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

NMR Spectra

  1. Mar 25, 2004 #1
    Recently in Chemistry we have been doing about how it is possible to identify carbon compounds by how many proton environments there are around a molecule.
    For this we look at graphs with various peaks dotted around them, this is not what bothers me, what bothers me is that some of these peaks have been split into a number of peaks. We have been taught about the n+1 rule.

    Consider ethanol:

    Code (Text):

        H   H
        |   |
    H - C - C - O
        |   |    \
        H   H     H
     
    There would be 3 proton environments around that molecule, one on the methyl group on the left, one for the two protons on the right hand carbon molecule and another for the proton bonded to the oxygen.

    The methyl group peak would be split into three separate peaks, the two protons on the right hand carbon peak would be split in to four separate peaks and the proton on the oxygen would just be a single peak.

    I am realy looking for as simple an explanation as possible as to why peaks split at all. Is that possible?
     
  2. jcsd
  3. Mar 29, 2004 #2
    Well, I've just done this in chemistry on wednesday so I think that I can help you. You see, although protons in the same chemical environment cannot have an effect on each other (don't ask me why cos apparently that goes into quantum physics- of which I have little to no knowledge) the protons on adjacent carbons CAN have an effect (though not the ones which are further away like the H of the hydroxy group in ethanol).

    The protons can either align themselves with the magnetic field (say > if the magnetic field is >) or against it (< for a > magnetic field).

    This means that they will have different net effects if different numbers of them are with or against the field. Example if in the CH2 group the protons are aligned >> the net effect is 2> if they are aligned << then the net effect on the protons in the CH3 group is 2< and if <> then the net effect is 0. This will give three different peaks (as in three different effects) for the CH3 groups adjacent to those respective CH2 groups (two smaller and one larger as one net effect will be more common in all the molecules of ethanol- I think it's <>). For CH3 there are four net effects which can be had on the protons in the CH2 group. This means that there are four peaks at CH2. As a result you can see that the CH3 group has a CH2 adjacent to it and that the CH2 group has a CH3 group adjacent to it.

    The H in the hydroxy group is too far from the other chemical environments and is not effected at all (one net effect which is 0). Hence there is only one peak.

    I hope I explained this well. (As much for my sake as for yours- if I can explain this after only learning it on Thursday then that bodes well for my A Levels. Heh.)

    If there are any nerds out there who are better qualified than I (a lowly A level student meaning that my actual qualification doesn't extend further than a double award GCSE in "Science") then please feel free to correct my work and make me look a fool...

    Rock and roll.
     
  4. Mar 29, 2004 #3
    How do you know if they have an effect on each other or not? They are in the same chemical environment so it would make sence for them to all show up the same on the NMR spectra.


    I don't get how three quarks can align themselves to a magnetic field but I can just gloss over this for now and accept it. :smile:


    Very well explaned, I can understand why the graph is interpreted the way it is now.


    Our teacher didn't go into any of this stuff, which is a shame because this is the kind of stuff that I like to learn about.

    I don't suppose you know why the signal for an OH group collapses when you add D2O do you?
     
  5. Mar 29, 2004 #4
    As for nuclei in the same chemical environment, they *do* show up in the NMR spectrum as exactly the same - this is what gives rise to varying signal intensities. For example, if you have six equivalent protons and then another four equivalent protons, you'll have (after integrating the signal) a ratio of 6:4 (3:2).

    The origin of nucleon spin is, as I last bothered to read, still an open question. There was a Scientific American article a few years back (1999?) that mentioned how it appeared - at time of that publication - that only about a third of a nucleon's spin is attributable to the quarks and the rest is derived from the full fledged dance of quantum chromodynamics. For many applications in NMR, this is not a terribly big deal. It's something for the nuclear physicists to worry about. :)

    D2O will exchange one of its deuterons for the proton, thereby nixing that proton resonance.
     
  6. Mar 30, 2004 #5

    Score! I still can't understand one when it's put in front of me though. Heh.

    Ah now that's something that will make it easier- forgot about that. Thanks.

    Heh- I kinda like this forum.
     
  7. Mar 30, 2004 #6

    Monique

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You know there is a Chemistry forum on this board..
     
  8. Mar 30, 2004 #7
    I'm not too sure why D2 should swop one of it's D's for an H (and only an H that's bonded to an O) but I don't think I'll inquire about that.

    Yeah I know there's a Chemistry forum but this is mainly to do with the Physics side of things.
     
  9. Mar 30, 2004 #8

    Monique

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    If this isn't Chemistry, I wouldn't know what is.
     
  10. Mar 30, 2004 #9
    What, detecting the spin of protons?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: NMR Spectra
  1. NMR explained (Replies: 11)

  2. Absorption Spectra (Replies: 7)

  3. NMR operators (Replies: 10)

  4. NMR mathematics (Replies: 7)

  5. Hydrogen spectra (Replies: 6)

Loading...