Characterising the ground level concentrations of harmful organic and inorganic substances released during major industrial fires, and implications for human health

We report on the concentration ranges and combustion source-related emission profiles of organic and inorganic species released during 34 major industrial fires in the UK. These episodic events tend to be acute in nature and demand a rapid public health risk assessment to indicate the likely impact on exposed populations. The objective of this paper is to improve our understanding of the nature, composition and potential health impacts of emissions from major incident fires and so support the risk assessment process. Real world monitoring data was obtained from portable Fourier Transform Infrared (FTIR) monitoring (Gasmet DX-4030/40) carried out as part of the UK’s Air Quality in Major Incidents service. The measured substances include carbon monoxide, sulphur dioxide, nitrogen dioxide, ammonia, hydrogen chloride, hydrogen bromide, hydrogen fluoride, hydrogen cyanide, formaldehyde, 1,3-butadiene, benzene, toluene, xylenes, ethyl benzene, acrolein, phosgene, arsine, phosphine and methyl isocyanate. We evaluate the reported concentrations against Acute Exposure Guideline Values (AEGLs) and Emergency Response Planning Guidelines (ERPGs), as well as against UK, EU and WHO short-term ambient guideline values. Most exceedances of AEGL or ERPG guideline values were at levels likely only to cause discomfort to exposed populations (hydrogen cyanide, hydrogen chloride, hydrogen fluoride and formaldehyde), though for several substances the exceedances could have potentially given rise to more serious health effects (acrolein, phosphine, phosgene and methyl isocyanate). In the latter cases, the observed high concentrations are likely to be due to cross-interference from other substances that absorb in the mid-range of the infrared spectrum, particularly when the ground level plume is very concentrated.

Abstract licence: CC BY-NC

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.