to keep the population alive. Three important food elements are carbon, nitrogen, and phosphorus. The optimum ratio for carbon to nitrogen tophosphorusis 100:5:1 (Lupton, 1998~.Because none of the hydrolysates meets this criterion, other materials are added, such as dextrose. The amounts of these extra materials vary depending on the particular hydrolysate mixture being fed, but for hydrolysate from nerve agents, the extra nutrients are a very large addition to the feed to the plant. Mixing chemical agent hydroly- sate, which is rich in phosphorus, with energetics hy- drolysate, which is rich in nitrogen, can also help real- ize the optimum feed. The bioreactor cannot eliminate organic material completely. Enough must remain in solution to sustain microbial life. The solution leaving the bioreactor is expected to have a dark color, due to organic color bod- ies. The depth of color depends on the feed material. The effluent from hydrolysate from munitions is par- ticularly dark. One product of the bioreaction is sludge, which is flocculated, dewatered, dried on a drum evaporator, and packaged for disposal in a landfill. The resultant solid will probably contain most of the heavy metals from the original feed and may be classified as toxic. There are two other product streams from the bioreactor: (1) the aeration air, mixed with product gases from the big-organisms and volatile materials (in- cluding low molecular weight chlorinated hydrocar- bons from mustard) and some liquid "spray;" and (2) a large stream of water containing dissolved salts and the remaining organic material. Catalytic Oxidation A large volume of gas leaves the bioreactor. This gas is then heated to 425 to 450°C (797 to 842°F) and passed to a catalytic oxidation unit for the removal of trace organics, oxidizable nitrogen, and chlorine com- pounds. The gas is then cooled and scrubbed using ei- ther a liquid or a solid soda scrubber. It may then be passed through a carbon filter before release to the air. A carbon filter has been installed on the Parsons- AlliedSignal ACWA demonstration system. If analyti- cal data show that the filter is necessary, it will be in- cluded in the final plant design. 123 Salt Recovery and Water Management Most of the water in the liquid bioproduct is re- cycled. Salts are recovered by evaporation, and most of the steam is condensed and recycled. A reverse osmo- sis unit is included in the technology provider' s ACWA demonstration system to reduce the volume of water that must be evaporated. There is no plan to include the reverse osmosis unit in a final plant design, however. The salt content of the reactor liquid is affected by the fraction of product liquid withdrawn and evapo- rated. If the fraction is large (e.g., approaching 1), the salt content can be maintained as low as 1 percent, but at the expense of a large evaporation requirement. The technology provider has suggested a salt content of 2 percent as a reasonable level, although the microbial population can tolerate levels as high as 3 or 4 percent. The organic matter remaining in the bioreactor liq- uid from VX or GB disposal may contain low levels (ppm) of Schedule 2 phosphonates. An ultraviolet /hy- drogen peroxide treatment is used to reduce these ma- terials to levels below detection limits. Brine from the bioreactor unit is first concentrated in an evaporator and then evaporated to dryness. The salt is recovered with a rotary-drum dryer. The steam from the evaporator is condensed and the water re- cycled. The steam from the drum dryer is released to the air. The salt, with a small residual organic compo- nent' is packaged in drums for disposal. Treatment of Metal Parts Metal parts are heat treated to produce metal cleaned to a 5X condition, which can be released from Army control. The important difference from the baseline approach is that the heating medium is superheated steam instead of combustion gas. A batch process is used to treat the parts to a 5X condition. The metal parts treater is an autoclave with induction heating. The metal parts are assembled in a basket and placed in the metal parts treater, which is then purged with nitrogen followed by low-pressure steam. The metal parts treater and its contents are quickly heated to about 650°C (1,202°F), and the sys- tem is held at that temperature for 15 minutes. At the end of the process, the steam is swept with air to a condenser, and the gases are passed on to a reheater and a catalytic oxidation unit. Organic materials driven

