Monitoring oil displacement processes with k‐t accelerated spin echo SPI

Abstract

Magnetic resonance imaging (MRI) is a robust tool to monitor oil displacement processes in porous media. Conventional MRI measurement times can be lengthy, which hinders monitoring time-dependent displacements. Knowledge of the oil and water microscopic distribution is important because their pore scale behavior reflects the oil trapping mechanisms. The oil and water pore scale distribution is reflected in the magnetic resonance T2 signal lifetime distribution. In this work, a pure phase-encoding MRI technique, spin echo SPI (SE-SPI), was employed to monitor oil displacement during water flooding and polymer flooding. A k-t acceleration method, with low-rank matrix completion, was employed to improve the temporal resolution of the SE-SPI MRI measurements. Comparison to conventional SE-SPI T2 mapping measurements revealed that the k-t accelerated measurement was more sensitive and provided higher-quality results. It was demonstrated that the k-t acceleration decreased the average measurement time from 66.7 to 20.3 min in this work. A perfluorinated oil, containing no (1) H, and H2 O brine were employed to distinguish oil and water phases in model flooding experiments. High-quality 1D water saturation profiles were acquired from the k-t accelerated SE-SPI measurements. Spatially and temporally resolved T2 distributions were extracted from the profile data. The shift in the (1) H T2 distribution of water in the pore space to longer lifetimes during water flooding and polymer flooding is consistent with increased water content in the pore space. Copyright © 2015 John Wiley & Sons, Ltd.

Polymer Flooding Enhanced Oil Recovery Evaluated with Magnetic Resonance Imaging and Relaxation Time Measurements

“…At a shear rate of 7 s –1 , the viscosity of the polymer solution was 130 mPas. Alcoflood 935 polymer has been widely employed for enhanced oil recovery in laboratory experiments and in field applications. ,− Control experiments and previous work showed that the contribution of 1 H in the polymer to the solution MR or MRI signal was negligible. , Brine and polymer solutions were degassed under vacuum to remove air before injection into the core plugs. A light mineral oil (Sigma-Aldrich, United States), largely nonpolar, and a crude oil (Stoney Creek Field, Canada) with polar components were employed.…”

“…16,31−33 Control experiments and previous work showed that the contribution of 1 H in the polymer to the solution MR or MRI signal was negligible. 22,24 Brine and polymer solutions were degassed under vacuum to remove air before injection into the core plugs. A light mineral oil (Sigma-Aldrich, United States), largely nonpolar, and a crude oil (Stoney Creek Field, Canada) with polar components 34 were employed.…”

“…Therefore, many T 2 distributions may be acquired while monitoring these core flooding processes. Spatially resolved T 2 distributions result in more data . The question then arises how best to analyze this data in a manner that reveals trends and permits comparison between experiments and samples.…”

“…Mitchell et al 23 applied MR pulsed field gradient diffusion measurements to investigate polymer flooding elastic turbulence mechanism in 3D porous structures. Li et al 24 employed k−t accelerated spin−echo single point imaging to monitor oil displacement processes and acquired spatially and temporally resolved T 2 relaxation distributions. A comprehensive review of MRI in rock core analysis has been provided by Mitchell et al 14 Wettability controls fluid flow behavior during flooding and is thought to affect oil recovery with polymer flooding.…”

“…Spatially resolved T 2 distributions result in more data. 24 The question then arises how best to analyze this data in a manner that reveals trends and permits comparison between experiments and samples. Plotting multiple distributions and attempting to describe changes in shape and/or peak position is clearly inadequate.…”