I have run calculations that predict atomic configurations. As a simple example, in a system that contains H and O, I might get a list like:

Configurations for the O atom
0 2
2 6

This would tell me that in my system, there's one O atom bonded to two H atoms and there's also a molecule made up of three O atoms bonded to each other and six H atoms. My real systems are more complex.

I am not really a chemist and have no idea whether some of the molecules I have calculated actually exist and what they would look like if they did. Does anyone know if there's software available that might help me with this; i.e., where I can enter a potential formula and the software draws the molecule and tells me whether it exists?

I have so far checked out ChemDraw Professional and Avogadro and it doesn't look like they do what I need. Thank you for any tips.

  • $\begingroup$ Would it be OK if the software does a database search? $\endgroup$
    – Karsten
    Commented Sep 3, 2019 at 11:30
  • 1
    $\begingroup$ Thank you for your reply, Karsten. Yes, that would be fine--although it would be great to know if there's any software that can maybe make predictions based on more fundamental phenomena. I know that databases might not contain all possible structures for a particular system, but any guidance would be helpful. My calculations are predicting polyatomic Na, Cl, H, and O structures and I have no idea whether they can exist. $\endgroup$ Commented Sep 3, 2019 at 11:33
  • $\begingroup$ They can't, unless they are NaCl, or H2O, or a handful of other known compounds. Then again, maybe you are interested in some transient clusters, in which case the databases won't help either. $\endgroup$ Commented Sep 3, 2019 at 11:40
  • 1
    $\begingroup$ I know NaCl, HCl, NaOCl, HClOx, NaOH and H2O exist. I am having troubling finding a structure made of Na, Cl, H, and O in any stoichiometry. I did find a paper with hydroxide ion (OH-) impurities in NaCl crystals: journals.jps.jp/doi/abs/10.1143/JPSJ.54.175 $\endgroup$
    – Karsten
    Commented Sep 3, 2019 at 11:43
  • 3
    $\begingroup$ Why list them? I don't think they even deserve to be called structures. True, many of them would pop up for a split microsecond, but that's hardly a fact of any consequence. $\endgroup$ Commented Sep 3, 2019 at 12:01

1 Answer 1


I believe this website will be of use to you


It appears to take names, formulas, smiles etc. If the name/formula/smile ID pops up in the search bar, it will draw it for you. Its database seems quite large, I use it for some pretty big drug molecules.

It draws 2D and 3D images, and it also does single/double/triple bonds.

  • $\begingroup$ Thank you very much, Charlie, this does look very useful and easy to use! I will explore it! $\endgroup$ Commented Sep 3, 2019 at 19:32
  • 1
    $\begingroup$ It may not help a ton for your specific problem, but it is quite useful in general $\endgroup$
    – B. Kelly
    Commented Sep 3, 2019 at 20:12
  • 1
    $\begingroup$ It's definitely more appropriate for my problem than anything else I have come across so far. I'll have to wait until tomorrow morning to check it out, so don't know yet how many of the structures I'm seeing appear in its database, but it looks like a great start no matter what. I'd just like to get some idea of whether my predictions are reasonable! Thank you very much indeed. $\endgroup$ Commented Sep 3, 2019 at 20:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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