Revisiting the Conditions and Consequences of CO2 Tank Explosions

Carbon dioxide is broadly used in a number of applications including food freezing, beverage production, enhanced oil recovery, supercritical extraction and welding.  It is commonly stored as a pressurized liquid (e.g.: -18 °C/0 °F and 2 Mpa/300 psia) in carbon steel tanks. 

A number of CO₂ tank explosions have been reported in the literature over the past 50 years.  In order to appropriately mitigate the risk associated with a liquid CO₂ storage tank, it is important to understand the potential consequences of a tank rupture.  Should the energy release be modeled as a PV blast or a Boiling Liquid Expanding Vapor Explosion (BLEVE)?  The question arises because of a common definition of a BLEVE as an event resulting from the sudden loss of containment of a liquid that is at or above its atmospheric superheat limit temperature.  Under the conditions listed above, CO₂ is above its atmospheric boiling point but below its superheat limit.  In this temperature range, will the vaporization rate be rapid enough to contribute to a blast overpressure in the event of a tank failure?

To answer this question, the authors made a critical review of information available in the literature including laboratory data and incident reports.  This paper will describe the analysis conducted and the results obtained.  It is concluded that the superheat limit is irrelevant in the determination of the force with which a ruptured CO₂ tank will explode.  The energy released by a CO2 tank rupture should be modeled as a BLEVE.