Improved Computer Modelling for Slow Decomposition Processes PhD

Make a difference: help to design safer explosives by developing predictive computer models. Slow chemical decomposition of explosives can eventually lead to ignition, and we need to understand that process in more detail so that we can assure and improve safety. In collaboration with laboratories in the UK and USA, this PhD opportunity aims to investigate and develop novel computer material models which more accurately capture the chemical and physical changes that occur in explosives at elevated temperatures.

When an energetic material is at elevated temperature the chemical reaction rate increases, leading to increased heat generation and eventual thermal runaway. Current finite element models are able to simulate this decomposition, but only in terms of the chemistry of decomposition, and they implement physical and microstructural changes relatively poorly. Workers in the UK and USA are actively researching this new area.

Existing models make two important and linked assumptions: that decomposition products are gaseous; and that the products are quickly transported away from the reaction zone and so no longer take part in global heat transfer. This means that existing computer models are best suited to fast decomposition at millisecond or microsecond time-scales, where the assumptions are valid, and historically the short timescale regime has been of most interest in safety management. For materials where the decomposition products are solid or where decomposition occurs over timescales of minutes or hours, however, these assumptions are invalid, and so existing computer models are less realistic and less usefully predictive.

The proposed PhD project aims to develop extensions to existing computer models of energetic materials so that they are useful at low temperatures and long timescales, with particular focus on the changing heat transfer properties of materials as they undergo decomposition. This studentship will be primarily theoretically based, involving computer model development and simulation, but validated against data obtained from experimental observation. There is opportunity for collaboration with research organisations in the UK and with USA national laboratories.