Nitrogen Crossover Research Articles

Experimental and numerical investigations of high-temperature PEM fuel cells under different anode dilution levels and varying temperatures

This work investigates the effect of nitrogen in the anode gas mixture on the performance of HT-PEMFCs using experiments and simulations. It explores four H2/N2 ratios (100/0, 80/20, 60/40 and 50/50) at three temperatures (140 °C, 160 °C, 180 °C), and four current densities (0.1 A cm−2 to 0.4 A cm−2). The results demonstrate that N2 dilution negatively impacts fuel cells performance, but higher temperatures can mitigate this effect. Temperature influences electrochemical impedance spectroscopy, with initial ohmic resistance increase followed by a slight decrease. N2 dilution effects charge and mass transfer resistances. Current density has minimal impact on ohmic resistance but improves charge transfer resistance, especially at high frequencies. This understanding can optimize fuel cell performance, especially in ammonia-fueled systems with hydrogen and nitrogen mixtures or during anode gas recycling with significant nitrogen crossover.

A 1 + 1-D Multiphase Proton Exchange Membrane Fuel Cell Model for Real-Time Simulation

A 1 + 1-dimensional (1 + 1-D) multiphase model of proton exchange membrane fuel cell (PEMFC) is developed to simulate the dynamic behaviors of the fuel cell under various operating conditions. Electrochemical, fluidic, and thermal physical phenomena are considered, and the model can accurately describe the multiphysical processes inside the PEMFC. Meanwhile, the water phase change and nitrogen crossover are also considered in this model. The developed model can simulate the fuel cell behaviors in both normal temperature and cold start conditions. Selection of the time step size is discussed, and the solution method of the real-time model is proposed. Computational efficiency is greatly improved by simplifying the multicomponent diffusion inside the membrane electrode assembly (MEA), which results in the shorter calculation time than the simulated physical time. The proposed model is suitable for embedded applications, such as real-time simulation or online diagnostic control. The comprehensive model validation is carried out by comparing the simulation results, including polarization curves, ohmic resistance, local current density distribution, and voltage evolution, under various working conditions of steady state, load change, and cold start, and a good agreement is achieved. Finally, some simulation cases are studied to further demonstrate the simulation ability of the model.

Comparison of state-of-the-art machine learning algorithms and data-driven optimization methods for mitigating nitrogen crossover in PEM fuel cells

Nitrogen gas crossover (NGC) and nitrogen accumulation at the anode of proton exchange membrane (PEM) fuel cells are ineluctable and it would lead to inferior performance and even irreversible damage to functional components. To mitigate this issue, multiphysics numerical models (MNMs) are established to describe NGC behaviors and further guide experimental studies. However, to obtain the optimized parameters that would suppress NGC and retain high performance, grid search conducted on MSMs would cost unaffordable computational resources and time. Therefore, we innovatively introduced a machine learning-assisted MNM (MSM-ML) as a surrogate model, in which 9 state-of-the-art machine learning algorithms were compared, to greatly boost the resolution of this engineering problem. Through the proposed MSM-ML workflow performed on an experimentally validated MSM, the cost for obtaining the best parameter combination is greatly reduced. Moreover, the impact of each parameter in this complex system is directly revealed through the application of black-box interpretation methods afterwards. As a result, a new approach was pioneered to solve engineering problems which was demonstrated to be more efficient and intelligent than traditional methods. The NGC coefficient is reduced by 49.5%, while the power density is improved by 20% through the multivariable optimization of the developed MSM-ML.

