Click for next page ( 54


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 53
Appendixes

OCR for page 53

OCR for page 53
Appendix A Autoignition and the M55 Rocket As a consequence of the applicability of the Army's broad experience with energetics in conventional weap- ons, there is no reason to expect autoignition of either the fuze or burster components of chemical munitions. However, the M55 rocket also contains 19.1 lb of M28 propellant, which is of a double-base composition, con- taining nitroglycerine (NG), nitrocellulose (NC), and 2-nitrodiphenylamine (NDPA) as a stabilizer. The NO and NC components degrade slowly under storage con- ditions, generating heat and releasing nitrogen oxides (NOX). These oxides react with trace moisture to form nitrogen acids, which lead to corrosion problems and heat generation. The NDPA stabilizer reacts with the NOX to form, successively, three disubstituted and two trisubstituted daughter products, each of which can fur- ther react with NOX, although with lower effectiveness. Each NDPA molecule (including subsequent daughter reactions) can theoretically combine with up to six NOX molecules. Because NDPA has only a limited capacity for absorbing NOX, which diminishes for its daughter species, the rate of propellant degradation increases as aging proceeds. This process also generates thermal energy. If the heat generation rate exceeds the rate of heat dissipation through the rocket and its packing, the propellant temperature will increase, thereby increas- ing degradation rates and, ultimately, the likelihood of autoignition. In addition to the stabilizer depletion that takes place normally in propellant under storage condi- tions, concerns have also been raised about GB leakage 55 into propellant grains and the impact that such leakage has on propellant stability. Further, when leaking GB M55 rockets are overpacked, the ability to dissipate heat from the rocket to the surroundings is reduced. Overpacking can thus contribute to heating of the pro- pellant and consequently to higher reaction rates, which might increase the possibility of autoignition. The original stabilizer concentration in the M28 pro- pellant was a nominal 1.8 percent, although, as noted earlier, a small sample of M55 rockets evaluated in 1985 showed propellant NDPA levels ranging from about 1.6 to 2.2 percent (NRC, 1994~. MITRE Corpo- ration studies done prior to 1994 suggested that if the rockets were stored at 95F, the primary NDPA would be depleted to 0.5 percent by 2007 (see, for example, Perry et al., 1993~. At the 0.5 percent concentration level, Army guidelines call for increasing surveillance of the munitions, since autoignition has been observed to occur when levels are 0.2 percent or lower. How- ever, this does not consider the additional stabilization capacity provided by the NDPA-daughter species. A subsequent Army study included the effect of stabili- zation by the daughter species and estimated that there would be a 50 percent likelihood of an autoignition in the M55 rocket stockpile in the time frame from 2041 to 2120 (U.S. Army, 1994~. In September 1996, the Army assembled a panel of outside experts to assess the likelihood of autoignition of M28 propellant in the M55 stockpile. This group,

