Table of contents

The development of complex mesoscale (nm - µm) materials used for electrochemical applications requires comparable progress in the analytical instruments and techniques in order to understand the physical and chemical structure-property relationships underlying their performance. Conventional "hard" X-ray (i.e. > 10 keV) scattering has received considerable attention due to the fact that it is a high-resolution nondestructive structural probe that can interrogate a statistically significant 3-dimensional sample area. The non-resonant nature of this scattering process limits its applicability to materials that possess significantly different electron densities. Unfortunately, the performance of many electrochemical materials hinges on subtle heterogeneities that do not possess a high electron density contrast such as interfacial nanostructures, impurities, and chemical composition gradients. To help address this challenge, resonant soft X-ray scattering (RSoXS) uses tunable "soft" X-rays (100 - 2000 eV) to dramatically enhance the scattering cross sections from heterogeneous materials when the X-ray photon energy is judiciously chosen to coincide with favored transitions near a material's absorption edges. The RSoXS results in Fig. 1a show an example of how we used the resonance-enhanced scattering signals at selected photon energies to isolate the scattering contribution from different polymers in a phase separated block copolymer in order to unambiguously define the complex morphology of a triblock copolymer sample with both chemical and nm-scale spatial sensitivity.¹

In this presentation, we reveal how operando RSoXS can be a powerful reciprocal space probe for mesoscale electrochemistry due to its chemical sensitivity, large accessible size scale, and polarization control.^{2, 3} We will convey how this technique can be applied under operando conditions to study pores, surfaces,⁴ and buried interfaces⁵ of low-Z element materials⁶ including many transition metals; the practical considerations of conducting such experiments will also be discussed. We will explain how the intrinsic combination of scattering and spectroscopy allows us to monitor spatio-chemical changes at a specific location by detecting the change in intensity at a specific scattering vector, q, for X-ray energies that are both ON and OFF resonance with the species of interest (Fig. 1b).

As an example of the utility of RSoXS to electrochemical applications, we present recent results on Nafion, a perfluorinated sulfonic acid (PFSA) membrane material that is considered to be a critical cost and performance-limiting component in many devices including fuel cells, electrolyzers, and redox-flow batteries. Recent RSoXS results acquired with a wet sample cell interrogated the Nafion films' partially orientated molecules inside ionomer domains. Using polarized X-rays with a photon energy tuned to the fluorine absorption edge (~690 eV), we observed a surprisingly strong scattering anisotropy that indicated preferred local crystalline grain orientation at the interface between different phases, an effect which is not visible when the X-ray photon energy is off-resonance with the fluorinated ionomers (Fig. 1c). These results enable us to develop a full electron density map that helps us understand why the pore structure of Nafion works so well, but may also yield insights into whether the development of porous separators as alternatives to PFSAs require pore sizes that are comparable to the hydrophilic channels in PFSAs (e.g., ≤ 3 nm).⁷ We will then expand on how combining such operando RSoXS data with electrochemical analytical methods could uncover important dynamic structure-property relationships underlying the interplay of various factors such as migration and electro-osmosis, chemical/physical stability, water uptake, permeability, etc.^8-10

References

1. C. Wang, D. H. Lee, A. Hexemer, M. I. Kim, W. Zhao, H. Hasegawa, H. Ade and T. P. Russell, Nano Lett, 2011, 11, 3906-3911.

2. B. A. Collins, J. E. Cochran, H. Yan, E. Gann, C. Hub, R. Fink, C. Wang, T. Schuettfort, C. R. McNeill, M. L. Chabinyc and H. Ade, Nat Mater, 2012, 11, 536-543.

3. S. C. B. Mannsfeld, Nat Mater, 2012, 11, 489-490.

4. J. Schlappa, C. F. Chang, Z. Hu, E. Schierle, H. Ott, E. Weschke, G. Kaindl, M. Huijben, G. Rijnders, D. H. A. Blank, L. H. Tjeng and C. Schussler-Langeheine, J Phys-Condens Mat, 2012, 24.

5. M. Nayak, P. C. Pradhan and G. S. Lodha, Sci Rep-Uk, 2015, 5.

6. C. Wang, T. Araki and H. Ade, Appl Phys Lett, 2005, 87.

7. M. L. Perry and A. Z. Weber, J Electrochem Soc, 2016, 163, A5064-A5067.

8. R. M. Darling, A. Z. Weber, M. C. Tucker and M. L. Perry, J Electrochem Soc, 2016, 163, A5014-A5022.

9. X. L. Wei, B. Li and W. Wang, Polym Rev, 2015, 55, 247-272.

10. Y. S. Kim and K. S. Lee, Polym Rev, 2015, 55, 330-370.

Figure 1

Land-channel geometry is necessary for proton exchange membrane fuel cells to transport electron and at the same time to transport reactants/products. However, this configuration causes the difference in transport distance between the flow channel to the catalyst layer, and results in the non-uniform distribution of various factors, such as species concentration, current generation, and rate of degradation. In order to investigate the distributions of various parameters in the land-channel direction, a small-scale segmented cell with about 300-micron resolution was developed by the authors' group, and successfully measured the current and high-frequency resistance distribution in the land-channel direction. Distribution of oxygen transport resistance from the flow channel to the catalyst layer was also measured using limiting current technique.

The oxygen transport mechanism is completely different in the conventional flow field and in the interdigitated flow field. In the conventional flow field, the oxygen transport is mostly driven by the duffusion due to the concentration gradient. In the interdigitated flow field, on the other hand, the convective flow through gas diffusion layer (GDL) makes the diffusion distance shorter than the conventional flow field case, and therefore the lower oxygen transport resistance is expected.

In this study the oxygen transport resistance distribution in land-channel direction is measured in two different flow field configurations, that is, conventional flow field and interdigitated flow field. The comparison of the results from these two flow fields reveals the impact of the convective transport through GDL on the oxygen transport resistance and on the distribution.

Since most of the PEMFCs have land-channel geometry in their flow field, the distance between the flow channel to the catalyst layer is not uniform. The distance is longer from the flow channel to the catalyst layer under the land area than to the catalyst layer under the channel area. This difference causes the non-uniform current density in land-channel direction, and the distribution largely depends on the operating conditions. As the inlet gas humidity in the operating condition decreases, the trend in the current density distribution changes from high local current under the channel to relatively uniform, and to higher local current under the land (Figure 1). This trend is due to the combined effects of (1) non-uniform oxygen transport resistance, (2) non-uniform liquid water transport distance causing the local flooding under the land area at wet condition, and (3) the tendency of faster membrane dryout under the channel area at dry condition.

In order to investigate the distributions of various parameters in the land-channel direction, a small-scale segmented cell with about 300-micron resolution was developed by the authors' group, and successfully measured the current and high-frequency resistance distribution in the land-channel direction. Distribution of oxygen transport resistance from the flow channel to the catalyst layer was also measured using limiting current technique.

In this study a new segmented cell is developed. Using the transparent cathode plate the behavior of liquid water in cathode flow channel can be observed, and simultaneously the current density distribution is measured in land-channel direction. With this cell the relation between the current density distribution in the land-channel direction and the presence of liquid water is investigated. First, it is studied that at what gas humidity and at what overall current density the liquid water starts emerging from the gas diffusion layer surface to the flow channel. This provides us the information about the dependency of the liquid water emergence location on the current density distribution is studied. Then, how the presence of liquid water in the flow channel changes the current distribution is studied.

Figure 1

In recent years phosphoric acid doped PBI-type fuel cells have drawn much attention as a promising candidate for energy storage and conversion applications at relatively high temperatures (100 – 200 °C). The elevated operating temperature of HT-PEMFCs (high-temperature polymer electrolyte membrane fuel cells) simplifies water and thermal management. In addition, it greatly enhances the fuel cell's tolerance against impurities (e.g. CO) in hydrogen, which is critical for operation with reformate hydrogen. However, durability and stability of high-temperature PEM-MEAs still need to improve for widespread commercialization [1], and better tools for identifying performance-loss related mechanisms are desirable. A very useful in-situ technique for performance analysis of HT-PEMFCs is electrochemical impedance spectroscopy (EIS). As previously reported, EIS can be used to characterize the impact of different parameters (e.g. stoichiometry, temperature, etc.) on cell kinetics [2, 3]. Equivalent circuit models are usually used to analyze the EIS data and identify the performance-loss related mechanisms.

A major drawback of this approach is that the assumptions for the model equivalent circuits are sometimes ambiguous or even misleading due to the lack of priori knowledge about the electrochemical system under study. Furthermore, a clear separation of various physiochemical processes is difficult if the time constants of these processes overlap significantly in the frequency domain. To overcome these drawbacks, advanced mathematical methods such as Distribution of Relaxation Times (DRT) can be applied. DRT relies on the representation of the polarization impedance by its characteristic time constants and is numerically approximated by a discrete distribution function. This method has been successfully demonstrated for process identification and separation in solid oxide fuel cells (SOFC) [4].

In this study, we applied this technique on high temperature PEM-MEAs for the first time. Electrochemical processes were identified and analyzed by varying cell parameters such as temperature and stoichiometry. The results offer a refined understanding of loss mechanisms and provide valuable guidance for fuel cell improvement and optimization.

[1] A. Chandan, M. Hattenberger, A. El-kharouf, S. Du, A. Dhir, V. Self, B.G. Pollet, A. Ingram, W. Bujalski, J. Power Sources 231 (2013) 264

[2] J. L. Jesperesen, E. Schaltz, S.K. Kær, J. of Power Sources 191 (2009) 289-296

[3] F. Mack, R. Laukenmann, S. Galbiati, J. A. Kerres, R. Zeis, ECS Transactions 69 (17), 1075-1087 (2015)

[4] A. Leonide, V. Sonn, A. Weber, and E. Ivers-Tiffée, Journal of the Electrochemical Society 155 (1) B36-B41 (2008)

Figure 1 Typical Nyquist plot of a phosphoric acid doped PBI-type HT-PEMFC at different current densities (a) and the corresponding distribution function (b).

Figure 1

A conventional polymer electrolyte fuel cell (PEFC) incorporates a membrane electrode assembly (MEA) which is comprised of the anode catalyst layer-polymer electrolyte -cathode catalyst layer sandwiched between two gas diffusion layers (GDLs), and bipolar plates. Conventional GDL is typically comprised of highly porous carbon paper or cloth to help diffuse the reactant gases onto the electrode, which is coated with a thin micro-porous layer (MPL). At high current densities, mass transport limitations of fuel or oxidizer in the PEFC occur in porous structures of the GDL, particularly at the cathode, which result in a sharp drop in the output voltage. The key to minimizing mass-transport losses is effective water management in the cell. Liquid water in macro-pores of GDL decreases the fuel cell performance at high current density due to the lack of oxygen reaching the catalyst layer.

A new MEA structure is recently introduced, where the carbon paper backing layer is eliminated and the entire gas diffusion layer consists of only the MPL [1]. We have further improved on this concept by directly depositing the MPL onto the CCM, resulting in an improvement in the interfacial contact between the MPL and the catalyst layer, as well as a simplified fabrication and assembly process. Spray deposition method was used for depositing this MPL onto a commercial CCM [2]. This concept was proven to work and perform better than a conventional cell with a micro-channel flow field used to provide the desired pathway for the reactant gases throughout the cell substituting for the macro porous carbon paper and the millimeter sized flow field.

In this work, a porous foam is utilized as the flow field to distribute the reactant gases over the micro-porous layer instead of the micro-channel flow field. Various pore sizes of the foam, i.e. 60-100 PPI, are used to compare with the conventional micro channel flow field. Contact resistance, permeability, electrochemical impedance spectroscopy, and mechanical properties are used to further characterize the new MEA structure incorporating the porous foam flow field. This method can be an effective way of enabling very high power density operation by reducing the mass transport limitations, and provides an improvement in the interfacial contact between the MEA and the flow field as well as improved mechanical properties.

[1] T. Kotaka, Y. Tabuchi, U. Pasaogullari, and C. Y. Wang, Electrochim. Acta, 146, 618 (2014).

[2] J. Park, U. Pasaogullari, L. J. Bonville, ECS Transactions, 69, 1355-1362 (2015).

There is much work in the literature related to the mathematical optimization of low-temperature polymer electrolyte membrane fuel cells (PEMFCs). In general, the optimization of PEMFCs involves (1) geometry optimization, which focuses on finding the optimum channel geometry and dimensions; and (2) microstructure optimization, which focuses on finding the optimum distributions of the porosity, catalyst, and electrolyte that maximize the power density of the cell. Depending on the complexity of the mathematical model used to describe the PEMFC, the number of variables that need to be optimized can become very large. For instance, let us consider a fuel cell that is discretized using a 3-D finite element mesh with 10⁶ nodes. If we would like to find the optimum values of the porosity, catalyst concentration, and electrolyte density at each mesh point, the total number of design parameters is 3 ×10⁶. Because the number of design parameters is so large, traditional heuristic techniques such as genetic algorithms, swarm-optimization, or other evolutionary algorithms are practically impossible to use even when implemented on massive computer clusters. (Notice that the number of optimization parameters in non-heuristic techniques is usually between 1-10, which is a few orders of magnitude smaller than the total number of design parameters for the PEMFC presented above.) Hence, so far, to make the problem computationally tractable, heuristic optimization techniques have been applied to optimize only a few number of PEMFC parameters.

In this presentation we develop a gradient-based technique for the optimization of PEMFCs, when the number of degrees of freedom, which is defined as the number of optimization parameters, is very large (the number of degrees of freedom in our work should not be confused with the number of nodes or elements in the discretization of the fuel cell) [1]. The technique is based on the computation of the sensitivity functions of the parameters that need to be optimized and is using these sensitivity functions to optimize the cell. The sensitivity functions of the parameters of interest are computed using an adjoint space method initially developed by the applied mathematics community to solve 1-D optimization problems in fluid dynamics, climate, and heat transfer problems [2]. Our group has also used a similar adjoint space technique to analyze the variability and optimize the doping profiles in 2-D and 3-D semiconductor devices [3]. The computational cost required to compute the sensitivity functions using the adjoint space method is relatively small since this method requires solving only one sparse system of linear equations instead of performing multiple fuel cell simulations.

Using the proposed technique we are able to predict the optimum 3-D distribution of platinum particles and porosity profile that maximizes the power density of the cell at different operating current densities. The optimum distributions and porosity profiles depends on the positions of the landings and openings, and on the geometry and dimensions of the layers. In agreement with existing experimental data and previous theoretical estimations [4], we obtain that, at large current densities, the catalyst density should be distributed non-uniformly inside the cell in order to increase the power density of the cell. At low operating current densities the optimum catalyst distribution is more or less uniform distributed inside the catalyst layer. In the case of the porosity distribution, at large operating current, it is more efficient to increase the porosity of the catalyst layer towards the gas diffusion layer side in order to increase the flow of the oxygen and water vapors. At low operating current densities the optimum porosity is uniformly distributed inside the catalyst and gas diffusion layers. More details about the technique, the numerical implementation, and a number of fully optimized structures will be presented at the meeting.

[1] P. Andrei and M. Mehta, "Large-scale optimization of polymer electrolyte membrane fuel cells," 227th ECS Meeting, Chicago, IL, 2015.

[2] D. Cacuci, Sensitivity and Uncertainty Analysis, Volume 1: Theory, Chapman & Hall/CRC, 2003.

[3] P. Andrei, I. Mayergoyz, "Quantum mechanical effects on random oxide thickness and random doping induced fluctuations in ultrasmall semiconductor devices", Journal of Applied Physics, 94 (2003) 7163-7172.

[4] P. Andrei, G. Mixon, M. Mehta, and V. Bevara, "Design of the catalyst layers in PEMFCs using an adjoint sensitivity analysis approach," 227th ECS Meeting, Chicago, IL, 2015.

Additional cost reductions are needed for proton exchange membrane fuel cells (PEMFC) to effectively compete with internal combustion engine powered vehicles. Recent trends have focused on the use of a high current density and a lower Pt catalyst loading to reduce the quantity of expensive materials. However, such strategies have led to higher mass transfer overpotentials (1). These performance losses are advantageously characterized by the mass transfer coefficient. Its separation into fundamental contributions will focus research efforts to improve cell performance. A separation method previously proposed (2) was extended, which provides information for a wider cell voltage range of relevance to duty cycling operation.

The extended model yields a closed form and implicit current distribution expression (Eq. 1) where X represents the dimensionless Cartesian coordinate along the flow field channel, i_e the inlet oxygen flow rate equivalent current density, i_k the kinetic current density, i_L,0 the inlet limiting current density and i the current density. Experimental data obtained with a segmented single cell of 100 cm² active area (Gore catalyst coated membrane, 18 μm thick membrane, 0.2 and 0.1 mg Pt cm^-2 catalyst loading, segmented and unsegmented Sigracet 25BC gas diffusion layer for the cathode and anode, respectively) validated Eq. 1 (Fig. 1a) and led to overall oxygen mass transfer coefficients k (derived from i_L,0 and operating conditions). A plot of 1/k versus M, the diluent molecular weight, yields a linear relation (Eq. 2) enabling the separation of the overall mass transfer coefficient into 2 contributions where k_e+K represents the oxygen mass transfer coefficient in the electrolyte and the gas phase (Knudsen diffusion), k_m the oxygen mass transfer coefficient in the gas phase (molecular diffusion) for the specified diluent (subscript He for helium, N for N₂, CF for C₃F₈) and b a parameter. The molecular diffusion mass transfer coefficient k_m is larger with a lighter diluent (Fig. 1b). Each mass transfer coefficient increases as the cell voltage is decreased. However, the increase in the mass transfer coefficient in the electrolyte and the gas phase (Knudsen diffusion) k_e+K is more substantial than for the other contribution. As a result, the k/k_e+K ratio decreases for lower cell voltages. Specifically for N₂, the mass transfer rate limiting step is associated with the electrolyte and the gas phase (Knudsen diffusion), which suggests a focus on the micro-porous and catalyst layers to increase performance. Two reasons are proposed to explain the cell voltage dependency observed in Fig. 1b. First, the local change in gas phase composition promotes convection. This effect is particularly important before reaching the mass transfer control regime, in the mixed kinetic and mass transfer control region, as the interfacial oxygen concentration changes from its bulk flow field channel value to near 0 in the 0.75 to 0.6 V cell potential range creating a diluent enriched layer. Second, the local temperature in the gas diffusion electrode (through plane direction) is also higher at a lower cell voltage owing to additional heat production. The larger local temperatures in turn promote larger diffusion coefficients in the gas phase (molecular and Knudsen) and permeabilities in the ionomer phase. A comparison with other mass transfer coefficient values derived from impedance data (3) and the model limits of validity will also be discussed.

Acknowledgments

We gratefully acknowledge funding from the Army Research Office (W911NF-15-1-0188). The authors are grateful to the Hawaiian Electric Company for their ongoing support of the operations of the Hawaii Sustainable Energy Research Facility.

References

1. A. Z. Weber, A. Kusoglu, J. Mater. Chem. A, 2, 17207 (2014).

2. T. V. Reshetenko, J. St-Pierre,J. Electrochem. Soc., 161, F1089 (2014).

3. S. Arisetty, X. Wang, R. K. Ahluwalia, R. Mukundan, R. Borup, J. Davey, D. Langlois, F. Gambini, O. Polevaya, S. Blanchet, J. Electrochem. Soc., 159, B455 (2012).

Fig. 1. a) Selected comparisons between current density distributions along the dimensionless flow field channel length obtained with He diluent and calculated model curves (Eq. 1), b) Gas phase molecular diffusion, and, gas phase Knudsen diffusion and ionomer phase diffusion oxygen mass transfer coefficients for different cell voltage values and gas diluents. 60 °C, 48.3/48.3 kPag, 100/100 % relative humidity and 5 % O₂ + diluents/H₂ for the cathode and anode respectively. 50 mL min^-1 O₂ + 958 mL min^-1 He or N₂/424 mL min^-1 H₂ and 25 mL min^-1 O₂ + 479 mL min^-1 C₃F₈/212 mL min^-1 H₂.

Figure 1

Cold-start (subzero startup) capability of polymer electrolyte fuel cell (PEFC) is of great importance. In this paper, the effects of the micro-porous-layer (MPL), varying start-up temperatures, start-up current density variation on the cold-start operation are investigated. We found PEFC with anode micro-porous-layer (AMPL), compared with one with CMPL, has higher possibility of successful cold start. By analyzing the anode and cathode pressure revolution of PEFC, the effect of MPL on the super cooled water removal and ice formation at different temperatures (-7,-10,-15 and-20℃) are discussed . In addition, we investigate the ice distribution in MEA through the X-ray device, by comparing the initial image when fuel cells shut down (before the produce the water) and final image after failed cold-start. We also explore the negative voltage phenomenon in the initial process of cold-start operation.

Cost has been one of the primary barriers to the commercialization of polymer electrolyte membrane fuel cells (PEMFCs), clean energy conversion devices used primarily for power generation applications. While system design improvements and the development of novel electrode materials with increased activity have helped lower costs, the primary driver of cost reduction will be manufacturing membrane electrode assemblies (MEAs) at high volume [1]. Recently developed in-line diagnostic tools to monitor quality have successfully detected a variety of electrode coating variations [2]. This work studies a subset of these electrode irregularities fabricated onto MEAs operating under accelerated stress tests (ASTs) to simulate stresses of real-world driving conditions while spatially monitoring the onset and development of failure points with a thermal camera. An understanding of the impact that electrode irregularities have on MEA lifetime will subsequently allow for (i) the potential classification of these irregularities as defects and (ii) the development of threshold detection limits for in-line quality control diagnostics.

Catalyst-coated membranes (CCMs) and gas-diffusion electrodes (GDEs) were fabricated with an ExactaCoat Sono-Tek ultrasonic spray system. The active area and nominal loading of these Pt/C electrodes were 50 cm² and 0.2/0.2 mg_Pt/cm², respectively. Membrane material was NRE-212. These layers were laminated into edge-protected MEA structures using a hot-pressing process. The MEA structures were exposed to a combined chemical and mechanical AST by operating the fuel cell at 80°C and 0.5/0.5 SLM H₂/Air under open circuit potential (OCV), and cycling between 0 and 80% relative humidity for a 30 second duty cycle. The hydrogen crossover limiting current and the OCV were indicators for a developing failure and the presence of a failure, respectively. A novel cell hardware shown in figure (a) was used to remove the cathode flow-field and conduct a spatial hydrogen crossover measurement. Any hydrogen crossing from the anode to the cathode exothermically reacts at the cathode catalyst with ambient air to produce a heat signature that is detected with the thermal camera [3]. Higher crossover rates at for example discrete failure points become apparent through a significant local temperature rise.

Sample results of a pristine MEA tested as described above are shown in the figures. The graph (b) shows the hydrogen crossover limiting current density rapidly increasing after the OCV declined with an increasing rate. This indicated that the MEA developed a failure point. The images show the thermal response during the spatial hydrogen crossover measurements for the pristine MEA (c) prior to and (d) subsequent to aging. The heat signature indicated a failure point developed near the inlet of the cell. Results will include studies on MEA structures with intentionally and systematically introduced coating variations of various sizes and severities.

References

1) S. Kamaruzzaman, and W.R. Daud. Renewable Energy 31.5 (2006).

2) M. Ulsh and B. Sopori and N.V. Aieta and G. Bender. ECS Transactions 50.2 (2013).

3) G. Bender and W. Felt, and M. Ulsh. Journal of Power Sources 253 (2014).

Figure 1

The polymer electrolyte membrane fuel cell (PEMFC) continues to develop as a viable alternative to the combustion engine in automotive applications. As this technology advances, it is critical that the PEMFC is capable of maintaining performance for long term operation. In an effort to understand the degradation of fuel cell performance over time, a study into the development of an accelerated gas diffusion layer degradation protocol was completed. As part of an effort to quantify the effects of this accelerated degradation, in situ synchrotron radiography imaging techniques were coupled with performance testing in order to quantify the effects of this degradation on water saturation profiles, transport resistance, and limiting current in an operating fuel cell.

Gas diffusion layer samples of SGL 25 BC and SGL 29 BC were artificially aged in a concentrated solution of hydrogen peroxide (30% wt.) for a period of 12 hours at an elevated temperature of 90 degrees Celsius. Hydrogen peroxide facilitates chemical corrosion of the carbon material in the gas diffusion layer and was found to significantly affect the wettability of the degraded samples. Hydrogen peroxide was selected to facilitate the degradation mechanism due to the fact that it is a recognized chemical species found in operating fuel cells and produced at the catalyst layer (1, 2).

The accelerated degradation procedure that was used in this study was found to primarily affect the wettability of the tested gas diffusion layers. Consequently, it was expected that the most significant impact of this degradation mechanism would be in the mass transport losses at high current densities. For this reason, a limiting current investigation was performed. Unlike other studies of limiting current in which the cathode oxygen concentration is varied (3), a limiting current study based on varying the relative humidity of reactant gases was performed. In this study, limiting current was measured for relative humidities of 0%, 50%, 80%, 90%, and 100% for several fuel cells with fresh and degraded GDLs. Simultaneous synchrotron imaging was used to quantify the distribution of liquid water in the anode and cathode of the operating cell during these limiting current studies.

Trends with respect to reactant transport resistance, water saturation profiles, and limiting current were quantified as a function of relative humidity and degree of degradation. For all tested samples, as the relative humidity was reduced, the corresponding limiting current increased. It was also found that the limiting current liquid water saturation profile was independent of the reactant gas relative humidity. Additionally, artificial degradation led to increased liquid water saturation levels in the operating fuel cell. This work illustrates the potential impacts of long term fuel cell operation on the water management of PEMFC gas diffusion layers.

References

1. C. Chen and T. Fuller, ECS Transactions., 11, 1 (2007).

2. W. Liu and D. Zuckerbrod, J.Electrochem.Soc., 152, 6 (2005).

3. D. R. Baker, D. A. Caulk, K. C. Neyerlin and M. W. Murphy, J.Electrochem.Soc., 156, 9 (2009).

R&D on Proton exchange membrane fuel cells (PEMFC's) technology has been accelerated in the last few years to reduce the dependency on fossil fuel. PEMFC's operate by internally combining oxygen from air with hydrogen to form water and generate electricity and waste heat. The most common hurdle for enhanced PEMFC durability and performance is still the water management: the proton exchange membrane in the center of these fuel cells has to be hydrated in order to keep its ability to conduct protons from anode to cathode side while on the other hand excessive liquid water can lead to cell flooding and increased degradation rates of the cell. Thus, a detailed understanding of all aspects of water management in PEMFC is important. This includes the fuel cell water balance, i.e. the question which fraction of the product water leaves the fuel cell via the anode channels versus the cathode channel. Our research group is currently developing a state of art technology to obtain an ad-hoc and real time electrical signal of the fuel cell water balance by employing a constant temperature hot wire anemometry [1]. The hot wire sensor is placed in the anode outlet of PEMFC, and the voltage signal received gives valuable insight into heat and mass transfer phenomena which can be interpreted directly to real time PEMFC water balance. So far, ex-situ experiments have been conducted to measure the voltage signal of the hot wire anemometer for a known gas stream that contains a binary mixture of hydrogen and water vapour as would ideally leave the fuel cell anode. However, in an operating fuel cell there are two main uncertainties in this method: (i) there is some internal hydrogen crossover that does not load to an external current, thus the exact amount of hydrogen leaving the fuel cell anode is not known, and (ii) that there is nitrogen crossing over from the cathode side to the anode side, and this nitrogen does slighty falsify the voltage signal that the hot wire yields. The effects of nitrogen-cross over on the hot wire voltage signal at the anode outlet and consequently on the measured water balance (r_d) is studied in this work.

In our previous work it was shown that the only unknown in the determination of the hot wire voltage signal is the equation that determines the heat transfer around the hot wire, and we have shown that it is necessary to employ a power-law equation as suggested by Hilpert:

Nu = C Pr^0.33 Re^m

where Re is the Reynolds number and P_r is the Prandtl number are calculated at film temperature. The constants C and m usually depend on the Re, and they will be determined out of the experimental data for the gas mixture, by plotting the measured (Nu/ Pr^0.333) versus the Re number of the gas mixture stream. The missing C and m from Hilpert equation can then be easily obtained from a power-fit of the general form Y = CX^m. Also, it was shown somewhere else that in fact only the exponent to the Reynolds number, of the gas mixture stream is important when determining the fuel cell water balance out of the hot wire signal [1].

 The nitrogen-cross over is experimentally demonstrated by introducing 1% of nitrogen to the dry hydrogen molar flow. The 99%H₂+1%N₂ of the dry mixture is humidified with water vapor by controlling the relative humidity (RH) of the dry mixture with in the range of (0-100)%RH, simulating the PEMFC anode outlet. The hot wire voltage is measured with and without nitrogen and it was slightly lower with the presence of 1 % nitrogen in the flow. The effect of the voltage reduction on the measured water balance can be neglected. This because the effect of 1% nitrogen on power law constant's m which is used in determining the water balance as explained somewhere else is extremely low. It might be concluded that, the hot wire technique for measuring the water balance is still accurate and can be used for ad-hoc and real time water balance measurements and the nitrogen cross-over affect can be neglected, knowing that the 1% N₂ is one order of magnitude higher than the actual concertation of 0.02 m² active area cell used for lab testing.

[1] Berning, T., & Al Shakhshir, S., Applying hot wire anemometry to directly measure the water balance in a proton exchange membrane fuel cell–Part 1: Theory. International Journal of Hydrogen Energy, 40(36) (2015), 12400-12412.

In PEM fuel cells, electrocatalyst interfaces including Pt-C, Pt-ionomer, C-ionomer interfaces, play a critical role for both performance and durability.^1-3 Pt-C interface influences the stability and activity of Pt dramatically; electrocatalyst/ionomer interfaces⁴ in fuel cell catalyst layers determine the electrochemical surface area (ECSA), local transport, and structural stability of catalyst layer. Delineating the structure evolution of the interfaces in response to PEM fuel cell operating environment will provide the science foundation for new electrocatalyst/catalyst layer design therefore improve the performance and durability of PEM fuel cells.

We have developed an aberration corrected environmental transmission electron microscopy (ETEM), which allows unprecedented spatial and temporal resolution in the relevant gas environment, such as O₂, H₂, H₂O, CO, CH₄, and CO₂. With customized intelligent gas delivery system, the gas environment relevant to PEM fuel cell anode and cathode (H₂/H₂O, O₂/H₂O) can be simulated and precisely controlled through the residual gas analyzer (RGA). The ETEM opens new avenue to study the dynamic changes of electrocatalyst interfaces.

In order to study the surface chemistry of catalysts under different environment, a specially-designed pre-treatment system is integrated into X-ray photoelectron spectroscopy (XPS), which allows us to move catalysts seamlessly between the pre-treatment system and XPS chamber so that surface contamination from external environment will be eliminated.

In this talk, we will present our recent progress on identifying both structure and surface chemistry evolution and their correlation in electrocatalysts under PEM fuel cell relevant environment through the combination of ETEM and XPS capabilities.

1. Y. Y. Shao, G. P. Yin and Y. Z. Gao, J. Power Sources, 2007, 171, 558-566.

2. S. Park, Y. Y. Shao, H. Y. Wan, V. V. Viswanathan, S. A. Towne, P. C. Rieke, J. Liu and Y. Wang, J. Phys. Chem. C, 2011, 115, 22633-22639.

3. R. Kou, Y. Y. Shao, D. H. Mei, Z. M. Nie, D. H. Wang, C. M. Wang, V. V. Viswanathan, S. Park, I. A. Aksay, Y. H. Lin, Y. Wang and J. Liu, J. Am. Chem. Soc., 2011, 133, 2541-2547.

4. A. Z. Weber and A. Kusoglu, J. Mater. Chem. A, 2014, 2, 17207-17211.

Hydrogen is regarded as a valuable high energy density, zero-emission fuel for the future, with the potential to reduce or eliminate fossil fuels. H₂ also has potential as an energy carrier or storage molecule: i.e. being generated, stored, and consumed to bridge the gap of high and low production cycles of renewable energy sources, such as solar cells and wind turbines. Because molecular hydrogen is not naturally available, the complications of generation, storage and transportation must be addressed. On Earth, hydrogen (H) mainly occurs in combination with oxygen, i.e. in water. One possible technology option for clean H₂ generation is low temperature water electrolysis, with the two half-reactions being the oxygen evolution reaction (OER) and hydrogen evolution reaction (HER). These reactions, particularly the OER, are mechanistically complex and dependent on the surface energy of multiple intermediate species, leading to high overpotentials and low efficiency. Precious metal-based catalyst materials are ultimately not viable, due to rarity and cost, and typically provide high catalytic activity for the OER or HER, but not both. More effective catalysts with the ability to catalyze both the OER and HER (i.e. OER/HER bi-functional behavior) could improve current electrolyzers and enable future technologies. Thus, there is a desire to design electrocatalysts based on earth- abundant, low-cost materials for overall water splitting.

In this talk we will present on the synthesis and electrochemical characterization cobalt phosphide-based materials. CoP-based electrocatalysts in both thin film and nanoparticle form have been prepared from their respective Co₃O₄ spinel precursors using phosphine gas, generated in situ from the thermal decomposition of sodium hypophosphite. These electrocatalysts were found to have performance comparable to that of the commercial precious metal benchmarks when examined in alkaline electrolyte. For example, ΔE(OER–HER) values were calculated by adding the overpotentials at which j_geo = 10 mA cm^-2 and j_geo = -10 mA cm^-2were attained for the OER and HER, respectively, to the thermodynamic OER–HER potential difference of 1.23 V. Values of ΔE(OER–HER) = 1.77 V for cobalt phosphide catalysts are within 70 mV of the idealized water electrolysis with 20% Ir/C (anode):20% Pt/C (cathode) where ΔE(OER–HER) = 1.70 V. Various aspects of this work will be presented.

This work was supported by the Laboratory Directed Research and Development program at Sandia National Laboratories, a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.

Introduction

Proton Exchange Membrane (PEM) water electrolysis provides a sustainable solution for the production of hydrogen, and is well suited to couple with the intermittent nature of energy sources such as wind and solar. To spur large-scale commercial application of PEM water electrolysis, more efficient and less expensive non-noble metal electrocatalysts towards the hydrogen evolution reaction (HER) are required. Recently, transition metal phosphides (TMPs) have become the typical representatives of the burgeoning non-noble metal HER electrocatalysts due to that their intrinsic structures meet the criteria of outstanding electrocatalysts. It should be noted that the currently studied TMPs are mainly focusing on improving their HER performance by a series of structural engineering methods [1-3]. To the best of our knowledge, there are very few reports on promoting the intrinsic activity of the TMPs in water electrolysis.

In this communication, we reported the scalable synthesis of nickel phosphides with a mixed crystalline structure (referred to as Ni₂P&Ni₃P) toward HER by a solution-phase reaction. To elucidate the superior catalytic activity of Ni₂P&Ni₃P, we also explored the preparation of the nickel phosphides in a single crystalline state (referred to as Ni₂P) and that in a single amorphous state (referred to as Ni-P) for comparison.

Result and Discussion

X-ray diffraction patterns of Ni₂P&Ni₃P、Ni₂P and Ni-P are shown in Fig 1a. The analysis of the diffraction data reveals that both tiny Ni₂P nanocrystallines and Ni₃P nanocrystallines are observed for Ni₂P&Ni₃P. As for Ni₂P, the Ni₃P tiny nanocrystallines are vanished and only the intensive diffraction peaks corresponding to Ni₂P are observed. In comparison with the crystalline state nickel phosphides, the Ni-P shows poor crystalline order with a very broad peak. The crystalline structures could take effects on the intermediate reaction of HER. In order to compare the hydrogen desorption ability of the three electrocatalysts, the hydrogen oxidation reaction (HOR) was investigated by LSV (Linear scan voltammetry) at 20 mA cm^-2 for 5 min immediately after the hydrogen evolution reaction (HER). As it is shown in Fig 1b, the characteristic mixed crystalline structure is expected to promote the hydrogen desorption ability. Thus, the largest peak current density is obtained with Ni₂P&Ni₃P. Furthermore, steady state polarization curves of different electrocatalysts were shown in Fig 1c. As expected, the Ni₂P&Ni₃P shows the best HER activity, and the overpotentials required for Ni₂P&Ni₃P、Ni₂P and Ni-P to produce cathodic current densities of 20 mA cm^-2 are 154 mV, 221 mV, 262 mV, respectively.

Fig 1 (a) XRD patterns of Ni₂P&Ni₃P, Ni₂P and Ni-P; (b) LSV and (c) Steady-state polarization curves of the three types of prepared catalysts at a scan rate of 1 mV s^-1. All the electrochemical characterizations are tested in the mixture of 0.1 M H₂SO₄ and 0.5 M Na₂SO₄ at 25 ^◦C and atmosphere pressure.

Reference

[1] Liu P., Rodriguez J. A. Catalysts for Hydrogen Evolution from the [NiFe] Hydrogenase to the Ni₂P (001) Surface: The Importance of Ensemble Effect [J]. Journal of the American Chemical Society, 2005,127(42):14871-14878.

[2] Popczun E. J., Read C. G., Roske C. W., Lewis S., Schaak R. E. Highly Active Electrocatalysis of the Hydrogen Evolution Reaction by Cobalt Phosphide Nanoparticles [J]. Angewandte Chemie International Edition, 2014,53(21):5427-5430.

[3] Liu Q., Pu Z. H., Asiri A. M., Sun X. P. Nitrogen-doped carbon nanotube supported iron phosphide nanocomposites for highly active electrocatalysis of the hydrogen evolution reaction [J]. Electrochim Acta, 2014,149(10):324-329.

Figure 1

Hydrogen is an ideal energy carrier with an energy density of 140 MJ kg^-1 and water as its final product, which is considered as an ideal candidate for the replacement of fossil fuels.¹ To date, most H₂ is produced today through nickel-catalyzed conversion of CH₄ to H₂ and CO followed by a water gas shift reaction to yield H₂ and CO₂. However, this approach is based on unsustainable fossil energy and discharges greenhouse gas. Water is an ideal source for H₂ production as it is carbon-free, plentiful and almost costless. Water electrolysis, with only H₂ and O₂ as the final product, is a highly promising route in environmental benign H₂ generation.^{2, 3} Although the ultimate goal is to couple hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) to an integrated efficient device for whole water electrolysis, one of the most crucial issues that must be solved is the development of highly efficient and durable catalysts for water oxidation and reduction.³

In this study, monocrystalline Ni₁₂P₅ hollow spheres with ultrahigh specific surface area (222.5 m^-2 g^-1) were prepared by a water-in-oil microemulsion method (Figure 1). The over potential required for 10, 20 50 and 100 mA cm^-2 current densities are 144, 173, 225 and 277 mV in strong acidic solution with a Tafel slope of only 45 mV dec^-1, ranking among the most active nonprecious HER catalysts. Moreover, the monocrystalline Ni₁₂P₅ hollow spheres catalyst exhibited a highly active and stable performance during a 3,000 cycling tests through cyclic voltammetry and a 12 hour duration chronopotentiometry test. The results indicate the highly promising application potential of the monocrystalline Ni₁₂P₅ hollow spheres catalysts in hydrogen generation to replace Pt. These promising features make Ni₁₂P₅ hollow spheres ideal candidates for the next generation HER catalysts. These catalysts with novel monocrystalline structure can be extended to electrochemical applications in various fields, such as supercapacitors, fuel cells, batteries and electrochemical sensors, and so on.

Figure 1. The illustration of synthesis procedure for the Ni₁₂P₅ hollow spheres and the water electrolysis behavior of the catalysts.

Acknowledgments 

This work is supported by the Strategic priority research program of CAS (Grant No. XDA09030104), the Jilin Province Science and Technology Development Program (Grant No. 20130206068GX, 20140203012SF, 20160622037JC) and the Recruitment Program of Foreign Experts (Grant No. WQ20122200077).

References

1. X. Zou and Y. Zhang, Chem. Soc. Rev., 2015, 44, 5148-5180.

2. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, E. A. S. Qixi Mi and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473.

3. S. D. Ebbesen, S. H. Jensen, A. Hauch and M. B. Mogensen, Chem. Rev., 2014, 114, 10697-10734.

Figure 1

There is an ongoing effort to develop alkaline stable anion exchange membranes to be employed as solid-electrolytes in alkaline membrane fuel cells (AMFCs) and solid-state alkaline water electrolyzers so as to take advantage of the distinct and attractive advantages of operation at high pH. These advantages, which are not accessible to acidic proton-exchange membrane fuel cells and water electrolyzers, include the possibility of using stainless steel flow fields and endplates (with the subsequent reduction in costs), more flexibility in the choice of fuels, and the possibility of using cheaper non-platinum-group-metal electrocatalysts for both the hydrogen evolution/oxidation and the oxygen evolution/reduction reactions (1).

However alkaline operation has an important drawback. The hydrogen evolution reaction (HER) is much more sluggish because of the low concentration of protons in alkaline media and the lower activity of noble metals for this reaction. Pt electrocatalysts show considerable lower HER activity in alkaline conditions than in acidic conditions.

The following mechanism for water electrolysis in alkaline conditions has been proposed:

a) Volmer: H₂O+e^- +* → H_ad + OH^- [1]

b1) Tafel: H_ad + H_ad → H_ad + 2OH^-+2* [2]

b2) Heyrovsky: H_ad + H₂O+e^- → H₂ + OH^-+ * [3]

 It is generally accepted that the Volmer step is the rate determining step (rds) for the overall reaction. This assumption implies that the presence of a bi-functional catalyst with hydrophilic domains able to stabilize the water in the transition complex will speed up the overall reaction, as observed by several authors for platinum, iridium, ruthenium and nickel electrocatalysts modified by deposition of transition metal hydroxides (i.e. nickel hydroxide) (2, 3).

In this work, we investigated the HOR/HER kinetics on Pt/C and Pt/C/Ni(OH)₂ bi-functional electrocatalysts. Our objective was to determine the optimum concentration of Ni(OH)₂ required for the best performance of the electrocatalyst for the HER under alkaline conditions. The catalysts were prepared by mixing the required amounts of colloidal dispersions of Ni(OH)₂ with commercial Pt/C catalyst previously dispersed (using ultrasonication) in water. The RDE measurements were done in 0.1M KOH, at temperatures ranging from 0°C to 30°C to extract useful kinetic information such as exchange current densities and activation energies for each catalyst. Kinetic currents were obtained after IR and mass transport corrections. The HOR/HER kinetic current densities (see Figure 1) were fitted using the Butler–Volmer equation to estimate the exchange current densities for each catalyst at each temperature. Arrhenius behavior was observed. The activation energy was the same for all Pt/C and the bi-functional catalysts, at 37.2 kJ/mol, which was in good agreement with previously published value for Pt (4).

The highest exchange current densities were obtained for the catalyst containing 10wt% Ni(OH)₂. Higher Ni(OH)₂ concentrations resulted in a poorer performance due to the coverage of the Pt active sites; a reduction in ECSA was observed.

The bi-functional catalysts were also tested in a solid-state alkaline water electrolyzer operated with ultrapure water. The membrane electrode assemblies were fabricated with a commercial anion exchange membrane The anode catalyst was iridium oxide (2.5 mg/cm²). The optimal bi-functional electrocatalyst outperformed Pt/C by 0.1-0.2 V at relevant current densities.

References

1. N. W. Li and M. D. Guiver, Macromolecules, 47, 2175 (2014).

2. N. Danilovic, R. Subbaraman, D. Strmcnik, K. C. Chang, A. P. Paulikas, V. R. Stamenkovic and N. M. Markovic, Angewandte Chemie, 51, 12495 (2012).

3. D. Strmcnik, M. Uchimura, C. Wang, R. Subbaraman, N. Danilovic, D. van der Vliet, A. P. Paulikas, V. R. Stamenkovic and N. M. Markovic, Nature chemistry, 5, 300 (2013).

4. W. Sheng, H. A. Gasteiger and Y. Shao-Horn, Journal of The Electrochemical Society, 157, B1529 (2010).

Figure 1

Hydrogen is poised to be a key energy carrier in a sustainable energy economy (1). Among the different approaches to produce hydrogen, water electrolysis has received special attention due to its unique advantages including reactant availability, safety, stable output and product purity (2). Alkaline water electrolysis offers additional advantages, including the possibility of using stainless steel flow fields and endplates (with the subsequent reduction in costs) and the possibility of using cheaper non-platinum-group-metal electrocatalysts for the oxygen evolution reaction. However, the slow kinetics for the hydrogen evolution reaction (HER) in alkaline media is currently a key drawback.

In this work we evaluated a mixed-metal-oxide composed of titanium dioxide and ruthenium dioxide as a support for Pt and measured the catalytic activity of this supported electrocatalys for the HER. The mixed-metal-oxide (TiO₂-RuO₂, RTO) was synthesized using a wet chemical synthesis procedure, and the Pt/RTO was synthesized by reduction of Pt precursor onto the support using an impregnation-reduction method. These materials were characterized by XRD, TEM and BET. The activity of Pt/RTO towards HER the was compared with a benchmark Pt/C catalyst (46%Pt; Tanaka, K. K.). Rotating disk electrode (RDE) measurements showed Pt/RTO outperforming (under all the conditions) the benchmark catalyst (Pt/C). The specific exchange current density for Pt/RTO was 2.51mA/cm²_Pt (295 K, 1600rpm, H₂-saturated 0.1M KOH), more than five times that of Pt/C. The catalysts were also evaluated in a solid-state water electrolyzer operated with ultrapure water at 50°C, wherein MEAs were prepared with IrO₂ as a common anode electrocatalyst. The MEAs fabricated with Pt/RTO as the cathode catalyst outperformed the benchmark catalyst by 0.1-0.2 V across relevant current densities (see Figure 1).

1. M. S. Dresselhaus and I. L. Thomas, Nature, 414, 332 (2001).

2. H. Yin, S. Zhao, K. Zhao, A. Muqsit, H. Tang, L. Chang, H. Zhao, Y. Gao and Z. Tang, Nat Commun, 6, 6430 (2015).

Figure 1

N

Introduction

Fuel cell technology has been extensively developed in order to create new clean sources of energy. In order to have a wide implementation of fuel cell devices powered by hydrogen (H₂), an inexpensive source of clean H₂ should be created. One of the technologies that produces hydrogen gas is the splitting of water through electrolysis. The majority of electrolyzers available on the market use proton exchange membrane (PEM) technology, which requires catalysts consisting of expensive and rare platinum group metals (PGMs). To increase penetration into the water electrolyzer and fuel cell markets, abundant and inexpensive materials should be incorporated. Alkaline exchange membrane water electrolysis (AEMWE) is an alternative technology that does not require PGM catalysts and, therefore, enables the use of base metals, alloys and oxides.

The goal of this research has been focused on the creation of a catalyst that is stable in alkaline media, has a high surface area, is easy to synthesize, and can be produced at a lower cost while maintaining a high oxygen evolution reaction (OER) activity.

Experimental

Unsupported ternary Ni-Mo-Cu materials were synthesized by the Sacrificial Support Method (SSM) using a high surface area silica [1-3]. The following process was followed, and produced a NiMoCo OER catalyst achieving a surface area (SA) of 25 m² g^-1.

15g of EH-5 (SA=400 m² g^-1) silica, 2.5g of nickel nitrate, 2.5g of copper nitrate and 3g of ammonium molybdate tetrahydrate were mixed by grinding them in a porcelain crucible. The crucible was placed into a tube furnace and heated to 100 °C with a ramp rate of 10ºC/min. The 100°C temperature was held for 50 minutes before an additional temperature ramp was conducted at an interval of 5ºC/min until 550°C was reached and held for four hours.. During this time, a reducing atmosphere was maintained using 7% H₂. The silica was later removed by rinsing with 7M potassium hydroxide (KOH) for 24 hours, followed by washing and drying.

Cell testing was performed using a Fuel Cell Technologies 25 cm2 fuel cell stack, modified for electrolysis operation. The carbon anode flow field was replaced with an anode piece fabricated at Proton OnSite, based on 316L SS, which was cleaned and passivated in a nitric solution prior test. ASTM Type II DI water was fed to the anode side of the cell via a diaphragm pump. A submersible heater, placed in the water reservoir, was used to control stack temperature at 50ºC. An image of the test station is shown below in Figure 1.

Figure 1. 25 cm² Test Stand

 Results and Discussion

SEM images show that the NiMoCu material has a well-developed porous structure, as well as primary particles on the nanometer scale (Figure 2). The surface area was measured at 25±3 m² g^-1 via BET analysis. Based on these results, the team decided to proceed with high surface area silica as the sacrificial support.

Figure 2. SEM on Ni-Mo-Cu catalyst Electrochemical activity of the OER is shown in Figure 3.

