Vapor cloud explosion blast prediction models, such as Baker-Strehlow and the Multi-Energy method, implicitly assume that the occurrence of a detonation in the vapour cloud is unlikely. Hence the prediction of far field blast is based on the combustion energy of the congested part(s) of the flammable cloud only. In case of transition of a deflagration into a detonation (DDT), those parts of the cloud that are not engulfing congested parts of a process installation may also contribute to blast generation since a detonation can propagate without “the support” of obstructions. Blast damage due to a vapor cloud explosion in case of a DDT may occur at far larger distances from the blast source.

Recently a new parameter has been introduced into the FLACS simulation tool that can be used to identify whether DDT is likely in a given scenario and indicate the regions where it might occur. The likelihood of DDT has been expressed in terms of spatial pressure gradients across the flame front and a criterion comparing geometric dimensions with the detonation cell size of the local gas mixture [1,2].

In the current study, this parameter that was originally developed and verified for hydrogen-air mixtures is verified against experimental data available on hydrocarbon-air mixtures including propane [3], ethylene [4] and acetylene [5]. The paper reports on this verification. Experiments performed for less reactive gas mixtures in the various geometries were also simulated to validate the methodology further for the cases where DDT is not observed experimentally, but high-speed deflagrations are seen.

Next, using a geometry representative for a petrochemical plant simulations are performed with various hydrocarbons to investigate the likelihood of detonations arising during vapour cloud explosions in this geometry. Both homogeneous stoichiometric clouds and realistic clouds (i.e. clouds developed from a dispersion) are considered. The likelihood of a transition to detonation in vapor cloud explosions depending on the reactivity of the fuel is discussed in the light of the results of the simulations and the results of the verification of the model against experimental results.

References:

1. Middha, P, Hansen, O. R., and Storvik, I. E. (2006). Proceedings of the 40th Annual Loss Prevention Symposium, Orlando, FL, April 23-27, 2006.

2. Middha, P. and Hansen, O.R. (2007). Proceedings of the 41st Annual Loss Prevention Symposium, Houston, TX, April 23-26, 2007.

3. Hjertager, B.H., Fuhre, K, Parker, S.J. and Bakke, J.R. (1984). Progress in Astronautics and Aeronautics, 94: 504-522, AIAA Inc. New York

4. McKay, D.J., Murray, S.B., Moen, I.O. and Thibault, P.A. (1989), Twenty-Second Symposium on Combustion, 1339-1353, The Combustion Institute, Pittsburgh

5. Moen, I.O., Bjerketvedt, D, Jenssen, A. And Thibault, P.A. (1985), Comb. and Flame, 61: 285-291