Magnetic resonance imaging (MRI) in the presence of metallic structures is very common in medical and non-medical fields. Metallic structures cause MRI image distortions by three mechanisms: (1) static field distortion through magnetic susceptibility mismatch, (2) eddy currents induced by switched magnetic field gradients and (3) radio frequency (RF) induced eddy currents. Single point ramped imaging with T1 enhancement (SPRITE) MRI measurements are largely immune to susceptibility and gradient induced eddy current artifacts. As a result, one can isolate the effects of metal objects on the RF field. The RF field affects both the excitation and detection of the magnetic resonance (MR) signal. This is challenging with conventional MRI methods, which cannot readily separate the three effects. RF induced MRI artifacts were investigated experimentally at 2.4 T by analyzing image distortions surrounding two geometrically identical metallic strips of aluminum and lead. The strips were immersed in agar gel doped with contrast agent and imaged employing the conical SPRITE sequence. B1 mapping with pure phase encode SPRITE was employed to measure the B1 field around the strips of metal. The strip geometry was chosen to mimic metal electrodes employed in electrochemistry studies. Simulations are employed to investigate the RF field induced eddy currents in the two metallic strips. The RF simulation results are in good agreement with experimental results. Experimental and simulation results show that the metal has a pronounced effect on the B1 distribution and B1 amplitude in the surrounding space. The electrical conductivity of the metal has a minimal effect.

Low salinity waterflooding (LSF) has been proposed to improve oil recovery, with major projects in progress worldwide. There is however no consensus on the mechanisms of LSF for enhanced oil recovery (EOR). Wettability change is the most widely accepted mechanism. In this work, magnetic resonance (MR) and magnetic resonance imaging (MRI) were employed to monitor oil displacement processes during model laboratory scale LSF experiments. The MR and MRI measurements permit evaluation of putative LSF mechanisms. Two clay-coated sand packs, one with nonswelling kaolinite, the other with swelling montmorillonite, were prepared as model porous media for LSF. The interactions between pore fluids (oil and water) and the clay-coated pore surfaces were evaluated with relaxation time measurements. A MRI methodology, spin echo single point imaging (SE-SPI), was employed to spatially resolve the T 2 distribution along the sand pack. The oil saturation profiles were determined from SE-SPI measurements. A new differential relaxation time distribution method is proposed in this work for oil saturation estimation. The pore fluid self-diffusion coefficients were measured. The mechanism of wettability change for LSF is suggested on the basis of the oil diffusion coefficient variation with LSF. The similarities and differences between the kaolinite and montmorillonite behaviors are discussed. This work demonstrates that MR and MRI are robust tools to monitor oil displacement processes, with the potential to reveal the mechanisms of LSF and other procedures for enhanced oil recovery.

No abstract

Magnetic resonance imaging (MRI) is increasingly employed as a core analysis technique by the oil and gas industry. In axial profiling of petroleum reservoir core samples and core plugs, the sample of interest may frequently be much longer than the natural field of view (FOV) defined by the radio frequency (RF) sensor and region of constant magnetic field gradient. Profiling such samples with a low field MRI will result in distorted, non-quantitative axial profiles near the edge of the FOV with data from outside the desired FOV folding back into the image, when the gradient magnetic field homogenity region is shorter than the region of RF excitation. The quality of MRI as a core analysis technique is increased if imaging can be performed on intact samples with the FOV reduced to the region of interest (ROI), either to increase the image resolution or to reduce the total time for imaging. A spatially selective adiabatic inversion pulse is applied in the presence of a slice selective magnetic field gradient to restrict the FOV to an ROI that is a small portion of a long sample. Slice selection is followed by a 1D centric-scan SPRITE measurement to yield an axial fluid density profile of the sample in the ROI. By employing adiabatic pulses, which are immune to RF field non-uniformities, it is possible to restrict the ROI to a region of homogeneous RF excitation, facilitating quantitative imaging. The method does not employ conventional selective excitation, but a subtraction based on images acquired with and without adiabatic inversion slice selection. The adiabatic slice selection lends itself to a selective T2 distribution measurement when a CPMG pulse sequence follows the slice selection. The inversion pulse selects a slice on the order of 1 cm at an arbitrary position. The local T2 distributions measured are of similar quality to bulk CPMG. This method is an alternative to MRI-based techniques for T2 mapping in short relaxation time samples in porous media when T2 is required to be measured at only a few positions along the sample, and a resolution of 1 cm is acceptable.

No abstract