Every chemical has a decomposition temperature. My understanding is that above that temperature molecular bonds are broken. And if we raise the heat high enough then all molecular bonds will break and we will be left with a "plasma" of ionized atoms, right?

Can the decomposition temperature of bonds be predicted by models, and if so on what data do those models depend?

Now suppose we heat a chemical in an inert (noble) atmosphere, or in a vacuum, to the point of total decomposition. Now we cool the plasma back to STP. Are there models that can predict the products we would expect to find – i.e., the right hand side of the reaction equation?

I'm guessing that thermodynamics stipulates we should find the lowest-"energy" compounds that can be created from the constituent elements on the right hand side.

In practice I imagine that achieving that result could require that cooling be done at some "sufficiently low rate." So, to increase the complexity of the question: Are there any models that can predict what products will be found as the cooling rate is increased? I believe that in the limit (a virtually instantaneous quench) one ends up with a "glass" (or combination of glass, condensate, and gas) but even then one can't be left with free ions so can we predict the composition of the quenched plasma?

  • $\begingroup$ I'm afraid that asking sth like that without being specific is much too broad $\endgroup$
    – Mithoron
    Aug 11, 2015 at 21:33
  • $\begingroup$ Also edited tile - it was very vague $\endgroup$
    – Mithoron
    Aug 11, 2015 at 21:39
  • $\begingroup$ @Mithoron: I disagree that the title was vague, but we can try your suggestion for a bit. It is a general question. The answer may simply be, "No, there are no general models of decomposition or molecular reconstruction." If there are, then I expect them to be well known by theorists. Would it help if I provided an example of a left-hand-side and then a bunch of right-hand sides that "balance" without any reason to believe they will actually occur? $\endgroup$
    – feetwet
    Aug 11, 2015 at 21:44
  • $\begingroup$ Two factors that are going to make it difficult to predict the outcome of this type of experiment (except perhaps for very simple systems like hydrogen gas or chlorine plus bromine) are (1) the complex combinatorics lead to a huge number of intermediates and final products, many of whose energies/reactivity are not known; (2) kinetic barriers are harder to predict that ground state energies. As is often the case, you may have to go into the laboratory and do some experiments. Possible approach: aim the plasma at a supercooled probe in a vacuum and spectrally characterize the resulting matrix. $\endgroup$
    – iad22agp
    Aug 12, 2015 at 12:42
  • $\begingroup$ @iad22agp: That sounds like the summary of a good answer. Could you expand points 1 and 2 a little bit and post as answer? The whole question is whether a general theoretical solution is possible. Obviously one can decompose a particular chemical and analyze the condensate in a lab, but I was wondering if theoretical chemistry offers models of recombinetrics of decomposed molecules. $\endgroup$
    – feetwet
    Aug 12, 2015 at 14:23

1 Answer 1


To begin with, dissociating molecules to a plasma at the atomic level will not only create ions but also neutral atoms. These will indeed reassociate on cooling, and the question is how to predict what the products will be. In the extreme case of a simple system such as pure hydrogen gas (dissociating, then recombining to H2) or, say chlorine plus bromine dissociating/recombining to give BrCl, it should be easy to predict the products formed and perhaps more about the behavior of the system as a function of temperature.

If you have a more complex system, including (for the sake of argument) atoms such as carbon that can form complex ring and chain structures, predicting the outcome will be impossible - since (1) the different ways to combine atoms may lead to a huge number of possible intermediates, many of whose energies/reactivity are not known; (2) even if you could predict these intermediate structures, the number of further possible reactions between them becomes astronomical, and (3) kinetic barriers (activation energies) for reactions are generally harder to predict that ground state energies. As is often the case when theory comes up short, you may have to go into the laboratory and do some experiments. Possible approach: aim the plasma (along with a stream of inert gas such as argon) at a supercooled probe in a vacuum and spectrally characterize the species in the resulting matrix. I imagine that you could end up with some pretty amazing chemical species that way.

I would urge those among my colleagues who are familiar with the current state of theoretical chemistry to add to this discussion. I love a question that makes you go back to basics and rethink everything from the ground up.

  • $\begingroup$ Rethinking theoretical chemistry from the basics? There is something they call the world formula. But that's as much a theoretical physics approach as it is chemistry. And it is very hard to get much grip on that. And I think they are not even close. And if/when it exists I would assume it is almost impossible to treat with or current hardware, even the most simplest systems, but that might change in a couple hundred years $\endgroup$ Sep 12, 2015 at 18:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.