Spontaneous ignition in storage and production lines: Investigation on wood pellets and protein powders

Abstract

SUMMARYThree materials were tested in small-scale baskets of cubic dimensions and analyses were performed based upon the results according to the Frank-Kamenetskii (FK) method and the crossing point (CP) method.Comparison of the FK method and the CP method results was performed additionally. The activation energy E and the QA value were found. The relationship between critical ambient temperature and the storage size was determined as well as the relationship between CAT and time to maximum thermal runaway for the materials. Frank-Kamenetskii; crossing point method; spontaneous ignition, self-heating; wood pellets, protein powders SCOPEThe purpose of the investigation was to find the behaviour of different materials with regard to igniting spontaneously and to investigate the usefulness of two general known methods i.e. the Frank-Kamenetskii (FK) method [1-3] and the crossing point (CP) method [4][5][6]. The underlying results are described in six reports issued by the Danish Institute of Fire and Security Technology (DBI) [7][8][9][10][11][12]. In three reports the results of the small-scale tests were compared to observations Hamlet Protein A/S, a producer of protein powders for the animal food industry, did during an incident in their drying process filter. In three reports the results of small-scale test were carried out for wood pellets.

Investigation of wood pellets self‐heating kinetics and thermal runaway phenomena using lab scale experiments

“…After tested at different oven temperatures, reactor sizes, and pellet samples, the mean activation energy of 78.7 ±0.8 kJ/mol and a mean ∆ r hA of (4.2 ±2.5) × 10 6 kJ/(kg s) are obtained. The activation energy from this study is consistent with the reported values of 76–93 kJ/mol at 100–200 °C and 60–80 kJ/mol at 50–80 °C for the Swedish wood pellets.…”

“…With a limited oxygen supply, the cellulosic materials underwent pyrolysis that produced active char and a large amount of heat, instead of the gas oxidation combustion, commonly known as burning or flaming. It should be noted that the smoldering phenomenon can easily develop to a rapid flame spreading combustion once the air flow increases or other conditions change that might increase the oxidation rate . As we observed, the pellets started smoking after they were exposed to air for a couple of minutes and could easily develop to flame combustion.…”

“…Therefore, the CP method by itself cannot be used as a tool to predict the critical temperature for self-ignition. [28]…”

“…It should be noted that the smoldering phenomenon can easily develop to a rapid flame spreading combustion once the air flow increases or other conditions change that might increase the oxidation rate. [28,29] As we observed, the pellets started smoking after they were exposed to air for a couple of minutes and could easily develop to flame combustion.…”

“…Theoretically, the CP method does not require as many tests as the FK method, however, the no‐conduction assumption will cause a systematic error. Therefore, the CP method by itself cannot be used as a tool to predict the critical temperature for self‐ignition …”