Transmitter power

Figure 6.206 Schematic circuit diagram of a transmitter power supply with galvanic isolation between input, output and auxiliary power.

A generalized CP pulse sequence is shown in Fig. 4.5.2, with vertical displacements indicating transmitter power and the horizontal axis indicating elapsed time. A CP pulse sequence begins with a pulse delay, td, to allow recovery of proton polarization along the static field of the NMR magnet. This can be achieved if td greatly exceeds T,(H), the time constant for recovery of equilibrium polarization. 

The large spectral widths required by some of the applications also put more severe demands on pulse power, if uniform excitation is to be achieved across the full width of the spectrum. If both components of the complex magnetization are detected (quadrature phase detection) the carrier can be placed at the centre of the spectrum without any rf carrier folding occurring as in single-channel detection better uniformity of excitation is thus achieved at a given transmitter power. 

Fig. 10. Effect of a droop of the transmitter power on m.p. spectra. An exponential decrease of all flip angles, which amounts to no more than 1% after 100 BR-24 cycles, that is, 2400 pulses, is stipulated. Note the wiggles at the feet of the lines.

Switch to CP mode and increment the transmitter power of the X nucleus until the maximum signal is observed (i.e. the optimum match is obtained). 

The transmitter power needs to be adjusted to provide pulses short enough to obtain nonselective 180° pulses. In applications requiring large coils, resonators can be constructed which are fed with the sine and the cosine component of the rf signal [Hou2]. By using such circularly polarized excitation a factor of two is saved in the rf power in comparison to linearly polarized excitation [Glol]. 

The first procedure for locking is to establish the resonance condition for the deuterium signal, which involves altering either the field or the frequency of the lock transmitter. Of these two options the latter is preferred since it avoids the need for changing transmitter frequencies and is now standard on modem instruments. Beyond this, there are three fundamental probe-dependent parameters that need to be considered for optimal lock performance. The first of these is the lock transmitter power used to excite the deuterium resonance. This needs to be set to the highest usable level to maximise the signal-to-noise ratio but must not be set so high that it leads to lock saturation. This is the... 

Although spectrometer transmitters are frequently used at full power when applying single pulses, there are many instances when lower transmitter powers are required, for example the application of decoupling sequences or of selective pulses. These lower powers are derived by attenuating the transmitter output, and the units used for defining the level of attenuation are the deciBel or dB. This is, in fact, a measure of the ratio between two power levels. Pi and P2, as defined by ... 

Having determined the necessary pulse duration, the transmitter power must be calibrated so that the pulse delivers the appropriate tip angle. This procedure differs from that for hard pulses where one uses a fixed pulse amplitude but varies its duration. For practical convenience, amplitude calibration is usually based on previously recorded calibrations for a soft rectangular pulse (as described below), from which an estimate of the required power change is calculated. Table 9.3 also summarises the necessary changes in transmitter attenuation for various envelopes of equivalent duration, with the more elaborate pulse shapes invariably requiring increased rf peak amplitudes (decreased attenuation of transmitter output). 

The simplest, most robust and most widely used technique is presaturation of the solvent [62]. This is simple to implement, may be readily added to existing experiments and leaves (non-exchangeable) resonances away from the presaturation frequency unperturbed. It involves the application of continuous, weak rf irradiation at the solvent frequency prior to excitation and acquisition (Fig. 9.23a), rendering the solvent spins saturated and therefore unobservable (Fig. 9.24). Invariably resonances close to the solvent frequency also experience some loss in intensity, with weaker irradiation leading to less spillover but reduced saturation of the solvent. Longer presaturation periods improve the suppression at the expense of extended experiments so a compromise is required and typically 1-3 s are used trial and error usually represents the best approach to optimisation. Wherever possible the same rf channel should be used for both the presaturation and subsequent proton pulsing, with appropriate transmitter power switching. 

In summary, quadrature detection is an extremely advantageous detection scheme for FT NMR and should be incorporated as a standard item in any FT spectrometer. Without it, 1) the transmitter power is used inefficiently leading to larger frequency dependent tip angles, 2) less than optimum S/N, the amount of which depends on the method of implementation and 3) inconvenience in doing experiments requiring irradiation of a part of the spectrum such as selective saturation experiments. 

Note that after spin-locking the protons, the proton transmitter is simply left on. During the times that the carbon transmitter is on, the two rf fields satisfy the Hartmann-Hahn condition and the two different kinds of spin reservoirs are in thermal contact as described earlier. When the carbon transmitter is off, the carbon FID s are being recorded and the proton transmitter must be left on so that it can perform as a decoupler. [The amplitude of the spin-locking "pulse" is reduced compared to the initial n/2 pulse in order to keep the transmitter power level at a reasonable value. ]... 

---