What happened?:
I was repeating an ampoule flame sealing procedure that I had previously performed without incident. The procedure involved sealing a 50 mL ampoule with an O2/CH4 torch (necessary to get the thick glass hot enough) in air on a benchtop. The contents of the ampoule were always purified deuterium oxide (non-flammable). Upon sealing the ampoule I would take the hot end of waste glass and drop it into the glass waste. In this case I was sealing 4 ampoules and upon sealing the third one the hot glass ignited the glass waste container. The fire was quickly put out with a class ABC fire extinguisher. Luckily it remained confined to the glass waste container so no permanent damage was done to the lab. It is important to note that the glass waste containers (thick plastic bag contained in cardboard box) often contain some waste paper products in our laboratory which may have served as fuel to eventually generate a fire large enough to burn the outer cardboard box. In the future our laboratory's procedure will be to dispose of hot glass by first placing it into a metal tray for cooling before putting it into the glass waste. Additionally, paper products will no longer be allowed in the glass waste.