Recording NMR spectra

Sample preparation and why we use deuterated solvents

To prepare a sample for NMR spectroscopy, a small amount of the material (10–50 mg) is dissolved in about 0.6 cm3 of a deuterated solvent, usually CDCl3 (deuterated trichloromethane, often referred to as deuterochloroform). Deuterated solvents are used in order to avoid ‘swamping’ the spectrum of the sample with peaks arising from the hydrogen atoms in the solvent itself because deuterium and hydrogen atoms have very different magnetic resonance frequencies, the deuterium peaks appear in a significantly different area of the spectrum and are therefore not seen in the region being looked at.

We can explain this with the analogy of tuning in a radio, as different nuclei (and isotopes) possess different resonance frequencies, it means that to observe their NMR signals you have to ‘listen’ at different frequencies, you don’t hear all the radio stations together, you tune to each individual stations frequency and only hear the broadcast from that station. Just as for radio stations, in NMR you ‘tune’ to the frequency of the observed nucleus and don’t ‘see’ any other nuclei – listening to ‘radio proton’ will not have any signals from ‘radio deuterium’ and vice versa!

The second reason why we use deuterated solvents to prepare samples is for stability. Modern Fourier transform (FT)-NMR spectrometers require samples to be run in a solvent containing deuterium because the instrument ‘locks’ on the deuterium resonance to achieve field–frequency stabilization. Although the superconducting magnets that allow us to observe NMR spectra are highly stable, there is still a slight drift in the magnetic field. If this drift is not compensated for, then, over the time taken to collect an NMR spectrum, the NMR signals will drift, causing them to smear out and become broad and featureless. Since the drift of the magnet is the same for all NMR nuclei, the spectrometer is designed to lock onto a nucleus that we aren’t observing (the deuterium in the deuterated solvent) and, as this drifts, use it to compensate for the drift of the nuclei that we are observing (typically proton).

The sample dissolved in an appropriate deuterated solvent is placed in a high-quality, precision made glass NMR tube ready for introducing into the magnet of the spectrometer.

Residual solvent peaks

The 1H NMR spectrum of a sample run in CDCl3 will always show a very small singlet peak at 7.26 ppm because chloroform is never 100% deuterated and a tiny amount of residual CHCl3 is therefore present in the sample since typically deuterated solvents are supplied with 99.8% level of deuteration. The smaller the amount of sample in the solution the more evident this residual solvent peak becomes.

Similarly, 13C NMR spectra of CDCl3 ­ solutions will show three peaks centred around 77.27 ppm for the natural abundance 13CDCl3. Why do we see three peaks? Deuterium is magnetically active and couples to the 13C nucleus giving rise to the observed triplet.

Residual solvent peaks are also seen in the 1H and 13C NMR spectra of solutions in the other common deuterated solvents and Tables 1 and 2 below lists the chemical shifts and multiplicities (number of lines) of solvents used in running the spectra in the compound library. More extensive compilations can be found at the following website www.chem.umd.edu/nmr/reference/isotope_solvent.pdf.

The second problem that arises is that all deuterated solvents are hygroscopic. This means NMR solvents will contain traces of water, when observing 1H NMR spectra, this additional trace of H2O will also produce a signal in the spectrum. NMR solvents will contain traces of water particularly when the samples contain exchangeable protons (eg OH and NH2 groups). This appears in the form of an HOD peak (water in which one of the protons has been exchanged for a deuteron). The chemical shift of this residual peak is very variable depending on the nature of the solvent see Table 1.

It should be noted that some solvents are highly hygroscopic (D2O and d6-DMSO) producing very large residual water signals however it is possible to remove all traces of water from some deuterated solvents – typically by using rigorously dried glassware and carrying out all procedures under an inert atmosphere (dry nitrogen gas).


Table 1: Residual solvent peaks in 1H NMR spectra

 

 

 

 

 

 

Solvent

1H Shift*

Multiplicity

1H shift of HOD*

 

Trichloromethane-d

7.27

1

1.5

 

Dimethyl sulfoxide-d6

2.5

5

3.3

 

Deuterium oxide

4.8

1

4.8

 

Methanol-d4

4.87

1

4.9

 

 

3.31

5

 

 

Propanone-d6

2.05

5

2.8

 

*  shift in ppm from TMS or in the case of D2O DSS


Table 2: Residual solvent peaks in 13C NMR spectra

 

 

 

Solvent

13C Shift*

Multiplicity

Trichloromethane-d

77.23

3

Dimethyl sulfoxide-d6

39.51

7

Deuterium oxide

N/A

N/A

Methanol-d4

49.15

7

Propanone-d6

206.68

13

29.92

7

*  shift in ppm from TMS

 

 

 

 

 

Accidental residual solvent peaks - Samples prepared in the laboratory have often been isolated from solutions in one of the common organic solvents.  If care is not taken to remove the last traces of this solvent, used for extraction or recrystallisation, peaks will often be seen for these ‘residual solvents’ as well.