Frequency selective rf pulse

Slice selection is accomplished by applying a frequency-selective rf pulse in the presence of a gradient. 

Figure 1 ISIS. In eight separate acquisitions combinations of up to three frequency-selective RF pulses invert the longitudinal magnetization in three orthogonal slices prior to a nonselective excitation pulse and collection of the free induction decay. The inversion pulses modify the resultant phase of the transverse magnetization existing after excitation. For three-dimensional localization, the inversion pulses are applied and the data added or subtracted according to the protocol in Table 1.

Figure 4 The chemical shift displacement artifact. (A) An applied linear magnetic gradient encodes spatial position in the resonant frequencies of two particular spectrum peaks (represented by the sloping lines). Peak 1 has a different chemical shift to peak 2. A selective RF pulse, centered on frequency and with bandwidth A/, will excite a slice at a different position for each peak as shown. (B) Increasing the strength of the linear magnetic gradient reduces the difference in slice position - however, to achieve slices with the same spatial width as in (A), a larger bandwidth RF pulse must be used.

The opposite extreme regime, when co coq, involves the so-called soft or selective RF pulses. In this case, pulses of long duration and low power can be used to excite just one of the transitions, which is achieved by a suitable choice of the resonance offset (72) and also of the pulse length. The main aspect to be considered here is that each transition has associated with it a specific frequency and a different effective nutation fi equency, which depends on the values of I and m. For selective excitation of a transition between the levels m + and m, it is found that the ideal soft jr/2 pulse must be shorter than the corresponding hard itjl pulse (which excites all transitions) by a factor of -3-1) —m m 1). For excitation of just the central transition (1 /2 -1 /2) for half-integer spin nuclei, the... 

Figure 2 RF and magnetic field gradient pulse sequence used in NMR microscopy along with the associated trajectory through Ir-space. The frequency-selective 180° pulse is used to select a layer of spins (the slice plane) to participate in the spin echo and hence to contribute to the image. NT=acquisition time TE=/echo time.

A 2D NMR experiment involves a selection of pulses, delays, frequencies, RF phases and amplitudes,... 

Siegel et al. showed that enhancement of the CT can also be obtained using hyperbolic secant (HS) pulses to invert selectively the STs [74], Unlike the DFS waveform, whose frequency sweep is generated by a constant rf-pulse phase while modulating the amplitude, the HS pulse utilizes both amplitude and phase modulation, yielding an enhancement exceeding that obtained by DFS or RAPT [61, 74, 75]. Most recently, the pulse sequence called wideband uniform-rate smooth truncation (WURST) [76] was introduced to achieve selective adiabatic inversion using a lower power of the rf-field than that required for the HS pulses [77,78]. One of its applications involved more efficient detection of insensitive nuclei, such as 33S [79]. 

Fig. 5. Pulse sequence for MR detection of vibration using a radiofrequency field gradient. A binomial 1331 radiofrequency pulse (pulse length D, interpulse delay r) is applied in-phase with the mechanical wave. Thus the vibration period 7V is equal to 4(D + r). The number of cycles can be increased to ensure a better frequency selectivity. The constant RF field gradient generated by a dedicated RF coil allows space encoding without using conventional static field gradients (from Ref. 16 with permission from Elsevier).

---