Prediction of the Minimum Ignition Energy and Minimum Ignition Temperature of Dust Clouds

Dust explosions are hazardous phenomena widely reported among industries such as grain handling and storage, food and feed processing, coal mining and storage, wood processing, metal working and pharmaceutical component manufacturing. Correct prediction of the dust explosion parameters is thus of great importance for industrial safety. Two common types of ignition sources of dust explosions are electrical sparks or arcs and hot surfaces. For a better understanding of the electric spark and hot surface sensitivity of dust clouds, the Minimum Ignition Energy (MIE) and the Minimum Ignition Temperature (MIT) were introduced to describe these two sensitivities. In recent years, experimental testing of the MIE and the MIT have become rather mature with the standardization on the testing equipment, while theoretical prediction of the MIE and MIT is still challenging.

First, in this thesis, a thermal model for the MIE calculation of single particle size pure dust clouds is developed and validated. This model considers the particle-to-particle propagation of ignition in the dust clouds for both cylindrical and spherical spark shapes. Variation trend of the MIE calculated with the cylindrical model fits well with the experimental data, but there are deviations between the absolute experimental and calculated MIE values. Possible reasons for the deviations between the absolute experimental and numerical MIE values, the influence of particle size and dust concentration on the MIE are discussed in detail.

Second, this complex thermal model with ignition propagation among dust particles is simplified considering the ignition of only a single dust particle. Comparison between the complex and simplified models verifies the feasibility of using the MIE of only a single particle at the ignition center as a prediction of the dust cloud MIE. A parametric study is then performed using the simplified single central particle model, which shows that most dust cloud and spark parameters have rather simple power function relationships with the MIE. However, influence of the ambient, the spark and the dust cloud minimum ignition temperature on the dust cloud MIE is too complex to be analytically expressed. Nevertheless, piecewise function approximations for the influence of the three temperatures on the MIE are derived. Based on these functions, simple MIE calculation equations are established as an alternative for the numerical solution of the simplified single central particle model.

Third, the thermal model for single particle size pure dust clouds is extended for the MIE calculation of pure dust clouds with particle size distribution, mixtures of combustible dust clouds, and mixtures of combustible and inert dust clouds. A particle size distribution approximation approach is introduced where the original particle size distribution is divided into multiple subdivisions. The median sizes of particle number based size distributions of these subdivisions are used to approximate the original distribution. As the number of subdivisions increases, the calculated MIE first fluctuates but gradually stabilizes at a certain value. For both pure dust clouds and mixture dust clouds, the calculated MIE variation trends agree with the experimental data. However, there are still deviations between the absolute experimental and calculated MIE data. A parametric study for the influence of the inert particle size, density and specific heat implies that only relying on the heat sink effect of the inert particles to increase the MIE of a combustible and inert dust mixture cloud is not satisfactory. A rather large proportion of the inert component is required for a significant increase of the MIE.

Last, A thermal model is established for the MIT calculation of single particle size pure dust clouds based on the Godbert-Greenwald furnace, considering heat transfer between the air and dust particles, the dust particle reaction kinetics, and the residence time of dust clouds in the furnace. In general, for the 13 dusts studied, the calculated MIE data are in agreement with the experimental values. There is also great accordance between the experimental and numerical MIT variation trends against particle size. Two different ignition modes of dust clouds are discovered, one is boundary thermal ignition of individual particles with rather short ignition delay times, the other one is thermal explosion starting from the center by self-heating of the whole dust cloud with longer ignition delay times. The numerical MIT results of magnesium dust clouds assuming infinite residence time are also compared with the analytical results with the classic Semenov and Frank- Kamenetskii thermal explosion theories. The numerical results and the analytical results from the classic thermal explosion theories are almost identical.