We would like to report an explosion that occurred in our laboratory last year while performing an oxidative coupling of trimethylsilylacetylene (TMSA) in a Glaser-Hay reaction. The explosion ruptured the 2-L reaction flask and seriously injured a researcher.
This reaction has been routinely used in our and many other laboratories to prepare 1,4-bis(trimethylsilyl)butadiyne-1,3 on a large scale (>100 g), and no dangerous or unusual behavior was previously noted.
The procedure involves purging oxygen through a solution of TMSA in acetone in the presence of a copper(I) chloride:tetramethylethylenediamine complex catalyst at room temperature as described by Andrew B. Holmes et al. (Org. Syntheses 1993, Coll. Vol. 8, 63). The authors of the procedure recommend a safety shield as a general precaution while working with flammable materials in the atmosphere of oxygen, although no hazard was ever encountered. In this incident, the explosion occurred as soon as we started adding the solution of catalyst in acetone to the reaction.
We have consulted with the pioneer of this reaction (Allan S. Hay) and the submitting author of the procedure (Holmes) and considered various scenarios to explain the explosion. Ignition of acetone/TMSA vapor by the external sources was hardly possible as the flask was well sealed and the outgoing gases were passed through a dry-ice condenser (lowering the vapor pressure below the explosive concentration) and brought to the back side of the fume hood through a 1-meter hose.
The reaction temperature (5 ºC) was noted by the researcher a few seconds before the explosion, thus ruling out unexpected reaction exothermy. The autoignition of the vapor on a hot stirring adapter (possibly heated by rotation-induced friction) was refuted, because joint lubrication was checked before setting up the experiment, and it would have required achieving an unrealistic temperature of greater than 300 ºC. Also, the explosion occurred upon adding the first few drops of copper catalyst, which makes crystallization of the explosive intermediate—copper bis(trimethylsilylacetylide)—highly improbable.
We speculate that a discharge of static electricity between the syringe needle and the digital thermometer inside the flask is the most likely cause of this explosion. A digital thermometer connected to a stirring hot plate (IKA) was used in the reaction, and a plastic syringe with a long metal needle was introduced through the same neck. An induced static voltage on the syringe through friction from handling (often observed in Montreal winter indoors while walking or even simply sitting) could then cause a sufficient differential potential on the needle for a discharge spark to occur close to the metallic body of the digital thermometer. The oxygen-rich atmosphere lowers the ignition energy and makes even a weak spark sufficient to cause a fire.
The incident emphasizes once more the potential danger of mixing oxygen gas with flammable solvents or reagents. More important, introducing two conductors into a flask brings a risk of static electricity discharge between the conductors, which is dangerous whenever a flammable solvent is used without inert gas. As wired metal-gauge digital thermometers are used more often in synthetic practice, precautions must be taken to avoid their contact with other metallic (conducting) parts inside the reaction flasks.
Dmitrii F. Perepichka