The rate of thermal decomposition of RDX has been investigated in the presence of its decomposition products and free radical traps. From the measurements, it is concluded that formaldehyde and nitrogen dioxide, presumably ?encaged? in the sample, catalyse the decomposition of RDX positively and negatively respectively. The non-volatile residue also acts as a positive catalyst. The other products have little or no effect on the rate, and the free radical traps did not reduce the rate.

The kinetics of the decomposition of RDX have been investigated in the presence of formaldehyde over the temperature range 170-197� with the RDX sample spread. This indicated a marked increase in the positive- catalytic effect of the formaldehyde with decreasing reaction temperature; however, the kinetics were not altered by the added formaldehyde. ��� The activation energy was about 44 kcal mol-1. It is suggested that the previously obtained activation energy of about this figure, for the decomposition of heaped samples of RDX in the absence of added formaldehyde, was due to catalysis of the reaction by the decomposition product formaldehyde.

Two mechanisms have recently been proposed to explain the behaviour of the initial rate of decomposition of RDX, with change in sample geometry. These are (i)that the decomposition proceeds by concurrent gas and liquid phase reactions, and (ii) that gaseous decomposition products influence the rate of decomposition of undecomposed RDX in the condensed phase. In this paper it is concluded that mechanism (ii) is the more probable when the reaction is carried out in the presence of nitrogen.

The activation energy has been determined in the temperature range 170-198�. If the sample was spread the activation energy was independent of the definition of the kinetic parameter substituted in the Arrhenius equation and was 63 kcal mole-1. In the case of the unspread samples the activation energies of the induction, acceleration, and maximum rates were 49, 43, and 62 kcal mole-1 respectively. The effect that sample geometry has on the activation energy is attributed to gaseous decomposition products influencing the reaction.

The rate of the direct reaction between a phenol and benzoyl peroxide increases with decreasing strength of the 0-H bond. This supports the view that the reaction occurs by homolytic transfer of the phenolic hydrogen atom to the carbonyl atom of the peroxide. Additional support is provided by the fact that steric effects which limit the accessibility of the hydrogen atom cause a decrease in the rate.The efficiency of a phenol in retarding the chain decomposition of benzoyl peroxide induced by dioxan increases as the 0-H bond strength decreases.THE preceding paper showed that the reaction between comparable concentrations of benzoyl peroxide and p-methoxyphenol in an inert solvent occurs mainly by direct interaction of the phenol and peroxide molecules. It was suggested that the reaction occurs by the transfer of the phenolic hydrogen atom to the carbonyl oxygen atom of the peroxide :This involves rupture of the 0-H bond of the phenol and the 0-0 bond of the peroxide, with formation of a molecule of benzoic acid. Since of these processes only the first depends on the structure of the phenol, the rate of reaction of the peroxide with different phenols would be expected to increase with decreasing strength of the 0-H bond. This paper describes an investigation of the effect of the structure of the phenol on the rate of reaction, carried out with the object of testing this conclusion. critical oxidation potential " (E,) of the phenol a s defined by Fieser,2 a high value of E, corresponding to a high bond strength (cf. p. 2960). The reaction rate should therefore increase with decreasing E,. The efficiency of a phenol as an antioxidant increases with decreasing A measure of the strength of the 0-H bond is given by the Batten and Mulcahy, preceding paper.