Flammable liquids – and therefore the majority of solvents – must not be left to stand unprotected in laboratories.
The only exceptions to this rule are containers for ordinary use of a nominal volume of no more than 1 litre. Further stipulations can be found in the guidelines for laboratories ("Working Safely in Laboratories") in connection with DIN EN 14470, DIN 12925 Part 1 and the Technical Regulations for Flammable Liquids (TRbF) 20 (Appendix J).
In this one case, the solvents were stored in one of the laboratory safety cabinets for flammable liquids, which had been successfully used in practice over many years. The stock included solvents in safety cans as well as dehydrated solvents kept in round-bottom flasks, which were stored with desiccants such as calcium hydride or potassium. After the cabinet had not been opened for around a week, a severe explosion occurred, forcing the doors including their bearings out of the cabinet. Large sections of the laboratory floor were devastated by the explosion and the subsequent fire or rendered unusable through reaction products from the fire.
It seems very unlikely that an ignition source entered the cabinet from outside. It can also be assumed that the cabinet itself did not function as an ignition source. After examining the incident, the following cause is probable: It is conceivable that an ignition took place in one of the flasks, for example because excess pressure caused a stopper to shoot out and subsequently, due to evaporation, the potassium or calcium hydride could have dried up. Local overheating could have caused the vapour-air mixture to ignite. A flask that had fallen off its cork ring may also have been responsible. A decomposition of peroxides appears unlikely in view of the harsh dehydration that took place previously, but is a possibility, particularly with cyclical peroxides that are very difficult to destroy, such as diisopropyl ether. This may have taken place if the liquid was shaken. This ignition would certainly have had far less serious consequences if an explosive mixture of vapour and air had not been able to form in the cabinet. Such a mixture is the only explanation for the enormous effect of the explosion, as a normal fire in the cabinet would not have forced it open. However, the formation of the explosive atmosphere in the cabinet would have been prevented if the cabinet had had tenfold air exchange through technical ventilation. A decision was made not to connect the cabinet to the exhaust air unit for cost reasons. Technical Regulations for Flammable Liquids (TRbF) 20 states under which circumstances ventilation is not required.
It can therefore be concluded that ignition sources, and also potential chemical sources of ignition, such as self-igniting substances, must not become active in such cabinets. If the flammable liquids are not in sealed, break-proof containers, technical ventilation of laboratory cabinets is recommended.
See: Sichere Chemiearbeit 3-1997, 32
You can reach the German-language version of the article with a single click on the flag-symbol.