Mass transport in a hydrogen gas diffusion electrode

Experimental data are presented concerning the diffusion-limited current density for hydrogen oxidation in a gas diffusion electrode (GDE) under various conditions. These current densities were obtained using mixtures of hydrogen and inert gases. To elucidate the dependence of the overall mass transport coefficient on the gas phase diffusion coefficient and the liquid phase diffusion coefficient of the hydrogen, a simplified model was derived to describe the transport of hydrogen in a GDE based on literature models. The GDE consists of a hydrophobic and a hydrophilic layer, namely a porous backing and a reaction layer. The model involves gas diffusion through the porous backing of the GDE combined with gas diffusion, gas dissolution and reaction in the reaction layer of the electrode. It was found that the transport rate of hydrogen under the experimental circumstances is determined by hydrogen gas diffusion in the pores of the porous backing, as well as in the macropores of the reaction layer. Diffusion of dissolved hydrogen in the micropores of the reaction layer, through the liquid, is shown to be of little significance.     

Mechanism of hydrogen oxidation on a platinum-loaded gas diffusion electrode

The mechanism of hydrogen oxidation on a platinum-loaded gas-diffusion electrode has been investigated. Experimental potential–current curves, especially in the low overpotential range, have been measured for H2–N2 mixtures with a small content of hydrogen and for pure H2. Theoretical relations have also been presented. Comparing the experimental and theoretical relations, it is concluded that the hydrogen oxidation occurs according to the Volmer–Tafel mechanism. The reactivity of the electrode has a large effect on the kinetic parameters for hydrogen oxidation. The limiting current is determined by diffusion of hydrogen for a very reactive gas diffusion electrode and by the Tafel reaction for a gas diffusion electrode with a low reactivity. The transfer coefficient for the Volmer reaction V is 0.5 and i0,V/i0,T 0.1 for a very reactive gas diffusion electrode. V increases and i0,V/i0,T ratio decreases with decreasing reactivity of the gas diffusion electrode.     

Mathematical Model of Proton Exchange Membrane Fuel Cell with Consideration of Water Management

One‐dimensional model on the membrane electrode assembly (MEA) of proton exchange membrane fuel cell is proposed, where the membrane hydration/dehydration and the possible water flooding of the respective cathode and anode gas diffusion layers are considered. A novel approach of phase‐equilibrium approximation is proposed to trace the water front and the detailed saturation profile once water emerges in either anode or cathode gas diffusion layer. The approach is validated by a semi‐analytical method published earlier. The novel approach is applicable to the polarization regime from open circuit voltage to the limiting current density under practical operation conditions. Oxygen diffusion is limited by water accumulation in the cathode gas diffusion layer as current increases, caused by excessive water generation at the cathode catalyst layer and the electro‐osmotic drag across the membrane. The existence of liquid water in the anode gas diffusion layer is predicted at low current densities if high degrees of humidification in both anode and cathode feeds are employed. The influences of inlet relative humidity, imposed pressure drop, and cell temperature are correlated well with the cell performance. In addition, the overpotentials attributed from individual components of the MEA are delineated against the cell current densities.     

Potential and current distribution on the surface of a hydrogen gas diffusion anode

The current and potential distribution for a hydrogen gas-diffusion disc electrode with a relatively high ohmic resistance are investigated. A theoretical model for these distributions is presented. Potential differences between the edge of the electrode and points on the electrode surface have been measured for a hydrogen gas-diffusion electrode loaded with various total currents. From the results it is concluded that the proposed model is very useful to obtain the potential and the current density distribution along a hydrogen-gas diffusion disc electrode. Moreover, the allowable size of cylindrical holes in a perforated plate placed against the rear of the gas diffusion electrode for its current supply, can be calculated to achieve a reasonably uniform current distribution along the gas-diffusion electrode.     

Electrochemical impedance spectra of solid-oxide fuel cells and polymer membrane fuel cells

Electrochemical impedance spectroscopy (EIS) is a very useful method for the characterization of fuel cells. The anode and cathode transfer functions have been determined independently without a reference electrode using symmetric gas supply of hydrogen or oxygen on both electrodes of the fuel cell at open circuit potential (OCP). EIS are given for both polymer electrolyte fuel cells (PEFC) and solid oxide fuel cells (SOFC) at current densities up to 0.76 A cm⁻² (PEFC) and 0.22 A cm⁻² (SOFC). With increasing current density the PEFC-impedance decreases significantly in the low frequency range reaching a minimum at 0.4 A cm⁻². At even higher current densities an increasing contribution of water diffusion is observed: the cell impedance increases again. From EIS of SOFC a finite diffusion behavior is observed even at OCP, depending on water partial pressure of the anodic gas supply. This additional element reflects the influence of water partial pressure on the cell potential. The simulation of the measured EIS with an equivalent circuit enables the calculation of the individual voltage losses in the fuel cell.     

