Multi-scale structural analysis of gas diffusion layers
Multi-scale structural analysis of gas diffusion layers
复制标题
DOI：
10.1016/j.jpowsour.2017.03.086
复制
发表时间：
2017-07-01
影响因子：
9.2
通讯作者：
Schladitz, Katja
中科院分区：
文献类型：
--
作者：
Goebel, Martin;Godehardt, Michael;Schladitz, Katja
材料的宏观性能在很大程度上取决于其微观结构。在这里，考虑用于燃料电池的气体扩散层（GDL）的传输特性。为了模拟流动和热性能，详细的微观结构信息是必不可少的。使用同步辐射和扫描电子显微镜（SEM）结合聚焦离子束（FIB）连续切片的高分辨率计算机断层扫描获得的3D图像。一个最近的方法，从FIB-SEM图像和复杂的形态图像变换的多孔结构的重建被应用到分割固体结构组件。描述和讨论了用于分割断层数据集中的不同分量的基本算法步骤。在本文中，两种类型的GDL，基于非织造基板层和纸基板层被认为是，分别。在同步辐射计算机断层扫描数据内分离出三个以上的分量。也就是说，纤维系统、聚四氟乙烯（PTFE）粘合剂/浸渍、微孔层（MPL）、微孔层内的内含物和孔隙空间被分段。由此导出的3D结构数据在不同的模拟应用中的使用可以被证明。根据GDL的淹没状态，可以模拟宏观特性，如热导率。(C)2017爱思唯尔B. V.保留所有权利。
The macroscopic properties of materials are strongly determined by their micro structure. Here, transport properties of gas diffusion layers (GDL) for fuel cells are considered. In order to simulate flow and thermal properties, detailed micro structural information is essential. 3D images obtained by high-resolution computed tomography using synchrotron radiation and scanning electron microscopy (SEM) combined with focused ion beam (FIB) serial slicing were used. A recent method for reconstruction of porous structures from FIB-SEM images and sophisticated morphological image transformations were applied to segment the solid structural components. The essential algorithmic steps for segmenting the different components in the tomographic data-sets are described and discussed. In this paper, two types of GDL, based on a non-woven substrate layer and a paper substrate layer were considered, respectively. More than three components are separated within the synchrotron radiation computed tomography data. That is, fiber system, polytetrafluoroethylene (PTFE) binder/impregnation, micro porous layer (MPL), inclusions within the latter, and pore space are segmented. The usage of the thus derived 3D structure data in different simulation applications can be demonstrated. Simulations of macroscopic properties such as thermal conductivity, depending on the flooding state of the GDL are possible. (C) 2017 Elsevier B.V. All rights reserved.