An Artificial Intelligence Solution for Predicting Short-Term Degradation Behaviors of Proton Exchange Membrane Fuel Cell

The dead-ended anode (DEA) and anode recirculation operations are commonly used to improve the hydrogen utilization of automotive proton exchange membrane (PEM) fuel cells. The cell performance will decline over time due to the nitrogen crossover and liquid water accumulation in the anode. Highly efficient prediction of the short-term degradation behaviors of the PEM fuel cell has great significance. In this paper, we propose a data-driven degradation prediction method based on multivariate polynomial regression (MPR) and artificial neural network (ANN). This method first predicts the initial value of cell performance, and then the cell performance variations over time are predicted to describe the degradation behaviors of the PEM fuel cell. Two cases of degradation data, the PEM fuel cell in the DEA and anode recirculation modes, are employed to train the model and demonstrate the validation of the proposed method. The results show that the mean relative errors predicted by the proposed method are much smaller than those by only using the ANN or MPR. The predictive performance of the two-hidden-layer ANN is significantly better than that of the one-hidden-layer ANN. The performance curves predicted by using the sigmoid activation function are smoother and more realistic than that by using rectified linear unit (ReLU) activation function.

Comparison of State-of-The-Art Machine Learning Algorithms and Data-Driven Optimization Methods for Mitigating Nitrogen Crossover in Pem Fuel Cells

Comparison of State-of-The-Art Machine Learning Algorithms and Data-Driven Optimization Methods for Mitigating Nitrogen Crossover in Pem Fuel Cells

Modeling and dynamic performance research on proton exchange membrane fuel cell system with hydrogen cycle and dead-ended anode

In order to improve the response speed and output power of the system to ensure the safe and stable operation of the proton exchange membrane fuel cell system, it is necessary to control the gas supply system reasonably and effectively. In the present study, a system simulation model considering nitrogen crossover and water migration is established, which is in good agreement with the experimental results. By using this model, the pressure difference between the two sides of the membrane during purging was simulated, showing that the pressure difference was all less than 20 kPa. Increasing the pump speed is advantageous because hydrogen recovery increases the relative humidity of the anode inlet and avoids membrane drying of the anode inlet due to the lack of an external humidifier. When the anode recovery rate is at most 90%, the corresponding maximum relative humidity at the anode inlet is 0.45. When the humidity is constantly with the operating temperature at 60, 70 and 80°, the higher the operating temperature, the lower the output voltage. These results reveal that the hydrogen recirculation fuel cell system with dead-ended anode has good performance and the control strategy is effective.

Uncertainty analysis and robust control of fuel delivery systems considering nitrogen crossover phenomenon

In this paper, the fuel delivery subsystem (FDS) with hydrogen recirculation and anode bleeding is applied in proton exchange membrane fuel cell (PEMFC) system, which is utilized to supply hydrogen to the anode of stack and recirculate fuel back to the supply line. As the diffusion of nitrogen from cathode to anode is inevitable in a real PEMFC during long-term operation. To prevent system performance decline due to nitrogen accumulation. Therefore, this paper firstly develops a control-oriented nonlinear dynamic FDS model involving gas diffusion. Additionally, the FDS is very sensitive to operating environment, uncontrolled operation conditions may cause stack degradation. Specifically, a method based on Monte Carlo simulation is proposed to identify the key parameter boundaries. Then the gas distribution in FDS due to nitrogen crossover is analyzed in detail. After this, a hybrid robust methodology based on sliding mode algorithm is also proposed to maintain adequate hydrogen pressure supply, suitable hydrogen and nitrogen content in the system in presence of nitrogen crossover and renewed uncertainties. Finally, the performance of the presented controller is compared with nonlinear PID (NPID) control and nonlinear multi-input-multi-output (NMIMO) control through a hardware-in-the-loop test bench. Experimental results show that the hybrid controller is accurate and suitable for control purpose, the nitrogen content is restricted to the given range and the variation of output voltage is limited to the desired boundaries, the feasibility and effectiveness are validated.

