Web Exclusive Article
Academic Focus : Long road ahead
Words: Nicola Bundschuh
A new detailed electrochemical approach and modeling strategy for solid oxide fuel cells has been developed that could lead to the vital breakthrough in fuel cell technology.
Fuel cells are a promising technology for electrical energy production in various application fields. For mobile applications, the automotive industry is working on differing power ranges to introduce fuel cell technology into vehicles.
As a replacement of the main engine, the vehicle would run with electricity generated by the fuel cell. But in addition to this, a replacement of the auxiliary power unit (APU) is needed to run the engine in an optimum regime and let the fuel cell produce as much electricity as needed in a modern car with electronic accessories. A possible augmentation of voltage is another advantage for the replacement of the APU. In larger trucks and commercial vehicles, the replacement of the APU is such that it already offers the use of systems such as air conditioning and other electric devices without need of running the engine.
In mobile applications, mass and volume of systems and fuel cells are as important as system efficiencies. Therefore the system design and the cell geometry have to be adapted to the specific application. For mobile applications such as vehicles, load changes are often demanded fast and sometimes of a big order. Generally, these demands are covered by battery or capacitor units in the system, but for larger amounts the fuel cell has to react fast. This implies fast load changes in the cell that essentially means short transport ways of electrons and a good gas supply to the reacting zones.
The longer the vehicle has to wait for the electrons to be supplied, the higher the cell internal ohmic resistance. If continuous gas supply to the whole cell surface can not be arranged, starvation zones will develop that cause a loss in immediate power production. If this occurs regularly, the end result will be a loss of cell power and an irreversible loss of performance of the cell called degradation.
As degradation zones can only be diagnosed post mortem, modeling can be used to not only get an insight in the cell performance, but also as a fast and efficient tool for cell geometry and flow field design optimization. The cell design is a compromise of short electron transport, which means a maximum contacting between the reaction zone and the bipolar plate. The knock-on effect is that the current collector and a maximum of free gas supply to the reaction zone results in broad gas channels in the bipolar plate.
A new electrochemical approach is integrated in the commercial CFD-code Star-CD by cd-adapco for evaluating the influence of gas the diffusion processes on the electrochemical conversion rate, and detecting starvation zones causing the degradation effects. This model development was based on measurements of the effective gas diffusion coefficients in the porous electrodes and support layers to monitor the real gas velocity levels to the active zones. The electrochemistry was gained by measurements with segmented cells and recording of the cell resistances, temperature and gas composition in each of the segments. Both the novel measurement devices are developed at the DLR.
The measurement of the segmented fuel cells is a common diagnostic tool for low temperature fuel cells such as PEM. For the SOFC, operating at high temperatures of 800°C cell segmentation is a challenging technology. Gas tight sealing between each of the sixteen segments are required as sealing to the environment around the input of the analysis devices. The cell resistances and polarizations are determined by impedance measurements both in open circuit and loaded mode. They are recorded for each segment independently and integrative for the whole fuel cell. In parallel to this development, the temperatures of the segments are measured with thermocouples at the segment surfaces, and the gas composition is determined via gas chromatography also at each segment.
From these measurements, an electrochemical approach with cell resistances dependent upon the temperature and gas species concentration is developed. Only the linear regime of the characteristic is regarded to eliminate the nonlinear effects of the diffusion polarization and the gas diffusion effects on the cell during measurements. A numerically fast and very stable electrochemical model was also received.
The effective gas diffusion coefficients through the porous anode and cathode are measured in a gas diffusion chamber positioned in a heating device. The gas supply allows for mixing of different fuel cell relevant gases in different concentration compositions. As a result, the measured gas diffusion coefficients are temperature, species and concentration dependent.
Implementing these effective diffusion coefficients in numerical codes lead to a high-resolution modeling approach. The regarded geometry consists of a SOFC segment of 4 x 4mm with an anode of 40µm and a cathode of 10µm thickness. The electrolyte has a thickness of 10µm and is regarded as a gas tight, solid layer. The thickness of the anodic and cathodic reaction layers are considered to be only 2µm flanking both sides of the electrolyte. By identifying symmetry planes in the gas channel and current collector, the ratio of computational complexity to calculation time has been brought to an acceptable level. Sensitivity studies with different inlet conditions have been performed in order to reflect different positions along the gas channel.
Polarization effects in the linear range of the U-I-characteristics are incorporated, namely ohmic resistances, cathodic and anodic activation polarizations. The diffusion polarization is determined by the locally resolved gas diffusion through the porous electrodes that are implemented by a discretised multilayer description.
The results show the influence of gas diffusion processes on the electrochemical conversion rate. By varying the current density, starvation effects have been observed both on oxidant and reactant side of the electrolyte depending on the bulk concentration each. The gas transport between current collector and electrolyte was identified as a limiting effect on the effective cell power.
Beside electrochemical conversion distribution, several other effects of locally differing temperature distribution and species concentration are shown in the model. As a result of this, the influence of diffusion effects on different stack designs, cell and layer structures are visualized. What’s important to note is that a shifting of the maximum conversion zone depending on inflow gas concentrations, temperature distribution and demanded power was found.
These results show that this model can be used as a tool for optimization of gas distribution structures of solid oxide fuel cells regarding the prevention of starvation and thereby caused degradation. Benchmarking
of different aspects in a development process like high fuel utilization, low temperature gradient levels or a homogenous power distribution across the reaction surface are other possible applications for this project.
Away from the automotive arena, even aircraft are regarded as a possible platform for fuel cell application. On the smaller scale, the ram air turbine (rat) has around 5kW for emergency power generation, but an APU replacement of about 100kW for electrical energy production on the ground would allow the aircraft to optimize the thrust of the turbines and therefore eliminate the need of turbines during take-off. Even a replacement of the main electric generation system (generators coupled to the turbines) seems a promising idea, so that the turbines could be optimized for thrust generation. For the aviation industry, this will result in a more efficient engine, cleaner propulsion and less air pollution by aircraft. In portable applications for powering computers,
mobile phones and other devices, fuel cells are already available, and in smaller markets such as caravan or yacht power supply, fuel cells are becoming common technology.
In all these application fields – including the automotive arena – different cell types are in use. For portable devices the direct methanol fuel cell (DMFC) is mainly used because its fuel methanol and the cell technology is easy to handle. Its low power density neglects this cell type for higher power ranges. A more common cell type is the polymer electrolyte membrane fuel cell (PEM) fuelled with pure hydrogen. It is used for all kind of applications and even for a breakthrough of this type of cell, a net of hydrogen supply has to be installed. A promising technology could be the solid oxide fuel cell (SOFC), which is able to convert various fuels such as hydrogen and carbon monoxide, into electricity, while other organic fuels need to be split by reforming. The SOFC works at high temperature levels between 650°C and 900°C, therefore fitting well to reforming technology at a similar level, thus enabling higher system efficiencies.