A Study of the Dual-Water Model Based on Log Data

Abstract

Summary This paper presents log evidence in support of the dual-water model. Included are basic studies that illustrate the model from the aspect of conductivity measurements, porosity and formation factor relationship, and the contrast to the prediction of Waxman and Smits' cation exchange capacity (CEC) model (W-S model). Introduction Since the introduction of the theoretical dual-water model in 1977, there has been considerable conjecture on the scientific aspects of- the CEC models. This paper intentionally avoids theoretical areas and concentrates instead on what can be observed from log, data with respect to the conductivity and porosity of shales, shaly sands, and sands. In addition, the dualwater hypothesis is expounded as a means of shedding new light on the troublesome problem of applying a formation factor relationship in shaly formations. The Dual Water Observed Studies of factors that influence the attenuation of electromagnetic propagation signals from EPT logs revealed that plots of attenuation rate as a function of the total porosity index always displayed similar trends, with the shales showing the greatest uniqueness. Fig. 1, which is a hand composite of several wells from a variety of geographic areas with largely different conditions, illustrates this characteristic. Further study determined that the basic plot shape was a reflection of changes in the conductivity of the formation fluids. This is shown in Fig. 2, where constant fluid conductivity trend lines have been added to a typical attenuation/porosity plot. This knowledge led directly to the idea of correcting the EPT log t-or shaliness by modifying Rw, proportionally. Although this was done successfully, a question was raised. Do other logs that react to the volume and conductivity of formation fluids also respond to dual waters? Micro later o log and Induction Log Data The familiar Pickett plot (log-log plot of conductivity vs. porosity) is a convenient form for making this evaluation because it allows a relatively pure look at the data. Pickett plots using a density/neutron-based porosity were constructed with Micro later o log (Fig. 3) and induction (Fig. 4) log data.* Two distinct data concentrations could be seen, one associated with low gamma ray values and the other with high gamma ray values. These plots have been enhanced with an illustration of a simple Archie-type transform of some known fluid conductivities by assuming that KR=1.0 and m=2.0 in the classic equations, FR=KR/om and Cw=FRCo, (Ro,=FRRw). The Cw values chosen reflect Rw, (Cw=30.0 mho. Rmf (Cw=1.0 mho), and the theoretical clay water conductivities for the existing temperature conditions. JPT P. 158^