Analytical Chemistry - 13-C NMR Chemical Shifts
Click to enlarge

In previous entries in the Analytical Chemistry series of graphics, we’ve looked at some of the tools that chemists can use to determine the identity of compounds in various samples, including infrared spectroscopy and hydrogen nuclear magnetic resonance (NMR). Today looks another similar method, that of carbon NMR; the graphic provides some general information on interpreting the resultant spectra, whilst we’ll briefly discuss how these signals are created below.

As mentioned, we’ve already discussed nuclear magnetic resonance, or NMR, in a previous post where hydrogen NMR was examined. That post also provides a more thorough overview of how the signals are generated, and the method for carbon-13 NMR is exactly the same – it’s just carbon atoms that are involved, instead of hydrogen atoms.

Like hydrogen atoms, some carbon atoms can have a property called ‘spin’. Spin is a rather abstract concept, but at a simplified level, nuclei that possess this property can be thought of as acting like very small magnets. These magnets, when place in a magnetic field, can align either with or against the field, and this is the basis of NMR. You’ll not I mentioned some carbon atoms, not all. This is because not all carbon atoms have spin; in fact, only carbon-13 atoms do.

Carbon-13 is slightly different from carbon-12, the more abundant isotope of carbon, in that it has an extra neutron in its nucleus, but otherwise its chemical properties are the same. It accounts for just 1% of all carbon atoms, with carbon-12 accounting for the other 99% (there are also some other very low abundance isotopes). It’s the carbon-13 atoms, then, that are responsible for the spectrum seen in carbon NMR, and carbon-12 atoms play no part.

By way of a brief overview of the process by which the spectrum is generated, the sample is put into a machine that can apply both an external magnetic field to the sample, and irradiate it with radio waves. Some machines vary the magnetic field strength and keep the radio frequency fixed, whilst other vary the radio frequency whilst the magnetic field remains the same. Either way, the result is that the carbon-13 nuclei in the sample can absorb energy from the radio waves and ‘flip’, so they are no longer aligned with the magnetic field, but against it. This can be used to generate a signal, which is where the peaks in the resultant spectrum come from.

Much like hydrogen NMR, the environment of the carbon atoms (i.e., the other atoms that it is in proximity to in the molecule) affect where a peak is seen in the spectrum. A closer proximity to electronegative atoms such as oxygen or nitrogen will result in the peak appearing further to the left of the spectrum. Thus, the carbonyl compounds, which all contain a C=O bond, appear on the left-hand side of the spectrum, whilst simpler, alkyl groups containing only C–H bonds appear on the right-hand side of the spectrum.

The low abundance of carbon-13 atoms means that the spectra generated look a little different from those seen in hydrogen NMR. First of all, you might wonder why spectra can be generated at all, considering its low abundance, but bear in mind that even a very small sample will contain millions upon millions of molecules. Therefore, even that 1% of all carbon atoms becomes a large number, meaning a spectrum can still be generated.

Secondly, if you’re familiar with hydrogen NMR spectra, you’ll know that the peaks generated are often split into a multitude of different peaks depending on how many other hydrogens are in proximity on adjacent carbons in the molecule. This can add an extra layer of complexity to interpreting the spectrum. With carbon-13, however, this problem doesn’t exist. Considering their 1% abundance, the probability of two carbon atoms appearing next to each other in the same molecule is extremely low, so for the whole sample, no splitting of peaks is seen. Actually, we would expect to see splitting as a result of interaction with hydrogen nuclei in the sample too, but a technique called ‘decoupling’ prevents this from being seen in the spectrum.

Why is this useful? Well, it’s another item in the chemist’s toolkit that allows us to identify unknown chemicals. Combined with both hydrogen NMR and infrared spectroscopy, identifying simple molecules becomes much easier. It can also give important pointers in more complex molecules.

As always, it’s worth noting that the graphic is designed to give a general overview of the rough points at which different carbon atoms in different environments appear on the spectrum. These are approximate values only, and can be affected by other substituent groups, as well as the temperature and the solvent in which the sample is placed. Also, if you want to read into the technique in more detail, there are some further links provided below.



The graphic in this article is licensed under a  Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Want to share it elsewhere? See the site’s content usage guidelines.

References & Further Reading

4 CommentsClose Comments


  • Trackback: Spectroscopy | Pearltrees
  • clemistry
    Posted April 7, 2015 at 8:51 pm 0Likes

    Another nice post. I would like to point out, though, that whilst 13C-13C splittings are not observed, 13C-1H splittings need to be (and typically are) suppressed through broadband decoupling of this other NMR-active nucleus. These couplings are quite large (one-bond H-C typically ~140 Hz), and if not decoupled, they make the spectrum quite difficult to interpret because they are spaced wider than the differences in chemical shift and are almost always overlapping one another. These couplings are also observed in the 1H NMR spectrum (called “carbon satellites), but are at 0.5% of the signal height (doublet) of the main 1H peak and are thus not generally a problem.
    This may be a nice segue into a third infographic in the NMR series, on the NMR of other nuclei, like 31P or 19F (both 100% natural abundance, and whose couplings can sometimes be observed in 1H and 13C), or on 2-D NMR, which takes advantage of these couplings to correlate 1H & 13C signals.

    • Compound Interest
      Posted April 7, 2015 at 11:32 pm 0Likes

      Very good point re: decoupling, that completely slipped my mind when writing this. I’ll add in mention when I get a minute. Thanks for the reminder!

      Certainly more planned for the analytical chem series. I’ll probably make one relating to mass spectrometry next, but there some on other forms of NMR would be nice too.

  • Yusuf Ooi
    Posted January 24, 2016 at 3:30 am 0Likes

    “bon-12. NMR doesn’t
    work for carbon-12, as its nucleus doesn’t have a ‘spin’. ” Explain detail

Comments are closed.