Why do tealights burn violently when nearly burnt out

Question

I'm not entirely sure if this is the right place for this question, but I decided that it's better than physics.se (and I had no other ideas). I hope this is a good match, otherwise please move the question where it would belong. :)

I often use a mildly scented tealight in my apartment to cover the lingering smell of food. I usually light them using a matchstick, and sometimes I would toss the matchsticks into the tealight after lighting it up.

Usually, the matchstick would burn out after a short while, and for most of the life of the tealight nothing interesting happens. However, when the tealight is very close to burning out with the remains of the matchstick inside, it starts burning very violently, with the flame forming fantastic shapes (it looks like the whole remaining volume of the wax is burning).

Once the flame was so violent it broke the heat resistant glass container I kept it in (I was careless and I let the side of the cup touch a thinner part of the glass, rather than just the bottom).

Why does this happen?

Answer 1

There are two factors that probably matter here. One is that, for most of the life of the tea light, only a small amount of wax is liquid and actually burning. Most of the flame is liquid wax burning in the wick with a small pool of liquid wax below the wick. But later, when most of the solid wax has gone all the wax will be liquid. The second is that the matchstick can act as a second wick.

When the wick is still vertical (and held there by the presence of solid wax) the area of melting is kept small as the flame is vertical and the area of melting is constrained. Later, the wick may fall over as there isn't enough solid wax left. Also the heat can spread wider as the volume of wax decreases as it is also conducted more effectively through the aluminium container (wax isn't a good conductor of heat). The net result is more liquid wax near the end of the candles life.

Then two other factors may intervene. If the wick falls over, there is now a larger area of wick where burning can happen and the whole wick (not just the tip) can burn. Sometimes the matchstick will now also act as a wick, increasing the area of flame. Both these effects make more flame and more heat in a sort of runaway reaction which might also cause enough heat to reach the flash point of the wax leading to an even bigger flame. Even if the flash point isn't reached, the flame from a bigger wick area and a match acting as a wick can be a lot bigger than the flame form a normal vertical wick.

Answer 2

When only about 3 mm of wax is left, the wax in the tea light becomes so hot that its flash point is reached, and it begins to burn independently of the wig as a yellow/orange flame, just as is well known to happen in a more bluish version with alcohol at room temperature. This effect has principally nothing to do with the matchstick, which has been thrown in (as mentioned in the question).

Given that the tea light is already burning independently of the wig, if one would now extinguish the flame by covering the vessel for a second, one could see that an extremely strong cloud of white fumes evaporates (which should not be inhaled!): this is inflammable gaseous wax, which may be torched again even if one removes the wick beforehand with tweezers.

In the case of a still full or half-full wax-vessel, it is not possible that the temperature of the wax raises to the flash point because the flame is then burning essentially above or on level with the rim of the vessel, so that the vessel and the wax are heated up much less. Additionally, when only few wax is left, this heating up is much more effective because the heat may then focus on less wax. (The heating effect is stronger if the wax is in an aluminium vessel, because metal reflects the heat back into the wax; however, the stronger effect is helpful to elucidate the real causal coherences.)