Estimating Health Effects of Accidental Releases of Toxic Chemicals such as Chlorine and Ammonia

When some toxic industrial chemicals are released to the air as a result of accidents or terrorist activities, they may become Toxic Inhalation Hazards (TIHs). Chlorine and anhydrous ammonia are relevant examples, as they are the two most widely manufactured and transported TIHs worldwide. Both are transported in road tankers and railcars as pressurized liquefied gasses, while anhydrous ammonia may also be transported and stored as a refrigerated liquid. The health effects of these chemicals do not follow the commonly-used Haber’s Law, which says that the health effect of a given dosage (concentration C integrated over time) is the same whether the exposure time is a day, an hour or a minute. Instead, the health effects of a given dosage are larger for smaller exposure times for these chemicals, and this effect is usually parameterized using a power law on the concentration (i.e., Cⁿ), where n is found to be greater than one for most TIHs. The widely accepted values of n are in the range of about 2 to 3 for ammonia and chlorine, although lately there has been work that suggests a lower value for chlorine. This formalism (the integral over time of Cⁿ) is known as the toxic load, which has been critical in improving the estimation of toxicity effects. There are a number of operational measures that have been developed based on the toxic load principle to describe lethal and non-lethal effects, including Acute Exposure Guideline Levels (AEGLs), Emergency Response Planning Guidelines (ERPGs), Temporary Exposure Limits (TEELs) and Protective Action Criteria (PACs). Toxic load formulas are also used to estimate Lethal Concentrations (LC_xs) where x percent of the population is expected to be affected.

Of course, there are very few TIH observations available to use to develop and test these human health effect relations. Data from past accidents and past wars are used, as well as data from other animals. But those data are limited, too. For example, although there are enough observations to develop the toxic load formula using laboratory experiments where an animal is exposed to a constant concentration over a range of exposure durations or averaging times (e.g., 10 min, 30 min, 60 min, 2 hr, 4 hr, and 8 hr), there are insufficient observations to test the formula for concentrations that are highly variable in time (as typically happens with accidental TIH chemical releases in the atmosphere). This paper reviews the widely used health effect parameterizations for chlorine and ammonia, as well as attempts to clarify the basic definitions of averaging time and exposure time and their use in defining averaged concentrations and dosages.