An analytical model for hydrogen and nitrogen crossover rates in proton exchange membrane fuel cells

In this paper, the hydrogen and nitrogen crossover through the membrane in proton exchange membrane fuel cells, are investigated by developing a semi-empirical analytical model. Different factors that affect the gas crossover rates were considered including pressure drop in gas diffusion layer (GDL) and catalyst layer (CL), operating temperature, relative humidity (RH) of the reactants, GDL compression, and the current density effect on the membrane temperature. The model is validated by published experimental data. It is found that RH is the most important parameter, followed by temperature. The hydrogen pressure drop through GDL and CL greatly depends on the GDL substrate properties, microporous layer (MPL) and CL. When permeability is low, an increase in current density reduces gas crossover. GDL compression, when MPL is used, was found to have a low impact on gas crossover. Gas crossover is improved with current density due to an increase in membrane temperature.

Two-dimensional simulation of cold start processes for proton exchange membrane fuel cell with different hydrogen flow arrangements

Proton exchange membrane (PEM) fuel cells with an off-gas recirculation anode (ORA) or dead-ended anode (DEA) are widely adopted in engineering. However, those two hydrogen flow arrangements may cause anodic water and nitrogen accumulation in comparison with the flow-through anode (FTA) mode, which causes significant performance degradation. In this paper, a two-dimensional cold-start model is developed with detailed consideration of water phase changes and the nitrogen crossover phenomenon. A simplified electrochemical module is built to calculate the current density distribution in the model. The simulation results are consistent with the experimental data at both subzero temperatures and normal operating temperatures. The effects of hydrogen flow arrangements, flow configurations, and startup strategies are investigated during startup process from subzero to normal operating temperatures. Much less ice is generated in counter-flow cases than in co-flow cases during constant current operation. A relatively lower startup voltage can effectively shorten the cold-start process and enhance the cold-start capacity for the PEM fuel cell. The ORA mode has the best hydrogen flow arrangement due to its general abilities, including higher hydrogen utilization efficiency, higher anodic nitrogen tolerance, better output performance and better startup capability.

Electrochemical Hydrogen Separation from Reformate Using High-Temperature Polybenzimidazole (PBI) Membranes: The Role of Chemistry

Various phosphoric acid (PA)-doped polybenzimidazole (PBI) membranes, para-PBI, m/p-PBI, and meta-PBI, were prepared via the poly(phosphoric acid) (PPA) process. These three membranes showed high levels of PA (10–32 PA/PBI repeat unit (r.u.)) and proton conductivity (0.14–0.26 S/cm at 180 °C) as compared with a conventionally imbibed meta-PBI membrane (6 PA/PBI r.u. and 0.08 S/cm at 180 °C). By controlling chemistry and increasing the polymer solid content to ∼18 wt %, m/p-PBI and meta-PBI membranes exhibited significantly improved creep resistance (<2 × 10–6 Pa–1) compared to para-PBI (10 × 10–6 Pa–1). In this work, various chemistries of PBI have been investigated to understand how the chemistry affected the electrochemical hydrogen separation (EHS) performance, including voltage requirement, power consumption, efficiency, hydrogen purities, and also long-term durability of the MEAs. Reformate streams containing H2, N2, and CO were used to validate the increased utility of this technique when operating at 160–200 °C due to the increased Pt tolerance to CO. The EHS device based on PBI membranes synthesized via the PPA process can be operated using dilute hydrogen feed streams with large amounts of CO (1–3%), producing fairly pure hydrogen products (>99.6% with <0.4% nitrogen crossover and ppm levels of CO) with very high power efficiencies of up to ∼72%.

Optimization of the performance, operation conditions and purge rate for a dead-ended anode proton exchange membrane fuel cell using an analytical model

Operating proton exchange membrane fuel cells in dead-ended anode mode results in fewer subsystem components and a lower cost and less complex system. However, dead-ended operation results in a gradual accumulation of water and impurities within the anode compartment, which leads to performance degradation. Therefore, anode purging is required to partially remove impurities and water from the anode and to recover performance. In the present study, a mathematical model, incorporating nitrogen crossover from the cathode to the anode and water build-up in the anode is developed. This model simulates the dead-ended anode proton exchange membrane fuel cell performance during the purge and the subsequent performance recovery. The model is in good agreement with experimental results. By using this model and introducing the concept of the ‘total wasted energy’, the purge parameters (purge interval and purge duration) can be optimized. The predicted optimum purge duration and purge interval for a sample single cell are 25 ms and 260 s, respectively. The effect of operating condition parameters on this optimization are investigated, showing that the hydrogen purity has a strong effect on the total wasted energy. By increasing the hydrogen purity from 99.5% to 99.99%, the efficiency increases by 2.4%.

