New fuel cells by using additive manufacturing

Motivation

The exhaustion of fossil energy sources such as natural gas, crude oil and coal as well as the danger of climate change due to CO2 emissions are pushing for alternative approaches to energy production. In addition to renewable energies, fuel cells offer a solution by converting the chemical energy bound in hydrogen into electricity and heat with a high degree of efficiency. The porous electrode on which the chemical reactions take place represents a challenge in terms of production technology, as it must have sufficient mechanical stability in addition to a defined porosity and electrical conductivity.

The generative manufacturing process “selective laser melting” (SLM) offers the possibility through an individual adaptation of the process parameters to combine porous as well as gas-tight and mechanically stable areas in only one work step.

Fuel cell electrode produced by SLM, right: Schematic representation of the SLM process.
Fuel cell electrode produced by SLM, right: Schematic representation of the SLM process.

Objective and procedure

Within the framework of the DFG-funded project “Selective laser melting of thin-walled porous electrodes for novel fuel cells”, a knowledge base for modelling and manufacturing novel tubular fuel cells (see picture) is being created in close cooperation with the Institute of Thermal Process Engineering (TVT). To achieve this, the effects of operating conditions such as temperature and pressure on the performance behaviour are first predicted with the help of numerical simulation. This knowledge in turn enables a definition of the porosity requirements. A realisation of the porous properties in the real component is carried out within the framework of statistically supported experiments. Here, the influence of the laser power, the speed of the laser beam, the distance between the individual scan paths and the thickness of a powder layer on the porosity is researched and transferred into an empirical model. The porosity can thus be specifically adjusted according to given requirements without time-consuming experiments.