10
$\begingroup$

Synthesis golf started over summer, and there is clear interest in a non-synthetic molecular design golf to explore the diversity of molecules.

This meta question is intended as a compilation of ideas for molecular design questions for future use.

Each month, we'll use one of the targets (that hopefully get posted) below.

Sensible Restrictions

The key is to propose a question that must be solved with a molecular structure design. In general, these should be solvable through use of chemical databases, calculations, or some combination of both. In general, the resources required should be available to a broad range of users (e.g., free, open source, etc.)

The catch is to ensure the question isn't too broad (e.g., What gas has the highest refractive index?)

So my suggestion is that the question must require some sort of limit, in molecular size or complexity.

Examples:

There have been a few examples that I think would qualify, e.g.:

$\endgroup$

3 Answers 3

Reset to default
8
$\begingroup$

What's the longest possible C=C bond?

There's been ample work and media coverage on how to make a long C-C single bond. To my knowledge, the longest stable C-C bond is currently 1.704 Å, prepared by Schreiner Nature 477, 308–311 in 2011.

All around us, there's considerable plastics, made from polymerizing alkenes like polyethylene. Clearly, longer C=C bonds would decrease the energy required to polymerize the alkane.

Using either chemical databases, like the Crystallographic Open Database, or computational methods, what's the longest C=C bond you can find?

$\endgroup$
5
  • $\begingroup$ Please post (or edit) any corrections or suggestions. Otherwise, I'll post this on Jan 8th. $\endgroup$
    – Geoff Hutchison
    Commented Jan 1, 2018 at 21:53
  • 2
    $\begingroup$ Hmm, "extreme" questions are problematic. At very least it should be longest found, but then that's a matter of time when longer is found. Also if bond is really double doesn't need to be so clear. $\endgroup$
    – Mithoron
    Commented Jan 1, 2018 at 23:40
  • 1
    $\begingroup$ @Mithoron A time limit on the challenge would provide a hard stop, at which point the accepted-answer 'award' could be granted. Perfectly fine for players to keep posting longer ones later on. $\endgroup$
    – hBy2Py
    Commented Jan 2, 2018 at 17:33
  • 2
    $\begingroup$ And yeah, per @Mithoron, a fairly clear set of guidelines for how to classify a C-C bond as double or not would probably be needed. IIRC from Dr. Rzepa's plumbings of the CCDC, there's no bond order information there. Something between a heuristic and a followup quantum bond-order calculation is probably necessary. $\endgroup$
    – hBy2Py
    Commented Jan 2, 2018 at 17:35
  • $\begingroup$ @hBy2Py Yes double bond is fuzzy, but calculating a bond order will probably not help, if you are stretching the bond to an extreme. However, you could define "double bond" as two trigonally coordinated carbons linked together, which at least provides a very restrictive geometric frame, no QC required. $\endgroup$
    – Martin - マーチン Mod
    Commented Jan 9, 2018 at 4:55
5
$\begingroup$

Is there a system capable of cleanly demonstrating a frontside SN2 reaction pathway experimantally?

Just this morning I discovered the theoretical existence of this kind of reaction pathway on the Chemistry subreddit (A. P. Bento and F. M. Bickelhaupt, J. Org. Chem. 2008, 73, 7290-7299; the relevant article discussing some of the fundamentals; DOI: 10.1021/jo801215z).
It seems there is no confirmed experimental observation of the process, because typical backside SN2 pathways have a fraction of the activation energy, and when it's not possible, SN1 can take over instead. How strong of a contender can we make to give unambiguous experimental proof?

$\endgroup$
3
  • $\begingroup$ I have zero knowledge on this matter, and I don't know to what extent we can tackle it. Nevertheless I thought it was an interesting example of the kinds of questions we can consider. $\endgroup$
    – Nicolau Saker Neto
    Commented Jan 4, 2018 at 0:03
  • $\begingroup$ This is awesome, and I never heard of this before, so it is definitely a good idea looking into this. $\endgroup$
    – Martin - マーチン Mod
    Commented Jan 9, 2018 at 5:06
  • $\begingroup$ Bickelhaupt often has great articles for fundamental concepts. :) $\endgroup$
    – Geoff Hutchison
    Commented Jan 9, 2018 at 17:44
3
$\begingroup$

What is the global minima?

I recently came across this post from the chemistry department at the University of Cambridge. The post itself is quite old, but the idea is one that I like.

Essentially, one chooses a fairly complex molecule, and the challenge is to computationally find a minima - whoever finds the lowest energy minima wins!!

In the website linked, the target is a long straight chain alkane, but something more interesting could be used. Given the relative ease of computational chemistry now (many open source packages for molecular mechanics and quantum chemistry) it should be a fairly accessible challenge.

Personally, I'd suggest a molecule with some more functionality, perhaps a recently reported natural product.

$\endgroup$
2
  • 1
    $\begingroup$ I think this would be a great second topic (i.e., February). Have an idea of a target? Ideally, I'd like to see at least 6-8 rotatable bonds, since the number of conformations goes up exponentially. $\endgroup$
    – Geoff Hutchison
    Commented Jan 23, 2018 at 20:48
  • $\begingroup$ I hadn't thought of anything specific. Something like scyphostatin could be okay - has rotatable bonds, a ring, an amide. That said, if I ignore my natural products bias there are a lot of interesting small supramolecular type structures $\endgroup$
    – NotEvans.
    Commented Jan 27, 2018 at 18:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .