Experimental investigation of effects of mica content, Fe and pressure on the pore size distribution and permeability of sandy sediment using proton nuclear magnetic resonance

Abstract

Sediment permeability, an important factor for evaluating groundwater flow in slope and/or ground and oil and natural gas reservoirs in sandy sediments, can be estimated using transverse magnetic relaxation time (T2) measurements by 1H nuclear magnetic resonance (NMR). However, the performance of this method is influenced by the mineralogy and the pore and grain size distributions of the tested material. Therefore, NMR estimations for sandy sediments must first be validated against flow-based measurements. A good relationship between the mercury porosimetry (DHG) and T2 peaks was obtained for samples with different mica content (0–10 wt%). However, the faster T2 relaxation and higher surface relaxivity coefficient depended on the mica content with increasing Fe concentration for the sand-mica mixtures (mica content >20 wt%). The measurement permeability results were compared with estimates from NMR spectra using the Schlumberger Doll Research and Timur–Coates models. While both models provided estimates consistent with flow measurements at 0.1 MPa, the permeabilities were overestimated by 1–2 orders of magnitude at 1.0 MPa for the samples of mica content 20–50 wt%. The results show that the permeability estimated by NMR in mica-rich sediments is overestimated at greater pressure (at approximately 100 m depth).