Sample Insertion

At the top center of each NMR magnet is a round vertical hole, called the bore, which extends all the way to the bottom of the magnet (Fig. 3.1). At the bottom, the bore is filled with the probe, which is inserted from below, and the room temperature shim coils, which form a concentric cylinder around the probe. The probe has a small vertical hole just large enough to admit the sample tube, and inside there is a set of RF coils that surround the sample. These are aligned with the center of the superconducting magnet. Wires connect these coils to the probe head, at the bottom of the probe, where connectors lead RF power into and out of the probe from cables. The probe coil acts as a radio transmitter antenna during the exciting pulse and as a radio receiver antenna during acquisition of the FID. The room temperature shims are just coils of wires wound in various directions and spacings around the probe so that adjusting the currents in these coils adds or subtracts magnetic field strength to the space occupied by the sample to make up for any lack of homogeneity in the main (superconducting) magnetic field. The sample tube is held by a plastic spinner turbine or "spinner," which is ejected from the probe out through the top of the bore with a cushion of air pressure, and inserted by gradually reducing the air pressure. When the spinner turbine and sample are resting in the probe, a small current of air can be directed at a skewed angle toward the spinner turbine, causing the spinner to lift slightly and spin on the vertical axis. For one-dimensional (1D) NMR spectra, samples are usually spun at about 20 Hz (revolutions per second) in order to average out any lack of magnetic field homogeneity along the X and Y (horizontal) axes.

Figure 3.1

To insert the sample, first push the sample tube gently into the spinner turbine and adjust the vertical position of the tube using a gauge to assure that the actual sample solution will be centered in the probe inside the RF coils. Place the sample on the air cushion at the top of the magnet bore and deactivate the eject air, and the sample will gently descend into the bore until the spinner rests on the probe with the bottom of the NMR tube inserted into the probe.


Although the magnetic field of a superconducting magnet is very stable, there is a tendency for the field strength to change gradually or "drift" by very small (parts per billion) amounts. If this tendency were not corrected, it would be impossible to sum a number of FID acquisitions because each FID would have a slightly different frequency than the previous one. Drift is prevented by a separate channel in the probe and the spectrometer that detects deuterium (2H). This can be done independently of proton or carbon acquisition because deuterium nuclei resonate at a very different frequency (e.g., 30.7 MHz compared to 200 MHz for !H on a 200 MHz instrument). The lock channel continuously detects the deuterium signal of the deuterated solvent and monitors its chemical shift position. You can think of this as a separate NMR spectrometer dedicated to 2H detection, which runs continuously in the background. Because the resonance frequency of any nucleus is proportional to the magnetic field strength (vo = yBo/2n), any drift in the magnetic field (decrease/increase)

Decrease Figure 3.2

will cause a shift (upfield: lower frequency/downfield: higher frequency) of the deuterium frequency detected. This shift in frequency is connected to a feedback loop that adjusts the field strength (by changing the current through a room temperature coil in the shim cylinder) so that the deuterium frequency does not change. This mechanism is called the "lock" system, and it maintains a constant magnetic field strength throughout your NMR acquisition. Regardless of the lock display presented to the user, the lock circuitry sees a dispersive (up/down) deuterium signal centered on the zero frequency (null point) of the feedback circuit (Fig. 3.2). The magnetic field strength is manually adjusted (Zo or field knob) to center the signal at the zero frequency. When the lock is turned on, the feedback loop is activated and control of the magnetic field strength (Bo) is given over to the control circuit. If the magnetic field increases slightly, the 2H signal is shifted to the left (higher frequency) leading to a positive error signal. This signal decreases the current in the Zo (field) coil in the shim stack, which decreases the magnetic field, correcting the drift. A slight decrease in magnetic field leads to the opposite error signal and a compensating increase in current sent to the shim coil. The system cannot achieve lock unless the null point is between the two maxima of the dispersive peak when the feedback loop is activated. The proper lock phase setting assures a symmetrical (equally up and down) dispersive signal in the feedback loop.