Figure 3. ORR activity of different Ni-Mo-Cu electrocatalysts in alkaline media.

Operational testing was conducted on the NiMoCo OER catalyst provided by UNM in the 25 cm² cell previously described. As shown below in Figure 4, the UNM supplied material experienced relative stability at the steady-state current density of 200 mA/cm2, as compared to the PGM based reference plot. Both tests used the same membrane and electrode binders.

Figure 4. Steady-state operation graph of non-PGM anode versus PGM reference

Conclusion

Synthesis of spinel-based catalysts by SSM was investigated. Upon examination of the physical characteristics, it can be assumed that further modification to the design of the electrocatalysts is necessary. The electrochemical characteristics from the RDE experimentation proved that the catalysts are active in OER and can act as a replacement for conventional Pt-group catalysts in the appropriate conditions. Short duration operational testing has shown some translation of the RDE measurements to in-cell performance. Additional testing is planned for evaluation of repeatability and longer term stability.

References

[1] U. Martinez, A. Serov, M. Padilla, P. Atanassov ChemSusChem 7(8) (2014) 2351–2357.

[2] A. Serov, K. Artyushkova, P. Atanassov Adv. Energy Mater. 4: 1301735 (2014) doi: 10.1002/aenm.201301735.

[3] A. Serov, U. Tylus, K. Artyushkova, S. Mukerjee, P. Appl. Catal. B: Environmental 150 (2014) 179-186.

Figure 1

The production of hydrogen via water electrolysis is currently recognized as the only option to store gigawatt electrical energy coming from renewable energy sources such as wind and sunlight. However, commercially available alkaline electrolyzers are still limited to low current densities (around 0.5 Acm^-2), primarily due to its internal resistance losses. In order to reach the excellent current densities of acidic polymer electrolyte membrane (PEM) water electrolysis (ranging between 2 and 4 Acm^-2), it is crucial to tailor electrolyte construction and cell design in order to overcome the high ohmic losses of conventional diaphragms and archaic cell designs used in alkaline electrolysis. Here, we demonstrate for the first time alkaline electrolysis achieving current densities as high as 2 A.cm^-2 at 2 volts, not only well overcoming the performances of lab-scale and commercial alkaline electrolyzers, but also closely matching the performance of state-of-the-art PEM water electrolysis.

This paper will describe the performance of alkaline water electrolyzers using newly developed Sustainion™-X5 membranes.

Sustainion™ membranes have a styrene base, and mixtures of other components to produce a high conductivity in alkaline systems. The membranes are stable when soaked in 1 M KOH at room temperature for months. A CO₂electrolyzer has run for over 4000 hours with no loss in performance. Initial tests also show stability in 1 M KOH at 60°C, See Figure 1, although no long term tests have yet been completed.

The work here considers the performance of the membranes in model alkaline membrane electrolyzers. We manufactured a small MEA using platinum coated carbon paper as a cathode and a IrO2 coated carbon paper as an anode and a Sustainion™-X5 membrane. The MEA was mounted into Fuel Cell Technologies 5 cm² cell hardware with carbon flowfields. The cell was heated to 60 °C and 1 M KOH was fed into the anode and cathode.

Figure 2 shows how the cell voltage varied with time at a fixed current of 0.8 A/cm² and 1 A/cm². The data were taken on a single cell, at a fixed current of 1 A/cm² during the day and 0.8 A/cm² at night. Initially the cell produced 0.8 A/cm² at 1.74 V and 1 A/cm² at 1.81 V, but the voltage rose to nearly 2V after 150 hours of operation. Disassembly of the cell showed that the carbon paper on the anode was corroding leading to a loss of anode catalyst.

We have also started work on nickel catalysts. The initial results were modest, but additional results will be presented at the meeting.

Acknowledgement

Parts of this work were supported by ARPA-E under contract DE-AR0000684 and by 3M. The opinions here are those of the authors and may not reflect the opinions of ARPA-E or 3M.

Figure 1

Polymer electrolyte membrane (PEM) fuel cells are an attractive energy conversion device because of the potential for extremely high efficiency and low emissions. Pure hydrogen is an ideal fuel for PEM fuel cells but it is energy intensive to produce. Currently, the main method of producing hydrogen is steam reforming of natural gas. This is a break from the renewable energy future promised by hydrogen fuel cells. A greener approach to producing hydrogen is electrochemical reforming, the most obvious feedstock is water. Water electrolysis occurs at a relatively high potential (1.23 V), which makes it expensive both in terms of materials required and electricity. To fulfill the promise of fuel cells, more efficient methods of producing hydrogen must be found.

One way to reduce the energy required is to depolarize the PEM electrolyzer anode by using an alternate feed. This feed must have a lower oxidation potential than water and be generated renewably. Two alternative feeds we have looked at are methanol and phosphomolybdic acid.

Oxidation of methanol-water solutions to hydrogen takes place at a theoretical voltage of 0.03 V. Methanol fuel cells are a well-studied problem and the issues are known. The main issue is crossover of methanol to the cathode, lowering the total cell potential. We looked to see if this would be as problematic in a hydrogen pumping environment since there would be no oxygen at the cathode for the methanol to react with.

The other feed we examined was phosphomolybdic acid (PMA). PMA is a keggin ion which is readily reduced by biomass substrates in the presence of sunlight or heat.¹Our idea was to apply the concept of a depolarized anode electrolyzer to a mediated electrochemical system. This system has the PMA in solution as a combination catalyst and charge carrier. The PMA is reduced by biomass and circulated to an anode where it is oxidized to release the hydrogen removed from the biomass. This gives many of the advantages of a flow battery, where power and capacity scale separately. Our focus has been on studying the anode performance characteristics of PMA.

 References.

1. Liu W, Mu W, Liu M, Zhang X, Cai H, Deng Y. Solar-induced direct biomass-to-electricity hybrid fuel cell using polyoxometalates as photocatalyst and charge carrier. Nat Commun. 2014;5:3208. doi:10.1038/ncomms4208.

An electrochemical hydrogen pump (EHP) is a device employed for compressing hydrogen gas. The compression efficiency of the EHP, which operates by an isothermal process in principle, is theoretically higher than that of a conventional mechanical compressor, which operates by an adiabatic process. The structure of an EHP cell is similar to that of polymer electrolyte membrane fuel cell (PEMFC). In the EHP, compressing process is realized by imposing a potential difference to catalyst coated membrane (CCM) and completed by 3 steps: hydrogen oxidation reaction (HOR) at anode, conduction of hydrogen protons through membrane and hydrogen evolution reaction (HER) at cathode.

Similar with PEMFC, overpotentials involved in EHP are categorized into two components: ohmic overpotential, mainly attributed to the conduction of protons through the membrane, and non-ohmic overpotential, which involves activation and concentration overpotentials. Due to the fast speed of hydrogen electrode reaction, ohmic overpotential is the main overpotential in the EHP. In order to reduce membrane resistance, bubbler humidifier same as that used in PEMFC system is conventionally utilized for humidifying membrane. However, different with PEMFC there is no water generation reaction occurs in the cathode side of EHP, the water at cathode dragged by hydrogen protons may be not enough for humidifying membrane at low current and high temperature conditions. Moreover, the effects of compression ratio (pressure of hydrogen at cathode) on overpotentials are important but not well discussed yet.

In order to solve the problems which mentioned above, we supposed a different humidification method and employed a reference electrode for separating overpotentials. An internal humidification method is employed which is realized by storing liquid water in the cathode end plate shown in Fig.1. The liquid water directly humidifies membrane, leading to higher conductivity of membrane than that using a conventional bubbler humidifier [1]. This method is expected to balance water transport through membrane and keep membrane hydrated at high temperature. Electrochemical impedance spectroscopy (EIS) is introduced to clarify the effects of compression ratio on ohmic overpotential and non-ohmic overpotential.

The compression ratio effects on overpotentials evaluated by EIS show following characteristics: For the part of ohmic overpotential, experimental results don't show any dependence on compression ratio which suggests that water transport in EHP is more like diffusion process rather than convection process. On the other hand, non-ohmic overpotential decreases with increasing of hydrogen pressure of cathode. Separation results with reference electrode clearly shows that decreasing of non-ohmic overpotential occurs at cathode and it is attributed to the increasing of exchange current density of HER by increasing hydrogen pressure.

Fitting I-V curves with a Volmer-Heyrovsy-Tafel expression [2] shows that the kinetic of HER and HOR are limited by diffusion process to/from electrode at anode/cathode. This is also confirmed by electrochemical impedance spectra data obtained at low frequency region. The concentration overpotential at anode is attributed to hydrogen diffusion through gas diffusion layer (GDL). In the EHP, in order to decrease hydrogen leakage, compression pressure supplied by end plates is set to 5MPa, which is much higher than common value (1~2MPa) used in PEMFC cell. At such a high compression pressure, GDL is compressed extremely, leading to small value of efficient gas diffusivity of hydrogen. Limiting current of anode calculated by hydrogen gas diffusion through GDL is less than 2A/cm²which agrees with the fitting results generally. The concentration overpotential at cathode is larger than that at anode. Frist, we suspected that water layer built at cathode may work as a mass transfer barrier for hydrogen gas and lead to large mass transfer resistance. However, an additional experiment carried out with bubbler humidification showed even larger mass transfer resistance at cathode. It is clear that mass transfer of hydrogen at cathode is not limited by water layer. It is possible that hydrogen diffusion in catalyst layer leads to the high concentration overpotential. The structure of catalyst layer is assumed to consist of gas pores and aggregates of Pt/C covered by ionomer layers [3]. Hydrogen that produced at reaction sites of cathode need to diffuse through this ionomer layer. With the liquid water humidification, water content of ionomer layer increases [4] which enhances the permeability of hydrogen and decreases mass transfer resistance.

Reference:

[1] T. A. Zawodzinski Jr., T. E. Springer, and S. Gottesfeld. Journal of The Electrochemical Society, 1993, 140(7) 1981-1985.

[2] P.M. Quaino, A.C. Chialvo, Electrochimica Acta, 2007(52) 7396–7403.

[3] Firat C, Ajay K. Prasad, Journal of The Electrochemical Society, 2014,161 (6) F803-F813.

[4] T. Sakai, E. Torikai. Journal of The Electrochemical Society, 1985, 132(6) 1328-1332.

Figure 1

New trends in PEM water electrolysis systems development open up new technology gaps and requirements that have not been discussed before with respect to PEM water electrolysis. For example, hydrogen is considered as one of the best solutions for large-scale energy storage that comes from renewable and intermittent power sources such as wind and solar electricity [1], therefore, new megawatt PEM electrolysers are required. These trends are evident through new large-scale recent installations, especially for Power-to-Gas projects in Europe and plan to treat contaminated water at Fukushima Nuclear Power Plant.

Even though PEM electrolysis has in fact been used for quite a few years now without undergoing substantial improvements over those years, however, with the new focus on hydrogen as the energy carrier there is much more interest in low-cost and high-efficiency H₂ production. There are two main ways to lower the cost of hydrogen production via PEM water electrolysis: to lower the capital expenses (CAPEX) and/or to lower the operating expenses (OPEX). 3M has recently demonstrated a very effective way to address reducing the high CAPEX by increasing the range of current densities where electrolyzers can operate from a maximum of about state-of-the-art 2.0 A/cm², to as much as 20 A/cm² by employing a novel 3M's proprietary Nano Structured Thin Film (NSTF) catalysts and more conductive 3M PFSA based electrolytes in the electrolyzer MEA [2].

What the cited work does not however addresses is an inherent limitation associated with 3M's Ir-NSTF based anode catalysts, namely their near complete inertness to HOR. An additional tradeoff between performance and gas-crossover also exists when thinner PEM membranes are used. Resulting high hydrogen cross-over creates a gap in the otherwise complete list of requirements for catalyst coated membrane (CCM) that has to be met to become successfully employed in PEM water electrolyzer [3-4]. This was plainly demonstrated in the early 3M catalyst trials where fuel cell derived Pt alloy based NSTF catalysts were used for water electrolysis. The reported durability of such electrolyzer catalysts was excellent, with performance however leaving room for improvement [5]. The performance aspect of 3M electrolyzer catalysts has since been addressed by switching to electrolyzer specific (and Ir based) catalyst compositions. Such approach has left a utility gap by eliminating hydrogen-crossover mitigating components out of the catalyst. To address this problem alternative means for mitigation strategies have to be devised.

In this work we will present some data, such as permeability (gas cross-over) of oxygen and hydrogen as a function of current density and other operational variables, aimed at establishing baselines for un-mitigated hydrogen cross-over of 3M electrolyzer MEAs based on 3M NSTF low-PGM loading catalyst and several types of PerFluoro Sulfonic Acid (PFSA) based PEM membranes, both widely commercially available (such as Nafion™ membranes) and their counterparts made by 3M. Dependence of such unmitigated hydrogen cross-over on applied pressure, pressure differentials, and temperature will be discussed and analyzed with reference to existing models of gas crossover [6-8](. In addition, we will also discuss challenges of in-situ and ex-situ gas –cross over measurements and also intend to present results of our initial attempts to employ alternative (to alloying Ir with Pt) mitigation strategies and gauge their respective effectiveness at various electrolyzer operating conditions and time frames. All proposed X-over mitigation strategies are selected with strong emphasis on compatibility of these candidate solutions with high speed/low cost roll-to-roll manufacturing process that will not be negatively affected by their properties and/or needed modifications.

• In: D. Bessarabov, H. Wang, H. Li, and N. Zhao (Eds): "PEM Electrolysis for Hydrogen Production: Principles and Applications", CRC Press., 2015.

• K. A. Lewinski, S. M. Luopa, "High Power Water Electrolysis as a New Paradigm for Operation of PEM Electrolyzer", Spring ECS Meeting, Chicago, IL, May 2015.

• K. A. Lewinski, et al., "NSTF Advances for PEM Electrolysis - the Effect of Alloying on Activity of NSTF Electrolyzer Catalysts and Performance of NSTF Based PEM Electrolyzers", Fall ECS Meeting, Phoenix, AZ, Oct 2015.

• K. A. Lewinski, et al., ECS Transactions, 69 (17) , p. 893-917 (2015).

• M.K. Debe, et al., Journal of The Electrochemical Society, 159 (6), (2012), p. K165-K176.

• C. Mittelsteadt, M. Umbrell, 207th ECS Meeting, Abstract #770

• D. Bessarabov, "Gas Permeability of Proton Exchange Membranes", Chapter 21, in: PEM Fuel Cell Diagnostic Tools , Editor(s): H. Wang, Xiao-Zi Yuan, Hui Li, CRC Press, Pages: 443-473, 2011

• M. Schalenbach at al., J. Phys. Chem. C 2015, 119, 25156−25169

Figure 1

Porous electrodes for PEM electrolyser and fuel cell technologies are optimized primarily to provide a high specific surface area for the desired electrochemical reactions. However, the increase in surface area comes at the cost of increased voltage losses due to the transport of reactant and product species through the porous medium. In gas evolving electrolyzer electrodes, the formation and transport of gas bubbles plays an important role: on the one hand, gas in the pores replaces the electrolyte phase, thus hindering ion transport and reducing the effective ionic conductivity in the porous electrode; on the other hand, gas bubbles cover and deactivate a portion of the internal surface area.

The impacts of gas formation and removal must be accounted for in the design of electrolyzer electrodes. The focus of the presented work is on the fundamental understanding of the relationships between structure, properties and performance of porous gas-evolving electrodes. The approach combines a macro-scale performance model of the electrode with a micro-scale model of gas evolution. At the macro-scale, a classic porous electrode model describes the transport of ions, electrons and produced oxygen using effective medium theory and accounting for the gas phase volume and distribution.

The micro-scale model describes the formation and growth of gas bubbles based on chemical energy considerations. It can explain the experimentally found high oversaturation that is necessary to nucleate bubbles [1]. Additionally, the size of the bubble nucleus in the experiment was estimated.

The model was used to study the influence of structural parameters on the operation of porous electrodes and to establish guidelines for an enhanced electrode design. It was found that the transport regime of the dissolved gas, viz. diffusion control vs. transfer control at the liquid-gas interface, determines the bubble growth law. Applications of the model to gas evolving porous electrodes in electrolyzers with liquid electrolyte as well as polymer electrolyte electrolyzers (PEEC) will be discussed.

[1] Chen, Q., Luo, L. and White, H. S. Electrochemical generation of a hydrogen bubble at a recessed platinum nanopore electrode. Langmuir 31, 4573–4581 (2015).

Figure 1: Bubble growth on an artificial nucleation site at 20 mA/cm². Comparison of model results with experimental data from C. Brussieux et al., Electrochimica Acta 56 (2011) 7194-7201. Deviation at large bubble radii due to deformation of the bubbles before detachment, which is neglected in the model.

Figure 1

Hydrogen has been identified as an attractive energy carrier for the future in terms of on-site energy generation using for example fuel cells. The production of hydrogen and oxygen (in this case the by-product) from water electrolysis has been commercialised, especially for on-site H₂ generation. For high volume H₂ production using electrolysis, a more intricate system has been proposed in the form of the sulfur dioxide depolarised electrolyser. The SO₂ electrolyser, consisting of a platinum anode and cathode separated by a proton exchange membrane, produces both H₂ and H₂SO₄ from H₂O and SO₂according to equation 1 (anode reaction) and equation 2 (cathode reaction).

SO₂ (g)+ 2H₂O (l) = 2H⁺ (aq.) + 2e^- + H₂SO₄ (aq.) E^o = 0.158 V_SHE(1)

2H⁺ (aq.) + 2e^- = H₂ (g) E^o = 0 V_SHE(2)

The anode is first depolarised by SO₂ to facilitate the electrochemical decomposition of water to protons and electrons. Recombination of the protons (permeating through the PEM) and electrons produces hydrogen at the cathode while H₂SO₄ is formed at the anode. The primary advantage of the SO₂ electrolyser over conventional water electrolysis is based on the theoretical potential needed to drive the reaction. By polarising the anode catalyst with SO₂, the voltage is reduced to E⁰ = 0.158 V_SHE, significantly less than the 1.23 V_SHErequired for normal water electrolysis.

This paper will review the current literature on the operating methods for the SO₂ electrolyser while including an in-depth study performed at HySA Infrastructure CoC, such as MEA manufacturing parameters, effect of H₂S on cell performance. Additionally, the use of PBI based membranes [1] (manufactured by University of Stuttgart through a joint collaboration) in the electrolyser is evaluated by focussing on membrane stability, MEA acid doping and cell performance.

Initially the SO₂ assisted water electrolyser was developed as part of the Hybrid Sulfur (HyS) cycle in the 1970's by the Westinghouse Corporation [2]. The SO₂ electrolyser was operated, for the HyS cycle, using SO₂saturated sulfuric acid as anode and clean sulfuric acid as cathode. The produced sulfuric acid is then returned to the decomposition step.

Sivasumbramanian et al. showed that the electrolyser can also be operated using a gaseous SO₂ anode and a liquid cathode with a proton exchange membrane (PEM) [3] as separator. By this configuration the cell performance could be increased significantly. Staser et al. [4] further investigated this operating method in-depth with regards to the influence of water transport across the PEM on cell performance and acid concentration produced. This operating method was also shown for the application of SO₂ usage in the flue gas compositions where H₂S gas is present. Although a reduced performance was observed it was significantly less than expected when compared to fuel cells for example.

The development of high temperature membranes, such as PBI based polymers[1,5], for operating the electrolyser in the 120 – 160 ᵒC range has been shown to perform either better (at 80 ᵒC) or comparable to that of commercial PFSA membranes in the 80 - 90ᵒC range [5]. However, at this stage, only one article is available REF where humidified SO₂was used as the anode feed using a sulfonated PBI membrane although low temperature (80 and 90 ᵒC) was used [6].

References

[1] Krüger AJ, Kerres J, Bessarabov D, Krieg HM. Evaluation of covalently and ionically cross-linked PBI-excess blends for application in SO2 electrolysis. Int J Hydrogen Energy 2015;40:8788–96. doi:10.1016/j.ijhydene.2015.05.063.

[2] Brecher LE, Spewock S, Warde CJ. The Westinghouse Sulfur Cycle for the thermochemical decomposition of water. Int J Hydrogen Energy 1977;2:7–15. doi:http://dx.doi.org/10.1016/0360-3199(77)90061-1.

[3] Sivasubramanian P, Ramasamy RP, Freire FJ, Holland CE, Weidner JW. Electrochemical hydrogen production from thermochemical cycles using proton exchange membrane electrolyzer. Int J Hydrogen Energy 2007;32:463–8.

[4] Staser JA, Weidner JW. Effect of Water Transport on the Production of Hydrogen and Sulfuric Acid in a PEM Electrolyzer. J Electrochem Soc 2009;156:B16–21.

[5] Kruger AJ, Cichon P, Kerres J, Bessarabov D, Krieg HM. Characterisation of a polaromatic PBI blend membrane for SO2 electrolysis. Int J Hydrogen Energy 2015;40:3122–33. doi:http://dx.doi.org/10.1016/j.ijhydene.2014.12.081.

[6] Jayakumar JV, Gulledge A, Staser JA, Kim C-H, Brian C B, Weidner JW. Polybenzimidazole Membranes for hydrogen and sulfuric acid production in the Hybrid Sulfur Electrolyzer. ECS Electrochem Lett 2012;1:F44–8.

In electrochemical systems, metal surface charging phenomena dictate the strength of electrostatic interactions between the electrified electrode and ions in solution. Historically, the potential of zero charge (pzc) of the metal has been employed to parameterize the surface charging relation [1][2].

However, for a Pt electrode, classical radiotracer experiments of Frumkin and Petrii revealed a non-monotonic charging behavior upon increasing the metal phase potential, with consecutive transitions from a negative to a positive surface charging region and further to another negative surface charging region [3]. Garcia-Araez et al. later reported a similar 'negative-positive-negative' charging relation derived from laser-pulsed experiments [4]. This alternating charging behavior defies the role of the pzc as a sufficient descriptor of electrode charging phenomena. The situation calls for an overhaul of the charging relation that must be derived from first principles. This study presents a refined structural model for electrified interfaces, as shown on the left side of Figure 1. The model accounts explicitly for charging effects caused by the formation of a surface oxide layer and a layer of oriented water molecules. An analytical expression for the charging relation is derived.

The charging relation is parameterized using experimental data for a Pt surface as well as DFT calculations. It exhibits the three different charging regions observed in the experiments cited above. To illustrate the impact of the modified charging relation, it is employed to study the oxygen reduction reaction in nanoporous electrodes of polymer electrolyte fuel cells.

References

[1]. Frumkin, A.; Gorodetzkaya A. Z. Phys. Chem. 1928, 136, 451.

[2]. Petrii, O. A. Zero charge potentials of platinum metals and electron work functions (Review). Russ. J. Electrochem. 2013, 49, 401.

[3]. Frumkin, A. N.; Petrii, O. A. Potentials of zero total and zero free charge of platinum group metals. Electrochim. Acta. 1975, 20, 347.

[4]. Garcia-Araez, N.; Climent, V.; Feliu, J. Potential-dependent water orientation on Pt (111), Pt (100), and Pt (110), as inferred from laser-pulsed experiments. Electrostatic and chemical effects. J. Phys. Chem. C. 2009, 113, 9290.

Figure 1

While first-principles investigations are suitable for exploring oxygen reduction reaction (ORR), they are not straightforward in estimating the final performance of a fuel cell. We are interested in how the important properties of ORR can be extracted and bridged to fuel cell performance. We have calculated the redox potentials and activation energies of the elementary electrochemical reactions of ORR, and evaluated the cyclic and linear sweep volatmograms from rate and diffusion equations. Our results demonstrate how ORR properties affect fuel cell performance and indicate a future direction for fuel cell investigations. The adsorption energy of the intermediates must not only be tuned (a point already made by many researchers from what is referred to as the volcano plot), but the activation energy of the elementary reactions must also be decreased to obtain much higher fuel cell performance beyond the volcano top performance. We took first-principles molecular dynamics (FPMD) to simulate two elementary steps of the electrochemical oxygen reduction reaction on Pt(111);

O(ad) + H⁺ + e⁻--> OH(ad)

and

OH(ad) + H⁺ + e⁻--> H2O

In order to control the electrode potential, i.e. Fermi energy, we used constant-µ scheme [1] of the effective screening medium (ESM) method [2]. The redox potentials of these reactions were determined by changing the OH(ad) and O(ad) coverages. The activation energies of the reactions were determined using the blue-moon ensemble (bond constrained) method [3]. Using the parameters obtained, cyclic and linear sweep voltammograms were obtained.

 We found that it is necessary not only to tune the adsorption energy of the intermediates, O(ad) and OH(ad), which are the origin of the so-called volcano plot, but also to reduce the activation energy of the second elementary step.

 FPMD calculations were performed with "STATE" code [4], which uses density functional theory for electronic structure calculation with plane wave basis, ultrasoft pseudo potentials, and the GGA-PBE correlation functional. A unit cell of 8.38Å×9.67Å×26.04Å in size was established under the periodic boundary conditions, consisting of a slab of three layers of the Pt(111) plane, each containing 12 Pt atoms and about 32 water molecules on it. The details of the calculations are given elsewhere [5].

Computations were performed using the supercomputers at HPCC in Hokkaido University, IMR in Tohoku University, and ITC in the University of Tokyo. This work was partly supported by the NEDO (New Energy and Industrial Technology Development Organization) projects.

[1] N. Bonnet, T. Morishita, O. Sugino, and M. Otani, Phys. Rev. Lett. 2012, 109, 266101.

[2] M. Otani and O. Sugino, Phys. Rev. B 2006, 73, 115407.

[3] M. Sprik and G. Ciccotti, J. Chem. Phys. 1998109, 7737.

[4] Y. Morikawa, K. Iwata, and K. Terakura, Appl. Surf. Sci. 2000. 169-170, 11.

[5] O. Sugino, I. Hamada, M. Otani, Y. Morikawa, T. Ikeshoji, and Y. Okamoto, Surf. Sci. 2007, 601, 5237; M. Otani, I. Hamada, O. Sugino, Y. Morikawa, Y. Okamoto, and T. Ikeshoji, J. Phys. Soc. Jpn. 2008 77, 024802; T. Ikeshoji, M. Otani, I. Hamada, O. Sugino, Y. Morikawa, Y. Okamoto, Y. Qian, and I. Yagi, AIP Adv. 2012, 2, 032182.

Certain aspects of the interpretation of rotating ring-disk electrode, RRDE, data for the oxygen reduction reaction, ORR, in aqueous electrolytes under forced convection have been examined. This presentation will highlight three specific topics:

• • The correct interpretation of the fraction of the oxygen consumption that generates H₂O₂(aq), as extracted from the ratio of ring current to disk current denoted as X_H2O2, A plot of this ratio vs ω^-1/2, where ω is the rotation rate of the disk, should be linear and, with a slope proportional to k₃/X_H2O2 where k₃ is the rate constant for the reduction of solution-phase hydrogen peroxide¹. 

• • Changes in the electrocatalytic activity of Pt for the oxygen and hydrogen peroxide reduction reactions (ORR, and HPRR, respectively) in an aqueous acidic electrolyte induced by the adsorption of bromide, as a model impurity, have been investigated using chronoamperometric techniques. Significant drops in the diffusion-limited currents for O₂ reduction, as well as an increase in the measured H₂O₂(aq) at the ring were induced by the presence of 10 M KBr, and could only be observed for _Br- > 0.25 regardless of E_step. This behavior was found to be consistent with a simple blockage of surface active sites by adsorbed bromide, as predicted by the attenuation model proposed by Levart.

• • An analysis of O₂^- generation at the disk and subsequent oxidation at the ring (see Figure 1) of a RRDE, which undergoes a subsequent second-order dismutation². 

Acknowledgements: This research was supported by a grant from NSF: CHE-1412060

References

1. N. S. Georgescu, A. J. J. Jebaraj and D. Scherson, ECS Electrochemistry Letters, 4, F39 (2015).

2. Z. Feng, N. S. Georgescu and D. A. Scherson, Analytical Chemistry, 88, 1088 (2016).

Figure 1

In order for hydrogen polymer electrolyte membrane fuel cell (PEMFC) technology to be economically competitive, overall system costs still need to be lowered. A key component of cost is the amount (grams) of Pt used in the membrane electrode assembly (MEA). Significant advances in electrocatalyst research, especially focused on the oxygen reduction reaction (ORR) for PEMFC cathodes, have thus been made and have led to the development of highly active catalysts capable of delivering target power at reduced Pt loadings.¹ Unfortunately, as the Pt loading i.e. electrode roughness factor, decreases, an additional resistance emerges leading to voltage loss. This resistance, associated with phenomena at or near the Pt surface, is poorly understood and has been shown to have the greatest impact on low loaded Pt electrodes (<0.1 mg_Pt/cm²) operating at high current densities (>1.0 A/cm²). Therefore, an understanding of the kinetic parameters governing the electrochemical reactions, specifically at the cathode, is essential in order to distinguish between voltage losses that are kinetically derived from those originating from other sources. An understanding of the nature and mechanisms of performance losses in PEMFCs is essential in order for industry to best direct R&D resources.

Discrepancies observed in theoretical kinetic predictions and real performance polarization curves for low Pt loaded MEAs (>0.1 mg Pt/cm2) have led to alterations in the simple Tafel kinetic model² by including an additional term related to oxide coverage.³ While the operating cell potential inherently drops as the catalyst loading decreases, oxide-coverage kinetics become increasingly important in modeling performance especially when transitioning between catalyst surfaces with little to no oxide coverage to those with high oxide coverage. Since some kinetic parameter values change considerably when incorporating an oxide coverage term into the simple tafel model, it is important to know how dependent these changes are in regards to catalyst type and support, as well as in how oxide coverage is defined. Because specific oxide species generated at the catalyst surface can differ with changing conditions e.g. Pt-OH (1e- process), Pt-O (2e- process), it is difficult to obtain a unified model which accurately describes oxide formation. Subsurface oxide generation has also been shown to occur,⁴ which can skew interpretations of oxide measurements. Additionally, oxide formation has been shown to differ substantially among different catalyst surfaces and supports.

This work attempts to clarify the impact that oxide species have on kinetic parameters (exchange current density (i_os), reaction order (γ), activation energy (E_A) etc.) for Pt-based catalysts. Experiments were performed under pure oxygen using a 5cm² differential cell with MEAs consisting of Pt/V, Pt/HSC, and Pt alloy/HSC at loadings of 0.2 – 0.05 mg Pt/cm². Sub-ambient system pressures were employed by means of a vacuum panel in order to lower pure oxygen partial pressure while minimizing gas transport effects. An example measurement is shown in Figure 1, with the oxide and transport free surface potential and current region highlighted. Interpretations of oxide measurement techniques as well as the impact of different catalyst surfaces and supports are discussed. As catalyst activity continues to improve, future implications and significance of oxide kinetics are also considered.

Figure 1. iR-free potential vs log current density as a function of oxygen partial pressure for Pt/Vu electrocatalyst.

The authors would like to acknowledge funding from the U.S. Department of Energy under CRADA #CRD-14-539.

References

1. A. Kongkanand and M. F. Mathias, The Journal of Physical Chemistry Letters, 7, 1127 (2016).

2. K. C. Neyerlin, W. Gu, J. Jorne and H. A. Gasteiger, Journal of the Electrochemical Society, 153, A1955 (2006).

3. N. Subramanian, T. Greszler, J. Zhang, W. Gu and R. Makharia, Journal of The Electrochemical Society, 159, B531 (2012).

4. B. Conway, B. Barnett, H. Angerstein‐Kozlowska and B. Tilak, The Journal of chemical physics, 93, 8361 (1990).

Figure 1

H₂-fed proton exchange membrane fuel cells (PEMFCs) represent the most advanced fuel cell technology and have a great deal of potential applications in low/zero-emission electric vehicles, distributed home power generators, and power sources for small and portable electronics. However, the commercialization of PEMFC technology has been greatly hindered by some challenges, mainly sluggish kinetics of the oxygen reduction reaction (ORR) at the cathode and the high cost of the noble metal Pt (1, 2).

Alloying platinum with non-noble metals such as Co is an effective approach to improve the catalytic performance and reduce the usage of Pt. In addition, through transferring Pt nanoparticles (NPs) into a hollow nanostructure, electrocatalytic performance can be improved greatly due to its relatively lower density and higher surface area-to-volume ratio than its solid counterpart (i.e., NPs).

In this work, Pt hollow nanospheres (HNSs) were fabricated through a replacement reaction of Co atoms by PtCl₆^2- ions to reduce Pt usage and improve the catalytic activity towards the ORR. Carbon black Vulcan XC-72R (VC) was introduced into a solution prior to the addition of Co(II) and the formation of Co NPs and the replacement of Co by PtCl₆^2-ions for a uniform dispersion of Co NPs and the Pt HNSs on the carbon support. Some Co atoms have alloyed Pt in the synthesis and exist in the Pt HNSs (3).

The hollow mesoporous core in the Pt HNSs can be utilized as an electrolyte solution buffering reservoir to minimize the diffusion distance to the interior surface of the porous shell of Pt crystallites while the porous nanochannels (i.e., the micropores between the Pt crystallites) in the shell open to the mesoporous hollow core form fast mass transport networks providing more accessible sites for oxygen transfer. In addition, triple phase boundaries (i.e., gas-electrolyte-Pt NPs) can be developed more easily in the Pt HNS enabling individual Pt crystallites in the shell to be accessible to electrolyte ions and oxygen, and thus catalytically active. Furthermore, alloying Pt with Co might result in a significant lattice shrinking because of the change in Pt-Pt bond distance which also contributes to the improved electrocatalytic activity. In contrast, for the state-of-the-art Pt NPs/VC catalyst, the Pt NPs can agglomerate more easily to form larger particles, resulting in reduced active sites. Besides, the interior voids between Pt NPs may not be accessible to electrolyte ions and do not contribute to the ORR activity due to the lack of triple phase boundaries. As a result, the as-developed Pt(20 wt%) HNS/VC catalyst outperforms significantly the state-of-the-art Pt(20 wt%)NP/VC catalyst.

References

1. B. Fang, M. Kim, J. Kim, M. Song, Y. Wang, H. Wang, D. Wilkinson and J.-S. Yu, J. Mater. Chem., 21, 8066 (2011).

2. B. Fang, N. Chaudhari, M. Kim, J. Kim and J.-S. Yu, J. Am. Chem. Soc., 131,15330 (2009).

3. B. Fang, B. Pinaud and D. Wilkinson, Electrocatalysis, DOI: 10.1007/s12678-016-0311-4 (2016).

Figure 1

One of the strategies to further increase PEFC performance is elevation of the operation temperature, reducing the reaction overvoltage of oxygen reduction reactions. On the other hand, durability of materials used in PEFC is assumed to be lowered at the higher temperature condition. In this study, understanding degradation phenomena of electrocatalysts at the higher temperature is particularly focused on in order to develop durable electrocatalysts at the elevated temperature.

MEAs were prepared using the standard catalyst and electrolyte, Pt/Ketjen black catalyst (TEC10E50E, TKK Corp.) and Nafion, respectively. Three different conditions of operation temperature and humidity (80 °C/RH100%, 105 °C/RH57%, and 80 °C/RH57%) were applied. Durability of MEAs in three conditions was studied through the potential cycles between 0.6 and 1.0 V assuming "load cycles" of FCVs.^{[1], [2]} 

Since Pt dissolution/re-precipitation is assumed to occur during the cycles, the changes in activation overvoltage and ECSA were carefully examined under three conditions. After 20000 potential cycles, activation overvoltage and ECSA increased and decreased, respectively, the most at 105 °C/RH57%. On the other hand, such degradation was much suppressed under the operation at 80 °C/RH57%. In order to understand the difference in degradation phenomena, the amount of water molecules was considered. With the presence of a large amount of water molecules, the rate of reducing ionized Pt (Pt²⁺) into Pt metal is assumed to be smaller. Then, in the case of 80 °C/RH57%, the amount of water molecules was smallest among three conditions, leading to fast and reversible reduction of Pt²⁺to Pt metal. Therefore, Pt particle growth was believed to be suppressed. However, if two factors, the amount of water and temperature, are considered, the temperature factor most likely influences more on Pt particle growth. 

Reference

[1] A. Ohma et al., ECS Trans, 41 (2011) 755.

[2] M. Kitamura et al., ECS Trans, 69 (2015) 701.

The durability of polymer electrolyte fuel cells (PEFCs) is a big concern for its applications like FCVs to be commercialized on a large scale. In particular, the instability of Pt-based cathode catalyst has been discussed intensively. Platinum (Pt) dissolution is recognized as one of the degradation mechanisms for carbon supported Pt catalysts (Pt/C) in PEFCs. It has been found that repeated potential cycles accelerate Pt dissolution.^[1-3] Fundamental studies in aqueous media revealed preliminary behaviors of Pt dissolution during one potential cycle. The up-to-date data were mostly obtained near room temperature, and a dominant cathodic dissolution during reduction of Pt oxide was understood. As reported by Fuller et al.,^[4] however, degradation of Pt/Cs at an elevated temperature near 60 to 80 °C, whereat a PEFC normally operates, was severer than that at room temperatures. Thus, it is necessary to evaluate Pt dissolution from Pt/Cs at such temperatures.

In prior studies,^[5-6] we compared Pt dissolution at 25 °C and 65 °C by inductively coupled plasma mass spectrometry (ICP-MS) and a channel flow double electrode (CFDE). The experiments had some difficulties, but the results showed a ca. 5-time enhancement from 25 to 65 °C. The cathodic dissolution was enhanced because larger amount of Pt oxide was formed at 65 °C. The effect of temperature on the anodic dissolutions and the place exchange process remains unclear yet.

In this study, we evaluate Pt dissolution under temperatures ranging from 20 to 80 °C. We test a series of commercial Pt/C catalysts produced by TKK that have Pt loadings from 20 to 70 wt% and average Pt particle size from ca. 1.5 to 5 nm. We limited the potential region between 0.6 and 1.4 V, which is closely related to the operation region of a PEFC cathode. We use ICP-MS to evaluate to overall dissolutions and a CFDE to analyze the dissolution behavior in one potential cycle. The attached Figure showed a typical CFDE data presenting the current on a working electrode loading a 30wt% Pt/C (TEC10E30A) under potential cycling (I_WE) and the current recorded on a collector electrode (I_CE) at 80 °C in 0.5 M H₂SO₄. The I_CE helps to understand the dissolution of Pt⁴⁺ and Pt²⁺ complex during one potential cycle. Similar analysis will be discussed at other temperatures like 20, 40, and 80 °C in the presentation.

Reference

• Z. Wang, E. Tada, and A. Nishikata, J. Electrochem. Soc., 161, F380 (2014).

• A.A. Topalov, S. Cherevko, A.R. Zeradjanin, J.C. Meier, I. Katsounaros, and K.J.J. Mayrhofer, Chem. Sci., 5, 631 (2014).

• P. Jovanovič, A. Pavlišič, V.S. Šelih, M. Šala, N. Hodnik, M. Bele, S. Hočevar, and M. Gaberšček, ChemCatChem, 6, 449 (2014).

• W. Bi and T.F. Fuller, J. Electrochem. Soc., 155, B215 (2008).

• Z. Wang, E. Tada, and A. Nishikata, J. Electrochem. Soc., 163, F1-F3 (2016).

• Z. Wang, E. Tada, and A. Nishikata, ECS Trans., 69, 255 (2015).

Figure 1

Introduction

Pt-M alloys catalysts have attracted attention over the years and expected to be used in proton exchange membrane fuel cells (PEMFCs) cathode catalysts due to their high oxygen reduction activity than pure Pt and reduction of Pt usage in PEMFCs [1]. However, Pt-M catalysts is easily degraded by both Pt and M dissolution from Pt-M alloys during on-off cycle and load cycle, which leads the performance loss of PEMFCs [2]. Thus, to understand the nature of dissolution mechanism of both Pt and M from Pt–M alloy catalysts is very significant to advance durability of PEMFCs under its operating conditions. In this study, we investigated that the dissolution mechanism of Pt-50 at% Fe alloy (Pt₅₀-Fe₅₀) and Pt₅₀-Cu₅₀ under potential cycling with a channel flow multi electrodes (CFME), and elucidate the effect of additive element to the dissolution behavior of Pt₅₀-M₅₀.

Experimental

Pt₅₀-Fe₅₀ and Pt₅₀-Cu₅₀ were subjected to potential cycling tests at 298 K using Ar-purged 0.5 M H₂SO₄ solution with a CFME. A double junction KCl-saturated Ag / Ag-Cl electrode was used as reference electrode, and Au coil was used as the counter electrode. Flow rate of the electrolyte was 10 cm s^-1 in order to establish laminar flow condition. Potential cycling tests were employed between 0.05 V and 1.4 V vs. SHE at 20 mV s^-1. Before potential cycling, Pt₅₀-M₅₀ was set at 0.45 V vs. SHE in order to remove Pt oxide layer formed on Pt–M alloy surface in ambient air and to measure baseline currents of collector electrodes (CEs). Reactions and potentials for detecting the dissolved Fe²⁺, Fe³⁺, and Cu²⁺ on Au-CEs are as follows,

Fe²⁺ → Fe³⁺ + e^- (E_c = 1.0 V vs. SHE) (1)

Fe³⁺ + e^-→ Fe²⁺ (E_c = 0 V vs. SHE) (2)

Cu²⁺ + 2e^-→ Cu (E_c = 0 V vs. SHE) (3)

Dissolved Fe²⁺, Fe³⁺, and Cu²⁺ from Pt₅₀-M₅₀ are monitored by the current change of CEs.

Results and discussion

 Figure 1 shows M detection current with a CFME during potential cycling, and red and blue line show the results of Pt₅₀-Fe₅₀ and Pt₅₀-Cu₅₀, respectively. As shown in inset, Fe dissolution from Pt₅₀-Fe₅₀ starts from around 0.3 V in anodic scan. Then Fe dissolution is clearly enhanced between 0.6 and 1.4 V in both anodic and cathodic scan, and finally terminate at 0.3 V in cathodic scan. Wang et al reported that Pt dissolution occur above approximately 0.6 V [3], accordingly major Fe dissolution is enhanced by Pt dissolution. On the contrary, minor Fe dissolution between 0.3 and 0.6 V in both anodic and cathodic scan is contributed by Pt surface diffusion, because Pt atoms are easily diffuse on Pt-M alloy surface in this potential range due to no adsorbed species on its surface [4]. Cu dissolution from Pt₅₀-Cu₅₀ only appears above 0.6 V, whereas Fe dissolution from Pt₅₀-Fe₅₀ starts from 0.3 V. Moreover, the amount of Cu dissolved from Pt₅₀-Cu₅₀ is smaller than that from Pt₅₀-Fe₅₀, which indicate that Cu is more durable elements than Fe. This is explained by the difference of standard electrode potentials between Cu and Fe. Pt-enriched layer formed on Pt₅₀-Cu₅₀ is denser and flatter than that on Pt₅₀-Fe₅₀, because the amount of dissolved Cu is small due to more noble standard electrode potential than Fe. Flat Pt-enriched layer inhibits Pt surface diffusion between 0.3 and 0.6 V, and Pt dissolution above 0.6 V.

Reference

[1] V. R. Stamenkovic, B. S. Mun, M. Arenz, K. J. J. Mayrhofer, C. A.Lucas, G. Wang, P. N. Ross, and N. M. Markovic: Nat. Mater.6 (2007) 241–247.

[2] H. A. Gasteiger, S. S. Kocha, B. Sompalli, and F. T. Wagner: Appl. Catal. B: Environ.56 (2005) 9–35.

[3] Z. Wang, E. Tada, and A. Nishikata: J. Electrochem. Soc.161 (4) (2014) F380–F385

[4] Q. Xu, E. Kreidler, and T. He: Electrochim. Acta55 (2010) 7551–7557.

Figure 1

We developed connected carbon-free platinum-iron (PtFe) nanoparticle catalysts with porous hollow capsule structure (Fig. 1) as oxygen-reduction-reaction (ORR) electrocatalysts for polymer electrolyte fuel cells (PEFCs).^[1] This catalysts consist of a beaded network by connected PtFe nanoparticles with a crystallite size of 6~7 nm and a chemically-ordered (face centered tetragonal) superlattice structure. The beaded network is electrically conductive; thus, carbon-supports can be removed from catalyst layers in PEFCs. We have demonstrated that an MEA prepared using a carbon-free cathode of connected PtFe-nanoparticle catalysts was highly durable against start-up/shut-down operations because of the elimination of carbon-corrosion problems. Moreover, an ORR specific activity of a connected PtFe catalyst is about 9 times higher than that of a commercial Pt-nanoparticle catalyst supported on carbon black (Pt/C).

In this study, the structural effects of a connected PtFe catalyst on an ORR activity were investigated in order to elucidate the factors of its high ORR activity. Note that the connected PtFe catalyst exhibited about twice higher ORR specific activity than the PtFe-nanoparticle catalyst (just nanoparticles without any connection between them) on carbon black. In addition, the ORR specific activity of the connected PtFe nanoparticles with hollow structure (with no supports) was 1.5 times higher than that of the connected PtFe nanoparticles on the electroconductive carbon support. These results indicate that the unique structures of connected nanoparticle catalysts, such as connected beaded network and hollow structure, enhance an ORR activity.

We also have performed the refined structural analyses of a connected PtFe catalyst by using in-situ X-ray absorption spectroscopy (XAS) combined with Rietveld structure refinement with powder X-ray diffraction data, and spherical aberration corrected scanning transmission electron microscope (Cs-corrected STEM). The in-situ XAS analysis disclosed shorter Pt‒Pt bond distance and more Pt 5d vacancy of the connected PtFe catalyst compared with the commercial Pt/C. In addition, the Cs-corrected STEM observations revealed that the beaded network formed by connected PtFe nanoparticles possessed polycrystalline structure. In this presentation, the effects of the geometric and electronic structures of a connected PtFe catalyst on an ORR activity will be discussed in details. The knowledge obtained in this study provides guidelines for the design of ORR catalysts with enhanced ORR activities to achieve a high performance PEFC.

[1] T. Tamaki, H. Kuroki, S. Ogura, T. Fuchigami, Y. Kitamoto, and T. Yamaguchi, "Connected nanoparticle catalysts possessing a porous, hollow capsule structure as carbon-free electrocatalysts for oxygen reduction in polymer electrolyte fuel cells", Energy Environ. Sci., 2015, 8, 3545-3549.

Figure 1

The widespread use of polymer-electrolyte fuel cells (PEFCs) requires more active, stable, and low cost electrocatalysts than platinum. Bimetallic electrocatalysts such as platinum-iron (PtFe) have been attracting a great deal of attention as an oxygen reduction reaction (ORR) catalyst since they have demonstrated both higher oxygen reduction activity and improved stability with much smaller amounts of platinum. Connected PtFe-nanoparticle catalyst with a beaded network by connected PtFe nanoparticles and a superlattice structure (a chemically-ordered face-centered tetragonal (fct) structure) exhibit specific ORR activity that is higher than those of the commercial Pt catalyst supported on carbon black (Pt/C) and even fct-PtFe nanoparticles (without any beaded networks) supported on carbon black. [1] Thus, the unique structures of connected PtFe-nanoparticle catalysts, such as connected beaded network and hollow structure, enhance an ORR activity.

Electrochemical x-ray photoelectron spectroscopy (EC-XPS) applied for the reaction mechanism analyses of PEFC catalysts. Electrochemical reactions occur at catalyst electrode surface, and then, the electrode is transferred to an UHV chamber immediately. Finally the electrode surface involving reaction intermediates is analyzed by using XPS without exposing to the air. Since structure of electrochemical-double layers is maintained during the measurement process, we can identify the intermediate species, and evaluate adsorption strength, etc., namely, we can analyze reaction mechanism. In this study, we performed electrochemical x-ray photoelectron spectroscopy (EC-XPS) measurements in order to understand the factors that influence the high ORR activity of the connected PtFe-nanoparticle catalyst. The EC-XPS results for connected PtFe-nanoparticle catalysts under ORR conditions showed the adsorption of oxygen species on the catalyst surfaces, indicating that ORR intermediates were detected. We could observe ORR processes of connected PtFe-nanoparticle catalysts. In this presentation, we will also show the EC-XPS measuring data for other platinum catalysts and discuss the factors that influence the high ORR activity of the connected PtFe-nanoparticle catalyst in detail.

[1] T. Tamaki, H. Kuroki, S. Ogura, T. Fuchigami, Y. Kitamoto, and T. Yamaguchi, "Connected nanoparticle catalysts possessing a porous, hollow capsule structure as carbon-free electrocatalysts for oxygen reduction in polymer electrolyte fuel cells", Energy Environ. Sci., 2015, 8, 3545-3549.

