Characterization of Filter Cake Generated by Water-Based Drilling Fluids Using CT Scan

Filter-cake characterization is very important in drilling and completion operations. The homogeneity of the filter cake affects the properties of the filtration process such as the volume of filtrate, the thickness of the filter cake, and the best method to remove it. Various models were used to determine the thickness and permeability of the filter cake. Most of these models assumed that the filter cake was homogeneous. The present study shows that the filter cake is not homogeneous, and consists of two layers of different properties. The objective of this study is to measure the filter-cake thickness and permeability of water-based drilling fluids by a new approach and compare the results with previous models. A highpressure/high-temperature (HP/HT) filter press was used to perform the filtration process under static conditions (225 F and 300 psi). A computed-tomography (CT) scan was used to measure the thickness and porosity of the filter cake. Scanning electron microscopy (SEM) was used to provide the morphology of the filter cake. The results obtained from the CT scan showed that the filter cake was heterogeneous and contained two layers with different properties under static and dynamic conditions. Under static conditions, the layer close to the rock surface had a 0.06-in. thickness, 10to 20-vol% porosity, and 0.087-ld permeability, while under dynamic conditions, this layer had a 0.04-in. thickness, 15-vol% porosity, and 0.068-ld permeability. The layer close to the drilling fluid had a 0.1-in. and 0.07-in. thickness under static and dynamic conditions, respectively, and it had zero porosity and permeability after 30 minutes under static and dynamic conditions. SEM results showed that the two layers contained large and small particles, but there was extremely poor sorting in the layer, that was close to the drilling fluid, which led to zero porosity in this layer. Previous models underestimated the thickness of the filter cake by almost 50%. A new method was developed to measure the thickness of the filter cake, and various models were screened to identify the best model that can predict our permeability measurements.