3.3.2 Locking

As soon as your sample drops into the probe, the 2H signal will become visible on the screen. You may need to increase the lock power and gain, and adjust the field (Bruker: Field, Varian: Zo) setting to see the lock signal. If the homogeneity of the magnetic field is very poor ("bad shims") you may not even see a signal! There is another major difference (Fig. 3.3) between Varian and Bruker in the way the lock signal is displayed: Varian shows a time-domain signal, which is a sine wave whose frequency is the audio frequency of the deuterium signal. Bruker shows a frequency-domain signal, which is "swept" by moving the deuterium excitation frequency back and forth repeatedly over a range of frequencies. When the excitation frequency matches the deuterium resonance frequency, you get a peak

Varian Bruker

Figure 3.3

Varian Bruker

Figure 3.3

that dies away in wiggles ("ringing") as the excitation frequency moves away from resonance. The same peak and ringing is observed as the excitation sweeps back the other way across the resonance position. This display is in the "unlocked" state: the feedback loop is inactivated and the deuterium signal is simply observed on the screen. If the deuterium frequency is far from the locking position, you will not see any signal. For Varian this is because the time-domain frequency is very high and the signal is weak; for Bruker it is because the deuterium resonance position is outside the range of frequencies being swept. On the Varian spectrometers, you adjust the field strength (Zo) until you begin to see a sine-wave signal ("wiggles") and continue to adjust until the frequency (number of cycles of sine wave displayed) decreases to zero and you have a horizontal line instead of a sine wave. On Bruker, you adjust the field strength (field) until the pattern of peaks and ringing is exactly centered on the screen. You are now ready to activate the lock feedback loop. Turning on the lock leads to a horizontal line that rises above the baseline. You can think of the height of this line as the peak height of the deuterium NMR peak of the solvent. Once locked, the deuterium frequency is no longer swept (Bruker) and the magnetic field strength (Bo) should be rock-stable over time. Mixed solvents (e.g., d6-acetone/CDCl3) or solvents with more than one deuterium resonance (e.g., CD3OD) can lead to problems if you lock on the wrong 2H signal, so be sure to verify which signal is centered in the 2H spectral window before turning on the lock.

3.3.3 Lock Parameters

The field setting required to center the lock signal depends on the deuterium chemical shift, which is roughly proportional to the proton chemical shift. Thus, the deuterium resonance of CDCl3 (!H 8 7.24 ppm) is downfield of the deuterium resonance of d6-acetone (!H 8 2.04 ppm) but very similar to that of d6-benzene (!H 8 7.15 ppm). The field settings for various common deuterated solvents are often posted near the spectrometer or in a logbook to allow easy access to "ballpark" settings. The lock level (the height of the lock signal on the screen) represents the height of the deuterium peak. As shimming (i.e., homogeneity of the magnetic field) is improved, the deuterium resonance becomes sharper and the height of the deuterium "peak" increases, since the area (amount of deuterium in the sample) remains constant. Two other factors affect the lock level: The lock power affects how much 2H signal is fed into the probe to excite the deuterium nuclei and the lock gain affects how much the detected signal is amplified in the receiver before being presented on the screen. Increasing either one will increase the lock signal level, but increasing the lock power will eventually "saturate" or overload the 2H nuclei with RF energy, causing the lock level to "breathe" or oscillate slowly up and down. Since the lock level is used to monitor changes in field homogeneity while shimming, a randomly oscillating level will interfere with shimming. If this happens you need to turn down the lock power and then pump up the lock gain as necessary to get a good lock level. A "good" lock level is about 80% of the maximum, allowing room for improvement during the shimming process. The lock signal should have a little bit of noise, indicating that the lock power is not excessive, but it should not have so much noise that small changes in the level cannot be readily observed. It is important to realize that the lock level is arbitrary; you can increase it or decrease it at any time by adjusting the lock gain and the lock power. It is only the changes in lock level resulting from changes in the shim settings that are important. If shimming brings the lock level above 100%, just reduce the lock gain to bring it back to 80% and continue shimming.

0 0

Post a comment