The dispersion of hazardous gases over flat terrain of uniform roughness in terms of time-averaged concentrations is generally well understood and predictive techniques for most situations are sufficiently accurate for risk and hazard assessment purposes.
However, certain hazardous substances when released from pressurised containment generate a two-phase (liquid and gas) flashing release. Part of the liquid may 'rain out' and form an evaporating pool on the ground. The remaining liquid is in the form of an aerosol which remains airborne and may persist within the dispersing cloud for some time. The aerosol droplets will gradualy evaporate, but may also react with water vapour entrained into the cloud. The thermodynamics of such interactions are complex and currently there are few data to assist in the development and validation of dispersion models which model these aerosol and humidity effects. Indeed one of the most ubiquitous hazardous materials, ammonia, when released from pressurised containment exhibits all the above characteristics. (Ammonia is toxic to the extent that concentrations of a few hundred ppm cause irritation of the eyes and throat, and concentrations of a few thousand ppm are lethal after a few minutes' exposure).
The storage of ammonia in pressurised containment is expected to increase in the future due to its use in the denitrification of power station emissions. The increase in its use will add to public and governmental concern about the effects on members of the public and workers of an uncontrolled release of ammonia from pressurised containment. Such concerns in the USA, led the Lawrence Livermore National Laboratory (LLNL) to conduct a series of large-scale field trials in 1983/4. These trials were conducted in low humidity desert conditions in open flat-terrain, and only a limited amount of date was collected.
The main thrust of the current proposal is therefore to extend the data base for ammonia and to provide information that will assist mathematical and physical modellers to enhance current models and validate them. Attention will be paid to aerosol/humidity interactions, obstacle effects and concentration fluctuations. Main emphasis will be given to the description of the transition of the heavy gas dispersion part to the passive dispersion part down to toxidity levels. Such data will also be relevant to other highly toxic substances which exhibit aerosol formation and interaction with water vapour such as anhydrous hydrofluoric acid(HF) and hydrochloric acid (HCl) mist. The project has been formulated to assist the harmonisation of the implication of the Seveso Directive within EC member states. The project will be carried out in collaboration with JRC-Ispra.
Funding SchemeCSC - Cost-sharing contracts
WA3 4NE Warrington
WD2 7JR Watford
S3 7HQ Sheffield
172 01 Sundbyberg
CB2 1PZ Cambridge