Pradeep Bhattad, Chemical Engineering, Louisiana State University, 110 Chemical Engineering, South Stadium Road, Baton Rouge, LA 70803, Karsten E. Thompson, Gordon & Mary Cain Department of Chemical Engineering, Louisiana State University, 110 Chemical Engineering, South Stadium Rd, Baton Rouge, LA 70803, and Clinton S. Willson, Department of Civil & Environmental Engineering, Louisiana State University, Baton Rouge, LA 70803.
Fluid-fluid interfaces play an important role in mass transfer and multiphase flow in many chemical and environmental engineering processes. However, quantitative characterization of interfaces can be difficult, especially in complex geometries such as porous materials. We use X-ray computed tomography to image interfaces in a variety of different porous materials. Quantitative characterization of the interfaces is performed in a two-step process: Starting from a voxelized image of a fluid-fluid interface, we generate an initial surface triangulation using the signed distance function calculated from the segmented images. Since the real fluid interfaces are smooth in nature we use energy minimization to smooth the triangulated surface. The energy minimization consists of two parts, an elastic energy function that minimizes the surface area, and an image energy function to ensure that the triangulated interface respects the original voxelized image. The resulting smooth triangulated surface is a powerful tool for characterizing or modeling interfacial behavior. We report the accuracy of interfacial curvature and surface area calculations as a function of image resolution and initial surface mesh resolution for both artificially generated and real tomography images.