Fuel cells and batteries depend on the transport of ions through a porous medium. The study of the 3D structure of these porous media is not straightforward. In general, a compromise must be made between resolution (which defines the smallest pore visible) and Field of View. There are different ways to create a 3D image of the porous medium; each with specific advantages and drawbacks.

MicroCT uses the absorption of X-rays to create a 3D reconstruction. The Field of View is large enough to be representative with regard to the grain sizes (1). However, the submicron porosity is typically not well captured by microCT because of its limited resolution.

The FIBSEM technique which uses a Galium ion beam for sectioning and an electron beam for imaging has high enough resolution to resolve all porosity. However, the Field of View is typically of the order of 10 micrometer; which is not big enough to create a representative model of the pore space.

In this study we have used a plasmabeam for sectioning and an electron beam for imaging. The plasmabeam typically removes a layer of 100 nm followed by imaging by the electron beam. The material removal rate with the plasmabeam is magnitudes higher than with the Ga focused ion beam, allowing to expand the field of view to hundreds of micrometers.

A 3D reconstruction of the pore space (Field of View 100 micrometer) make it possible to calculate parameters such as the tortuosity on a representative volume.

Segmentation of the pore space is not always straightforward due to minimal grey level differences between the different phases. Also, it is not uncommon to see details from layer N+1 through the pores in layer N. This so-called pore-back effect must be taken into account otherwise the permeability will be greatly underestimated.

1. Cai et al, Electrochimica Acta 56 (2011) 5804–5814