The current-voltage performance of polymer electrolyte fuel cells (PEFCs) in the intermediate voltage region has received attention from the standpoint of increasing the maximum power of the cell. The performance-limiting factors in this kinetic-mass transport mixed control region were first considered as a protonic resistance in the polymer electrolyte, an oxygen diffusion resistance in the pores of the catalyst layer (CL) and gas diffusion layer (GDL), and a diffusion resistance of the dissolved oxygen in the polymer electrolyte that covers the catalyst. These three factors were taken into consideration in mathematical models that predict the performance, which resulted in the predictions much better than experimental results. To bridge the gap, the diffusion resistance in the catalyst agglomerates, which are composed of carbon-supported catalysts and an ionomer, was proposed. However, huge agglomerates that the models assumed were not observed in the scanning electron micrographs of the cross sections of the catalyst layers. In the meanwhile, Kudo and Morimoto measured the oxygen diffusion resistance using Nafion thin films and concluded that the interfacial resistance at the Nafion-Pt or Nafion-gas interface gave rise to the diffusion resistance [1]. The existence of the resistance at the Nafion-Pt interface was supported by molecular dynamics simulations [2]. In this work, the interfacial resistance estimated from the measurement of limiting current densities was incorporated into the performance model and the model predictions were compared with experimental results. The discrepancy between them was assumed to stem from a dependence of diffusion resistance on the electrode potential and then the dependency was analyzed.

In the experiments, the cell was operated at 65 °C with a large excess flow rate of hydrogen and 1% oxygen balanced with nitrogen, both humidified to 80% relative humidity, under atmospheric pressure. The condition was selected to reduce the overpotential variation in the through-plane direction caused by the protonic resistance and to avoid the effect of the produced water and generated heat, and thus to emphasize the effect of oxygen diffusion. Following the seminal studies of Mashio et al. [3], we decomposed the diffusion resistance into two resistances in series, namely, the one caused by molecular diffusion (R_molec) and the remainder (R_other). R_molec was estimated with the aid of limiting-current density measurements using 1% oxygen balanced with helium instead of 1% O₂-N₂. R_other was calculated by subtracting R_molec from the overall diffusion resistance. We well call R_other determined by this procedure as R_other^ref.

The main features of the model to analyze the experimental results are as follows: (1) One-dimensional through-plane distribution is considered; (2) The anode CL and GDL are neglected; (3) Overpotential distribution and hence the reaction rate distribution is taken into account; (4) The oxygen reduction reaction rate is expressed by Tafel equation and is proportional to the oxygen concentration at the catalyst surface; (5) Oxygen concentration at the catalyst surface is calculated using the diffusion resistance and reaction rate.

In Figure 1a, an experimental result (dotted line) and a model prediction (solid line) of the performance of a cell are compared. In the model, constants in Tafel equation are selected so that the prediction fits to the experiment in the activation-controlled region; the membrane and CL resistances were the values measured by impedance spectroscopy; R_molec and R_other^ref were used as the diffusion resistances. The model prediction gives larger current density than the experiment in the intermediate cathode potential region. We assumed that the discrepancy stemmed from the constant R_other (= R_other^ref), and therefore hypothesized that R_other is a function of electrode potential. We therefore increased R_other from R_other^ref at each potential to find the value that the current density from model prediction matched the experimental value. The resulting R_other is shown in Fig. 1b as a function of cathode potential (R_other^ref is shown as a dashed line). R_other increased with potential except for high potential region, where the effect of diffusion resistance on the performance is less significant and hence the accuracy of the fitting is deteriorated. The effect of catalyst loading will be discussed and a mathematical formulation of R_other will be presented.

[1] K. Kudo, Y. Morimoto, ECS Trans., 50, 1487 (2012)

[2] R. Jinnouchi, K. Kudo, N. Kitano, Y. Morimoto, Electrochim. Acta, 188, 767 (2016)

[3] T. Mashio, A. Ohma, S. Yamamoto, K. Shinohara, ECS Trans., 11, 529 (2007)

Figure 1

Polymer electrolyte fuel cells (PEFCs) are expected as a next-generation power sources due to their low emissions and high efficiencies. However, it is required to improve the power density at lower Pt loading. It is known that the losses of the power density are caused by activation losses, ohmic losses and mass transport losses, and that the mass transport losses are dominant at higher current density. Moreover, it is suggested that one of the causes of the mass transport losses is the shortage of oxygen in the cathode side of PEFC. In the cathodic catalyst layer, there are Pt catalysts on supported carbon microparticles, and those particles are covered with ionomer films which are composed of polymer electrolytes and water molecules. The ionomer has two properties for the water-generating reactions in the cathodic catalyst layer: the proton conductivity and the oxygen permeability. In particular, the dependence of the oxygen permeability on water content has not been clear.

In this study, we analyzed water content dependence of oxygen permeation properties in ionomer on Pt surface using molecular dynamics simulations. Regarding the calculation system, Nafion was adopted to the polymer electrolyte in the ionomer, and water content, λ, was defined as the ratio of the number of sulfo groups in Nafion to the number of water molecules. We set λ = 3, 7, 11 and constructed a system of an equilibrium state of ionomer on Pt surface at each water content. Then oxygen molecules were inserted above the ionomer and permeate the ionomer along the thickness direction at steady state.

Firstly, the density distributions were obtained to evaluate the structural properties of the ionomer. As a result, the region of the ionomer is divided into three region: ionomer/gas interface, bulk region, ionomer/Pt interface. Next, the number of permeated oxygen molecules through the ionomer was counted, and the oxygen permeability of the ionomer was estimated at different water content. As a result, the oxygen permeability decreases with increasing water content. Gas permeability is estimated using the permeation coefficient which is obtained by the product of the diffusion coefficient and solubility coefficient of gas molecules, indicating that the diffusivity and solubility are governing factors of the permeability. Therefore, the oxygen diffusivity and solubility was evaluated to discuss the oxygen permeation mechanism in the ionomer. The oxygen diffusion coefficient was obtained by the density distribution of the oxygen molecules along the thickness direction using the Fick's law. As a result, the oxygen diffusivity in each region of the ionomer decreases at lower water content while slightly increases at higher water content, with the increase in water content. On the other hand, the oxygen solubility coefficient was calculated using the test-particle-insertion method, and the solubility in each region of the ionomer decreases with increasing water content. Finally, the oxygen permeability was evaluated using the diffusivity and solubility in each region. Consequently, the oxygen permeability decreases with increasing water content in each region, and the permeability in the ionomer/Pt interface is the smallest, which indicates that the oxygen dissolution in the ionomer/Pt interface is dominant in the oxygen permeation through the ionomer.

The objective of this work is to gain the mechanical understanding of the water management in the proton exchange membrane fuel cell (PEMFC). Water is a by-product of the fuel cell reaction and its amount is proportion to the current of fuel cell output. Water is used to retain the proper hydration level in the PEM. At the same time, condensed water in the flowfield and the gas diffusion layer (GDL) reduces oxygen transport to the oxygen reduction reaction (ORR) area. Therefore, excess accumulation of condensed water causes to lower the performance, particularly at the higher current densities. Detailed simulation of two-phase water in the GDL is significantly important to understand the water management.

This work shows the successful in development of a multi-scale calculation technique that incorporates various scales of numerical model in a fuel cell and simultaneously performed a prediction. The outcomes of this work are, (i) development of two-phase fluid model in the micro-scale porous structure of GDL, and (ii) integration of this micro-scale GDL fluid model with a state-of-the-art macro-scale flowfield fluid model to predict two-phase water in the fuel cell. Figure 1 gives the concept of modeling approach at each component of the fuel cell. The flowfield in the bipolar plate and MEA models are calculated using traditional computational fluid dynamics (CFD) method with existing PEMFC model [1-3]. The two-phase fluid in the micro-scale porous structure of GDL is numerically predicted by Lattice Boltzmann Method (LBM) [4-5]. The solution of each iteration from CFD and LBM are required to simultaneously exchange for the next iteration until all solutions are converged, especially at the interfaces. In this figure, it shows the flowfield is transformed into the macro-scale CFD model and the image of detail structured GDL is converted into micro-scale LBM model. Actual porous structure imaging of GDL is obtained by the nano-scale high special resolution x-ray computed tomography. Figure 2 shows an example of the predictions from the simulation when the macroscopic flowfield model is combined with the microscopic GDL model while performing the calculation. The predictions give the detail distributions of variables in every components of the multiscale model. In this figure, the current density distribution on MEA surface, the liquid water transport inside the GDL, and the temperature of solid and fluid phases inside the GDL are presented.

References:

1. S. Shimpalee, D. Spuckler, and J. W. Van Zee J. Power Sources, 167 (2007) 130-138.

2. S. Shimpalee, M. Ohashi, C. Ziegler, C. Stoeckmann, C. Sadeler, C. Hebling, J. W. Van Zee, Electrochemica Acta, 54 (2009) 2899-2911.

3. S. Shimpalee, J. of Electrochem. Soc., 161 (2014) E3138-E3148.

4. U. Frisch, B. Hasslacher, Y. Pomeau, Physical review letters 56 (14) (1986) 1505-1508.

5. G. R. McNamara, G. Zanetti, Physical Review Letters 61 (1988) 2332-2335.

Figure 1

Polymer electrolyte fuel cells (PEFCs) have been developed for automotive, residential, and portable power applications. Typically, PEFC stacks use platinum (Pt) or Pt alloys as the oxygen reduction reaction (ORR) catalyst in the cathode. Although Pt performs well, its raw material cost is significant and limits future cost reductions. As an alternative, platinum group metal-free (PGM-free) catalyst may have the potential to significantly reduce PEFC costs. Despite recent advances in PGM-free catalyst ORR activity,¹ the volumetric activity remains well below that of Pt cathodes and results in thick electrodes with significant transport losses.

To advance PGM-free catalyst electrodes, it is important to understand the coupling between electrochemical reactions, transport phenomena, and microstructure in the catalyst layer. Here, we expand this understanding using a combination of nano-scale X-ray computed tomography (nano-CT) and direct simulations on the CT images. Nano-CT imaging was done for cyanamide-polyaniline-iron PGM-free catalyst cathodes with three different Nafion^® loadings (35, 50 and 60 wt%). Using Zernike phase contrast we differentiated the solids (Nafion^®, carbon, and iron) and the pore phase,² and by staining the electrode with cesium, we distinguished the Nafion^® phase by absorption contrast in a second scan. The two data sets were overlaid to complete the structure of the whole cathode.³ In other words, at each voxel we have the volume fractions of pore, Nafion^®, and carbon catalyst phases. To illustrate, Figure 1a shows the Nafion^® fraction over the simulation domain where the Nafion^® had agglomerated and was heterogeneously distributed. The spatially-resolved volume fraction data is then input to the simulation of oxygen, proton, and electron transport as well as the local ORR volumetric current. Figure 1b shows the ionic potential and current streamlines in Nafion^® phase resulting from the distributed reaction.

Acknowledgement

The authors gratefully acknowledge the support the support of the technology development manager Nancy Garland and funding from the US Department of Energy, the Office of Energy Efficiency and Renewable Energy, Office of Fuel Cell Technologies. The nano-CT instrument was acquired with the support of a Major Research Infrastructure award from the National Science Foundation under Grant No. 1229090.

References

1. G. Wu, K. L. More, C. M. Johnston, and P. Zelenay, Science, 332, 443–447 (2011).

2. S. Komini Babu, H. T. Chung, G. Wu, P. Zelenay, and S. Litster, ECS Trans., 64, 281–292 (2014).

3. S. Komini Babu, H. T. Chung, P. Zelenay, and S. Litster, ECS Trans., 69, 22–33 (2015).

Figure 1

INTRODUCTION

Although extensive studies have been conducted on PEFC behavior and a lot of knowledge has been published, understanding the PEFC is still not easy since the system is considerably complicated. By investigating a dimensionless form of equations describing the cathode catalyst layer (CL), the cathode model was simplified and a few intrinsic moduli which control the dimensionless rate profile were proposed (1). The impact of convection on the oxygen reduction reaction (ORR) rate and some case studies are demonstrated.

DIMENSIONLESS MODEL

The ORR rate is proportional to the oxygen partial pressure at fixed cathode emf. The emf dependency follows a Tafel equation. Oxygen balance in cathode catalyst layer is expressed as follows:

d(N_gy_O – C_gD_eOdy_O/dz) / dz = –r_vc[1]

where N_g is the total gas flux, y_O is the oxygen mole fraction, and D_eOis the effective oxygen diffusivity (2). Proton flux can be calculated by the reaction stoichiometry. Proton potential follows the Ohm's law.

With the 1D model of cathode CL, 12 variables including boundary conditions have to be specified so as to determine the profiles of partial pressure, emf, reaction rate, etc. Dimensionless forms of model equations were derived, which include the following dimensionless moduli we propose:

M_O^(C)_m =  δ^(C) (k_vcm RT / D_eO)^1/2[2]

M_p^(C)_m =  δ^(C) {4F k_vcm p_Oc / (σ_epb_c)}^1/2 [3]

where σ_ep is the effective proton conductivity, b_c is the Tafel slope, δ^(C) is the CL thickness, k_vcm is the ORR rate constant per unit volume of cathode CL at the PEM–CL boundary and p_Oc is oxygen partial pressure at the CL–GDL boundary. k_vcm is the greatest rate constant and p_Oc is the highest partial pressure in the CL. Therefore, k_vcmp_Oc represents the intrinsic ORR rate without transport resistance. M_O^(C)_m and M_p^(C)_mrespectively represent the ratio of reaction performance to oxygen transport performance and the ratio of reaction to proton transport.

A Peclet number which represents the ratio of convection to diffusion of oxygen is also required for formulation:

P_O^(C)_m =  δ^(C)N_A^(M) /(C_g D_eO) [4]

where N_A^(M)is the net water flux through the PEM, which equals the total gas flux at the PEM–CL boundary.

The dimensionless ORR rate profile is determined by only 4 dimensionless parameters, M_O^(C)_m, M_p^(C)_m, P_O^(C)_m, and y_Oc. The dimensionless system greatly reduces the complication of system.

RESULTS AND DISCUSSION

Define the cathode effectiveness factor F_ec as a ratio of the ORR rate to the intrinsic ORR rate without transport resistance. Figure 1 shows that increase in transport resistance reduces the effectiveness factor. Typical values of M_O^(C)_m and M_p^(C)_m in usual cells are around 1. When M_p^(C)_m is high, proton transport is slow, which increases the cathode emf and reduces the ORR rate. If M_O^(C)_m >> M_p^(C)_m, oxygen transport is suppressed and ORR takes place near the oxygen inlet, the GDL side.

As the convective flow from PEM to GDL increases, oxygen distribution is shifted towards the GDL, reducing the effectiveness factor as shown in Fig. 2. Even at quite low M_O^(C)_m and M_p^(C)_m of 0.01, the convection can reduce the effectiveness factor. When M_O^(C)_m and M_p^(C)_m are high, the impact of convection is not remarkable since the reaction takes place only near the boundaries. When back diffusion of water in the PEM is dominant, P_O^(C)_m can be negative. F_ec reaches 1 under reaction control.

Figure 3 shows a case study in which the CL thickness δ^(C) is varied with a fixed Pt loading per volume. As δ^(C) increases, the Pt amount increases proportionally but the transport resistance increases M_O^(C)_m and M_p^(C)_m and the effectiveness factor decreases as shown in Fig. 1. As a result, the increase in the current density is limited. Since the increase in δ^(C)raises both moduli, it overtakes the increase of Pt; a maximum current appears in Fig. 3.

Employing the proposed dimensionless moduli, the effects of dimensions and operating conditions can be estimated quantitatively without expensive simulation.

Acknowledgement This work is a part of the project by the New Energy and Industrial Technology Development Organization (NEDO), Japan.

REFERENCES

1. M. Kawase et al., 24th Int. Symp. Chem. React Eng. (Minneapolis, June 2016).

2. M. Kawase et al., ECS Trans. 16(2), 563–573 (2008).

Figure 1

Abstract: The polymer electrolyte membrane fuel cell (PEMFC) possesses many advantages for both automotive and stationary application, including high energy efficiency, low operating temperature, zero emission, and so on. Aside from the fact that great improvement should be made on its key materials, i.e., the ORR electrocatalysts and proton exchange membrane, it is believed that the PEM fuel cell performance is greatly affected by the operating temperature, gas inlet humidity as well as the flow pattern, and so on [1,2]. Thus, in this work, a 3D steady state model is established to investigate detailedly the effects of the flow pattern (co-flow and counter-flow), the anode and cathode gas inlet relative humidity (RH) on the cell performance. The governing equations of the 3D model result from careful analysis on the electro-chemical reactions, current conservation, membrane proton migration, membrane water transport and water-vapor phase transition. The membrane water transport takes into accounts the electro-osmatic drag and water back diffusion, and the conservation of momentum, species and energy is applied to all components of the PEM fuel cell. Experimental validation is also performed and fits very well with the simulation. Figure 1 shows the current density distribution for different flow patterns: (a), co-flow and (b), counter-flow. The corresponding achievement will offer an efficient guide on the design and performance optimization of PEMFC.

Acknowledgements

This work was supported in part by National Natural Science Foundation of China (Grant No. 21373135 and 21533005) and Science Foundation of Ministry of Education of China ( Grant No. 413064).

References

• K. Dannenberg, P. Ekdunge, G. Lindbergh, Mathematical model of the PEMFC, Journal of Applied Electrochemistry, 2000, 30: 1377-1387.

• S.H. Ge, B.L. Yi, A mathematical model for PEMFC in different flow modes, Journal of power sources, 2003, 124: 1-11.

Figure 1

Building a catalyst layer includes many elements that make understanding and studying it a complex problem. These systems have different elements with multiscale, multiphase and multicomponent issues, as Figure 1 illustrates. Typical PEMFC catalyst layers are composed of three elements: porous carbon to provide electron and gas/liquid transport, ionomer to provide protonic transport and catalyst to facilitate the reduction or oxidation reactions.

The replacement of Pt with non-precious-metal-based catalysts (NPMCs) for the ORR is considered to be one key to cost-effective power generation in PEMFCs. Accordingly, developing NPMCs with high performance for ORR (1-4) has been a focal point of research. A major complication in analysis of the NPMC systems under study lies in the unknown nature of the active site. It is difficult to identify structural and electronic parameters based on most characterization techniques. Many questions arise when compared with Pt-based systems such as: Which elements do Pt and NPMC catalyst layers have in common? What makes one metal more suitable than another in the NPMC? What specific catalyst layer composition and structure are optimal? How do we (or can we) correlate energetics, structure and transport issues?

Modeling catalyst layers is a complicated task because we need to take into account the phenomena occurring over different time and size scales as represent in Figure 1: (i) heterogeneous electrochemical reactions in the microscale; (ii) proton, electron and gases transport in the mesoscale; and (iii) gases and water transport through membranes and porous media in the macroscale (5-10).

The first steps are to identify and classify the main macroscopic parameters that affect the performance by using complex and adequate experimental data sets and to reduce the degrees of freedom for fitting. Once the parameters are identified, we can work at the right scale, with the appropriate physics and computational methodology that the macroscopic results point out in order to complete the picture and resolve the unknowns.

The aim of this study is to extend previous results from a 1D model in which we identified the active area and permeability as the most distinguishing elements between Pt-based and NPM-based catalyst layers. On one hand, the active area enters into macroscopic models convolutes with the intrinsic activity of the catalyst site. The latter is related to the specific catalytic properties of the active site; understanding it requires a deep study of the catalyst nature, which is not well known due to the pyrolysis of NPM materials. In this case, a coupled modeling approach using DFT (Density Functional Theory) and MD (Molecular Dynamics) techniques could give us insight in the active site behavior. On the other hand, transport properties and differences can be study by using CFD (Computational Fluid Dynamics) tools to compare fluid transport and local distributions within the catalyst layers with parameters obtained via DFT-MD simulations. This contribution will summarize the previous modeling of transport and describe calculations to probe the final configuration after high temperature treatment of porphyrin and iron porphyrin carbon structures. We aspire to describe interactions of the porphyrin with the carbon substrate during the heating process and give insight into the active site final configuration from a theoretical perspective by combining MD with DFT calculations.

Acknowledgement

We gratefully acknowledge the support of the NSF EPSCoR program and Colciencias for support of this work.

References

1. F. Jaouen, E. Proietti, M. Lefevre, R. Chenitz, J.-P. Dodelet, G. Wu, H. T. Chung, C. M. Johnston and P. Zelenay, Energy & Environmental Science, 4, 114 (2011).

2. G. A. Goenaga, J. Brooksbank, C. Dabke, A. Belapure, A. Papandrew, S. Foister and T. A. Zawodzinski, ECS Transactions, 50, 1749 (2013).

3. M. Ferrandon, X. Wang, A. J. Kropf, D. J. Myers, G. Wu, C. M. Johnston and P. Zelenay, Electrochimica Acta, 110, 282 (2013).

4. F. Jaouen, V. Goellner, M. Lefèvre, J. Herranz, E. Proietti and J. P. Dodelet, Electrochimica Acta, 87, 619 (2013).

5. A. Z. Weber and J. Newman, Chemical Reviews, 104, 4679 (2004).

6. T. A. Zawodzinski Jr, T. E. Springer, F. Uribe and S. Gottesfeld, Solid State Ionics, 60, 199 (1993).

7. A. Ohma, T. Mashio, K. Sato, H. Iden, Y. Ono, K. Sakai, K. Akizuki, S. Takaichi and K. Shinohara, Electrochimica Acta, 56, 10832 (2011).

8. T. S. Olson, K. Chapman and P. Atanassov, Journal of Power Sources, 183, 557 (2008).

9. C. Song and J. Zhang, in PEM Fuel Cell Electrocatalysts and Catalyst Layers, J. Zhang Editor, p. 89, Springer London (2008).

10. F. Charreteur, F. Jaouen, S. Ruggeri and J.-P. Dodelet, Electrochimica Acta, 53, 2925 (2008).

Figure 1

Depending on the operating conditions, the overall performance and durability of polymer electrolyte membrane fuel cells (PEMFC) can be strongly influenced by liquid water saturation of pores within the gas diffusion layer (GDL). Water agglomerations in the GDL may occur due to material morphology, surface characteristics and local boundary conditions. In the catalyst layer (CL) and microporous layer (MPL) these parameters are considerably different than those of the fibre substrate. The MPL and CL feature by orders of magnitude smaller pores than the substrate and the boundary conditions, e.g. temperature and relative humidity, are different than those in the substrate near the bipolar plate. Thus, different physical behavior within each layer can be observed, e.g. local liquid sorption. Therefore, finding an optimal strategy regarding water management requires knowledge on virtually all the factors involved.

To approach this target, we are using and further developing a 3D Monte Carlo (MC) model which simulates water distribution within the GDL, employing the real GDL structure and operating conditions, exploited respectively from tomography and computational fluid dynamics (CFD) studies¹. The 3D images are obtained from an assembled fuel cell and therefore, the results include compression and inhomogeneity information of the GDL. Furthermore, recent progress of the model has enabled simulations on the length scales relevant for the MPL and CL in addition to the fibre substrate.

GDL substrates usually have an average pore size in the micrometer scale, whereas this value for the MPLs and CLs may go down to the nanometer scale. Therefore, a complete simulation of the water distribution in GDL includes working on at least two scales. Although obtaining accurate results on all three sub-layers is equally important, MPL and CL are the more challenging ones from the point of view of both structure and surface properties acquisition. Furthermore, simulating relevant domains within the material requires expensive computational effort. In this study, we report first results on MC simulations of the water distribution within the MPL and CL pores for different operating conditions.

The 3D reconstruction of the material is obtained from the focused ion beam - scanning electron microscope (FIB-SEM) tomography². By using atomic layer deposition (ALD) of ZnO, a good material contrast between pores and solid particles is achieved and therefore a reliable binarization can be obtained³. The voxel size is set to 45 nm in order to achieve both, an accurate computation with respect to the applied equations and an appropriately sized simulation domain.

Our MC model provides the water distribution within the actual MPL and CL structures based on the local boundary conditions. The results show how the water distribution depends on parameters such as contact angle, porosity, local temperature and relative humidity. An example of simulation results is illustrated in the Figure 1.

References

1. K. Seidenberger et al., J. Power Sources, 239, 628–641 (2013)

2. S. Thiele et al., J. Power Sources, 228, 185–192 (2013)

3. S. Vierrath et al., J. Power Sources, 285, 413–417 (2015)

Figures

Figure 1

The work elucidates the effect of micro-morphology parameter. i.e. catalyst distribution, on macroscopic mass transport properties of a catalyst electrode of a PEM fuel cell. Indeed, the micro-morphology parameters have significant effect on transport properties, such as permeability, tortuosity. Pore-scale analysis of fluid flow has been done using Lattice Boltzmann method. Besides other morphological parameters, like porosity, clusterization, geometry those are common for a porous medium, the mass transport properties of a porous electrode of a fuel cell is governed by catalyst particle deposition distribution. This fact enables to control flow field inside of the most vulnerable membrane electrode assembly compartment. Redistribution of catalyst might be required to design novel MEA more robust prone to degradation processes, especially for transport application. In particular, the influence of a catalyst loading on permeability has been studied. It is explored that increasing the catalyst loading definitely improves transport parameter, but this improvement has maximum effect up to certain (ω_Pt≈0.2) value of catalyst loading. Moreover, how permeability changes depending on a distribution of the same amount catalyst towards membrane has been considered. Pore-scale simulation of different catalyst layer configuration with regards to catalyst deposition shows that redistribution of catalyst in exponential decaying towards membrane allows three times improve permeability of catalyst electrode compared to conventional/homogeneous distribution.

Over the last decade, research team from Daihatsu Motor Co. has introduced the concept of anion-exchange membrane fuel cell with liquid fuels for automotive applications. Switching from more commonly used proton exchange membranes (PEM) to alkaline, anion-exchange membranes (AEM) has several kinetic and materials benefits and opens the possibility of developing Fuel Cell Vehicle (FCV) entirely based on Platinum Group Metal-free (PGM-free) technology.¹ This has been made possible by the adoption of hydrazine hydrate as a liquid fuel, which affords technology deployment while using the existing fuel distribution and dispensing technology. This concept pioneered the use of AEM Fuel Cells for automotive applications worldwide and created the state of the art AEM Fuel Cell stack technology².

UNM team has been a long-term participant in the multi-institutional program led by Daihatsu Motor Co., focusing on the development of Oxygen Reduction Reaction (ORR) catalysts and selective hydrazine electro-oxidation catalysts used as cathode and anode material in AEM membrane electrode assembly (MEA)³. The theoretical electromotive force of such direct hydrazine fuel cell is 1.56V and hydrazine hydrate as a fuel can be oxidized by number of catalysts using exclusively Earth-abundant metals, such as NiZn unsupported catalysts developed at UNM⁴. This line of research has been continued in NiLa⁵ and supported Ni-allowed catalysts⁶ aiming not only high activity, but also extreme selectivity of hydrazine electro-oxidation to N₂ and H₂O only. Elucidating the mechanism of hydrazine oxidation included EXAFS/XANES study⁷ and let to formulating of the hypothesis of the reaction mechanism⁸ that involves surface hydroxyl groups on Ni as critical participants in the hydrazine dehydrogenation reaction step (see Fig. 1). The role of the nickel, secondary metal phase and the carbon support in the activity and durability of Ni-allow catalysts will be discussed.

ORR catalysts developed under this program included initially Co-polypyrrole materials system⁹. The mechanism of oxygen reduction and the structure-to-property relationships have been described in a recent EXAFS/XANES study¹⁰. This paper will present a family of Fe-Nitrogen-Carbon ORR catalysts that were used in the successful stack development and Daihatsu FCV prototype tests. This new generation Fe-containing PGM-free catalysts are synthesized by UNM Sacrificial Support Method (SSM) a type of for templated synthesis of hierarchically structured electrocatalysts materials.^11-13 In this method the catalysts precursors are being absorbed on, impregnated within or mechanically mixed with the support (usually mono-dispersed or meso-structured structured silica), thermally processed (pyrolyzed) and then the silica support is removed by etching (in KOH or HF) to live the open frame structure of a "self-supported" material that consists of the catalysts only. Such hierarchical structures are advantageous in enhancement of the fuel cell performance since they correspond to the different levels of transport in the corrugated electrode matrixes. A wide variety of materials can be made by these methods in which not only the composition but also the microstructure can be varied. It is the combination of these attributes - control over microstructure at a number of different length scales and composition, simultaneously - that is extremely important to the performance of the electrocatalyst materials in a fuel cells.

The high activity of these Fe-N-C cathode catalysts was confirmed in both RDE and MEA tests. As synthesized materials were extensively studied by XPS, SEM, TEM, BET and other methods, in order to elucidate the structure-to-properties correlations. This paper describes a new class of templated, self-supported PGM-free catalysts derived by sacrificial support method (SSM) and their activity in free alkaline electrolyte (KOH) and in anion-exchange MEA.

• P. Atanassov and A. Serov, Japanese Society of Automotive Engineers, 67 (2013) 68-71

• J. Varcoe, P. Atanassov, et al., Energy & Environmental Science, 7 (2014) 3135-3191

• A. Serov, M. Padilla, et al., AngewandteChemie Intern. Ed., 53 (2014) 10336 –10339

• U. Martinez, K. Asazawa, et al., Physical Chemistry & Chemical Physics, 14 (2012) 5512-5517

• T. Sakamoto, K. Asazawa, et al., J. Power Sources, 234 (2013) 252-259

• T. Sakamoto, K. Asazawa, et al., J. Power Sources, 247 (2014) 605-611

• T. Sakamoto, D. Matsumura, et al., Electrochimica Acta, 163 (2015) 116-122

• T. Sakamoto, H. Kishi, et al., Electrochimica Acta, (2016) in press

• T.S. Olson, S. Pylypenko, et al., J. Phys. Chem. C., 114 (2010) 5049–5059

• K. Asazawa, H. Kishi, et al., Journal of Physical Chemistry - C, 118 (2014) 25480–25486

• S. Pylypenko et al., Electrochimica Acta, 53 (2008) 7875-7883

• A. Serov et al., Electrochem. Comm., 22 (2012) 53-56

• Serov et al., Advanced Energy Materials, 4 (2014) DOI: 10.1002/aenm.201301735

• Serov, et al., Nano Energy, 16 (2015) 293-300

Figure 1

Among the existing fuel-cell types, alkaline-exchange-membrane fuel cells (AEMFC) have intriguing features as compared to proton-exchange-membrane fuel cells (PEMFC). Their major advantage is the possibility of using non-noble catalysts due to faster oxygen-reduction reaction (ORR) kinetics in alkaline than in acidic media. However, water management is a more serious concern in AEMFCs because OH- conductivity is more highly dependent on water content and the ORR consumes water. Compared to PEMFCs, the lower performance of AEMFCs is mostly caused by extremely nonuniform distribution of water in the ionomer phase between the anode and cathode as well as the increased overpotential for the hydrogen oxidation reaction. In this presentation, we will discuss the performance-limiting mechanisms specific to different operating conditions (e.g. varying inlet relative humidity (RH)) based on a cell-level mathematical model. For example, anode flooding can be a critical issue at 100% RH, whereas at lower RH high ionic transport resistance in the cathode dominates due to dehydrated ionomer phase at high current, where the impact of water-consuming ORR kinetic polarization is also critical. Low AEMFC performance at high current is not simply due to mass transport issues with vapor/membrane water, but a consequence of poor water distribution leading to sluggish OH- conduction and ORR kinetics. A sensitivity analysis of design parameters including the humidifying condition and membrane property is performed to identify the most significant factors controlling performance. Overall, water management and ionic/mass transport characteristics of an AEMFC assembly are discussed in detailed. The developed model will be used to examine and elucidate performance bottlenecks and enable strategies to overcome them, significantly increasing the possibility of AEMFC commercialization.

Acknowledgements

This work was funded by the Fuel Cell Technologies Office, Office of Energy Efficiency and Renewable Energy, of the U. S. Department of Energy under contract number DE-AC02-05CH11231, program manager David Peterson.

In searching for an exemplary carbon neutral fuel, dimethyl ether (DME) may be the most appealing. This simplest of the ethers can be readily produced from renewably sourced hydrogen and CO₂, making it essentially a hydrogen carrier (1). Both nontoxic and easy to liquefy under moderate pressure, DME closely matches diesel and has been run in trucks (2). The ubiquitous propane grill tanks can store DME without modification. Recently, Los Alamos National Laboratory (LANL) demonstrated the potential for direct oxidation of DME in a fuel cell.The breakthrough of our work is LANL's highly active catalyst for direct oxidation of DME that in the early phase of development shows similar performance to direct methanol when using typical low-temperature membranes. However, the output is not sufficient to approach commercial acceptance targets for higher power applications or precious metal cost. High-temperature membrane electrode assemblies (HT-MEAs), based on phosphoric-acid-imbibed membranes, operate at 160 °C to 180 °C without additional water and are highly tolerant to carbon monoxide – an intermediate of DME oxidation. This work exploits a novel ternary LANL anode catalyst with the features of high-temperature operation to produce high-power, low-cost direct DME MEAs.

In support of direct DME oxidation, scientists at LANL have recently shown a direct liquid fuel cell operating with DME as part of a program to develop technology for direct methanol (MeOH), ethanol (EtOH), and DME fuel cells (3). In this work, it has been shown that DME can be directly oxidized via a PtRuPd ternary catalyst that not only activates the ether bonds but facilitates the removal of CO as a reaction by product (Figure 1). Even at this early stage, the team led by LANL has shown results that approach a highly developed DMFC system with regards to performance and precious metal loading (3). One highly relevant aspect of this work is the sharp temperature dependency of the current-voltage behavior, with an increase in the cell temperature from 80 °C to 90 °C leading to an increase in the current density at 0.5 V by ca. 50%, from 140 mA cm^-2 to 220 mA cm^-2

High-temperature PEM MEAs developed by BASF for the polybenzylimide (PBI) system or pyridine polymer based TPS^®system of Advent Technologies, Inc. have demonstrated utility and robustness when operating with highly impure reformates of around 1-3% CO (4). Similarly, cathode sensitivity to air contaminants is also ameliorated. The electrolyte in both systems is phosphoric acid imbibed into a stable and inert polymer matrix. Phosphoric acid is notable in that it does not need water to conduct protons, and can do so over a wide temperature range from ~120 °C to more than 200 °C. Advent is a licensee of BASF's MEA and gas-diffusion-electrode technology.

This presentation will summarize our on-going efforts at facilitating direct oxidation of DME by using high temperature PEM MEAs with both PBI and TPS combined with the LANL catalyst operating at 160 ^oC to 180 ^oC.

• Olah, G.A., Goeppert, A. and Prakash, G.K. Surya in J. Org. Chem., 2009, 74, 487-498.

• Hutchinson, H., 2013 in ASME web article www.asme.org/engineering-topics/articles/transportation/diesel-alternative-hits-the-road.

• Advanced Materials and Concepts for Portable Power Fuel Cells, 2014 Hydrogen and Fuel Cells Program AMR presentation.

• Technical Brochure, 2014 Advent Technologies, Inc. at www.Advent-Energy.com.

Figure 1

An anion exchange membrane fuel cell (AEMFC) using hydrazine hydrate liquid fuel has been investigated¹ and the concept vehicle was exhibited at the 42^nd and 43^rdTokyo Motor Show.

For the improvement of the AEMFCs, technological breakthrough is necessary for anion exchange membranes. Especially, it is important to overcome the counteracting relationship of high ion conductance and low fuel permeability.

In addition, technology to optimize triple-phase interface formed by catalyst and ionomer is important to improve the performance.

Through many efforts to control the triple-phase interface, state-of-the-art proton exchange membrane fuel cell electrode that consists of platinum group electrocatalysts and perfluorinated ionomer has been investigated. However, the catalyst utilization is still insufficient. In addition, it is more challenging to control the triple phase interface for PGM-free electrocatalysts and anion exchange ionomers.

Our approach to develop the advanced AEMFC is using anion conductive quaternized aromatic multiblock copolymers, poly(arylene ether)s (QPEs)²as shown in Fig.1 and controling the interface and PGM-free electrocatalysts both for anode and cathode.

In this preliminary study, we focus on ionomer layer design for anode and cathode evaluated by ionomer thickness covered on the electrocatalysts. (TEM image shown in Fig.2)

MEA consisting of the QPE membranes and ionomers was fabricated with PGM-free electrocatalysts and subjected to performance evaluation. The MEA performance of direct hydrazine fuel cells will be shown and discussed in the meeting.

Figure captions

Fig. 1. Representative structure of QPEs.³

Fig. 2. TEM Image of ionomer layer (red line) on the cathode catalysts.

References

[1] K. Asazawa, K. Yamada, H. Tanaka, A. Oka, M. Taniguchi, and T. Kobayashi, Angew. Chem. Int. Ed., 46, 8024, 2007.

[2] M. Tanaka, K. Fukasawa, E. Nishino, S. Yamaguchi, K. Yamada, H. Tanaka, B. Bae, K. Miyatake, and M. Watanabe, J. Am. Chem. Soc., 133, 10646–10654, 2011

[3] R. Akiyama, N. Yokota, E. Nishino, K. Asazawa, and K. Miyatake, submitted.

Acknowledgements

This work was supported by CREST, JST.

Figure 1

Water management in anion exchange membrane fuel cells (AEMFCs) is more complex than proton exchange membrane fuel cells (PEMFCs). In the PEM system, water is generated at the cathode and otherwise only serves as the membrane proton transport medium. In the hydroxide AEMFC, water is generated at the anode, consumed at the cathode, and is greatly needed for transport of the larger hydroxide ion. The challenge this introduces is the need to provide adequate water to maintain the membrane humidity without flooding the catalyst or gas diffusion layers.¹ One method being used to achieve the necessary high membrane water content is utilizing hydrophilic gas diffusion layers without microporous layers.² Others have used gas diffusion layers with a hydrophobic coated microporous layer while raising the temperature of the humidifiers well above the cell temperature.³ Gas stream dew points above 100% relative humidity and back pressure are common approaches in increasing the membrane water content. However there is instability in the cell performance and conditions that generate high performance levels are not easily repeatable. The problem is the line between proper membrane hydration and flooded catalysts layers is very thin, and non-existent in some cases.

In this study, the influence of the membrane, ionomer and gas diffusion layer as well as the flow rate and dew points of the anode and cathode gases on AEMFC performance were explored. Tokuyama A201 membranes and AS-4 ionomer were investigated alongside quaternary-ammonium-functionalized radiation-grafted ETFE alkaline anion-exchange membranes and ionomers.⁴ Using a hydrophilic gas diffusion layer without a microporous layer increased membrane hydration, but as expected increased the potential for flooding. Manipulating the dew points led to the counter-intuitive discovery that the cell performs better with the humidity higher at the anode than the cathode, despite water generation and electro-osmotic drag towards that electrode. In fact, removing too much water from the anode caused instability in the cell, while increasing the water at the anode decreased the membrane resistivity. Back diffusion likely plays an important role in membrane hydration and hydroxide transport through the membrane. A very high flow rate (1.0 L/min) also increased cell performance, despite being several orders of magnitude above the stoichiometric need.

Combining all areas of improvement resulted in a very high performing AEMFC with a maximum current density of 2.2 A/cm² (at 0.1 V) and max power density of 670 mW/cm² (880 mW/cm² iR-corrected) with a membrane resistivity of 75 mOhms*cm² (Figure 1a). Only a minor drop in the current was observed using air at the cathode with the same 1.0 L/min flow rate as oxygen, giving max current density of 1.7 A/cm² and a max power density of 580 mW/cm² with a resistivity of 74 mOhms*cm² (Figure 1b). This near identical behavior confirms that the amount of reactant present supplied by the higher flow rate is not necessary, but the volumetric flow rate is needed for water management. It is likely that the pressure drop along the single pass cell hardware allows the gas to "jump the bar" only at very high flowrates, which results in better water removal and limits cell flooding in the cell.

References:

1. T. D. Myles, A. M. Kiss, K. N. Grew, A. A. Peracchio, G. J. Nelson and W. K. Chiu, J.Electrochem.Soc., 158, 7 (2011).

2. R. B. Kaspar, M. P. Letterio, J. A. Wittkopf, K. Gong, S. Gu and Y. Yan, J.Electrochem.Soc., 162, 6 (2015).

3. M. Mamlouk, J. Horsfall, C. Williams and K. Scott, Int J Hydrogen Energy., 37, 16 (2012).

4. S. D. Poynton, R. C. Slade, T. J. Omasta, W. E. Mustain, R. Escudero-Cid, P. Ocón and J. R. Varcoe, Journal of Materials Chemistry A., 2, 14 (2014).

Figure 1

Catalyst for polymer electrolyte fuel cell (PEFC) remains a central issue toward a large-scale penetration of technologies especially for automobile applications. Along with the development of synthesis as well as analytical methodologies, a few nano-meter scale nano-particle with manipulated structure or compositions, such as core-shell or skin-layer alloy catalysts, have been intensively developed (1). Because of the complexity of catalytic reactions as well as the porous microstructure of catalyst layer with multi-components, theoretical methods are also intensively applied (2). While majority of theoretical papers adopts slab models under periodic boundary condition for simpler computation, properties of nano-particles such as strain, stability, or electronic structures should be discussed by using a cluster model. In discussing electronic structure of nano-particles, it is reported that small cluster models show discrete electronic structure, which may lead to a molecular properties rather than metallic properties (3). Thus, it is necessary to adopt models with a realistic diameter. While we can find a few papers adopting models over 2 nm (4), we often find ca. 3 nm particles as PEFC electrocatalyst (5). In our preceding study, we have discussed the stability and electronic structures of Pd-Pt alloys with different atomic configurations by using ca. 3 nm model consisting of 711 atoms (6). In this study, we extend our efforts to Pt-based electrocatalyst with skin-layer configuration.

For density functional theory calculations, we used Vienna ab initio simulation package (VASP) (7) with the projector augmented wave method (8). The Perdew-Burke-Ernzerhof under generalized gradient approximations (9) was used as the exchange and correlation functional. The cutoff energy was set to be 400 eV. Pt-Co alloy nano-particle models with solid solution and Pt-skin configurations are prepared. To discuss the stability of nano-particles with different configuration, excess energy (10) are evaluated.

By comparing the excess energies of the Pt-Co models with solid solution and Pt-skin configurations, we found that the Pt-skin configuration is more stable than solid solution configuration at the same Pt-Co composition. To have insight on the activity toward oxygen reduction reaction, we have analyzed interatomic distances to find shorter distances for Pt-skin configuration. Details will be discussed in the presentation.

• For example, J. Wu, H. Yang, Acc. Chem. Res., 46, 1848 (2013); S. Guo, S. Zhang, S. Sun, Angew. Chem. Int. Ed., 52, 8526 (2013); M. Watanabe, D. A. Tryk, M. Wakisaka, H. Yano, H. Uchida, Electrochim. Acta, 84, 187 (2012).

• For example, J. R. De Lile, S. Zhou, Electrochim. Acta, 177, 4 (2015); F. Calle-Vallejo, M. T. M. Koper, Electrochim. Acta, 84, 3 (2012); Z. Shi, J. Zhang, Z.-S. Liu, H. Wang, D. P. Wilkinson, Electrochim. Acta, 51, 1905 (2006); J. Greeley, M. Mavrikakis, Nature Mat., 3, 810 (2004).

• L. Li, A. H. Larsen, N. A. Romero, V. A. Morozov, C. Glinsvad, F. Abild-Pedersen, J. Greeley, K. W. Jacobsen, J. K. Nørskov, J. Phys. Chem. Lett., 4, 222 (2013).

• W. An, P. Liu, ACS Catal., 5, 6328 (2015); S. H. Noh, B. Han, T. Ohsaka, Nano Res., 8, 3394 (2015).

• M. Watanabe, H. Yano, D. A. Tryk, H. Uchida, J. Electrochem. Soc., 163, F455 (2016); M. Chiwata, H. Yano, S. Ogawa, M. Watanabe, A. Iiyama, H. Uchida, Electrochem., 84, 133 (2016).

• T. Ishimoto, M. Koyama, J. Phys. Chem. Lett., 7, 736 (2016).

• G. Kresse, J. Furthmüller, Phys. Rev. B, 54, 11169 (1996); G. Kresse, J. Hafner, Phys. Rev. B, 47, 558 (1993).

• G. Kresse, J. Joubert, Phys. Rev. B, 59, 1758 (1999); P. E. Blöchl, Phys. Rev. B, 50, 17953 (1994).

• J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett., 77, 3865 (1996).

• R. Ferrando, A. Fortunelli, G. Rossi, Phys. Rev. B, 72, 085449 (2005).

Binary and ternary alloys often show improved activity, selectivity or durability as electrocatalysts compared to elemental catalysts. For practical applications, nanoparticle catalysts are required to maximize utilization of precious metals such as platinum. Theoretical simulations based on density functional theory (DFT) have helped rationalize experimental observations and suggested new design directions for experimentalists.¹ Commonly, theoretical studies are, however, limited to small clusters or extended surfaces under periodic boundary conditions. Although DFT simulations of nanoparticles with hundreds of atoms have become possible in recent years,² prediction of properties such as morphology, atomic distribution and catalytic activity of nanoalloys remains a challenge.

Here, we report recent theoretical work on the global optimization of atomic distribution within nanoalloy catalysts using genetic algorithms and we predict their activity using density functional theory. We investigate the activity of Pt-Pd-Au binary and ternary alloys for the oxygen reduction reaction (ORR) and the stability of these ternary alloys under ORR conditions. Our findings suggest subsurface Pd, in binary Pt-Pd films, results in improved activity over pure Pt, while addition of small amounts of Au will hinder Pt dissolution and might also enhance activity. We study the ORR activity of 2 nm ternary Pt-Au-M Mackay icosahedral core-shell nano-particles in order to identify core elements, which increase activity and particle stability through strain and ligand effects.

Acknowledgements

Financial support from the European Commission under the FP7 Fuel Cells and Hydrogen Joint Technology Initiative grant agreement FP7-2012-JTI-FCH-325327 for the SMARTcat project is gratefully acknowledged.

References

1. I. E. L. Stephens, A. S. Bondarenko, U. Grønbjerg, J. Rossmeisl and I. Chorkendorff, Energy Environ. Sci., 2012, 5, 6744.

2. L. Li, A. H. Larsen, N. A. Romero, V. A. Morozov, C. Glinsvad, F. Abild-Pedersen, J. Greeley, K. W. Jacobsen and J. K. Nørskov, J. Phys. Chem. Lett., 2013, 4, 222–226.

Fuel cells are expected to be a key next-generation energy source used for vehicles and homes, offering high energy conversion efficiency and minimal pollutant emissions. The high efficiency arises from the fact that fuel cells convert chemical energy directly into electrical energy without the Carnot limitation of thermal engines. However, due to the extreme operating environment, Pt was almost exclusively the only practical catalyst for molecular oxygen electroreduction in PEM fuel cells. Recently, considerable efforts have been made to investigate synergy effects of platinum alloyed with base metals to improve the sluggish kinetics of oxygen reduction reaction of Pt catalyst. The more active Pt-alloy catalysts, in return, may arise the durability issue under the extreme environment of PEM fuel cell operation.

In this presentation, results of a more active yet durable PtTi alloy catalyst will be discussed for oxygen reduction reaction in acidic media, including (i) combinatorial high-throughput discovery of PtTi alloy composition; (ii) synthesis and characterization of nano-scale PtTi particles with controllable size, phase, and individual particle composition; and (iii) electrochemical STM and ICP characterization of coarsening and corrosion of PtTi alloy catalyst in comparison with Pt and PtCo catalysts.

A novel combinatorial high-throughput electrocatalyst discovery workflow and associated tools were developed [1-4], where thin films of alloys can be made through a Multi-sources Physical Vapor Deposition system [5], and the resulted materials can be electrochemically screened by a Multi-channel Rotating Disk Electrode system [6]. PtTi binary system has been screened using this methodology for activity-stability-composition relationship and the best alloy composition was identified.

In addition to the discovery of alloy compositions, engineering the PtTi alloy particles in nanoscale has been a challenge. Several synthesis technologies were studied and developed to achieve capabilities of controlling particle size and particle microcomposition [7]. The results show that by careful engineering the particle size and microcomposition in nanoscale, it is able to achieve the superior electrocatalytic activity predicted by the combinatorial high-throughput discovery.