A study of current distribution in a DEM cell during bromate formation

The anodic oxidation of potassium bromide to potassium bromate is performed in an undivided cell with hydrogen evolution the major reaction at the counter electrode. The cell used is a dished electrode membrane (DEM) cell. Current density distribution, measured using a segmented electrode, shows a variation in the two principle dimensions; along the length of the electrode and over the width of the electrode. Current densities are highest at the electrolyte flow inlet and also exhibit a localized maximum along the electrode length. The variation in current density is due to the influence of electrolytic gas evolution on the effective electrolyte conductivity and mass transport and also due to the change in shape of the dished electrode, which influences mass transport, electrical potential field and flow at the cell inlet and exit.     

Gas diffusion electrodes improve hydrogen gas mass transfer for a hydrogen oxidizing bioanode

Background

Bioelectrochemical systems (BESs) are capable of recovery of metals at a cathode through oxidation of organic substrate at an anode. Recently, also hydrogen gas was used as an electron donor for recovery of copper in BESs. Oxidation of hydrogen gas produced a current density of 0.8 A m^‐2 and combined with Cu²⁺ reduction at the cathode, produced 0.25 W m^‐2. The main factor limiting current production was the mass transfer of hydrogen to the biofilm due to the low solubility of hydrogen in the anolyte. Here, the mass transfer of hydrogen gas to the bioanode was improved by use of a gas diffusion electrode (GDE).

Results

With the GDE, hydrogen was oxidized to produce a current density of 2.9 A m^‐2 at an anode potential of –0.2 V. Addition of bicarbonate to the influent led to production of acetate, in addition to current. At a bicarbonate concentration of 50 mmol L^‐1, current density increased to 10.7 A m^‐2 at an anode potential of –0.2 V. This increase in current density could be due to oxidation of formed acetate in addition to oxidation of hydrogen, or enhanced growth of hydrogen oxidizing bacteria due to the availability of acetate as carbon source. The effect of mass transfer was further assessed through enhanced mixing and in combination with the addition of bicarbonate (50 mmol L^‐1) current density increased further to 17.1 A m^‐2.

Conclusion

Hydrogen gas may offer opportunities as electron donor for bioanodes, with acetate as potential intermediate, at locations where excess hydrogen and no organics are available. © 2017 The Authors. Journal of Chemical Technology & Biotechnology published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.     

Reduction of metal ions in dilute solutions using a GBC-reactor: Part I: Experimental study on the laboratory scale reduction of ferric ions

To reduce metal ions in dilute solutions a new type of electrochemical reactor has been developed: the GBC-reactor. This reactor consists of a gas diffusion electrode coupled with a packed bed electrode. The working principle of the reactor is based upon two main reactions: the catalytic oxidation of hydrogen gas in the gas diffusion electrode and the simultaneous reduction of metal ions on the packed bed electrode. This process occurs spontaneously without the need for an external power supply when the Gibbs free energy of the total reaction is negative. To study the behaviour of the reactor the reduction of ferric ions was used as a model system. The experimental results, obtained from varying a number of key process parameters, could be described using a very simple macroscopic rate equation. It is concluded that the reduction of ferric ions in a GBC-reactor is controlled by both mass transfer and electrochemical kinetics. To carry out scale-up and optimization studies a reactor model incorporating the potential distribution in the packed bed electrode is, however, necessary.     