OCR for page 53
56 EFFECTS OF DEGRADED AGENT AND MUNITIONS ANOMALIES ON CHEMICAL STOCKPILE DISPOSAL OPERATIONS TABLE A-1 Calculated Best-Estimate Autoignition Probabilities for M55 Rocket Sites Year Tooele Anniston Umatilla Pine Bluff Blue Grass 2010 7.8x 10-~7 9.0x 10-~6 1.4x 10-~5 1.2x 10-~5 1.7x 10-~5 2015 5.11 x 10-~4 5.2 x 10-~3 8.3 x 10-~3 6.6 X 10-~3 9.1 X 10-~3 2020 5.2 x 10-~2 5.2 x 10-~i 8.2 x 10-~i 4.9 x 10-~i 8.5 x 10-~i 2025 2.0x 10-~ 1.7x 10-9 2.7x 10-9 1.8x 10-9 2.7x 10-9 2030 3.8 x 10-8 3.1 x 10-8 4.8 x 10-8 3.0 x 10-8 4.7 x 10-8 Source: U.S. Army (2002b). through expert elicitation techniques that considered detailed mechanisms and uncertainties, developed fur- ther insights on the effectiveness of NDPA and its daughter products in capturing NOX generated during M28 propellant degradation, estimated that NDPA lev- els would have to be less than 0.2 percent to allow autoignition, and suggested that autoignition might not occur immediately after depletion of the stabilizer (U.S. Army, 1997~. Using all the information developed, es- timates of the likelihood of autoignition were prepared for all sites having M55 rockets (Table A-1~. The study estimated that in 2010, the frequencies of autoignition at the sites were on the order of 10-~5 per year or less; by 2020, the frequencies had risen to around 10-~ per year. Science Applications Interna- tional Corporation (SAIC) uses a frequency of 10-8 per year as the criterion for inclusion of a risk contributor in a quantitative risk analysis (QRA), and autoignition risks would not be above this threshold until 2025. It is common practice in QRAs to establish a quantitative value to screen out scenarios that do not singularly or collectively contribute significantly to the risk metrics of interest. According to the available analyses, autoignition will not significantly contribute to risk for the anticipated operational life of the sites. Further, the contribution of autoignition to overall storage risk is negligible, since the frequencies of ignition by a light- ning strike at the sites storing M55 rockets vary from 6 x 10-4 to 5 x 10-3 per year. The expert group also advised the Army to focus special attention on the effect that leaking GB may have on the stability of M28 propellant. Experience has shown that about 0.5 percent of GB M55 rockets have leaked (in contrast to 0.02 percent of the VX M55 rock- ets). As leakers are discovered, they are overpacked. When 99 overpacked GB leakers at TOCDF were sampled 11 to 32 years after overpacking, some of the overpacks showed pressure buildup due to continuing NOX generation. Further, NDPA levels were measured and found to be lower than would be expected in the absence of GB contamination. A second expert elicita- tion conducted in November 1998 evaluated the effect of agent contamination on the likelihood of autoignition Since VX rocket leakers are rare, atten- tion was focused on GB rockets. The expert panel con- cluded that M28 propellant degradation was enhanced by the presence of GB, but that additional work was needed to clarify the characteristics of the interactions. The Army selected Midwest Research Institute to perform both experimental studies and analysis of leaker sample tests to assess the effects of agent con- tamination on M28 stability (U.S. Army, 2000, 2002a). Experiments were conducted at elevated temperatures (50C and 65.5C) to increase reaction rates. Only modest increases in stabilizer depletion rates occurred with small amounts of GB contamination, and NO and NC also were depleted along with the stabilizer. How- ever, at concentrations above about 6 percent GB in the propellant, the stabilizer depletion rate increased rap- idly. A detailed model was developed to predict degra- dation (in terms of remaining effective stabilizer) and thermal behavior for GB-contaminated propellant grains. A significant feature of this model was that va- por transport was a major pathway by which GB agent could reach the propellant in addition to direct liquid leakage. This pathway became increasingly important at higher temperatures (due to higher vapor pressures) and in sealed containers. Thermal modeling included detailed heat transfer paths through the rocket and in the two types of over- pack systems. Heat generation from internal chemical iNotes from conference Expert Elicitation on Autoignition of Agent-Contaminated M28 Rocket Propellant, Science Applications International Corporation, November 18-20, 1998.

OCR for page 53
APPENDIX A reactions was also modeled. The possibility of leakage that was not externally detected, i.e., leakers that were not overpacked, was also considered. Where the analy- sis showed that the peak heat generation rates exceeded the rate of heat transfer out of the system, autoignition was assumed to occur. The analysis confirmed that the degradation process over time results in increasing NOX generation, which in turn further increases degradation as the NOX pres- sure increases. Overpacking may be accelerating deg- radation. Controlled venting of overpacks may reduce degradation rates and heat buildup. A comprehensive Army report in March 2002 con- cluded that the probability of an autoignition event from propellant stabilizer depletion (including consequent thermal effects) is as follows (U.S. Army, 2002b): for nonoverpacked (undetected) leaking GB M55 rockets, Pa = 1.8 x 10-7 to 1.4 x 10-6 per year, and for over- packed leaking GB M55 rockets, Pai = 1 x 10-5 to 6 X 10-5 per year. Estimates were reported for the years 2000, 2005, 2010, and 2020; the above ranges include the median values to the year 2020 (U.S. Army, 2002b). The probabilities of propellant ignition by other 57 means range from 6 X 10-4 to 5 x 10-3 per year. By comparison then, the degradation of GB-contaminated propellant leading to autoignition appears to be a fairly small part of the total storage risk. All of these values are subject to uncertainties of about an order of magnitude. REFERENCES NRC (National Research Council). 1994. Evaluation of the Johnston Atoll Chemical Agency Disposal System, Operational Verification Testing: Part II. Washington, D.C.: National Academy Press. Perry, J., G. Nelson, and D. Tripler. 1993. Assessment of the U.S. Chemical Weapons Stockpile: Integrity and Risk Analysis, April. McLean, Va.: MITRE. U.S. Army. 1994. M55 Rocket Storage Life Evaluation, December. Aber- deen Proving Ground, Md.: Program Manager for Chemical Demilitari- zation. U.S. Army. 1997. Report on the Expert Elicitation on Autoignition of Nonleaking Rockets, May. Aberdeen Proving Ground, Md.: Program Manager for Chemical Demilitarization. U.S. Army. 2000. Evaluation of GB Effects on M28 Propellant. Aberdeen Proving Ground, Md.: Program Manager for Chemical Demilitariza- tion. U.S. Army. 2002a. Preparation and Analysis of M55 Leaker Rocket Propel- lant Samples Final Test Report. Aberdeen Proving Ground, Md.: Pro- gram Manager for Chemical Demilitarization. U.S. Army. 2002b. M55 Rocket Assessment Summary Report. Aberdeen Proving Ground, Md.: Program Manager for Chemical Demilitarization.