A Quasi-2D Transient Multiphase Modeling of Cold Start Processes in Proton Exchange Membrane Fuel Cell

&lt;div class="section abstract"&gt;&lt;div class="htmlview paragraph"&gt;It’s well known that startup process of proton exchange membrane fuel cells (PEMFCs) under subzero temperature is extremely significant because of its influence on fuel cell performance and durability. In the study, a quasi-2D numerical model is developed and dynamic equations of mass conservation, energy conservation, membrane water conservation, ice conservation, species conservation are all considered. Three different hydrogen supply modes are studied in detail: flow-through anode (FTA) mode, dead-ended anode (DEA) mode and off-gas recirculation (OR) mode. It is found that the local current density (LCD) and temperature distribution vary remarkably along flow channel in OR mode as t &amp;gt; 500s due to nitrogen crossover and accumulation. During the cold start operation, the DEA mode and OR mode hold more water in anode catalyst layer (ACL) which reduces the effects of hydraulic permeation, resulting in more ice formation in cathode catalyst layer (CCL) and slower temperature rising. In term of hydrogen utilization ratio, the efficiencies in DEA mode and OR mode are close and much better than that in FTA mode. Compared with DEA mode and FTA mode, characteristics of OR mode can be summarized as the highest operating efficiency, almost the best output performance and the second shortest startup time.&lt;/div&gt;&lt;/div&gt;

Anode bleeding experiments to improve the performance and durability of proton exchange membrane fuel cells

Cost, durability, efficiency and fuel utilization are important issues that remain to be resolved for commercialization of proton exchange membrane fuel cells (PEMFC). Anode flow mode, which includes recirculation, dead-ended and exit bleeding operation, plays an important role in fuel utilization, durability, performance and the overall cost of the fuel cell system. Depending on the flow mode, water and nitrogen accumulation in the anode leads to voltage transients and local fuel starvation, which causes cell potential reversal and carbon corrosion in the cathode catalyst layers. Controlled anode exit bleeding can avoid the accumulation of nitrogen and water and improve fuel utilization. In this study, we present a method to control the bleed rate with high precision in experiments and demonstrate that hydrogen utilization as high as 0.9988 for a 25 cm2 single cell and 0.9974 for an 8.17 cm2 single cell can be achieved without significant performance loss. In the experiments, anode pressure is kept at 1 bar higher than the cathode pressure to decrease nitrogen crossover from the cathode, decreasing the crossover from the cathode. Moreover, four load cycle profiles are applied to observe the cumulative loss in the electrochemical surface area (ECSA), which are acquired from cyclic voltammetry (CV) analysis. Experiments confirm that the ECSA loss and severe voltage transients are indicative of fuel starvation induced by prolonged dead-ended or low exit-bleed operation modes whereas bleed rates that are larger than the predicted crossover rate are sufficient to operate the fuel cell without voltage transients and detrimental ECSA loss.

Effect of operating conditions on performance of proton exchange membrane fuel cell with anode recirculation

Anode recirculation operation is one method to both achieve self-humidification and increase hydrogen utilization for proton exchange membrane fuel cells (PEMFCs) automotive application. In this study, effect of operating conditions, including cathode inlet relative humidity, anode and cathode inlet pressure, and anode stoichiometry on performance of PEMFC with anode recirculation are analyzed by simulation work. Generally, the performance variation when the PEMFC starts anode recirculation can be divided into two stages. The performance increases first due to the self-humidification effect and then decreases due to the nitrogen crossover. The results show that performance enhancement caused by self-humidification becomes less significant with increasing cathode inlet humidity. Increasing anode inlet pressure can retard the performance decline caused by nitrogen crossover and increasing cathode inlet pressure will exacerbate it. Increasing anode stoichiometry can enhance the self-humidification by anode recirculation.

