ALBERT

Description: 〈span〉〈div〉Summary〈/div〉An integral component of the surface nuclear magnetic resonance forward model involves predicting the magnitude of the transverse magnetization following excitation. To predict the transverse magnetization the Bloch equation must be solved. Traditional surface NMR forward models solve a simplified version of the Bloch equation where the relaxation terms are neglected. A shortcoming of this approach is that it can struggle to accurately describe the impact of relaxation during pulse effects. To address this concern, an alternative forward model based on solution of the full Bloch equation is proposed. The advantage of the proposed scheme is that it implicitly accounts for relaxation during pulse effects, increases the flexibility to implement alternative parameterizations of the inverse model, and can readily describe an arbitrary excitation protocol given that it no longer requires closed form expressions of the transverse magnetizations. To demonstrate the potential of the updated forward modelling scheme, a novel approach for the inversion of complex-valued FID data is presented. The inverse model is reparametrized in order to produce depth profiles of the water content, T〈sub〉2〈/sub〉* and T〈sub〉2〈/sub〉. This approach has great potential to enhance the ability of FID measurements to provide insights into pore size and permeability as it can provide direct sensitivity to T〈sub〉2〈/sub〉. In contrast, traditional approaches that employ a forward model based on the simplified Bloch equation and estimate only T〈sub〉2〈/sub〉* are plagued by uncertainty surrounding the link between T〈sub〉2〈/sub〉* and pore size/permeability. Synthetic and field results are presented to demonstrate the feasibility of the proposed forward model and FID inversion framework.〈/span〉