Toxic combustion products from pesticide fires. Executive summary

In order to obtain reliable data on the generation of toxic combustion products and to get more insight into the risks of fires in pesticide warehouses TNO performed the research project 'Toxic combustion products from pesticide fires'. The following research activities have been performed during the project:
— A study of the casuistry of pesticide fires;
— A literature study of small scale combustion test methods;
— Execution of medium scale combustion experiments with pesticides;
— A study on the extrapolation of the emissions of small and medium scale combustion tests to real fire conditions;
— A study on the prevention and control of fires in pesticide storages. Most important results and conclusions of the project.
— The medium scale combustion experiments with parathion, chlorfenvinphos, dichlobenil and 2,4-D indicate the following:
— Emissions of solids (particles and soot), carbon monoxide and hydrocarbons are higher than assumed in present hazard assessment
studies;
- Emissions of nitrogen oxides and chlorinated dioxins and furans are lower than assumed in present hazard assessment studies;
- Chlorine and sulphur in the pesticide are nearly fully converted into hydrogen chloride and sulphur dioxide;
- The emissions of ammonia, chlorine and phosgene are below or just above detection limits;
- Large quantities of solid residues remain in all fires. The residues contain up to 10% unburnt pesticide;
- Simulated fire extinguishing or oxygen deficiency do not significantly influence the emissions during the medium scale experiments.
— Similarity requirements to extrapolate the emissions to pesticides other than those investigated have been determined. As these similarity requirements will not be completely fulfilled in many cases the uncertainties in the extrapolated emissions are large.
— The conservative conversion rates are proposed for a pesticide fire.
Fire protection systems
— A methodology has been developed to define the required fire protection level of a pesticide storage:
- The required fire protection level for a pesticide storage depends on the product fire hazard, the packaging fire hazard, and the environmental and health hazard of the stored materials;
- Four fire protection levels for a pesticide storage are distinguished in the methodology, with fire protection levels having different requirements. For level one only the basic minimum requirements like good facility lay out, ignition source control, good housekeeping and regular training of employees are necessary. The highest protection level is level four. For level four automatic fire alarms, sprinkler systems and a basin for extinguishing water are additional requirements.
— Application of this methodology is recommended to obtain a cost effective protection level for a pesticide storage taking into account the hazards of the stored pesticides.
Small scale test methods
— Based on the results of the study on the characteristics of small scale test methods it can be concluded that the applicability of small scale test methods to predict the emission of toxic products of a pesticide fire is very limited.
— Small scale methods should be further developed in order to improve the simulation of the fire conditions observed during the medium scale
experiments.
Case histories
— Due to the very general description of the case histories it was impossible to get reliable data on the causes, development and consequences of pesticide fires.