BLEVE - Effect modeling methods

Gas liquefaction is commonly used in the oil and chemical industries (but not only). This process enables larger quantities of a product to be stored or transported, given the same volume. Liquefaction involves lowering the temperature of the gas below its boiling point (liquefaction by cooling) and/or increasing its pressure to above its saturation pressure (liquefaction by compression). Liquefaction of natural gas, for example, is generally achieved by lowering the temperature, whereas for products such as butane or propane, the pressurized storage solution is generally favored. Pressurized storage is relatively simple to implement, and enables the liquid to be stored in dedicated spheres or cylindrical tanks, as well as transported by railcar, tanker or ship.

One of the major drawbacks of this type of storage is the risk of BLEVE (Boiling Liquid Expanding Vapour Explosion). BLEVE is one of the most serious phenomena to be encountered in an accident situation. It results from a specific loss of tank containment. Its occurrence can potentially generate mechanical effects (overpressure), projections affecting the environment over several hundred meters and, depending on the product stored, thermal effects (fireball radiation) and toxic effects. Due to its characteristics, it is also a source of domino effects. The destructive potential of a BLEVE is clear from a simple accident analysis. The 80 major BLEVEs that occurred between 1940 and 2005 claimed over 1,000 victims, injured over 10,000 and caused billions of euros in damage. It should also be noted that the vast majority of BLEVEs occurred during the transport phase, which means that the risk is not limited to the immediate vicinity of storage or production sites.

From an industrial risk management perspective, it is therefore essential to be able to characterize the effects of such a phenomenon. However, even though it has been the subject of much research and publication, our understanding of the various mechanisms involved in a BLEVE is still insufficient to propose a definitive and unambiguous method for modeling these effects. The problem is all the more complex in that the effects generated are of different natures (mechanical, thermal, etc.) and call for specific modeling methods. The aim of this article is therefore to shed some light on the BLEVE phenomenon and, above all, to review the state of the art of the main modeling approaches currently used in industry, approaches which are very often based on empirical observations or accident analyses.