Atmospheric Chemistry Knowledge Gaps for Accidental Releases of Chlorine from Railcars

Chlorine releases to the atmosphere due to accidents involving railcars, such as the incident in Graniteville, SC, on 6 January 2005, are unique and can be extremely hazardous to health, the environment, and man-made materials. To assist in better understanding the complex nature of large-scale releases of pressurized liquefied gases to the atmosphere, the 2010 Jack Rabbit I (JR I) field experiments at Dugway Proving Ground, Utah, included releases of 1 or 2 tons of pressurized liquefied chlorine or anhydrous ammonia. Interestingly, during JR I, chlorine concentrations were observed to exceed 1 % at distances out to 100 to 200 m, and a dense aerosol cloud was observed at distances out to 50 to 100 m. So planning is underway for a campaign of field experiments in 2015 and 2016 called Jack Rabbit II (JR II), where much larger quantities (up to 20 tons) of pressurized liquefied chlorine will be released, which is consistent with the likely masses released during railcar accidents. This paper describes the status of the JR II planning, with emphasis on atmospheric chemistry knowledge gaps (challenges) related to chlorine releases.

In an accident, the pressurized liquefied chlorine is released through a relatively large hole (perhaps about 10 cm), where 20 tons can be released in a few minutes or less. The flashing process at the source aperture produces a very dense chlorine cloud that is initially about 20 % gas and 80 % small (median diameter 20 to 30 μm) aerosol drops, with a temperature near the boiling point of chlorine (- 34 C). The first challenge is to identify samplers (in situ or remote) that can accurately measure the chlorine concentration in a two phase cloud at very high concentrations (exceeding 1 %). The second challenge is to measure the aerosol chemical composition and size distribution; there is uncertainty concerning how fast the chlorine evaporates and how the chlorine gas and aerosols interact with ambient water vapor and drops that have condensed in the very cold (-34 C) cloud. The third challenge is to measure the chemical conversion of chlorine to reaction products in the near field (where there is an aerosol cloud) and farther downwind (past a few hundred meters) where the cloud is all gas and is invisible. Since chlorine is very reactive with water and organic matter on surfaces, the fourth challenge is to measure removal of the chlorine at the ground surface and by vegetation. Of course in all cases, care is needed to avoid damage to the sampling instruments by the corrosive chlorine cloud.