INVESTIGATION OF AN EXPLOSION AND FLASH FIRE IN A FIXED-BED REACTOR

Citation
Ra. Ogle et Jl. Schumacher, INVESTIGATION OF AN EXPLOSION AND FLASH FIRE IN A FIXED-BED REACTOR, Process safety progress, 17(2), 1998, pp. 127-133
Citations number
3
Categorie Soggetti
Engineering, Chemical
Journal title
ISSN journal
10668527
Volume
17
Issue
2
Year of publication
1998
Pages
127 - 133
Database
ISI
SICI code
1066-8527(1998)17:2<127:IOAEAF>2.0.ZU;2-A
Abstract
An explosion and flash fire in a fixed bed reactor occurred at a munic ipal wastewater treatment plant (WWTP). Two employees were injured in the accident. The accident occurred in an ozone treatment building whe re ozone was used to treat odors from the offgas of the sludge concent ration units. Excess ozone manually was routed to the fixed bed reacto r (ozone destruct unit) where the ozone is catalytically transformed i nto oxy gen before being discharged to the atmosphere. An investigatio n of the accident was conducted to determine the root cause of the exp losion and flash fire and identify corrective actions which the WWTP m anagement could undertake to prevent a recurrence, This investigation included site inspections, interview with the injured employees, sampl ing and analysis of various materials, an explosion dynamics analysis, and a root cause analysis. It was concluded that cooling oil from one of the ozone generation units entered the main ozone gas line due to a crack in one of the reactor's dielectric tubes. The cooling oil was vented into the ozone destruct unit when an employee opened a ball val ve on the main ozone gas line. The cooling oil, essentially a saturate d hydrocarbon mixture, reacted exothermically when it contacted the ma nganese dioxide catalyst. The exothermic reaction resulted in an explo sion which propelled the access panel outwards and dispersed the catal yst pellets. A flash fire followed the explosion. The flash fire burne d two employees and caused thermal damage to a nearby control panel. A lthough this accident was the first of its kind at this facility, this was not the first time that the ozone generator had experienced a fai lure of a dielectric tube. Thus, there was a significant probability t hat a dielectric tube failure could leak cooling oil into the main ozo ne gas line. This failure event could, in turn, result in another expl osion and flash fire. The WWTP staff neither designed nor fabricated t he ozone generator-destructor system. therefore, it did not seem appro priate for the WWTP staff to modify the ozone system. Instead, it was recommended that the ozone destruct unit be taken out of service. The WWTP management acted on this recommendation.