The abnormally high conductivity and relative permittivity of graphite-water-glass beads mixtures at low frequencies has been documented. The abnormally large conductivity has been attributed to an interfacial polarization phenomenon that describes the dispersive losses caused by the migration of a charge buildup in the presence of the applied low-frequency electromagnetic field. The charge buildup is created in a layer between the graphite particle and saline connate water and occurs because of the discontinuity of the potential between the ionic charge transportation in the connate water and the electronic transport in the graphite particle in the presence of an induced low-frequency electromagnetic field.
The interfacial polarization losses create an abnormally large dielectric effect on induction measurements. These dielectric losses themselves are complex which results in the imaginary component of the dielectric constant to be in phase with the real component of the conductivity. This imaginary component is given the term "dielectric loss factor". Dielectric loss factors are impossible to measure directly since they are indistinguishable from the real part of the conductivity. They are problematic to model with petrophysical parameters because of the difficulty in measuring them.
We use a unique approach to find the value for the dielectric loss factor. We take the same lab mixtures that were measured with the low-frequency tri-axial induction electromagnetic field and measure them with a four-electrode cell capable of measuring a wide range of frequencies spanning six orders of magnitude. Since the real part of the conductivity is not influenced by frequency, the difference in the lowest frequency and higher frequency electromagnetic lab measurements is attributed to the dielectric loss factor.
As a result of these new experiments, we can characterize the dielectric loss factor in terms of the lab measurement parameters: frequency, real portion of the induction relative dielectric constant, and salinity. This enables us to determine the relationship between dielectric loss factor and the petrophysical properties normally used in petrophysics for any given measurement frequency. We have named this relationship the Petrophysical Dielectric Loss Model (PDLM).
When logging measurements are available, the workflow is reversed. The necessary petrophysical parameters and induction dielectric constant are computed from the field logs and the dielectric loss factor is calculated using the new PDLM. This dielectric loss factor approaches 10 Siemens/m in some unconventional wells. The measured induction is corrected for dielectric effects then the dielectric loss factor is subtracted from this value. The result is a dielectric and dielectric loss corrected induction measurement. This conductivity measurement is then free from the graphitic kerogen induced conductivity effects and allows this measured induction log to be used in petrophysical calculations to accurately calculate water saturations in organic mudrocks. In summary, we have removed the parasitic graphite effect and returned the formation to an Archie like rook.