Multi-variable optimization of PEMFC cathodes using an agglomerate model

A comprehensive numerical framework for cathode electrode design is presented and applied to predict the catalyst layer and the gas diffusion layer parameters that lead to an optimal electrode performance at different operating conditions. The design and optimization framework couples an agglomerate cathode catalyst layer model to a numerical gradient-based optimization algorithm. The set of optimal parameters is obtained by solving a multi-variable optimization problem. The parameters are the catalyst layer platinum loading, platinum to carbon ratio, amount of electrolyte in the agglomerate and the gas diffusion layer porosity. The results show that the optimal catalyst layer composition and gas diffusion layer porosity depend on operating conditions. At low current densities, performance is mainly improved by increasing platinum loading to values above 1 mg cm⁻², moderate values of electrolyte volume fraction, 0.5, and low porosity, 0.1. At higher current densities, performance is improved by reducing the platinum loading to values below 0.35 mg cm⁻² and increasing both electrolyte volume fraction, 0.55, and porosity 0.32. The underlying improvements due to the optimized compositions are analyzed in terms of the spatial distribution of the various overpotentials, and the effect of the agglomerate structure parameters (radius and electrolyte film) are investigated. The paper closes with a discussion of the optimized composition obtained in this study in the context of available experimental data. The analysis suggests that reducing the solid phase volume fraction inside the catalyst layer might lead to improved electrode performance.     

Reduction of metal ions in dilute solutions using a GBC-reactor: Part I: Experimental study on the laboratory scale reduction of ferric ions

To reduce metal ions in dilute solutions a new type of electrochemical reactor has been developed: the GBC-reactor. This reactor consists of a gas diffusion electrode coupled with a packed bed electrode. The working principle of the reactor is based upon two main reactions: the catalytic oxidation of hydrogen gas in the gas diffusion electrode and the simultaneous reduction of metal ions on the packed bed electrode. This process occurs spontaneously without the need for an external power supply when the Gibbs free energy of the total reaction is negative. To study the behaviour of the reactor the reduction of ferric ions was used as a model system. The experimental results, obtained from varying a number of key process parameters, could be described using a very simple macroscopic rate equation. It is concluded that the reduction of ferric ions in a GBC-reactor is controlled by both mass transfer and electrochemical kinetics. To carry out scale-up and optimization studies a reactor model incorporating the potential distribution in the packed bed electrode is, however, necessary.     

Influence of gas phase mass transfer limitations on molten carbonate fuel cell cathodes

The purpose of this paper is to elucidate to what extent mass transfer limitations in the gas phase affect the performance of porous molten carbonate fuel cell cathodes. Experimental data from porous nickel oxide cathodes and calculated data are presented. One and two-dimensional models for the current collector and electrode region have been used. Shielding effects of the current collector are taken into account. The mass balance in the gas phase is taken into account by using the Stefan–Maxwell equation. For standard gas composition and normal operating current density, the effect of gas-phase diffusion is small. The diffusion in the gaseous phase must be considered at operation at higher current densities. For low oxygen partial pressures, the influence of mass transfer limitations is large, even at low current densities. To eliminate the influence of conversion on polarization curves recorded on laboratory cell units, measurements should always be performed with a five to tenfold stoichiometric excess of oxygen. Two-dimensional calculations show rather large concentration gradients in directions parallel to the current collector. However, the influence on electrode performance is still small, which is explained by the fact that most of the current is produced close to the electrolyte matrix.     

Identification of electrode surface blocking by means of thin-layer cell: 1. The model

A new method for the identification of electrode surface blocking, based on the measured differences of the diffusion limiting current densities in a thin-layer cell and in a cell with rotating-disk electrode is presented. An equation describing the diffusion limiting current density dependence on the inter-electrode distance in a thin-layer cell was derived. Experimental results were obtained from measurements with model electrodes to demonstrate the possibilities for unambiguous identification of surface blocking phenomena. A model electrode, covered with a partially conductive blocking layer was also discussed and the appearance of two diffusion limiting current density levels on the voltammograms was experimentally verified.     

A model for Kolbe electrolysis in a parallel plate reactor

A parallel-plate reactor model is developed for the Kolbe electrolysis of acetate to ethane and carbon dioxide with hydrogen evolution as the counterelectrode reaction. The parallel-plate reactor is considered to consist of three zones: a turbulent bulk region in which streamwise convection is the dominant mass-transport mechanism (plug-flow model) and a thin diffusion layer at each electrode where diffusion and migration mass transport are dominant (Nernst diffusion-layer model). The acetic acid solution is supported with sodium hydroxide, and the reactor is under steady cell-potential control. Gaseous products are tracked by a hypothetical gas layer which increases in thickness in the streamwise direction. The gas phase is assumed to be an ideal, three-component mixture of hydrogen, carbon dioxide and ethane; the liquid phase consists of acetate, proton, acetic acid, and sodium and hydroxyl ions. The model predicts streamwise profiles of concentration, current density, gas-void fraction, and gas and liquid velocities in addition to reactant conversion, and cell-polarization characteristics. The average current density exhibits a maximum at a base-to-acid ratio of 0.96 due to the weak-acid/strong-base chemistry and a broad maximum at an interelectrode spacing of 0.37 cm resulting from minimized ohmic losses.     

