Safe Storage of Combustible Solids

Whilst the flammable nature of gases and liquids have some definable characteristics, that is, flammability limits and flash points respectively, it is not so easy to classify the burning behaviour of solid materials. Solid materials have historically been identified by the UK Fire Service on the basis of involvement in past fires or data from a range of small scale tests. Guidance relating to the storage of flammable materials was issued to Fire Brigades in Fire Service Circular 59/1976. This circular contained details of hazardous materials and the quantities and conditions, under which they could be stored before requiring the need for a fire certificate. Shortcomings and omissions relating to solid materials have been highlighted by a number of incidents. The solid materials appearing in the above Circular and its Amendment were originally defined as "highly flammable" and "flammable". However, to initiate a scientific basis, the materials have been re-classified during this research programme by collating their physical and chemical properties and these were then used in conjunction with the UN Committee of Experts' "Recommendations on the Transport of Dangerous Goods" classification scheme. This permitted the revision of the lists previously forming the two categories of solid materials. A detailed survey covering a 10 year period, 1981 - 1990 analysed 290 warehouse and other storage area fires, involving solid materials only. The analysis showed that packaging materials are the most frequently involved goods in storage fires and arson represents the major cause of fire with known ignition sources. The analysis also provided information on the success, or otherwise, of active fire protection systems. Synthetic, polymeric materials were absent from the original classification scheme and a means of classifying such materials was sought by considering the fire hazard posed. The Health and Safety Executive had previously employed a scaled room-corridor fire test to classify such materials. Ignitability and rate of heat release were parameters chosen to enable the classification of the burning behaviour of solid materials. A cone calorimeter was constructed and the time to ignition, average rate of heat release and peak rate of heat release were determined for a range of materials. Criteria were selected to enable classification into "highly flammable" and "flammable" categories. The selection of criteria based on the peak rate of heat release provided a rank order of materials similar to that obtained in the scaled room-corridor fire test. Average rate of heat release data were used in a field model computer package, FLUENT to predict the temperature contours from stacks of the selected materials. The ceiling temperatures and rate of rise of temperature of the warehouse roof/ceiling provided an excellent guide to the effects caused by varying stack size, configuration, environment, and separation distance between neighbouring stacks. The combination of experimental data on rate of heat release and use of data from a computational’ fluid dynamics package provides a methodology for including materials into a classification scheme for the definition of their burning behaviour. The work reported provides a scientific basis for the revision of guidance given to Fire Brigades.