Aqueous foam concentrate (AFC) 380 foam was developed by Sandia National Laboratory as a blast mitigation foam for unexploded ordnance (UXO) and its ''engineered foam structure'' is reported to be able to ''envelop chemical or biological aerosols'' [1]. It is similar to commercial fire-fighting foams, consisting mostly of water with small amounts of two alcohols, an ether and surfactant. It also contains xanthan gum, probably, to strengthen the foam film and delay drainage. The concentrate is normally diluted in a 6:94 ratio with water for foaming applications. The diluted solution is normally foamed with air to an expansion factor of about 100 (density 0.01 g/cc), which is called ''dry'' foam. Higher density foam (0.18> {rho}> 0.03 g/cc) was discovered which had quite different characteristics from ''dry'' foam and was called ''wet'' foam. Some characterization of these foams has also been carried out, but the major effort described in this document is the evaluation, at the small and medium scale, of chemical, mechanical and thermal approaches to defoaming AFC 380 foam. Several chemical approaches to defoaming were evaluated including oxidation and precipitation of the xanthan, use of commercial oil-emulsion or suspension defoamers, pH modification, and cation exchange with the surfactant. Of these the commercial defoamers were most effective. Two mechanical approaches to defoaming were evaluated: pressure and foam rupture with very fine particles. Pressure and vacuum techniques were considered too difficult for field applications but high surface area silica particles worked very well on dry foam. Finally simple thermal techniques were evaluated. An order-disorder transition occurs in xanthan solutions at about 60 C, which may be responsible for the effectiveness of hot air as a defoamer. During defoaming of 55 gallons of foam with hot air, after about 70% of the AFC 380 foam had been defoamed, the effectiveness of hot air was dramatically reduced. Approximately 15 gal of residual foam containing mostly small bubbles was resistant to further defoaming by methods that had been effective on the original, dry foam. In this paper the residual foam is referred to as ''wet'' and the original foam is referred to as ''dry''. Methods for generating ''wet'' foam in small to moderate quantities for defoaming experiments have been developed. Methods for defoaming wet foam are currently under study.

The Chemical Reactivity Test, or CRT, has been the workhorse for determining short-to-medium term compatibility and thermal stability for energetic materials since the mid 1960s. The concept behind the CRT is quite simple. 0.25 g of material is heated in a 17 cm{sup 3} vessel for 22 hours at 80, 100, or 120 C, and the yield of gaseous products are analyzed by gas chromatography to determine its thermal stability. The instrumentation is shown in Figure 1, and the vessel configuration is shown in Figure 2. For compatibility purposes, two materials, normally 0.25 g of each, are analyzed as a mixture. Recently, data from the past 4 decades have been compiled in an Excel spreadsheet and inspected for reliability and internal consistency. The resulting processed data will be added this year to the LLNL HE Reference Guide. Also recently, we have begun to assess the suitability of the CRT to answer new compatibility issues, especially in view of more modern instrumentation now available commercially. One issue that needs to be addressed is the definition of thermal stability and compatibility from the CRT. Prokosch and Garcia (and the associated MIL-STD-1751A) state that the criterion for thermal stability is a gas yield of less than 4 cm{sup 3}/g for a single material for 22 hours at 120 C. The gases from energetic materials of interest ordinarily have an average molecular weight of about 36 g/mol, so this represents decomposition of 0.5-1.0% of the sample. This is a reasonable value, and a relatively unstable energetic material such as PETN has no problem passing. PBX 9404, which yields 1.5 to 2.0 cm{sup 3}/g historically, is used as a periodic check standard. This is interesting in itself, since the nitrocellulose in the 9404 is unstable and probably has partially decomposed over the decades. However, it is not clear whether this aging of the standard would lead to more or less gas, since the initial gaseous degradation products are captured by the DPA stabilizer. Clearly this is an issue that needs reconsideration. The criterion for compatibility is less clearly correct. Although some LLNL reports say that generation of gas in excess of the materials by themselves is an indication of incompatibility, LLNL reports invariably say that materials are compatible if they generate less than 1 cm{sup 3}/g of gas. There are two problems with this criterion. First, it is not stated whether the gas yield is per gram of energetic material or mixture. Second, a material that generates>2 cm{sup 3}/g by itself could never pass the compatibility tests as stated, because even a mixture of equal masses of that material with a completely inert material would generate>1 cm{sup 3}/g of gas per mixture. Furthermore, Prokosch states that a yield equal to or less than from the materials individually means that no reaction has occurred. Clearly, less gas can not be generated unless some type of interaction has occurred. An obvious example would be mixing CaO with a CO{sub 2}-generating energetic material. In the absence of any direct action of the CaO on the energetic material, the CO{sub 2} product would be captured by the CaO, thereby decreasing the gas yield and liberating considerable heat. In a large, closed volume, this could tip the balance to thermal runaway.