Anode recirculation and purge strategies for PEM fuel cell operation with diluted hydrogen feed gas

Commercial polymer electrolyte membrane (PEM) fuel cell systems require pure hydrogen feed gas (ISO 14687-2), otherwise impurities and inert gases would accumulate. Inert gases are difficult to remove, but do not hazard the fuel cell stack itself. Therefore, two purge strategies are introduced and experimentally investigated which enable fuel cell operation with up to 30 vol.% nitrogen content in the feed gas. Both strategies use a commercial on-line hydrogen sensor at the stack outlet either to trigger a discontinuous purge or to control the purge valve continuously. The experimental results show that the discontinuous purge strategy can be applied up to 10 vol.% nitrogen content in the feed gas. The continuous purge strategy was successfully operated with up to 30 vol.% nitrogen content and achieved the theoretical maximum fuel efficiency between 80 and 100%. The influence of nitrogen crossover on fuel efficiency and operating performance was investigated and found negligible. To sum up, the new continuous purge strategy offers an efficient, easy-to-implement, and robust solution to operate polymer electrolyte membrane fuel cell systems with up to 30 vol.% nitrogen content in the feed gas.

A quasi-2D transient model of proton exchange membrane fuel cell with anode recirculation

Anode recirculation of proton exchange membrane fuel cell (PEMFC) is considered as an effective way for self-humidification and efficient hydrogen utilization, but nitrogen crossover presents a problem. To investigate the transient characteristics of PEMFC with anode recirculation, a comprehensive quasi-2D non-isothermal transient model is developed. The simulation results show that for a PEMFC initially operated with dry hydrogen, anode recirculation leads to certain level of performance increase in the beginning due to self-humidification effect, and then the performance decreases caused by nitrogen crossover, hydrogen dilution and the increase of anode activation loss. The nitrogen accumulation in anode downstream is severer than upstream, and the performance decline rate decreases with increasing cathode inlet velocity under the simulated operating conditions. The performance and current density distribution uniformity of counter-flow configuration is better than co-flow under low inlet humidity, but the difference becomes smaller with anode recirculation. Anode recirculation and dead-ended anode (DEA) operation both have self-humidification effect, and compared to DEA, the advantages of anode recirculation can be summarized as lower performance decline rate, less chance of local hydrogen starvation and better current density distribution uniformity.

Purge strategy optimization of proton exchange membrane fuel cell with anode recirculation

Anode recirculation increases hydrogen utilization but also causes nitrogen crossover and accumulation in proton exchange membrane fuel cell (PEMFC) anode. Purge is necessary to remove the impurity. Voltage-based and nitrogen-based are the two basic purge strategies. For voltage-based purge, purge interval is defined by the voltage drop rate of the voltage peak. The voltage recovers first and then drops in the excess purge duration, so the optimal purge duration is defined as the purge stops when the voltage starts falling. The optimal purge duration is mainly determined by scavenging velocity, and it decreases with increasing scavenging velocity. Energy efficiency and fuel loss rate both increase with decreasing purge interval for the simulated operating conditions. Scavenging velocity significantly affects the fuel loss rate but has little effect on energy efficiency under the optimal purge duration. For nitrogen-based purge, the effect of purge duration on energy efficiency is much less significant than purge interval. Due to the difficulty of the real-time nitrogen fraction measurement, the voltage-based purge is more recommended. An optimal bleed rate for energy efficiency exists and 3% bleed rate is the optimal for the simulated operating conditions.

