Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Are Chemical Reactions Quantumly Indeterminate?

  1. Apr 4, 2011 #1
    When one atom chemically combines to another atom is it totally determinate? I mean if one atom gradually approaches the other would the reaction occur at exactly the same place or does HUP play a part and give it some randomness?

    I think there is a known answer to this but I cannot find it in google and do not know it.

    (I am thinking if it is totally determinate then every thought we have must be predetermined and there would be no 'free will'. -Just like a random ball computer game always does the same thing.)
  2. jcsd
  3. Apr 4, 2011 #2


    User Avatar
    Science Advisor

    Quantum mechanics doesn't say anything about determinism either way, that's an interpretational issue.
    (And the connection between non-determinism and free will is, IMO, even more tenuous)

    That said, in the interpretations that do hold quantum mechanics to be fundamentally indeterministic, then
    those indeterministic properties do carry over to chemical reactions. Although this doesn't mean much for chemistry, which is statistical anyway.

    Also, AFAIK there are no physiological-scale processes (such as the firing of a neuron) that depend on the outcome of any single chemical reaction.
    We've got much more robustness and redundancy than that. In fact, if there's any 'design principle' to biochemistry,
    it's that it does small steps very many times in parallel, rather than single things doing all the work.
    The only thing there's one of in a cell is the DNA molecule itself.
  4. Apr 4, 2011 #3


    User Avatar
    Gold Member

    Look into Prigogine who contributed to Advances in Chemistry. He saw his dissipative systems as a blow to deteriminism. Of course. As axlm says, a deterministic universe is a philosophical interpretation of the data.

    As for myself, I'm not so concerned about free will. I find human behavior to be mostly predictable. And of course, we still have willpower without free-will (ie we still get to subjectvely enjoy the decision-making process even if it's following a deterministic path)
  5. Apr 5, 2011 #4
    I think this isn't nearly as deep as you're thinking; this basically comes down to when you consider the "reaction" to have happened. As one atom approaches the other, nothing will happen until their wave functions significantly overlap. What a "bond" is is just that overlap, and the "reaction" is the rearrangement of the electron density to the lowest energy state for that new combined system. The only "uncertainty" in this calculation is the uncertainty in the wave functions. If you decide that you know them exactly, then yes, the reaction, as defined by you, would occur at the same distance between the center of mass each time because the overlap would always be the same.

    Now, if you're not doing this as a toy model where you know the exact location of the atoms, then there is the uncertainty in that coordinate, but most importantly probability comes in when you start looking at ensembles that you need to model a real reaction. Look up "reaction cross section"; maybe that's what you're after.
  6. Apr 5, 2011 #5
    Cannot quite follow all that, sorry.
    Does this mean that if an atom approaches another atom in exactly the same way every time that they will combine at exactly the same time and place every time? Or is their randomness here?

    Still worried about my bouncing balls game that does the same thing every time I switch it on except now I am more confused than at the start of this thread. If they were combining atoms would they do the same thing every time?

    Can anyone give me a straight answer?
  7. Apr 7, 2011 #6
    It's not clear to me what you're asking. Here's my confusion:

    What you're describing can be written down on paper, and a calculation can be done to see if the overlap of the orbitals is such that electrons reorganize and you can say a reaction has taken place. THAT will happen the same way every time. No "randomness" here. This is simply defining some point where you say a "reaction" has occurred and a bond is formed or broken and calculating when that happens.

    The problem is, that's absolutely NOT what happens in nature. In nature (or the laboratory if you'd prefer), if you have atoms combining in a bi-molecular reaction, then you have a distribution of the densities of each atom (it's probabilistic how many you'll have in a given volume to react with each other) and a distribution of velocities of each one (obeying the Maxwell-Boltzmann equation), so what you get is a reaction CROSS SECTION that tells you on average how many reaction causing collisions occur at that temperature and density etc. From that you can get things like rate constants that will tell you about how frequently reactive collisions take place etc. If that's what you're talking about, then the whole thing is probabilistic.
  8. Apr 7, 2011 #7
    I am only interested in finding out if chemical process have quantum randomness (I do not care about average rates of reaction for many atoms in a chem lab).
    From the above a single atom will react exactly the same way if it follows exactly the same route to its partner atom every time. I mean the exact time and place and result would repeat every time. I thought the Heisenberg UP would add a randomness to the process. Seems I am wrong?
    That worries me, because it would mean our thoughts (chemical processes) are on a fixed treadmill from which there is no escape. Or am I missing something here?
  9. Apr 7, 2011 #8

    You're creating your own problem here. You're specifying that these atoms will follow the "exact same route" but that only exists on paper. In real life, there is no exact route because there's always uncertainty in the position/momentum trajectory and such. ON PAPER AND ON PAPER ONLY can we specify some conditions that count as a "reaction" and so a calculation and no matter what, for a certain distance apart the wave functions will overlap the same way and you'll always get the same answer to "did it react, yes or no" based on what you decide "react" means. Of course, if you wanted to you could calculate the uncertainty in the position or momentum and get some spread around your chosen value that will say "here is the chance that the reaction actually took place a thousands of an angstrom to the right" or whatever, but that's not really necessary. The thing you're worrying about only exists on paper, not in nature. Shouldn't that be comforting if you're worried about free will stuff?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Similar Discussions: Are Chemical Reactions Quantumly Indeterminate?