Yeah, I know, I bumped the thread...
First off, paul.skyrme is correct - the research was done at the BAM facility in Germany and was "sponsored" (if that's the right word) by Transport for London, HSE and a couple of others. The belief was that an acetylene cylinder required cooloing water sprayed onto it for 24 hours, after it had been involved in a fire.
The research was initiated after one particular year in London saw something like (happy to be corrected...) 42 incidents involving acetylene, which probably caused many hold ups in a city that busy.
As has been stated by Carl.Riva in the last post, acetylene is unstable (cylinders are filled with acetone, which keeps acetylene from spontaneously decomposing) and when it does split, it emits heat, which destabilises surrounding acetylene molecules causing them to also split. This leads to a chain reaction scenario, leading to pressure building up in the cylinder.
Also correctly stated, a fire man died (in Hampshire, IIRC, in 2008) while moving an acetylene cylinder that had been involved in a fire and which had this chain reaction happening inside it.
The BAM research concluded that 15 minutes of cooling was sufficient (leaving the cylinder in situ), but recommended 1 hour to be on the safe side, with 1 hour of monitoring.
Acetylene cylinders (and pretty well most cylinders with liquids inside) have a pressure relief valve on them somewhere because if all the liquid turned into a gas, you'd have a fair bit of pressure to contain.
Previously mentioned is the BCGA website, with their (at the time of press) free to download guidance notes and codes of practice, which give examplary advice on the use of gases.