There are a lot of research subjects to investigate the factors to improve the fuel cell performance, such as air movement, system considerations, system integration, load handling, fuel delivery, water removal, uniform temperature distribution, homogeneous fluid distribution, good current transport, high conductivity and materials.
Meyers and Maynard [4] used a miniaturized fuel cell integrated on a silicon substrate. There were two designs, one was a bilayer design, which was like sandwich structure that the anode and cathode was separated; another design was a monolithic design that the anode and cathode was on the same substrate. Comparing with the two designs, they found that the power density of a bilayer design is much better as shown in Fig.1.3.
Noponen et al. [5] introduced a measurement system for mapping of current distribution in a free-breathing PEMFC. The results showed that the operating temperature has a significant influence on the performance of the fuel cell. Furthermore, the free convection is weak at low cell temperature, and the membrane is dry at high cell temperatures. They also showed that under some
conditions the fuel cell has homogenous current distribution and does not need any auxiliary pumps or fans to increase the airflow.
The paper by Shah, Shin and Besser [6] was the first report using PDMS as the base material in micro fuel cell. They manufactured micro flow channels on PDMS by soft lithography, because PDMS was cheaper material comparing to silicon. They also compared Pd catalyst with Pt one and found that Pt catalyst is better than Pd one due to the poorer reaction kinetics of Pd catalyst. The performance of fuel cell is proportional to the humidification of hydrogen stream, the conversion of hydrogen, and the catalyst porosity in electrodes. The loading of catalyst has no influence on the conversion of hydrogen; the conversion of hydrogen is determined by the partial pressure of gases at the outlet measured by the mass spectrometer and the temperature on catalysts. When the temperature on catalysts increases, the conversion of hydrogen is decreased, leading to a poor performance.
Shimpalee et al. [7] established a model of 200 cm2 serpentine flow-fields with different patterns to make the gas distribution uniform. In this paper, they employed the 3-channel, 6-channel, 13-channel, 26-channel serpentine flow-fields and 26-channel complex flow-field on 200 cm2 PEMFC as shown in Fig. 1.4. The conclusion was that the performance changes with the number of parallel channels, which is related to the path length. The shorter path length (the higher number of parallel channels) has better performance, less water production and more uniform distribution of current density. However, the performance of 13-channel serpentine flow-field is better than that of the 26-channel one, because it has higher water content in membrane, which leads to a higher proton conductivity. The performances of 26-channel serpentine and 26-channel complex flow-fields are similar, indicating that the performance
6 performance than the one with the hydrophilic. The latter absorbs water in the GDL, so the membrane is dry out. On the contrary, the hydrophobic GDL cannot absorb water in the GDL, so it keeps the membrane humidified and blocks the oxygen supply. The best performance for air-breathing PEMFCs is an untreated Toray○R GDL, which has a lower contact resistance and higher porosity; it keeps the membrane humidified, but does not block the supply of oxygen.
Jung et al. [9] improved the water management problem and the performances of the air-breathing and air-blowing PEMFCs at low temperature by adding hydrophilic SiO2 particles to the anode catalyst layer. The conclusion was that the performance of air-blowing PEMFC is higher than that of the air-breathing one due to the good transportation of air at the cathode.
The performance becomes highest with 100% humidification at the anode; on the other hand, the performance becomes the lowest with flooding at the cathode.
The advantages of SiO2 are that the back diffusion of water to the anode can be enhanced and the water at the cathode can be removed by absorbing water to the anode.
Chen et al. [10] used rapid prototyping (RP) technology to build a new 10-cell air-breathing miniature planar array fuel cell stack, which has a volume
of 6cm×6cm×0.9cm, and the active area was 1.3cm×1.3cm in each individual MEA. The flow field plate was made of acrylonitrile-butadiene-styrene (ABS) by plastic injection molding technology. The RP technology is much faster and cheaper than the conventional CNC and MEMS. The peak powers of the parallel connected and serial connected stack are 99mWcm-2 at 0.425V and 92mWcm-2 at 4.25V under free convection (at 70℃), and 123mWcm-2 at 0.425V and 105mWcm-2 at 5.25V under forced convection. The parallel connected stack has higher power density than the serial one, and the performance under forced convection is higher than that of free convection.
Ito et al. [11] developed an evaluation method based the ratio of liquid water to pore volume in GDL. They measured the differential pressure through the interdigitated fuel cell to estimate the liquid water ratio in GDL. The differential pressure was measured by ac impedance method that measuring the ionic resistance in polymer electrolyte membrane, and the water saturation was related to the operation condition of fuel cell. The variables included the material of GDL, load current, gas utilization ratio and humidification temperature. The conclusion was that the performance of cloth-type GDL is higher than that of paper-type one, and the water saturation is proportional to the load current, gas utilization ratio and humidification temperature.
Wen and Huang [12] presented a new method to discharge the waste heat to ambiance by using the pyrolytic graphite sheet (PGS) in a single fuel cell.
The PGS is a thermal conduction material, playing an important role as a heat spreader, and the advantages are less volume, light weight and cost reduction by reducing the ancillary components. The heat transportation and temperature distribution are better with PGS than the ones without PGS. When increasing
8
the oxygen flow rates, more water is removed from the membrane and it becomes dehydrated. The use of PGS increases the water condensation due to the temperature decrease by conduction, so the membrane becomes hydrated.
The conclusion was that the PGS increases the water droplets under the high flow rates and dry conditions.
Pandiyan et al. [13] developed an analytical method to understand the thermal and electrical resistances of PEMFC, which were determined by the mass balance and polarization curve, respectively. Thermal and electrical resistances of the electrode are 67.7 and 52mΩ in a four cells stack, respectively.
When the increase of thermal resistance is three times, the current increases 50%
and the temperature change is 10℃. The electrode fabrication process can change the internal resistance of the fuel cell stack, which is a major key in thermal management.
Siu and Chiao [14] used PDMS as the material of gasket and electrode in microbial fuel cell. They used MEMS technology, such as etching and evaporation processes, to form the microchannel pattern on the silicon wafer, and to mold PDMS on the wafer. Comparing with the recent silicon micromachined microbial fuel cell, their result showed the better performances in the average power density and average current density.
Song et al. [15] used PDMS to fabricate planar PEM fuel cells. They measured current density at different flow rates. The result showed that the current density is a decreasing function of flow rate because the residence times for protons reaching the cathode are shorter at higher flow rates. Since the distance between anode and cathode must be as short as possible, they used Nafion membrane, whose thickness was only 200~400nm. The current density
increases 166% compared to the other planar membrane device.
According to [12], Wen et al. [16] used PGS in a 10-cell stack with 100 cm2 active area. The temperature variation was measured by four thermocouples on the cathode gas channel plate. The result showed that the temperature distribution becomes uniform in fuel cell stack with the PGS, indicating that PGS is a good thermal management material. And the maximum power of fuel cell stack with the PGS increases 15%.
Bussayajarn et al. [17] applied three different cathode geometries with the same opening ratio: parallel slit, circular open and oblique slit as shown in Fig.
1.5. The performance and stability were investigated with the different cathode geometries. The circular open design is found to be the best in performance and current density, and the performances of parallel slit and oblique slit design are similar, because the shorter rib distance and hydraulic diameter can result in better oxygen transportation and uniform oxygen distribution. The oblique slit design shows high stability, on the contrary, the parallel slit and circular open design are unstable. Both the performance and stability are better in forced convection condition than in self-breathing condition.
Dai et al. [18] surveyed the papers related to water transport and balance in the MEA of PEMFCs. The parameters included operating conditions, component designs and material properties. The major rule of the water balance depends on the materials of water management properties and the components matching operating conditions and load requirements. The data of materials and components in MEA were insufficient, so it was difficult to optimize the design of MEA. It needs to develop a new material for water management capability and to design a component structure and water balance models. Water balance influences not only the performance but also the
10
durability in both experiment and modeling.
Karst et al. [19] used different cathodic cover opening ratio to manage water content for micro air-breathing PEMFCs. It was a method which decreased the flooding at the cathode and increased the water back-diffusion at the anode by varying the cover opening ratio. The new method didn’t require any control tool and didn’t increase the volume of cell, either. The total closure of the cover maintains the water accumulated at the cathode. The 5% of cover opening ratio maximizes the back-diffused water and produces 33% of total water at 150mA cm-2.
Pomfret et al. [20] used a Si-charge-coupled device (CCD), camera-based and near-infrared imaging system, to observe the anode processes in the solid oxide fuel cell. A Si-CCD camera-based NIR imagining system was the first equipment using at the Ni/YSZ cermet anodes. The benefit of the NIR imagining system was cheaper, lighter and more sensitive than mid-IR imaging system. Most importantly, it was easier and quicker to observe the operation and processes in SOFCs under various conditions. And the result showed that the temperature drop is due to the presence of oxides and water in SOFCs.
Kim et al. [21] used the thin flexible printed circuit board (FPCB) as a current collector in order to reduce an air-breathing monopolar stack's volume.
They also designed different patterns of air-breathing holes on the cathode to find the effect of varying the geometry and opening ratios on stack performance as shown in Fig. 1.6. They found that in cathode, the circular-hole pattern with opening ratio of 38% has the best performance; on the other hand, the rectangular cathode opening pattern with 65% opening ratio causes cathode flooding and unstable output problems.
