RELAXATION DURING PULSE EFFECTS FOR ADIABATIC PULSES IN SURFACE HMR

摘要

To improve the signal to noise ratio of the surface nuclear magnetic resonance (NMR) technique recent work has focused on the development of an alternative transmit scheme employing an adiabatic excitation pulse (Grunewald et al., 2016). The use an adiabatic pulse requires that the excitation modeling in the standard forward model be altered. One factor impacting the accuracy of excitation modeling is relaxation during pulse (RDP) effects. A robust approach to deal with RDP has been developed in the context of an on-resonance pulse (Walbrecker et al., 2009), but in order to ensure reliable water content estimates we must also understand how RDP impacts an adiabatic pulse. A sensitivity analysis is presented to investigate how RDP effects an adiabatic pulse and whether the existing approach to accommodate for RDP is effective for an adiabatic pulse. The magnitude of RDP during an adiabatic pulse is observed to depend strongly on the underlying relaxation mechanism controlling T2* (compared to an on-resonance pulse); i.e. whether T2* is controlled by T2 processes (such as surface or bulk relaxation) or by background magnetic field inhomogeneity related dephasing. Each scenario leads to very different magnitude transverse magnetizations. The challenge is that given only free induction decay measurements (the standard measurement in surface NMR) the two scenarios cannot be differentiated. As such, water content estimates produced using adiabatic pulses may correspond to larger uncertainties due to potential modeling errors associated with incomplete knowledge about which scenario is present. The standard approach to compensate for relaxation during pulse effects is demonstrated to produce biased results using adiabatic pulses in the fast to intermediate T2* range (<~few 100 ms).