The Link Between Measured NMR Parameters and Clay Content and Distribution

Clay content and distribution impact the hydraulic conductivity (K) of a sediment, with the distribution having a significant impact if the clay forms layers that act as hydraulic barriers. Nuclear Magnetic Resonance (NMR) is a geophysical method that can be used in the laboratory and in the field to obtain information about the geometry of the pore space in a water-saturated geological material. The question motivating this study: Can we extract from NMR data information about the clay distribution in a sample? We conducted laboratory NMR experiments on two sets of samples with mixed and layered sand and kaolinite clay distributions. For five different clay contents we prepared five layered samples and five homogeneously mixed samples; plus a pure sand and a pure clay sample. Relaxation time (T2) was measured for each of the samples; T2 is primarily governed by the surface-area-to-volume ratio (S/V) of the pores with the distribution that is observed represented by the mean log value, T_2ML. The pure sand sample had T_2ML=586 ms; the pure clay sample had T_2ML=77 ms. In the mixed samples' NMR relaxation spectra we see a single peak in T2 for each sample that moves from 586 ms (pure sand) to 148 ms in the 5% clay sample and stays relatively constant at about 80 ms for the rest of the samples. This is because the clay grains infilled all the sand porosity and thus the pore space geometry for the higher clay-content samples is that of the clay. In the layered samples' spectra we see two peaks in T2, corresponding to the T_2ML values of pure sand and pure clay. The relative signal strength of the sand peak decreases significantly as clay content increases. Because there were experimental challenges with clay entering the sand layers, the T_2ML values of these sand peaks ranged from 420 ms to 780 ms. For any clay content, the averaged T_2ML values of the layered sample is higher than that of the mixed sample due to the presence of the pure sand layer. The present study demonstrates that the differences observed in NMR data make it possible to differentiate between layered and mixed clay. In our data the percentage of clay could be estimated by the relative signal strength between the two peaks in the layered samples, but could be estimated only at low clay content in the mixed samples due to the way in which the clay filled in the pore space and thus dominated the NMR response.