voltage, power density obtained and the heat generation as a function of the operating current density. .................................................................................................................... 7 Figure 5: Typical polarization curve for PEMFC characterization showing the three different dominant loss regions - activation, ohmic and mass transport losses. ................. 7 Figure 6: Transport resistance in two perpendicular directions due to presence of liquidwater in reactantchannels. ................................................................................................ 10 Figure 7: Flow patterns commonly observed in PEMFC reactantchannels. (a) slug flow (b) film flow (c) mist flow. ............................................................................................... 11 Figure 8: Pressure drop signatures associated with flow patterns commonly observed in PEMFC reactantchannels. (a) slug flow (b) film flow (c) mist flow . ...................... 12 Figure 9: Flow pattern map over a range superficial water and gas velocities at dry and fully humidified inlet conditions . .............................................................................. 28 Figure 10: Two-phase pressure drop multiplier from experimental investigations by See  (left) and Anderson (right). ................................................................................. 28 Figure 11: Saturation pressure of water in air as a function of temperature. .................... 48 Figure 12: Scaling active area for single cell geometry – 50 cm 2 . .................................. 51 Figure 13: Schematic of the in situ test setup. A Greenlight G-40 test stand controls the reactant conditions, while external controls are setup for controlling cell temperature. .. 53 Figure 14: Polarization curve for validating performance of PEM fuel cell. Operated at 60°C with 95% inlet RH. Stable operation up to 1.2 A/cm 2 without mass transport
By means of the VOF model, between 2006 and 2008, a series of numerical simulations of two-phase flow in several different types of cathode flow field design were conducted [25-29]. The water removal characteristics [25-27], the effects of electrode wettability on liquidwater behaviour , and accelerated numerical test of liquid behaviour  were investigated, but neither conventional porous layers nor electrochemical reactions were included in these analyses. In 2008, a general model for PEMFC has been proposed  and applied in the following years to investigate two-phase flow coupled with electrochemical reactions, watertransport through the membrane, and heat and mass transfer for single cells with different flow field designs [30, 31] and a 3-cell stack . However, this technique requires excessive computational time. In order to minimize computational time, a simplified model  was developed by neglecting electrochemical reactions and heat transfer effects since computational results were hardly affected for the aim of simulating liquidwatertransport. And detailed experimental validation  was performed by direct optical visualization method to capture the motion and deformation of liquidwater. Then liquidwater emerging and flooding process inside a PEMFC cathode with straight parallel  and interdigitated  flow field designs were studied by means of this validated simplified model.
In this study, the effect of electrical contact resistance on cell performance and local transport characteristics of protonexchangemembranefuelcells (PEMFCs) are numerically investigated by using a two-dimensional, non-isothermal and two-phase flow fuel cell model. The conservation equations of species, temperature, charge, liquidwater and dissolved water were solved to investigate the transport processes of heat and mass transfer, electron and proton transports, liquidwater formation and transport, and watertransport through the membrane. The mathematical model was validated against the experimental data reported in the open literature. Results showed that the performance is significantly affected by the electrical contact resistance, especially at low cell voltages. In addition, the temperature, liquidwater saturation and solid phase potential distribution profiles are greatly influenced by the existence of electrical contact resistance.
Cao and Djilali  developed a 2-D, nonisothermal, nonisobaric model for PEMFC. In their research, conservation laws for water and current, together with an empirical relationship between electroosmotic drag and water content, were applied to obtain a transport equation for water molar concentration and to derive a new equation for the electric potential that strictly accounted for variable water content. The model was coupled with a computational fluid dynamics model that included the porous gas diffusion electrodes and the reactant flow channels. The resulting coupled model accounted for multi-species diffusion (Stefan-Maxwell equation), first-order reaction kinetics (Butler-Volmer equation), protontransport (Nernst-Planck equation), and watertransport in the membrane (Schlogl equation). The results showed that water content distributions can be improved in the membrane when the cell was operated at a higher pressure on the cathode side than on the anode side.
Fuel cell technology, particularly for transport applications, would take a leap forward if a viable system were to be developed that could use a liquidfuel without the need for reformation. The prospects for anode catalysts being developed having the activity to operate on petroleum derived hydrocarbon fuels are poor. However, Shell and others in the 1960s established that Methanol, with anode catalysts such as Pt/Ru, had some potential. The early work utilized sulphuric acid as the electrolyte. With the introduction of proton conducting membranes, interest in DMFC systems in the 1990s has been renewed with projects in America, Japan and Europe. Of particular significance has been the work of Los Alamos National Laboratory. If the power density required for vehicle applications are to be achieved, further improvements to anode catalyst performance are necessary. In addition, existing membrane materials are subject to what is known as ‘methanol crossover’, which in turn contributes to poor cell performance. In this context, it is interesting to speculate on how high temperature membranes such as that developed by Celanese would perform in a DMFC fuel cell (23) .
