tl;dr There are two types of combustion that lead to different rates of burning. The reaction rates are completely different, due to the material being burned, temperature and oxygen levels, and biochemical structural composition.
This article is dedicated to understanding the science behind the following video. Watch:
We note that all the match tips burn out within the first 90 seconds, as the fire propagates quickly across the assembled pattern. So what happens in the next 14 minutes? Didn’t we already see all the matches burn out in the first phase? How are they continuing to burn?
We first breakdown the burning of the match into three categories: ignition of the match head, burning of the match stick, and burning of the match charcoal. Phase I takes roughly 33 seconds (at 0:09 to 0:42), Phase II takes close to 50 seconds (0:42 to 1:32), and Phase III takes the longest time at roughly 12.75 minutes (1:32 to 14:19). We start with understanding the reaction in Phase I, ignition of the match.
Phases I and II: Burn notice
Now we transition to understand what a match is made out of. The match head consists of antimony trisulfide- an oxidant that vigorously initiates the burning, potassium chlorate to extend the duration of the flame, ammonium phosphate to prevent excessive smoke from burning, and paraffin wax along the match stick to allow the fire to travel along the length of the match. As a side note, the wax coating also prevents smoke by creating a hydrophobic layer to reduce water release. The diagram below describes this composition:
The key to the controlled burning of the match comes actually comes from the matchbox. This contains powdered glass, which acts as an abrasive to heat via friction, along with red phosphorus, which converts to white phosphorus when frictional heat is applied. The white phosphorus then spontaneously ignites the antimony trisulfide and potassium chlorate on the match head. One can see the slow motion ignition in the video below; the yellow bubbles forming are sulfur bubbles from the antimony trisulfide.
So everything above has explained how the match ignites. But we haven’t addressed how exactly the match continues to burn long after the match has died down. What happens next?
Phase III: How slow can you go?
As we enter phase III, the basic material has been ‘burnt’, but there’s still a lot of carbon content left to burn! To really understand what is going on here, we need to understand what we mean by ‘burning’. This is really just the addition of oxygen to some hydrocarbon to spontaneously release CO2 and H2O. The most basic of examples is the combustion of hydrogen, a reaction used in rocket engines:
2 H2 + O2 -> 2 H2O , ΔH = -141.9 kJ/g
This is a very high energy reaction- that’s why hydrogen fuel cells run a very high safety risk. In comparison, wood has a heat of combustion close to 15 kJ/g. But this is an irrelevant standard of comparison- what’s more important to understand is the difference in types of combustion.
In phases I and II, rapid combustion was observed. As the name suggests, this implies a quick release of a lot of energy- usually in the form of a fire, but in some cases with insufficient material, an explosion. We saw a flame form on the match heads, and then quickly progress down the match stick length.
In phase III, smouldering is observed, which is slower, incomplete combustion. Rapid combustion involves a quick surface burning of the gasses released by the pyrolysis of the match. Smouldering involves slower interior burning of the char residue of the ‘burnt’ wood at a lower temperature. Smouldering is roughly 10 times slower than rapid combustion. The conditions for smouldering arise in low oxygen, low temperature environments. We recall the below fire triangle to understand the appropriate balance required for any fire:
The key to smouldering: cellulose!
So now we understand that wood actually burns in a different way (lower temperature/oxygen environment) in Phase III. But what component of the match is the real cause behind this slow burning?
We first recognize that normal wood is mostly comprised of water, in fact in logs the water content is roughly 50%. This is not the case with matches, but it helps understand why logs in a campfire smoke heavily during burning. What’s more interesting is the ~40% composition of cellulose within the wood that played a critical role in the tree that the wood was chopped from. Cellulose is a polymer of glucose monomers like starch, but unlike starch, cellulose forms a stiff straight chain that is bound very strongly. Starch easily hydrolyzes into sucrose and fructose, but cellulolysis is extremely difficult in the absence of a strong ionic solvent or enzymes.
At high temperatures (> 350ºC), however, breaking down cellulose is possible. And this is exactly what happens in matches, which leads to the smouldering effect. However, this rate of cellulose breakdown is very slow, as you can see in this video below:
The key is that a higher temperature is required to activate the cellulose breakdown, which does in fact happen with the match. This is why Phases I and II are important before considering the smouldering in Phase III; as you can see in the below image, the same flame contains different temperature ranges due to the types of combustion taking place:
So the first two phases are important to raise the temperature high enough inside the match to continue the fire. This is similar to starting a campfire by first adding small tinder, then adding in larger kindling, and finally laying down the larger wood fuel. Even after the larger wood fuel is done showing visible flame, the char and ash still smoulder in a low oxygen environment long after. The same happens in the above match domino set; a virtual campfire has been created with the sheer quantity of matches incorporated, and the cumulative mass is enough to sustain a fire for 14 minutes.
And there you have it! A long winded explanation to understand why a match domino took so long to be put out. I’ll leave you to consider the following: the orientation and quantity of matches play a role in determining time of burning. Take a look at the following two links: the first is a hemisphere of 3000 matches that took 10 minutes to burn, while the second is a three layer tower of 7000 matches that took 45 minutes to burn. Clearly geometry of the matches matters; is there some optimal packing that extends the burn time of the system? This type of discussion will become more mathematical, and will be left to the interested reader/aspiring mathematician to pursue. I hope you enjoyed the article; kudos to you for making it this far!