dilute samples these two spectra are usually accumulated in an interleaved mode by an automation program. You have to set:

decoupler attenuation for selective presaturation [60 dB]
td	32 k
sw	10 ppm
o1	middle of ¹H NMR spectrum
o2	on resonance for the methyl protons in the first experiment and on resonance of the TMS signal in the reference spectrum
p1	90° ¹H transmitter pulse
d1	0.1 s
d2	10 s
ds	4
ns	16

5. Processing

NOE difference spectra can be processed in different ways. Since one wants to observe signal intensity changes of 2 to 10%, one should use an exponential window function with lb = 2 Hz to minimize artefacts of subtraction. One can either transform the two spectra separately using a digitally identical phase correction and subtract the two spectra, or, more conveniently, subtract the two FIDs directly from each other. In the difference spectrum, adjust the phase of the methyl group signal to be negative and the phase of the reference signal (TMS) to be positive. Evaluate only signals which have a correct phase and have therefore not been affected by inadequate spectrometer stability.

6. Result

The figure shows the results obtained on a DRX-400 spectrometer. In a the normal ¹H NMR spectrum is given while in b the NOE difference result is shown with a vertical expansion in the olefinic region. Integration of the methyl group signal and comparison with the integral of the cis olefinic proton (δ_H = 5.54) gives an NOE effect of 5%, taking into account the threefold number of protons within the methyl group. Note the subtraction artefacts for the trans olefinic proton. The methylene group of the ethyl ester side chain at δ_H = 4.2 probably displays a small Overhauser effect in addition to some subtraction artefact.

7. Comments

The exact theory of the NOE effect is complicated and is described in detail in the references given. In contrast to Experiment 4.8, where the saturation for two isolated protons was described, one obtains, in typical organic compounds with different relaxation processes and a molecular weight of about 200 to 1000, positive NOE effects in the order of 5%. One should avoid placing too much weight on a quantitative interpretation of the results of these measurements, since the relaxation times of all neighbor protons are important; furthermore, three spin effects and other complications