The probe-head should be tuned to the sample. A doped sample (proton relaxation time of about 50 ms) is chosen to avoid saturation effects. Load standard ¹H parameters, record a normal ¹H NMR spectrum and note the offset of CHCl₃. You have to set:

transmitter attenuation [0 dB]
td	4 k
sw	500 Hz
o1	on resonance of CHCl₃ signal
p1	1 μs ¹H transmitter pulse, to be increased in 1 μs steps
d1	5 s
rg	receiver gain for correct ADC input (with 90° pulse)
ns	1

Record one spectrum with a 90° pulse and check on all parameters. Reset the pulse duration to 1 μs and use an automation routine, which performs this experiment 150 times in sequence by increasing the transmitter pulse with an increment of 1 μs.

5. Processing

Use standard 1D processing (see Exp. 3.1) applying an exponential window with a line-broadening factor lb = 1 Hz. Adjust the phase of the spectrum obtained with the 90° pulse and always use the same digital phase correction.

6. Result

The figure shows the signals obtained on a DRX-400 spectrometer with a 5 mm inverse multinuclear probe-head with z-gradient. Note the decay of the signal intensities after several full rotations by 360°. As a measure for the r.f. homogeneity one compares the intensities of the signals after 90°, 450°, and 810° pulses.

Here, ratios of 83% at 450° and 73% at 810° were measured, which are acceptable values for modern probe-heads.

7. Comments

The decay shown in the figure can be explained by assuming that the precession in the r.f. field is not homogeneous, because different spins at different locations in the NMR tube experience slightly different r.f. fields. Thus, during the r.f. pulse we already have a dephasing of the magnetization, which will increase with the increasing duration of the pulse and hence lead to a smaller signal. Other factors such as relaxation during the pulse also have to be considered. The r.f. homogeneity is most important for 180° pulses so as to achieve a complete inversion for all spins. For some applications, however, extreme r.f. homogeneity is not desired. For example, purging spin-lock pulses (see Exp. 6.17) rely on a complete dephasing during the spin-lock, and this is more effective with poorer r.f. homogeneity. There is also a solvent dependence of the r.f. homogeneity due to different magnetic susceptibilities.