Key Instruction Points
In November 1999, a graduate student conducted an experiment to purify by sublimation a colorant intermediate. An oil bath used for heating purposes caught fire. No physical injury resulted and minor damage to the ceiling tiles above the fume hood where the experiment was conducted was caused by smoke. The incident was immediately reported to EH&S. The following day, the Environmental Health and Safety department, the Chief Deputy Fire Marshall, and the faculty advisor to the student running the experiment reviewed the scene of the incident. Since then, corrective action was identified and has been implemented. This report outlines the procedure used for the sublimation experiment, provides a hypothesis of how the fire originated, and details the corrective action.
The equipment employed for the sublimation experiment consists of a hot plate onto which is placed an oil bath. In this particular experiment only a small amount of oil was placed in the bath. A sublimation apparatus was mounted so that only the bottom of the apparatus was submerged in the oil bath. The sublimation apparatus consists of two glass chambers of differing size. The smaller chamber is designed to be inserted into the larger chamber and sealed with a ground glass joint and grease. The pressure in the outer chamber is reduced during the experiment, and this chamber becomes hot when in contact with the oil. The inner chamber is the ‘cold finger’, or condenser, which is cooled by running a water stream through it. The oil employed for heating was purchased from Aldrich Chemical, and is designed specifically as a heating medium. Flash point of the oil is 310-325 degrees C, fire point is 360 degrees C, and melting point is 60 degrees C. The flammability properties of the oil used were appropriate for operation of a cold finger sublimation experiment at less than 150 degrees C.
A small amount (approximately 200 mg) of the compound to be sublimed was placed in the outer glass chamber of the sublimation apparatus. The glass joint between the head of the cold finger and the main chamber was lubricated. The cold finger unit was positioned so that the bottom of the outer chamber was slightly submerged under the surface of the pre-heated oil. Input/output hoses were connected to the cold finger head and water was passed through the hoses in a slow stream. A vacuum hose was connected to the outer chamber and the pressure reduced using in-house vacuum. The normal procedure results in vaporization of the test compound in the outer chamber and slowly condensation of the vapor onto the outer wall of the cold finger. The slow condensation promotes purification and slow crystal growth of the compound. This was the objective of the experiment. However, in this case, some time after heating of the cold finger began, a water hose became disconnected from the cold finger and water ran into the oil bath. A fire commenced immediately afterwards, and oil burned in the fume hood for several seconds. Use of the fire extinguisher was not attempted (students present were not trained in the use of the fire extinguisher).
The purpose of the experiment was to purify a compound by sublimation. Particularly high purity was required in the present compound, as it was to be employed in x-ray crystallographic analysis. While purification by sublimation is not a procedure commonly employed in our laboratory, the graduate student had conducted similar experiments in another University previously. The equipment is simple to operate and the procedure is straightforward, if performed correctly.
Examination of the equipment after the incident revealed that old and weak hoses were employed.