Determining capillary-pressure curve, pore-size distribution, and permeability from induced polarization of shaley sand

Abstract

An appropriate form of induced polarization (IP) acts as a bridge between the structure of a water-saturated core plug and its transport properties. The induced-polarization decay curves of natural rocks can be modeled as a weighted superposition of exponential relaxations. A singular-value decomposition method makes it possible to transform the induced-polarization decay data of the shaley sands into relaxation-time spectrum, defined as plot of weight versus the relaxation-time constant. We measured the induced-polarization decay curves of core samples from a formation of Daqing oil field using a four-electrode method. The decay curves were transformed to relaxation-time spectra that were used to estimate the capillary-pressure curves, the pore-size distribution, and the permeability of the shaley sands. The results show that salinity ranges from [Formula: see text] have little effect on the IP relaxation-time spectra. A pseudocapillary pressure curve can be derived from the IP relaxation-time spectrum by matching the pseudocapillary curve with that from HgAir. The best-matching coefficients of the studied cores change slightly for the samples. Defined as the value of pressure at which the injected mercury saturation is 5%, entry pressures of the cores can be estimated well from IP-derived capillary-pressure curves. Pore-size distributions generated from induced polarization and mercury capillary-pressure curves are comparable. Permeability can be predicted from IP measurements in the form of [Formula: see text], where [Formula: see text] is the estimated permeability from IP relaxation spectrum in millidarcies (md), [Formula: see text] is the porosity in percentage, and [Formula: see text] is average time constant of IP relaxation-time spectra in millis (ms). The constants and exponents from various rock formations are slightly different.

A comparison between Gauss-Newton and Markov-chain Monte Carlo–based methods for inverting spectral induced-polarization data for Cole-Cole parameters

“…The different peaks reflect relaxation processes at different scales. Such behavior was also observed in time-domain measurements of induced polarization (Tong et al, 2006;Tarasov and Titov, 2007). …”

Dependence of spectral‐induced polarization response of sandstone on temperature and its relevance to permeability estimation

“…When taking b = 3.85, for instance, as reported by Binley et al [2005] and a = 0.90, the apparent permeability varies by more than 2 orders of magnitude in the temperature range from 0°C to 80°C. The temperature dependence of k a becomes significantly weaker when using lower values of b as reported, for example, by Tong et al [2006] and Zisser et al [2010]. Note that using a different exponent a (0.7 ≤ a ≤ 1) in equation (55) causes only minor variance of the computed k a (cf.…”

“…More details regarding implementation are given in Appendix B. Nonnegative constraints were imposed on j by applying a procedure similar to that reported by Tong et al [2006]. The discrete approximation of the RTD is obtained from the inverted j values by means of normalization:…”

“…Empirical relationships between hydraulic permeability k and t in the form of a power law were correspondingly reported [e.g., Binley et al, 2005;Tong et al, 2006;Zisser et al, 2010]:…”

The time dependence of rock healing as a universal relaxation process, a tutorial

“…In that approach techniques of linear inversion (Menke 1984;Tarantola 1987;Aster et al 2004) can be utilized to estimate the distribution of relaxation times P(τ ). Methods for estimating relaxation spectra are an essential part for interpreting measurements of induced polarization (Morgan & Lesmes 1994;Tong et al 2006;Tarasov & Titov 2007) and NMR (Whittall & MacKay 1989). In NMR the relaxation time may, in fact, depend on location (Blümich 2000).…”