Pressure Drop Predictions in Microfibrous Materials Using Computational Fluid Dynamics

Abstract

Three-dimensional computational fluid dynamics simulations are performed for the flow of air through microfibrous materials for void fractions of 0.41 and 0.47 and face velocities ranging between 0.04ms and 1.29m∕s. The microfibrous materials consist of activated carbon powder with diameters of 137×10−6m entrapped in a matrix of cylindrical fibers with diameters of 8×10−6m. These sintered microfibrous materials are a new class of patented materials with properties that are advantageous compared to traditional packed beds or monoliths. Microfibrous materials have demonstrated enhanced heat and mass transfer compared to packed beds of particles of similar dimensions. In this paper, the simulations are used to predict the pressure drop per unit length through the materials and to analyze the details of the flow that are difficult to interrogate experimentally. Various geometric approximations are employed in order to allow the simulations to be performed in an efficient manner. The Knudsen number, defined as the ratio of the mean free path between molecular collisions to the fiber diameter, is 0.011; thus, velocity-slip boundary conditions are employed and shown to have only a minor effect on the pressure drop predictions. Significant effort is made to estimate numerical errors associated with the discretization process, and these errors are shown to be negligible (less than 3%). The computational predictions for pressure drop are compared to available experimental data as well as to two theory-based correlations: Ergun’s equation and the porous media permeability equation. The agreement between the simulations and the experiments is within 30% and is reasonable considering the significant geometric approximations employed. The errors in the simulations and correlations with respect to experimental data exhibit the same trend with face velocity for both void fractions. This consistent trend suggests the presence of experimental bias errors that correlate with the face velocity. The simulations generally underpredict the experimental pressure drop for the low void fraction case and overpredict the experimental pressure drop for the high void fraction case.