Electrochemical STM technique, in combination with electrochemical potentiostating, potential cycling, and ICP analysis, was used to monitor the coarsening and base metal corrosion under dynamic fuel cell operation conditions and compared with phenomena observed from Pt and PtCo catalysts [8-10]. The results show that PtTi not only is more active than Pt and much stable than PtCo, but also can mitigate the catalyst coarsening.

Reference

[1] T. He and E. Kreidler, Phys. Chem. Chem. Phys. 10, 3731 (2008). "Combinatorial screening of PtTiMe ternary alloys for oxygen electroreduction".

[2] E. Kreidler and T. He, in Catalysts for Oxygen Electroreduction – Recent Developments and New Directions (Ed. T. He, Transworld Research Network) Chapter 4, p.63 (2009). "Oxygen reduction on Pt alloys: high throughput combinatorial screening of Pt binary alloys".

[3] T. He, E. Kreidler, L. Xiong and E. Ding, J. Power Sources 165, 87 (2007). "Combinatorial screening and nano-synthesis of platinum binary alloys for oxygen electroreduction".

[4] T. He, E. Kreidler, L. Xiong, J. Luo and C.J. Zhong, J. Electrochem. Soc. 153, A1637 (2006). "Alloy electrocatalysts: combinatorial discovery and nano-synthesis".

[5] T. He, E.R. Kreidler and T. Nomura, US 8,944,002 (2015). "High throughput physical vapor deposition system for material combinatorial studies".

[6] T. He, US 7,077,946 (2006). "High throughput multi-channel rotating disk or ring-disk electrode assembly and method".

[7] E. Ding, K.L. More and T. He, J. Power Sources 175, 794 (2008). "Preparation and characterization of carbon-supported PtTi alloy electrocatalysts".

[8] L. Tang, B. Han, K. Persson, C. Friesen, T. He, K. Sieradzki and G. Ceder, J. Am. Chem. Soc. 132, 596 (2010). "Electrochemical stability of nanometer-scale Pt particles in acidic environments".

[9] Q. Xu, E. Kreidler, D.O. Wipf and T. He, J. Electrochem. Soc. 155, B228 (2008). "An in-situ electrochemical STM study of potential-induced coarsening and corrosion of platinum nano crystals".

[10] Q. Xu, T. He and D. Wipf, Langmuir 32, 9098 (2007). "An in-situ electrochemical STM study of the coarsening of platinum islands at double-layer potentials".

In current polymer electrolyte fuel cells (PEMFC), high amounts of platinum catalysts is required due to the sluggish kinetics of the oxygen reduction reaction (ORR). As such it is important to find ways to increase the activity of the ORR catalyst. Previous studies have shown that alloying platinum with yttrium results in a catalyst which is more than six times as active as sole Pt in half cell rotating disk electrode measurements [1].

The model electrodes in this study are prepared by sputtering a well-defined layer of alloy catalyst onto a gas diffusion layer (GDL). In this manner the composition and thickness of the film can be controlled. These electrodes are studied with single cell PEMFC measurements to characterize the materials, similar to previous work done with these types of electrodes [2]. Using these methods the electrochemical surface area (ECSA), ORR activity, and stability are evaluated. The effects of different MEA preparations will also be investigated.

Scanning electron microscopy (SEM) with energy dispersive X-ray (EDX) analysis as well and X-ray photoelectron spectroscopy (XPS) are used to characterize the structure of the alloy catalyst as well as the bulk and surface compositions, i.e. platinum to yttrium ratio. This characterization is performed on both as-prepared samples and electrodes after being used in the PEMFC.

The main focus is to determine the role of yttrium in platinum thin-film platinum electrodes, and if the excellent activity in half cell experiments will also be seen in real PEMFC measurements.

References

[1] J. Greeley, I. E. L. Stephens, A. S. Bondarenko, T. P. Johansson, H. A. Hansen, T. F. Jaramillo, J. Rossmeisl, I. Chorkendorff and J. K. Nørskov , Nature Chemistry, 2009, 1, 552-556

[2] M. Wesselmark, B.Wickman,C.Lagergren,G.Lindbergh, Electrochimica Acta, 2013, 111, 152-159

The high platinum loadings required to compensate for the slow kinetics of the oxygen reduction reaction (ORR) impede the widespread uptake of polymer electrolyte membrane fuel cells. In order to improve the ORR kinetics and reduce the Pt loading, we can tailor the electronic properties of the Pt surface atoms by means of alloying Pt with other metals. Researchers have intensively studied alloys of Pt with late transition metals such as Ni and Co during the last decades. However, these compounds typically degrade under fuel cell reaction conditions, due to dealloying. In contrast, alloys of Pt and lanthanides present very negative enthalpy of formation [1,2], which should increase their resistance to degradation.

Herein we present eight novel Pt-lanthanide and Pt-alkaline earth ORR electrocatalysts: Pt₅La, Pt₅Ce, Pt₅Sm, Pt₅Gd, Pt₅Tb, Pt₅Dy, Pt₅Tm and Pt₅Ca [3]. All the materials are highly active, presenting a 3 to 6-fold activity enhancement over Pt. Pt₅Tb is the most active polycrystalline Pt-based catalyst reported in the literature. A Pt overlayer with a thickness of few Pt layers is formed onto the bulk alloys by acid leaching [1-3]. Notably, the experimental ORR activity as a function of the bulk lattice parameter and the Pt-Pt distance follows a "volcano" relation [3], with Pt₅Tb presenting the highest initial activity while Pt₅Gd is the most active after 10 000 cycles stability test between 0.6 and 1.0 V versus the reversible hydrogen electrode. We use the lanthanide contraction to control strain effects and tune the electrocatalytic activity, stability and reactivity of Pt [3].

References

[1] M. Escudero-Escribano, A. Verdaguer-Casadevall, P. Malacrida, U. Grønbjerg, B.P. Knudsen, A.K. Jepsen, J. Rossmeisl, I.E.L. Stephens, I. Chorkendorff, J. Am. Chem. Soc. 2012, 130, 16476.

[2] P. Malacrida, M. Escudero-Escribano, A. Verdaguer-Casadevall, I.E.L. Stephens, I. Chorkendorff, J. Mater. Chem. A2014, 2, 4234.

[3] M. Escudero-Escribano, P. Malacrida, M.H. Hansen, U.G. Vej-Hansen, A. Velázquez-Palenzuela, V. Tripkovic, J. Schiøtz, J. Rossmeisl, I.E.L. Stephens, I. Chorkendorff, Science2016, 352, 73.

The conversion of electric energy into pressurized hydrogen by means of water electrolysis is currently discussed as promising option to handle intermittent electricity supply that occurs with the increasing use of renewable energy sources. Subsequently, this hydrogen could be (1) either reconverted into electricity, or (2) converted into other energetic molecules, such as methane (power to gas) or liquid fuels, or (3) utilized as a feed stock for the chemical industry.

Besides conventional alkaline water electrolysis, a water electrolysis technology based on a Polymer Electrolyte Membrane (PEM) is discussed since several years, which is the focus of the present contribution. The particular advantages of PEM water electrolysis are higher attainable current densities, higher voltage efficiency, and higher load flexibility. Current disadvantages include far lower technical maturity, shorter life time, and higher specific costs compared to alkaline water electrolysis.

The formulation of appropriate physico-chemical models for solid polymer electrolysis can help to contribute to the solution of the above mentioned problems. In the present contribution state of the art of modelling of solid polymer electrolyzers is summarized and discussed.

Proton exchange membrane (PEM) water electrolysis has become increasingly attractive due to the penetration of renewable energy (e.g, solar and wind). Hydrogen production from PEM water electrolysis is advantageous over other technologies due to its simple and clean nature. Membrane and electrode assemblies (MEAs) of PEM electrolyzers typically use iridium (Ir) as an anode catalyst and Pt as a cathode catalyst.

Performance and durability of the MEAs play an essential role for the cost and viable commercialization of PEM water electrolysis. However, unlike the well-established MEA benchmarks of PEM fuel cells, the performance and durability of PEM electrolyzer MEAs have not been thoroughly studied. The objective of this work is to establish benchmark MEA performance and durability for PEM water electrolysis. For this purpose, a series of oxygen evolution reaction (OER) catalysts, which includes commercial Ir black and various Ir nanostructures, has been evaluated under test protocols established at Giner Inc. These approaches include high-voltage hold (>1.8 V), accelerated stress test (e.g., voltage cycling from 1.4 to 2.0 V), and constant low-current operations. The polarization curves of the MEAs will be obtained after each test. The morphology and structure of MEAs after durability tests will be characterized to correlate to their performance and durability.

The established performance and durability may provide metrics and guidance to the community of PEM water electrolysis.

Acknowledgement: The financial support is from the Department of Energy under the Contract Grant DE-SC0007471.

Water electrolyzers are predominantly operated at elevated pressures. This is especially favorable in the context of the subsequent utilization of the produced hydrogen e.g. due to the more efficient, electrochemical and isothermal hydrogen compression or less afford for hydrogen drying. However, higher pressures lead to higher cross-permeation of the product gases in electrolysis cells.

In the present contribution, hydrogen permeation across a fumea EF-40 membrane is systematically investigated for pressure differences between 5-35 bar and for four different temperatures between 22 °C and 60 °C. The measured permeation fluxes show a quadratic dependency on the pressure difference. A permeation model combining a diffusive and convective transport can describe the experimental data quantitatively. For the investigated membrane, the diffusive as well as the proposed convective permeability coefficient and their temperature dependencies are obtained.

The diffusive permeability coefficient and its temperature dependency agrees very well with published data for Nafion membranes. The slightly lower value for the EF-40 can be explained by the lower water volume fraction. The obtained convective permeability coefficient indicates a high hydraulic permeability, especially in comparison to recently reported values for Nafion membranes. However, the reported values of the hydraulic permeability show high variations of about two orders of magnitude, with strong dependence on investigated materials, that show the importance of this parameter with regard to convective hydrogen permeation.

The present work discusses two potential reasons for the high obtained hydraulic permeability. The first one is related to the membrane properties. In particular, the lower equivalent weight of the fumea EF-40 membrane could cause the higher hydraulic permeability. Second the electrolysis conditions, in particular the applied small current, might increase the hydraulic permeability due to wider membrane water channels and a change to a hydrophilic membrane surface.

Figure: Parameterized equation in comparison to the experimental results.

Figure 1

To store or transport hydrogen (and optionally oxygen) in high pressure tanks or pipelines pressures up to 1000 bar are needed. Pressurized polymer electrolyte electrolyzers are typically operated at around 30 bar to reduce downstream mechanical compression and gas drying stages. Interestingly, often the cell voltage does not show the expected thermodynamic voltage increase [1,2]. Consequently beneficial influences are compensating the compression work, at the expense of higher faradaic losses due to increased gas crossover.

In this work we analyze cell voltages at gas pressures up to 100 bar as function of temperature up to 70 °C and current densities up to 4 A/cm², considering the three main overpotentials (ohmic, kinetic, mass transport) using a zero-order analysis based on the Tafel kinetics model.

As expected, the ohmic overpotential is mainly a function of temperature and almost independent from pressure. Therefore the overpotential gains with increasing pressure, observed for current densities above about 0.8 A/cm², stem from the kinetics and/or the mass transport. Both overpotentials decrease with increasing pressure, but over a wide current density region kinetics contribute about two thirds to the voltage gain. The improved kinetics can be attributed to an increased apparent exchange current density, whereas the increased mass transport might be related to smaller gas volumes or velocities resulting in an improved two-phase flow in the porous structures.

Literature:

[1] S.A. Grigoriev, V.I. Porembsky, V.N. Fateev, Pure hydrogen production by PEM electrolysis for hydrogen energy, Int J of Hydrogen Energy, 31 (2006) 171-175.

[2] P. Millet, R. Ngameni, S.A. Grigoriev, N. Mbemba, F. Brisset, A. Ranjbari, C. Etiévant, PEM water electrolyzers: From electrocatalysis to stack development, Int J of Hydrogen Energy, 35 (2010) 5043-5052.

This conference contribution touches upon electrochemical characterization of operating polymer electrolyte membrane electrolysis cells (PEMECs) by the application of electrochemical impedance spectroscopy (EIS). Analysis of differences in impedance spectra (ADIS) (Jensen et al., 2007) can be applied to gain insight into the relative magnitudes of resistance contributions from the two electrodes, the electrolyte and of mass transfer limitations and can help identifying the time scale of the respective processes. The gained knowledge may facilitate further development of the PEMECs.

A state-of-the-art PEMEC obtained from IRD with a 2.9 cm² active electrode area and loaded with 0.3 mg/cm² IrO₂ on the anode and 0.5 mg/cm² Pt on the cathode was operated at current densities ranging from 0.07 to 1 A/cm² while examined with EIS. The cell temperature at operation was 61 ˚C at all current densities, and the cell was conditioned for 20 minutes at each current step prior to EIS measurements. A Solartron 1260 was used for the data acquisition, the frequency range was 100 kHz - 0.01 Hz, the amplitude was 17.3 A/cm² and 12 points were measured per decade. Nyquist plots of the EIS results are shown in the upper graph in the figure. One arc is evident in the spectrum at 0.07 A/cm², whereas the spectra at 0.35 and 0.69 A/cm² include two arcs, and two overlapping arcs are observed at 1 A/cm². ADIS plots, with the impedance measured at 0.07 A/cm² subtracted, are shown in the lower graph in the figure. Based on the ADIS plots it can be concluded that the impedance is independent of the current density at frequencies above 100 Hz, whereas two processes observed at approximately 0.4 and 10 Hz are current density dependent, the one increasing and the other decreasing in size with increasing current density.

Acknowledgement

The work is part of the research project e-STORE funded by the Innovation Fund Denmark.

Reference

S. H. Jensen, A. Hauch, P. V. Hendriksen, M. Mogensen, N. Bonanos, T. Jacobsen. (2007). A Method to Separate Process Contributions in Impedance Spectra by Variation of Test Conditions. Journal of The Electrochemical Society, 154 (12), B1325-B1330.

Figure

Nyquist plots (upper graph) and ADIS plots (lower graph) of EIS measured on a PEMEC operating at 0.07, 0.35, 0.69 and 1 A/cm². The ADIS plots represent the difference between the differentiated real part of the impedance at each current density and the differentiated real part of the impedance collected at 0.07 A/cm².

Figure 1

Honda launched new fuel cell vehicle "CLARITY FUEL CELL" in March 10, 2016. In this presentation, I explain the fuel cell vehicle (FCV) development status of Honda including the various main technologies of "CLARITY FUEL CELL". Honda does not only develop the FCV but also small hydrogen station and mobile inverter for power output from FCV. Honda has efforts toward the hydrogen society with concept of "Produce" "use" and "Connect". I explain the small hydrogen station named Smart Hydrogen Stration (SHS) and mobile inverter named "Power Exportor 9000". This explanation will be explained technical conceopts and actual demonstration test resylts. In terms of FCV development, I will explain the FCV development history and future technology direction. Of courese, FCV is not expanded without prepareation of hydrogen infrastructure (Hydrogen Refueling Station : HRS). So this presentation will be explained the situration of HRS preparation and recent direction of movement. Especially, I will introduce the most aggressive Japanese situation. And I expain the future strategy of Honda.

Renewable hydrogen is becoming an increasingly important component of the transition away from fossil fuel use and towards reduction in carbon dioxide production. Hydrogen is the intermediary between primary energy sources and end products in many chemical processes such as ammonia generation, refining, and biogas processing, and is currently mainly produced by reforming of natural gas. Hydrogen from electrolysis can both make a strong environmental impact on these industries and also improve utilization of intermittent renewable energy sources such as wind and solar by leveraging otherwise stranded resources. Hydrogen can also enable improved integration of the electrical grid with the transportation sector and industrial processes, by serving as a storage medium until needed (Figure 1). Because of hydrogen's end use flexibility, electrolysis is a key component of Europe's strategy for energy storage and management of these renewable energy sources on the electrical grid. Proton exchange membrane (PEM) electrolysis is especially well suited to energy capture because of the dynamic range and ability to quickly ramp up and down from near zero output to full capacity.

PEM electrolysis was largely developed in the 70's by GE for life support applications where reliability and reproducibility were key. As the technology was translated to commercial markets it started to compete with the incumbent alkaline water electrolysis on cost, safety (no mixing of O₂ and H₂), ease of maintenance (no caustic) and operational flexibility (differential pressure operation, idling). While the resulting commercial success of PEM electrolysis suggests that the current state of the technology is sufficiently mature, the truth is that while PEM electrolyzers have demonstrated > 50,000 hrs lifetimes and > 99% reliability in the field, the technology and manufacturing techniques have not changed significantly from those early days. In traditional industrial applications the customer rarely cares about efficiency as much as reliability. At the same time, the robustness and manufacturability of the technology (cell, stack and system engineering), have allowed the systems to be scaled from laboratory scale (0.01kg/day) to megawatt scale (100kg/day) with no loss in performance or reliability. To date the research community has mostly focused on advancing PEM fuel cell technology, producing staggering PGM reduction, improvement in efficiency and translation to high throughput manufacturing. Thus electrolysis technology development has lagged behind PEM fuel cells, providing significant opportunities for continued improvements. Thinner membranes, reduced PGM usage, and improved manufacturing techniques have all demonstrated viability for electrolysis, and have the potential to make higher cost and performance impact.

As the application areas for PEM electrolysis start to shift from traditional markets to fueling and renewable energy capture, efficiency, capital cost and ethical resource use become significant factors. Thus advancing the state of the art, translating laboratory advances to commercial PEM stacks and then manufacturing them at scale, while maintaining reliability becomes paramount. Questions that arise are: How do you test reliability, without testing for 50,000 hrs? How do you pick research "winners" at the rotating disk electrode (RDE) level and develop them to large scale PEM membrane electrode assemblies (MEAs). Where is the gap in research at the laboratory scale based on commercial failures? This talk will discuss the challenges in continued scale up, translating laboratory scale findings to commercial systems as well as some of recent advancements and impact on cost.

Figure 1

Transition metal oxides (TMOs) are attracting an increasing attention as promising materials for oxygen electrode in polymer electrolyte fuel cells (PEMFCs) and electrolyzers (PEMEC). Despite significant number of publications oxygen electrocatalysis over TMOs is still insufficiently understood. Pending questions relate to the reaction mechanisms, the state of the surface during the oxygen reduction (ORR) and oxygen evolution reaction (OER), structure-activity relationships, and degradation phenomena which are particularly detrimental during the OER. This slows down development of potent and durable materials for the oxygen electrode of PEMFCs and PEMECs.

This presentation will consist of two parts. The first deals with the ORR on noble-metal free single and complex Mn, Co and other TMOs in view of their application for PEMFCs based on anion-exchange membranes. In this work we seek to understand the relations between the structure and the composition of TMOs oxides on the one hand, and the kinetics and the mechanism of the electrocatalytic ORR on the other hand. An experimental rotating ring disc electrode investigation of the ORR is complemented with the rotating disc electrode study of the oxidation/reduction of the stable ORR intermediate, hydrogen peroxide, and combined with microkinetic modeling in order to arrive at a self-consistent model of the ORR on TMOs oxides. For Mn oxides we conclude that the potential of the Mn(III/IV) red-ox transition of surface cations can be used as a descriptor of the catalytic activity of Mn oxides in the ORR. We further note that the reaction mechanism and the selectivity for the 4e ORR depend on the structure and composition of Mn oxides. Finally, we propose a tentative explanation for the discovered relationship between the catalytic activity and the crystalline structure.

The second part of the presentation is related to the OER on mixed Ir_xRu_1-xO₂ oxide in acid medium and on noble-metal free TMOs in alkaline media. We apply synchrotron-radiation-based near-ambient pressure X-ray photoelectron spectroscopy (NAP-XPS) for studying OER on Ir_xRu_1-xO₂ anodes of PEM electrolyzers with the objective to shed light on the long time known but still insufficiently understood stabilization effect of Ir. We conclude that various forms of Ru coexist in the anode during the OER, and their relative contributions strongly depend on the electrode potential and on the presence of Ir. Based on the in situ spectroscopic data we propose a tentative mechanism of the OER on mixed Ir(Ru) anodes and the stabilization effect of Ir. Finally, we will discuss stability of various oxide materials in alkaline and in acid media during the OER.

Acknowledgements The author is indebted to Anna S. Ryabova, Ivan S. Filimonenkov, Galina A. Tsirlina, Sergey Y. Istomin, Evgeny V. Antipov, Filipp S. Napolskiy, Artem M. Abakumov, Tiphaine Poux, Antoine Bonnefont, Gwenaelle Kerangueven, Spyridon Zafeiratos, Viktoriia A. Saveleva, Li Wang, Aldo S. Gago, K. Andreas Friedrich, M. Haevecker and A. Knop-Gericke for their contribution to the work. Financial support received in the framework of ERA.NET.RUS.PLUS (project ID # 270 – NANO-MORF) and European Union's Seventh Framework Programme (FP7/2007-2013) for Fuel Cell and Hydrogen Joint Technology Initiative under Grant No. 621237 (INSIDE) is gratefully appreciated.

Me-N-C catalysts are today's most active non-precious metal catalysts for the oxygen reduction reaction (ORR) in proton exchange membrane fuel cells (PEMFC). The application of these catalysts dates back to 1964 when Jasinski was the first demonstrating ORR activity of different phthalocyanines in alkaline electrolyte. The most important steps in improvement of ORR activity and stability were made by the heat-treatment of MeN₄-macrocycles in 1976 and the pyrolysis of independent metal, nitrogen and carbon sources in 1989 as shown by Jahnke et al. respectively Gupta et al..

However, especially during the last decade important progress in the development and fundamental understanding of these catalysts was gained. Based on these findings the ORR activity was boosted significantly. This motivated researchers worldwide to contribute to the development of Me-N-C catalysts and to the controversy of the origin of ORR activity.

In this contribution the most important milestones in the development of Me-N-C catalysts will be summarized. Based on this rational criteria can be defined for the further optimization of these catalysts with respect to a final implementation in the car.

Polymer electrolyte membrane (PEM) fuel cell performance and materials degradation, particularly associated with the cathode catalyst layer (CCL), can be directly attributed to the structure and chemistry of individual material components, as well as their uniformity/homogeneity within a CCL. The individual material constituents used to form the CCL within the membrane electrode assemblies (MEAs), which include the electrocatalyst, catalyst support, and ionomer films, and especially the critical interfaces that are formed between these various constituents, are critically importance in controlling fuel cell perfomance. Understanding the specific microstructural characteristics of the individual materials within the CCL, and how the materials interact to "form" the CCL, is important for identifying materials optimization parameters that can significantly enhance performance and durability. Materials in several states/conditions, e.g., prior to incorporation in the CCL (as-synthesized), after MEA preparation (CCL), and after fuel cell testing, are beig evaluated and quantified using a combination of advanced electron microscopy methods, which are used to interrogate the materials constituents and interfacial structures and chemistries from the μm- to the Å-level. The as-processed (prior to and following incorporation into a CCL) microstructural evidence is directly correlated with observations of materials-specific degradation mechanisms that contribute to fuel cell performance loss, and are then used to identify potential processing variables (materials-based mitigation strategies) to improve the microstructure and compositional homogeneity within the electrode structure, and enhance MEA durability and stability during fuel cell operation.

Research efforts at Oak Ridge National Laboratory are focused on the high-resolution microstructural and microchemical characterization of MEAs fabricated using different electocatalysts (typically Pt-based) and catalyst loadings, carbon-based support materials, and ionomer solutions, as well as the same MEAs subjected to accelerated stress tests (ASTs) designed to degrade specific MEA components. While a significant microscopy effort has been aimed towards understanding catalyst degradation (e.g., coarsening, de-alloying), recently, high-resolution analytical microscopy methods have been used to directly image/map the distribution and chemistry of the ionomer films/layers within CCLs, results of which are being combined with high-resolution imaging and 3-D tomography data of powder materials and MEAs, to provide unprecedented insight into the MEA architecture and interfaces (ionomer/support, ionomer/catalyst, catalyst/support, ionomer/pore). This presentation will focus on understanding materials distributions within CCLs as a function of materials used and ink/MEA processing variables, e.g., initial ionomer and/or ink chemistry, electrocatalyst (type, content, and dispersion), and the type of carbon support used. Additionally, the stability of the ionomer films, electrocatalysts, and support structures in CCLs following ASTs designed specifically for either catalyst degradation or carbon corrosion, will be evaluated.

 ________________________________________________________________________

Research sponsored by (1) the Fuel Cell Technologies Office, Office of Energy Efficiency and Renewable Energy, U.S. Department of Energy and (2) Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility.

Widespread of fuel cells are strongly desired from the point of reducing the GHG and for clean environment. In order to achieve these, reducing the cost of the fuel cell vehicle is needed and from this point, higher operating temperatures of PEFC can contribute for designing compact and simple cooling system. We have been developing variety of new Perfluorosulfonic acid (PFSA) ionomers having high ion exchange capacity (IEC) and other properties required for each the membrane and the electrolyte in the electrodes. The ionomer with high proton conductivity was tested by copolymerization of tetrafluoroethylene with a new bifunctional fluorosulfonyl monomer ¹. The ion exchange capacity was 1.58 meq/g and the water uptake after immersion in hot water at 80 °C for 16 hours was much lower compared to that of conventional ionomer.

Reduction in Pt loading in the cathode is another requirement for PEFC to reduce the cost. Especially at higher temperature and lower RH conditions under which the flux of oxygen to the Pt surface through the ionomer in the electrode could be the rate limiting factor for the oxygen reduction reaction (ORR). In order to increase the oxygen flux to Pt surface, the use of ionomers having higher oxygen permeability can be a solution. We have designed and synthesized various ionomers based on a general rule that the lower the density of a fluorinated polymer, the higher the oxygen permeability. One of the newly prepared ionomer having lower density showed more than 2 times higher the oxygen permeability ² at low RH conditions. The use of this high oxygen permeable ionomer at cathode showed the possibility of achieving the same cell voltage with half the amount of Pt loading.

We see that the ionomer plays an important row in developing PEFC achieving the requirements from the market. The ionomer can also be used in the micro porous layers (MPLs) between the catalyst layers and the Gas diffusion substrate to create hydrophilic MPLs. Conventionally, hydrophobic MPLs were applied between the GDM and the catalyst layer to prevent flooding due to excess water in the electrode. But the hydrophobicity of the MPL is not favorable when the cell is operating under a dry condition. We have been reporting the use of hydrophilic MPL between the GDM and the catalyst layer ³. It can control the water balance in the MEA, achieving higher performance under low RH conditions without facing the low performance due to mass transfer issue at wet conditions. This hydrophilic MPL was used in the cathode in combination with a high proton conductive membrane, and the MEA showed excellent performance under a very dry condition.

We are continuously making efforts developing PEFC materials getting into the next stage.

References

1. A.Watakabe, H Yamamoto, M Iwaya, S Saito, K Yamada, S Hommura and T Shimohira, Fluoropolymer (2012)

2. K. Yamada, S. Hommura, and T. Shimohira, ECS Trans., 50 (2), 1495 (2013).

3. T. Tanuma and S. Kinoshita, J. Electrochem. Soc., 161 (1), F94 (2014).

Background and objective

One of the important challenges in the current fuel cell research is to develop an alternative membrane to state-of-the-art perfluorinated sulfonic acid (PFSA) ionomers. While the PFSA ionomer membranes are highly proton conductive and chemically and physically stable, high gas permeability, high cost, and environmental incompatibility of the fluorinated materials are drawbacks for the wide-spread dissemination of fuel cells. In the past couple of decades, a variety of proton conductive materials have been proposed as alternative membranes. Nonfluorinated hydrocarbon polymers, acid-doped polymers, inorganic/organic nanohybrids, solid acids with superprotonic phase transition, and acid/base ionic liquids fall into this category. Some of them are claimed to exhibit high proton conductivity, very low gas permeability, and reasonable stability. However, none of them can compete with PFSAs under a wide range of fuel cell operating conditions. The most critical issues of these alternative membranes are insufficient durability and significant dependence of the proton conductivity upon humidity. In addition, interfaces between these emerging proton conductive materials and the catalyst layers have not been well explored. In this communication, current issues (in particular, chemical and mechanical stabilities and interfacial problems) and future possibilities of aromatic ionomer membranes will be discussed.

Stability issues

In pursuit of better performing and more stable aromatic ionomer membranes, we have investigated the effect of a simple structure, sulfo-1,4-phenylene unit, as a hydrophilic component on the membrane properties. Unlike typical aromatic ionomer membranes, the newly designed ionomer membranes exhibited reduced water uptake and excellent mechanical stability under humidified conditions due to the absence of polar groups such as ether, ketone, or sulfone groups in the hydrophilic component. The membrane has survived several tens of thousands cycles in humidity cycling test of US DOE protocol. The high local ion concentration in the hydrophilic segments contributed to an increase in the proton conductivity especially at low humidity. Consequently, the high ion exchange capacity (IEC = 2.67 meq/g) membrane exhibited high proton conductivity under low humidity (20% RH, 7.3 mS/cm) conditions at 80 °C, which are one of the highest values reported thus far for aromatic ionomer membrane.¹⁾

Interfacial issues

In the literature, a number of hydrocarbon ionomer membranes that exhibit comparable proton conductivity to PFSA membranes have been reported, however, they underperformed in operating fuel cells compared to PFSA membranes in most cases. This is partly because of the cathode performance was lower even when the same cathode catalyst layers were used. We have revealed that the double ionomer membrane composed of thin layer Nafion attached on our aromatic ionomer membrane showed higher fuel cell performance than the parent (single layer) ionomer membrane.^2,3) The Nafion inter-layer improved the interfacial contact between the aromatic ionomer membrane and the catalyst layer. Based on the results, we have then designed and synthesized novel ionomer membranes composed of perfluoroalkyl and the sulfonated phenylene groups. The resulting membranes could support good electrocatalytic performance of the catalyst layer at the membrane-electrode interface due to the well-controlled finely phase-separated morphology.

Acknowledgement

This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO) through the SPer-FC Project.

References

1) J. Miyake, T. Mochizuki, K. Miyatake, ACS Macro Lett., 4, 750-754 (2015).

2) T. Mochizuki, K. Kakinuma, M. Uchida, S. Deki, M. Watanabe, K. Miyatake, ChemSusChem, 7, 729-733 (2014).

3) T. Mochizuki, M. Uchida, H. Uchida, M. Watanabe, K. Miyatake, ACS Appl. Mater. Interfaces, 6, 13894-13899 (2014).

Proton exchange membrane (PEM) fuel cells are being developed as alternative energy sources for both residential and automotive application. In order for this technology to become fully commercial, the reduction of cost and improvements in performance and durability of PEM fuel cells membrane electrode assemblies (MEAs) are still required. To address the requirement for further cost reduction the Pt loading of the cathode catalyst layer (CCL) needs to be reduced to 0.2- 0.1 mg/cm² while maintaining high efficiency of Pt-utilization at high power densities. Consequently, it becomes increasingly important to not only develop new catalyst materials, but also optimize the 3D structural arrangement of the CCL components such as catalyst, ionomer and void space so that all critical functionalities can be achieved simultaneously and optimize PGM utilization. In general terms this entails to provide sites catalytically active for ORR, and to further provide transport to/these sites for the reactants O₂, protons, electrons and products H₂O and heat, respectively. In order to be able to design CCL structures that meet the performance and durability requirements, it is necessary to obtain a better understanding of structure versus performance relationships. This requires the capability to fabricate different CCL structures, to characterize the spatial distribution of all components within the catalyst layer^1,2 (carbon, Pt, ionomer and void), to measure the physico-chemical properties (both ex-situ and in-situ) and finally to use these experimental data as inputs for the development a model based understanding of the relationship between CCL structure and CCL performance and durability³.

References

• A. P. Hitchcock, V. Berejnov, V. Lee, D. Susac and J. Stumper, "In situ methods for analysis of polymer electrolyte membrane fuel cell materials by soft X-ray scanning transmission x-ray microscopy", Microscopy & Microanalysis 20(S3) 1532 (2014)

• V. Lee, V. Berejnov, M. West, S. Kundu, D. Susac, J. Stumper, R.T. Atanasoski, M. Debe and A.P. Hitchcock, "Scanning Transmission X-ray Microscopy of Nano Structured Thin Film Catalysts for Proton-Exchange-Membrane Fuel Cells" J. Power Sources 263 163 (2014)

• S. G. Rinaldo, W. Lee, J Stumper, M. Eikerling: "Nonmonotonic dynamics in Lifshitz-Slyozov-Wagner theory: Ostwald ripening in nanoparticle catalysts" Physical Review E 86, 041601 (2012)

One key challenge to enable a wide spread adoption of proton-exchange membrane fuel cell vehicles is the reduction of Pt use to a level comparable to the incumbent internal combustion engine vehicles (<5 g_PM/vehicle). Great progress in recent years in improving the activity and durability of Pt-based catalysts, particularly Pt alloy catalysts, has enabled a reduction of Pt loading in the cathode approaching 0.15 mg_Pt/cm² or about 15 g_Pt/vehicle. However, as we attempt to go below 10-15 g/vehicle, a large performance loss is often observed at high-current density (high power).

In the cathode, where oxygen molecules meet protons and electrons to produce water, one must create as much interface as possible in order to minimize the transport resistances. As Pt loadings are reduced, performance losses due to these transport phenomena become more localized to the Pt and ionomer interface, posing new challenges to diagnose and understand these electrodes.

In this talk, we will discuss how we use in-situ electrochemical diagnostics and modeling to understand the performance of low-Pt electrodes, identify their key limiting factors, and guide our catalyst layer development. Diagnostics used to quantify the kinetic terms, proton conduction, oxygen transport resistance, and Pt-ionomer interaction will be introduced. Moreover, because non-noble metals leach out from Pt alloy catalysts and can replace protons in the ionomer, we will also consider the resulting impact on low-Pt electrode performance.

This work was partially supported by the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy under grant DE-EE0007271.

The United Nations Framework Convention on Climate Change (COP21) in Paris last year has underlined the urgency of a fast and consequent transformation of the various energy systems from a fossil based energy feedstock towards renewable energies like photovoltaics, wind- and hydropower, in order to drastically decrease the greenhouse gas emissions (GHG). Such a transformation requires a thorough restructuring of the energy systems with all energy carriers and consumer sectors including its intersectoral dependencies, grid structures and energy storage needs.

At Fraunhofer ISE we performed time-resolved studies on how to reach the main goal of the German energy transformation to decrease its GHG emissions by at least 80% until 2050 and thus, to widely decarbonize all energy related sectors. Potential transformation pathways based on various scenarios were studied and the required costs were modelled for each scenario. The results of this modelling made very obvious that the ambitious GHG reduction targets can only be fulfilled by a strong rise in electricity generation by fluctuating renewable energies and on the other hand by the installation of large plants for producing synthetic energy carriers from renewable energies.

The most versatile energy carrier for such large amounts of energy is hydrogen produced from water electrolysis. Hydrogen can be stored in large amounts in gas tube fields or salt caverns and can be used directly either in fuel cell cars and buses as a fuel or in stationary combined heat and power fuel cell systems. Furthermore the hydrogen can be used in order to hydrogenate CO2 and thus to generate green liquid fuels such as oxymehylenether (OME) or other chemicals in order to replace oil based products. The conversion of electricity into hydrogen is called Power-to-Gas, the further processing of the hydrogen and CO2 towards liquid energy carriers and chemicals is called Power-to-Liquid.

With the advent of various Power-to-Gas demonstration projects it becomes clear how important this technology will be in the future energy system. Distributed water electrolyzers in the Megawatt power range will stabilize the grid frequency and voltage or will deliver other ancillary services and thus, will contribute to the secondary balancing market.

Alkaline water electrolyzers are applied for more than 100 years in industrial applications and are nowadays complemented by proton exchange membrane (PEM) electrolysers in order to cover the needs for a highly dynamic load in the grid. However, main challenges of the PEM technology are high costs associated with expensive materials and the durability of cells and stacks. With ongoing technological development it is expected that PEM electrolysis systems become a competitive alternative to alkaline systems in the next years.

In this talk, the technological base of water electrolyzers will be described and the related performance data, life-time expectation and cost reduction potential will be discussed. Furthermore the impact of distributed water electrolyzers in the Megawatt power range as part of an increasingly renewable energy system in terms of Power-to-Gas modelling scenarios will be given.

Figure 1

 PEM water electrolysis is one of the promising technologies for hydrogen production from renewable energy sources. Iridium and their oxides are recognized as the most suitable catalyst showing high activity and stability for oxygen evolution reaction (OER) in acidic media [1], however, overpotential is still high and its scarcity and price are also drawbacks. To overcome these problems, a number of studies concerning particle size reduction, catalyst compositions, and utilization of catalyst support have been reported [2]. Among them, use of corrosion-resistant and highly electro-conductive oxides, especially Magneli phase Ti₄O₇ as catalyst support would be beneficial for reducing Ir loading and increasing durability. We have developed Ti₄O₇-supported Ir catalyst and examined activity and durability as an anode catalyst of PEM electrolyzer. 

Titanium oxide support was prepared by UV laser technique [3]. Ir/Ti₄O₇ catalysts obtained by using H₂IrCl₆ precursor showed conductivity comparable with that of commercial Pt/C. MEAs with the Ir/Ti₄O₇ anode can be operated up to 4 A cm^-2 at low cell voltage < 2.0 V (with DuPont NR212 membrane) with very low Ir loadings (0.25 mg-Ir cm^-2). Considering fluctuating nature of the renewable energy inputs, catalyst stability against voltage change should be another important issue. To simulate and accelerate such degradation modes, potential sweep between 1.0 V and 2.0 V at a sweep rate of 500 mV s^-1 was applied up to 10,000 cycles. The Ir/Ti₄O₇ MEA shows limited degradation even after 10,000 cycles. Stability of the cell performance and the catalyst structure will be discussed at the Meeting.

References

1. M. H. Miles and M. A. Thomason, J. Electrochem. Soc.,123, 1459 (1976).

2. M. Carmo, D. L. Fritz, J. Mergel, D. Stolten, Int. J. Hydrogen Energy, 38, 4901 (2013).

3. T. Ioroi, H. Kageyama, T. Akita, K. Yasuda, Phys. Chem. Chem. Phys., 12, 7529 (2010).

1. Introduction

Renewable energy is an alternative choice to solve the universal problem of energy depletion and environmental pollution. Among various energy carriers, hydrogen is a most competitive one, because it is friendly to environment and can be produced from renewable energy through water electrolysis. Among types of water electrolysis, polymer electrolyte membrane water electrolysis (PEMWE) attracts most attention for its advantages because of simple structure and high energy conversion efficiency.

Usually, water electrolysis needs higher voltage supply than theoretical value, due to high overvoltage in oxygen evolution reaction at anode. To minimize the anodic overvoltage, it is recently suggested that annealing catalyst powder at certain temperature and time can decrease both the HFR and activation overvoltage, and the annealing effects are proven in the conventional PEMWE, where operation temperature is around 80℃ [1-4]. The current mechanisms about how annealing impacts electrolysis performance are as followings: (i) annealing can largely decrease electrolysis voltage by reducing HFR [1]; (ii) annealing will enlarge the specific surface area [2]; (iii) annealing increases the catalyst activity by changing the structure of catalyst particle [3~4]. In order to examine the above three mechanisms, we evaluated I-V and I-HFR characteristics of IrO2 samples which have different specific surface area and were annealed under different temperatures and time. This study also challenges to find the optimum condition of annealing time and temperature, to minimize electrolysis voltage under rather high temperature operation of 100℃.

2. Experiment

PEMWE cell used in this study consists of a homemade CCM (with electro-catalyst area of 4cm2), porous current collectors, and separators with flow channels. For the CCM fabrication, anode and cathode catalytic electrodes are hot pressed to a membrane (Nafion117). Iridium oxide powder (IrO2 powder, type IV, Tokuriki Co., Japan) as for anode catalyst is annealed in air at different time and temperatures shown in table 1. Rasten et al [1] suggests that annealing at 490℃ and 8hrs results in the lowest electrolysis voltage. Siracusano et al [3~4] shows that the optimum annealing condition should be 350℃ and 1hr. Based on these, Annealing condition of the powder is at 350℃ and 490℃ from 1 to 8 hrs. To examine the effect of enlarging specific surface area, IrO2 samples with two different specific surface area were selected: sample 1 and sample 2 originally has the specific surface area of 42m2/g and 84m2/g respectively. A commercial 46% Pt/C (Tanaka Kikinzoku Japan) catalyst is used for cathode electrode. Ti mesh (Nikko Techno Co., Japan, fiber diameter 20μm) is used for anode current collector. As for cathode current corrector, carbon paper (hydrophobic, SGL Co., Germany, 34BC) is chosen for comparing with each other. Water at room temperature is fed into only anode channel at a flow rate of 1mL/min, which corresponds to the water utilization of 2% at 1A/cm2. All the experiments are conducted at atmospheric pressure and 100℃.

3. Result and discussion

Fig. 1 (a) and (b) shows how specific surface area and annealing conditions impact on I-V characteristics and high frequency resistance at 10 kHz. The IV and I-HFR characteristics revealed that: (i) enlarging specific surface area to two times has little impact on electrolysis performance; (ii) annealing catalyst powder at 350℃ and 1hr introduced the best electrolysis performance and annealing at 490℃ and 8hrs introduced the highest electrolysis voltage. When annealing temperature is 350℃, long annealing time will raise the electrolysis voltage. But when annealing temperature is 490℃, annealing time makes a limited impact on electrolysis voltage. It also should be noticed that the OCV changes with annealing temperature, annealing catalyst annealed at 490℃ generates much higher OCV (about 80mV) than that annealed at 350℃. Such phenomenon means annealing at 490℃ will decrease the catalyst activity. Annealing catalyst at certain temperature can decrease the HFR, but the decreased ohmic overvoltage can't explain the total electrolysis voltage difference. Therefore, among the above three main mechanisms, annealing method should impact the electrolysis performance by increasing the catalyst activity via changing the structure of catalyst particle.

References

[1]. Egil Rasten, Georg Hagen, Reidar Tunold. Electrochimica Acta (2003) 3945-3952.

[2]. J. C. Cruz, V. Baglio, S. Siracusano, R. Ornelas, L. Oritiz-Frade, L. G. Arriaga, V. Antonucci, A.S. Aricò. J Nanopart Res (2011) 13: 1639-1646.

[3]. S. Siracusano, V. Baglio, A. Stassi, R. Ornelas, V. Antonucci, A.S. Aricò. International journal of hydrogen energy 36 (2011) 7822-7831.

[4]. S. Siracusano, S. Siracusano, A. Di Blasi, N. Briguglio, A. Stassi, R. Ornelas, E. Trifoni, V. Antonucci, A.S. Aricò. International journal of hydrogen energy 35 (2010) 5558-5568.

Figure 1

The development of highly active and stable electrocatalysts for OER is urgent due to the slow kinetics and high overpotential of oxygen evolution reaction (OER) [1]. To date, the most promising catalysts for OER are IrO₂ based catalysts with ordered porous structures, where a templating method was generally used to ensure the high utilization of the catalysts. However, the synthesis route for such catalysts is generally of high technique complexation, where the templates need to be effectively removed after synthesis [2,3]. Hence, a simple template-free method that leads to the formation of highly porous IrO₂ catalysts with high surface area is highly attractive.

Herein, a satisfactory template-free ammoniating method was developed in which NH₃ ligand was successfully kept until pyrolysis treatment, thus acting as the pore-forming agent to achieve highly porous IrO₂. The specific and electrochemical surface areas of as-prepared IrO₂ catalysts increased with the increasing quantity of NH₃·H₂O additive. Specifically, IrO₂ (1:100)-450 ^°C catalyst synthesized with the highest ratio of NH₃ ligand exhibited high specific surface area of 363.3 m² g^-1 as metal oxide. Furthermore, IrO₂ (1:100)-450 ^°C showed remarkable electrocatalytic activity for acidic OER in the whole potential window, which can be ascribed to the high catalysts utilization. As shown in Figure 1, the over-potential for the IrO₂ (1:100)-450 ^°C catalysts to attain current at 10 mA cm^-2 was only 282 mV, which was shifted negatively by 31 mV and 46 mV in comparison to the home-made IrO₂ (without NH₃) and commercial IrO₂ (CM) catalysts, respectively. Thus, the NH₃ mediating method was found a superior method for increasing catalysts utilization, where the IrO₂can be conveniently synthesized in large scale.

Figure 1. CV curves (Left, the inset shows the specific surface area) and LSV curves (Right, the inset shows the TEM of IrO₂ (1:100)-450^°C) of IrO₂ (1:100)-450^°C, IrO₂ (1:10)-450^°C, IrO₂-450^°C, and IrO₂(CM) catalysts.

Acknowledgments

This work is supported by National Basic Research Program of China (973 Program, 2012CB932800), National Natural Science Foundation of China (21433003, 21373199), the Science & Technology Research Programs of Jilin Province (20150101066JC, 20160622037JC).

References

[1] S. Park, Y. Shao, J. Liu, and Y. Wang, Energy Environ. Sci., 5, 9331 (2012).

[2] W. Hu, Y. Wang, X. Hu, Y. Zhou, and S. Chen, J. Mater. Chem., 22, 6010 (2012).

[3] E. Ortel, T. Reier, P. Strasser, and R. Kraehnert, Chem. Mater., 23, 3201 (2011).

Figure 1

A bifunctional catalyst for both oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) could be employed in a unitized regenerative fuel cell, an energy storage device that can be coupled to intermittent renewable energy such as wind or solar to peak-shift electricity to the grid. Though significant work has been done over the past few decades, no catalyst for both ORR and OER with high activity for both reactions have been discovered. It is well know that platinum or platinum based alloys exhibit the highest activities for ORR among all other catalyst, however they display poor activity for OER. In contrary, transition metal (hydro)oxides, such as iron, nickel, or cobalt (hydro)oxides, always possess highly activities for OER but poor activity for ORR. Due to "migration-effect" of transition metal in ORR, in which the free transition metal (hydro)oxides will move to the surfaces of platinum during electrochemical process and results in the dramatically decrease of the ORR activity of Pt eventually, it is impossible to achieve high activity for both ORR and OER by simply mix platinum or platinum based alloy with transition metal (hydro)oxides together. Inspired by this, we developed a novel method to prepare 7nm platinum nanoparticles doped with atomic sized cobalt oxides for the applications of ORR and OER. Our primary results showed that these hetero-structured materials exhibited higher activity than state-of-the-art commercial platinum for ORR, but also 10 times higher activity for OER than Co3O4. Most importantly, our catalysts also demonstrated excellent stability. Apparently, the interfaces must play a dominated role in both ORR and OER. Investigation of the interfaces between platinum and Co3O4 of our catalyst in atomic scale is crucial to understand how the interfaces worked and what kinds of interfaces are better for both ORR and OER. Our work will advanced the fundamental understanding of the metal-oxide interface in enhancing the ORR and OER, which is envisioned to shed light on the design of advanced catalysts for catalyzing complex chemical processes.

A PEM (Polymer Electrolyte Membrane) water electrolysis cell is a concept of zero-gap cell that uses a thin (100-200 micrometers thick) film of proton-conducting polymer as solid electrolyte [1]. Elevated operating current densities (in the multi A.cm^-2 range) can be reached efficiently. Most popular polymer electrolytes operating between 40 and 80°C are perfluoro-sulfonated materials (PFSA like Nafion^® or Aquivion^® [2]). Attempts to use phosphoric acid doped PBI (polybenzimidazole) electrolytes operating at higher (150-200°C) temperature have also been reported [3]. Platinum group metals (PGM) electrocatalysts are extensively used because only them can sustain the highly acidic environments [4]. Regarding electrocatalysis, conventional PEM water electrolysis cells are using platinum black (unsupported Pt particles) at the cathode for the hydrogen evolution reaction (HER) and unsupported iridium dioxide at the anode for the oxygen evolution reaction (OER). Typical PGM loadings vary from 0.5 – 1.0 mg.cm^-2 at the cathode to 1-2 mg.cm^-2 at the anode. Whereas the relative cost of platinum group metal (PGM) electrocatalysts in industrial PEM systems is limited to a few percent, costs constraints (especially in view of the development of electrolysers at the multi MW scale, for example for energy storage applications) are calling for cheaper solutions. A first option is to reduce PGM loading. A second option is to develop non-PGM electrocatalysts.

Regarding the reduction of PGMs, carbon-supported Pt nano-particles can be advantageously used to reduce Pt loadings from 1 down to 0.1 mg.cm^-2 at the cathode. Reduced IrO₂ loadings (down to 0.5 – 1.0 mg.cm^-2) have been demonstrated [5] but addition of an inert metal [6] or alloying (to form ternary or quaternary mixed oxides [7]) are also viable alternatives. Regarding the development of non-PGM materials, most advances concern the cathode. Ideally, catalyst containing transition metals (Ni, Co, Fe) that maximize the exchange current density as a function of the M-H bond strength should be used [8]. They are significantly less expensive than PGMs and still adequately electro-active. Whereas in alkaline water electrolysis technology, Ni and Co bulk particles are commonly used, their implementation at the cathode of PEM water electrolyzers is not a straightforward task because they are rapidly corroded. New approaches based on molecular chemistry have been developed to bypass the problem. For example, metallic molecular complexes such as Co, Ni, Fe clathrochelates have been successfully implemented at the surface carbonaceous substrates (carbon powder and fibers) and implemented in PEM water electrolyzers [9]. They offer several advantages compared to nanoparticles: (i) non noble metals can be used as active centres; (ii) only limited amounts of metals are required; (iii) redox properties of active centres can be tuned to desirable values by selecting appropriate organic ligands; (iv) ligands can be used as chemical linkers for efficient surface functionalization.

The purpose of this communication is to review existing and innovative electrocatalysts for PEM water electrolysis applications and to compare their efficiency in relation with microstructural aspects.

[1] W.T. Grubb, L.W. Niedrach, Batteries with Solid Ion-Exchange Membane Electrolytes. II. Low Temperature H₂-O₂ Fuel Cells, J. Electrochem. Soc., 107(2) (1960) 131 – 134.

[2] K.A. Mauritz, R.B. Moore, 'State of understanding of Nafion', Chemical Reviews, 104 (2004) 4535 – 4585.

[3] D. Aili, M.K. Hansen, C. Pan, Q. Li, E. Christensen, J.O. Jensen, N. Bjerrum, Phosphoric acid doped membranes based on Nafion, PBI and their blends, Int. J. Hydrogen Energy, 36 (2011) 6985 – 6993

[4] P. Millet, 'PEM water Electrolysis for hydrogen production: Principles and Applications', chapter 10, PEM electrolyzer characterization tools, D. Bessarabov, H. Wand, H. Li, N. Zhao, CRC Press (2015).

[5] C. Rozain, E. Mayousse, N. Guillet, P. Millet, Influence of iridium oxide loadings on the performance of PEM water electrolysis cells: Part I – Pure IrO₂-based anodes, J. Appl. Catalysis B: Environmental, 182 (2016) 153 – 160.

[6] C. Rozain, N. Guillet, E. Mayousse, P. Millet, Influence of iridium oxide loadings on the performance of PEM water electrolysis cells: Part II – Advanced anodic electrodes, J. Appl. Catalysis B: Environmental, 182 (2016) 123 – 131.

[7] A.T. Marshall, S. Sunde, M. Tsypkin, R. Tunold, Performances of a PEM water electrolysis cell using IrxRuyTazO2 electrocatalysts for the oxygen evolution electrode, Int. J. Hydrogen Energy, 32(13) (2007) 2320-2324.

[8] S. Trasatti, Work function, electronegativity, and electrochemical behaviour of metals: III. Electrolytic hydrogen evolution in acid solutions, J. Electroanal. Chem. 39 (1972) 163-184.

[9] M-T. Dinh Nguyen, A. Ranjbari, L. Catala, F. Brisset, P. Millet and A. Aukauloo, Implementing Molecular Catalysts for Hydrogen Production in Proton Exchange Membrane Water Electrolysers, Coord. Chem. Review, 256 (2012) 2435 – 2444.

The interest in harnessing energy from renewable sources and a push to achieve environmental cleanliness in the energy industry keeps gaining momentum. The need to be able to convert and store all that renewable energy has rekindled interest in hydrogen as a clean and environmentally benign energy carrier. In fact, hydrogen seems to be the only energy storage medium capable of handling the vast grid-scale energy storage in a reasonable manner. Likewise, recent accomplishments in auto-motive fuel cells adds further impetus to advance simultaneously renewable and economical hydrogen generation technology, which will be the ultimate key to success for hydrogen as fuel in societal everyday transportation solutions. All this is done with the ultimate goal of supplanting the current hydrocarbon based economy by hydrogen economy as means to achieve globally environmentally friendly energy industry and regional/local energy independence.

This recent increase in significance placed on the age-old water electrolysis process gave rise to new approaches to generate hydrogen with increased efficiency and lower cost. Polymer Electrolyte Membrane (PEM) water electrolysis emerged as one of the best matched choices today for the highly variable and unpredictable nature of the renewable energy.

There are two main ways to lower the cost of hydrogen production via PEM water electrolysis: to lower the capital expenses (CAPEX) and/or to lower the operating expenses (OPEX). We at 3M have recently showed a way to address reducing the first (high CAPEX ) by successfully demonstrating the ability to widen the range of current densities where electrolyzers can operate from a previous maximum of about 2.0 A/cm² as used today in commercial electrolyzers, to as much as 20 A/cm² in our novel constructions. While the research on high power density operation of PEM water electrolysis still continues and many questions remain as yet unanswered, the 3M's proprietary Nano Structured Thin Film (NSTF) catalyst with its highly conductive, compact, and hydrophilic catalyst layer seems uniquely fitting to the task. Indeed, its hydrophilicity, which is perhaps a major challenge in fuel cells, is a tremendous strength and asset in PEM water electrolysis and one of the features enabling the extremely high virtually mass transport free current densities we reported previously¹.

In this work, we intend to present results from our attempts to tackle the second means to reduce the cost of hydrogen production (the OPEX), by increasing the intrinsic activity of the catalyst as means to increase the kinetics of oxygen evolution and hence the efficiency of water electrolysis. In the preceding work we have explored the Pt/Ir compositional parameter space (in the NSTF alloy catalyst format) and the compositional effects on fundamental catalyst activity, as determined by Rotating Disk Electrode (RDE) measurements and the evaluation of single electrolyzer cells^2,3. That work has shown, rather unexpectedly we must add, quite detrimental effects of alloying basic Ir-NSTF catalyst with Pt in RDE tests. The results were confirmed, and perhaps even more drastic, when catalyst compositions were tested for performance in the actual electrolyzer hardware. Now, we intent to show the effect of alloying Ir with sample of non-PGM transition metals (again, using 3M's proprietary Ir-NSTF catalyst format) and the effects of such alloys on the basic catalytic activity. Subsequently, we will correlate this to the in-situ performance measurements in a PEM electrolysis cell. Finally, we will discuss the possibilities in other promising alloying compositions and structures.

1.    Krzysztof A. Lewinski, Sean M. Luopa, (invited) "High Power Water Electrolysis as a New Paradigm for Operation of PEM Electrolyzer" (abstract 1948), Spring ECS Meeting, Chicago, IL, May 2015.

2.    Krzysztof A. Lewinski, Dennis van der Vliet, and Sean M. Luopa, "NSTF Advances for PEM Electrolysis - the Effect of Alloying on Activity of NSTF Electrolyzer Catalysts and Performance of NSTF Based PEM Electrolyzers" (Abstract 1457), Fall ECS Meeting, Phoenix, AZ, Oct 2015.

3.    Krzysztof A. Lewinski, Dennis van der Vliet, and Sean M. Luopa, "NSTF Advances for PEM Electrolysis - the Effect of Alloying on Activity of NSTF Electrolyzer Catalysts and Performance of NSTF Based PEM Electrolyzers", (10.1149/06917.0893ecst), ECS Transactions, 69 (17) , p. 893-917 (2015).

The development of cost-effective polymer electrolyte membrane (PEM) electrolyzers is key to the implementation of a hydrogen-based infrastructure. It cannot be denied that there is an ever-increasing demand to shift away from the use of nuclear power and the traditional burning of fossil fuels as primary energy systems. The only foreseeable long-term solution has been increasingly focused on the implementation of clean, renewable power supplies, such as wind farms and solar power stations. In this regard, the electrochemical conversion of water to hydrogen is expected to play a key role in the development of scalable energy storage that is required for such intermittent power supplies(1).

The cathodic generation of hydrogen from water splitting is known to be extremely facile on Pt-based catalysts(2). The simultaneous oxygen evolution reaction (OER) occurring at the anode, however, is limited by sluggish kinetics and requires a considerable overpotential to achieve modest current densities. Moreover, the harsh acidic environment and high anodic operating potentials limits the choice of stable electrocatalyst materials to those of the noble metal oxides. Reduction of the noble metal loading at the anode and enhancing catalyst stability for OER in PEM electrolyzers remains a challenge. Perhaps the most widely implemented approach for reducing the noble metal content is that which considers reducing the catalyst particle size(3, 4). A major issue that arises from this approach, however, is that it becomes increasingly difficult to establish structure-activity relationships for the OER due to the transformation processes, e.g. changes in microstructure and crystallinity, that commonly occur during the preparation of the materials.

The research reported herein is focused on expanding the fundamental understanding of the influence of crystallinity, particle size, and microstructure on the electrochemical OER activity of nanocrystalline IrO₂ with particular emphasis pointed towards formation of a hydrous surface layer. Chlorine−free iridium oxide nanoparticles are synthesized using the modified Adams fusion method(5), which is capable of producing spherical 1.7 ± 0.4 nm particles with a specific surface area of 150 m²/g using a low temperature synthesis (350 °C). Increasing the synthesis temperature to 600 °C results in the formation of larger, rod−shaped particles mostly terminated by highly ordered non−defective (110) surfaces. X-ray absorption spectroscopy (XAS), X-ray photoelectron spectroscopy (XPS), and electrochemical studies indicate the presence of a hydrous surface layer, i.e. Ir(O)OH, that leads to an enhanced OER activity. We report that it is possible to create a larger hydrous layer on smaller nanoparticles and thus increase the specific OER activity.

References:

1. L. Bertuccioli, A. Chan, D. Hart, F. Lehner, B. Madden and E. Standen, Development of Water Electrolysis in the European Union, in, p. 83, Fuel Cells and Hydrogen Joint Undertaking (2014).

2. W. Sheng, H. A. Gasteiger and Y. Shao-Horn, Journal of The Electrochemical Society, 157, B1529 (2010).

3. J. C. Cruz, V. Baglio, S. Siracusano, R. Ornelas, L. Ortiz-Frade, L. G. Arriaga, V. Antonucci and A. S. Aricò, J. Nanopart. Res., 13, 1639 (2011).

4. Y. Lee, J. Suntivich, K. J. May, E. E. Perry and Y. Shao-Horn, The Journal of Physical Chemistry Letters, 3, 399 (2012).

5. E. Oakton, D. Lebedev, A. Fedorov, F. Krumeich, J. Tillier, O. Sereda, T. J. Schmidt and C. Coperet, New Journal of Chemistry, 40, 1834 (2016).

Within the United States, hydrogen is a major chemical commodity for the production of ammonia in agriculture and the upgrading of crude oil in transportation. Hydrogen in the US is produced largely from natural gas by steam methane reformation.[1] Although electrochemical water splitting currently accounts for a small amount of hydrogen production, it is expected to have a larger role in the future, particularly when using renewable energy sources as input. Studies in proton exchange membrane (PEM) electrolysis are usually focused on catalysts in the oxygen evolution reaction, due to the high overpotential relative to the hydrogen evolution reaction. PEM electrolyzers typically use iridium (Ir) or Ir oxide as the anode catalyst, due to reasonable activity and stability. Although platinum (Pt) and ruthenium (Ru) are often examined as potential alternatives, Pt requires a higher overpotential and Ru is prone to dissolution at elevated potential.[2, 3]

Determining the electrochemical surface area (ECA) of Ir allows for the quantification of the number of sites available to participate in oxygen evolution. It also allows for the evaluation of site-specific activity, or site quality. ECAs are of interest in Ir catalyst development, to compare different catalyst types and direct future research. ECAs are also of interest in durability studies to evaluate the modes of Ir degradation, and to determine whether activity losses are due to deteriorating surface area or site quality.

The ECAs of Ir catalysts have typically been evaluated using hydrogen underpotential deposition, carbon monoxide oxidation, or capacitance. These methods are generally limited to predurability measurements and catalyst type: hydrogen underpotential deposition and carbon monoxide for Ir metals; capacitance for Ir oxides. Hydrogen underpotential deposition and carbon monoxide oxidation can be used to determine the ECA of Ir metals, but not Ir oxides, and are sensitive to surface oxides formed on Ir metals following oxygen evolution characterization and durability testing. To date, no method is available to determine the ECAs of Ir and Ir oxide, prior to and following durability testing. Mercury underpotential deposition is presented in this study as an alternative, able to produce reasonable ECAs on Ir and Ir oxide nanoparticles, and able to produce similar ECAs prior to and following characterization in oxygen evolution. This method was previously developed for the study of polycrystalline Ir, and expanded in this study to include nanoparticle catalysts, oxides, and oxygen evolution relevant testing condition.[4]

Figure 1. Cyclic voltammograms of (a) Ir nanoparticles and (b) Ir oxide nanoparticles in 0.1 m perchloric acid (blue) and 0.1 m perchloric acid containing 1 mm mercury nitrate (red).

[1] A. Milbrandt, M. Mann, in: U.S. Department of Energy (Ed.), http://www.nrel.gov/docs/fy09osti/42773.pdf, 2009.

[2] T. Reier, M. Oezaslan, P. Strasser, ACS Catalysis, 2 (2012) 1765-1772.

[3] M. Pourbaix, Atlas of electrochemical equilibria in aqueous solutions, National Association of Corrosion Engineers, Houston, Texas, 1974.

[4] S.P. Kounaves, J. Buffle, Journal of The Electrochemical Society, 133 (1986) 2495-2498.

Figure 1

Renewable H₂ production is a prerequisite for successfully establishing fuel cell-based electromobility and hydrogen infrastructure. In this respect, proton exchange membrane water electrolysis (PEM-WE) has attracted much interest, not least due to the enormous power densities possible in these devices (1). Research is mainly focused on the anode side, where the oxygen evolution reaction (OER) takes place, causing the vast majority of kinetic losses.

OER catalysts of choice are usually iridium oxide (IrO_x) based, owing to its decent activity and acceptable stability (2,3). Recent studies showed a strong dependence of the OER activity and dissolution resistance of IrO_x on its surface morphology and hydration state, which were controlled by the calcination temperature during the synthesis. They found that crystalline, thermal IrO₂is the most stable but least active species (4,5).

TGA-analysis from our lab reveals that IrO_x can easily be reduced to metallic Ir in dilute H₂ at a typical PEM-WE operation temperature of 80 °C. This is also observed for IrO_x based membrane electrode assemblies (MEAs) held at OCV in a PEM-WE, where crossover H₂ from the cathode side reduces the surface of the IrO_x catalyst at the anode within hours, as evident from the formation of H-UPD features in the cyclic voltammogram (see Figure 1b, red vs. blue CV). At the same time, polarization curves after this reduction show significantly decreased cell voltage at slightly reduced Tafel slopes (Figure 1a, red vs. blue curves), corresponding to an increased OER activity. However, a subsequent CV following these polarization curves (see Figure 1b, black CV) reveals that the H-UPD features have disappeared again and that the catalyst surface properties have transformed to a state closer to the less crystalline, hydrous IrO_x reported in ref. 4.

This suggests that partial IrO_x reduction and re-oxidation into (hydrous) IrO_x can occur during cycles of extended OCV periods and electrolyzer operation. In analogy to voltage cycling degradation observed in fuel cells, the here described reduction/oxidation cycles might also lead to iridium dissolution. Therefore, we will study the effect of transient operation conditions in PEM-WEs on the OER activity, surface properties, and stability of IrO_x based anodes. We will also provide a systematic analysis of OER kinetic parameters such as exchange current density and activation energy as well as their dependence on relative humidity for a commercial iridium oxide catalyst.

References

(1) K. A. Lewinski, D. F. van der Vliet, and S. M. Luopa, ECS Transactions, 69, 893 (2015).

(2) C. Rozain, E. Mayousse, N. Guillet, and P. Millet, Appl. Catal. B, 182, 123 (2016).

(3) E. Fabbri, A. Habereder, K. Waltar, R. Kötz, and T. J. Schmidt, Catal. Sci. Technol., 4, 3800 (2014).

(4) T. Reier, D. Teschner, T. Lunkenbein, A. Bergmann, S. Selve, R. Kraehnert, R. Schlögl, and P. Strasser, J. Electrochem. Soc., 161, F876 (2014).

(5) S. Cherevko, T. Reier, A. R. Zeradjanin, Z. Pawolek, P. Strasser, and K. J. J. Mayrhofer, Electrochem. Commun., 48, 81 (2014).

Figure 1a. PEM-WE polarization curves recorded at 80 °C under dynamic O₂ (anode)/H₂ (cathode) at 147 kPa_abs and 200 % RH (inlet). Blue curves were taken after a 12 h conditioning at 1 Acm^-2 while red curves signify the status after a 15 h in-situ reduction of the anode catalyst in an H₂ atmosphere. Hollow circles and crosses are subsequent repetitions.

Figure 1b. Anode CVs recorded under dry N₂ at 100mVs^-1. Blue CV: before the polarization curves in blue, red and black CVs: before and after the polarization curves in red.

The anode was loaded with 0.66 mg_Ircm^-2 of a commercial IrO_x/TiO₂ using 12 wt-% of ionomer in the catalyst layer, cathode was loaded with Pt/C at 0.35 mg_Ptcm^-2 and a Nafion^® XL membrane was used. MEAs with 5 cm² active area were tested in a house-made single cell hardware with gold-plated titanium flowfields using porous Ti sheets and carbon fiber paper as GDLs on the anode and cathode side, respectively.

Acknowledgements:

P. J. Rheinländer would like to acknowledge financial support from Greenerity GmbH. M. Bernt would like to acknowledge funding from the Bavarian Ministry of Economic Affairs and Media, Energy and Technology through the project ZAE-ST (storage technologies).

Figure 1

Water electrolysis technology has attracted much attention because it serves as the key part of a carbon-neutral energy storage system as follows: electricity generated from renewable resources such as sunlight, water power, and wind, is used to produce hydrogen by water electrolysis. Then, the hydrogen is supplied to fuel cell system for power generation, which only generates electricity and water. Proton exchange membrane water electrolyzers (PEMWEs) are currently the most promising candidate because polymer electrolyte membrane provides higher ionic conductivity, better load flexibility, and purer hydrogen than alkaline electrolyte solution, anion exchange membrane, and solid oxide electrolyte, etc.

The key challenge for PEMWEs is to develop efficient anode electrocatalyst because of the high catalytic efficiency loss related with the oxygen evolution reaction (OER) on anode. In addition, the use of PEM requires the anode material durable under long-term anodic polarization in acid environment, which significantly limits the options of materials. At present stage, iridium oxide (IrO_2­) is widely regarded as the most promising anode electrocatalyst for PEMWEs owing to its high activity for OER and chemical stability in strong acid environment. Considering the price and scarcity of Ir, however, it is highly desired to lower the amount of Ir in the anode. In view of this, many approaches have focused on enhancing the mass activity of Ir-based catalysts towards OER by designing novel nanostructures. Another effective way is to disperse IrO₂ on electronic conductive support with large surface area. Unfortunately, the most common support material for electrocatalysts, carbon materials, are regarded as not appropriate for PEMWEs because it would encounter severe degradation at high oxidizing potential region. Instead, metal oxides with large surface area and resistance to acid are used. However, their low electrical conductivity limits the electrochemical performance for OER.

In previous studies, our group wrapped pristine multi-wall carbon nanotubes (MWNTs) with polybenzimidazole (PBI). The thin layer of PBI (thickness<1 nm) provides abundant binding sites for cations. As a result, uniform Pt nanoparticles could be deposited homogeneously on the PBI-wrapped pristine MWNTs. The resultant composite catalyst exhibited ehnhanced electrocatalytic activity and extraordinary long-term durability even polarized at high potential region of 1.0-1.5 V vs. RHE because (1) the aggregation of Pt nanoparticles was significantly suppressed due to the constraint effect of polymer and (2) the highly crystallized graphitic surface of pristine MWNTs refrained the electro-oxidation of carbon. The results provide a new insight into carbon nanotubes as an excellent support material with high corrosion-resistance under anodic polarization in acid environment.

In present work, a composite catalyst of pristine MWNT, PBI, and IrO₂ prepared by similar strategy. For comparison, Ir was also deposited on carbon black (CB) and PBI-wrapped CB. The products were characterized by thermogravimetry (TG), X-ray diffraction (XRD) analysis, X-ray photoelectron spectroscopy (XPS), Raman spectroscopy, and scanning transmission electron microscope (STEM). The electrochemical properties of the catalysts were investigated by half-cell tests.

Alkaline environment offers the increased electrocatalystic activity and stability of non-precious metal group catalysts for the oxygen reduction reaction. However, sluggish hydrogen oxidation reaction (HOR) activity of electrocatalysts under alkaline environment has remained an issue to overcome. In this presentation, we report the hydrogen oxidation inhibition by organic cation adsoprtion to expalin the low HOR activity of Pt based electrocatalyst. The HOR activity of Pt based electrocatalyst in several organic cationic solutions including tetramethylammonium, tetrabuthylammonium, and tetrabutylphosphonium is investigated using rotating disk electrodes. Time-dependent and potential driven adsorption behaviors of the cationic groups suggest that this adsorption has rather complex chemisorption nature rather than just simple adsorption. The new proposed inhibition mechanism may give useful insites on how to design ionomeric binders and electrocatalysts for advanced alkaline membrane fuel cells.

It has long been recognized that the reaction rates of the hydrogen oxidation and hydrogen evolution reactions (HOR and HER) are slower in basic than acidic electrolytes, even though the surface intermediate of adsorbed hydrogen is independent of solution pH. Understanding the root of this observation is critical to designing catalysts for a multitude of electrochemical reactions with relevance to energy conversion and storage. In this work, we undertake a fundamental investigation relating interfacial water structure to the reaction barrier for hydrogen underpotential adsorption and to the HER/HOR kinetics. The pH-dependence of HOR/HER kinetics has been attributed to a variety of reasons. The research groups of Gasteiger and Yan have both suggested that in base, hydroxide ions stabilize the Pt-H bond for stronger binding and slower catalysis^1,2. Central to this hypothesis is an experimentally measured shift with pH in the peak potential for hydrogen underpotential deposition on the (100) and (110) steps on the Pt surface. Although it is unclear why pH would stabilize adsorbed hydrogen, stronger binding would push platinum further to the right of the volcano peak and decrease the overall activity. Markovic et al proposed instead that in base, the water recombination/dissociation step of HOR/HER requires specific adsorption of hydroxide ions and therefore optimal binding to two adsorbates³. One observation that has been less discussed than either hydrogen or hydroxide adsorption is the effect of pH not only on the apparent hydrogen adsorption energy, but the kinetic barrier to adsorption. Koper et al attributed this observation to a pH-dependent water orientation at the surface, and hypothesized that "H-down" in acid vs "H-up" in base resulted in slower transfer of hydrogen to and from the surface⁴. The same pH-dependence of water orientation was recently supported computationally by Rossmeisl's group⁵.

 In this work, we examine specifically the hypothesis that water orientation governs the rate of hydrogen adsorption and thus the overall HER/HOR kinetics. The native oxide formed on chromium metal is known from colloidal literature to have a potential of zero charge (Epzc) of 1.22V vs. RHE⁶. We hypothesize that if water orientation truly governs the alkaline HOR/HER rate, the negative oxide surface charge on chromium oxide will induce an H-down water orientation adjacent to platinum nanoparticles supported on the surface. This will result in faster hydrogen adsorption kinetics and thus, higher activity for alkaline HOR/HER than extended platinum surfaces. To test this hypothesis, platinum nanoparticles were synthesized via pulsed electrodeposition on chromium oxide. The nanoparticle loading was controlled by the duration of deposition pulse, with longer times corresponding to higher coverage as well as larger particles. The hydrogen adsorption kinetics were measured by cyclic voltammetry in the hydrogen underpotential deposition region. Varying the sweep rate results in greater peak separation that can be used to extract rate constants for the adsorption reaction via traditional electroanalysis⁷. A typical voltammogram of Pt/CrOx in 0.1M KOH is shown in Fig. 1a. As the scan rate increases, kinetic barriers cause greater peak separation (dEp) for hydrogen adsorption on the 100 and 110 facets of platinum. For lower loadings with shorter pulse periods, both smaller currents and smaller peak separations are observed due to the lower Pt surface area. The effects of surface area can be accounted for by normalizing current to Pt surface area as measured by Hupd coulometry. The resulting surface-adsorption Tafel plot is shown in Fig 1b. The peak potential separation dEp is much greater for the larger nanoparticles with higher loading. This difference suggests that smaller particles have qualitatively more rapid adsorption kinetics due to the advantageous water orientation induced by negatively charged CrOx. The results of this study contribute to resolving a long-standing paradox in electrocatalysis and surface science by highlighting the importance of kinetic barriers, as well as adsorption energies, and suggest the ability to modulate alkaline hydrogen activity through metal-support interactions. Future work will discuss the relation of hydrogen underpotential deposition kinetics to HER/HOR activity and the effects of different oxide supports.

(1) Durst et al. Energy Environ. Sci.2014, 2255.

(2) Sheng et al. Nat. Commun.2015, 5848.

(3) Strmcnik et al, Nat. Chem.2013, 1.

(4) Van Der Niet et al. Catal. Today2013, 105.

(5) Rossmeis et al. Phys. Chem. Chem. Phys.2013, 10321.

(6) McCafferty, E. Electrochim. Acta2010, 1630.

(7) Angerstein-Kozlowska et al. J. Electroanal. Chem. Interfacial Electrochem.1979, 1.

Figure 1

Anion Exchange Membrane Fuel Cells (AEMFCs) are fast growing alternative to Proton Exchange Fuel Cell (PEMFC) technology. It was shown that several classes of non-platinum group metal (non-PGM) catalysts outperformed platinum in the oxygen reduction in alkaline media.[1] In contrast the published data on alternative to platinum anodes for hydrogen oxidation reaction in alkaline media is scarce.[2]One of the possible non-PGM catalysts for the HOR in alkaline media are Ni- and Pd-based materials. We have prepared Ni-Mo-Cu catalysts that shows HOR activities in the rotating disk electrode (Figure 1). The preliminary results on MEA tests show that the Ni-based alloys as an anode have shown extremely high open-circuit voltages (OCVs) in both oxygen and air (up to 1.1V), which is significantly higher compared to PGM-free anodes in PEMFCs. However, the effort in integration of those materials into MEA with most appropriate ionomer and MEA fabrication technique is needed. In this talk, we will discuss the preparation and characterization of the novel electrocatalysts for HOR in alkaline media.

Figure 1. HOR on Ni-Mo-Cu catalysts in 1M KOH saturated with H₂, 1600RPM, 10 mV s^-1.

Acknowledgements: Department of Energy, Hydrogen Oxidation Reaction in Alkaline Media, Control Number: 0966-1624, Award Number: DE-EE0006962 (PI A. Serov).

[1] M. H. Robson, A. Serov, K. Artyushkova, P. Atanassov, Electrochim. Acta, 90, 2013, 656–665.

[2] A. Serov, C. Kwak, Applied Catalysis B: Environmental, 91, 2009, 1–10.

Figure 1

Electrochemical energy productions are considered as an alternative energy sources because those energy consumptions are recognized to be more environmentally friendly and sustainable. The fuel cells, especially the direct methanol fuel cells (DMFCs), have received a lot of attention due to the higher energy density (5.04 KWh/L), easy storage and transportation compared with the hydrogen fuel cells (0.53 KWh/L). Also the direct conversion of methanol has a voltage similar (2CH₃OH+3O₂→2CO₂+4H₂O+6e^-, E=1.19 V) to that of hydrogen (E=1.23 V). However, the DMFCs still suffer from three problems on the anode side; namely, i) the carbon monoxide (CO) poisoning of the platinum nanoparticles (Pt-NPs), which is generated from the uncompleted methanol oxidation reaction (MOR), ii) the sluggish methanol oxidation reaction (MOR) compared to the hydrogen oxidation reaction (HOR) and iii) the low durability of the electrocatalyst in terms of Pt stability and carbon corrosion, which degrades the fuel cell performance.

To commercialize the direct methanol fuel cells (DMFCs), the durability of the anodic electrocatalyst needs to be highly considered, especially under high temperature and methanol concentration conditions. The low durability caused by the carbon corrosion as well as the carbon monoxide (CO) poisoning of the platinum nanoparticles (Pt-NP) leading to the decrease of the active Pt-NPs and increase in the inactive Pt-NPs covered by CO species. We previously reported that poly(vinylphosphonic acid) (PVPA) plays an important role in enhanced CO tolerance of the electrocatalyst due to the acceleration of the reaction between Pt and H₂O to form the Pt(OH)_ads that consumes the CO poisoned Pt, namely, Pt(CO)_ads. ¹ In this study, we deposited Pt-NPs on poly[2,2'-(2,6-pyridine)-5,5'-bibenzimidazole] (PyPBI)-wrapped nanoporous carbon (NanoPC)^2,3) and coated the as-synthesized electrocatalyst with poly(vinylphosphonic acid) (PVPA). The durability of the as-synthesized NanoPC/PyPBI/Pt/PVPA was tested in 0.1M HClO₄ electrolyte at 60 ^oC by cycling the potential from 1.0 to 1.5 V vs. RHE, and the results indicated that the NanoPC/PyPBI/Pt/PVPA showed an ~5 times better durability by comparison with that the commercial CB/Pt. The methanol oxidation reaction (MOR) of the electrocatalyst was tested before and after the potential cycling in the presence of 4M or 8M methanol at 60 ^oC and found that the CO-tolerance of the electrocatalyst was ~3 times higher than that of the commercial CB/Pt. Such a higher CO tolerance is due to the coating of the PVPA, which was proved by an EDX mapping measurement. The NanoPC/PyPBI/Pt/PVPA showed a high durability and CO tolerance under high temperature and high methanol concentration conditions indicating that the electrocatalyst would be used in real fuel applications.

References

1) N. Nakashima et al., J. Mater. Chem. A 2014,2, 18875-18880.

2) N. Nakashima et al., ACS Appl. Mater. Interfaces, 2016, in press.

3) N. Nakashima et al., ACS Appl. Mater. Interfaces, 2015, 7, 9800-9806.

Alkaline fuel cells have some advantages as compared with conventional acid fuel cells. Our research group has investigated the possibility of alkaline fuel cells using anion-exchange membrane as an electrolyte. In the presentation, I would like to introduce and discuss about some topics relating to alkaline fuel cells as follows:

1) Direct oxidation fuels: alcohols

We have been focusing on ethylene glycol as a direct fuel in anion-exchange membrane fuel cells (AEMFCs). Ethylene glycol has superior energy density (7.56 kWh dm⁻³) and higher boiling point (471 K) than some typical alcohol fuels such as methanol and ethanol. The oxidation of ethylene glycol in alkaline medium is faster than that in acid medium. And, surprisingly, ethylene glycol provides larger oxidation currents than methanol and polyols such as glycerol, erythritol, and xylitol in alkaline solutions. Thus, ethylene glycol is a promising fuel for AEMFCs, which overcomes a conventional direct methanol fuel cell (DMFC).

Ethylene glycol has the above-mentioned advantageous features as DAFCs' fuel, but it has a crucial problem derived from its inherent molecular structure. Ethylene glycol is a C2 molecule having a C−C bond in its structure. C−C bond is a stumbling block for electroorganic chemists since thus C−C bond is comparatively strong and cannot be easily cleaved. As well as ethanol oxidation, ordinal Pt catalyst cannot achieve the complete oxidation of ethylene glycol, and leaves some intermediate products. Therefore, in order to increase the efficiency of fuel utilization, active catalysts, having sufficient ability to break C−C bond in ethylene glycol, are strongly required. To our best knowledge, there is no available literature concerning the oxidation of ethylene glycol to CO₂from an electrocatalytical view point. Here we introduce an effective C−C bond cleavage electrocatalyst for ethylene glycol.

2) Oxygen reduction catalysts: non-precious metal catalysts

Electrochemically reduction of oxygen is an essential reaction involving in fuel cells. Among the various kinds of electrocatalysts, Pt, Pt-alloy, Ag are known to be fairly active for oxygen reduction in alkaline solutions. Precious metals, however, have some serious problems to be overcome before worldwide spread of fuel cells, such as economical cost, reproducibility and limited resources. These problems motivated many researchers to study for alternative catalysts besides a standard catalyst of Pt/C. Among proposed alternative catalysts, perovskite-type oxide is one of the promising candidates as a non-precious metal catalyst. Since Meadowcroft pointed out the catalytic activity of lanthanum-cobalt oxide for oxygen reduction, perovskite-type oxides become attractive materials as electrocatalysts for many electrochemists. However, the detailed mechanism for oxygen reduction on perovskite-type oxide electrodes has not been fully clear yet. We investigated the oxygen reduction on perovskite-type oxides using thin film electrodes and single crystal electrodes, and shed lights on the catalytic roles of oxides and carbon additives.

Small alcohol molecules such as methanol and ethanol represent promising fuels to power fuel cells for mobile electronic devices, transportation and stationary applications. Ethanol is of particular interest due to its renewability by generation from biomass feedstocks, large energy density and compatibility with existing infrastructures for fuel storage and delivery. Although substantial progress has been made in the recent years for the development of direct ethanol fuel cell (DEFC) technology, obstacles are still present for the practical implication, largely in the lack of efficient and selective electrocatalysts for complete oxidation of ethanol to CO₂.

Platinum (Pt) is commonly used as electrocatalyst in fuel cells, including DEFCs. However, ethanol oxidation on Pt is found to be incomplete and produces undesired partial oxidation products such as acetaldehyde and acetic acid. It is thereby considered that Pt itself cannot cleave the C-C bond in ethanol molecules and a second transition metal such as Rh and Ir is needed to facilitate this rate- and selectivity-limiting step. However, evidences are also present in the literature for the formation of strongly binding adsorbates, such as -CO and -CH_x, at low potentials (<0.5 V vs. RHE), from both in situ molecular spectroscopy (e.g., Infrared (IR) and surface enhanced Raman spectroscopy (SERS)) and differential electrochemical mass spectroscopy (DEMS) studies. These results suggest that oxidative removal of these C₁ species, instead of C-C bond cleavage, is the sole rate-limiting factor on Pt catalysts.

In attempt to resolve this controversy, we have performed systematic studies of methanol, ethanol and ethylene glycol oxidation on polycrystalline Pt electrodes. In particular, surface-specific sum frequency generation (SFG) spectroscopy is combined with electrochemistry to determine the reaction intermediates and evaluate their dependence on electrode potentials. It is found that C₁ adsorbates form at low potentials, and the coverages increase as the electrode potential is raised from 0.1 V to ~0.4 V. While the -CO feature drops above 0.4 V for methanol and ethylene glycol oxidation, it persists up to 0.5 V and then drops for ethanol oxidation. Our results confirm the presence of active Pt sites for C-C bond cleavage and underline the role of -CH_x species generated from the β-carbon in governing the kinetics of alcohol oxidation reactions. It is proposed that the slow transition from -CH_x to -CO gives rise to the higher overpotentials for ethanol oxidation compared to methanol and ethylene glycol oxidation.

In this work, novel Pd-(metal oxide)/C electrocatalysts were synthesized and evaluated as anodes for the oxidation of ethanol, methanol, ethylene glycol, glycerol and isopropanol in alkaline electrolyte. The metal oxides proposed to enhance the catalytic activity of Pd nanoparticles were Fe₃O₄, Fe₂O₃ and cerium oxide nanorods (CeO_2-NR) pyrolized at 200 and 400 °C. The Pd-(metal oxide)/C anodes, as well as a Pd/C electrocatalyst as comparison, were synthesized using NaBH₄ as reducing agent and Vulcan XC-72 as the support. The electrochemical characterization revealed a higher performance of the Pd-(metal oxide)/C electrocatalysts compared to Pd/C for all the oxidation reactions, at several concentrations of the fuels. Among the novel electrocatalysts, that containing CeO_2-NR thermally treated at 400 °C (Pd-CeO_2-NR400/C) showed the highest catalytic activity. Only in the case of the Ethylene Glycol Oxidation Reaction, Pd-Fe₂O₃/C showed a performance as high as that of Pd-CeO_2-NR400/C.

It is of fundamental importance to understand the factors controlling trends in activity for electrocatalytic reactions as a function of pH. In the case of the oxygen reduction reaction, numerous reports suggest significant divergences between noble metals surface catalytic performances in acid and base.^[1,2]

In our earlier studies, we mapped out the experimental Sabatier volcano for the oxygen reduction reaction in 0.1 M HClO₄ using the Cu/Pt(111) near-surface alloy system, see Figure 1 for near-surface alloy schematic.^[3,4]

In this study, as those of ^[3,4], we found that by changing the subsurface coverage of Cu we could tune the surface binding of the key reaction intermediate, OH; we thus monitored the OH binding energy shift through the observable shifts in the base voltammograms in both acidic and alkaline media.

Further, we elucidate the experimental oxygen reduction volcano in 0.1 M KOH for the Cu/Pt(111) near-surface alloy system. Remarkably, we observe that the same trend persists between OH binding shifts and Cu/Pt(111) oxygen reduction activities between acid and alkaline electrolyte, with the optimum catalyst in alkaline exhibiting an 8-fold improvement in activity, relative to Pt(111). However, all surfaces show a ~4 fold improvement in activity in 0.1 M KOH, relative to the same surface in 0.1 M HClO₄. At the peak of the volcano the surface exhibits an exceptionally high specific activity of 90 mA/cm² at 0.9 V with respect to the reversible hydrogen electrode. Thus, our results confirm that OH binding energy is the key descriptor in both alkaline and acid electrolytes.

[1] R. Rizo, E. Herrero, J. M. Feliu, Phys. Chem. Chem. Phys. 2013, 15, 15416–25.

[2] J. Staszak-Jirkovský, R. Subbaraman, D. Strmcnik, K. L. Harrison, C. E. Diesendruck, R. Assary, O. Frank, L. Kobr, G. K. H. Wiberg, B. Genorio, et al., ACS Catal. 2015, 5, 6600–6607.

[3] I. E. L. Stephens, A. S. Bondarenko, F. J. Perez-alonso, F. Calle-vallejo, L. Bech, T. P. Johansson, A. K. Jepsen, R. Frydendal, B. P. Knudsen, J. Rossmeisl, et al., 2011, 5485–5491.

[4] A. S. Bondarenko, I. E. L. Stephens, I. Chorkendorff, Electrochem. commun. 2012, 23, 33–36.

Figure 1

a, Key Laboratory of Material Chemistry for Energy Conversion and Storage (Huazhong University of Science and Technology), Ministry of Education, Hubei Key Laboratory of Material Chemistry and Service Failure, School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan, 430074, P.R. China. *Email: wangdl81125@hust.edu.cn

b, Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY, USA

Graphene, a single-atom-thick hexagonally arrayed sp² carbon atoms bonded sheet, has attracted tremendous interest in renewable energy conversion and storage devices due to its extraordinary properties, especially ultrahigh specific surface area and predominant electron conductivity¹. However, graphene based two-dimensional catalysts suffer from an irreversible re-stacking of the nanosheets due to the strong π-interaction during thermal annealing process and electrochemical measurements which would result in a lower surface area, limiting the mass transfer rate and even decreasing the catalytic activity as well as electrochemical stability.

Figure 1 (a) TEM image of NSGCB; (b) ORR activity of NSGCB and the corresponding comparative catalysts; (c) TEM image of NSCNT-3; (d) ORR activity of NSCNT-3 and the corresponding comparative catalysts.

Herein, we report two strategies for the synthesis of nitrogen and sulfur co-doped three-dimensional graphene intercalated structure nanomaterials as electro-catalysts for ORR in alkaline medium. One is the insertion of Vulcan XC-72 carbon spheres into the graphene layer, forming a graphene-Vulcan XC-72- grapheme "sandwich-like" structure². The other way is partially unzipping of multi-walled carbon nanotubes to form a graphene nanoribbon-carbon nanotube-graphene nanoribbon structure. Both of the prepared three-dimensional structured nanocomposites were experienced to nitrogen and sulfur co-doping process, and moreover, the resulting doped electro-catalysts exhibited outstanding ORR performance.

Acknowledgements

This work was supported by the National Natural Science Foundation (21306060, 21573083), the Program for New Century Excellent Talents in Universities of China (NCET-13-0237), the Doctoral Fund of Ministry of Education of China (20130142120039).

References

(1) Li, Y., Zhao, Y., Cheng, H., Hu, Y., Shi, G., Dai, L., & Qu, L. Nitrogen-doped graphene quantum dots with oxygen-rich functional groups. Journal of the American Chemical Society, 2011, 134, 15-18.

(2) Wu, M., Wang, J., Wu, Z., Xin, H. L., & Wang, D. Synergistic enhancement of nitrogen and sulfur co-doped graphene with carbon nanosphere insertion for the electrocatalytic oxygen reduction reaction. Journal of Materials Chemistry A, 2015, 3, 7727-7731.

Figure 1

Carbon-based electrodes are known to undergo corrosion in acidic and alkaline media during the oxygen evolution reaction (OER), yet studies on the efficacy of OER catalysts frequently employ carbon as catalyst support. Carbon remains a material of interest in the development of bifunctional air electrodes, particularly due to its utility in supporting efficient oxygen reduction reaction (ORR) catalysis. Methods to quantify carbon corrosion rates are needed in order to assess the practical utility of carbonaceous materials in bifunctional air electrodes and other OER electrodes. In-situ electrochemical mass spectrometry enables direct and precise assessment of carbon corrosion in acidic media by quantifying gaseous CO₂ that is generated as a product of the corrosion reaction. For alkaline media, this in-situ approach is unsuitable since CO₂ generated during corrosion reacts with and is sequestered within the alkaline electrolyte as carbonate.

In this presentation, we describe a method to precisely quantify carbon corrosion in alkaline media. The method is direct and sufficiently sensitive to provide rapid (1-2 days) characterization of carbon corrosion, whereas methods based on electrochemical characterization or mass measurement may yield ambiguous results and take months. We report on the use of this method to quantitatively compare the rate of corrosion using different types of carbon, including reduced graphene oxide, doped carbon and carbons with various coatings and catalysts. A critical assessment of the use of carbon in bifunctional air electrodes will be provided.

In Pt-transition metal (TM) alloy nanoparticles for proton exchange membrane fuel cells (PEMFCs), the TM surface can be easily oxidized by air, water, and acidic electrolytes. Accordingly, the electron transfer from the TM to Pt is retarded owing to the electronegative O species on the TM surface, and the dissolution of the surface TM oxide is accelerated during the fuel cell operation, which results in catalyst degradation. Therefore, we propose a novel hybrid catalyst concept that selectively modifies the surface Co atoms with N-containing polymers, resulting in highly active and durable PtCo nanoparticles useful for the oxygen reduction reaction (ORR). The N-containing polymer functionalized on carbon black selectively interacts with the Co precursor, resulting in Co-N bond formation on the PtCo nanoparticle surface. Electron transfer from Co to Pt in the PtCo nanoparticles is therefore enhanced by the increase in the difference in electronegativity between Pt and Co. In addition, the dissolution of Co and Pt is prevented by the selective passivation of surface Co atoms and the decrease in the O binding energy of surface Pt atoms, respectively. As a result, the activity and durability of organic-inorganic hybrid PtCo nanocatalysts for the ORR are significantly enhanced by the electronic ensemble effects.

Demand on the practical synthetic approach to the high performance electrocatalyst is rapidly increasing for the fuel cell commercialization. However, sluggish kinetics hindered high performance. Therefore methods to improve the activity of catalysts that facilitate reaction kinetic have been and continue to be a popular area of research. Since electrocatalytic reactions occur on the catalyst surface, nanoparticles (NPs) with high surface area are promising candidates to overcome problems. However, their low stability at harsh operating condition impedes nanoparticle utilization. Here we present a synthesis of highly durable and active nanoparticle based electrocatalyst for fuel cell electrocatalyst. One example is ordered intermetallic face-centered tetragonal (fct)-PtFe NPs encapsulated by thin-layer carbon shell. Only few nanometers of carbon shells which is in situ formed from dopamine coating could effectively prevent the coalescence of NPs. This carbon shell also protects the NPs from detachment and agglomeration as well as dissolution throughout the harsh fuel cell operating conditions. The other is carbon encapsulated metal phosphide NPs for robust hydrogen evolution reaction. Anti-oxidative carbon shell prevent surface oxidation of NPs, leading exceptionally long-term durability. We believe that thin-layer carbon shell encapsulation of NPs can open a new possibility for the development of highly active and durable electrocatalyst in near future.

The progress in the ORR electrocatalysis over the last two decades has been achieved primarily through the fundamental understanding of processes that are taking place at well-defined surfaces. For instance, the nature of active sites is a common topic in the literature, however, despite decades of persistent studies aimed to reveal the fundamentals, insight at atomic level is still lacking. Properties such as surface crystallographic orientation, morphology, composition and defects are yet to be assigned to the correlation between the atomic structure and catalyst activity. For the first time, we report on the atomic structure that has been investigated in combination with durability. The ultimate precision that goes beyond a part per million of a single atomic layer has been achieved in determining electrocatalytic properties of the ORR catalysts. Obtained knowledge is of paramount importance in design of advanced highly functional nanoscale materials. Surfaces of Pt-based materials have been characterized by AES, LEED and UPS, which was followed by controlled transfer to electrochemical, in-situ FTIR and STM cells. These findings have been further used to optimize a unique RDE coupled ICP-MS system. Such effort has led towards the design and synthesis of nanoscale materials with superior electrocatalytic properties. Fine tuning of the surface properties induced unprecedented improvements in functionality of real world catalyst for the ORR in fuel cells that will be reported for the first time.

In case of polymer electrolyte fuel cells (PEFCs) Platinum is still the most used catalyst for the oxygen reduction reaction (ORR) at the cathode – where PtNi alloys are the state of the art [1, 2]. Following a study by Stamenkovic et al. in 2007 [3] several groups dealt with octahedral shaped nanoparticles (NPs) during the last decade to enhance both specific and mass activity [4]. Well-defined octahedral shaped NPs can be obtained e.g. by solvothermal synthesis using DMF as solvent and reduction agent [5]. Unfortunately, such PtNi octahedra show a major disadvantage in lacking of morphological stability during electrochemical cycling [6]. In the present work, we combine the DMF-based synthesis with a thermal post-treatment as annealing step toward an improved PtNi alloy structure with higher stability. The study is primarily based on the morphological investigations by transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM) combined with energy dispersive X-ray analysis (EDX), both ex situ as well as in situ mode, to gain insights into the structure-activity-stability relationship.

In detail, based on a previous work [7], we exposed octahedral PtNi particles (supported on carbon) to a post-treatment in hydrogen using a tubular furnace. First off all, an increased annealing temperature results in a better defined PtNi alloy – proven by XRD. Whereas after 300°C mainly octahedral NPs were present, a complete loss of shape was observed after annealing at 500°C, where nearly spherical particles resulted. Additionally, the morphological changes during heating were also observed in situ using a TEM heating nano-chip (i.e. heating in vacuum). Both annealing in hydrogen as well as in vacuum show indications for Pt diffusion to (111) facets before the octahedral shape collapsed. Indeed, a rather mild treatment in hydrogen can improve the alloy structure and, thus, lead to higher activity in ORR, whereas a complete PtNi alloying at higher temperatures is diminishing the ORR activity due to the loss of octahedral shape and, accordingly, (111) facets.

References

[1] Y. Bing et al., Chem. Soc. Rev 39 (2010) 2184-2202.

[2] H. A. Gasteiger, N. M. Marković, Science 324 (2009) 3791.

[3] V. R. Stamenkovic et al., Science 315 (2007) 493.

[4] P. Strasser, Science349, 379-380 (2015).

[5] M. K. Carpenter et al., J. Am. Chem. Soc. 134 (2012) 8535-8542.

[6] L. Gan et al., Science 346 (6216) (2014) 1502-1506.

[7] M. Ahmadi et al., ACS Nano 7 (2013) 9195-9204.

The oxygen reduction reaction (ORR) is a corner stone for energy conversion techniques such as fuel cells and metal-air batteries[1]. The use of Pt unfortunately generate a number of obstacles including the high cost, scarcity and readily poisoning by impurity and the cross-over of fuel⁹. Shaped bimetallic nanocrystals (NCs) with precisely controlled structure and composition at the atomic level, e.g. Pt_xNi_1−x alloy NCs[2], intermetallic PtCo core–shell nanoparticles[3] and Pt₃Ni nanoframes[4], represent a class of exciting and promising ORR catalysts.

