Case histories Of 65 incidents in runaway reactions and emelgency relief in Taiwan were analyzed and classified into several categories according to their causes, materials involved, equipment types, reaction types, and ignition sources. The cases in reactors and storage tanks were examined in more detail owing to the higherprob ability or lalgerpotential hazard in these two types of equipments. The most common consequence of the incidents are explosions, fires, and atmospheric release of toxic chemicals. The most severe case was a thermal explosion from an organic peroxide storage area which caused the death of 33 persons. Popping and direct releasing of process chemicals to the atmosphere from relieving devices cause the greatest environmental concerns to the community close to the plants. Runaway reactions in batch type reactors occur frequently due to various operational mistakes. Heat of reaction is the most frequent ignition source of runaway reactions and emergency relie$

An algorithm for calculating the parameters of a steady one-dimensional detonation wave in mixtures of a gas with chemically inert particles and estimating the detonation-cell size in such mixtures is proposed. The calculated detonation parameters and cell size in stoichiometric hydrogen-oxygen mixtures with W, Al 2 O 3 , and SiO 2 particles are used to analyze the method of suppression of multifront gas detonation by injecting chemically inert particles ahead of the leading wave front. The ratio between the channel diameter and the detonation-cell size is used to estimate the limit of heterogeneous detonation in the mixtures considered. The minimum mass of particles and the characteristic cloud size necessary for detonation suppression are calculated. The effect of thermodynamic parameters of particles on the detonation suppression process is analyzed for the first time. Particles with a high specific heat and (if melting occurs) a high phase-transition heat are found to exert the most pronounced effect.

Dynamic and steady release tests have been performed to uncover the precise condition for pure silane ignition upon its release into air. Two regimes (dynamic and steady) are studied. In dynamic tests, silane is released from a vessel with a known pressure. It is found that prompt ignition is not observed even with the source pressure down to 0.15 MPa. In steady flow tests, a reproducible critical exit velocity is found above which silane can be released indefinitely into air without any ignition. With the aid of the laminar boundary layer theory, it is found that ignition always occurs in a well-defined mixture fraction called the most reactive mixture fraction. The critical exit velocity and the most reactive mixture fraction suggest that silane release without prompt ignition is most likely caused by flow strains or by scalar dissipations, which prevent chemical reactions of silane oxidation.