Mass transfer in AC electrolysis part I: Theoretical analysis using a film model for sinusoidal current on a rotating hemispherical electrode

A film model is presented for the analysis of mass transfer to a rotating hemispherical electrode when sinusoidal alternating current (AC) together with direct current (DC) are flowing across the electrode surface. The concentration of a diffusing ion is separated into two independent components: a constant DC component and a periodic AC component. The DC concentration is obtained by solving the steady-state convective mass transport equation with the perturbation method. The periodic AC concentration distribution is analyzed by the solution to the one-dimensional transient diffusion equation based on the concept of Nernst diffusion layer. The limiting AC current densities corresponding to a zero surface concentration of a reactive ion are investigated for various DC current densities and AC frequencies. The resulting periodic concentration overpotential wave and its phase shift with respect to the applied AC are examined. A comparison with a previous rigorous model indicates that the film model is a good approximation to the mass transfer calculation in the regimes of a dimensionless AC frequency K = (ω/Ω)Sc^1/3 greater than 2 and less than 0.01.     

An experimental comparison of a graphite electrode and a gas diffusion electrode for the cathodic production of hydrogen peroxide

This work studies the production of hydrogen peroxide through the cathodic reduction of oxygen in acidic medium, by comparing the results obtained using a commercial graphite and a gas diffusion electrode. A low pH was required to allow the application of hydrogen peroxide generation to an electro-Fenton process. The influence of applied potential and the gas flow composition were investigated. The gas diffusion electrode demonstrates a higher selectivity for hydrogen peroxide production, without significantly compromising the iron regeneration, thus making its successful application to a cathodic Fenton-like treatment, possible. Unlike the graphite cathode, the gas diffusion cathode also proved to be effective in the air flow.     

Ohmic resistance of solution in a vertical gas-evolving cell

Vertical electrolysers with a narrow cell gap between a gas-evolving electrode and a membrane or diaphragm are used to produce industrial gases. Generally, the local current density decreases with height in the cell. Electrolyses are carried out with a KOH solution in a tall vertical divided rectangular cell with two gas-evolving electrodes. Either the hydrogen or the oxygen bubbles containing solution from the divided cell are passed through a small measuring cell. Ohmic resistance experiments are carried out in the small measuring cell with a gas-evolving electrode and a gas diffusion electrode, on which no gas bubbles are evolved. The effect of various parameters, viz. current density, solution flow rate and temperature, on the ohmic resistance of solution in the measuring cell are determined. It is found that the normalized ohmic resistance of the solution in the measuring cell during electrolysis increases with current density and with the gas voidage in the bulk of solution, decreases with increasing solution flow rate and is practically independent of temperature at 25 to 60 OC. Moreover, it is found that for an oxygen evolving electrode in a solution containing only oxygen bubbles, as well as for a hydrogen evolving electrode in a solution containing only hydrogen bubbles, the normalized resistance of the solution between the gas-evolving electrode and the nongas evolving electrode is given by a relatively simple empirical relation. A relation is derived describing the gas voidage in the solution as a function of the distance from the gas-evolving electrode in the presence and the absence of gas bubbles in the bulk solution.     

Struktur und eigenschaften von Raney-Nickel-katalysatoren mit zulegierungen für alkalische brennstoffzellen

As is well known the catalytic activity of Raney nickel may be increased by alloying with some percentage of transition metals eg Fe, Ti, or Mo. Electrocatalysts activated in this way may reach exchange current densities up to 5 A/g, and in the anodic oxidation of H₂ current densities up to 4 A/g at 23°C. Sorption measurements and pore distributions calculated from the sorption isotherms show that the structure of the Raney sponge is changed but little by these transition metal additions; by the observed small structure differences the different properties of electrocatalysts cannot be explained. Discussion of these results shows that the measured high exchange current densities alone do not account for the high anodic current densities, because the diffusion of molecular hydrogen through the transitional pore system of the Raney sponge filled by electrolyte to the reaction zone does not admit current densities of more than 1 A/g. Only additional diffusion of chemisorbed H-Atoms on the catalyst surface allows the high current densities observed. Therefore an essential condition of a good electrocatalyst surface is a large coefficient D₀ of the surface diffusion of hydrogen exceeding 5 × 10⁻⁹ cm²/s.     