124 ALTERNATIVE TECHNOLOGIES FOR DEMILITARIZATION OF ASSEMBLED CHEMICAL WEAPONS off the metal parts and broken down by the heat are oxidized to CO2, H2O, and possible acids, such as P2O5. The gas from the catalytic oxidation unit is scrubbed with lime or caustic solution, possibly passed through a carbon filter, and exhausted to the environment. Fuzes and associated small metal parts are also heated in steam in a rotary metal parts treater where the fuzes ignite or explode. The small metal parts are treated to a 5X condition for disposal. Treatment of Dunnage Some contaminated dunnage is treated in the metal parts treater. For example, DPE suits are shredded and then vaporized in the metal parts treater, leaving be hind a small ash residue. Some dunnage may be held until the plant is closed. Some may be carefully exam ined for contamination and reused (e.g., pentachlo rophenol-treated wooden pallets). Grit from waterjet cutting is heated to a 5X condition in the rotary metal Gas Streams parts treater. Process Instrumentation, Monitoring, and Control On-line chemical analysis during baseline disassem bly and during operations unique to the Parsons AlliedSignal process is limited to (1) a determination of pH and specific gravity at critical control points; (2) chemical oxygen demand and levels of all nutrients for the immobilized cell bioreactor; (3) ACAMS monitor ing of exhaust gases and of ventilated spaces and criti cal operations, such as munition overpacks; and (4) monitoring of sulfur and phosphorus content of the catalytic oxidizer inlet stream. The bioreactor is instru mented to monitor pH, temperature, and inlet and out let chemical oxygen demand. Feed Streams The following materials will be required by the plant: · grit for waterjet cutting · caustic for agent and energetics hydrolysis and for gas scrubber solutions · nitric acid for pH control · nutrients for the bioreactor, with dextrose in the largest amount · a flocculating agent, probably a polymeric amine · hydrogen peroxide for final polishing of the bioliquid product · carbon adsorbent for gas and ventilation air cleanup · fresh water (no wastewater leaves the plant, but some water is lost as water of crystallization of salts, water with the biosolids, humidity added to the air, etc.) · decontamination solution (sodium hypochlorite) Waste Streams There are three gaseous and four solid waste streams. There is no liquid effluent from this process. The largest gas stream is estimated to be 30,000 ac- tual cubic feet per minute (ACFM) of slightly depleted air, with oxygen content reduced from 21 percent to about 19 percent. This stream is treated in a catalytic oxidation unit. A second, much smaller stream, also treated in a catalytic oxidation unit, is a product of the metal parts treaters. A third gas stream is steam from the drum dryers used to dry salts and biomass. This stream is released to the atmosphere. So/id Streams Metal Parts. All of the metal parts from the original munitions are cleaned to a 5X condition. Biosludge. The dried biosludge, together with a small amount of a flocculating agent, probably a poly- meric amine, is expected to amount to 10 to 20 percent of the mass of material fed to the reactor (agent plus energetics plus added nutrients). Sodium Salts of Various Acids. Sodium salts are pro- duced from the heteroatoms (i.e., fluoride, chloride, sulfate, nitrate and nitrite, and phosphate) and are a solid waste. Chemical agents yield a large amount of salt (e.g., VX yields salt equal to about 150 percent of the mass of the original VX). The yield of solid salts from the munitions varies with the munition but will

PARSONS-ALLIEDSIGNAL TECHNOLOGY PACKAGE probably be roughly double the weight of the agent and energetics in the munition. The salts may have a small organic residue left from the biotreatment and the treat- ment with ultraviolet light/peroxide. This remains to be demonstrated. Grit (Garnet). This material, used in waterjet cut- ting, is treated at 1,000°F for at least 15 minutes in steam. Mass Balance Mass balances for two munitions were provided by Parsons-AlliedSignal: an 8-inch GB projectile and a 4.2-inch mortar with mustard (Tables 8-2 and 8-3~. (Both tables show Fenton's solution being used, al- though that is now considered unlikely.) These mass balances cannot be considered exact because the prod- ucts from the bioreactor are uncertain. The fraction of the feed (C, H. and N) that will be oxidized to gaseous products and the fraction oxidized to solid biomass are not known with certainty. Experience suggests that about 10 to 20 percent of the feed (C, H. N) to the bioreactor, including added nutrients, will end up as biomass; the rest will be oxi- dized to gas products. The values shown in Table 8-3 demonstrate the large amounts of nutrients and dex- trose required for bioprocessing nerve agent (GB) hydrolysate. TABLE 8-2 Mass Balance for Processing HD 4.2-inch Mortars (lb/lb HD) Component Amount (lb) Input Streams Flocculent + Fenton's reagent Air Nutrients HD Energetic materials Sulfuric acid NaOH Water Total input Output Streams Air Sludge (wet) Water (evaporated) Salt Total Output 0.2 278.0 0.3 1.0 0.1 o.o 1.0 9.4 290.0 280.0 0.6 7.8 1.6 290.0 Source: Parsons-AlliedSignal, 1999a. 125 TABLE 8-3 Mass Balance for Processing GB 8-inch Projectiles (lb/lb GB) Component Amount (lb) Input Streams Flocculent + Fenton's reagent Air Nutrients Dextrose GB