Whenever a sample is placed inside the detection coils of an NMR spectrometer the field is distorted. Every sample affects the homogeneity of the field and this is why every sample needs to be shimmed, as discussed previously. The sample also affects the radio frequency circuit that delivers the pulses and detects the signals. To optimise signal detection, the frequency of the circuit needs to be tuned so that it matches the frequency produced by the spectrometer's amplifiers. In practice this is like tuning a radio to the frequency of a particular station. A variable capacitor is adjusted until the desired frequency is obtained. As well as tuning the circuit to the correct frequency, the resistance of the circuit must be adjusted so that it matches the resistance of the spectrometer hardware. This maximises transfer of power to the sample and gives more uniform excitation. Again, a variable capacitor is adjusted.
On our Bruker spectrometers tuning and matching is done using a "wobble curve". This displays the probe response as a function of frequency along the horizontal axis and reflected power along the vertical. Using the tuning capacitor to change the circuit's resonant frequency allows the dip in the wobble curve to be moved horizontally. Using the matching capacitor to adjust the reflected power moves the wobble curve's dip vertically. The figure below shows poor tuning and matching in the left panel and good in the right one. Typically, there is some interaction between tuning and matching so one needs to go back and forth between tuning and matching to find the optimum.
Since tuning is optimising the frequency it is possible to build a probe where the frequency can be adjusted over a large enough range that the frequencies of many different nuclei can be accessed. These types of probes are called "broad band" probes. We have a 5mm room temperature broad band inverse-detected (BBI) probe. Tuning and matching to different heteronuclei is done using sliders that move up and down, as shown on the right panel of the figure below, while tuning and matching the 1H channel is done with the familiar rods, seen in pale yellow in the left panel. To simplify the process of tuning to different nuclei the probe has a card listing the tuning and matching frequencies so that it is easy to get close to the correct tuning and matching for various nuclei.
Optimising the tuning and matching maximises the efficiency with which the radio frequency pulses are transferred from the coil to the sample. If the probe is not well tuned and matched then 90o and 180o pulses will not be calibrated properly. For simple experiments this may not reduce sensitivity much, but for experiments that rely on accurate pulse calibration poor tuning and matching will reduce sensitivity and may prevent the experiment from working properly.
Always enjoy your blogposts, Brendan. Whether one is tuning an NMR probe or a guitar, having a good ‘sense’ of whether an optimum resonant circuit (essentially that’s what the NMR probe is) has been achieved is good for the S/N of the experiment; shimming and tuning affect both. A sense of field homogeneity is better observed when shimming (one should know a shim when it’s achieved - ‘is my residual CHCl3 signal a sharp line?’) but good tune-match is just as important. Fun fact: back in the early days at UCSD, there was only one ‘high field’ NMR - a 360 MHz and it had no 2H channel. Tuning was done in a similar way, but there no way to shim other than observe the FID ‘ping’ of the TMS signal (always, one needed to always add a boatload of TMS to the sample) on an oscilloscope, more like manually tuning a guitar by ear (not an auto tune!). When the FID rang out across the tiny circular screen without significant decay, you KNEW you had a good shim and your 1D or 2D NMR experiment would work! A sense of S/N
ReplyDelete