A scheme for the classification of explosions in the chemical process industry.

All process industry accidents fall under three broad categories-fire, explosion, and toxic release. Of these fire is the most common, followed by explosions. Within these broad categories occur a large number of sub-categories, each depicting a specific sub-type of a fire/explosion/toxic release. But whereas clear and self-consistent sub-classifications exist for fires and toxic releases, the situation is not as clear vis a vis explosions. In this paper the inconsistencies and/or shortcomings associated with the classification of different types of explosions, which are seen even in otherwise highly authentic and useful reference books on process safety, are reviewed. In its context a new classification is attempted which may, hopefully, provide a frame-of-reference for the future.

Explosive decomposition of ethylene oxide at elevated condition: effect of ignition energy, nitrogen dilution, and turbulence.

Experimental and theoretical investigation of explosive decomposition of ethylene oxide (EO) at fixed initial experimental parameters (T=100 degrees C, P=4 bar) in a 20-l sphere was conducted. Safety-related parameters, namely the maximum explosion pressure, the maximum rate of pressure rise, and the Kd values, were experimentally determined for pure ethylene oxide and ethylene oxide diluted with nitrogen. The influence of the ignition energy on the explosion parameters was also studied. All these dependencies are quantified in empirical formulas. Additionally, the effect of turbulence on explosive decomposition of ethylene oxide was investigated. In contrast to previous studies, it is found that turbulence significantly influences the explosion severity parameters, mostly the rate of pressure rise. Thermodynamic models are used to calculate the maximum explosion pressure of pure and of nitrogen-diluted ethylene oxide, at different initial temperatures. Soot formation was experimentally observed. Relation between the amounts of soot formed and the explosion pressure was experimentally observed and was calculated.

The relation of cool flames and auto-ignition phenomena to process safety at elevated pressure and temperature.

The cool-flame phenomenon can occur in fuel-oxygen (air) mixtures within the flammable range and outside the flammable range, at fuel-rich compositions, at temperatures below the auto-ignition temperature (AIT). It is caused by chemical reactions occurring spontaneously at relatively low temperatures and is favoured by elevated pressure. The hazards that cool flames generate are described. These vary from spoiling a product specification through contamination and explosive decomposition of condensed peroxides to the appearance of unexpected normal (hot) flame (two-stage ignition).

How is it possible? Why didn't we do anything? A case history!

Two similar serious accidents occurred at a metal refining process installation within a 6-month time interval. The first one killed ten people, and the second accident one person. The accidents provide a typical case history of how a safety management system and a corresponding organisation could have prevented the occurrence of such an accident or, at least, have reduced its effects. The case is also interesting because it illustrates the physico-chemical complexities forming the root cause of the accidents.

Risk informed resource allocation policy: safety can save costs.

During economic doldrums, decision making on investments for safety is even more difficult than it already is when funds are abundant. This paper attempts to offer some guidance. After stating the present challenge to prevention of losses in the process industries, the systematic approach of quantified risk assessment is briefly reviewed and improvements in the methodology are mentioned. In addition, attention is given to the use of a risk matrix to survey a plant and to derive a plan of action. Subsequently, the reduction of risk is reviewed. Measures for prevention, protection, and mitigation are discussed. The organization of safety has become at least as important as technical safety of equipment and standards. It is reflected in the introduction of a safety management system. Furthermore, the design process in a pro-active approach is described and the concept of inherent safety is briefly addressed. The concept of Layer of Protection Analysis is explained and also the reason why it is relevant to provide a cost-benefit analysis. Finally, after comments regarding the cost of accidents, the basics of costing and profitability are summarized and a way is suggested to apply this approach to risk-reducing measures. An example is provided on how a selection can be made from a number of alternatives.