Energetic materials Sulfuric acid NaOH Water Total input Water (evaporated) Salt Total Output 7.8 2,897.0 6.3 44.4 1.0 0.5 2.8 2.7 50.1 3,012.6 2,930.0 53.0 24.3 4.9 3,012.2 Source: Parsons-Allied Signal, 1999a. Start-up and Shutdown The bioreactor is the largest volume unit in the Par- sons-AlliedSignal technology package. This unit will be run continuously but with some possible changes in the feed. Experience suggests that the time required for acclimation of the big-organism to a mustard hydroly- sate feed is only a few hours. Acclimation to a nerve agent hydrolysate that contains phosphorus in the form of a phosphonate compound may take several weeks. Disposal campaigns are planned accordingly. In the event of a nonroutine shutdown, most pro- cessing units, which are batch or semibatch operations, can be shut down and held on stand-by status. The bioreactor can withstand a shutdown of the air supply for only a few hours, however. For a longer shutdown, the bioreactor will require an auxiliary feed and air to maintain the microbial population. EVALUATION OF TH E TECH NOLOGY PACKAG E Process Efficacy Effectiveness of Munitions Disassembly The technology provider claims that munitions han- dling, disassembly, and plant safety design and practices

126 ALTERNATIVE TECHNOLOGIES FOR DEMILITARIZATION OF ASSEMBLED CHEMICAL WEAPONS will be modeled closely after the baseline system for which significant operating experience is available. These technologies have all been demonstrated and can be considered to be mature. Two new technologies will be introduced in the dis- assembly process: waterjet cutting and waterjet clean- out. Both have been used in the demilitarization of con- ventional munitions. A brief overview of the state of the art for demilitarizing ordnance using high-pressure jet cutting and clean-out of the energetic materials is given in Appendix G. In view of the significant devel- opments in waterjet cutting technology and its tested use for cutting high-explosive casings, the technique can be considered suitable for application in disassem- bly operations. Explosive and propellant will be recovered from the munitions by high- pressure waterjet clean-out, which has been used on a substantial scale in conventional demilitarization operations for many years. However, it has only been used for removing energetic materials from ordnance and not for minimizing the particle size of the energetic material. The committee believes that a small particle size (e.g., 0.25 inch) will be necessary for the hydrolysis reaction. Therefore, the simultaneous removal and size reduction of energetic materials will have to be demonstrated. Effectiveness of Hydrolysis In the Parsons AlliedSignal Technology package, hydrolysis will be the main technology for achieving a 99.9999 percent agent destruction. Subsequent treat- ment in the bioreactor is expected to destroy the hydro- lyzed materials, but the bioreactor should handle little if any agent. The hydrolysis reactions of chemical agents have been studied extensively (see Appendix D), and hydrolysis for VX and HD should have accu- mulated many hours of demonstration (at Aberdeen and Newport) before an ACWA-based plant starts up. The Army has already hydrolyzed several hundred pounds of GB and VX to prepare hydrolysate for ACWA tech- nology demonstrations of SCWO and biotreatment. The hydrolysis of energetic materials is considered a less mature technology than the hydrolysis of agent (see Appendix E). The design will have to allow for various reaction times and for various quantities, de- pending on the type of munition being processed. Effectiveness of Biotreatment The use of natural microbial consortia for the degra- dation and mineralization in a biotreatment system de- pends mainly on providing organic food sources and nutrients to the microorganisms. A sustainable food- to-microorganism ratio must be maintained in the bioreactor to ensure microbial viability. In theory, mi- croorganisms can be made to mineralize almost any organic contaminant, but in practice the toxicity of or- ganic and inorganic constituents in the feed can be a major problem that requires close monitoring and con- trol. Furthermore, biotreatment alone cannot remove all of the organics in the hydrolysate. A final polishing step may be required to meet regulatory levels. Biotreatment of the hydrolysate from mustard will be used at the Aberdeen Proving Ground facility. Tests using a "sequencing batch reactor" system have been quite successful. The liquid product from this process will go to the sewage treatment plant of the Aberdeen Army base before final release to the environment. Similar testing of a batch reactor was conducted for biotreatment of VX hydrolysate for possible use at the Newport, Indiana, bulk storage site. However, difficul- ties were encountered that were not completely re- solved (DeFrank et al., 1996~. VX hydrolysate contains phosphorus in the form of phosphonate (i.e., with a C- P bond), which appears to be difficult to metabolize. Recent tests by the technology provider, however, have been successful in removing the phosphorus to very low levels (more than 95 percent removal) (Parsons- AlliedSignal, l999b). Removal of phosphonate appears to depend on the following factors: . . . The bacteria will use phosphonate as long as no other form of phosphorus is available. Therefore, other materials, such as phosphates must be rigor- ously excluded. The optimum ratio of major nutrients for the bio- mass is in the approximate ratio C:N:P = 100:5:1. Therefore, relatively large amounts of nutrients containing C and N have to be added to make use of the phosphorus in the feed. For example, dex- trose was added in a ratio of 44 lb per lb of origi- nal agent in the hydrolysate. Phosphorus is used to produce biomass, rather than for metabolism. Therefore, a combination of reactor types must be used for optimum results

PARSONS-ALLIEDSIGNAL TECHNOLOGY PACKAGE (e.g., an immobilized-cell reactor followed by a free-floating biomass reactor). In contrast to the difficulties experienced with the bioreaction of nerve agent hydrolysates, the bioreaction of mustard hydrolysate has worked well, although food supplements are required. Mustard does contain low molecular weight chlorinated hydrocarbons that are difficult to biodegrade, and they will be sent to the cata- lytic oxidizer, either in the effluent air from the bioplant or in gases leaving the evaporator. The catalytic oxida- tion unit is expected to be effective in destroying them, but this will have to be demonstrated. The committee anticipates that the ACWA demon- stration tests being conducted during the preparation of this report will address a number of questions concern- ing the bioprocess. First, the pH of GB hydrolysate will be adjusted to 8.5 before it goes to the bioreactor. This pH is low enough for GB to re-form (see Appendix E). This possibility will be monitored during the demon- stration. Second, the biosludge, amounting to 10 to 20 percent of the carbon in the feed, will have to be tested for toxicity. Third, the technology provider has indicated that the final effluent from the bioreactor after post-treatment, should have a biological oxygen demand of < 100 mg/L, a chemical oxygen demand of < 1,000 mg/L, and a total organic carbon of < 100 mglL. This chemical oxy- gen demand is somewhat lower than reported in the technology provider's proposal and will have to be demonstrated. (The biological oxygen demand and to- tal organic carbon are usually much lower than the chemical oxygen demand, though exceptions can oc- cur.) A "polishing" step (hydrogen peroxide with ultra- violet light ~ to reduce them further is provided for the nerve agent hydrolysates. Fourth, the airflow rate through the bioreactors is far above the stoichiometric requirement for mineralizing the feed. In early laboratory tests reported by the tech- nology provider, the air flow was 25 to 50 times the stoichiometric requirement. In more recent larger scale work, however, the technology provider has demon- strated satisfactory operation with 12-fold stoichiomet- ric requirements (i.e., about 8 percent of the oxygen in the inlet air was used). Operations should require as small an air supply as feasible for the bioreactors be- cause the product air must be treated further (i.e., 127 heated to 425°C t797°F] for the catalytic oxidation unit, cooled for acid gas scrubbing, and possibly reduced in relative humidity for activated carbon adsorption). Fifth, material vaporized or entrained in the air from the bioreactor may affect the performance of the cata- lytic oxidation unit. For example, entrained phospho- rus could deactivate the catalyst. The gas stream from the bioreactors will have to be characterized. Also, tests will have to show whether entrained material will be a problem. Sixth, the salt content of the feed to the bioreactor must be kept low for the microbial population. The technology provider has stated that it will be main- tained at less than 3 percent. It appears that a salt con- tent of about 1 percent is being used in the ACWA demonstrations. The level will affect the requirements for the salt-recovery evaporator. A low salt content of about 1 percent (which will be demonstrated) appears to be a conservative choice for operation. Effectiveness of Evaporation for the Production of Sa/ts and Bioso/ids The evaporation processes for recovering biosolids and salts are well established. The products will prob- ably be considered toxic, however, because most heavy metals in the feed will end up in the biosolids. Polysac- charides formed in the big-operation are known to be good sequestering agents for heavy metals. (See Tables D-3 and D-9 in Appendix D for a list of heavy metals found in some samples of VX and HD.) Some metals will also appear in the salts from the evaporation process. Heavy metals in the biosludge will probably prove to be nonleachable as defined by EPA's TCLP test be- cause they are usually tightly bound to the polysaccha- rides. Heavy metals in the salts from nerve agent pro- cessing will probably also be relatively nonleachable because phosphates are present. However, heavy met- als in the salt product from mustard may be more soluble. Effectiveness of Cata/ytic Oxidation Two separate catalytic oxidation units, with subse- quent alkaline scrubbing and carbon filtration, are pro- posed. One treats gas from the metal parts treaters, the