In this manuscript Dynamic behavior of a fuel cell which is a complex phenomenon which is modeled in mathematical equation including all loses which arises in PEM Fuel cell and simulated using mat lab / Simulink package. A PEM fuel cell has been modelled with this model exhaustively using more parameters. Dynamical structure of the model can be obtained to change the input parameters as required in this model the factors in the dynamic behavior of a PEM fuel cell are the reactant gases humidity change, various load changes and liquidwater formation in cathode channel.
The enhanced automotive fuel cell model developed in this research provided an effective tool for developing fault recognition techniques on protonexchangemembranefuel cell technology. This research effectively incorporated cell deviation for analyzing a single cell. The single cell model tracks with negligible error (on the order of 10 −7 ) in nominal conditions. The charge double layer completes the dynamic electrical effects in the en- hanced model. The addition of a thermal model introduces long term transients as the stack is no longer idealized at 80 ◦ C. The control of cathode humidification by manipulating the thermal model’s target temperature effectively prevents liquidwater from forming in the cathode channels. Inert gas build-up due to nitrogen permeation and impure fuel sources is successfully modeled, reducing the output voltage as the nitrogen accumulates. The anode outlet manifold successfully models the purging transients, introducing yet another auto- motive fuel cell transient. Each of these effects have the ability to increase and decrease the operational fuel cell voltage making fuel cell fault classification more complicated.
In order to explain the reason that mechanical vibration causes the voltage fluctuation, the experiments watertransport in an anode flow channel were conducted for the transparent PEMFC under mechanical vibration and no mechanical vibration, respectively. The watertransport progress in the anode flow channels under no vibration During this process, the gas was not humidified at the anode and the cathode, the flow rate was 70 mL/min at the anode and 1 L/min at the cathode, the temperatures was 40 _C, the transparent PEMFC was loaded at a constant current 5 A. At first, a water droplet suddenly migrates into and through the diffusion layer at the 5 min, From the enlarged picture shows, the water droplet is appearing close to the channel. After 8 min the water droplet grow larger, which lead to a complete blockage of the channel. The hydrogen-flow in channel is interrupted while a decline of voltage cannot be observed during the 8 min. This phenomenon can be explained that although gas channel was blocked by water droplet, the hydrogen flow could be distributed uniformly by diffusion through the gas diffusion layer. Then, the droplet keeps growing and changes shape gradually by blowing of hydrogen. Finally, the hydrogen pressure in the channel overcome the capillary force and the wall viscous wall force of the droplet, and the droplet was broken and flowed away. Before long an enormous undershoot and overshoot of voltage can be observed. The voltage falls from 0.50 to 0.47 V and overshoots to 0.51 V. The main reason for the fluctuation of the voltage is due to the decrease of the hydrogen pressure and the fluctuation of the hydrogen when the droplet was blown away. Notably, this phenomenon is very similar the voltage fluctuations under mechanical vibration the water column exudation and growth progress in the anode flow channel under no vibration. After 14 min, the water column was so larger that it was difficulty to be blown out of the channel when the hydrogen flow rate was 70 ml/min.
The use of compaction pressure during the assembly of fuelcells plays a crucial role particularly at the interface between the GDL and the reactant-flow plates. It reduces the interfacial contact resistance between those two parts as well as it serves as sealant to ensure proper delivery of reactants to the active flow channels. However, the increase in pressure must be controlled accurately since it may cause damage to some components of the cell. If the reactant-flow plate breaks, the reactants can escape from the reaction channels to the outer side of the cell; alternatively they can cross from channel to channel making less use of the active area of the catalyst. In addition, the bulk resistance of the reactant-flow plate becomes higher and the same applies to the GDL. Any hole or broken strip on the surface of the GDL allows the reactants to cross more easily from the anode to the cathode, or visa versa, increasing the amount of fuel crossover. The reactants can also flow directly from the channels into the catalyst layer without being uniformly distributed on the surface of the catalyst. Finally, the conducting area between the GDL and the catalysts will be reduced.
For PEMFC single cell tests, membrane electrode assemblies (MEAs) with active area of 5.0 cm 2 were prepared. Pt/C was introduced as anodic catalyst, while Pt/C and PtM/Cs were assigned as cathodic catalysts, respectively. The catalyst powders of Pt/C and PtM/Cs were ultrasonically mixed with 5 % Nafion solution (Solution Technology, Inc., 1100EW) and deionized with water and isopropyl alcohol for the preparation of catalyst ink. For making MEA, CCM (catalyst coated membrane) method was used. In the method, the catalysts inks were directly air-sprayed onto both sides of Nafion 212 (Dupont) membrane. The loading of metal catalysts were 0.2 mg cm -2 in both electrodes, respectively. The MEAs were then assembled with gas diffusion layers (GDL, E-Tek division, plane weave, 0.35 mm cloth thickness and 30% polytetrafluoroethylene (PTFE)) and gasket (TF 350, CNL, Inc.). The completed MEAs were then placed between two graphite plates, which have engraved serpentine flow channels.
