Simplified Production Method for PEM Fuel Cells

MPL-GDLThe heart of PEM fuel cells is the membrane electrode assembly (MEA), which has so far been produced by using only polymer electrolyte membranes. The manufacture of these membranes, however, is highly complex and expensive, limiting MEA production to a few companies around the globe. An innovative method discovered by researchers from the Department of Microsystems Technology (IMTEK) at the University of Freiburg makes the membrane process technology obsolete: A polymer electrolyte solution is directly processed between electrode and membrane. This technology is not only easy to employ and inexpensive, but it increases fuel cell performance compared to currently commercially available technologies. The extremely simplified process for PEM fuel cell production earned the MEMS Applications research team the f-cell award research & development in Stuttgart.

Fuel cell performance is highly influenced by the functional characteristics of the individual layers, their thickness and even their borders (see figure 2). All these performance factors will require optimization if one wants to produce a commercially viable fuel cell in the near future. A focus area of research into this field has always been the catalyst layer based on platinum. Ideally, it should display excellent electric and proton conductivity, create a high specific surface and contain as little as possible of the expensive and precious metal.

Hydrogenics

Another component moving into the focus of research is the polymer electrolyte membrane. Here, efforts are concentrated mainly on synthesizing new polymers with high chemical and mechanical stabilities combined with excellent proton conductivity. The border areas of the catalyst layer and membrane, however, have so far received not more than a little attention, although they harbor great potential for reducing the electric resistance of the cell. And despite the great cost saving opportunities, innovative methods to manufacture this tripartite layer combination have sparked little research efforts up until now.

CCM-MCEFig. 2: MEA of a low-temperature fuel cell with anode, cathode and proton exchange membrane as well as the related electrochemical reaction. A: Catalyst coated membrane (CCM, conventional technology): The catalyst layers are applied onto a relatively thick foil-type membrane. B: Recently presented membrane coated electrode (MCE): Thin membrane layers are applied directly to the catalyst layers of anode and cathode.

Development at IMTEK
These were exactly the two factors considered by the prized invention from the Freiburg scientists: In April 2015, they published a method to produce MEAs with optimized border areas of catalyst layer and membrane. The highlight: The entire MEA can be produced using only a single device.

Through ink jet print, a polymer electrolyte dispersion is printed in liquid form directly onto the catalyst layers of a fuel cell. The ink jet print can also process catalyst layers and microporous gas diffusion layers, so that every functional layer of the MEA can be processed with the same device. Thanks to the initial liquid state, the presented method creates a perfect link between the printed membrane layer and the catalyst layers. The liquid electrolyte enters the porous catalyst layer at up to a few hundred nanometers before it hardens, which strongly decreases the proton contact resistance of both. The direct deposition of the membrane on the electrode also allows for extremely thin membrane layers, which further contributes to reducing proton resistance.

The combination of the two effects has resulted in a fuel cell with extraordinary high proton conductivity, leading to power densities of above 4 W/cm² (300 kPa of absolute pressure, including H2 & O2 as combustion gases). Another benefit of the thin membrane was evidenced by a favorable self-humidification of the cell. Measurements showed less than a ten per cent reduction in performance between entirely wet and dry reactive gases, which renders the use of active humidifiers unnecessary.

Simplified production
Besides the high performance, removal of the conventional foil-type membrane offers another advantage: simplicity. Future mass production could entail spraying the microporous gas diffusion layer, the catalyst layer and finally the membrane subsequently and directly onto the gas diffusion layer as a substrate (see figure 1): Drawing showing the mass production of fuel cell half-cells: Subsequently, the microporous layer (MPL), the catalyst layer (CL) and the membrane (PEM) are sprayed in form of a substrate onto a gas diffusion layer (GDL). This process saves costs and time during production. The fuel cell can be produced by combining any of the half cells available.

Spray coating has already been a large-scale industrial method for manufacturing gas diffusion electrodes. Since the spraying process for the diffusion and the catalyst layer has already been state of the art, it would now need to be complemented only by spraying the membrane onto the fuel cell as well. The researchers from Freiburg have already tested and proven the feasibility of such a process. Together with cooperation partners from the Canadian Simon Fraser University, a first trial shows that an entirely spray-coated fuel cell shows competitive power densities in comparison with current processes (see figure 3).

graph_performance_characteristics_engFig. 3: Performance characteristics of an entirely spray-coated fuel cell with a cathode loading of 0.3 mgPt/cm².

The use of the reactive gases hydrogen and oxygen resulted in a maximum power density of 1.5 W/cm² (at 80°C, 92% relative humidity, atmospheric pressure). Additionally, it showed that entirely spray-coated cells can be manufactured with large surfaces in mind, meaning a transfer to industrial throughput is possible. The new manufacturing method would also no longer require any knowledge about foil-type membranes. Instead the importance would be on expertise in the production of functional dispersions, suspensions or solutions as well as their processing as ink for coating systems, such as spray coaters or screen printers. This means that the new technique is especially interesting to all those companies which have know-how in one of these areas and want to establish a presence in the fuel cell market as lateral entrants.

Room for novelties
This invention can also be the foundation of other innovative approaches in manufacturing composite membranes (e.g., use in mid-temperature fuel cells up to 120°C). The layer structure of the fuel cell enables the simple integration of reinforcing elements into the membrane layer, which manufacturers had so far only been able to do by putting a lot of effort into it (see figure 4).

Standard MEAFig. 4: Standard MEA without reinforcement; B: Fiber-reinforced MEA

An obvious application would be the addition of colloidal nanoparticles to the membrane ink before processing. This method has already been in use in foil-type membrane production, but can as well be applied to the new technology. At IMTEK, research showed that TiO2 nanoparticles enable the fuel cell to operate at higher temperatures without a significant loss of performance.

Another forward-looking method to reinforce the membrane has already been under development: A stable fiber compound consisting of nanofibers (e.g., known from bullet-proof vests or cut-proof pants) is planned to be incorporated into the membrane through electric spinning, in order to stabilize them mechanically (see figure 4B). This is especially noteworthy in the case of processing polymer electrolytes, which do have very high proton conductivity but rather low mechanical stability on their own. In the future, the process of creating powerful and durable composite membranes could be a very inexpensive one.

There are plans for future projects on the aging and upscaling of the new fuel cells. Here, the questions posed would revolve around the durability and transfer at stack level.

This research was subsidized by the BMBF in the framework of the GECKO project (Subsidy ID: 03SF0454C).

Authors:
Dr. Simon Thiele, Matthias Klingele, Matthias Breitwieser, Roland Zengerle
All from IMTEK, University of Freiburg

Leave a Reply