128 . , , ALTERNATIVE TECHNOLOGIES FOR DEMILITARIZATION OF ASSEMBLED CHEMICAL WEAPONS other treats gas from the bioreactors. The catalytic oxi- dat~on units resemble automobile catalytic converters and use proprietary AlliedSignal technology. They must handle a large gas flow, estimated at 30,000 ACFM, from the bioreactors, and the oxidizable mate- rial will be very dilute. The proposed catalyst has been shown to be effective at residence times as low as 0.1 seconds and should oxidize most of the chlorinated materials in the gas stream (arising from mustard). Small molecules are difficult to oxidize, however, and the efficiency of the catalyst in oxidizing materials such as methylene chloride (which might be in the gas) must be demonstrated. The possible presence of chlorinated dioxins and furans in the product gas should be checked. If they are present, the proposed carbon filters could effectively remove them. A small amount of liquid mist in the gas leaving the bioreactors should be expected. Phosphorus and sulfur (derived from nerve agents and mustard, respectively) in this mist may negatively affect the catalyst. The cata- lyst has been shown to destroy nerve agents and mus- tard in short-term tests, but the technology provider recognizes that long-term data are limited, and durabil- ity has yet to be demonstrated. This is one area being investigated during the ACWA demonstration tests. A caustic scrubber and an activated carbon filter to treat the effluent from the catalytic oxidizer are included as an extra safeguard. (The presence of agent in the gas stream from the bioreactors appears to be highly un- likely. Much smaller gas streams Efrom the metal parts treaters or venting of feed tanks] might contain some agent.) The technology provider's proposal states that sul- fur and phosphorus concentrations in the gases flowing to the catalytic oxidation units will be monitored. Con- tinuous monitoring of the exhaust-gas stream is used in place of a hold-test-release monitoring process. Effectiveness of Peroxide/U/travio/et Oxidation The hydrogen peroxide/ultraviolet light treatment process must be tested to demonstrate its effectiveness. The solution to be treated is colored (brown) and un- doubtedly contains finely divided solids in suspension. Ultraviolet radiation is directed into the solution through quartz windows, and it may be difficult to keep the window clean during VX processing. The technol- ogy provider's goal is to reduce the total organic car- bon to less than 50 ppm and to reduce residual phospho- nate to 3 ppm. Effectiveness of the Meta/ Parts Treater The metal parts treater and rotary metal parts treater will heat the metal parts (as in the baseline system) but will use steam (at 650°C t1,202°F]), rather than com- bustion gas. The process will qualify as a 5X treatment. Sampling and Analysis The proposed sampling and analysis methods are generally well established and straightforward and should not pose significant difficulties. Maturity The overall process is a combination of many (at least 10) different technologies, all of which have sub- stantial operational backgrounds, although some will require demonstration for their application to chemical weapons destruction. Hydrolysis of mustard and VX are planned at Aberdeen, Maryland, and Newport, In- diana, respectively. Extensive testing and development of the hydrolysis processes has been done for the de- sign of these two facilities. Biotreatment of mustard hydrolysate is planned for the Aberdeen facility, and a similar level of testing and development has taken place, albeit on a different type of bioreactor than the immobilized cell bioreactor. Tests of the biological treatment test work on mustard hydrolysate have been quite successful. Biotreatment of energetic hydroly- sates has also been successful, though the full range of materials has not been demonstrated. Ultraviolet/per- oxide treatment of the mustard hydrolysate to remove low molecular weight chlorinated hydrocarbons is also being investigated at Aberdeen, but this process differs from the treatment proposed by Parsons-AlliedSignal for ACWA. The concentration of chlorinated hydro- carbons at Aberdeen should be much lower than the concentration of organics to be treated in this applica- tion. The bioreaction of a feed with high phosphorus content, particularly with the phosphorus present as a "phosphonate" (i.e., C-P bond), has not been proven.

