Relationships between Non-wetting Phase Invasion and Magnetization Evolution in Connected Pore Systems as Revealed by Network Simulation

Pore network models are used to relate fundamental pore structure parameters (pore body and pore throat size distributions) to the distribution of water at the pore level and the decay of proton magnetization under conditions of primary drainage in water-wet rocks. The simulations reveal conditions under which diffusive coupling between pores has a significant effect on the decay spectra and provide insight into the estimation of primary drainage capillary pressure curves from NMR T2 distributions. The conversion factor κ = Pc T2 required to map a NMR cumulative T2 distribution at Sw = 100% onto a primary drainage capillary pressure curve furnishes important information about the pore-tothroat size aspect ratio, a parameter that critically affects recovery efficiency by waterflooding in water-wet media. INTRODUCTION With the advent of NMR wireline logging tools, the feasibility of estimating primary drainage capillary pressure curves from the NMR T2 distributions of water-saturated rock samples has recently attracted significant attention (e.g., Volokitin et al., 1999). In a NMR measurement, hydrogen atoms on water molecules probe the pore space by diffusive motion. The magnetic moment carried by these atoms relaxes due to dipolar interactions with other molecules and with the fluid-solid interface. For a single pore in the fast diffusion limit, the surface relaxation rate is slow compared to the rate of magnetization equilibration by diffusive motion (Brownstein and Tarr, 1979). Consequently, magnetization is approximately uniform across the pore and decays with a characteristic time T2 that is inversely proportional to the pore’s surface area-to-volume ratio: