Imaging in electrically conductive porous media without frequency encoding

Understanding multi-phase fluid flow and transport processes under various pressure, temperature, and salinity conditions is a key feature in many remote monitoring applications, such as long-term storage of carbon dioxide (CO(2)) or nuclear waste in geological formations. We propose a low-field NMR tomographic method to non-invasively image the water-content distribution in electrically conductive formations in relatively large-scale experiments (∼1 m(3) sample volumes). Operating in the weak magnetic field of Earth entails low Larmor frequencies at which electromagnetic fields can penetrate electrically conductive material. The low signal strengths associated with NMR in Earth's field are enhanced by pre-polarization before signal recording. To localize the origin of the NMR signal in the sample region we do not employ magnetic field gradients, as is done in conventional NMR imaging, because they can be difficult to control in the large sample volumes that we are concerned with, and may be biased by magnetic materials in the sample. Instead, we utilize the spatially dependent inhomogeneity of fields generated by surface coils that are installed around the sample volume. This relatively simple setup makes the instrument inexpensive and mobile (it can be potentially installed in remote locations outside of a laboratory), while allowing spatial resolution of the order of 10 cm. We demonstrate the general feasibility of our approach in a simulated CO(2) injection experiment, where we locate and quantify the drop in water content following gas injection into a water-saturated cylindrical sample of 0.45 m radius and 0.9 m height. Our setup comprises four surface coils and an array consisting of three volume coils surrounding the sample. The proposed tomographic NMR methodology provides a more direct estimate of fluid content and properties than can be achieved with acoustic or electromagnetic methods alone. Therefore, we expect that our proposed method is relevant for geophysical applications, such as for monitoring CO(2) injections in saline aquifers or detecting water leakage into nuclear waste deposit sites installed in electrically conductive formations.