| SETTING UP YOUR EXPERIMENT | |||||
| Designing an NMR experiment: | |||||
| Sweep width. | |||||
|
As indicated in the previous section, the range of precessional frequencies to be observed in an NMR experiment depends on the offset frequency and sweep width. While the offset defines the center of the range, its breadth is governed by the sweep width. So, a sweep width of 2400 Hz (2.4 kHz) covers 12 ppm of chemical shift if the 1H Larmor frequency is 200 MHz (2400 Hz/200 MHz). Whether these 12 ppm go from 0 to +12, from -2 to +10, or even from +5 to +20 ppm is determined by the offset frequency. Roll your mouse over the images below to see the effect of changing the offset (red curves), the sweep width (blue), and both (yellow). |
|||||
| A chemical-shift range of 12 ppm is reasonable for 1H, but other nuclides may exhibit much broader ranges. For example, typical 13C shifts appear over 250 ppm, so, for the spectrometer cited above, the sweep width would have to be set to 12500 Hz or 12.5 kHz (the Larmor frequency of 13C is only 1/4 that of 1H, so 12500 Hz/50 MHz = 250 ppm). | |||||
|
It is important that the sweep width be set broad enough to encompass all the resonances in the spectrum. If some peaks occur outside of the range, they will be aliased, or folded over, and appear at ncorrect chemical shifts. Furthermore, aliased peaks are usually unphasable. Generally, standard experiments have appropriate sweep widths setv as default values, so this parameter will not often need to be changed. However, if your sample is inorganic, or an unusual organic compound (such as an organometallic), you should check literature references to insure that the default sweep-width value will cover the expected chemical-shift range. |
|||||
| NOTE: The term "sweep width" is somewhat misleading, because it implies that the frequency is being varied over the specificed range. This is exactly what was done in the the early days of NMR, but the pulsed, or Fourier-Transform, approach has superceded the swept-field experiment. The term persists, however. | |||||