A real-time capable quasi-2D proton exchange membrane fuel cell model

In this paper a dynamic proton exchange membrane fuel cell model for real-time applications is presented. Following a quasi-2D approach, effects such as multicomponent diffusion in porous layers, membrane water transport driven by diffusion and electro-osmotic drag as well as membrane nitrogen crossover forced by partial pressure differences, are considered. A linearisation of the governing equations with respect to the previous time step is applied to avoid numerically expensive Newton iterations and to speed up the simulation. Furthermore, a solution method based on Chebyshev collocation minimises the required number of nodes and assures real-time capability. The model is validated in terms of polarisation curves, current density and species distribution versus steady-state computational fluid dynamics simulations of a 3D fuel cell performed in AVL Fire™. The transient behaviour is found to be in good qualitative agreement with results published by other authors. Due to the fast computation capability of the presented model, it is suitable for widespread parameter studies, control unit adjustments or state predictions, e.g. fuel starvation or membrane drying and flooding.

Local potential evolutions during proton exchange membrane fuel cell operation with dead-ended anode – Part I: Impact of water diffusion and nitrogen crossover

Abstract Operating a PEMFC with a dead-ended anode may lead to local fuel-starvation because of water and possibly nitrogen accumulation in the anode compartment. In previous works, we used a segmented linear cell with reference electrodes to monitor simultaneously the local potentials and current densities during dead-ended anode operation. The results indicated that water transport as well as nitrogen crossover through the membrane were most probably the two key factors governing fuel starvation. In this first from a set of two papers, we evaluated with more details the contributions of nitrogen crossover and water transport to hydrogen starvation. To assess nitrogen contribution, the fuel cell cathode compartment was first supplied with pure oxygen instead of air. The results showed that in the absence of nitrogen (in the cathode side) the fuel starvation was much slower than with air, suggesting that nitrogen contribution cannot be neglected. On the other hand, the contribution of water flooding to hydrogen starvation was investigated by using different cooling temperature on the cathode and anode sides in order to drive water toward the colder plate. The results showed that with a colder anode side, fuel starvation was faster. In the opposite case of a hotter anode plate, water accumulation in the anode compartment was limited, nitrogen crossover through the membrane was the main reason for hydrogen starvation in this case. To fully assess the impact of the thermal configurations on membrane-electrode assembly (MEA) degradation, aging protocols with a dead-ended anode and a fixed closing time were also performed. The results showed that operation with a hotter anode could help to limit significantly cathode ElectroChemical Surface Area (ECSA) losses along the cell area and performance degradation induced by hydrogen starvation.

Local potential evolutions during proton exchange membrane fuel cell operation with dead-ended anode – Part II: Aging mitigation strategies based on water management and nitrogen crossover

Abstract Proton exchange membrane (PEM) fuel cells operate with dead-ended anode in order to reduce system cost and complexity when compared with hydrogen re-circulation systems. In the first part of this work, we showed that localized fuel starvation events may occur, because of water and nitrogen accumulation in the anode side, which could be particularly damaging to the cell performance. To prevent these degradations, the anode compartment must be purged which may lead to an overall system efficiency decrease because of significant hydrogen waste. In the second part, we present several purge strategies in order to minimize both hydrogen waste and membrane-electrode assembly degradations during dead-ended anode operation. A linear segmented cell with reference electrodes was used to monitor simultaneously the current density distribution along the gas channel and the time evolution of local anode and cathode potentials. To asses MEA damages, Platinum ElectroChemical Surface Area (ECSA) and cell performance were periodically measured. The results showed that dead-end mode operation with an anode plate maintained at a temperature 5 °C hotter than the cathode plate limits water accumulation in the anode side, reducing significantly purge frequency (and thus hydrogen losses) as well as MEA damages. As nitrogen contribution to hydrogen starvation is predominant in this thermal configuration, we also tested a microleakage solution to discharge continuously most the nitrogen accumulating in the anode side while ensuring low hydrogen losses and minimum ECSA losses provided the right microleakage flow rate is chosen.