Mechanism of combustion of hydrocarbons

Question

While this question comes close, it doesn't quite answer my question.

Adding heat to oxygen & a hydrocarbon breaks the atoms apart and they reform into carbon dioxide and water. I get that.

I want to know specifically at what point the water and carbon dioxide are formed?

As I understand it the heat vibrates the atoms, breaking bonds. Those atoms reform with oxygen forming stronger molecules ($\ce{H2O}$ and $\ce{CO2}$) which can withstand the combustion heat. I reached this far with help from a Reddit thread.

Where that thread seemed to disagree is in my main question, on where/when the formation actually happens, it also raised another question of how long atoms are broken apart for.

In this article, it would suggest that free atoms are created which find oxygen:

$\ce{CH4 + 2O2 → CO2 + 2H2O + energy}$

[...] the arrow represents the moment the reaction happens. But there are a lot of interesting things happening that are hidden behind that arrow. A more detailed equation would look something like this:

$\ce{CH4 + 2O2 + a little energy →C + 4H + 4O →CO2 + 2H2O + lots of energy}$

The first arrow represents the breaking of the bonds, which requires energy. On the middle line are the atoms, now broken out of molecules and free to react. [...]

Which at least one of the Reddit thread contributors disagreed with.

So! (TL;DR)

• When are the new molecules formed (cylindar, afterwards or both)?
• How long are atoms "free", how long do they take to form new molecules, is it instantaneous or ... what?

Answer 1

Combustion reactions are complicated even when they result in simple products

Part of the problem with answering the question "at what point are water and carbon dioxide formed" is that the reactions occuring in combustion are very complicated and not at all well represented by the simple versions shown in the question.

We can say for sure that the picture where all the components are atomised and then recombine is not the way the reaction works. This simplistic idea is part of the reason for disagreement on the reddit thread.

It is an incorrect view for many reasons but mostly because the reaction doesn't all happen at once. Take petrol (that's gasoline for americans) engines. The initiation of combustion is started by a spark and then (relatively slowly in chemical terms) progresses through the rest of the fuel/air mixture (if it were very fast the engine would explode rather than smoothly driving the cylinder, if it were only a little too fast the engine would "knock", damaging the cylinder). So we need to think–even before considering the detailed chemistry–about the (slow) steps involved in the reaction.

The major steps involved in most combustion reactions involve radical chains. Though the full detail is not understood (though this answer gives some more detail), it is thought that the initiating step involves a hydrocarbon reacting with an oxygen molecule to give a hydroperoxide radical and a hydrocarbon radical. Very simplistically this looks like:

$\ce{O2 + CH3(CH2)nCH2-H -> HO2. +CH3(CH2)nCH2.} $

But neither of those radicals is stable or long-lived. In an environment flooded with more hydrocarbons and oxygen, they can both react further to give more products and more radicals. For example, the hydroperoxide may also fall apart to hydroxide radicals which can react with parts of other hydrocarbon molecules to give water and yet more radicals. In the early stages, most reactions create more radicals; later in the process some radicals can combine into stable products (like water and carbon dioxide). Once the reaction is started it can be self-sustaining as the formation of those stable products release a lot of heat, enough to keep creating new radicals and new chains until the source hydrocarbons or the oxygen is depleted.

So, even thinking about the detailed, step by step, chemical reactions suggests that the process is messy. Even early on in the process some final products are being created, though this continues until the fuel/air is mostly used up. if you plotted the amount of final product over time you would see some very early and a rising amount until the reaction had finished. It is pretty meaningless to ask "at which point" the products are formed.

In short, the initiation does not "atomise" the components: it creates radicals (this doesn't need as much energy input as "atomisation"). They can continue to react creating chains of yet more radicals and some chains can stop when stable products are formed, even early in the reaction. The net reaction releases enough heat to sustain combustion while fuel remains. Final products start to form early in the process and continue to be formed until the air and fuel have gone.

Answer 2

The hydrocarbon and the oxygen begin to combine when the spark initiates and they continue to combine in a burning, and not homogeneously (not an explosion - except if the combustion begins too early because of knocking, which occurs because the hydrocarbon is too sensitive to the heat and pressure build-up from the compression in the cylinder). Typically, essentially all of the hydrocarbon is burned before the exhaust valves open up, unless the rpm is very high.

That's the beginning. Then, because some of the hydrocarbon escaped burning, or was transformed into other unburned material, and undesirable CO and NO$_x$ was formed, further reactions occur over the catalytic converter, and finally, CO$_2$ and H$_2$O come out the exhaust pipe.

"Specifically at what point" is a question that does not have a short answer. The actual combustion in one cycle is a time line, a chain reaction over which many different types of reaction occur. A common thought is that explosions occur in the cylinders of an internal combustion engine, but the goal is to produce a rapid, even burning over several milliseconds so as to develop even pressure gradients to push the pistons. Knocking is really an explosion; not a terribly fast one, but one that could eventually overstress the engine. Even that explosion isn't a point in time.

In a car engine, we would be talking about femtosecond reaction times (10$^{-15}$ seconds) for individual molecules or perhaps less for specific reactions like ionization. The time and distance scale of reaction for an individual molecule could be plotted down to less than a micron and a microsecond, but in any real reaction, there is so much material involved that the "point" is totally spread out in space and time. The overall combustion reaction from beginning to end is a bit more than the few milliseconds in the cylinder because of the catalytic converter, but for most purposes, you could consider the reaction to take the whole time between the spark and the exhaust.

There is an extensive discussion of combustion on Wikipedia: https://en.wikipedia.org/wiki/Combustion

https://corels.ibs.re.kr/html/corels_en/research/research_0305.html