The influence of oxygen additions to hydrogen in their electrode reactions at Pt/Nafion interface

The influence of oxygen gas added to hydrogen in their electrode reactions at the Pt/Nafion interface was investigated using ac impedance method. The electrochemical cell was arranged in either electrolytic (hydrogen enrichment) or galvanic (fuel cell) mode. The impedance spectra of the electrode reaction of a H₂/O₂ gas mixture were taken in each mode as a function of the gas composition, electrode surface roughness and the cell potential. The spectrum taken for the anodic reaction of electrolytic arrangement confirmed the anodic oxygen reduction reaction (AOR, the local consumption of hydrogen by the added oxygen) by showing an independent arc distinguishable from that for hydrogen oxidation. But the independent arc was not revealed in the spectrum taken on a smooth (low surface area) electrode or on a Pt/C anode of the galvanic cell. At any cell current density, the electrolytic mode showed its anodic overpotential much higher (nearly three times higher at the current density of 100 mA cm⁻²) than the potential registered in galvanic mode implying that the oxygen gas in the mixture engages more active and independent AOR at the anode of the electrolytic cell.     

Determination of overall mass transport coefficients in gas diffusion electrodes

Gas diffusion electrodes are used for many purposes, for example in fuel cells, in synthesis and as anodes in electrodeposition processes. The behaviour of gas diffusion electrodes has been the subject of many studies. In this work the transport of gas in the gas diffusion electrode, characterized by the overall mass transport coefficient, has been investigated using hydrogen-nitrogen mixtures. A reactor model for the gas compartment of the gas diffusion electrode test cell is proposed to calculate the concentration of hydrogen in the gas compartment as a function of the input concentration of hydrogen and the total volumetric gas flow rate. The mass transport coefficient is found to be independent of variations in hydrogen concentration and volumetric gas flow rate. The temperature dependence of the mass transport coefficient has been determined. A maximum was found at 40°C.Notation Agd geometric electrode surface area (m²) - C _in concentration of reactive component at the inlet of the gas compartment (mol m^–3) - c _out concentration of reactive component at the outlet of the gas compartment (mol m^–3) - E potential (V) - E _e equilibrium potential (V) - E _t upper limit potential (V) - F _v volumetric flow rate (m^–3 s^–1) - F _v,H volumetric flow rate of hydrogen (m^–3 s^–1) - F _v,N volumetric flow rate of nitrogen (m^–3 s^–1) - F _vin volumetric flow rate at the inlet of the gas compartment (m^–3 s^–1) - F _v,out volumetric flow rate at the outlet of the gas compartment (in ^–3 s^–1) - F _v,reaction volumetric flow rate of reactive component into the gas diffusion electrode (m^–3 s^–1) - Faraday constant (A s mo^–1) - I _gd current for gas diffusion electrode (A) - i _gd current density for gas diffusion electrode (A m^–2) - I _gd,1 diffusion limited current for gas diffusion electrode (A) - i _gd,1 diffusion limited current density for gas diffusion electrode (A m^–2) - I _gd,1,calc calculated diffusion limited current for gas diffusion electrode (A) - i _gd,1,calc calculated diffusion limited current density for gas diffusion electrode (A m^–2) - I _hp current for hydrogen production (A) - k _s mass transport coefficient calculated from c _out (m s^–1) - n number of electrons involved in electrode reaction - T temperature (°C) - V _m molar volume of gas (m³ mol^–1) - overpotential (V)     

Microcalorimetric study of the self-discharge of the NiOOH/Ni(OH)2 electrode in a hydrogen environment

A microcalorimetric method has been used to investigate the self-discharge behaviour of nickel oxyhydroxide electrodes in a pressurized gaseous hydrogen environment. It was found that the heat generation rate is proportional to hydrogen pressure, and is significantly dependent on the immersion state of the electrode in the electrolyte. Hence, diffusion of dissolved hydrogen gas towards or within the electrode controls, at least partially, the self-discharge rate. However, the heat generation decreases exponentially with time, indicating that self-discharge is also proportional to the amount of the charged active material available for the reaction. The presence of Mg, Co and Cd oxides or hydroxides appears to inhibit self-discharge. It was found that direct chemical reaction between dissolved hydrogen and the active material dominates, while in addition, electrochemical oxidation of hydrogen coupled with electrochemical reduction of the active material might also occur at a much smaller rate than the direct reaction.