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

How can there be back-bonding in PF3?

  1. Jul 10, 2018 #1
    Phosphorus: sp3 hybridized
    Fluorine (each one of three): sp3 hybridized
    Back bonding can occur between only PURE orbitals ie those which are unhyridized.
    Let's look at e- config. of both of them to be a little clear of what I'm asking:
    P: upload_2018-7-10_21-18-43.png
    F: upload_2018-7-10_21-19-4.png

    Since fluorine is sp3 hybridized, the orbitals ie either of px, py or pz (2 of which are filled and are lone pairs on F of the 3 total lone pairs) can't participate in back bonding with the unhiybridized d orbital of Phosphorus.

    My question is, then how does it show back bonding? I know it does, but how?
     
  2. jcsd
  3. Jul 10, 2018 #2

    HAYAO

    User Avatar
    Gold Member

    I don't know. Is it?
     
  4. Jul 10, 2018 #3
    In general , every compound in which central atom has vacant orbitals and adjacent atoms(or even molecules) have lone pair of electrons back bonding occurs.
    Here P has vacant 3d orbitals even when it is sp3 hybridized to form normal covalent bonds with p orbitals of F atoms( note that the hybridization occurs only to the orbitals of the central atom ). Also there are 3 lone-pairs available for each F atom. This satisfies the general condition and thus back bonding is likely to exist in PF3 molecule.
    Hope you are clear.
     
  5. Jul 11, 2018 #4

    DrDu

    User Avatar
    Science Advisor

    Hybridization is not a property of an atom but a way to explain certain types of bonding, and geometries more easily. In fluorine, the energetic difference between s and p orbitals is so high, that hybridization makes no sense and is also not necessary as a linear bond geometry can be explained without hybridization.
    What is even more problematic is that bonding and backbonding involving d-orbitals in main group compounds has since long been disproven. Nevertheless chemistry teachers seem still to be widely ignorant about this fact. See for example, https://onlinelibrary.wiley.com/doi/abs/10.1002/anie.198402721
     
  6. Jul 11, 2018 #5

    HAYAO

    User Avatar
    Gold Member

    As a reference, here's what happens when you calculate PF3 in Gaussian 09W with NBO (natural bonding orbitals). Essentially, NBO analysis gets you as close as possible to the Lewis structure of a molecule, which would help you understand the bonds.

    Computational details: DFT (B3LYP), aug-ccPVDZ, Structure optimization, full NBO analysis.

    The result yielded that PF3 have 8 core electron pairs, 3 two-center bonds, 0 three-center bonds, and 10 lone pairs.
    Lewis occupancy was 99.420% and non-Lewis occupancy was 0.580% (Valence 0.408%, Rydberg 0.173%). Basically, this means that the molecule is 99.420% Lewis structure, roughly speaking.

    The 3 two-center bonds (which is what we are looking at) was this:
    1st P-F bond:
    P (14.56%) s(11.36%) p(84.45%) d (4.20%)
    F (85.44%) s(28.28%) p(71.55%) d (0.17%)
    1.jpg
    2nd P-F bond:
    P (14.56%) s(11.36%) p(84.45%) d (4.20%)
    F (85.44%) s(28.28%) p(71.55%) d (0.17%)
    2.jpg
    3rd P-F bond:
    P (14.56%) s(11.36%) p(84.45%) d (4.20%)
    F (85.44%) s(28.28%) p(71.55%) d (0.17%)
    3.jpg

    Basically, 14.56% of the electrons (of a single bond) lies on the P atom and 85.44% of the electrons lies on the F atoms.
    The percentage of s, p, and d shows the degree of hybridization of each atomic-like orbitals. You can see here that the hybridization of P is not sp3 but something a little more complicated than that. On the other hand hybridization of F seems to be close to being sp3.



    As for the backbonding, I have never tried NBO analysis for analyzing backbonding so I do not know precisely of the procedures. I am assuming that the choice of the reference bonds (turning off donor-NBO-to-acceptor-NBO interaction) is crucial in this case to correctly account for how strong the backbonding energy is. I'll try it if I have the time since I am also interested.
     
    Last edited: Jul 11, 2018
  7. Jul 11, 2018 #6

    DrDu

    User Avatar
    Science Advisor

    First I want to mention that this is a density functional study and thus not directly comparable to valence bond. The small d-orbital contribution is not a sign of backbonding. Rather, the d-orbitals serve as polarisation functions. GAMESS contains the possibility to do true VB calculations, however, I don't have it installed. Maybe somebody wants to try out.
     
  8. Jul 11, 2018 #7

    HAYAO

    User Avatar
    Gold Member

    Hence the reason why I performed NBO calculation. But yes, of course, DFT is not a valence bond calculation.

    I didn't claim that d-orbital contribution is a sign of backbonding. Just a hybridization, or in a sense polarization like you said. Analyzing backbonding requires to turn off certain interactions and is not simple in this case. I think I have written that quite explicitly above...
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted