Experimental Design of High-Performing Open-Cathode Polymer Electrolyte Membrane Fuel Cells

Open-cathode polymer electrolyte membrane fuel cells (PEMFCs) utilize a unique air-cooled system design to eliminate the humidifiers, air compressor, and liquid cooling loop of conventional, liquid-cooled PEMFC systems, thereby greatly reducing system cost. However, the open-cathode PEMFC performance is restricted by poor humidification, high membrane and charge transfer resistances, and overheating due to inefficient thermal and water management. This work aims to strategically modify the membrane electrode assembly (MEA) design to overcome these issues and achieve high open-cathode PEMFC performance that approaches that of liquid-cooled systems. The use of thinner membrane along with short side chain ionomer is found to elevate the cell performance due to increased water retention at the cathode catalyst layer (CCL) and decreased ohmic losses. Thinner gas diffusion layers with high porosity enable additional cell performance increment by improving oxygen availability at the CCL. An overall current density rise of 88% at 0.6 V and 53% at 0.4 V is achieved by the strategically designed MEA for open-cathode cells. The enhanced power density enabled by the custom MEA can both reduce the stack cost and expand the power range of open-cathode PEMFCs, thus expanding their potential use for low-cost fuel cell system applications.


CCL
Conventional polymer electrolyte membrane fuel cell (PEMFC) systems operate at moderate pressure with fully humidified reactant gases and liquid coolant to maintain the stack temperature. 1For efficient fuel cell operation, this system requires various balance of plant (BoP) components including an air compressor, humidifiers, and heat exchangers to meet the pressure, temperature, and thermal balance demands.However, these BoP components are costly and consume additional power which falls under the auxiliary losses of the system. 2 The total cost includes the capital cost for installing such BoP components and the associated operational and maintenance costs.The BoP components constitute nearly 35% of the total cost of the system for transportation applications. 3pen-cathode fuel cells have been gaining popularity for portable applications and other mid-range power applications as an alternative low-cost system. 4In open-cathode cells, dry hydrogen is generally fed on the anode side and one or more fans in front of the cells blow ambient air on the cathode side by forced convection. 5he air blown into the system through the open channels at the cathode not only provides oxygen (O 2 ) for the electrochemical reaction but also cools down the cell and maintains the internal temperature through heat extraction. 6The need for liquid cooling, air compression, and humidification is thus eliminated in contrast to conventional PEMFCs and the system design is greatly simplified. 7owever, the cell overheating caused by inefficient heat removal, 8,9 membrane dehydration due to low humidity operations, 10 high ohmic resistance possessed by increased membrane ionic resistance, and lower cell kinetics 11 are some of the key challenges associated with open-cathode fuel cell systems.This adversely affects the power performance of the fuel cell stack and also warrants more customized stack designs, provided that the airflow rate is the only means of active system control.
In general for PEMFCs, the humidification of the ionomer in the membrane and catalyst layers (CLs) plays an important role in the overall cell performance.The water uptake dynamics of the ionomer are strongly temperature dependent. 12At high relative humidity (RH) conditions where humidification is sufficient; a CL with a high electrochemical surface area (ECSA) catalyst is generally expected to perform better than a low ECSA catalyst given that the Pt loading and ionomer to carbon ratio remain the same for both cases.Whereas, at dry operating conditions even the increase in ECSA is not able to boost the cell performance due to poor protonic conduction. 13The water sorption/desorption rate constant (γ) of the ionomer was therefore recently hypothesized to be an influential factor for the overall performance of open-cathode PEMFCs. 14ased on predictions from computational modelling of an opencathode cell, lower γ ionomer is anticipated to support the cell performance by providing better water retention capability within the ionomer phase of the CL. 15 To date however, there are no experimental reports available to corroborate these predictions.The equivalent weight (EW) of the ionomer has also been reported to influence the cell performance for open-cathode systems.The use of a high EW ionomer with 1100 EW over 850 EW was found to possess improved cell performance due to lower mass transport resistance as determined by electrochemical impedance spectroscopy (EIS), but at the expense of lower proton conductivity. 16Also, low EW ionomers were reported to have high O 2 transport resistance and high ECSA at the same time as compared to high EW ionomers z E-mail: ekjeang@sfu.ca*Electrochemical Society Member.
ECS Advances, 2024 3 014504 while operating under both dry and wet conditions in conventional liquid-cooled PEMFCs. 17,18Similarly, Garsany et al. reported that short side chain (SSC) ionomers with lower EW produce superior cell performance compared to long side chain (LSC) ionomers with higher EW for RH conditions ranging from 50% to 100%. 19While these reports are important for conventional liquid-cooled PEMFCs, the findings may not be transferable to the unique operating conditions and local environment within open-cathode cells.The effect of membrane selection is also lacking in the literature for open-cathode cells, whereas it is relatively well established that thin, reinforced membranes are favourable for high performing liquidcooled cells due to reduced ohmic resistance. 20,21as diffusion layers (GDLs) may also play an important role in determining the overall cell performance by facilitating water and gas transport across the cell. 225][26] The inclusion of an MPL was also found to provide improved voltage stability. 27However, the extent of potential MPL benefits for open-cathode cells has not yet been established.Within the GDL, the porosity available for gas phase transport is dependent on the presence of liquid water, which may cause flooding in conventional liquid-cooled PEMFCs.Hence, high GDL porosity is found to support the cell performance by increasing the O 2 transport at high current densities.However, the GDL porosity typically has less influence on the polarization level at medium or low current density (CD). 28,29The GDL thickness is also an important parameter.Reduction in GDL thickness was found to support the cell performance of conventional PEMFCs as a consequence of reduced mass transport resistance of the liquid and gaseous flow. 25On the contrary, Zhou et al. 30 suggested the use of thinner GDLs to facilitate higher water content and improved hydration for the membrane.A similar modelling study by Jeng et al. 31 reported the dependence on PEMFC performance based on the GDL porosity and thickness.At low GDL porosity, a reduced thickness was found supportive of the cell performance.Whereas, at high GDL porosity the requirement of an optimal thickness of the GDL showed a reverse trend.3][34][35][36] Overall, the literature is rich on contributions investigating the effect of various GDL parameters such as porosity, thickness, hydrophobicity, and pore size on the performance of conventional liquid-cooled PEMFCs.8][39] In the case of open-cathode PEMFCs with combined cathode air flow and cooling channels, Atkinson et al. 40 reported the impact of GDL porosity on the hydration levels of the cell, electrical resistance, and thermal profile.Cathode charge transfer, cell hydration, and thermal management were reported to be improved with small decrement in GDL porosity.However, recent modelling results from our group 41 suggested the contrary effect of MPL selection in the GDL with highly porous, thin MPL being advantageous for open-cathode PEMFC operation due to reduced O 2 transport resistance.The need for further research to explore the GDL influence on various cell related parameters for open-cathode PEMFCs is therefore evident.

Materials
A total of five MEAs are fabricated for evaluating the effect of different materials by changing one of the components of the MEA at a time.The baseline MEA for this work, MEA-1, is taken as the commercial MEA procured from Ion Power Inc. with specifications given in Table I.It is a catalyst coated membrane (CCM) based MEA with an active area of 25 cm 2 having 60% Pt/C on Vulcan type carbon support and Nafion type LSC ionomer and membrane.MEA-2, MEA-3, MEA-4, and MEA-5 are prepared in-house by following similar methods with specifications as in Table I.The ionomer-tocarbon ratio is fixed at 1.75 for all these MEAs.The MEA-2 is prepared with similar composition as MEA-1 to establish the inhouse MEA preparation capability and validate the baseline results.ECS Advances, 2024 3 014504 with eight bolts across the plates which are tightened by providing manual torque of 4 Nm at each of the bolts.The MEA is placed between the two graphite plates consisting of spray coated CCM with GDL on both sides.The edges of the MEA are sealed by using a Teflon gasket on both sides of the membrane having a suitable thickness.

The effect of SSC ionomer on the cell performance of open-cathode
The MEAs are tested in the open-cathode single-cell setup at an operating ambient air condition of 40 °C and 40% RH.The test setup is placed in an environmental chamber connected with a Greenlight Innovation G400 test station inside which the ambient air condition is maintained.A small air compression device is installed inside the environmental chamber which delivers the air to the open-cathode channels through pipe fittings connected to the duct at ambient pressure.A manual rotameter is connected along the pipe fittings to control the air flow rate at 2.5 nlpm for all the experimental conditions at all current densities.This cathode side flow rate corresponds to air stoichiometry of nearly 12 at the operating current density of 0.5 A cm −2 for the current MEA active area of 25 cm 2 .The anode side uses 99.999% pure H 2 fed at an operating temperature of 40 °C and ambient pressure.The flow is kept at 0.5 nlpm with RH of the incoming gas maintained at 60% mimicking dead-end mode operating conditions, which are commonly used in open-cathode PEMFC systems. 9,46The single-cell setup is enclosed by glass wool across all the peripheries to suppress any loss of heat generated within the system and thus direct the heat dissipation to the air channels.The fuel cell tests are performed by measuring the steady state current density at fixed cell voltages of 0.6 V and 0.4 V until the thermal equilibrium is achieved for each case in 40-60 min.The thermal equilibrium is monitored by the temperature recorded as T1, T2, and T3 at the central open cathode channel at the cathode-GDL interface, center of the cathode side solid graphite plate, and the cathode endplate using thermocouples TC1, TC2, and TC3 respectively as reported in Fig. 2a.
The conditioning of individual MEAs is performed using the same active area liquid-cooled setup operating at 60 °C with three sets of 100 cyclic voltammetry scans performed at the scan rate of 50 mV s −1 , constant voltage operation at 0.6 V for 4-5 h, and polarization curve measurement before actual data collection on the open-cathode setup.The fuel cell current density is recorded for the open-cathode single cell for each of the MEAs while operating at 0.6 V and 0.4 V respectively.This method follows best practices for fuel cell polarization curve measurements to measure current at a fixed cell voltage until the current reaches an equilibrated condition, thus ensuring true steady state conditions for every point reported on the "curve."For open-cathode PEMFCs, this further implies reaching thermal and hygral equilibrium at each point, which is ensured in the present work.Three repeated measurements are taken on each individual MEA to ensure reproducibility of the results.EIS is performed in potentiostatic mode at 0.6 V for each of the MEAs with the frequency range scanned from 1 kHz to 0.1 Hz with AC perturbation voltage of 5 mV using Gamry Reference 5000 E with a booster connected to the Greenlight Innovation G400 test station for data acquisition.

Results and Discussion
Figure 2a shows the thermal equilibration data for MEA-3 which depicts the trends of T1, T2, and T3 measured using TC1, TC2, and TC3 while operating at 0.4 V where, T1, T2, and T3 are the temperature at the central open cathode channel at the cathode-GDL interface, center of the cathode side solid graphite plate, and the cathode endplate respectively.The internal temperatures at the cathode channel-GDL interface and graphite plate increase due to self-heating and reach nearly 52 °C while operating at the given current density, whereas the external cathode endplate remains at a proportionally lower temperature of around 40 °C which is near the ambient operating condition.The system is found to be equilibrated after approximately 50 min while operating at 0.4 V with a rise below 0.05 °C min −1 .The current density starts from an initial value of 0.44 A cm −2 and rises slightly until it equilibrates at 0.52 A cm −2 as depicted in Fig. 2b.As expected, due to self-heating, most notably within the MEA at the center of the cell, the rise in internal cell temperature appears to lag the rise and equilibration in CD, until a new thermal equilibrium is established under the steady state opencathode fuel cell operating condition.The internal cell temperature reaches a higher value at 0.4 V than at 0.6 V due to the higher CD and increased heat generation.Similar thermal equilibrium is achieved for each of the MEAs before recording the fuel cell data.The measured CDs for all the tested MEAs are captured in Fig. 3 while being operated at 0.6 V and 0.4 V along with error bars for the three repeats of each polarization data captured.
The average CD and the percentage change in CD with respect to the baseline MEA test data are presented in Table III for further understanding.For the in-house fabricated MEA-2 with a similar composition to that of the commercial MEA-1, the average CD is found to be nearly the same with a deviation of 6% at operating cell voltage of 0.6 V and 5% at 0.4 V.This trend validates the repeatability of the MEA performance at the same operating condition and similar MEA composition.The slight deviation can be attributed to the different methods of MEA preparation.MEA-3 is found to give a CD increment of 31% at 0.6 V and 30% at 0.4 V respectively with reference to the baseline performance of MEA-1.This increase in CD is attributed to the change in ionomer as compared to MEA-2, the Aquivion ionomer which has lower EW and short-side chain polymer structure with reference to the Nafion ionomer used in the baseline MEA.This improvement is likely due to better water retention capability at the ionomer sites in the CL 47 and may also be related to the low water sorption/desorption rate constant inherent to SSC ionomer. 14Next, the change in membrane from Nafion 212 to Aquivion 720 along with the ionomer change with respect to the baseline for MEA-4 leads to a significant improvement in CD by 75% at 0.6 V and 38% at 0.4 V respectively.The higher rate of increment observed at medium current density operation at 0.6 V is attributed to the decreased ohmic resistance due to the lesser membrane thickness of 20 μm used for MEA-4 as compared to 50 μm in the baseline MEA.Also, the use of Aquivion membrane in MEA-4 compared to Nafion membrane for the baseline may benefit from the improved water absorption capability of the Aquivion membrane pertaining to its SSC structure and low EW. 48,49This also adds to the increased water content at the membrane-CL region of the MEA which is contributing to the elevated CD.A further boost in CD by 13% at 0.6 V is found for MEA-5 as compared to MEA-4, whereas the increment at 0.4 V is 15%.This increase in CD is attributed to the modified GDL used in MEA-5, which is thinner and has higher porosity than the baseline GDL used in the other MEAs.More specifically, the modified GDL features a thickness reduction from 235 to 215 μm and a reduced areal weight from 90 to 70 g m −2 .The reduced GDL thickness and increased porosity are expected to jointly reduce the O 2 diffusion resistance in the MEA with increased O 2 availability at the active sites of the cathode CL (CCL). 15,41This leads to an enhanced rate of reaction and thus resulting in increased CD.
The measured average temperatures T1, T2, and T3 after reaching thermal equilibrium at 0.4 V operation are shown in Fig. 4. Overall, the cell temperature is observed to be higher for the MEAs with higher CDs and closely follows the trend established in the cell performance data.Accordingly, the highest temperature is observed for MEA-5, which is the highest performing MEA out of the five MEAs being tested.Similarly, the highest incremental temperature rise is observed for MEA-3 versus MEA-2, in agreement with its high CD increment stemming from the improved ionomer type.This indicates the direct dependence of temperature distribution on the CD which is further associated with the improvement in the MEA design.These findings are consistent with theoretical expectations, as the waste heat generation at a given cell voltage is proportional to CD, as well as numerical modelling predictions for open-cathode PEMFCs. 11,14The temperature increment may also benefit the overall cell performance through enhanced kinetics, but only if membrane hydration can be maintained.At 0.6 V, the 0.16 A cm −2 current density reached with the baseline MEA-1 in the open-cathode cell is roughly 60% below that of a conventional liquid-cooled and externally humidified closed-cathode cell operated at 50 °C, 50% anode RH, and 90% cathode RH with similar MEA architecture, as reported in our previous work. 11With the improved design of MEA-5, the open-cathode cell performance at 0.6 V is merely 30% below that of the closed-cathode cell, effectively reducing the gap in half.The corresponding gap in current density at 0.4 V is similarly reduced from 50% to 20%, showing that the improved MEA can effectively raise the opencathode cell performance closer to the expected level for conventional, closed-cathode cells. 11he Nyquist plot obtained from the EIS data for the five different MEAs is shown in Fig. 5 while operating at 0.6 V.The highfrequency resistance (HFR) decreases subsequently from MEA-1 to MEA-5.This trend is in agreement with the CD data obtained for these cases and asserts the performance increment for the subsequent MEA design.The HFR values for MEA-1 and MEA-2 are essentially equal with only 2% deviation and justifies the similar composition of the two MEAs being operated at the same ambient conditions.However, the second x-axis intercept representing the charge transfer resistance (R ct ) is lower for MEA-2 compared to MEA-1.This can be attributed to the difference in kinetic overpotential possessed by these two MEAs which is likely related to variations in catalyst microstructure.
The HFR reduces from 9.39 mΩ for MEA-2 to 7.26 mΩ for MEA-3, following the change in ionomer from Nafion D521 to Aquivion D72-25BS.However, the protonic resistance of the ionomer in the CLs is only a minor contribution to the combined ohmic cell resistance measured by HFR.Hence, the 23% HFR decrement for open-cathode PEMFC operation is more likely due to improved membrane hydration induced by ionomer related water retention under dry conditions.This is accompanied by a moderate reduction in R ct as another indication of ionomer related improvements in the CL performance.A further reduction in HFR to 3.45 mΩ is observed by replacing the 50 μm Nafion 212 membrane with a 20 μm Aquivion membrane in MEA-4.The low HFR is primarily attributed to the reduced membrane thickness which offers less protonic resistance when combined with good water retention facilitated by the same SSC ionomer with low EW in both membrane and CLs.This trend in HFR is believed to be a major contributing factor for the CD enhancement for the improved MEA designs.Consequently, the successive R ct decrements from MEA-2 to MEA-3 and further to MEA-4 by 23% and 9% respectively can be attributed to the increased cell temperature stemming from the enhanced CD with SSC ionomer (Fig. 4).The HFR of MEA-5 is similar to that of MEA-4; however, the R ct is found to decrease for MEA-5 as compared to MEA-4.This outcome is likely a result of enhanced O 2 availability at the active sites of the CCL in the presence of a more porous and thinner GDL. 15,41hese experimental results are comparable to our previously published modelling predictions 11,14 where a 3D computational fuel  cell model was used to establish the effect of ionomer property changes, membrane changes, and MPL/GDL modifications for opencathode PEMFCs.The model revealed the theoretical benefits of an ionomer with low water sorption/desorption rate constant (γ) to aid water retention at the CCL and thereby increasing the CD performance by up to 130% at 0.6 V. 14 Similar water retention capability is featured experimentally in the present work by use of SSC ionomer with low EW over conventional LSC ionomers.Both results point toward enhanced water retention capability and pave pathways for designing novel materials with low γ and EW.The experimentally observed benefits of a thin membrane, when paired with the same ionomer, also follows the trend predicted by modelling results due to less ohmic resistance offered by such membranes.The present experimental findings for the thinner, low-density GDL (MEA-5) are also corroborated by the modelling results, wherein a highly porous, thinner MPL was predicted to enhance CD performance by means of increased mole fraction of O 2 in the CCL.The magnitude of decrement in R ct achieved experimentally with MEA-5 is however constrained by the limited availability of desired commercial GDLs with high porosity and low thickness.It further opens the scope for designing novel materials for high-performing open-cathode fuel cell systems based on the collective predictions from both modelling and experimental works reported herein.

Conclusions
The effect of experimentally tuned membrane electrode assembly design for open-cathode PEMFCs was evaluated in this work for a 25 cm 2 single-cell open-cathode setup operated at the ambient condition of 40 °C and 40% RH.A total of five MEAs were tested at the given operating condition with a commercial MEA taken as the baseline.The other four MEAs were fabricated in-house and the subsequent effects of ionomer, membrane, and GDL changes were analyzed by comparing the average current densities obtained at fixed cell voltages of 0.6 V and 0.4 V.The three MEAs having SSC ionomer with low EW consistently achieved major performance improvement for the open-cathode cell by inducing better water retention capability at the CCL site which leads to improved protonic conduction in an otherwise dry environment.The subsequent introduction of a thinner membrane with the same low-EW SSC ionomer contributed a further CD increment of 44% at 0.6 V owing to drastically reduced ohmic cell resistance as measured by EIS.Lastly, a thinner GDL with a more porous structure was found to elevate the cell performance by an additional 15% at 0.4 V through improved oxygen transport and thus leads to a suitable MEA design for high-performing open cathode PEMFCs.Importantly, the tuned MEA designs were able to leverage the incremental temperature rise at high CDs toward improved kinetics, as manifested from reduced charge transfer resistance measured by EIS.This capability was enabled by the improved water retention of the modified ionomer and membrane in order to avoid or delay membrane dry out.The results obtained experimentally in this work were also corroborated by theoretical predictions from computational modelling results for similarly designed and operated open-cathode fuel cells.In our future work, efforts will be made to develop novel materials designed for the specific requirements of high-performing opencathode PEMFC systems using the insights obtained from this work, with the goal to approach the performance of conventional, liquidcooled PEMFCs.The enhanced power density enabled by the custom MEA can both reduce the stack cost and expand the power range of open-cathode PEMFCs, thus expanding their potential use for low-cost fuel cell system applications.
the present work addresses the critical gap in the literature on custom MEA design and component material selection specifically for opencathode PEMFCs.The objective of this work is to experimentally evaluate the opportunities for customized MEA design for opencathode PEMFC systems and determine the explicit role of strategic selection of CL ionomer type, membrane thickness, and GDL/MPL design on the overall cell performance of open-cathode fuel cells.The fabrication and testing of several MEAs are carried out and a tailored MEA configuration is proposed to enhance the performance of open-cathode cells by minimizing the limitations offered by a baseline MEA originating from conventional MEA design for liquidcooled fuel cells.A total of five different MEAs are tested to understand the individual as well as combined effects of ionomer, membrane, and GDL changes on the open-cathode cell performance using a single cell experimental setup with a 25 cm 2 active area.The results obtained for all the MEAs in the form of achieved CD at fixed cell voltages, impedance data, and temperature profiles are compared and assessed.The results are also compared with theoretical predictions obtained from numerical modelling of open-cathode fuel cells. 11,14 TableI.Details of the different MEAs prepared for open-cathode fuel cell testing.The modifications in each case are underlined.= 0.5 mg cm−2   m Pt = 0.5 mg cm −2 Nafion 212 (50 μm) (Commercial) GDL-SGL 29BC (235 μm; 90 g m −2 ) GDL-SGL 29BC (235 μm; 90 g m −2 ) Ionomer-Nafion (D521-1100 EW) Ionomer-Nafion (D521-1100 EW) MEA-2 m Pt = 0.5 mg cm −2 m Pt = 0.5 mg cm −2 Nafion 212 (50 μm) (In-house) GDL-SGL 29BC (235 μm; 90 g m −2 ) GDL-SGL 29BC (235 μm; 90 g m −2 ) Ionomer-Nafion (D521-1100 EW) Ionomer-Nafion (D521-1100 EW) MEA-3 m Pt = 0.5 mg cm −2 m Pt = 0.5 mg cm −2 Nafion 212 (50 μm) GDL-SGL 29BC (235 μm; 90 g m −2 ) GDL-SGL 29BC (235 μm; 90 g m −2 ) Ionomer-Aquivion (D72-25BS-720 EW) Ionomer-Aquivion (D72-25BS-720 EW) MEA-4 m Pt = 0.5 mg cm −2 m Pt = 0.5 mg cm −2 Aquivion-720-20 (20 μm) GDL-SGL 29BC (235 μm; 90 g m −2 ) GDL-SGL 29BC (235 μm; 90 g m −2 ) Ionomer-Aquivion (D72-25BS-720 EW) Ionomer-Aquivion (D72-25BS-720 EW) MEA-5 m Pt = 0.5 mg cm −2 m Pt = 0.5 mg cm −2 Aquivion-720-20 (20 μm) GDL-SGL 29BC (235 μm; 90 g m −2 ) GDL-SGL 22BB (215 μm; 70 g m −2 ) Ionomer-Aquivion (D72-25BS-720 EW) Ionomer-Aquivion (D72-25BS-720 EW) ECS Advances, 2024 3 014504 systems is studied by changing the ionomer material for MEA-3 as compared to MEA-2 while keeping other compositional parameters the same as for MEA-2.Subsequent change in the membrane by using Aquivion-720-20 is done for MEA-4 to analyze the effect of SSC membrane in conjunction with similar ionomer.As an additional step, in MEA-5 the GDL is changed from SGL 29BC to SGL 22BB on the cathode electrode to access the effect of thinner GDL with high porosity on open-cathode cell performance.The slurry ink is prepared by mixing Vulcan carbon supported 60% Pt/C of PK catalyst along with ionomer and solvent.The same catalyst is used for preparing all the in-house MEAs with fixed Pt loading of 0.5 mg cm −2 on both anode and cathode CLs.The catalyst is measured as per the loading of individual MEAs and wetted with 2-3 drops of deionized water with mixing done using a glass rod before preparation of the catalyst ink slurry.The respective ionomer is measured separately and added to the slurry dropwise.5 wt% of Nafion D521-1100 EW is taken as the initial ingredient for ionomer solution while preparing MEA-2 whereas 25 wt% of Aquivion D72-25BS is used for MEA-3, 4, and 5.A mixture of 2 ml of deionized water and 18 ml of ethanol is used as solvent for preparing the slurry ink solution required for coating each side of the membrane.The process of slurry ink formation involves the first step of adding half of the solvent to the weighed catalyst and ultrasonic stirring for 15 min.In the next step, the ionomer is added with the remaining solvent and the solution is then subjected to ultrasonic stirring for another 20 min.The same process is followed for each MEA preparation.Once the catalyst ink slurry is ready, it is sprayed on the desired PEM using Flair Stainless Steel Multipurpose Air Brush Paint Spray Gun on both sides.While spraying the ink solution on the PEM, a temperature of 60 °C is maintained at the base plate where the PEM is placed to ensure uniform drying of the ink.The spray gun is operated at a pressure of 20 psi using dry nitrogen gas for a uniform flow of slurry and to achieve uniform contact of the slurry onto the membrane.Both anode and cathode CLs are spray coated using the same procedure.Experimental Setup and Test ProcedureThe performance of the five MEAs is evaluated by performing single-cell tests on an open-cathode PEMFC with an active area of 25 cm 2 .The present small-scale experimental setup is inherently designed to represent the fuel cell and internal thermal equilibrium conditions of larger open-cathode fuel cell stacks with airflow from ambient air driven by a fan.The single-cell test setup shown in Fig.1is equipped with an external duct made up of PA2200 Nylon to carry the ambient air onto the open channels on the cathode side.Ethylene propylene diene monomer rubber gaskets are used at the junction of the duct and the graphite plates open channels to arrest leaks across the duct.The single-cell setup consists of graphite plates with serpentine channels on the anode side and open channels grooved on the cathode side plate with dimensions as listed in TableII.The graphite plates are compressed in a single cell fuel cell hardware consisting of gold plated current collectors and aluminium end plates

1 .Figure 2 .
Figure 2. (a) Thermal equilibration data for MEA-3 while operating at 0.4 V where, T1, T2, and T3 are the temperature at the central open cathode channel at the cathode-GDL interface, center of the cathode side solid graphite plate, and the cathode endplate respectively.(b) Current density at 0.4 V for MEA-3 during thermal equilibration.

Figure 3 .
Figure 3. Current density for the five MEAs operated at cell voltages of 0.6 V and 0.4 V at steady state, following thermal equilibration.

Figure 4 .
Figure 4. Average temperatures (T1, T2, and T3) reached at thermal equilibrium for the five MEAs at 0.4 V operation.

Figure 5 .
Figure 5. Comparative Nyquist plot of impedance for the five MEAs obtained by in situ EIS on the single cell open-cathode fuel cell operated at 0.6 V.