Herein, we demonstrate a one-pot protocol for the controlled synthesis of icosahedral Pt₃M nanocrystal with twenty active Pt₃M (111) facets using glucosamine as both reducing and structure-directing agent. Among the investigated catalysts, the Pt₃Ni icosahedral with nano-segregated Pt-skin displays the unexpected 32-fold enhancement in specific activity and 12-fold improvement in mass activity relative to state-of-the-art Pt/C catalyst. Besides, the catalyst was found ultra-robust during 20,000 potential cycling tests. The Pt₃Ni icosahedral thus represents one class of the most efficient ORR electrocatalysts ever reported and demonstrates an effect technique in mimicing the active extended surface in nano-scale.

Fig.1 (a) TEM image of the typical icosahedral Pt₃Ni nanocrystal, (b) HRTEM micrograph showing the twin boundaries of five-fold symmetry, (c) the (111) and (200) planes, (d)cyclic voltammetry curves of Pt/C, Pt NI/C and Pt₃Ni/C catalysts, (e) ORR polarization curves for Pt/C, Pt NI/C and Pt₃Ni/C catalysts in O₂-saturated 0.1M HClO₄, (f) comparative ORR activities of Pt/C and Pt₃Ni/C catalysts before and after 20,000 potential cycles.

Acknowledgments

The study was financed by the National Basic Research Program of China (973 Program, 2012CB215500), the National Natural Science Foundation of China (21373199, 21433003) and the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA09030104). 

References

[1] Stamenkovic, V.R., B. Fowler, B.S. Mun, G. Wang, P.N. Ross, C.A. Lucas, and N.M. Marković, Science, 315(5811), 493-497,(2007).

[2] Zhang, C., S.Y. Hwang, A. Trout, and Z. Peng, Journal of the American Chemical Society, 136(22), 7805-7808,(2014).

[3] Wang, D., H.L. Xin, R. Hovden, H. Wang, Y. Yu, D.A. Muller, F.J. DiSalvo, and H.D. Abruña, Nat Mater, 12(1), 81-87,(2013).

[4] Chen, C., Y. Kang, Z. Huo, Z. Zhu, W. Huang, H.L. Xin, J.D. Snyder, D. Li, J.A. Herron, M. Mavrikakis, M. Chi, K.L. More, Y. Li, N.M. Markovic, G.A. Somorjai, P. Yang, and V.R. Stamenkovic, Science, 343(6177), 1339-1343,(2014).

Figure 1

Improving design and/or reducing noble metal content in electrocatalysts for fuel cell electrodes while maintaining and/or increasing proton exchange membrane fuel cell (PEMFC) performance in terms of durability and power density are crucial challenges for PEMFC mass market applications. A possible way consists in combining noble metals (Pt, Pd, Au ...) and non-noble metals for preparing binary and ternary nanocatalysts.

The formation of platinum alloys with transition metals such as Ni, Co, Fe, Cr represents a promising way for improving the activity of Pt-based catalysts [1-3].  Pt based alloys have indeed often demonstrated higher electrocatalytic activity towards ORR than pure platinum.

However, while non-noble metals are interesting from a cost reduction point of view, their presence may involve lower stability of the catalyst than that of pure platinum due to their dissolution [4.5], which leads to a loss of performances in PEMFC. Au addition was proposed to improve durability of the nanocatalysts [6-9].

On the other hand, the atomic structure and morphology [10] also play very important roles in the electrocatalytic efficiency for ORR.

In this context, the present study aims at systematically comparing the catalytic activity and the selectivity towards the ORR of binary and ternary catalysts based on Pt, Ni, Co and Au in order to determine the best atomic ratio for this reaction, in terms of kinetics current density and of number of exchanged electrons per oxygen molecule reduced. For this purpose, monometallic (Pt/C), binary (Pt_xNi_10-x/C and Pt_xCo_10-x/C) and ternary (Pt_xAu_yNi_z/C and Pt_xAu_yCo_z/C) nanocatalysts supported on carbon Vulcan XC-72 have been first synthesized by a wet chemical method and comprehensively characterized. The morphologies, compositions and structures of the particles were characterized by physical methods (transmission electron microscopy, X-ray diffraction and atomic absorption), whereas metal loadings on the carbon support was determined by thermogravimetric analysis. Electrochemical active surface areas and surface compositions were estimated by cyclic voltammetry. Electrocatalytic activity, selectivity and durability of catalysts were studied by the rotating disc (Figure 1) and rotating ring disc electrodes.

After having determined the best atomic ratios for the different metals, i. e. Pt₇Ni₃, Pt₆Ni₂Au₂/C and Pt₅Ni_1.7Au_3.3/C, plasma sputtering method was use to prepare alloyed Pt_xAu_yNi_z/C and Au_y@Pt_xNi_z core-shell electrodes with ultra-low metal loadings (10 µg_metal cm^-2). This method is indeed very convenient for controlling the materials loading, composition and structure by controlling the physical parameters of deposition. The electrocatalytic behavior of these different electrode structures was evaluated and compared to that of chemical catalysts.

Acknowledgement:

The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) for the Fuel Cells and Hydrogen Joint Technology Initiative under grant agreement #325327 (SMARTCat project). 

Reference

[1] M.T. Paffet, J.G. Beery, S. Gottesfeld, J. Electrochem. Soc. 135 (1988) 1431–1436.

[2] C. Wang, M. Chi, D. Li, D. van der Vliet, G. Wang, Q. Lin, J. F. Mitchell, K. L. More, N. M. Markovic, V. R. Stamenkovic, ACS Catal. 1 (2011) 1355–1359.

[3] T. Toda, H. Igarashi, H. Uchida, M. Watanabe, J. Electrochem. Soc. 146 (1999) 3750–3756.

[4] M. Watanabe, K. Tsurumi, T. Mizukami, T. Nakamura, P. Stonehart, J. Electrochem. Soc. 141 (1994) 2659–2668.

[5] C.F. Yu, S. Koh, J.E. Leisch, M.F. Toney, P. Strasser, Faraday Discuss. 140 (2009)

283–296.

[6] D.F. Yancey, E.V. Carino, R.M. Crooks, J. Am. Chem. Soc. 132 (2010) 10988–10989.

[7] J. Zhang, K. Sasaki, E. Sutter, R.R. Adzic, Science 315 (2007) 220–222.

[8] K. Sasaki, H. Naohara, Y. Cai, Y.M. Choi, P. Liu, M.B. Vukmirovic, J.X. Wang, R.R.

Adzic, Angew. Chem. Int. Ed. 49 (2010) 8602–8607.

[9] V. Tripkovic, H.A. Hansen, J. Rossmeisl, Tejs Vegge, Phys. Chem. Chem. Phys. 17 (2015) 11647–11657.

[10] D. Alloyeau, C. Mottet, C. Ricolleau, Nanoalloys, London: Springer-Verlag; 2012.

Figure 1

The amount of platinum (Pt) in the catalyst layer accounts for a significant portion of fuel cell cost and limits the commercial deployment of proton exchange membrane fuel cells.[1] Catalyst studies typically focus on the oxygen reduction reaction (ORR), since the reaction is orders of magnitude slower kinetically than hydrogen oxidation. Extended surface nanomaterials offer key advantages in ORR, including an order of magnitude higher site-specific activity, long range conductivity, and long term durability.[2] Traditionally, however, the catalyst type has been limited by low surface areas.

Spontaneous galvanic displacement occurs when a more noble metal cation contacts a less noble metal template, and combines aspects of corrosion and electrodeposition. In ORR, catalysts formed by spontaneous galvanic displacement are ideally situated, taking advantage of the specific activities generally associated with the catalyst type while significantly improving upon their surface area.[3] Galvanic displacement has been used to deposit small amounts of Pt onto extended templates, and in the case of Pt-nickel (Ni) nanowires, has produced materials with surface areas in excess of 90 m² g_Pt^‒1.[4] Post-synthesis processing has been used to improve the activity and durability of Pt-Ni nanowires in rotating disk electrode (RDE) half-cells. Thermal annealing integrated Pt-rich and Ni-rich zones, compressing the Pt lattice and improving ORR activity.[5] Acid leaching and oxidation of the Pt-Ni nanowires also removed surface Ni and formed Ni oxides near the nanowire surface, improving catalyst durability and reducing Ni dissolution in durability tests (30,000 cycles, 0.6‒1.0 V).

Recently, Pt-Ni nanowires were studied for their ORR activity in RDE half-cells using updated testing protocols. A rotational air drying method formed thin, uniform coatings onto the RDE working electrodes.[6] Thin catalyst ink dispersions also allowed for the ORR diffusion-limited current to be reached while maintaining a low Pt loading. Using these methods improved Pt-Ni nanowire activity in RDE half-cells by 60%. The optimized Pt-Ni nanowires produced an ORR mass activity 9 times greater than Pt/HSC, while losing less than 3% of that activity and less than 0.5% catalyst mass to dissolution in durability testing.

Atomic layer deposition (ALD) has also been used in an effort to replicate the Pt-Ni nanowire synthesis by galvanic displacement. Works has included different chemistries (oxygen and hydrogen routes), as well as efforts to disperse the nanowire template without the benefit of a liquid medium. Successful use of ALD can potentially produce catalysts on a much larger, more commercially viable, scale.

Figure 1. Surface areas (x-axis) and site-specific oxygen reduction activities at 0.95 V (y-axis) of Pt/HSC and Pt-Ni nanowires, as-synthesized and modified by post-synthesis processing. The solid black line denote constant mass activities of 100, 400, and 800 mA mg_Pt^‒1.

[1] D. Papageorgopoulos, in: U.S. Department of Energy (Ed.), http://www.hydrogen.energy.gov/pdfs/review14/fc000_papageorgopoulos_2014_o.pdf, 2014.

[2] M. Debe, in: U.S. Department of Energy (Ed.), http://www.hydrogen.energy.gov/pdfs/review09/fc_17_debe.pdf, 2009.

[3] S.M. Alia, Y.S. Yan, B.S. Pivovar, Catalysis Science & Technology, 4 (2014) 3589-3600.

[4] S.M. Alia, B.A. Larsen, S. Pylypenko, D.A. Cullen, D.R. Diercks, K.C. Neyerlin, S.S. Kocha, B.S. Pivovar, ACS Catalysis, 4 (2014) 1114-1119.

[5] B. Pivovar, Extended, Continuous Pt Nanostructures in Thick, Dispersed Electrodes, in: U.S. Department of Energy (Ed.), http://www.hydrogen.energy.gov/pdfs/review14/fc007_pivovar_2014_o.pdf, 2014.

[6] K. Shinozaki, J.W. Zack, R.M. Richards, B.S. Pivovar, S.S. Kocha, Journal of The Electrochemical Society, 162 (2015) F1144-F1158.

Figure 1

State-of-the-art polymer electrolyte fuel cells (PEFCs) require large amounts of carbon-supported platinum nanoparticle (Pt/C) catalysts (~ 0.4 mg_Pt/cm²_geometric) to account for the large overpotential of the oxygen reduction reaction (ORR).¹ These excessive Pt-loadings that impede widespread commercialization of PEFCs can be mitigated by increasing the catalysts' ORR activity, e.g. by alloying platinum with other metals like Ni, Cu and Co, to form materials which show up to one order of magnitude higher mass-specific activity than commercial Pt/C catalysts.² On the other hand, state-of-the art carbon-supported materials suffer from significant carbon and Pt corrosion during the normal operation of PEFCs, gradually compromising their performance. To partially overcome these stability issues, a lot of research effort is dedicated to the development of unsupported ORR catalysts. Among these materials, bimetallic alloy aerogels consisting of nanoparticles interconnected to nanochains³ present an interesting option, since their extended 3D structure should facilitate transfer to actual PEFC cathodes.

With this motivation in mind, we have modified a previously published synthesis³ in aqueous environment to prepare 'clean' bimetallic Pt-Ni aerogels with different stoichiometries, among which Pt₃Ni and Pt_1.5Ni were characterized in depth and tested for ORR activity. The alloy formation and extended 3D nanochain structure were confirmed by X-ray diffraction and transmission electron miscroscopy, respectively. Moreover, X-ray photoelectron and X-ray absorption spectroscopy indicated the existence of a separate Ni-(hydr)oxide phase in the Pt_1.5Ni aerogel that was not present in the Pt₃Ni sample. This hypothesis was further verified by complementing electrochemical measurements in alkaline electrolyte. Despite this difference in composition, both Pt-Ni aerogels showed comparable ORR activities in rotating disk electrode measurements in 0.1 M HClO₄ electrolyte, reaching the activity target of 440 A/g_Pt at 0.9 V_RHE set by the U.S. Department of Energy⁴ for automotive PEFCs. However, accelerated stress tests⁵ using the above mentioned RDE setup did not reveal significant durability improvements with respect to a commercial Pt/C benchmark catalyst. Consequently, these results were compared to those obtained in a differential PEFC setup to better understand the degradation behavior in an environment closer to real application.

In summary, this contribution reports the detailed investigation of surface and bulk properties of bimetallic Pt-Ni aerogels alongside with performance and stability assessments in aqueous electrolyte and a differential PEFC setup.

Acknowledgement

Funding from the Swiss National Science Foundation (20001E_151122/1), the German Research Foundation (EY 16/18-1) and the European Research Council (ERC AdG 2013 AEROCAT) is greatly acknowledged.

References

1. F. T. Wagner, B. Lakshmanan and M. F. Mathias, J. Phys. Chem. Lett., 1, 2204 (2010).

2. C. Wang, M. Chi, D. Li, D. Strmcnik, D. van der Vliet, G. Wang, V. Komanicky, K. C. Chang, A. P. Paulikas, D. Tripkovic, J. Pearson, K. L. More, N. M. Markovic and V. R. Stamenkovic, J. Am. Chem. Soc., 133, 14396 (2011).

3.  W. Liu, P. Rodriguez, L. Borchardt, A. Foelske, J. Yuan, A. K. Herrmann, D. Geiger, Z. Zheng, S. Kaskel, N. Gaponik, R. Kotz, T. J. Schmidt and A. Eychmüller, Angew. Chem. Int. Ed., 52, 9849 (2013).

4. D. Papageorgopoulos, http://www.hydrogen.energy.gov/pdfs/review15/fc000_papageorgopoulos_2015_o.pdf (accessed 09.03.2016).

5. F. Hasché, M. Oezaslan and P. Strasser, ChemCatChem, 3, 1805 (2011).

Platinum-coated nickel nanowire (PtNiNW) catalysts have shown high mass activity, specific activity, and exceptionally high surface areas in oxygen-reduction reaction (ORR) rotating disc-electrode experiments, all several times higher than platinum on carbon and well above DOE targets.^1,2 However, their subsequent incorporation into proton exchange membrane fuel cell (PEMFC) membrane electrode assemblies (MEA) has been exceedingly difficult, resulting in substantially lower mass and specific activities and lower than anticipated electrochemical areas (ECAs).² This presentation will focus on the challenges associated with the incorporation of extended surface Pt alloys in PEMFC MEAs, the ensuing conditioning protocols and the large discrepancy between RDE and MEA data. It is found that soaking the MEA in dilute sulfuric acid to remove nickel and increase the platinum weight fraction increases cell performance metrics such as open-circuit voltage, mass activity, and specific activity, as shown in Figure 1. Decreases in membrane resistance are also observed. The effects of ionomer, carbon and unsupported electrocatalyst ratios on MEA performance will also be discussed. Scanning transmission electron microscopy (STEM) and transmission x-ray microscopy (TXM) tomography will be examined as tools to elucidate the relationship between electrocatalyst/electrode structure and MEA performance.

1. Alia, S. M. et al. Platinum-Coated Nickel Nanowires as Oxygen-Reducing Electrocatalysts. ACS Catal.4,1114–1119 (2014).

2. Pivovar, B. S. Extended, Continuous Pt Nanostructures in Thick, Dispersed Electrodes. 2015 DOE Hydrogen and Fuel Cells Program Review (2015).

Figure 1

Proton Exchange Membrane Fuel Cell (PEMFC) technologies have focused much attention over the last decades as energy sources for both stationary and transportation applications. Their electrical performances have now reached a sufficient level for the deployment in specific markets. However, higher power densities are needed to decrease the size of the stack, and to reduce the total cost and/or to take into account the high degradation of the electrical performance during operation [1]. PEMFC performances are usually ruled out by the cathode side where the sluggish Oxygen Reduction Reaction (ORR) is processed. Moreover, the stability of the nanomaterials is also a concern in the harsh operating conditions of a PEMFC cathode [1].

The formation of hollow Pt/C particles was reported after degradation of Pt₃Co/C alloys in PEMFC on-site operation [2]. Having unveiled the enhanced ORR activity of such nanostructures, LEPMI recently developed a simple "one pot" method to synthesize hollow PtNi nanoparticles supported on high surface area carbon (PtNi/C). The best porous hollow PtNi/C nanocatalysts achieved 6-fold and 9-fold enhancement in mass and specific activity for the ORR, respectively over standard solid Pt/C nanocrystallites of the same size [3] (liquid electrolyte).

Upscaling the synthesis process is now a key-step to integrate these hollow nanoparticles into Membrane Electrode Assemblies (MEAs). Step-by-step development of the synthesis enabled to reach a 10-fold upscale of the synthesis corresponding to ca. 4 g of catalyst per batch (Figure 1). Combined physical and electrochemical characterizations were performed to ensure that the initial structure and the intrinsic properties of the hollow PtNi/C nanoparticles were maintained (Figure2). In parallel, different experimental conditions were also experimented to optimize the overall manufacturing process. This material upscaling also rendered possible the use of classical electrodes manufacturing processes for MEAs such as bar coating and screen printing, similarly to commercial catalysts. The first measurements performed in 25 cm² single cells demonstrated promising results for MEA using hollow PtNi/C at the cathode (Figure 3). Moreover, after an accelerated stress test designed to test the robustness of the metal nanoparticles [4], the MEA performance became higher than that of the Pt/C-based reference MEA. This is particularly true at high efficiency (up to medium current densities), which tends to confirm the interest of such nanostructures on the long-term. Complementary characterizations revealed that the utilization factor of the hollow catalyst is only 50%, and can be improved by optimizing the cathode catalyst layer formulation before integrating these new materials in larger MEAs and finally into representative PEMFC stacks.

References :

[1]: L. Dubau, L. Castanheira, F. Maillard, M. Chatenet, O. Lottin, G. Maranzana, J. Dillet, A. Lamibrac, J.-C. Perrin, E. Moukheiber, A. Elkaddouri, G. De Moor, C. Bas, L. Flandin, N. Caqué., Wiley Interdisciplinary Reviews: Energy and Environment,2014, 3, 540-560.

[2]: L. Dubau, J. Durst, F. Maillard, L. Guétaz, M. Chatenet, J. André, E. Rossinot, Electrochim. Acta 2011, 56, 10658-10667.

[3]: L. Dubau, T. Asset, R. Chattot, C. Bonnaud, V. Vanpeene, J. Nelayah, F. Maillard, ACS Catal. 2015, 5, 5333-5341

[4]: L. Castanheira, W.O. Silva, F.H. Lima, A. Crisci, L. Dubau, F. Maillard, ACS Catal. 2015, 5, 2184-2194

Figure 1

Proton exchange membrane fuel cells (PEMFCs) are characterized with many advantages including high power density, low operating temperature (~80 ^oC), quick start-up and quick match to shifting demands for power, and regarded as a very promising technology for powering transportation and stationary applications¹. However, the kinetics of the cathodic oxygen reduction reaction (ORR) in PEMFCs is sluggish and the most efficient material is platinum (Pt) that is rare on the earth and has a prohibitive cost. To make PEMFCs economically available, it is imperative to maximize the utilization of Pt or rather noble metal by improving catalyst layer's mass transport of reactants and products, and by increasing noble-metal-mass activities toward the ORR in cathode.

One of the strategies to improve mass transport in cathode is to thin the catalyst layer through developing high-content and high-dispersion Pt electrocatalysts. However, it is difficult to prepare Pt supported catalysts with a high Pt content because of the agglomeration of Pt nanoparticles. It was reported that the commercial Pt/VC catalysts had a particle size of 2.0 nm in Pt(10 wt%)/VC, 3.2 nm in Pt(30 wt%)/VC, and 8.8 nm in Pt(60 wt%)/VC². Although great efforts have been made to improve traditional methods, such as impregnation method and colloidal method, it remains very challenging to find an economical and controllable way to synthesize Pt/VC electrocatalysts with a high Pt content for the ORR in PEMFCs. This can be attributed to several facts, including the poor size control, the contaminants from stabilizer residue, tedious procedures and so on. In this work, we propose a simple and controllable approach to deposit highly-dispersed Pt nanoparticles on Vulcan XC-72 carbon black. The as-synthesized Pt/VC possesses a high Pt content as well as high-dispersion Pt NPs via controlling the nucleation and growth process, and was investigated as the cathodic catalyst in PEMFC.

Ideally highest Pt utilization with an extraordinary Pt-mass activity has been achieved by the Pt-monolayer-shell electrocatalysts which possess monoatomic thick Pt shells on noble-metal substrates³. However, the noble-metal-mass activities for these Pt-monolayer-shell electrocatalysts were also compromised, since the noble-metal substrates are quite essential for fabricating such electrocatalysts. In this regard, with the highest Pt utilization, improving the non-Pt noble-metal utilization in the substrates is critical for developing the Pt-monolayer-shell electrocatalysts. In this work, a surfactant-based, composition- and size-tunable method has been demonstrated to synthesize the monodispersed Pd-Ni as the substrates for fabricating the carbon-supported Pd-Ni@Pt nanospheres with Pt monolayer shells. And an excellent balance among activity, durability and noble-metal utilization was acquired.

Acknowledgements

This work was supported in part by National Natural Science Foundation of China (Grant No. 21373135 and 21533005) and Science Foundation of Ministry of Education of China ( Grant No. 413064).

Reference

1. M.L. Perry, F.T. Fuller, J Electrochem Soc, 149 S59-S67 (2002).

2. B. L. Gratiet, H. Remita, G. Picq, M. Delcourt, J Catal, 164, 36 (1996).

3. (a) J. Zhang, M. B. Vukmirovic, Y. Xu, M. Mavrikakis, R. R. Adzic, Angewandte Chemie International Edition, 44 2132-2135, (2005); (b) M. Shao, K. Shoemaker, A. Peles, K. Kaneko, L. Protsailo, J Am Chem Soc, 132 9253-9255, (2010). (c) J. X. Wang, H. Inada, L. Wu, Y. Zhu, Y. Choi, P. Liu, W. P. Zhou, R. R. Adzic, J Am Chem Soc, 131,17298( 2009) .

High-temperature polymer electrolyte fuel cells (HT-PEFCs) are a suitable technology for decentralized small scale electricity and heat production. HT-PEFCs do not require hydrogen infrastructure and are characterized by a simpler and therefore less expensive system compared to available fuel cell systems. In combination with a reformer unit, HT-PEFC systems offer an efficiency gain over conventional combustion of hydrocarbons, such as natural gas. In near future, HT-PEFCs will be even more effective and efficient when renewable biofuels and hydrogen are widespread available.

We present an efficient HT-PEFC based combined heat and power (µ-CHP) system for the provision of electrical energy and hot water in single family households (see Table 1). Due to the optimized design and layout of the fuel cell based µ-CHP system and the respective manufacturing processes of the catalysts and the membrane electrode assemblies (MEA), the demand for cost effective and greenhouse gas efficient energy at customer level is addressed.

Major efforts were devoted to the establishment of scalable catalyst deposition methods, which enable a loss-free utilization of precious metals. By using appropriate multimetallic catalyst systems at anode and cathode, the precious metal loading was reduced by approx. 20% in comparison to commercially available electrodes without compromising performance (see Figure 1) [1]. Furthermore, by introducing post-preparation treatments, the stability of the catalysts was enhanced over commercial Pt/C. The activity and stability of the catalyst systems were evaluated ex situ by means of cyclic voltammetry and accelerated stress tests using a rotating disk electrode (RDE) setup. Furthermore, the catalyst systems were characterized in situ by means of polarization curves, continuous operation, accelerated stress tests and electrochemical impedance spectroscopy measurements at single cell, stack and at system level (see Figure 1).

Acknowledgment

Financial support was provided by The Climate and Energy Fund of the Austrian Federal Government and The Austrian Research Promotion Agency (FFG) through the program Energieforschung (e!Mission).

[1] A. Schenk, C. Grimmer, M. Perchthaler, S. Weinberger, B. Pichler, C. Heinzl, C. Scheu, F.-A. Mautner, B. Bitschnau, V. Hacker, Platinum–cobalt catalysts for the oxygen reduction reaction in high temperature proton exchange membrane fuel cells – Long term behavior under ex-situ and in-situ conditions, J. Power Sources. 266 (2014) 313–322.

Figure 1

A hydrogen/bromine regenerative fuel cell involves complex water and heat transport phenomena during charge and discharge that significantly affect the degree of electrolyte dehydration, electrochemical reactions, proton transport, the concentrations of bromide complexes at the hydrogen electrode, etc. Therefore, developing innovative water and heat management schemes is critical to improve the uniformity of key distributions inside a cell as well as to enhance cell performance and durability. In this work, a three-dimensional, transient, two-phase, non-isothermal hydrogen/bromine fuel cell model is developed by rigorously accounting for two-phase transport and various heat generation mechanisms including irreversible heat, entropic heat, and Joule heating. The model is applied to a 25 cm² cell in order to precisely investigate water transport and thermal aspects under various charge and discharge conditions. A parallel computing methodology is employed to handle large-scale simulations involving millions of grid points. The large-scale simulations are able to provide extensive multidimensional contours of species concentration, temperature, and current density, assisting in identifying optimal water and thermal management strategies based on a typical geometry of a real-scale hydrogen/bromine cell.

One of the most important degradation mechanisms, which cause a reduction of lifetime and performance of high temperature (HT) polymer electrolyte membrane (PEM) fuel cells, is phosphoric acid loss. This paper presents the effect of different operation strategies on acid loss of HT-PEM FCs compared with fresh membrane electrode assemblies (MEA).

These investigations have been executed during the run-time of the European Project CISTEM (Construction of Improved HT-PEM MEAs and Stacks for Long Term Stable Modular CHP Units, GA-No. 325262). The vision of this project is the development of a new HT-PEM FC based CHP (combined heat and power) technology with high efficiency and long lifetime.

Degradation investigations of single components, MEAs, FC stacks and complete CHP units are main objectives within the project. With this paper we provide a more detailed insight to the phosphoric acid loss driven degradation process which causes shorter lifetimes and lower performances [1-2].

In dependence on the typical operational occurrences of the HT-PEM CHP-systems, the following operational strategies have been tested:

• Fuel switching between pure H₂ and synthetic reformate (Test 1 - grey)

• Start-stop-cycling with an idling temperature of 25 °C (Test 2 - red)

• Long term tests at constant current density (0.3 A/cm²) with different reactant gas compositions:

  • Synthetic reformate/pure O₂ (Test 3 - blue)

  • Synthetic wet reformate/O₂ enriched air (Test 4 - black)

All tests have been performed with Dapozol®-G55 MEAs, which consist of thermally cured polybenzimidazole (PBI) membranes and Pt/C based electrodes. During operation, the MEAs have been electrochemically characterized in-situ at Beginning of Life (BoL), once per week and End of Test (EoT) via polarization curves, electrochemical impedance spectroscopy (EIS), cyclic voltammetry (CV) and linear sweep voltammetry (LSV).

In addition, product water samples have been weekly collected from both cathode and anode gas outlets, and their acid content has been determined with ion chromatography (IC). The results of the total acid leaching in product water for every operation strategy are shown in Figure 1. After EoT, the remaining H₃PO₄within the tested MEAs was identified by titration with sodium hydroxide [3].

By ex-situ ante- and post-mortem micro-computed tomography (µ-CT) investigations, mechanical transformations can be visualized and the thickness changes of all layers can be verified.

Fresh MEAs are characterized by mirror symmetry through all layers, while MEAs, operated under dry reactant gases, disclose a reduction of catalyst layer thicknesses on anode sides. The operation with pure O₂results into increased water production, which therefore causes higher acid-leaching. The MEA tested with synthetic reformate and pure oxygen has revealed the highest phosphoric acid loss (Figure 1) on one hand and lowest degradation rates (-6.4 µV/h) on the other hand.

Figure 1: Total H₃PO₄ loss in product water after EoT and degradation rates (DR) of each operation strategy

References:

1. S. Yu, L. Xiao, B.C. Benicewicz, Fuel Cells, 08, 3-4, 165-174 (2008).

2. Y. Zhai, H. Zhang, G. Liu, J. Hu and B. Yi, J. Electrochem. Soc., 154, B72 (2007).

3. N. Pilinski, M. Rastedt, P. Wagner, ECS Trans., 69, 323-335 (2015)

Figure 1

Alkaline anion exchange membrane fuel cells have become a topic of substantial interest in recent years, opening up a new electrochemical environment for hydrogen fuel cells. Facile kinetics for the oxygen reduction reaction open the promise of non-PGM or even non-metal AEMFCs, and radical stability. The most challenging aspect for the field, as defined by the 2016 DOE AMFC III Workshop, is membrane and ionomer stability in alkaline conditions at elevated operating temperatures. Few papers report endurance data, and best practices for in situ fuel cell conditioning and electrochemical characterization are still being developed by the community.

Here, we report that HMT-PMBI exhibits membrane and ionomer stability in situ as AEMFCs in relevant conditions for device operation. These fuel cells demonstrate re-equilibration from extensive carbonation and complete re-conditioning in a shut-down / start-up cycle. We further report operation in typically challenging conditions, e.g. increased temperature and reduced humidity. Finally, we report on our attempts to define best-practices for electrochemical characterization.

With the advancement of portable technology, suitable power sources must be developed. Alternatives to standard battery technology, such as fuel cells, have shown promise; however such fuel cells are in their infancy with regards to industrial and consumer adoption [1-3]. In order for fuel cells to become a feasible alternative to traditional dry-cell alkaline and lithium-ion batteries, their performance must be optimized for their intended application via manipulation of parameters such as fuel molarity, flow rate, temperature, and electrode geometry. Here, we present a mathematical model of a microscale direct methanol fuel cell (DMFC), validated with experimental data. In the past, an iterative fabrication approach was undertaken to optimize DMFC performance, involving the fabrication and testing of many individual components to assess the effects of individual parameters. This model provides a means to automate the design process and remove the fabrication requirement.

Prior parametric studies by Thorson et al. [4] show that shorter and wider electrodes yield higher current densities, however a recent review by Goulet et al. [5] highlights the need for a combined modeling and experimental approach to evaluating electrode design in a microscale fuel cell. Here we present a mathematical model to analyze the fuel flow pattern and performance of a direct methanol microscale fuel cell as a function of electrode geometry. Localized current densities are calculated over the electrode surface to elucidate the differences between electrode geometries. In the present work, the dimensions of the fuel channel are constant (0.5 cm (L) x 1.0 cm (W) x 0.125 cm (H)), while the length and width of the catalyst deposition region are varied. The model is validated with experimental data taken as a function of electrode geometry, fuel concentration, fuel temperature, and fuel flow rate. The performance of our experimental fuel cell is consistent with our modeling studies, achieving a maximum power density greater than 25 mW/cm² at room temperature with 1 M methanol. The model presented here, in conjunction with the supporting experimental data, expands prior parametric studies [4] to predict optimal electrode design with regard to methanol concentration, temperature, and flow rate.

[1] C.K. Dyer, Fuel Cells Bulletin, 2002 (2002) 8-9.

[2] E. Kjeang, N. Djilali, D. Sinton, Journal of Power Sources, 186 (2009) 353-369.

[3] A.S. Hollinger, P.J.A. Kenis, Journal of Power Sources, 240 (2013) 486-493.

[4] M.R. Thorson, F.R. Brushett, C.J. Timberg, P.J.A. Kenis, Journal of Power Sources, 218 (2012) 28-33.

[5] M.A. Goulet, E. Kjeang, Journal of Power Sources, 260 (2014) 186-196.

Figure 1

An alternative approach to the rotating disk electrode (RDE) for characterising fuel cell electrocatalysts has been recently demonstrated¹. The approach combines high mass transport with a flat, uniform, and homogeneous catalyst deposition process, well suited for studying intrinsic catalyst properties at realistic operating conditions of a polymer electrolyte fuel cell (PEFC). Uniform catalyst layers were produced with loadings as low as 0.16 μg_Pt cm^-2 and thicknesses as low as 200 nm. Such ultra thin catalyst layers are considered advantageous to minimize internal resistances and mass transport limitations leading to very high performance at low platinum loadings. Modelling of the associated diffusion field suggests that such high performance is enabled by fast lateral diffusion within the electrode. The electrodes operate over a wide potential range with insignificant mass transport losses, allowing the study of the hydrogen oxidation reaction (HOR) and hydrogen evolution reaction (HER) at high overpotentials. For the HOR, geometric current densities as high as 5.7 A cm^-2_Geo were experimentally achieved at a loading of 10.15 μg_Pt cm^-2 at room temperature (561 A mg_Pt^-1), which is three orders of magnitude higher than current densities achievable with the RDE. For the HER specific current densities greater than 5 A cm^-2_Specific  have been achieved. The latter corresponds to a turnover frequency of almost 20,000 hydrogen molecules per surface platinum site per second.

We have taken these results and applied a new microkinetic model for hydrogen evolution/oxidation based around the classical Heyrovsky-Tafel-Volmer mechanistic steps. This new mechanistic formalism is easy to implement and provides insights into the HOR/HER. For each of the different mechanisms, an "atlas" of H_ads coverage with overpotential and corresponding current density is provided, allowing an understanding of all possible responses depending on the dimensionless parameters. Analysis of these mechanisms provides the limiting reaction orders of the exchange current density for protons and bimolecular hydrogen for each of the different mechanisms, as well as the possible Tafel slopes as a function of the molecular symmetry factor, b. Only the HV mechanism is influenced by pH whereas the TV,HT, and HTV mechanisms are not. The cases where the equations simplify to limiting forms are discussed. Analysis of the exchange current density from experimental data is discussed, and it is shown that fitting the linear region around the equilibrium potential underestimates the true exchange current density for all of the mechanisms studied. Furthermore, estimates of exchange current density via back-extrapolation from large overpotentials is also shown to be highly inaccurate. Analysis of Tafel slopes is discussed along with the mechanistic information which can and cannot be determined.

Finally, some interesting effects are discussed which show that the HER can be accelerated under certain conditions.

Ohmic shorting through the membrane has been identified as one of the major failure modes in polymer electrolyte fuel cells (PEFC) [1]. Shorting occurs when electrons flow directly from the anode to the cathode instead of through the device being powered. Ohmic shorting not only can reduce the fuel cell performance, but also can lead to local heat generation in the vicinity of the short, causing membrane damage that can ultimately result in gas crossover failure. Two types of membrane shorts have been observed during the fuel cell testing: soft shorts and hard shorts. A soft short is a sub-critical short that results from the conductive carbon fibers or debris penetrating through the membrane. The soft shorts do not immediately lead to fuel cell failure; however, significant accumulation of soft shorts can reduce the overall cell resistance and compromise fuel cell durability through cell voltage degradation. A hard short is a critical short that is the result of significant heat release from an existing soft short. A hard short can directly lead to membrane crossover and cell failure. Hard shorts occur suddenly in an operating fuel cell stack where one cell develops a cell voltage reversal well below -1 V [2].

The present study is composed of two parts. The first is on the development of a test method to determine the density of soft shorts and their ohmic resistance that may lead to hard shorts when the adverse fuel cell operating conditions are met. The second part of the study is to investigate these adverse fuel cell operating conditions that lead to cell reversal and the potential hard shorts. The ultimate objective of the study is to prevent the fuel cell shorting failure through better material selection/designs and fuel cell operation controls.

In the first part of this study, we use a current distribution circuit board [3] in a fixture that induces uniform compression over a 32 cm² area of a sample that consists of a piece of proton exchange membrane or a membrane electrode assembly sandwiched between two gas diffusion layers. By incrementally increasing the compressive pressure and subjecting the sample to a voltage of 0.6 V, we can measure increasing shorting currents in some of the 0.5 cm² current distribution segments. By de-convoluting the current density distribution measured in this experiment, the number and severity of each short can be identified and the robustness of various material sets against soft shorts can be compared.

In the second part of the study, we stepwise dried out a single large-active-area fuel cell by decreasing the inlet gas dew points while under galvanostatic control. At a critical dew point, the cell potential dropped rapidly to reach a negative voltage that was limited to -1 V. Tests were conducted at various cell current densities and inlet temperatures, which showed that the corresponding cell resistance at the onset of rapid cell voltage drop decreases as the current density increases or as the temperature decreases. Good correlation is found between the cell resistance and the onset of cell reversal at various current densities. When the cell reversal limit of -1V was removed, we found that hard shorts can develop at a cell voltage around -3V, confirming the importance of operating the fuel cell within the operating window developed in this study.

• A.B. LaConti, M. Hamdan, R.C. McDonald, Mechanism of Membrane Degradation for PEMFCs, in Mechanisms of Membrane Degradation for PEMFCs, in: W. Vielstich, A. Lamn, H.A. Gasteiger (Eds), Handbook of Fuel Cells: Fundamentals, Technology and Applications, Vol. 3, John Wiley & Sons, New York, 2003, pp. 647–662.

• Gittleman, C. S., Coms, F. D., and Lai, Y. H., "Chapter 2: Membrane Durability: Physical and Chemical Degradation", Polymer Electrolyte Fuel Cell Degradation, Editors: Matthew M. Mench, E. Caglan Kumbur, T. Nejat Veziroglu, Elsevier (2012).

• J.J. Gagliardo, J.P. Owejan, T.A.Trabold, and T.W.Tighe, Nucl. Instrum. Meth. A, 605 (2009) 115-118.

The successful commercialization of Polymer Electrolyte Fuel Cells (PEFCs) system requires highly active oxygen-reduction reaction (ORR) electrocatalysts to meet the cost, performance, and durability targets for automotive applications. Although significant progress has been made in the past decade, Pt/C catalyst has the limitation of high material cost and insufficient ORR activity. Platinum-nickel alloy catalysts with core-shell structure have been demonstrated by multiple research groups to have higher intrinsic ORR activity, which can exceed the U.S. DOE 2015 ORR activity target, and also have promising stability in PEFC conditions [1, 2]. Additionally, Membrane Electrode Assemblies (MEAs) with low loadings of highly active Pt-group metal (PGM) catalysts have already been demonstrated to exceed the high-efficiency target of > 0.3 A/cm² at 0.8 V [3]; however, the rated-power target of 1 W/cm² cannot be met with these MEAs due to transport losses that are unique to MEAs with ultra-low catalyst loadings [4, 5].

United Technologies Research Center (UTRC) is a part of a DOE-supported research project, led by Argonne National Laboratory (ANL), focused on improving the understanding, performance, and durability of advanced Pt-alloy catalysts operating in complete PEFCs. In the current study, advanced MEAs with high ORR activity Pt-Ni alloy catalysts and ultra-low Pt loading (total ~ 0.13 mg-Pt/cm²) are prepared by Johnson Matthey Fuel Cells (JMFC) and tested by UTRC.

The focus of this talk will be on using a variety of in-cell diagnostics and analysis methods to investigate both the initial performance and durability of these state-of-the-art MEAs. One example diagnostic that will be utilized are Polarization Change Curves (PCCs), which were originally developed for durability studies [6], but can also be used to investigate changes in performance due to operating conditions or MEA composition. For example, Fig 1a is a PCC comparison of the initial performance of the PtNi MEA and a Pt-only MEA, which have similar average catalyst particle size and the same carbon support, metal/carbon ratio, and ionomer/carbon ratio. Clearly, in addition to the differences in catalyst activity, there are also differences in the transport losses between these two different electrocatalysts. The durability of ultra-low Pt loading MEAs will also be discussed. Fig. 1b is a durability example of a PCC result, which is after applying a recent DOE-suggested AST protocol (0.6V-0.95V trapezoid potential cycle, with 700mV/s ramp rate and 6s/cycle). Other diagnostics and analysis beyond PCCs will also be included. The results obtained from these in-cell diagnostics will be used to hypothesize the potential mechanisms responsible for the differences in polarization curves, as well as project possible pathways to future performance and durability improvements.

Acknowledgements

This work is partially funded by the U. S. Department of Energy (DOE) under contract number DE-AC02-06CH11357. The authors would like to thank their DOE-project colleagues, especially those at ANL and JMFC.

References

1. V. R. Stamenkovic, et al. Science, 315, 493 (2007).

2. B. Han, et al, Energy Environ. Sci., 8, 258, (2015)

3. D. Myers, et. al., DOE AMR, ID# FC106 (2015).

4. W. Yoon & A. Z. Weber, J. Electrochem. Soc., 158, B1007 (2011).

5. A. Kongkanand, et al., ACS Catalysis, 6, 1578 (2016).

6. R. Darling, et al., Modern Topics in Polymer Electrolyte Fuel Cell Degradation, Chap. 7, M. Mench, et.al., Ed.; Elsevier, Denmark (2011).

Figure 1

It is important to ensure that the water content of a catalyst coated membrane (CCM) has equilibrated before it is used in the manufacturing of a proton exchange membrane (PEM) fuel cell stack. Equilibration is achieved when the water content in the CCM reaches equilibrium with controlled relative humidity (RH) and temperature conditions. A CCM consists of a membrane coated with a catalyst layer (CL) electrode on both sides. The ionomer, which is a component of both the membrane and the CLs, absorbs/desorbs water as the CCM equilibrates, and as such, the ionic conductivity of the CCM is directly impacted. The electronic conductivity is indirectly impacted as dimensional changes in the ionomer result in changes to the network of electronically conducting materials in the CLs. Alternating current (AC) and direct current (DC) methods were investigated to measure the through-plane ionic resistance and in-plane electronic resistance as the CCM equilibrated.

When CCM is manufactured it is wound around a plastic core with an impermeable leader and trailer portion protecting the surface of the CCM from the core and from the environment. As such, only the two edges of the roll are exposed to environmental conditions. It is in this state that the CCM must equilibrate to the manufacturing environment. To simulate this, discrete 4 samples of a commercially available CCM (Ion Power) were tested in a sample holder with pressure applied to replicate the roll winding tension. The sample holder covered the top, bottom, and two sides (roll direction) of the CCM, leaving the other two sides exposed to controlled conditions in an environmental chamber. The measurements described below were taken as the samples equilibrated from 50% RH to 45% RH at a constant temperature.

The first method is an AC technique used to measure the through-plane sample impedance. The correlation between the RH of ionomers and their ionic conductivity is well established in literature [Cooper 2009, Maréchal 2007, Anantaraman 1996]. To measure the ionic conductivity of the sample, a sinusoidal alternating voltage was applied to silver foil strips on the top and bottom of the sample. This results in an alternating current passing through the sample. Electrochemical impedance spectroscopy (EIS) was used to measure the impedance as a function of frequency. From this data it was possible to extract the conductivity of the membrane and the equilibration time was investigated by tracking the impedance of the samples at 1 kHz.

The second method is a DC technique, which employs more readily available equipment and reduces the data analysis complexity compared to the AC technique. The DC setup used four probes made of gold plated foil strips contacting each corners of the sample: two probes to measure a voltage difference and two to apply current to the sample. Current flows through the CL via a carbon/platinum network, which is only indirectly affected by water content as the ionomer expands or contracts due to water absorption or desorption [Morris 2014]. A constant current was applied to the sample undergoing equilibration and the voltage difference was monitored to determine the equilibration time.

Both methods showed sensitivity to changing RH and exhibited reasonably constant results after a sufficient time, indicating that the samples had equilibrated to the environment. Figure 1 shows the impedance (1 kHz) of a CCM initially equilibrated to 50% RH as it equilibrates to 45% RH. The impedance increased rapidly for the first few hours and then slowed down before eventually reaching an approximately constant value, indicating equilibrium with the environment was achieved after approximately 18 h. The results of the DC method were consistent with those of the AC method.

The novelty of this research is in providing quantitative methods to determine when a CCM roll has equilibrated with its environment. Once equilibrated, the CCM roll can be used in the manufacturing of fuel cell stacks. This research will help increase the efficiency of the manufacturing process and improve the quality assurance of the final product.

Research funding from both MITACS (Accelerate program) and NSERC (Engage Plus program) is gratefully acknowledged, as is early and significant work by Anne Moore and Jeff Laflamme.

References

Anantamaran, A. V.; Gardner, C. L. J. Electroanal. Chem., 414, 115 (1996)

Cooper K. R. ECS Transactions, 25, 995 (2009)

Maréchal, M; Souquet, J. -L.; Guindet, J; Sanchez, J. -Y. Electrochem. Comm., 9, 1023 (2007)

Morris, D.; Liu, S.; Gonzalez, D.; & Gostick, J. ACS Applied Materials & Interfaces, 18609 (2014)

Figure 1

Proton-exchange-membrane fuel cells (PEMFC) have been greatly advanced towards commercially viable hydrogen-powered fuel cell systems for automotive application in recent years. Significant reduction in cost has been achieved by reducing the loading of Pt catalyst in the electrodes. Unfortunately lower Pt catalyst loadings accompany higher losses in cell voltage at high power. To quantify various voltage loss terms and thus identify the opportunity for improvement, mathematical models with parameters extracted from experiments are commonly employed. A differential cell operates at high stoichiometry flow with minimal pressure drop down the channel, making it almost one-dimensional across the cell. Hence the experiments are conducted on it to extract model parameters. The cell performance can be described by:

        E_cell = E_rev - iR_Ω - η_HOR - η_ORR - i(R_H⁺,a +R_H⁺,c) - η_{tx O2}           (1)

 where E_cell is the cell voltage; E_rev is the reversible cell voltage, i is the current density; R_Ω is the sum of the Ohmic resistances of proton conduction through the membrane and of electron conduction (commonly referred to as high frequency resistance or HFR); η_HOR and η_ORR are the charge transfer overpotentials for the hydrogen oxidation reaction (HOR) and the oxygen reduction reaction (ORR), respectively;  is the effective resistance to proton conduction in an electrode with subscripts a and c denoting anode and cathode, respectively; and lastly, is the oxygen transport loss which consists of two parts: one results from oxygen transport through gas phase, and the other is associated with the oxygen transport resistance local to the Pt surface, namely

 η_{tx O2} = η_{tx O2 (gas)} + η_{tx O2 (Pt-local)}  (2)

 Limiting current densities are commonly used to evaluate gas phase and local oxygen transport resistances. Measurements with oxygen pressure and concentration variation can be used to evaluate oxygen transport resistance through the bulk phase, ¹ and limiting current density measured on MEAs with varying Pt loading can be used to estimate local transport resistance.² While both gas phase and local transport resistances are coupled, models with 1D transport lengths (across the gas diffusion layer (GDL), catalyst layer etc.) are used to de-convolute the resistances. The differential cells used are usually land-channel configuration which also has second diffusion length under the land. Transport resistance due to oxygen under the land across the GDL can dominate the total transport resistance. The p_O2 is also largely decreased under the land and in an actual (non-differential) cell its impact on local oxygen transport resistance is enhanced down the channel where oxygen concentrations are lower. Therefore an accurate description of transport resistance under the land is needed to decouple the transport resistances. Differential cell tests with an active area of 5 cm² were designed to study the effects of O₂ diffusivity under the land by varying land/channel dimensions or flow field pitch. Cathode land/channel dimensions (mm/mm) of 0.2/0.2, 0.5/0.5, 0.7/0.7 and 1.0/1.0 were studied. In addition to the flow field dimensions, membrane electrode assemblies (MEAs) with different carbon supports and various Pt loadings were fabricated and tested. Catalyst such as Pt/HSC (high surface area carbon) and Pt/Vulcan at different Pt wt% were chosen to provide wide contrast in local oxygen transport resistance owing to the Pt particle distribution and location.  Electrochemical performance of the electrodes were characterized with electrochemical surface area, mass activity and polarization curve measurements. Additionally, protonic resistance of the catalyst layer were determined with electrochemical impedance spectroscopy in H_­2-N₂.

 An electrochemical model coupled with transport across the cell was developed to interpret the differential cell data. The oxygen transport resistance local to the Pt surface, hypothesized to originate from ionomer-Pt interaction, is also evaluated by model-data comparison. The model parameters perceived to be critical for accurate voltage-loss and cell performance prediction will be examined by sensitivity analysis to design and operating variations

 References

1) D. R. Baker, D. A. Caulk, K. C. Neyerlin, and M. W. Murphy, J. Electrochem. Soc., 156, B991 (2009).

2) T.A. Greszler, D.A. Caulk, P. Sinha, J. Electrochem. Soc., 159, F831 (2012).