PARSONS-ALLIEDSIGNAL TECHNOLOGY PACKAGE Individually, all unit operations can be considered mature technologies, and some have been applied to the treatment of assembled chemical weapons materi- als. Nevertheless, past industrial experience has shown that starting up a process with many steps in series can- not be accomplished without significant difficulties. Robustness All of the technologies appear to be reasonably op- erable and robust in that they can be readily controlled to desired set points and can accommodate modest fluc- tuations in feed composition. Munitions disassembly, energetics wash-out, hydrolysis, and biotreatment will already have been applied to chemical weapons dis- posal at Aberdeen and Newport. The sensitivity of the bioreactor to fluctuations in feed or contamination is being addressed in the demonstration. Biological con- versions bring their own special problems, however, particularly in dealing with living materials. Problems with feed toxicity or predators for the microorganisms may develop. The problems previously mentioned all appear to be solvable. However, one aspect of robust- ness that cannot be determined at this time is the ability of the integrated process to handle feed variations through the entire set of technologies. Monitoring and Contro/ Overall, the proposed monitoring, control, and in- strumentation system appears to be modern in design and well thought out. Parsons-AlliedSignal proposes using as much of the design as possible of the Army's current baseline technology, taking advantage of the design and lessons learned from that experience. The following monitoring and control technologies are new to chemical weapons destruction: · the design of monitoring, control, and instrumen- tation of waterjet cutting and wash-out systems · the demonstration of a biomass accumulation in the reactor that can achieve the necessary conversions · control of foaming in the evaporator (a small amount of organic material from the bioreactor in the brine would affect the liquid surface property and could lead to foaming in the evaporator, a 129 common problem that can probably be handled by additives) · demonstration of the steady-state ultraviolet/per- oxide oxidation step Applicability Many of the technologies in this technology pack- age will have been applied to chemical weapons before the start up of an ACWA plant based on this overall process. Catalytic oxidation (on a large scale), metal decontamination (to 5X) with steam, and waterjet cut- ting applied to rockets will be new. In the committee's opinion, the technology package is conceptually appli- cable to the treatment of all assembled chemical weap- ons. However, mine processing was not addressed by the technology provider, and successful biotreatment of nerve agent hydrolysate must still be demonstrated. Process Safety The unique equipment proposed by the Parsons- AlliedSignal team is associated with the following technologies: · waterjet cutting · wash-out of agent and energetics · shredding of energetics · base hydrolysis · biological treatment · ultraviolet/peroxide oxidation · catalytic oxidation · decontamination of metal parts with high-tem- perature steam in the metal parts treaters · decontamination of grit and munition fuzes in high-temperature steam in the rotary metal parts treater Waterjet cutting, wash-out, and energetics hydroly- sis will be done in explosion-containment areas. The hydrolysis process operates at temperatures up to 90°C (194°F); the bioreactor processes operate at ambient temperature. Both processes operate at ambient pres- sure. The catalytic oxidizer and the metal parts treaters operate at elevated temperatures, 425°C (797°F) and 649°C (1,200°F), respectively. The hydrolysis reactors and the 5X treaters, which represent the primary detoxi- fying processes, operate in a batch mode; thus, the

130 ALTERNATIVE TECHNOLOGIES FOR DEMILITARIZATION OF ASSEMBLED CHEMICAL WEAPONS effectiveness of treatment can be ascertained prior to release of the processed material to the next step. The remaining systems are routine chemical pro- cesses, and in this application, they occur down- stream of the primary (hydrolysis) and secondary (biotreatment) detoxifying processes. These systems should pose no unique hazards. The equipment con- sists of caustic scrubbers, carbon filters, evapora- tors, and dryers. Worker Health and Safety If a process upset occurs, the primary destruction components (the hydrolysis reactors and bioreactors) cannot be shut down quickly because incomplete hy- drolysis products, are extremely hazardous. Procedures are expected to be established for safe shutdown and restarting of the system. The air effluent during an up- set will continue to be treated, first in the catalytic oxi- dizer, then in the caustic scrubbers, and, potentially, in activated carbon filters. The low-speed shredder for energetics poses a po- tential worker safety issue. Friction, shear, or heat may result from the inadvertent introduction of metal, an excessive feed rate, or some other cause and could ini- tiate the energetic material. Workers are not expected to be present, however, during normal operations. The rotary metal parts treater can be designed to ac- commodate detonations of fuzes (fuze detonation chambers are not unique). Workers are not expected to be present during normal operations. Only trace amounts of energetics will be present in the metal parts treater under expected operating condi- tions. Scenarios for the introduction of energetics be- yond design conditions will be evaluated to ensure that they are extremely unlikely before the design is com- pleted. Potentially flammable dunnage pyrolysis prod- ucts are being characterized during ACWA demonstra- tion testing, and the impact of these and other effluents should be considered as the design develops. The technology provider plans to hydrolyze