Model Synthetic Samples for Validation of NMR Signal Simulations

Abstract Simulations of nuclear magnetic resonance (NMR) signal from fluids contained in porous media (such as rock cores) need to account for both enhanced surface relaxation and the presence internal field gradients due susceptibility contrast between matrix fluid phase. Such simulations are typically focussed on extraction NMR T 2 distribution which can be related pore size indirectly system permeability. Discrepancies such digital cores associated experimental measurements however frequently reported; these generally attributed spatial variations matric composition resulting heterogeneously distributed relaxivities (?) gradients. To this end, a range synthetic sediments composed variable mixtures quartz garnet sands were studied. These two constituents selected following reasons: they have different densities allowing ready phase differentiation 3D ?CT images samples use simulation lattices distinctly ? values allow rigorous test simulations. Here used calculate samples, data then compared against corresponding measurements. Agreement was reasonably good with largest discrepancy being predicting weak (in vicinity sand mixed samples) not detected experimentally. The suite all publicly available development validation efforts.