The durability of fuel cells is closely related to water content of the catalyst coated membrane (CCM). When the water content is high, the electrode catalysts become swollen due to dissolution, agglomeration, and redeposition, and the decline in the catalyst surface area accelerates. When the water content is low, impurities and hydrogen peroxide accumulate in the PEM, promoting chemical degradation and reducing the thickness of the PEM.

To help ensure durability of automotive fuel cells, it is therefore necessary to identify the in-plane distribution of the water content of CCM in the cells and to control it within the proper range. The in-plane distribution of water content is also affected by the operating condition parameters of the cells. In addition to the fact that a large number of combinations of operating condition parameters are possible, measurements also take time, and it has been challenging to determine parameters with an adequate level of accuracy.

 The development of the fuel cell stack for the 2016 model FCV sought to realize increased fuel cell stack durability by clarifying the distribution of the water content of the CCM in the cells through the development of a multi-segment impedance sensor system and the prediction of the optimum water content distribution using simulations.

 There are various measurement methods of water content inside fuel cell. For example nuclear magnetic resonance (NMR), neutron radiography, X-ray and so on. In order to clarify CCM water content in an actual vehicle, we need a method that would make it possible to obtain measurements without influencing the in-plane humidity distribution and other such factors during electric generation.

 Based on the results of examination, the impedance method was selected to measure in-plane distribution of water content　during actual vehicle operation and development carried out.  The sensor is intended for incorporation in a fuel cell stack and was therefore designed with the same shape as the MEA. As a result, it becomes possible to obtain CCM water content measurements without any need to prepare special separators or seals. The　surface on either side of the sensor has 75 square sensor pads which used to make impedance measurements of the CCM arranged. With measurement using the sensor, the distribution of CCM water content could be measured.

 In order to develop a method of simulation of the in-plane water content of CCM in cells, original Honda functional modifications were made to commercially available polymer electrolyte fuel cell simulation software. Simulation results correspond well with results of measurements of multi-segment impedance sensor system and the average predictive accuracy for the 75 segments was within 5.0%. Following the optimization of the operating condition parameters with the simulation, variations in the distribution of water content in the CCM were reduced by about 30%. In addition, the use of the developed simulation made it possible to determine operating conditions in several days that would previously have necessitated several weeks using tests.

The in-plane water content distribution of the CCM varies due to instantaneous changes in operating conditions, considerably during running mode, increasing during acceleration, and decreasing during deceleration. Because of this, it was necessary to verify whether the in-plane water content distribution of the CCM, which was optimized by simulation above, was maintained within the scope of the upper and lower limit values during transient current generation in an actual on-board generation system. During acceleration, In-plane water content of the CCM reaches its maximum value in the center of the cell in the direction of gas flow but remains below the upper limit value. In the transition from acceleration to deceleration, the in-plane water content of the CCM at the cathode inlet declines to close to the lower limit value. It was verified, however, that the in-plane water content of the CCM during deceleration remained within the upper and lower limit values.

The effect of the microporous layer (MPL) on membrane electrode assemblies (MEAs) has been investigated. The performance of an MEA employing an MPL with a larger pore volume was better under 100%RH condition, which we reported in 2015 (1). Under 30%RH, no distinct relationship was found between the pore volume of the MPL and the MEA performance. However, the MEA employing an MPL with a larger median pore diameter showed higher cell voltage.

We reported that the MEA using a hydrophilic MPL showed much better performance in a wide range of pressure and humidity conditions than that using a conventional hydrophobic MPL. We applied our technique of preparing a carbon fiber dispersion to the fabrication of MPLs using carbon particles and carbon fibers with various fiber diameters. Properties of carbon materials and MPLs are shown in Table 1. The pore size distribution (Fig. 1) was measured using mercury intrusion porosimetry.

Figure 2 shows the surface SEM images of the MPLs: The MPL made of carbon black (Vulcan XC) has the smallest pore diameter (0.068 μm) and that made of carbon fiber (VGCF-H) has the largest pore diameter (0.77 μm). Although carbon fiber CF-X has the largest fiber diameter of 250 nm, the MPL made of it has the smallest pore volume. This is probably because fibers of CF-X are straight and longer than other carbon fibers, tending not to become tangled and form larger pores.

MEAs having an MPL made of four different carbon materials (Vulcan, 24PS, CF-X, and VGCF-H) were prepared, and polarization curves were measured under the conditions of 80 ºC, 30%RH. The IR free cell voltage (@0.1, 0.5, 1.0 and 2.0 A/cm²) for the four MEAs are plotted against the median pore diameter of the MPL in Fig. 3. The larger the pore diameter of the MPL, the higher the cell voltage becomes. The MEA employing an MPL made of VGCF-H, which has the largest median pore diameter, shows the best performance under the dry condition of 30%RH. This is probably because there is little liquid water in the MPL under the dry condition, and the cell voltage is largely dependent on the oxygen diffusivity of the MPL: The larger median pore diameter of the MPL, the better oxygen diffusivity and the higher the cell voltage.

References

1.　T. Tanuma and M. Kawamoto, ECS Transactions, 69, 1323 (2015).

Figure 1

Manufacturing methods for fuel cell catalyst layers are critically important for achieving high material utilization and performance. Typically, electrode systems are cast from an 'ink' or 'colloidal' state, where the solid material is suspended in a solution containing a solvent and a polymer. For polymer-electrolyte fuel cells, a perfluorinated sulfonic-acid (PFSA) ionomer, typically Nafion, is used in the ink and catalyst layer, where its roles are to act as a binding agent and conduit for proton conduction to the reaction site.¹ Recently, it has been suggested that the ionomer film that surrounds the platinum reaction site may contribute to increased resistances that limit performance at low Pt loadings.^1,2 Thus, it is of great interest to understand the factors controlling the film formation in catalyst layers to optimize material properties and cell performance. The nature of the dispersion and casting process could have significant influence on the morphology and properties of the dispersion-cast membranes.^3-5 To isolate these effects, the morphology of PFSA dispersed in various solvents (glycerol, water/2-propanol, and NMP) was reported previously using small angle neutron scattering (SANS), where catalyst layers from the different solvents exhibit different performance.⁶ The understanding of local'ink' structures as a function of active material (polymer and solid) volume fraction (Φ) and across solvent space and length scales can aid in understanding bulk properties and the impact of ink formulation in cast systems. In this talk, a direct observation of the colloidal suspension of the ionomer in different solvents will be described using confocal laser scanning microscopy (CLSM). CLSM has long been utilized as a tool for characterizing the structure of biphasic colloidal gels. We combine dynamic light scattering, ultra-small angle x-ray scattering, and confocal microscopy to understand PFSA size and structure in dilute (1,3, and 5 wt%) ionomer solutions. The structural formation of nafion membranes are observed in-situ with small angle x-ray scattering. Finally, we contrive a model 'ink' composed of attractive and repulsive silica spheres and PFSA suspended in different solvents (water, NMP, and isopropyl-alcohol) and use CLSM to gain a greater understanding about solid|PFSA interactions in model inks.

1. A. Z. Weber and A. Kusoglu, Journal of Materials Chemistry A, 2, 17207 (2014).

2. A. Kongkanand and M. F. Mathias, The Journal of Physical Chemistry Letters, 1127 (2016).

3. T. T. Ngo, T. L. Yu and H.-L. Lin, Journal of Power Sources, 225, 293 (2013).

4. C.-H. Ma, T. L. Yu, H.-L. Lin, Y.-T. Huang, Y.-L. Chen, U. S. Jeng, Y.-H. Lai and Y.-S. Sun, Polymer, 50, 1764 (2009).

5. Y. S. Kim, C. F. Welch, R. P. Hjelm, N. H. Mack, A. Labouriau and E. B. Orler, Macromolecules, 48, 2161 (2015).

6. C. Welch, A. Labouriau, R. Hjelm, B. Orler, C. Johnston and Y. S. Kim, Acs Macro Letters, 1, 1403 (2012).

Aviation industry and governing bodies have set out to significantly reduce the environmental footprint of aviation. The visions brought forward request such low emissions that continuous optimization and advancement of current technology alone are not sufficient. This has led research groups to explore disruptive paths of innovation for aircraft and a trend to the electrification of onboard systems in more electric aircraft. Providing the required electric power using fuel cells and batteries not only promises the reduction of carbon emissions and noise but also allows for system level optimizations using synergy effects from a multi-functional system-integration that aircraft design could benefit from. At the same time using fuel cells in aircraft environments provides great challenges towards system design. System weight and size constraints require the use of air breathing fuel cells in commercial aircraft and airliners. This results in operation at low pressure and temperature which in term confronts the system design with two problems. First the low air density makes cooling of power systems difficult especially considering the low operating temperatures of PEM Fuel Cells. Second the low ambient temperature means the ingested air will heat up much more than during ground operation and in term will remove more water from the fuel cells, making water management and fuel cell operation strategies more important and complicated than in other applications. Also ambient conditions very rapidly change in air, adding even more complexity to the application.

Taking into consideration possibilities of advanced aerodynamic concepts like distributed propulsion and propulsion/wing vortex interaction, that electric propulsion could enable, Fuel cells could be of even larger benefit for aviation when considering a larger time-frame. All-electric aircraft could potentially be emissions-free and very low noise, thus enabling sustainable, environmentally friendly aviation. While not yet feasible for large passenger aircraft, application of such systems is currently being researched in aircraft like the HY4 enroute to this goal. This presentation will discuss the challenges of fuel cell operation in-flight, as well as introduce the concepts for multifunctional fuel cell APU replacements for A320 type aircraft and drive-trains for all-electric aircraft. Finally results from ground- and flight-testing of the all-electric fuel cell powered HY4 aircraft will be discussed.

Hydrogen fuel cells materials and systems have attracted billions of dollars of government and commercial research such that they are now reliable and cost effective enough to be commercialized as the power source for automobile propulsion and for materials handling. Also known as polymer electrolyte membrane fuel cells (PEMFCs), hydrogen fuel cells are attractive because they convert high-energy hydrogen fuel and oxygen (typically from air) efficiently to electricity and water through electrochemical processes, and produce relatively little heat and noise. They also respond rapidly to changes in load. The result is that they can have higher specific energy and energy density than battery systems for certain applications. One such application for commercial automotive hydrogen fuel cells is for large unmanned undersea vehicles (UUVs). This paper will describe an Office of Naval Research effort by the U. S. Naval Research Laboratory, Naval Surface Warfare Center Carderock Division, University of Hawaii, and General Motors (GM) to integrate an automotive fuel cell from GM into an unmanned undersea vehicle. The advantages and challenges with taking commercial fuel cell technology and adapting it to undersea use will be discussed.

In this work a design process for the development of an air independent fuel cell system operating a commercial fuel cell stack as a range extender in an AUV is presented. The development follows an optimization strategy which has been derived from the universal fuel cell development plan proposed by Ang et al.[1] The first step is defining the constraints and the objectives in the design process. While constraints set the boundary conditions of the system the identified objectives are used as optimization parameters. The design plan follows the schematic in figure1:

Figure 1: design concept for an optimized system used in an AUV

The development of fuel cell systems used in a submarine environment will need to consider different boundary conditions than the development of a fuel cell system for automobiles. These constraints result from the sea water environment which can't be used as an oxygen source and additionally hinders mass exchange with the fuel cell system because of pressure differences. As a result a closed fuel cell system relying on a pure oxygen supply has to be developed. The fuel cell is powering the AUV with a stationary output while a battery is buffering the power differential between the output und the consumption of the AUV. The AUV is used as a testing platform by the German navy and thereby sets very unique constraints on the system development. Additionally the commercial fuel cell stack is setting boundary conditions for the system development, which has to meet the stacks individual operating conditions. After identifying the boundary conditions the system size was identified as the optimization parameter.

To optimize the system size different setups for the cooling system as well as the gas supply system found in the literature are evaluated. Setups meeting the boundary conditions are validated in a test bench with hardware-in-the-loop tests. Based on these measurements a volume-optimized system can be found. Challenges which deviate from this setup are identified and can be tackled in the subsequent design step.

The next step is a discussion of the different concepts and an evaluation of their impact on the system operation. While the discussed solutions are still meeting the maximum system size, an evaluation of their influence on other design parameters can be made. From this different system designs are derived and the resulting operation strategies are compared to further improve the volume-optimized system. Also solutions targeting specific challenges like standby mode and system shutdown are evaluated and presented.

In a last step the qualitative connections between the found system designs and the boundary conditions in this system are discussed leading to a more general approach for the development of air independent fuel cell systems for AUVs.

[1] S. M. C. Ang, E. S. Fraga, N. P. Brandon, N. J. Samsatli, and D. J. L. Brett, "Fuel cell systems optimisation – Methods and strategies," Int. J. Hydrogen Energy, vol. 36, no. 22, pp. 14678–14703, Nov. 2011.

Figure 1

Lower cost materials for stack hardware and system components help reduce the overall cost of the automotive and stationary fuel cell systems and make these competitive in the market. However, low-cost system component materials need to provide similar function, performance and durability. Intelligently selecting low cost materials for application in polymer electrolyte membrane fuel cell (PEMFC) systems requires understanding the potential adverse effects that system contaminants may have on the fuel cell performance and durability.

There are many prospective balance of plant (BOP) materials that can be used in fuel cell systems. Families of material, based on input from OEMs, fuel cell system manufacturers and other attributes like cost, physical properties, have been studied. These include structural materials, elastomers for seals and (sub)gaskets, and assembly aids (adhesives, lubricants).

The contaminants from the BOP materials were leached out via an accelerated aging procedure. The leachates obtained from these plastics were a mixture of organics, inorganics, and ions. The sulfate anion was one of the anions identified in the leachates and was chosen as the model compound for further studies. Sulfate anions can also come from membrane degradation. Ex-situ electrochemical measurements were carried out to understand the impact of sulfate anion concentration on oxygen reduction reaction (ORR). The ORR scan rate and direction (low to high potential and vice versa) were studied to determine the effect of the initial state of the Pt surface on sulfate contamination. In-situ fuel cell experiments were also carried out. Sulfate anion was introduced to a working fuel cell to determine their effect on fuel cells performance. Our previous work showed that sulfate anion did not result in performance loss when it was infused into the cathode at 0.2 A/cm². The cathode potential was greater than 0.8 V at this current density. It is postulated that sulfate contamination is potential dependent and that sulfate adsorption only occurs when Pt is in its metallic state. In the current study, the effect of potential on sulfate contamination will be studied by infusing sulfate into the cathode at several constant current densities. Several in-situ diagnostics such as infusion, cyclic voltammetry, electrochemical impedance spectroscopy, and I-V curves were carried out to better characterize the contaminant effects of sulfate anion. The goal is to better understand the contamination mechanisms of specific species so that mitigation strategies can be determined.

The authors would like to acknowledge funding from the U.S. Department of Energy EERE Fuel Cell Technologies Office, under Contract No. AC36-08GO28308 with the National Renewable Energy Laboratory and collaborations with colleagues at GM. Structural plastic materials and leachates were provided by GM for this study.

Research has shown that contaminants in the hydrogen fuel stream have an impact on Polymer Electrolyte Membrane Fuel Cells (PEMFCs) performance that is influenced by the platinum loading, membrane thickness, the impurity, its concentration and PEMFC operating conditions such as cell temperature, relative humidity, and back pressure. A significant portion of previous reported results were conducted with the fuel cell being operated in single-pass mode where the exhaust hydrogen is vented from the fuel cell. While these findings have proven useful in improving the fundamental understanding of the poisoning mechanisms of non-hydrogen species (contaminants) in operating single fuel cells; the question remains whether or not these findings correlate to PEMFC systems that use a re-circulating fuel stream as being designed by the OEMs. Some advantages of re-circulating the anode exhaust gas back into the anode inlet are: 1) the anode outlet water is returned into the dry H₂ fuel stream to obtain proper humidification and also 2) the excess H₂ fuel is returned to enhance fuel utilization. While these advantages are attractive, there are some inherent challenges that are introduced in fuel re-circulation mode. Two of the more notable drawbacks involve inert gas build-up due to crossover and the accumulation of contaminants in the anode. In this study, we focus our efforts on correlating the impact of CO and H₂S on fuel cell performance in H₂ single-pass versus H₂ re-circulation mode using the SAE J2719¹ and ISO 14687-2 Hydrogen Fuel Product Specification² limits (4 ppb H₂S and 200 ppb CO). This study was conducted on Membrane Electrode Assemblies (MEAs) at the 2015 DOE target loadings for platinum (anode: 0.05 mg/cm² and cathode: 0.1 mg/cm²).

References:

• SAE J2719: Hydrogen Fuel Quality for Fuel Cell Vehicles, www.sae.org

• ISO 14687-2, Hydrogen Fuel – Product Specification, Part 2: PEM fuel cell applications for road vehicles, http://www.iso.org/iso/catalogue_detail.htm?csnumber=55083

Acknowledgements:

The authors gratefully acknowledge the financial support of the DOE Fuel Cell Technologies Office and the support of Technology Development Manager, Charles (Will) James, Jr.

In Hawaii, very high penetration levels of intermittent renewable energy sources such as wind and solar are causing challenges in regulation of grid frequency, and as the percentage of these intermittent resources increases can result in their curtailment. It has been postulated that, similar to a battery, the optimized use of an electrolyzer as a variable load may help regulate grid frequency and increase the penetration of renewable energy resources on the grid. The use of an electrolyzer in this way could provide an "ancillary service" to the grid that can be assigned a monetary value. This monetary value can be used to offset the cost of hydrogen production. The hydrogen in turn can be used in high value applications such as a transportation fuel. The Hawai'i Natural Energy Institute (HNEI) is conducting research to assess the technical potential and economic value of using an electrolyzer-based hydrogen production and storage system as a demand response tool for grid management.   A 65 kg/day hydrogen energy system (H₂ES) consisting of a PEM electrolyzer, 35 bar buffer tank, 450 bar compressor, and associated chiller systems, has been purchased and installed at the Hawaii Natural Energy Laboratory Hawaii Authority (NELHA) to demonstrate long-term durability of the electrolyzer under cyclic operation required for frequency regulation on an island grid system. A secondary objective is to supply hydrogen for three fuel-cell buses to be operated at Hawai'i Volcanoes National Park (HAVO) and by the County of Hawai'i Mass Transit Authority. The hydrogen will be transported from the production site by tube trailers to the HAVO dispensing site. A second dispensing site is located at the NELHA production site.     

A comprehensive test plan has been developed to characterize the performance and the durability of the electrolyzer under dynamic load conditions and initial testing was conducted at Powertech Labs in Vancouver Canada. A main objective of the test plan is to determine the operating envelope and dynamic limits of the electrolyzer and the overall H₂ES.  A long-term study shall evaluate the H₂ES performance under a load profile developed from a fast acting Battery Energy Storage System (BESS) that is currently in use as a grid management tool on the Big Island grid.  These tests shall demonstrate whether the electrolyzer has the capability to effectively mitigate the impacts of intermittent solar and wind power on the grid, while continuously generating 90-80% of its designed hydrogen productioncapacity.

The presentation will describe the H₂ES configuration, plans for operation in the field, and results of initial tests to determine the dynamic response of the eletrolyzer and overall system

These are exciting times to be pioneering Electrochemical Hydrogen Compression (EHC) as the call for linking renewable electrical energy and chemical energy storage is becoming stronger. Hydrogen is ideally suited for electrochemical conversion, thanks to its relative abundance in compounds on earth, extremely fast redox reaction and highest energy content by weight. Although we do not produce hydrogen, we are utilising the fast redox reaction to convert hydrogen gas at the membrane interface to protons, and back to hydrogen gas again.

The principles of Electrochemical Hydrogen Compression have been disseminated before and demonstrated the capability to achieve high pressures of 100 MPa single stage (14,503 psi), while simultaneously purifying the hydrogen gas. The company HyET is developing a product that will be suitable for multiple applications, ranging from (home) refuelling hydrogen vehicles to purification and recycling of industrial hydrogen (bio)gas.

Here we present the outcome of the project PHAEDRUS and DONQUICHOTE, where EHC technology was developed and validated as compressor for a Hydrogen Refuelling Station, further compressing the hydrogen gas supplied directly from a PEM electrolyser. Both technologies exhibit dynamic operation, fast response and modularity making them well compatible. The most effective system configuration to capture renewable wind and solar energy and deliver high pressure hydrogen has been investigated.

In addition, we review the purification ability of the EHC technology, extracting hydrogen selectively from mixed gasses in low and high concentrations. Hydrogen gas is typically produced from compounds (e.g. methane), its value is highly determined by its quality and any residual contaminations. The solid membrane was design to hold back high pressure hydrogen, and therefore shows superior permselectivity limiting passage of larger molecules. As for the active hydrogen extraction process, the membrane selection should be viewed in tandem with the catalyst system, which may be affected by specific contaminations.

Technical working principles are elaborated on in this presentation, pointing out key benefits and scientific challenges that were encountered and to some degree resolved, finally discussing the roadmap ahead leading towards very attractive applications.

Figure 1

One of the main concerns in the commercialization of fuel cell powered electrical vehicles is the on-board storage of hydrogen. Although pressurized vessels used in conventional systems provide higher hydrogen feeding rates, they have disadvantageous due to their extra weight and higher volumes in the systems. In this study, we report direct hydrogen generation from chemical hydrides, which does not require pressurized vessels. Self-dehydrogenation of chemical hydrides is a very slow process. In order to increase hydrogen generation kinetics, highly active and stable catalysts should be used. Although precious catalysts such as platinum and ruthenium meet these demands, their higher cost and recycling problems make them less useful in the fuel cell area.

A novel approach will be presented here for the fast, safe and stable hydrogen generation for fuel cells. This was achieved by circulating the acid and the chemical hydride solutions which were separated by the disulfonated poly(arylene ether sulfone) copolymer membrane. Since the rate of proton transfer is a function of ion exchange capacity (IEC) and scales with the degree of disulfonation of membranes, tailoring the IEC of membranes is crucial during the hydrogen generation. Therefore, membranes with varying sulfonation degrees (25-45 molar %) have been prepared and their hydrogen generation rates have been evaluated. Additionally, it was found that the type of acid and acid concentration influenced the hydrogen generation rates.

The effect of temperature on hydrogen generation performances was also investigated. When temperature exceeded the 80 °C, the hydrogen generation performance of N212 (the state of art membrane, Nafion™) was decreased due to the morphological changes occurred at this temperature. However the performances of our membranes without morphological relaxations continued to increase at higher temperatures. Furthermore, we run a proton exchange membrane fuel cell with our hydrogen generator and demonstrated continuous hydrogen generation more than 400 h. As a result, highly efficient on-board hydrogen generator for electrical vehicles powered by fuel cells presented here completely eliminates the need for separate facilities for producing hydrogen.

The properties of nanomaterials are known to be size depend. Such size depend effects could offer powerful means to finally control both the thermodynamic and kinetic properties of hydride materials at the molecular level [1] and thus enable the practical design of hydrogen storage materials from these effects. However, investigations through such an approach require new strategies to both synthesise and stabilise nanoparticles of highly reactive hydride materials and the understanding to control the key parameters that will allow reversibility with high storage capacity. The use of porous host structures offers such a route, however the storage capacity usually remains limited by the intrinsic difficulty of entirely fill the porosity.

Herein, the potential of a new core-shell confinement approach and recent progress we have made through this nanosizing method will be discussed [2]. Since, complex hydrides still undergo phase transitions including melting at the nanoscale, this core-shell approach has proven to be an very effective strategy to stabilise and simultaneously catalyse the reversible storage of hydrogen with hydrides including borohydrides and alanates (Figure 1). It also provides us with new means to stabilise against oxidation highly reactive nanoparticles of elements such as lithium.

References

[1] K.-F. Aguey-Zinsou and J.-R. Ares-Fernandez, Energy Environ. Sci., 3 (2010) 526.

[2] M. Christian and K.F. Aguey-Zinsou, ACS Nano, 6, 9, (2012) 7739.

Figure 1. (a) Approach for the synthesis of core-shell borohydrides and (b) TEM and associated elemental mapping of NBH₄@Ni, and (c) corresponding hydrogen cycling.

Figure 1

In rural areas, networkable environmental monitoring equipment is often used to provide continuous, real-time information without the need for a dedicated operator for maintenance or monitoring. However, power and energy availability are often controlling factors for operational lifetime when these systems are put on location. In these situations, batteries are often used as the main energy storage medium and require servicing or replacement when exhausted. This can lead high costs related to man-power and logistics to return and maintain the equipment at those locations.

Sodium borohydride hydrolysis was once looked at as a promising way to store hydrogen for use in PEM fuel cells. Concerns regarding hydrogen storage capacity and recyclability of reaction products have prevented its use in larger transportation or energy storage applications. However, the energy density of sodium borohydride still makes the reaction suitable for specific applications where reuse is not needed. As such, present here a system utilizing solid sodium borohydride as a storage medium to meet on demand power needs of these types of systems. The system utilizes a PEM fuel cell integrated with an electro-mechanical system to control chemical metering and power management.

We have previously shown results from our initial testing with solid sodium borohydride using commercial-off-the-shelf tablets for generating hydrogen when added to a reaction vessel containing water. Conceptually, a tablet would be introduced to the reaction vessel whenever additional hydrogen gas would be required. Operating in this manner would require good stability and reproducibility over multiple tablet reaction events.

Initial testing showed good reaction rates when a cobalt catalyst was employed, either pre-doped into the solid tablet mixture or when the catalyst was pre-mixed with the water in the reaction vessel. Analysis of the gas mixture showed high quality hydrogen gas flows could be produced and sent to the PEM fuel cell for power production. However, concerns with the stability, safety, and handling of the catalyst have led to a need to identify a more green chemistry approach.

We present here an alternative approach using common mineral and organic acids as hydrolysis accelerators that could be considered a greener approach to hydrogen generation for remote power applications. These include hydrochloric acid, phosphoric acid, citric acid, and acetic acid. Initial testing shows good reproducibility over multiple hydrogen generation events and reasonable reaction control. A comparison of hydrogen yield and gas composition is also shown. Finally, cost comparison between these accelerators is performed in order to assess their suitability in practical application.

The present study refers to a novel and unique approach of fabrication and deposition of gold nanoparticles on the surfaces of reduced graphene oxide and multi-walled carbon nanotubes. Among important issues is application of inorganic (rather than organic) capping ligand, heteropolymolybdate, to modify and stabilize (as well as probably also to link with the oxygen or hydroxyl groups on graphene surfaces) gold nanoparticles. During operation in alkaline medium or neutral media, polyoxometallates disappear but catalytically highly active gold remains and it exhibits excellent stability. The resulting material has occurred to show highly potent electrocatalytic properties toward electroreductions of carbon dioxide in neutral medium and oxygen in alkaline solution. What is even more important is that both carbon nanotubes and graphene have occurred to act effectively as carriers for gold nanostructures. Mutual activating interactions are feasible. The conclusions are reached on the basis of diagnostic electrochemical (e.g. rotating ring disk voltammetry), spectroscopic (FTIR) and microscopic (SEM, TEM) experiments. A series of comparative experiments with different carbon carriers and model catalytic materials (e.g. Vulcan-supported platinum) have also been performed. With respect to oxygen reduction, our diagnostic experiments at different concentrations of H₂O₂, support a view that the effect of the fast following chemical (H₂O₂-reductive-decomposition) reaction could be the dominating factor in explaining the observed positive potential shift observed during the oxygen reduction. The fact, that the optimum graphene-based catalytic system produced the oxygen reduction peak current comparable to that observed at the model platinum containing catalyst, would imply the efficient four-electron-type reduction mechanism.

Graphene and other distinct carbon nanostructures are explored as supports (carriers) for dispersed metallic silver nanocenters which exhibit electrocatalytic activity toward reduction of oxygen in alkaline media. To facilitate fabrication, immobilization and distribution of Ag(0) nanoparticles, various types of graphene (e.g. reduced graphene oxide, RGO, and CF_x graphene with carbon black) have been modified with the silver analogue of polynuclear Prussian Blue, namely with ultra-thin silver(I) hexacyanoferrate(II,III) layers. Following the heat-treatment step at temperatures as high as 400-600^oC, some thermal decomposition of the cyanometallate network occurs and, consequently, metallic silver sites are generated. Their formation and distribution are facilitated by the voltammetric potential cycling in KOH electrolyte. The most promising electrocatalytic results with respect to the reduction of oxygen (the highest currents and the most positive electroreduction potentials) have been obtained when graphene nanostructures are combined or intermixed with Vulcan XC-72R nanoparticles. What is even more important that, due to the presence of the polynuclear cyanoferrate modifier or linker, the amounts of the undesirable hydrogen peroxide intermediate are significantly decreased.

Understanding of the catalyst-ionomer interfacial behavior is critically important for fuel cell research. In fact, alkaline membrane fuel cells (AMFCs) have shown limited performance compared to proton exchange membrane fuel cells due to relatively poor hydrogen oxidation reaction (HOR) kinetics [1]. Previous work has shown that cationic adsorption at the surface of the catalyst was closely related to the HOR activity [2, 3]. Here, we present the results of a study of the benzyltrimethyl ammonium electrolyte-Pt interface in hydrogen oxidation reaction (HOR). More specifically, we will discuss the adsorption of the cationic group that may inhibit the HOR activity of Pt electrocatalysts. Microelectrode, Infrared Reflection Adsorption Spectroscopy (IRRAS) and Neutron Reflectometry are complementary tools to investigate the catalyst-ionomer interface.

Acknowledgement

US department of Energy Fuel Cell Technologies Program (Program Manager: Nancy Garland) supports this research through Fuel Cell Incubator Project. We also wish to acknowledge Oak Ridge National Laboratory Spallation Neutron Source, the NIST Center for Neutron Research and the Los Alamos Neutron Science Center.

References

[1] W.C. Sheng, H.A. Gasteiger, Y. Shao-Horn, "Hydrogen Oxidation and Evolution Reaction Kinetics on Platinum: Acid vs Alkaline Electrolytes," J. Electrochem. Soc., 157, B1529 (2010).

[2] H. T. Chung, U. Martinez, J. Chlistunoff, Y. S. Kim, Y. K. Choe, and I. Matanovic, "Impact of Organic Cation Adsorption on the Hydrogen Oxidation Reaction of Pt in Alkaline Fuel Cells", 2015, ECS Meeting Abstracts, 1510-1510

[3] Sung-Dae Yim, Hoon T. Chung, Jerzy Chlistunoff, Dae-Sik Kim, Cy Fujimoto, Tae-Hyun Yang, Yu Seung Kim, "A Microelectrode Study of Interfacial Reactions at the Platinum-Alkaline Polymer Interface," J. Electrochem. Soc., 162, F499 (2015).

Non-precious metal catalysts (NPMCs) for the oxygen reduction reaction (ORR) present an opportunity to improve the industrial feasibility of low temperature fuel cells by replacing expensive precious metal catalysts. However, improvements in performance of NPMCs are needed before they can realize widespread commercial adoption. Since the catalyst is projected to be the largest single contributor to the overall cost of the fuel cell, this area has seen heavy research activity. Many NPMC chemistries have been studied for the ORR in alkaline media, with perovskite oxides (of the form ABO₃) being of particular interest due to the large variety of possible compositions that form perovskite structures, and the ability to tune material properties through doping of the A and B sites.¹ Optimization of the bulk and surface chemistry to maximize intrinsic catalytic activity is of major importance to drive the development of perovskite oxide NPMCs. Synthesis strategies that balance surface chemistry and surface area optimization are necessary to improve the active site density while maintaining high catalytic activity. In addition, formation of effective perovskite oxide/carbon composites is essential to catalyze the ORR by a 2 step, 4 electron (2 per step) mechanism.²

This work focuses on the development of high performance perovskite/carbon composites by tuning surface chemistry, surface area, and optimizing the ratio of perovskite oxide to carbon. Through an aerogel synthesis process, a high surface area Ca_0.9La_0.1Al_0.1Mn_0.9O_3-δ perovskite oxide catalyst was produced. Calcination temperature was varied between 500 and 1000C to study the interplay between phase purity and the catalyst surface area. Rotating disk electrode (RDE) studies showed that the optimum balance between phase, purity, surface composition and surface area is obtained using calcination at 800C. Rotating ring disk electrode studies (RRDE) were used to evaluate the effect of perovskite oxide to carbon ratio with further investigations in membrane electrode assemblies (MEAs) also demonstrating a significant impact on the performance (Figure 1). Additional improvements in performance were realized by varying the catalyst loading and using carbon functionalized with nitrogen.

References

1. Hardin, W. G., Mefford, J. T., Slanac, D. A., Patel, B. B., Wang, X., Dai, S., ... Stevenson, K. J. (2014). Tuning the electrocatalytic activity of perovskites through active site variation and support interactions. Chemistry of Materials, 26(11), 3368–3376. http://doi.org/10.1021/cm403785q

2. Poux, T., Napolskiy, F. S., Dintzer, T., Kéranguéven, G., Istomin, S. Y., Tsirlina, G. A., ... Savinova, E. R. (2012). Dual role of carbon in the catalytic layers of perovskite/carbon composites for the electrocatalytic oxygen reduction reaction. Catalysis Today, 189(1), 83–92. http://doi.org/10.1016/j.cattod.2012.04.046

Figure 1

Introduction

Alkaline fuel cells (AFC) have attracted much attention because non-Pt-based catalysts can be potentially applied.

Metallocomplex-based catalysts are the one of the most prospective candidates of non-Pt-based oxygen reduction reaction (ORR) catalysts. These catalysts have several advantages over Pt catalysts. First, since active sites consist of isolated metal atoms in these catalysts, metal resource can be used economically. Second, the catalytic activity can be improved by modifying structure of the ligand. Among these metallocomplexes, iron complexes-based catalysts are intensively studied because of low cost and relatively high activity.

We have been particularly interested in tetraazaannulene (TAA) iron complexes. The TAA complex is one of the N₄-type macrocyclic complexes, as shown in Fig. 1. The main advantage of TAA ligands is that the molecular size is smaller than those of other conventional N₄macrocyclic ligands such as porphyrins and phthalocyanines. The density of small molecules on an electrode can be increased. Furthermore, the substituents would have strong electronic effects on the active sites due to the small molecular size. Especially, the introduction of a substituent into α- or β- carbon located at diketiminate motif (Fig. 1) is believed to give large electronic effects on the active center. This indicates that the ORR activity of TAA complexes might be enhanced by the introduction of suitable substituents.

Experimental studies on ORR activities of iron complexes of TAA have not been conducted to date, while computational studies reported that iron complexes of TAA and its derivatives have good ORR activity¹. In this work, we investigate the redox property and ORR of the iron complexes ligated by TAA (Fig. 1 (a)) and TAA-NO₂ (an electron withdrawing derivative, Fig. 1 (b)) in alkaline conditions, and discuss the effect of the ligand structure on the electrochemical behavior.

Experiments

The ligands of TAA² and TAA-NO₂³ were synthesized according to the reported procedures. The iron complexes abbreviated as Fe-TAA⁴ and Fe-TAA-NO₂ were synthesized by refluxing the ligand with FeCl₂･4H₂O in an organic solvent in an inert atmosphere.

The iron complex-modified carbon materials were prepared as follows. Ketjenblack (KB) was added to the solution of the iron complexes. The mixture was evaporated to dryness.

The hydrodynamic voltammetry using a rotating ring(Pt)-disk(GC) electrode was performed in 0.1 M NaOH under 25 °C. A Pt coil and a reversible hydrogen electrode (RHE) were used as a counter electrode and a reference electrode, respectively. The iron complex-modified KB (Fe-TAA/KB or Fe-TAA-NO₂/KB) was immobilized on a glassy carbon (GC) electrode.

Results and discussion

The ORR activities of the Fe-TAA/KB and Fe-TAA-NO₂ were evaluated in the alkaline aqueous solution. Strong activities were observed; the onset potential, at which the value of the catalytic current is 0.2 mA cm^-2, reached 0.94 V (Fe-TAA) and 0.97 V (Fe-TAA-NO₂) vs. RHE. The onset potential was increased by the introduction of the -NO₂ substituents. In the both of the cases, few H₂O₂was detected at ring electrode. This indicates that the numbers of electron transfer number of ORR by these catalysts are close to 4. The correlation between the onset potentials and redox potentials will also be discussed.

Acknowledgement

The authors are grateful to Tokuyama Corporation for providing the anion exchange resin (AS-4).

References

1) A.L. Pereira Silva, L.F. de Almeida, A.L. Brandes Marques, H.R. Costa, A.A. Tanaka, A.B. Ferreira da Silva, and J.d.J. Gomes Varela Junior, Journal of Molecular Modeling, 20, 2131-2140 (2014).

2) Y.G. Yatluk, and A.L. Suvorov, Chemistry of Heterocyclic Compounds, 23, 316-320 (1987).

3) N. Nishiwaki, T. Ogihara, T. Takami, M. Tamura, and M. Ariga, J. Org. Chem., 69, 8382-8386 (2004)

4) Sustmann, R., Korth, H. G., Kobus, D., Baute, J., Seiffert, K. H., Verheggen, E., Bill, E., Kirsch, M., de Groot, H. Inorg. Chem., 46, 11416-11430 (2007)

Figure 1

Introduction

Biodiesel fuel (BDF) attracts attention as a carbon-neutral fuel as well as bioethanol. As the production of BDF increases, the production of the glycerol byproduct is also increasing. It is important to develop new uses for glycerol to cope with the mass production of BDF. The use of glycerol as a fuel for direct alcohol fuel cells is promising, because anodic oxidation of glycerol produced by bioprocesses occurs in carbon-neutral cycles and direct glycerol fuel cells are expected to produce electricity with a low environmental load and to be energy efficient.

It is well-known that Pd is cheaper than Pt and Au, and Pd-based alloy electrodes are highly active for the oxidation of alcohols in alkaline media. We also have reported the PdAg alloys-loaded carbon exhibited higher glycerol oxidation reaction (GOR) activity and tolerance to poisoning species than Pd.¹ The mechanism for GOR, however, is not so clear. In this study we prepared an Ag atomic layer-modified Pd model electrode, and investigated the potential dependence of GOR products by in situ infrared reflectance-absorption spectroscopy (IRAS). In addition, we discussed the GOR mechanism based on these data.

Experimental

The Ag atomic layer-loaded Pd (Ag/Pd) electrode was prepared by underpotential deposition of Cu (Cu-upd) on a Pt polycrystalline substrate and the following galvanic replacement with Ag. The coverage of Ag (θ_Ag) on the Pd substrate was evaluated to be 0.5 from the charges for Cu-upd before and after the Ag modification. 1 M KOH and (1 M KOH + 0.5 M glycerol) solution were used for electrochemical measurements. In situ infrared reflection-absorption spectroscopy (IRAS) was used to qualitatively analyze the products of GOR on the Pd and Ag/Pd electrodes in alkaline solution. All electrochemical measurements were performed at room temperature.

Results and Discussion

The cyclic voltammograms of the Pd and Ag/Pd electrodes in a (1 M KOH + 0.5 M glycerol) solution is shown in Fig. 1. The Ag/Pd electrode had higher oxidation current and more negative onset potential of the oxidation current than the Pd electrode, clearly indicating that the GOR activity was enhanced by the modification of the Ag atomic layer. In addition, the peak potential of the oxidation current for the Ag/Pd electrode was more positive than that for the Pd electrode, suggesting that the former was superior in tolerance to poisoning species to the latter.

IRAS spectra for the Pt electrode exhibited that the absorption peak at 1335 cm^-1 assigned to the formation of dihydroxyacetone was mainly increased at lower potentials, but the absorption peak at 1310 cm^-1 assigned to the formation of glycerate was increased at higher potentials. This suggests that the secondary OH group in a glycerol molecule is preferentially oxidized at smaller overpotentials, and the primary OH group is greatly oxidized at larger overpotentials. For the Ag/Pd electrode, the oxidation of the primary OH group occurred even at smaller overpotentials, and hydroxypyruvate was also formed at larger overpotentials. 

Acknowledgement

This work was partially supported by JSPS KAKENHI Grant Number 15H04162.

Reference

1 B. T. X. Lam, M. Chiku, E. Higuchi, H. Inoue, J. Power Sources, 297, 149 (2015).

Figure 1

The rapidly increasing energy demand for human activities stimulates the lasting research interests to develop renewable energy alternatives worldwide. The electrochemical reduction of oxygen is one of key steps in controlling the performance of various next-generation energy conversion and storage devices, such as fuel cells, metal-air batteries. The electrochemical reduction of oxygen is one of key steps in controlling the performance of various next-generation energy conversion and storage devices, such as fuel cells, metal-air batteries. The commercialization of these technologies prominently depends on the development of low-cost high-performance electrocatalysts for oxygen reduction reaction (ORR) to replace the precious metal-based catalysts.

In this presentation, several reasonable ways for designing new low-cost nanocatalysts with high electrocatalytic activities and superior stability for ORR will be discussed. By focusing on the creation and the enrichment of highly active sites for ORR and simultaneously considering the mass transfer and electron transportation, we have developed several efficient ORR nanocatalysts.^1-7 The further improvement of the performance can be achieved by introducing transition metal or nanostructures into these nanocatalysts. Furthermore, understanding the origin of high activity of these electrocatalysts in ORR is also critical for developing efficient non-precious metal catalysts but still challenging. We developed a new highly active Fe-N-C ORR catalyst containing Fe-N_x coordination sites and Fe/Fe₃C nanocrystals, and revealed the origin of its activity by intensively investigating the composition and the structure of the catalyst and their correlations with the electrochemical performance. Based on our experimental and theoretical results, it can be concluded that the high ORR activity in this type of Fe-N-C catalysts should be ascribed to that Fe/Fe₃C nanocrystals boost the activity of Fe-N_x. These new findings open an avenue for the rational design and bottom-up synthesis of low-cost highly active ORR electrocatalysts.

References

[1] W. Ding, Z.D. Wei, S.G. Chen, X.Q., T. Yang, J.S. Hu, D. Wang, L.J. Wan, S.F. Alvi, L. Li, Angew. Chem. Int. Ed., 2013, 52, 11755.

[2] Y.P. Xiao, W.J. Jiang, S. Wan, X. Zhang, J.S. Hu, Z.D. Wei, L.J. Wan, J. Mater. Chem. A.2013, 1, 7463.

[3] W.J. Jiang, J.S. Hu, X. Zhang, Y. Jiang, B.B. Yu, Z.D. Wei, L.J. Wan, J. Mater. Chem. A.2014, 2, 10154.

[4] R.J. Huo, W.J. Jiang, S.L. Xu, F.Z. Zhang, J.S. Hu, Nanoscale, 2014, 6, 203.

[5] Y. Zhang, W.J. Jiang, X. Zhang, L. Guo, J.S. Hu, Z.D. Wei, L.J. Wan, Phys. Chem. Chem. Phys., 2014, 6, 13605.

[6] L. Guo, W.J. Jiang, Y. Zhang, J.S. Hu, Z.D. Wei, L.J. Wan, ACS Catal.,2015, 5, 2903.

[7] Y. Zhang, W.J. Jiang, X. Zhang, L. Guo, J.S. Hu, Z.D. Wei, L.J. Wan, ACS Appl. Mater. Interfaces,2015, 7, 11508.

[8] Y. Zhang, W.J. Jiang, X. Zhang, L. Guo, J.S. Hu, Z.D. Wei, L.J. Wan, J. Mater. Chem. A.2016, DOI: 10.1039/C6TA01655C.

[9] W.J. Jiang, L. Gu, L. Li, Y. Zhang, X. Zhang, L.J. Zhang, J.Q. Wang, J.S. Hu, Z.D. Wei, L.J. Wan, J. Am. Chem. Soc., 2016, 138, 3570.

Fuel cells are power sources converting chemical energy into electrical energy through electrochemical reactions occurring on catalyst surfaces, in general, on Pt or Pt alloy surfaces. Hydrogen is mostly and widely used as the fuel, but other fuels such as methanol, ethanol and formic acid are investigated as alternatives because of their higher energy density as well as easy transport in our society. Among these alternative systems, direct methanol fuel cells (DMFCs) have been developed and considered for applications in electronics because of their high theoretical energy density, simple systems and ease of handling. In contrast to these merits even over the hydrogen-based systems, the complicated and slow electrooxidation reactions of methanol have hampered their commercialization. In this light, a variety of electrocatalysts for methanol oxidation reactions have been synthesized and evaluated in order to enhance the sluggish reactions, in other words, to reduce a fuel cell stack cost by decreasing the amount of Pt usage (e.g., alloying with other metals, making Pt monolayer on other metals, making composites with other substances to have synergy effects, and controlling the shapes to utilize more active surface structures).

The carbon supported Pt nanoparticles (Pt/C) have been recognized as the most reasonable starting point for each fundamental research, though alloys are widely used in the research and commercial products for achieving better performance. Pt and its alloys have the best activities in methanol oxidation reactions (MOR) and oxygen reduction reactions (ORR), and, in order to utilize Pt surface effectively, Pt/C catalysts have been generally prepared, e.g., by an impregnation method involving a wet-chemical reduction (i.e., impregnation of the carbon supports withsolutions containing the metal salts to be deposited). This method is quite simple and is easy for large-scale production. Very recently, we have proposed a new approach for preparation of Pt/C catalysts by arc plasma deposition (APD) method, where Pt nanoparticles can be deposited on carbon in vacuum and has been applied for various fields other than fuel cell catalyst. Herein, we focus on a preparation of new MOR and ORR catalysts by using our coaxial pulse APD method, where two different metal sources (Pt and Ti) can be deposited on carbon support through vacuum processes.

The morphology of the obtained sample was characterized by the scanning electron microscope (SEM) and the transmission electron microscopy (TEM). The SEM images show a typical morphology of an activated carbon, and the TEM images show that the deposited nanoparticles are well dispersed on the carbon support and have an average size of ~3 nm. The elemental mapping data show that the Pt and Ti are homogeneously distributed on the entire of carbon support. X-ray photoelectron spectroscopy (XPS) measurement is carried out to ascertain the electronic state of Pt and Ti, and the Pt 4f_7/2 and Pt 4f_5/2 doublet peaks are observed at ~71.5 and ~74.5 eV, which mean the Pt is pure metallic Pt. Regarding core-level Ti XPS, the peaks located at 458.9 eV and 464.8 eV can be assigned to Ti(IV) 2p_3/2 and Ti(IV) 2p_1/2, respectively. Another peak located at 455 eV is probably attributed to the Ti³⁺. Considering the fact that the XRD pattern does not show any crystalline TiO_x phases, the Ti should form amorphous TiO_x phases. Thus, the materials are found to be Pt nanoparticles (~3 nm) and amorphous TiO_x phases on carbon black.

Compared with other Pt-based catalysts, our catalysts show superior electrochemical performances in MOR and ORR, probably owing to the presence of the secondary metal oxide phase. The cyclic voltammograms (CVs) of methanol oxidation reaction in 0.5 M H₂SO₄ containing 0.5 M methanol solution were scanned for the synthesized catalyst, compared with pure Pt catalyst on carbon support prepared by the APD method, commercially available Pt/C catalyst, and Pt black. All the current densities were normalized by Pt mass (i.e., Pt mass activity), and our catalyst exhibits higher activities than the other catalysts, probably because of bifunctional mechanism as found in PtRu or other systems, where the metal oxide surface provides active oxygen for removal of intermediates, such as CO, on the Pt surface effectively. The ORR performance of our catalysts were preliminarily investigated by linear sweep voltammetry (LSV) using a rotating disk electrode in O₂-saturated 0.1 M HClO₄ solution at a scan rate of 10 mV s^-1 and 1600 rpm. The synthesized PtTi/C and the synthesized Pt/C show almost the same onset potentials, and show the better mass activity than the commercially available products. More details will be presented on the day.

With the dramatically increased market demands of new energy materials due to the development of a large number of electronic devices and environmental awareness of the awakening, catalysts for oxygen reduction reaction (ORR) which are the most critical parts for high energy capacity and cleaning power sources including fuel cells and metal-air batteries have attracted highly attention and intense researches recently. Despite massive efforts, developing catalysts of high efficiency, long-term durability and excellent methanol tolerance with low cost for ORR is still a big challenge. Pt and Pt–based alloys are still the best choices for catalyzing oxygen reduction process by now. However, their high cost, low abundance, inferior stability and declining activity hinder their large-scale uses in industry. Therefore, selecting alternative materials to replace Pt and improve its performance is of great concern.

Chitosan, as the second most abundant biopolymer in nature, is rich in the outer shell of some arthropod (prawns, crabs, insects) and the inner shell and cartilage of mollusks (squid, cuttlefish) with some attractive properties, such as non-toxic, biocompatibility and safety, especially, its natural nitrogen rich, strong chelation effects, sustainable and environmental friendly and low cost depend that it can be heavy used to synthesize metal and nitrogen co-doped carbon (MNC) nanomaterials , which can provide an enhanced catalytic activities.

