Foreign gas effects in the oxidation of phosphene : the mechanism of the transformation of white to red phosphorus in presence of tungsten : its relationship to the heterogeneous initiation of chains in phosphorus-oxygen mixtures

• Main Author: Gray, Stewart C.
• Published: University of Edinburgh 1935
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.651752

Part I. Foreign gas effects in the oxidation of phosphine The effect of foreign gases on the lower explosion limit of PH3 -02 mixtures has been determined as accurately as the experimental conditions will permit. H2, Ne, A, N2, CO2, N20 and SO2 all lower the limit in the normal way by impeding the diffusion of the chains to the walls. There is no measurable gas phase inhibition when the pressure of the gas is of the sam order as that of the phosphine and oxygen. This agrees with observations on the effect of some of these gases on the upper limit. C2H4, C5H6, PbMe4, and CC14 all raise the lower limit in marked contrast to the lowering of the limit observed with P4 -02 mixtures. It is shown that the reason for the difference is that the probability of branching of chains in the PH3 -02 reaction is less than that in P4 -02 mixtures, and that a given inhibit +r exerts a more powerful influence on the former reaction. The value of the inhibition coefficient has been calculated and, in the case of ethylene, checked by measuring the inhibition of the stable photo-oxidatio of phosphine. An extended table of inert gas coefficients has been compiled. Part II. The mechanism of the transformation of white to red phosphorus in presence of tungsten. Its relationship to the heterogeneous initiation of chains in phosphorus-oxygen mixtures. The transformation of white to red phosphorus via the gas phase has been studied. Phosphorus vapour was introduced into reaction vessels containing tungsten filaments which could be heated to various temperatures. At filament temperatures between 500 and 2000 ?0 red phosphorus was deposited on the reaction vessel walls. The range of reaction rates was no less than 106: 1. All the evidence points conclusive- ly to the fact that P4 molecules are dissociated into P2 molecules by the filament. The reaction is approximately unimolecular. The energy of activation is 41 kg. cals. at low collision efficiences. Attention is drawn to certain discrepancies in the heats of vaporisation of white and of red phos- phorus. These can be removed by postulating that P2 molecules evaporate from and are in true equilibrium with red phosphorus. An attempt was made to evaporate red phosphorus in vacuo to demonstrate that only P2 molecules leave the surface but the experimental conditions were not sufficiently well defined to yield conclusive evidence in support of the hypothesis. Ra?sing the temperature of the reaction vessel walls leads to reflection of the P2 molecules from the walls. A fraction also recombines on the heated walls to form P4. P2 molecules do not migrate along the walls. Decreasing the stream density of P2 molecules on the surface 100 fold makes no difference to the rate of deposition of red phosphorus. By measuring the watts input to the filament, and the number of molecules reacting, it has been found that the heat of dissociation of P4 is 25 kg. cals. which compares favourably, having regard to the method, with the value calculated from the variation of equilibrium constant with temperature. Oxygen poisons the tungsten surface for the dissociation. When phosphorus is present, however, the tungsten is not attacked. The rate of oxidation of phosphorus is slower than the reaction P4--p 2P2 hence P2 molecules cannot start the chains. These results point to the conclusion that a lower oxide evaporates from the surface to initiate chains. This is a zero order reaction and it appears that the 'orientation of the phosphorus oxide is, diagrammatically, W - P - 0, the oxygen adsorbed on top of the phosphorus. The reaction may go on an oxygen- treated filament. Oxygen is removed during the reaction at higher temperatures, but not at all at lower temperatures. There is no evidence to disprove, however, that the reaction, in this case, really takes place on the parts of the 'filament uncovered by oxygen.