Summary of Meeting Paper

The 1996 Annual Meeting of the Society for Risk Analysis-Europe

Fires in Chemical Plants - How to Reduce the Risks. P. van Donkelaar, Greentech Research, B 6887 Herbeumont, Belgium

INTRODUCTION

Accidents in the chemical industry, such as those that took place in Seveso (1976) and Bhopal (1984) can kill or injure thousands of people and many, such as the Sandoz fire in Schweitzerhalle-CH, can also result in considerable environmental damage (1), mostly caused by substantial amounts of contaminated fire fighting water, not adequately contained.

CAUSES

We studied eight fires in chemical plants in detail (2) and found that most of the fires were either caused by unusual circumstances or by very unlikely combinations of materials. As the chemical industry grows in size and complexity, the probability of such unpredictable circumstances occurring is also likely to grow.

A recent EU Report (3) argues that 95% of the fires could have been prevented by 'the application of existing knowledge, proper management and operational procedures'. Our study indicates, however, that it is much easier to apply this 'existing knowledge' to explain a fire after it has taken place (as in the EU Report), than it is to have it readily available in the minds of the people, responsible for taking the crucial decisions in the split-seconds available when a real fire breaks out.

This explains why, despite strong efforts by industry and government, fires in chemical plants do occur from time to time. It is this reality we have to deal with.

FIRE FIGHTING

In all of the eight fires studied, fire fighters were on the scene in a matter of minutes, but in two of the fires the supply of fire fighting water was a problem as it could only be found at distant sources. Precious time was lost. We therefore think that for chemical plants, a reliable supply of fire fighting water should be made mandatory.

When available, fire fighters tend to use water in abundance. Usage ranged from 1,000 m3 to as much as 20,000 m3. in nearly ail cases, these large quantities of fire fighting water, often polluted with high concentrations of various toxic chemicals, ended up in the environment, causing extensive and, in some cases, catastrophic damage (4).

CONSEQUENCES

On top of the direct fire damage, this environmental damage led to legal proceedings, heavy fines, high remedial costs and the closure or expensive improvements of facilities, sometimes worldwide. In addition to these financial and material consequences, there is often also severe damage to the public trust in the company involved. As Dr. R. Schweitzer of Sandoz put it:

....the material damage was outweighted by the loss of confidence that was suffered in the region and abroad. The accident caused widespread feelings of insecurity in large sections of the population and their confidence can only be regained by verifiable deeds, including concrete measures to prevent recurrence of such disastrous events'.

LEGISLATION

Following the Seweso disaster, the EU issued a Directive on the control of major accident hazards involving dangerous substances (COMAH), usually called the 'Seveso-Directive'. According to the 'subsidiarity'-principle, each Member State is required to implement the basic objectives of the Seveso Directive in its own legislation. Several Member States have, indeed, done so and a listing can be found in (2).

EMERGENCY SPILL BASINS

As we found in our study of eight fires, by far the most important source of environmental damage was contaminated fire fighting water entering the environment in very large quantities. In only one case, all of the fire fighting water could be contained in an emergency spill basin and, in that case, no environmental damage occurred.

Emergency spill basins are now required in many countries but there is not much uniformity in the required sizes. This is a direct (and undesirable) consequence of the EU subsidiarity principle. For three hypothetical cases, the table below compares the requirements of Germany, France, the Netherlands and Switzerland:

required size of emergency spill basins in m3

case 1 2 3
Germany 140 300 1480
France 1250 1500 3000
Netherlands 260 215 1460
Switzerland 400 250 2000

We see that the Dutch and German sizes are about equal, both tend to be lower than the Swiss and much lower than the French sizes.

The largest basin is 3,000 m3, which is much smaller than the largest actual amount of fire fighting water of 20,000 m3 we found in our study.

CONSTRUCTION

Existing legislation usually specifies constructions such as cellars under or just outside storage areas for chemicals. These are structures quite similar to swimming pools with a relatively small depth and a large surface. Such constructions demand a large surface space, which is often not available at existing sites. An even more important disadvantage occurs when burning chemicals are floating on fire fighting water. In such cases, the fire would extend itself over the entire surface of the basin, increasing the risk for damage. Even if the chemicals are not burning, evaporation over a large surface area could lead to a higher risk of explosions.

For all of these disadvantages there is an elegant solution, patented by the Dutch firm FUNDEX (5). These patents cover a 'CALEX' emergency spill basin with a relatively small diameter (up to 9 m) and large depth (up to 80 m), which can be 'drilled' into the ground with special machinery, using a number of 6-step sequences as described in Figure 1.

This construction is vibration-free and the shafts can be placed very close to existing buildings or production facilities, without unduly disturbing the normal operation of the manufacturing plant. Additional interesting features of this construction are that it can be used as a foundation element and that the basin is earthquake-resistant. A patented C02 supply extinguishes and cools any burning substances and ceramic granules floating on top of the chemicals minimize their evaporation. it also allows for easy separation of chemicals of different specific weights which helps proper disposal after the fire. This is in contrast to swimming pool type basins, where separation is much more difficult.

If the special drilling equipment is there anyhow, several shafts can be drilled into the ground to form a 'cluster'. Some of these could be filled with fire fighting water to ensure sufficient supply. As the basins are earthquake-resistant, this particularly applies to fires caused by this type of disaster.

CONCLUSION

With CALEX emergency spill basins of sufficient size, the risk for environmental damage by contaminated fire fighting water can be completely eliminated and the extend of the total damage can be considerably reduced by the ready availability of sufficient quantities of fire fighting water to quell the fire in its initial stages. The high costs of direct and indirect fire damage can thus be considerably reduced, leading to more favorable insurance premiums.

LITERATURE

  1. Firpo de Souza Porto, M. and Machado de Freitas, C., Major Chemical Accidents in Industrializing Countries: The SocioPolitical Amplification Risk, Risk Analysis, Vol. 16, No. 1, 1996.

  2. Donkelaar, P. Van, Environmental impact of firefighting--Protecting waterways from run-off pollution, Fire International, Winter 1995.

  3. EUR 13885, Lessons learned from accidents notified, Community Centre on Industrial Risk, Major Accident Reporting System, ENDS Report 197, June 1991.

  4. HSE, The fire at ALLIED COLLOIDS Ltd, a report of HSE's investigation into the fire at Allied Colloids Ltd, Low Moor, Bradford on July 21, 1992, HSE Books 1993.

  5. FUNDEX, Postbus 55, NL 4500 AB OOSTBURG, tel: + 31 117 457575, fax: + 31 117 457550.