What happened? A researcher was performing a Johnson-Claisen Rearrangement (see figure below) using 66 mL of triethyl orthoacetate (aka 1,1,1-triethoxyethane, CAS # 78-39-7) and propionic acid as a catalyst. The reaction was being performed in a 210 mL, 150 psi pressure-rated glass bottle at 180oC. The reaction was set up in a hood with a blast shield in place. The reaction was being heated using a silicon oil bath and a Variac transformer with alligator clips attached to a nichrome wire.
After about 15 hours, the reaction vessel exploded with a tremendous bang, shattering the blast shield, cracking the fume hood sash, and igniting the solvent in a fireball. The researcher was in the lab but away from the hood when it exploded. He used a fire extinguisher to put out the fire. He was shaken up but fortunately suffered no injuries other than temporary ringing in his ears.
 |
 |
 |
Why did it happen? No obvious flaws in the glass bottle were noted before the reaction was started. There is suspicion that generation of volatile by-products (possibly ethoxyacetylene, 1,1-diethoxyethylene, and ethanol) increased the internal pressure of the glass reaction bottle beyond the pressure rating. A pressure gauge that fit a similar bottle was later acquired. When 66 mL of triethyl orthoacetate alone was heated to 180oC (without the catalyst), the pressure in the bottle was only 60 psi. The ignition source for the solvent was likely the oil bath heating equipment.
How can a similar occurrence be avoided? Whenever possible, metal reactors with glass liners should be used instead of sealed glass bottles. However, it is sometimes convenient to run very small scale reactions at low pressures in small sealed glass tubes or in thick-walled pressure bottles. Some bottles are equipped with a head containing inlet and exhaust gas valves, a pressure gauge, and a pressure-relief valve.
For any such reaction, one should be fully prepared for the significant possibility that the sealed vessel will burst. Any reaction of this type should be carried out in a hood labeled with signs that indicate the contents of the reaction vessel and the explosion risk, and every precaution to prevent injury from flying glass, hazardous chemicals, or hot liquids should be taken by using suitable shielding (Note – the blast shield alone was not sufficient shielding for this explosion. The addition of the fume hood sash was necessary). Also, resistive heating using a nichrome wire coil (submerged in the oil) connected to an electrical plug using alligator clips should never be used. It is dangerous due to the potential for electrical shock, sparks igniting flammable vapors (as occurred in this incident), and fires due to overheating oil baths.
All contents copyright © 2008. Arizona Board of Regents.