Explosion in a refinery
This accident took place in different phases and in two different units of a refinery:It consists first in a non-hydrogen accident, followed by a hydrogen-related accident. Before the hydrogen-related accident took place, a local fire occurred near the atmospheric vessel, containing a solution of NaOH in water used for neutralizing the acidity of the raw oil contained in the desalter. The fire caused damage to the tank and to the electrical cables running near the vessel. This fire was completely extinguished before the second accident. Half-hour later, an explosion occurred at the exit of the tube from the preheating oven of gasoil + hydrogen of the desulphurisation plant. This explosion damaged, besides the oven, various equipment in a radius of about 50 m from the explosion point. The fire following this explosion was of short duration. The incident occurred in a zone in which numerous personnel, either internal or external, were present for the modifications going on the near-by plants. Its consequences could have been more serious not only to structures, but also to people, if the "all clear" alarm was given immediately after the end of the first accident; in that case many persons would have returned to their job and would have been taken by surprise by the second accident. This conclusion is supported by the considerable damages received by the materials (crane with extensible arm, van for material transport, hut from sheets of metals, materials for changes of parts, etc. . . ) in a radius of 50 m from the explosion point.
Event Initiating System
Classification of the Physical Effects
Nature of the Consequences
Cause Comments
The INITIATING CAUSE of the first accident was the release of the hydrocarbons contained in a tank. Its root CAUSE was probably a human error, during the manual operation of the dilution of the sodium hydroxide solution (routine operation), protracted for an excessive time, leading to the overflow and emission of hydrocarbons in the canals and their flushing shafts collecting white waters. One of the consequence of the first accident was the interruption of the electrical supply which provoked the loos of gasoil supply loss in the oven in which the second accident occurred. The increase in the temperature and pressure inside the pipes in which gasoil + H2 should have circulated occurred due to this supply loss, with consequent overheating of the pipes resulting in rupture and explosion. Also in this second case, the root CAUSE was probably a human error consisting of the undue silencing of alarms due to insufficient supply and of high temperature and pressure and the subsequent omitted interventions as foreseen by the existing procedures. An alternative or contributing root CAUSE would be inadequate training on the procedures.