differ- ent types of energetic materials simultaneously in the same reactors. As discussed in Appendix E, the com- mittee is concerned that this could lead to the forma- tion of compounds that are both energetic and sensi- tive. Therefore, different energetic materials should be processed in separate reactors unless tests shows that the formation of sensitive compounds does not occur. Small amounts of aluminum particulates created during waterjet cutting of the rocket warhead sections will generate hydrogen during the hydrolysis step. The hydrogen and hydrolysis gases will be vented to the bioreactor off-gas stream and then heated to 425°C (797°F) for treatment in the catalytic oxidizer. The large bioreactor off-gas flow rate will dilute the hydro- gen to well below the flammability limit before heating in the oxidizer. Standard operational procedures and designs for flammable gas systems, (e.g., maintaining a negative pressure to avoid release to air spaces) should be adequate to minimize explosion hazards. The biosludge produced in the bioreactor could con- tain some pathogenic microorganisms. The potential for worker exposure to these microorganisms is ex- pected to be minimized by appropriate protective gear. Waterjet technology is commonly used in the de- militarization of conventional munitions and should not pose unique safety issues. The ACWA demonstration includes tests of the capability of this technology to separate the fuze and the rocket motor from the war- head. Even if an ignition occurs, there will be little risk to workers because the cutting is performed remotely in an explosion-containment area. The energetics hydrolyzer incorporates an external circulation and cooling loop. Pumping an aqueous slurry of energetic materials can be done safely under the proper conditions. If an accident occurs during nor- mal operations, there would be little risk to workers ~. ~ . . . . because the loop is in an explosion-containment area. The loop will have to be designed to ensure that ener- getic material does not precipitate and accumulate in the piping, which could result in an accident during maintenance procedures. The primary hazardous materials used are sodium hydroxide, nitric acid, sodium hypochlorite, and hy- drogen peroxide. These chemicals are used routinely at many industrial facilities and are not unique to the Par- sons-AlliedSignal process for demilitarization. Public Safety The release of agent and other regulated substances in plant effluents is judged to be extremely unlikely. The destruction of agent and energetics will be verified

PARSONS-ALLIEDSIGNAL TECHNOLOGY PACKAGE by hold-test-release operations before the transfer of hydrolysate from the hydrolysis reactors to the bio- reactors and before the transfer of bioreactor sludge to the sludge containerization step. The gaseous effluent from the bioreactors will be continuously released through activated carbon filters, if the demonstration test results indicate that this is desirable or necessary. No hold-test-release operation is provided for the gas- eous effluent stream from the bioreactors. Because this stream will be continuously monitored for hazardous materials, the release of hazardous materials is consid- ered extremely unlikely. There is a low probability of agent release in case of a failure of the rotary metal parts treater. This hazard can be mitigated by good design and operational pro- cedures that are confirmed by risk analyses. GB reformation in the bioreactor because of low pH (see Appendix D) is being investigated during demon- stration testing. Any GB vented from the feed tank or the bioreactor is expected to be destroyed in the cata- lytic oxidizer or captured in the backup activated carbon filters. Human Health and the Environment Eff/uent Characterization and Impact on Human Health and Environment In the absence of health risk and environmental as- sessments, a precise statement on the impact of the ef- fluents on human health and the environment cannot be made at this time. However, the gas flow leaving the plant should be free of hazardous material. It will have been exposed to a very high temperature (about 425°C [797°F] in the catalytic oxidation unit), and it will have been through extensive cleanup processes to remove traces of organic materials (including any agent). Two of the solid materials leaving the plant will be treated as hazardous waste: salts with traces of organic material; and biomass with small amounts of salts. Completeness of Eff/uent Characterization The very large gas flow, primarily from the bio- reactor, will have gone through catalytic oxidation and ... . . . 131 then be tested routinely for chemical agent, oxygen, carbon dioxide, and carbon monoxide on a real-time basis. It should also be characterized for low concen- trations of hazardous materials, such as dioxins. Biomass will be tested periodically for leachability and for toxicity. Salt residue will also be tested for leachability. Other effluents that have been treated to a 5X condition will not require further characterization. Eff/uent Management Strategy Salts. Dried salt, probably containing some organic materials, will contain sodium salts of fluoride, chlo- ride, sulfate, nitrate, and nitrite. The technology pro- vider expects this waste stream to be hazardous. Sta- bilization, either at the plant or at a commercial hazardous-waste treatment facility, may be required. Experimental studies will be necessary to determine the leaching levels of hazardous constituents in the salt and then, if required, determine the additives needed to stabilize the salt. Stabilization and burial in a hazard- ous-waste landfill should provide adequate protection to human health and the environment. Biosludge. The composition of the biosludge pro- duced in the bioreactor is unknown at this point, al- though the technology provider expects