Yu et al. [22] attempted to reduce the cell resistance and improved the
performance of miniature silicon wafer fuel cells. Three different thicknesses of current collector were selected. They found that the thicker the current collector, the better the cell performance. It is because an increment of the current collector thickness increases the area of conductor, which decreases the cell resistance and improves the cell performance.
Zhang et al. [23] used two effective methods, FEM analysis and simplified
prediction method, for estimating the contact resistance between the bipolar plate and GDL. The predicted results by both methods show the good agreement with the experimental ones. The contact resistance is influenced by the average clamping pressure and the assembly clamping pressure distributions.
Chang et al. [24] studied the effects of the clamping pressure on the performance of PEMFC. The contact electrical resistance is a function of the clamping pressure, so it is necessary to assemble a fuel cell with a proper force.
The results showed that increasing the clamping pressure reduces the interfacial resistance and enhances the electrochemical performance of a PEMFC at the low clamping pressure levels. In contrast, increasing the clamping pressure reduces the Ohmic resistance, but meanwhile narrows down the mass diffusion path from gas channels to the catalyst layers at the high clamping pressure levels.
The above two effects make the power density not to rise due to the lower mass-transfer limitation for higher current density.
Zhou et al. [25] investigated the effect of clamping force on the performance of PEMFC that directly affects the interfacial contact resistance, non-uniform porosity distribution of GDL, GDL deformation and reactant transport in GDL. They used finite element method (FEM) to analyze the elastic deformation of GDL and porosity distribution, and finite volume method (two-phase flow model) to analyze the mass transport of reactants and products.
12
The results showed that the contact resistance plays an important role at a low clamping force in determining the power density because the contact resistance decreases obviously with increasing clamping force. But at a high clamping force, the contact resistance decreases slightly with an increasing clamping force, and now the GDL deformation plays a main role. As the GDL porosity observe the internal contact pressure distributions under three different clamping torques (8, 12 and 16 N-m) and three different bolt configurations (2, 4 and 6-bolt configurations). When the torque was applied, the pressure-sensitive film was broken and a color-forming material was released and absorbed by the film. The pressure film was then transferred into a color image file that was compared with the reference color bar to obtain the pressure values. The results showed that the larger mean contact pressure and more bolt numbers, the higher maximum power. The uniformity of the contact pressure distribution is improved and the contact ohmic resistance is reduced when increasing the mean contact pressure and bolt numbers. However, the maximum power does not increase linearly with the bolt numbers and torques. In fact, it increases until a certain torque point (e.g. >10 bar) is reached, and further increasing the clamping pressure not only reduces the contact ohmic resistance but also narrows down the mass transfer path from gas channels to the catalyst layers.
Lee and Chu [27] applied a finite volume-based CFD approach to
investigate the behavior in a fuel cell. The effects of cell temperature and
humidification temperature influence the location of the gas-liquid interface (defined as the location where the liquid water begins to condense), the cell performance and the distribution of liquid water saturation. The results indicated that the humidification temperature slightly higher than the cell one is the best working condition. It is because that in such condition, liquid water forms when the inlet gases enter the channel, some of the liquid water keeps the membrane moist to enhance its ionic conductivity.
Matian et al. [28] used a thermal imaging camera to study the temperature distribution and variation on the outer surfaces of PEM fuel cell. One important parameter, namely the surface emissivity factor, that needed to be identified in advance. A calibrated thermocouple was put on surface and the temperature was recorded, then, emissivity factor in the camera settings was altered until the temperature measured by the camera agreed with the one by the thermocouple. They obtained the emissivity factor of 0.88, and this value did not change throughout the experiments. The results showed that the temperature distribution in the stack is not only affected by convection in the gas flow channels but also by natural convection and conduction; about 50~60% of the heat is dissipated by natural convection.
Zhang et al. [29] operated a PEMFC without external humidification (0%
relative humidity) to eliminate the gas humidification system and decrease the complexity of the fuel cell system. The performance at 100% RH is higher than that at 0% RH, because the membrane needs to be humidified to carry hydrogen ions. The results showed that the cell performance at 0% RH decreases with the increasing operation temperature and reactant gas flow rate and a decreasing operation pressure.
Ous and Arcoumanis [30] used 25 cm2 reaction area of the air-breathing
14
single PEM fuel cell to investigate the formation of water droplets and their aggregation in the cathode under various operating parameters, such as air and hydrogen stoichiometry, cell temperature and external load. The stoichiometry of air and hydrogen causes droplets aggregation and makes fewer droplets extraction. In contrast, the temperature and external load have the effects on removing water, and the temperature is especially an obvious operating parameter. The formation of water droplets is reduced as the temperature increases, but the over-temperature can cause the membrane dehydrated.