In the present work, taking advantage of strong chelation effects between hydroxyl group and metal ions and abundant natural nitrogen of chitosan, we developed a novel Co and N co-doped carbon (Co-N-C) nanocomposites as high performance electrocatalysts for ORR by using a simple and facile hydrothermal method, followed by high-temperature calcination at inert gas atmosphere. This material shows a delectable catalytic activity compared to the commercial 20% Pt/C electrode in the alkaline solution for ORR with a relative positive onset potential of -0.045V (vs. Hg/HgO), a much improved by 16.7% current density than Pt/C (loading ~ 0.4mg/cm²) as well as excellent durability performance and superior methanol tolerance. In addition, the method we used in this research is also suitable for preparing other MNC, which demonstrates that we found a new, straightforward and low-cost way to synthesis a series of catalysts for ORR which are comparable to commercial Pt/C.

Cyclic voltammetry (CV) was employed to test the ORR catalytic activity of Co-N-C-800 and the commercial 20% Pt/C catalyst in saturated O₂ alkaline solution (0.1 M KOH) and shown in Figure 1a, an obvious cathodic peak can be found at -0.10 V vs. Hg/HgO which implies excellent ORR performance, and according to figure 1b, we can find that our Co-N-C-800 has a very close onset potential (-0.045V) and much higher current density (16.7% improved) in comparison to that of the commercial Pt/C, respectively, superior to the performances of pure chitosan without Co (C-800), which can confirm that Co-doping is crucial to the outstanding activity performance of Co-N-C-800. Owning to small metal nanocrystals of cobalt can improve the conductivity of materials and charge transfer efficiency between carbon and cobalt and give more active sites¹. RDE measurements at various rotating speeds at a scan rate of 10 mV s^-1 in O₂-saturated solution were carried out for Co-N-C-800 and shown in Figure 1c. The current density shows a typical increase with rotation rate due to the shortened diffusion layer and the linearity of K-L plots for Co-N-C-800 indicates a good four-electron reaction path selectivity with the electron transfer numbers (n) is ≈ 3.9, which is close to that of Pt/C catalyst and further demonstrates an outstanding catalytic performance of this catalyst.

Acknowledgments

This work was financially supported by the Shenzhen Peacock Plan (KQCX20140522150815065), the Natural Science Foundation of Shenzhen (JCYJ20150331101823677) and the Science and Technology Innovation Foundation for the Undergraduates of SUSTC (2015x19 and 2015x12).

Reference

1. Xie, S.; Huang, S.; Wei, W.; Yang, X.; Liu, Y.; Lu, X.; Tong, Y., Chitosan Waste-Derived Co and N Co-doped Carbon Electrocatalyst for Efficient Oxygen Reduction Reaction. ChemElectroChem 2015,2 (11), 1806-1812.

Figure 1

The development of energy storage and conversion devices provides a beneficial approach for the renewable energy application, which can help relieve the severe reliance on fossil fuels and also address problems related to global climate change. Currently, the efficiencies of energy conversion devices such as the metal–air batteries and fuel cells are mainly limited by the sluggish kinetics of the oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) processes. Developing catalytically active and cost effective catalyst for the ORR and OER is of prime importance. So far, noble metals and their alloys, such as Pt and Pt-Pd, have been exclusively used as electrochemical bifunctional catalyst in ORR and OER due to their superior catalytic activity. However, the high cost, limited availability, poor durability and sluggish electron transfer kinetics of noble metal based bifunctional catalysts have impeded the practical application of metal-air batteries. Therefore, the discovery of cost effective, catalytically active alternative bifunctional catalyst such as non-precious metals, carbonaceous materials, and transition metal oxides is highly desirable.

Perovskite oxide (ABO₃) possesses a unique electronic structure and chemical defect properties, and has been demonstrated to be a promising non-precious metal catalyst among the transition metal oxides in the application of metal air batteries and alkaline fuel cells. It is known that the intrinsic electrochemical catalytic activity is mainly determined by the B site cation in perovskite oxide. Extensive research conducted by Yang et al. demonstrated that the ORR and OER catalytic activities of perovskite oxides follow a volcanic relationship with the filling of electrons in antibonding orbitals [1]. Among the typical LaMO₃ (M= Mn, Co, Fe, Ni, Cr), LaMnO₃ and LaCoO₃ have exhibited highest ORR and OER catalytic activities, respectively. In addition, due to the desirable electronic properties of the perovskite oxides, strategies such as cation partial substitution and oxygen non-stoichiometry formation could, therefore, be utilized as the effective approaches to fine-tune the catalytic properties and to achieve a better bifunctional activity [2,3].

To enhance the bifunctional catalytic performance, improved specific surface area as well as enhanced oxygen channel are desired. Researchers have found that electrochemical catalysts with porous nanofiber structures could favor the ORR and OER pathways by providing uniform O₂ electrolyte distribution, and beneficial oxygen diffusion channels. Zhao et al. have synthesized mesoporous La_0.5Sr_0.5CoO_2.91 nanowires through the multistep microemulsion method, showing significant enhanced catalytic performance [4]. Xu et al. have demonstrated that the porous La_0.75Sr_0.25MnO₃ nanotube catalyst fabricated through facile electrospinning technologies provides favorable advantages to the availability of the catalytic active sites in the organic solvent electrolyte in Li-O₂ batteries [5].

Herein, we developed a bifunctional perovskite catalyst towards both ORR and OER in alkaline solution. Cobalt cations were doped into Pr_0.5Ba_0.5MnO_3-δ perovskite to achieve the higher intrinsic ORR and OER activities by engineering the structure symmetry, electronic and the oxygen vacancy defects of the material. The bifunctional catalyst with a unique porous nanofiber structure was fabricated by electrospinning technology (Figure.1). A single phase perovskite oxide Co doped Pr_0.5Ba_0.5MnO_3-δ was obtained after sintering the electrospun precursor at high temperature. The ORR and OER activities of the composite catalysts consisting 50 wt% of the as-synthesized perovskite oxides and 50 wt% carbon black were investigated in 0.1 KOH with rotating disk electrode (RDE). Significant enhancement in ORR and OER performance was achieved via using the composite catalyst with Co doped Pr_0.5Ba_0.5MnO_3-δ nanofiber/carbon black with respect to the reduced overpotential and improved current density. In particular, the Co doped Pr_0.5Ba_0.5MnO_3-δ nanofiber/carbon black composite demonstrated enhanced electron transfer number in the ORR process, indicating preferable dominance of four electron transfer pathway. Moreover, the Co doped Pr_0.5Ba_0.5MnO_3-δ composite exhibited high stability during the cycling tests, indicating its promising applications in metal-air batteries.

References:

(1) Suntivich, Jin, et al. Nature chemistry 3.7 (2011): 546-550.

(2) Zhu Yinlong, et al. Chemistry of Materials, 28.6 (2016):1691−1697

(3) Jung, Jae-ll, et al. Angewandte Chemie 126.18 (2014): 4670-4674.

(4) Zhao, Yunlong, et al. Proceedings of the National Academy of Sciences 109.48 (2012): 19569-19574.

(5) Xu, Jing Ji, et al. Angewandte Chemie International Edition, 52.14 (2013): 3887-3890.

Figure 1

Manganese oxide/poly(3,4-ethylenedioxythiophene) (MnOx/PEDOT) nanostructured hybrid thin films are readily prepared using a simple anodic electrodeposition process from aqueous solution. Using this approach, MnOx/PEDOT films with thicknesses and mass loadings up to 100 nm and 40 μg cm-2 were prepared, then tested for oxygen reduction reaction (ORR) activity in alkaline electrolyte using rotating disk electrode and rotating ring disk electrode methods. MnOx/PEDOT provided improvements over MnOx-only and PEDOT-only control films, with > 0.2 V decrease in onset and half-wave overpotentials and > 1.5 times increase in current density. The MnOx/PEDOT film exhibited only a slightly lower reaction order (n = 3.86-3.92) than the 20% Pt/C benchmark electrocatalyst (n = 3.98) across all potentials. MnOx/PEDOT also displayed a more positive half-wave potential and superior electrocatalytic selectivity for the ORR upon methanol exposure than 20% Pt/C. The high activity and synergism of MnOx/PEDOT towards the ORR is attributed to effective intermixing/dispersion of the two materials, intimate substrate contact and the improved charge transfer processes attained by co-electrodepositing MnOx with PEDOT. PEDOT has previously been demonstrated in fuel cells to limit methanol crossover. PEDOT has also been used as a catalyst support in fuel cells. These facts, combined with the substantially lower cost of Mn versus Pt (Mn is ~ 17,000 times cheaper), augurs for further development and testing of MnOx/PEDOT hybrid materials as ORR electrocatalysts with application in alkaline fuel cells.

This work was supported by the Laboratory Directed Research and Development program at Sandia National Laboratories, a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.

The slow kinetics of the oxygen reduction reaction (ORR) at the cathode and high cost of electrocatalysts (commonly Pt and Pt-based alloys) for this reaction are major hurdles that hinder the large scale production of proton exchange membrane fuel cells (PEMFCs).[1] A core-shell structure consisting of an atomically thin layer (up to several atoms thick) of Pt deposited on Pd has been recognized as a promising ORR electrocatalysts to replace bulk Pt-based materials.[2] Ideally, all of the Pt atoms will be fully utilized since they are all on the surface in the case of an atomically thin layer of Pt on a core material. The large scale synthesis of Pd@Pt core-shell catalysts, however, has not been fully realized. This study reported two synthesis methods to prepare gram scale batch of Pd@Pt/C electrocatalysts with the assistance of citric acid.

The first is Cu-mediated-Pt-displacement method involving the displacement of an underpotentially deposited (UPD) Cu monolayer by Pt.[3] In this method, the Cu monolayer deposited on Pd then is displaced by Pt via a surface limited redox replacement (SLRR) reaction: Pd@Cu + PtCl₄^2- → Pd@Pt + Cu²⁺ + 4Cl^-. Recent studies have shown that the SLRR reaction is rather complicated and poorly-controlled resulting in formation of 3D Pt islands instead of a uniform overlayer, especially in the large batch size synthesis. By adding citric acid in the displacement reaction, the activity of Pd@Pt/C can be enhanced by 2 times compared with that synthesized without citric acid.

The second method is direct chemcial reduction of PtCl₄^2- with the assistance of citric acid without preformation of a Cu UPD layer. Comparing with other reducing agents, the reducing power of citric acid is neither too strong nor too week and allows the reduction reaction to occur at the room temperature, which significantly simplifies the reaction process. In addition, the strong interaction of citric acid with metals may reduce the Pd dissolution and introduce charge transfer from adsorbed citric acid molecules to Pd surface, which can also be used to reduce Pt cations. 

Both methods resulted good quality of core-shell catalyts in terms of activity and durability. The Pt activities of Pd@Pt/C preapred by the Cu-mediated-Pt-displacement and chemcial reduction methods were 0.95 and 0.78 A mg^-1 at 0.9 V, respectively. Fuel cell performance will be also reported and the role of citric acid will be discussed in the presentation. 

References

[1] in: M. Shao (Ed.) Electrocatalysis in Fuel Cells: A Non- and Low- Platinum Approach, Springer, London, 2013.

[2] R.R. Adzic, Electrocatalysis, 3 (2012) 163-169.

[3] R.R. Adzic, J. Zhang, K. Sasaki, M.B. Vukmirovic, M. Shao, J.X. Wang, A.U. Nilekar, M. Mavrikakis, J.A. Valerio, F. Uribe, Top. Catal., 46 (2007) 249-262.

Introduction

Carbon supported Pd core-Pt shell catalyst (Pt/Pd/C) and Pt-Pd alloy catalyst (PtPd/C) are attractive alternatives to a carbon supported Pt catalyst (Pt/C) because their ORR specific activities are enhanced with an accelerated durability test (ADT) performed at 80°C [1, 2]. However, ECSA of the catalysts drastically decreased with the ADT, resulting in slight increase of ORR mass activities. We explored the ADT protocol and developed a high activation protocol (HAP) to mitigate the ECSA decay and enhance the ORR mass activities. Furthermore, we developed a Cu-O₂ treatment which mimics the HAP carried out on GC electrode and is suitable for mass-production of highly activated catalysts. We also applied SiO₂ coating technique [3, 4] to prevent agglomeration of the catalyst particles with the ADT and improve durability of the catalysts.

Experimental

Pt/Pd/C was synthesized with a modified Cu-UPD/Pt replacement process [1]. A carbon supported Pd core (Pd/C, Pd size: 4.6 nm, Pd loading: 30 wt.%, Ishifuku Metal Industry) was dispersed in 50 mM H₂SO₄ containing 10 mM CuSO₄ and the solution was stirred with co-existence of a metallic Cu sheet at 5°C under Ar atmosphere. After stirring for 5 h, the Cu sheet was removed and K₂PtCl₄ was added to replace under potentially deposited Cu shell on the Pd core surface with Pt shell. PtPd/C alloy catalyst was synthesized with an impregnation method. Pt(acac)₂ and Pd(acac)₂ were impregnated onto a carbon support using t-butylamine as a solvent (Pt₂₀Pd₈₀ in at.%), followed by reduction under 15% H₂/Ar atmosphere at 600°C for 4 h. The catalysts were characterized by TG, XRF, XRD, TEM, CV and XAFS techniques. ORR activity of the catalysts was evaluated with RDE technique in O₂ saturated 0.1 M HClO₄ at 25°C. ADT was conducted using rectangular wave potential cycling of 0.6 V (3 s)-1.0 V (3 s) vs. RHE in Ar saturated 0.1 M HClO₄ at 80°C for 10,000 cycles.

Results and Discussion

Changes in ECSA and ORR mass activity of the Pt/Pd/C and PtPd/C catalysts with ADT, HAP and Cu-O₂ treatment are summarized in Fig. 1 (dotted black lines show Pt/C catalyst; Pt size: 2.8 nm, Pt loading: 46 wt.%, TEC10E50E, TKK). The ECSA of the catalysts drastically decreased with the ADT, by which the ORR mass activity showed slight increases. On the contrary, the ECSA decay was mitigated with the HAP (rectangular wave potential cycling of 0.4V (300 s)-1.0 V (300 s) vs. RHE in Ar saturated 0.1 M HClO₄ at 80°C for 30-40 cycles) and the ORR mass activity was highly enhanced, showing ca. 3-fold values of the Pt/C catalyst.

Since the HAP is performed on the GC electrode, treated amount of the catalysts is extremely small (ca. 30 µg), we developed a Cu-O₂ treatment (Fig. 2) to scale-up the HAP. In the Cu-O₂ treatment, the catalyst powders are stirred for 300 s in 2 M H₂SO₄ containing 0.1 M CuSO₄ at 80°C under an inert atmosphere (N₂) with co-existence of a metallic Cu sheet, where the equilibrium potential of Cu²⁺/Cu (ca. 0.3 V) is applied to the catalysts when they contact with the Cu sheet. Next, the Cu sheet is removed and O₂ gas is introduced in the solution, where the equilibrium potential of the ORR (ca. 1.0 V) is applied to the catalysts. As shown in Fig. 1, the ECSA decay by the Cu-O₂ treatment is equivalent to that by the HAP and the ORR mass activity of the catalysts were equivalently enhanced, indicating that the Cu-O₂ treatment mimics the HAP on the GC electrode and is suitable for mass-production of highly activated Pt/Pd/C and PtPd/C catalysts.

However, the highly activated Pd@Pt NPs were severely agglomerated with the ADT and the ORR mass activity was decayed. In order to prevent the NPs agglomeration with the ADT, we applied SiO₂ coating onto the Pd@Pt NPs [3, 4]. The SiO₂ coating suppressed the NPs agglomeration and the ECSA decay was mitigated (Fig. 3), which highly enhanced the ORR mass activity of the Pt/Pd/C catalyst even after the ADT (Fig. 4).

Acknowledgement

This work was supported by NEDO, Japan.

References

[1] M. Inaba and H. Daimon, J. Jpn. Petrol. Inst., 58(2), 55 (2015).

[2] K. Okuno et al., The 228^th Electrochemical Society Meeting, #1405, Phoenix, USA, October 2015.

[3] S. Takenaka et al., J. Phys. Chem. C, 111, 15133 (2007).

[4] S. Takenaka et al., Appl. Catal. A Gen., 409- 410, 248 (2011).

Figure 1

Introduction

Pd@Pt core-shell catalyst is one of the promising cathode electrode materials for polymer electrolyte fuel cell (PEFC) from the view-points of a great potential to reduce Pt usage and to improve the oxygen reduction reaction (ORR) activity of Pt [1]. Recently, some researchers have reported that ORR activities of the core-shell catalysts increase during applying potential cycles, as a result of structural changes at the near surface regions [2]. Their results suggest that comprehensive understandings of dynamic behavior of the surface atomic structures during potential cycles are crucial for developing highly-active and highly-durable Pd@Pt core-shell catalysts. In this study, we investigate the relation between surface structures and electrochemical stabilities for the Pt/Pd(111) model electrocatalysts prepared by molecular beam epitaxy (MBE).

Experimental

Sample fabrication processes of the Pt/Pd(111) were conducted in UHV. Pd(111) single crystal substrate was cleaned by repeating cycles of Ar⁺ sputtering and subsequent annealing at 800 °C. 4ML-thick Pt was deposited onto the cleaned Pd(111) (4ML-Pt/Pd(111)) by an electron-beam evaporation method at the substrate temperatures of 300 °C. The resulting surface structures were verified with reflection high-energy electron diffraction (RHEED), scanning tunneling microscope in UHV (UHV-STM), and low-energy ion scattering (LEIS). Then, the prepared samples were transferred without being exposed to air to an electrochemical system set in a N₂-purged glove box. Cyclic voltammogram (CV) of the samples were recorded in N₂-purged 0.1 M HClO₄, and, then, linear sweep voltammetry (LSV) was conducted by using a rotating electrode (RDE) method after saturating the solution with O₂. The ORR activities were estimated by kinetic-controlled current density (j_k) at 0.9V vs. RHE by using Koutecky-Levich plots. The electrochemical stabilities were evaluated by applying potential cycles (PCs) between 0.6 V and xV (x = 0.8, 0.85, 0.9 and 1.0V) vs. RHE for each 3 seconds in O₂-saturated 0.1 M HClO₄ at 80 °C. The potential cycle conditions are hereafter denoted as "xV-PCs".

Results and Discussion

Surface structures for the clean Pd(111) and 4ML-Pt/Pd(111) are summarized in Fig.1 (a) and (b). The UHV-STM image indicates that the clean Pd(111) surface has atomically flat terraces. On the other hand, UHV-STM image of the 4ML-Pt/Pd(111) shows islands-like structures having hexagonal-shaped terraces with ca. 20 nm width. However, the corresponding RHEED pattern indicates sharp streaks, suggesting that deposited Pt atoms have epitaxially grown (111) lattice on the substrate.

Initial ORR activities for the 4ML-Pt/Pd(111), clean Pt(111) surfaces are shown in Fig.1 (c). The results indicates that the 4ML-Pt/Pd(111) exhibits ca. 4.5 times higher ORR activity than the clean Pt(111).Fig.1 (d) summarizes changes in ORR activities of the 4ML-Pt/Pd(111) during applying PCs at 80 °C. The activity of 1.0V-PCs applied sample drastically decreased with increasing PC numbers and dropped to less than that of the clean Pt(111) at 1000 PCs. However, it can be seen that decrease in upper limit potential of PCs suppress activity deterioration of the 4ML-Pt/Pd(111). In particular, the activities of 0.8V- and 0.85V-PCs applied samples after 5000 PCs were 120 % and 90 % of the initials, respectively. To investigate effects of surface structural changes in the ORR activity, the surface structures and compositions were evaluated by UHV-STM and LEIS after 5000 PCs. LEIS spectra (Fig.1 (e)) obtained after the PCs indicate that Pd surface concentration of 0.9V-PCs applied sample is higher by ca. 30 % than 0.8V-PCs applied sample, suggesting that activity deterioration during PCs is derived from increase in surface composition of Pd which show much lower ORR activity than Pt. Furthermore, judging from the corresponding UHV-STM images, although height roughness of the 0.8V-PCs applied surface is suppressed to lower than 0.7 nm, the 0.9V-PCs applied surface comprises agglomerated particle-like structures with 20 nm width and 3 nm heights. These results suggest that the structural and electrochemical stabilities for Pd@Pt core-shell catalysts strongly depend on operating conditions of PEFC.

Acknowledgement

This study was supported by the new energy and industrial technology development organization (NEDO) of Japan.

References

[1] R. Adzic, et al., Top. Catal., 46, 249 (2007).

[2] a) H. Daimon et al., 226^th ECS Meeting, Abstract #1063, Cancun, Mexico (2014); b) K.A. Kuttiyiel, et al., Electrochim. Acta, 110, 267 (2013); c) W. Liu, et al., Angew. Chem. Int. Ed., 52, 9849 (2013).

Figure 1

Developing a novel catalyst for oxygen reduction reaction (ORR) for real world applications should be focused on not only optimizing its activity and stability, but also synthesizing it by a simple process and thus being capable for mass production cost effectively. By taking a reliable recipe and facile one-pot synthesis method, we fabricated a wide range of unique nanoframe structures with high complexity, including the octopod, polyhedral, and hierarchical nanoframe architectures. Specially, the octopod nanoframe architectures possess three-dimensional catalytic surfaces, favorable platinum atomic arrangement and beneficial high-index facets (HIFs), thus can serve as models of advanced catalysts. The PtCu octopod nanoframe catalyst achieved 20-fold enhancement in mass activity for ORR compared to the commercial platinum-carbon catalyst. More importantly, this nanoframe catalyst maintained its structure over storing for months, after rigorous electrochemical reactions, and heat-treatment process. We also demonstrated that the one-pot synthesis of high-performance octopod nanoframe catalysts can be easily scaled up in high quality.

Fig. 1.Structural and compositional analysis of typical Pt-based nanoframe architectures.

Figure 1

Fuel cells are a promising source of renewable, clean energy. Widespread commercialization of fuel cell technology, however, has been hampered by technological and economic hurdles due to inadequate device efficiencies and expensive Pt electrocatalysts. To address these challenges, considerable research has focused on enhancing the kinetics of the oxygen reduction reaction (ORR) at the cathode which is the dominant source of voltage loss in state-of-the-art fuel cells.¹ Recent work has sought to increase intrinsic catalyst activity and reduce Pt loading via core-shell nanoparticles where a Pt shell surrounds the core of a different metal or alloy. These nanostructures provide the ability to increase Pt mass activity through a larger Pt surface area to volume ratio as well as enhance the Pt intrinsic activity by altering its electronic structure though stain and ligand effects. These electronic changes dictate the binding energies of reaction intermediates to the catalyst surface and thus reactivity for the ORR. To identify a core material capable of improving ORR activity, we utilized a systematic approach for catalyst design developed in a previous study. This study found that activity can be improved by selecting a core material that is theoretically predicted to over-weaken the oxygen binding energy when covered with a Pt monolayer.² The oxygen binding energy can then be strengthened towards its optimum by increasing the Pt shell thickness and through nanoscale effects where undercoordinated sites demonstrate stronger binding to adsorbates. Through this method, iridium was identified as a promising core material. Therefore, this work focuses on the synthesis, characterization, activity, and stability of Ir-Pt core-shell nanoparticles. Ir-Pt nanoparticles with three different compositions were synthesized using a highly scalable, inexpensive polyol method. TEM, STEM-EDS, and XPS were used to characterize particle size and composition. ORR activities of the Ir-Pt catalysts were compared to synthesized Ir-only and Pt-only nanoparticles as well as to the state-of-the-art commercial standard Pt catalyst, TKK. Electrochemical analysis demonstrates the high activity of the Ir-Pt catalyst with a specific activity of approximately 2 times that of TKK. This catalyst also demonstrates high stability after 10,000 stability cycles showing improvements in both mass and specific activity upon cycling with the latter increasing to 2.6 times that of TKK.

1. Stephens, I.E.L., et al., Energy & Environ. Sci. 5, 6744-6762, (2012).

2. A. Jackson et al., ChemElectroChem 1, 67-71 (2014).

In order to simultaneously meet the automotive PEMFC targets of cost, performance and durability, Pt-alloys electrocatalysts are being intensively investigated.^1,2,3 Pt-alloys show promise of higher ORR mass activities that will permit the use of lower Pt loadings (~0.1 mgPt/cm²) in the cathode catalyst layers. In this work we have studied several Pt-Co/C catalysts that exhibit mass activities that exceed the DOE target of 440 mA/mg_Pt at 0.90V, 80^oC, 100% RH for ORR activity. The performance at rated power density of 1W/cm² at 0.60V is more difficult to achieve at low loadings due to unanticipated transport resistances. In addition to carrying out ORR activity studies both in TF-RDE set-ups^4,5 as well as in MEAs of subscale cells, we have also applied complementary diagnostics of wet and dry H₂-Air curves, microscopy of catalyst powders and catalyst layer films, EIS and dilute oxygen limiting current studies⁶ on these materials. Rigorous base-lining was carried out to obtain the ORR activity of a commercial Pt/C catalyst in order to benchmark the Pt-Co/C catalysts under study using identical protocols.^7,8 Although we easily achieve the mass activity targets at 0.90V, we only approach the rated power targets indicating that further understanding of the transport resistances at high current densities and low Pt loadings is necessary to identify and mitigate the losses.

Acknowledgements: This work was funded through the DOE FC-PAD Consortium.

References Mathias MF, Makharia R, Gasteiger HA, Conley JJ, Fuller TJ, Gittleman CJ, Kocha SS, Miller DP, Mittelsteadt CK, Xie T, Van SG. Two fuel cell cars in every garage. Electrochem. Soc. Interface. 2005;14(3):24-35.

• Gasteiger HA, Kocha SS, Sompalli B, Wagner FT. Activity benchmarks and requirements for Pt, Pt-alloy, and non-Pt oxygen reduction catalysts for PEMFCs. Applied Catalysis B: Environmental. 2005 Mar 10;56(1):9-35.

• Kongkanand A, Mathias MF. The Priority and Challenge of High-Power Performance of Low-Platinum Proton-Exchange Membrane Fuel Cells. The journal of physical chemistry letters. 2016 Mar 11;7:1127-37.

• Kocha SS, Zack JW, Alia SM, Neyerlin KC, Pivovar BS. Influence of ink composition on the electrochemical properties of Pt/C electrocatalysts. ECS Transactions. 2013 Mar 15;50(2):1475-85.

• Shinozaki K, Zack JW, Pylypenko S, Pivovar BS, Kocha SS. Oxygen Reduction Reaction Measurements on Platinum Electrocatalysts Utilizing Rotating Disk Electrode Technique II.

• Baker DR, Caulk DA, Neyerlin KC, Murphy MW. Measurement of oxygen transport resistance in PEM fuel cells by limiting current methods. Journal of The Electrochemical Society. 2009 Sep 1;156(9):B991-1003.

• Kocha SS. Principles of MEA Preparation in Handbook of Fuel Cells: Fundamentals, Technology, and Applications; Vol. 3, edited by W. Vielstich, A. Lamm, and HA Gasteiger. Ch.;43:538.

• Neyerlin KC, Gu W, Jorne J, Clark A, Gasteiger HA. Cathode catalyst utilization for the ORR in a PEMFC analytical model and experimental validation. Journal of The Electrochemical Society. 2007 Feb 1;154(2):B279-87.

Polymer electrolyte fuel cells (PEFCs) are regarded as a promising alternative power source for automobiles (fuel cell electric vehicle; FCEV). Cost reduction and increasing power density of PEFCs are the key challenges for commercialization of FCEVs, and reduction of Platinum amount in the catalyst layer (CL) is essential to achieve cost reduction of FCEVs. In order to accomplish the reduction of Platinum loading without sacrificing PEFC performance under high current density operation, it is important to make more effective use of Pt used in the CL of the membrane electrode assembly (MEA), and the mass transport of reactants within the CL should be enhanced significantly.

From this background, fabrication process of the CL should be investigated in more detail. The fabrication process of the CL includes an ink preparation process and a coating and dry process. In the former process, the materials composing the CL are mixed and dispersed in the solvent. A blend of water and alcohol is typically used as a dispersion medium for Pt-supported carbon and ionomer. The composition of the dispersion medium, the choice of alcohol, solid contents, and mixing conditions are major controlling parameters. The catalyst ink is deposited on a substrate or directly onto a PEM. The screen printing, die coating, or spray-coating is a major choice for the deposition method. The CL structure is finally formed after drying the deposited CL ink under elevated temperature. Extensive number of studies have been conducted to understand the impact of each parameters on PEFC performance, but the mechanism of structural formation of the CL has not been fully understood.

In this study, Impact of stirring condition of the catalyst ink on the CL structure, cell performance and oxygen transport phenomena was investigated by using cyclic voltammetry (CV) measurement, limiting current density measurement, and soft X-ray imaging. To investigate the impact of stirring condition, we prepared two catalyst ink with the same composition but the different stirring condition; stirred with Zr ball (ink A) and without Zr ball (ink B).

Figure 1 shows the result of CV measurement, and it showed that the CL prepared from the ink A have larger electrochemical surface area. Because the amount of Pt-supported carbon contained within the MEA should be the same, this result clearly suggested that the stirring condition of the catalyst ink affect the CL structure and its Pt utilization ratio. In addition, the i-V characteristics of these CLs (Figure 2) clearly showed that the CL prepared from the ink A have better performance than the CL prepared from the ink B, and the performance drop of the CL prepared from ink B was serious under the cell temperature of 30 deg.C. To understand this performance drop in more detail, the soft X-ray imaging technique and the oxygen transport resistance measurement were carried out. Results of the soft X-ray imaging (Figure 3) showed that more liquid water was accumulated within the MEA prepared from ink B. In addition, the oxygen transport resistance suggested that the oxygen transport resistance within the MEA prepared by using the ink A is smaller than the MEA prepared by using the ink B.

These results clearly showed that the stirring condition of the catalyst ink strongly affect the structure of the CL and the transport phenomena within the MEA. The stirring condition might affect not only the aggregation of the Pt-supported carbon but also the ionomer coverage on it, and their impact on the PEFC performance is still not clear. It is highly important to understand the CL fabrication process more scientifically, and it helps to develop high-performance CL.

Acknowledgement:

This work was supported by the New Energy and Industrial Technology Development Organization (NEDO), Japan.

Figure 1

The development of stratified catalyst layers promises an increase in catalyst layer performance compared to conventional flat catalyst layers.^1,2 Irregular catalyst layer thickness and porosity can lead to enhanced gas and water transport in and out of the catalyst layer, respectively. The stratified structure is expected to have the same performance in the kinetic region where the performance is controlled by the overall Pt loading. However at high current densities, the thinner sections of the stratified catalyst layers should allow for better mass transport properties.

Electrode fabrication is done in house with a custom designed spray coating procedure and catalyst ink recipe. Various approaches are being explored to achieve the desired electrode structure. One approach is to densify the catalyst layer in localized regions, and is based on Ion Power proprietary manufacturing techniques. This can involve the use of glass epoxy masks during the spray coating process to directly create a patterned electrode on the membrane that has thicker and thinner regions. Figure 1 illustrates the effect of stratification on the fuel cell performance of a MEA.

The second approach is to pattern thicker and thinner sections of an electrode using masks. During spray coating, one mask exposes the lands and the other mask exposes the channels. This way the two extreme cases can be investigated and the resulting performance compared. The concept of having more catalyst material in the channels may be beneficial due to rapid reaction times during influx of reactant gases through the channels. However, the presence of a thicker catalyst layer under the channel may later become an issue for product water removal and oxygen diffusion. On the other hand, spray coating more of the electrode in the land areas may support better product water removal and oxygen diffusion by leaving a thinner catalyst layer under the channels. The best performance is expected from a combination of ordered thin and thick regions in the catalyst layer. Results from various MEAs tested in a single cell on a fully automated test station to evaluate catalyst performance will be presented.

The evaluation of new catalyst ink recipes with reduced ionomer to carbon ratios is included in this study to further increase mass transport by reducing ionomer swelling in the catalyst layer at high relative humidity.

Acknowledgments

This research is supported by DOE Fuel Cell Technologies Office, through the Fuel Cell Performance and Durability (FC-PAD) Consortium; Fuel Cells program manager: Dimitrios Papageorgopoulos.

References

1. T. E. Springer, M. S. Wilson, and S. Gottesfeld,  J. Electrochem. Soc., 140 (12), 3513–3526 (1993).

2. R. Borup and T. Rockward, US Dep. Energy Annu. Merrit Rev., Project ID: FC052(2015). https://www.hydrogen.energy.gov/pdfs/review15/fc052_rockward_2015_p.pdf

Figure 1. Performance of textured and baseline MEA at 275kPa in H₂/Air @ 80^oC and 100%RH.                 

Figure 1

For polymer electrolyte fuel cells (PEFCs) to achieve broad commercialization several technological hurdles have to be resolved. These include the high cost associated with the use of Pt as electrocatalyst in the electrodes, and durability problems due to carbon corrosion. PEFCs' conventional carbon-supported platinum (Pt/C) electrodes use 0.25 mg/cm² of Pt loading, whereas the set target for 2020 by the U.S. Department of Energy (DOE) is 0.125 mg/cm². Non-conventional thin-film electrodes prepared by atomic layer deposition (ALD) are a promising alternative to conventional designs^1-2.

In the ALD process, monolayers of Pt are deposited onto the substrate by dissociative chemisorption in a self-assembling reaction. In this work, we propose an ALD thermal exposure mode to deposit Pt nanostructures onto a sacrificial anodized aluminum oxide (AAO) substrate using the trimethylcyclopentadienylmethylplatinum (IV) and oxygen gas as precursors. Thermal exposure mode provides additional time for the reactions in the ALD chamber and allows higher penetration of the precursor into the substrate. After the deposition the ALD electrodes were hot-pressed onto Nafion XL membrane and sacrificial AAO template was etched. We have demonstrated the feasibility of long free-standing structures (6-8 μm), as shown by Figure 1a.

The ALD cathode electrodes were electrochemically characterized within a fuel cell hardware where the cell had custom-made anode gas diffusion electrodes (GDE) with a loading of 0.2 mg/cm². Polarization curves, cyclic voltammetry and electrochemical impedance spectroscopy (EIS) were performed at dry and wet conditions to assess activity and water management of these electrodes. A cyclic voltammogram is shown in Fig 1c, measured at a sweep rate of 40mV/s, 100% RH at 35° C and an inset to the figure shows a Nyquist plot of EIS under 60° C, 100%RH, H₂/N₂ at 0.8V, with a 5 mV perturbation measured between a range of frequencies from 200 MHz to 100 mHz.

In this presentation we will compare various ALD electrode fabrication conditions, ionomer presence and provide detailed electrochemical analysis (electrochemical surface area, polarization curves, cyclic voltammetry) of these nanoelectrode arrays.

Acknowledgements

This work was performed in part at the Center for Nanoscale Systems (CNS), a member of the National Nanotechnology Coordinated Infrustructure Network (NNCI), which is supported by the National Science Foundation under NSF award no. 1541959. CNS is part of Harvard University. We thank Mr. Jake Berliner for helping in ALD electrodes characterization and Mr. Berney Peng for SEM sample preparation. Fabrication of the structures was carried out in part at the Tufts Micro and Nano Fabrication Facility.

References

1. Galbiati, S.; Morin, A.; Pauc, N., Supportless Platinum Nanotubes Array by Atomic Layer Deposition as Pem Fuel Cell Electrode. Electrochimica Acta 2014, 125, 107-116.

2. Galbiati, S.; Morin, A.; Pauc, N., Nanotubes Array Electrodes by Pt Evaporation: Half–Cell Characterization and Pem Fuel Cell Demonstration. Applied Catalysis B: Environmental 2015, 165, 149-157.

Figure 1

Significant advances in the development of electrocatalysts that exceed the DOE target of 440 mA/mg_Pt (0.90V, 80^oC, 100% RH, p_total=150 kPa) for ORR activity have placed fuel cells on a promising path towards achieving the DOE target of a 0.1 mg_Pt/cm²_elec cathode loading by 2020. However, unanticipated voltage losses that manifest at high current density and low Pt loading have prevented the attainment of the 0.125 g_PGM/kW_rated 2020 target.¹ While some of the observed voltage loss for low Pt electrodes (< 0.1 mg_Pt/cm²_elec) has been attributed to oxide dependent kinetics that manifest at the lower iR-free potentials (<0.75V),² it has been observed that a significant portion of this unanticipated loss stems from a transport resistance local to or associated with electrochemically available Pt surface area.¹ While the exact cause of this phenomenon remains unknown, studies have demonstrated that this loss both; 1) scales with total Pt surface area¹ and 2) can be associated with the incorporation of ionomer into the cathode electrode.³Several diagnostics have been developed and applied to PEMFC materials in an attempt to isolate the exact cause and/or quantify the magnitude of this so termed "local Pt resistance".^4-6 However insightful the results of these studies may be, the idealized conditions under which the diagnostics have been performed leave questions as to the relevance of the extracted resistance values. In this work we will expand the parameter space used previously for oxygen limiting current measurements⁴as a means to elucidate trends and gain further insight into the cause and magnitude of the local Pt resistance at relevant PEMFC operating conditions (e.g. cell temperature, oxygen partial pressure and relative humidity).

Experiments were performed using a 5cm² differential cell with MEAs consisting of Pt/V and Pt/HSC at loadings of 0.2 – 0.05 mg Pt/cm². An example measurement is shown in Figure 1 where clear increases in local Pt resistance (R_O2,local) can be observed commensurate with; 1) an increase in oxygen partial pressure (p_O2) and 2) a decrease in relative humidity (RH).

Figure 1. Local Pt resistance (R_O2,local) as a function of oxygen partial pressure (p_O2) and relative humidity (RH). Data was obtained on 50 wt% Pt/HSC electrodes of 0.05, 0.10 and 0.25 mg_Pt/cm²nominal loading, which were diluted with carbon to maintain a uniform thickness.

Acknowledgements: This work was funded in part under the DOE FC-PAD consortium and also by the U.S. Department of Energy under CRADA #CRD-14-539.

References

1. A. Kongkanand and M. F. Mathias, The Journal of Physical Chemistry Letters, 7, 1127 (2016).

2. N. Subramanian, T. Greszler, J. Zhang, W. Gu and R. Makharia, Journal of The Electrochemical Society, 159, B531 (2012)

3. A. Kongkanand, J.E. Owejan, S. Moose, M. Dioguardi, M. Biradar, R. Makharia, Journal of TheElectrochemical Society, 159, F676 (2012).

4. T. A. Greszler, D. Caulk, P. Sinha, Journal of The Electrochemical Society 159, F831 (2012).

5. H. Liu, W.K. Epting, S. Lister, Langmuir, 93, 8361 (1990).

6. H. Iden, S. Takaichi, Y. Furuya, T. Mashio, Y. Ono, A. Ohma, Journal of Electroanalytical Chemistry 694, 37-44 (2013).

Figure 1

Commercially viable polymer electrolyte membrane fuel cell (PEMFC) systems for transportation and stationary power applications require optimal combination of performance and durability at a competitive cost. Optimized water management across a fuel cell is key to improving performance and maintaining high conductivity of the ionomer in PEM and catalyst layers (CLs), while preventing excessive flooding in CLs, gas diffusion layers (GDLs), channels, and manifolds. Durability is also greatly affected by water transport as certain degradation mechanisms are accelerated depending on the humidity levels. As well, catalyst layers that suffered carbon corrosion due to long operation and/or high number of starts/stops are more susceptible to flooding.

Imaging in situ and ex situ enables crucial insight into the microstructure of the fuel cell components and materials, associated water transport, and their effect on cell performance, dynamic response, and degradation mechanisms. Several case studies are presented to highlight the use of optical imaging, neutron imaging, and micro X-ray computed tomography (microXCT) to investigate fuel cell components across a range of length scales. Laboratory-scale microXCT instruments are used ex situ to resolve the three-dimensional structure of the GDL and electrodes, before and after various accelerated stress tests.

Optical imaging in operating fuel cells offers a cost-effective combination of high temporal resolution (tens of frames per second) and high spatial resolution (several µm). The imaging requires special cell design with visual access to the imaged area (GDL surface and channels [1] or catalyst layer surface [2]). We employed optical visualization in a variety of studies, such as to visualize liquid water dynamics in the cathode and anode flow fields, or at the interface between the CL and the GDL. Besides water transport studies, optical imaging can quantify in-plane gas distribution in the flow fields during startups and shutdowns [3].

Neutron imaging is a powerful tool to visualize and quantify water transport in operating fuel cells. It has a high sensitivity to small amounts of water inside a cell, while having high transmission through the common fuel cell hardware. At a lower spatial resolution of 250 µm, neutron imaging is used to measure in-plane water distribution with high temporal resolution and large field of view of up to 20 cm by 20 cm. We studied the performance with novel NSTF (nano structured thin film) electrodes and GDL materials [4], and also visualized water and ice formation when cell was subjected to repeated starts at sub-freezing temperatures [5]. Since neutron imaging provides water content integrated along the neutron beam, we combined neutron imaging with the optical visualization to study water transport in different flow field geometries [6] as well as in an operating electrolyzer [7]. Such approach with simultaneous imaging provides additional information about the water location, and in certain situations allows distinguishing between channel and GDL water, or between anode and cathode channels and manifolds.

High-resolution neutron imaging is able to measure water distribution across the cell thickness at spatial resolution as high as 10 µm. Image processing procedure is developed to measure the water uptake and observe Schroeder's paradox in situ in PEMs [8,9]. Further, properties of the microporous layer (MPL) of the GDL were manipulated to prevent excessive flooding in cathode catalyst layers.

Anode flooding is evidenced by optical visualization, simultaneous imaging, and high-resolution neutron imaging. Increased water transport across the membrane, from cathode to anode, can be detrimental as liquid water on the anode side may cause localized fuel starvation and cause irrecoverable degradation, similar to accelerated carbon corrosion during unassisted startups and shutdowns. However, for certain novel cathodes, which are prone to flooding, e.g. nano-structured thin film (NSTF) electrodes and non-precious group metal (non-PGM) catalysts, water removal through the anode has proven to be a viable option to improve the performance by reducing cathode water content.

The authors acknowledge support from US Department of Energy EERE FCTO, Los Alamos National Laboratory LDRD (Laboratory Directed Research and Development) program, Federal Transit Administration, US Department of Commerce, and NIST Center for Neutron Research.

References:

• D. Spernjak et al., J. Power Sources 170 pp334 (2007)

• F.-Y. Zhang et al., J. Electrochem. Soc. 154 (11) B1152 (2007)

• Y. Ishigami et al., J. Power Sources 269 pp556 (2014)

• A. J. Steinbach et al., ECS Trans. 33 (1) 1179 (2010)

• N.Macauley et al, J. Electrochem. Soc. submitted (2016)

• D. Spernjak et al., J. Power Sources 195 pp3553 (2010)

• O.F. Selamet et al., Int. J. Hydrogen Energy 38 pp5823 (2013)

• D. Hussey et al., J. Appl. Phys. 112 (10), 104906 (2012)

• D. Spernjak et al., ECS Trans. 33 1451 (2010)

Polymer electrolyte fuel cells (PEFC) require an effective water management to operate efficiently, especially at high current densities which are needed to reach system cost targets [1]. The description of the complicated two-phase water transport remains a challenge in PEFC models and requires experimental validation on various length scales. In this work, operando X-ray tomographic microscopy (XTM) with scan times of 10 s was used to depict the liquid water at defined differential conditions at the technically relevant cell temperature of 80 °C at the TOMCAT beamline of the Swiss Light Source.

Cells with Toray TGP-H060 GDL with MPL were operated at high feed gas humidity (0.75 A/cm², 108 % rH, H₂/Air) or high current density (3.0 A/cm², 78 % rH, H₂/O₂) conditions. The liquid water in the cathode GDL was analysed in terms of the local and average saturation, water cluster sizes, connectivity of the water clusters and the permeability of the water filled GDL pores. While fully through-plane connected water clusters contribute to the transport of water in the liquid phase, only partly connected water clusters do not (see Figure 1a). The majority of liquid water volume (> 80 %) is found in a limited number of fully through-plane connected water clusters (< 20 %) (see Figure 1b). Even though the water phase occupies in average the larger pores of the GDL, as reported earlier [2], still some large pores and percolation paths remain liquid free. Nevertheless, the available fully through-plane liquid paths provide liquid permeabilities that result in no noteworthy pressure losses to drain the product water of today's and upcoming current densities.

As research in industry and academia strive for the separation of the transport paths of the liquid and the vapour phase [3], the effect of the presence or absence of the partly connected water clusters on the gas phase transport parameters (permeability & effective diffusivity) of the GDL is discussed as well.

References

[1] O. Gröger et al., J. Electrochem. Soc. 162 (2015), A2605-A2622.

[2] T. Rosén et al., J. Electrochem. Soc. 159 (2012), F536-F544.

[3] A. Forner et al., Advanced Materials 27 (2015), 6317-6322.

Figure 1: a) Sketch of the different cluster types that can be found in the GDL domain; blue: full connected water clusters (FC); green: top connected (TC); red: bottom connected (BC); purple: non connected (NC). b) Volume and cluster number fractions of the different cluster types for the high humidity condition.

Figure 1

Neutron imaging is an ideal method to study water transport phenomena in proton exchange membrane fuel cells (PEMFCs). Neutrons readily penetrate aluminum, carbon and platinum, yet have a high sensitivity to hydrogenous materials, including water. In addition, unlike x-rays, neutrons are a non-destructive probe and do not alter the performance of PEMFCs due to irradiation. As such, there have been numerous studies that revealed the heat and mass transport phenomena in the flow fields and gas diffusion layers [1], freeze operation [2], impacts of carbon corrosion [3], and the impact of porosity in thick non-precious metal catalyst layers [4].

The primary limitation of neutron imaging is the achievable spatial and temporal resolutions. Current state of the art spatial resolution is about 20 µm, with a 20 minute image acquisition time. The spatial resolution is primarily limited by the range of the charged particles that are the result of the neutron capture reaction and are used to detect the neutron. This range is of order 5 µm for many common detector schemes. The temporal resolution is limited by the brightness of neutron sources, which are in general many orders of magnitude less bright than synchrotron x-ray sources; thus to have reasonable time resolution conventional neutron imaging requires the use of beam defining apertures that are of order 1 mm for high resolution imaging of PEMFCs, so that one cannot make use of geometric image magnification.

We are pursuing three efforts to improve the achievable image spatial resolution and time resolution. The simplest approach to improving the spatial resolution is to place a narrow slit in front of the test section and scan it across the active area. The challenge of this method is to create a narrow, well-defined, completely opaque slit with opening area ~1 µm. KAERI and Pusan National University have devised a Gadox powder filling method that enables creating such slits, enabling imaging with 1 µm resolution [5]. By amplifying the scintillation light and using a high speed camera, it is possible to record individual neutron scintillation events and using a centroid algorithm determine the position of the neutron with high precision (less than 10 µm). The measured through-plane water content of a PEMFC is shown in figure 1 for the present state of both methods. The final effort is the development of a neutron microscope objective based on reflective neutron optics [6]. With a lens, one no longer requires the use of small beam defining apertures and thus the image temporal resolution can increase by at least a factor of 100. We will report on the progress of the first optics to reach 20 µm resolution and discuss the possibility of achieving 1 µm spatial resolution through neutron image magnification.

References

• T.A Trabold et al, "Use of neutron imaging for proton exchange membrane fuel cell (PEMFC) performance analysis and design", Handbook of fuel cells v. 5, 2009.

•  R Mukundan et al, "Performance of PEM fuel cells at sub-freezing temperatures", ECS Transactions 11 (1), 543-552, 2007.

• JD Fairweather et al, "Effects of cathode corrosion on through-plane water transport in proton exchange membrane fuel cells", JECS 160 (9), F980-F993, 2013.

• D Spernjak et al, "Water Management in PEM Fuel Cells with Non-Precious Metal Catalyst Electrodes", 228^th ECS Meeting abstracts, p. 1540, 2015.

• J. Kim et al, "Fabrication and characterization of the source grating for visibility improvement of neutron phase imaging with gratings", RSI 84(6):063705, 2013.

• D. Liu et al "Demonstration of achromatic cold-neutron microscope utilizing axisymmetric focusing mirrors", Applied Physics Letters 102 (18), 183508, 2013.

Figure 1 Comparison of the image quality of the current state of the art (a), the slit imaging method with 4 µm resolution (b), and the reconstructed image from centroiding (c).

Figure 1