5
$\begingroup$

I'm reading up on molecular modelling and have come across some terms I should probably have learnt a long time ago but always scared me off a bit with the Greek letters. I guess I'm a big boy now so it's probably time to clarify...

The textbook uses the lowercase symbol omega ($\omega$) for torsion (a.k.a. dihedral) angle while Wikipedia gives the uppercase, $\Omega$, plus both upper and lower case phi ($\Phi$, $\phi$) and psi ($\Psi$, $\psi$) in its description which from reading I'm starting to understand designate particular geometrical cases from the article, though it looks like it might be a little difficult to remember at first.

My textbook also gives in the list of symbols and physical constants [lower case] tau, $\tau$, but I can't see any mention of this on the Wiki page.

Firstly I wanted to ask is the use of upper/lower case totally arbitrary, and secondly is tau just a non-specific way of expressing torsion angle compared to the other 3 (which is the only use I can imagine for it?) Or is this use totally confined to molecular modelling?

$\endgroup$
4
  • 1
    $\begingroup$ From personal experience as a computational chemist, we have always use the symbol $\tau$ when referring to a torsional angle. Also, a torsional angle is not equivalent to a dihedral angle (even though a lot of people interchange the two). Whatever symbol you use, just be sure to define it explicitly somewhere and you're golden. Secondly, as far as I'm aware, casing is more situational than anything else. $\endgroup$ Commented Oct 23, 2013 at 19:30
  • $\begingroup$ Right, I've not been introduced to a formal definition of either, I had a working understanding of torsion angle from a diagram of ethane in a staggered conformation and just Googled 'torsion angle', the first result for which was Wiki:Dihedral angle... Care to explain the difference as you see it? (or anyone else reading this!) $\endgroup$ Commented Oct 23, 2013 at 19:35
  • $\begingroup$ I've written out about 4 halfway attempts trying to explain this only to realize its really difficult because I have never really given it much thought. I can give an example however. Imagine a CH$_4$ molecule with atom labels H1,C2,H3,H4. The angle between the following four atoms H1-C2-H3-H4 would be a dihedral angle. Imagine Glycylglycine (Glycine dipeptide) and you wanted to describe the cis- or trans-like nature of the peptide bond. You would measure the angle C-C-N-C which would be a torsion angle (where the C-N bond is the peptide bond). I could explain this better with pictures. $\endgroup$ Commented Oct 23, 2013 at 19:45
  • 1
    $\begingroup$ For the record, I just went and asked a couple of professors still lingering around here and they were both unable to explain it to me. Will follow up on this once my boss is back in town next week. $\endgroup$ Commented Oct 23, 2013 at 20:09

1 Answer 1

5
$\begingroup$

It is a habit to assign greek letters (most often lowercase) to angles. In case of "normal" angles, it's mostly $\alpha, \beta, \gamma$, whereas for dihedrals some higher letters, as you mentioned, $\phi, \psi, \omega$. As noted in comments, whatever you use, clearly state what you mean.

The difference between dihedral and torsional angle is more conceptual, then formal. In torsion you discuss rotation around a bond, dihedral is general angle of two planes, as mentioned in the $\ce{CH_4}$ example. The formal expression for both of them is the same, indeed.

To be noted, there are some well known widely used dihedral angles, where you should stick to the common notation. The most prominent example is the protein backbone arrangement, see Ramachandran plot. In this case you should not invent your own notation.

$\endgroup$
3
  • $\begingroup$ "In torsion you discuss rotation around a bond" I didn't know this was the actual defining factor for a torsion even though its how I've always implemented it. A torsion is still the angle of two planes no? $\endgroup$ Commented Oct 24, 2013 at 12:06
  • 2
    $\begingroup$ You are right, torsion angle is measured as angle of planes ABC x BCD, where you expect bond B-C to be the one, which rotates. I.e. central bond in butane. $\endgroup$
    – ssavec
    Commented Oct 24, 2013 at 19:42
  • $\begingroup$ I liked your answer. I just never realized it was not a 'formal' definition. $\endgroup$ Commented Oct 24, 2013 at 19:46

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.