Lumped model has been developed in this study. Lumped model is model with zero dimensions. The model was developed with some assumptions such as, the transport process is steady- state which resolves coupled transport in membrane, all gases obey the ideal gas low, the gas flow channels is laminar ,the catalysts is very thin ,the change phase of water was neglected ,the heat transport cross solid medium as gas medium, the output temperature is cell temperature. The developed model has five major sections. Theses included mass transport, heat transport, electrical characterization, water management and losses product.
water management. Bernardi et al.  developed a hydraulic model with assumptions that the membrane is fully hydrated with liquidwater. However, this model is not suitable for a partially dry membrane. Yi et al.  developed one model for interdigitated flow fields considering liquidwater. However the liquidwater flow induced by capillary forces and gas flow induced drag was addressed by semi-heuristic equations. Natarajan et al.  developed a pseudo-three-dimensional model by extending their two-dimensional isothermal model. This model can be used to obtain qualitative insight into the distribution of liquidwater in the backing layer. Two-phase flow modeling in the porous air cathode of a PEMFC was attempted by Wang et al. . The transport coefficients in the stationary numerical two-phase model are parameterized as functions of the liquidwater saturation. However, this model only considered the cathode half cell. Berning et al.  developed a 3D, two- phase PEMFC model based on computational fluid dynamics multiphase. Transport of liquidwater inside the gas-diffusion layers is modeled using viscous forces and capillary pressure terms. Shimpalee et al.  developed a complete three-dimensional model, which treated the liquidwater as component of the gas mixture. Steinkamp et al.  presented a 2D dynamic two-phase flow model accounting for all important transport processes in a PEM fuel cell. Unfortunately, the computational cost for simulation was quite high because of the high complexity of the model. Siegal et al.  presented a 2D model to account for the transport of liquidwater in the electrode and treat all three forms of water as a separate phase and allow mass transport among them. Zhang et al.  developed a model including three forms of water: dissolved water
The chemical energy of fuel is directly be converted into the electricity through a device named fuel cell. In the fuelcells, oxygen and fuel are combined to generate electricity along with heat and water as the by-products . Compared to the internal combustion engines, the fuelcells have higher efficiency as they directly convert the chemical energy to electricity, without any combustion process . However, there are several issues associated with the proton-exchangemembranefuel cell (PEMFC) technology, particularly the heat dissipation process. As the PEMFC performance highly depends on its operation temperature; the heat removals and thermal management of the fuel cell are very important [3, 4]. There are several ways for improving thermal performance of mechanical equipment [5-10] and cooling system of a PEM fuel cell, such as heat spreaders, air flow or liquid cooling, and the cooling associated with the phase change such as evaporation[11, 12]. Among these approaches, the water-cooled system is more of interest as it provides a better temperature
cell is considered to be isothermal, the membrane is assumed to be well-hydrated and the treating of liquidwater phase is unevident. To overcome the limitations of Springer model and Bernardi and Verbrugge model, Nguyen and co-workers presented a pseudo-two dimensional mathematical model that considers the heat and mass transport in the fuel cell. In Nguyen model, watertransport through the membrane caused by electro- osmotic drag effect and back diffusion is described as a function of water activity in the anode flow channel and gradient of water concentration across the membrane is assumed to be linear. The model results show the influences of various operating parameters and conditions on the fuel cell performance and the changes of current density, net water flux and electro-osmotic drag coefficient along the channel for different humid- ification circumstances. The voltage loss, current density, water and temperature profiles along the channel were also described. Out of these models, the degrees of the flooding in the cathode diffusion layer and cata- lyst layers were investigated in a one-dimensional, steady state fuel cell model developed by Baschuk and Li . The model results demonstrated that the overpotential of the cathode catalyst layer is increased as the oxygen concentration decreases and at the low current density, the maximum flooding will be reached as the cell pressure significantly increases. By introducing the fraction of electrode flooded, the polarization curves gained from the numerical model also show a good agreement with the experimental data from the literature. Zhou and co-workers  developed a one-dimensional, along-flow-channel model combined with catalyst layer model that consider the effects of pressure, temperature, anode inlet humidification, anode inlet water content and the membrane thickness on the fuel cell performance. Zhou also defines the presence of liquidwater distribution along the channel by the local relative
The commercialisation of fuelcells is mainly affected by their high cost and the lack of a hydrogen infrastructure. Attempts to reduce the cost of fuelcells and to develop a global hydrogen network for their use are still ongoing. The primary focus is to minimise the loading of Platinum (Pt) in the catalyst of the Membrane Electrode Assembly (MEA) which is considered to be the main contributor to the high cost . Some studies have suggested the use of more effective techniques for Pt deposition onto the catalyst to reduce its loading [3-7], others concentrated on enhancing the polarisation performance of the cell to compensate for further Pt reduction [8-10]. Although some studies  claimed that reduction of up to 50% (e.g. 0.15 mg-Pt cm -2 ) is achievable with marginal compromise to the fuel cell stack efficiency (<1%), further reduction is still needed . The manufacturing cost of fuel cell components, particularly the MEA and the Gas Diffusion Layer (GDL), is another significant contributor to the overall cost . An approach which can make them more affordable is to increase the manufacturing volume of fuel cell stacks and to reduce the complexity of the manufacturing processes of some of their components . It is expected that, as fuel cell stacks become commercially viable, hydrogen fuelling infrastructure will be established that is easily accessible for supporting the penetration of fuelcells into the market place [13-14]. This remains a major concern particularly for the transport sector where issues associated with hydrogen production [15-16], safety [17-19], on-board and off-board storage [20-22] have surfaced.
