Page 436 - Dust Explosions in the Process Industries
P. 436

Ignition of Dust Clouds and Dust Deposits  403


               of temperaturerise was affected significantlyby the activation energy and frequency factor
                of the coal oxidation. Measurement of the moisture absorbed on the oxidized coal sam-
               ples showed that the loss in mass due to oxidation increased markedly at temperatures
                above 120°C.By assuming that the limiting temperature for significant self-heating in
                coal storageis 12O"C, a maximum permissible size of stored coal depositto prevent self-
               ignition was estimated for various types of coal.
                  Hensel(1988) was concerned with a similar problem, predicting maximum permissi-
               ble storage periods for large coal piles. He extrapolated empirical laboratory-scalecor-
               relationsbetween the volume/surface area ratio of the dust deposit and the induction time
               to ignition.An induction time of 10years was predicted for some 20-.year-old,large coal
               piles in Berlin, in which self-ignitionhad been observed repeatedly over the last years.
               By extrapolatingthe laboratory-scaledata, Hensel also confirmedthat the size of the actual
                coal piles was larger than the maximum permissible size for preventing self-ignitionat
                average ambient air temperatures in the Berlin region.

                5.2.4.2
                Natural Organic Materials

               Raemy and Loliger (1982) used a heat flow calorimeterfor studying the thermal behav-
               ior of cereals above 20°C. When the sampleswere heated in sealed measuring cells, strong
                exothermic reactions were observed at about 170°C. These reactions were attributed
                mainly to carbonizationof the carbohydratesin the cereals. Raemy and Lambelet (1982)
               conducted a similar heat flow calorimetric study of  self-heating in coffee and chicory
                above 20°C.
                  In a study of the thermal behavior of milk powders, Raemy, Hurrel, and Loliger (1983)
               used both heat fl.owcalorimetry and differentialthermal analysis. They found that four
               main types of reactions are involved in the thermal degradation of milk powders. In order
                of increasing temperature, they are crystallizationof amorphous lactose, Maillard reac-
               tions, fat oxidation, and lactose decomposition.
                  Self-ignition properties of  fish meals were studied by Alfert and coworkers at CMI,
               Bergen, Norway, by storing the samples, supported by metal gauze blaskets, in air at con-
                stant temperatures in the range 100-250°C.  Some results were reported by Hostmark
                (1989).For 1-and 2-liter samples,the minimum ambient airtemperaturesfor self-ignition
               were 140 and 13O"C,respectively. The corresponding induction times to ignition were
                5-6  and 8 hours. At ambient air temperatures exceeding 20O-24O0C,  the dust samples
                ignited close to the surface after induction times on the order of 2 hours (see trend in
               Figure 5.7.)

                5.2.4.3
               Corrosion of  Direct-Reduced Iron

               Birks and Alabi (1986, 1987, 1988) were concerned with the special problem of  self-
               ignition in piles of direct-reducediron when exposed to water. The problem arose because
               direct-reduced iron is stored and transported in charges of considerable size, and it had
               been observed that the bulk material has a tendency to oxidize to an extent leading to
                self-ignition.Bilrks and Alabi investigated the various chemical reactions operating when
                direct-reduced iron reacts with water and the oxygen in the air.
   431   432   433   434   435   436   437   438   439   440   441