it to be hazard- ous because it will be the sink for most heavy metals from the process. The sludge will also contain micro- organisms, some of which might be pathogenic. Test- ing will be necessary to determine whether or not the waste is hazardous as defined by the EPA. If so, dis- posal in a hazardous-waste landfill may not be possible because of the biological activity. Incineration of the waste is an alternative. If it is not hazardous, it can most likely be sent to a municipal solid-waste landfill without threat to human health or the environment. Gas. Exhaust gas from the catalytic oxidizer will pass through an acid gas scrubber (and possibly through an activated carbon adsorber). Whether or not any con- stituents of concern will be present in this stream is not known at this point. Analyses will be necessary to con- firm the presence or absence of low molecular weight hydrocarbons and chlorinated hydrocarbons, oxides of nitrogen, and chlorinated dioxins and furans. acid gas scrubbing. The composition will have to beMetal Parts and Garnet Grit. Metal parts and grit determined in detail during initial trials. The gas shouldare cleaned and deactivated to the 5X condition in the

132 ALTERNATIVE TECHNOLOGIES FOR DEMILITARIZATION OF ASSEMBLED CHEMICAL WEAPONS metal parts treaters. The cleaned parts are not expected to pose any threat to human health or the environment. Resource Requirements The Parsons-AlliedSignal technology package has three steps that will consume large amounts of energy: (1) producing 650°C (1,202°F) steam for the metal parts treater; (2) heating the vent gas from the bio- reactor to about 425°C (797°F) for the catalytic oxi- dizer; and (3) evaporating the salt solution from the bioreactor. The power requirement has been estimated to be a few megawatts. None of the resource require- ments appears to be excessive. Environmenta/ Compliance and Permitting There are no apparent reasons that the combination of technologies selected by the technology provider should lead to unusual permitting or compliance prob- lems. The absence of liquid emissions is an important advantage of the process. However, the catalytic oxi- dation operations are close enough to incineration in concept that regulatory (and public) concerns could be raised. STEPS REQUIRED FOR IMPLEMENTATION The following steps would have to be taken to imple- ment this technology package: 1. demonstration of the effectiveness of the bio- treatment of various combinations of agent and energetics hydrolysates of sufficient length to give reasonable assurance of long-term perfor- mance operation of the bioreactor at the planned salt- content a. characterization of the off-gas from the bioreactor to evaluate the extent of air-stripping from the re- actor and the possible poisoning of the catalyst in the catalytic oxidation unit 4. demonstration of the effectiveness and long-term performance of the catalytic oxidation system in destroying organic constituents in the bioreactor off-gas 5. quantification and characterization of the sludge from the biological process to ascertain if Sched- ule 2 compounds or other hazardous constituents are present 6. demonstration of unproven steps in the proposed process, including ultraviolet/peroxide oxidation and evaporation operations 7. quantification and characterization of the salts from the evaporation operations to ascertain what organic compounds are present FINDINGS Finding PA-1. The biological treatment operation will require further demonstration to prove its ability (1) to handle a variety of feed stocks with reasonable accli- mation times between changes, and (2) to achieve high levels of conversion of the Schedule 2 compounds in the hydrolysate. The demonstration will have to last long enough to give confidence in the long-term opera- tion ability of the process. Finding PA-2. The relative effects of biological treat- ment and air-stripping on the destruction of organic materials in the bioreactor have not been established. This will affect the composition of the off-gas from the bioreactor. Finding PA-3. The effectiveness of ultraviolet/ hydrogen peroxide oxidation in reducing Schedule 2 compounds to an acceptably low level has not been demonstrated. Finding PA-4. The bioreactor has been operated only at very low salt concentrations. Operation at design concentrations has not been demonstrated. Finding PA-5. Additional data should be gathered on the effectiveness of the catalytic oxidation system in destroying organic materials in the biotreatment off-gas. Finding PA-6. The sludge from the biological process has not been completely characterized. Finding PA-7. Even though the evaporation operations involve conventional technologies, they have not been tested for this application. Finding PA-X. The dried salts from the evaporation operations have not been characterized for leachability and toxicity.

Next: 9 Teledyne-Commodore Solvated Electron Technology Package »
Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons Get This Book
×
 Review and Evaluation of Alternative Technologies for Demilitarization of Assembled Chemical Weapons
Buy Paperback | $75.00 Buy Ebook | $59.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

This report examines seven disposal technologies being considered by the U.S. government as alternative methods to the process of incineration for destroying mortars, rockets, land mines, and other weapons that contain chemical warfare agents, such as mustard gas. These weapons are considered especially dangerous because they contain both chemical warfare agent and explosive materials in an assembled package that must be disassembled for destruction. The study identifies the strengths and weaknesses and advantages and disadvantages of each technology and assesses their potential for full-scale implementation.

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!