The wetted bend ratio and wetted area ratio for both sides have been calculated from the captured images and plotted in Fig. 5. It can be seen that they both decrease as the air flow rate increases. As indicated earlier, this is most likely to be due to the increase in the ability to remove liquidwater from the flow channel with increasing air flow rate. Since water is produced at the cathode, the wetted ratio numbers at the anode are smaller than those of the cathode. However, their trends are similar to each other even though the hydrogen flow rate has been kept constant. This may be attributed to the net watertransport across the membrane through the electro-osmotic drag and the back diffusion. This indicates that the wetted ratio numbers at the anode side can also be employed to interpret the relationship between the fuel cell performance and the air flow rate.
over the respective GDL and remove the water produced during the reaction. Since the performance of the PEM fuel cell is strongly affected by the flow field design, several numerical models have been developed to analyse the coupled transport process and electrochemical reaction in PEM fuelcells . Typically, the flow field configurations can be divided into five types: parallel, interdigitated, pin-type, spiral, and serpentine, which are schematically illustrated in Figure 2. The serpentine flow field, either containing single (Figure 2e) or multiple channels (Figure 2f), is the most commonly used design in commercial fuelcells . The effect of the gas flow fields design on the fuel cell performance was investigated experimentally by Dhahad et al. . On the other hand, the modelling of an optimum flow field design was presented by Kahraman and Orhan , including a parametric study with respect to different design and performance parameters in a flow field plate. In addition to the conventional flow field patterns, some nature-inspired flow field designs have been studied recently . The results show that the bio-inspired interdigitated designs improve the fuel cell performance by about 20–25%, in comparison with the conventional designs . However, the manufacturing complexity leads to significant costs because typically these BPPs are made from graphite composite and produced by a selective laser sintering process.
safety are some issues which needed to be accounted for designing a fuel cell assembly. There is a need for development in catalyst layer. Further studies are needed in characterization of pore size distribution as well as hydrophilicity and hydrophobicity distributions and using this information to develop pore level models. The realistic and accurate simulation of liquidwater and gas transport through gas diffusion layers with highly non-uniform pore sizes and wettability can be done by such studies. For success in direction of wide areas of applications, a number of important rather complex problems must first be solved including cost, environmental stability and longer lifetime of the cell.
Another issue is the field of fuelcells turns around of the water management. Poor water quantity in the system decreases the proton conductivity and increases the cell resistance and the cell voltage decreases. Unnecessary water amounts in the cathode causes flooding; the water blocks the supply gases in the porous media resulting in mass – transport limitations, increasing the cathode overvoltage, and in considerable power losses in the PEMFC [39-41]. One alternative is the operation of PEMFC’s at high temperatures (above 100 °C). At this condition the water management can be simplified when a single phase of water is considered and the reaction gases can be introduced with lows amounts of water [42-44]. More advantages of work at high temperature and low relative humidity are (i) the improvement on the kinetics of the oxygen reduction reaction ; (ii) higher tolerance to CO [45, 46]; (iii) the omission of cooling systems ; and (iv) waste heat can be recovered . In order to work at high temperature, Nafion membrane must be modified. One effective procedure consist in the incorporation of inorganic materials as Al 2 O 3 , SiO 2 , ZrO 2
The mesoporous inorganic materials have a large internal surface area leading to a capillary condensation of water molecules that occurs at relatively low relative humidity (RH), which leads to easy transport of protons without excessive humidification . As porous framework materials, mesoporous structures have many ordered or disordered nano-size channels. Therefore, the mesoporous structure could improve their water adsorption capacity and proton conductivity under high temperature. The proton conductor such as P 2 O 5 could dissolve in water and presence in these