ABSTRACT
Global warming by greenhouse gas emissions is one of the main threats of our modern society, and efficient CO2 capture processes are needed to solve this problem. Membrane separation processes have been identified among the most promising technologies for CO2 capture, and these require the development of highly efficient membrane materials which, in turn, requires detailed understanding of their operation mechanism. In the last decades, molecular modeling studies have become an extremely powerful tool to understand and anticipate the gas transport properties of polymeric membranes. This work presents a study on the correlation of the structural features of different membrane materials, analyzed by means of molecular dynamics simulation, and their gas diffusivity/selectivity. We propose a simplified method to determine the void size distribution via an automatic image recognition tool, along with a consolidated Connolly probe sensing of space, without the need of demanding computational procedures. Based on a picture of the void shape and width, automatic image recognition tests the dimensions of the void elements, reducing them to ellipses. Comparison of the minor axis of the obtained ellipses with the diameters of the gases yields a qualitative estimation of non-accessible paths in the geometrical arrangement of polymeric chains. A second tool, the Connolly probe sensing of space, gives more details on the complexity of voids. The combination of the two proposed tools can be used for a qualitative and rapid screening of material models and for an estimation of the trend in their diffusivity selectivity. The main differences in the structural features of three different classes of polymers are investigated in this work (glassy polymers, superglassy perfluoropolymers and high free volume polymers of intrinsic microporosity), and the results show how the proposed computationally less demanding analysis can be linked with their selectivities.
ABSTRACT
Objective To investigate the effects of element size and type, material property distributions of vertebral cancellous bone and simulation methods of cortical bone structure on the finite element (FE) results during the finite element modeling of lumbar vertebral body. Methods Based on QCT images of lumbar spine, 22 FE models of L2 without posterior structure were built by 6 element sizes (0.5, 1.0, 1.5, 2.0, 2.5, 3.0 mm), 2 heterogeneous material distribution methods of cancellous bone (300, 150) and 2 cortical bone modeling methods. The maximum displacement, strain energy, average stress and axial stiffness of these models were obtained to analyze and verify the results. Results When the element size was 0.5 mm, the axial stiffness of models with 10, 150 and 300 kinds of heterogeneous materials showed obvious differences; for the vertebral cancellous bone with 150 kinds of materials, the variation of average stress was not distinct under different element sizes; the average stress of the model using the outermost hexahedral elements to simulate the cortical bone structure was larger than that appending the skin to the outmost of the model. Conclusions It is more reasonable and effective to build the FE model of lumbar vertebral body with the method by 0.5 mm element size, 8-noded hexahedral elements, 150 kinds of heterogeneous materials, and using the outermost hexahedral elements to simulate the cortical bone structure. The research findings will lay a foundation for building subject-specific FE models of lumbar vertebral body on a large scale in future.
