Reidar Tunold

Influence of Ammonium on Conductivity and Water Content of Nafion 117 Membranes

Ion-exchange equilibria of ammonium between an aqueous phase and Nafion 117 were measured at 10, 25, 40, and 60°C by equilibrating the membrane in 0. 1 M chloride electrolytes of known cation composition. The water content in the membrane phase decreased linearly with increasing cation fraction of ammonium in the membrane phase (y NH + 4 ) from λ H2O = 21.2 (moles of water per mole-sulfonic acid groups) in proton form Nafion in pure water to λ H2O = 13.2 in ammonium form Nafion in a 0.1 M chloride solution. The conductivity was measured by ac impedance in a two-electrode setup using a stack of membranes. The conductivity also decreased linearly with increasing y NH+4 from 97 to 25 mS/cm at 25.0°C. Our results indicated that the conductivity of Nafion was isotropic, however, available literature is not conclusive on this matter. The temperature dependence of the conductivity was measured, and the fitted activation energy in an Arrhenius-type equation was found to depend on membrane composition and hence water content.

Morphological Changes at the Interface of the Nickel‐Yttria Stabilized Zirconia Point Electrode

The H 2 -H 2 O, Ni/YSZ point electrode has been investigated using long-term potential step measurements and impedance spectroscopy at 1273 K. Morphological and structural changes at the electrode interface were evaluated by electron microscopy, energy dispersive X-ray analysis, and Raman spectroscopy ex situ. The anodic current was found to induce a self-catalytic effect on the electrode, an the anodic steady state current increased to more than twice the initial value with a time constant of about 40 h. In contrast, cathodic polarization reduced the performance of the electrode, and the cathodic current decreased significantly with a time constant of about 20 h. Redistribution of material in the reaction zone is suggested to control most of the changes in electrode activity. At anodic overpotentials it was observed that Ni was transported to the electrolyte surface, forming a necklace of Ni particles around the electrode/electrolyte contact. This is believed to increase the three-phase boundary (TPB) length and account for the higher activity of the electrode. At cathodic overpotentials the transfer of Ni to the YSZ was found to be restricted, and it is proposed that agglomeration of dispersed metal particles reduced the TPB length, and accordingly the cathodic current. In addition to the morphological modifications, the catalytic properties of the surfaces were significantly altered as the electrode was polarized. Transformation from cubic to tetragonal YSZ, due to segregation of the material, was observed on the surface of the electrolyte when the sample was kept at working conditions for long periods of time (135 days). The passage of current was not found to generate any permanent phase transformation in the YSZ.

Electrochemical Oxidation of CO on Pt and Ni Point Electrodes in Contact with an Yttria-Stabilized Zirconia Electrolyte I. Modeling of Steady-State and Impedance Behavior

The steady-state and impedance response of a solid metal point electrode in contact with a solid, oxygen-ion conducting electrolyte in CO-CO 2 atmospheres was derived for three reaction mechanisms, involving gaseous species or species adsorbed on the surface of the electrode and/or the electrolyte. The overall electrochemical reaction was assumed to proceed in elementary steps, such as adsorption, diffusion of adsorbed species, and charge transfer. Eliciting the reaction mechanism from the steady-state response may require an extensive analysis in terms of temperature, gas-phase composition, and overpotential dependence. The impedance response, on the other hand, can in favorable cases be more distinctively dependent on the number of adsorbed species involved in the reaction and on the role of diffusion. Thus, if only one adsorbed intermediate species is involved, the impedance spectrum will always appear in the first quadrant of the impedance plane plot irrespective of experimental conditions. If two adsorbed intermediates are involved, however, this may lead to appearance of fourth-quadrant data.

Structure and related properties of (La,Ce,Nd,Pr)Ni5 alloys

Abstract The effects of substitution of La by Ce, Nd and Pr in the LaNi5-based alloys were systematically studied, with the use of factorial analysis. The changes of equilibrium pressure, hydrogen capacity and hysteresis were analyzed in relation to the structure changes of the hexagonal unit cell of (La,Ce,Nd,Pr)Ni5. Both individual effects of each alloy component, and their interaction in the alloy were studied. It has been shown that variations in the basal plane parameter (a) can be used as an indication for the plateau pressure changes resulting from substitution, giving better fits than variations in the unit cell volume. Structural anisotropy is discussed as a possible tool for the prediction of the decrepitation resistance of the alloy. Methods for optimization of the electrode alloys (including reduction of the cobalt content) are proposed.

Electrodeposition of magnesium from halide melts—charge transfer and diffusion kinetics

Electrodeposition of magnesium from molten MgCl2 and MgCl2-MgF2 (78-22 mol%) mixtures was studied by electrochemical techniques. The process was found to be quasi-reversible with a cathodic electrochemical rate constant for the total charge transfer reaction of about 10−3 cm/s at 780 °C. Underpotential deposition of an adsorbed layer of reduced Mg(II)-species was observed during cyclic voltammetry and potential step measurements. The electrode capacitance was determined from galvanostatic pulses and found to depend on the applied current density. This has been explained by a pseudo capacitance due to the formation of an adsorbed layer. At potentials positive to nucleation a diffusion controlled process was identified. The determined concentration of the diffusing species, as well as the fact that Mg(II) species are the only cations in the system, indicate that the process is diffusion of dissolved metal (the product) from the electrode interface to the electrolyte.

Iridium oxide-based nanocrystalline particles as oxygen evolution electrocatalysts

Iridium-based oxides are highly active as oxygen evolving electrocatalysts in PEM water electrolyzers. In this work XRD reveals that Ir-Sn oxides contain a single rutile phase with lattice parameters between those of pure IrO2 and SnO2. Addition of Ru leads to the synthesis of a core-shell type material due to the strong agglomeration of Ru colloids during the preparation procedure. The shell of this material consists of an Ir-Sn-Ru oxide deficient in Ru relative to the bulk. This leads to a decrease in the surface noble metal concentration (as found by XPS), which in turn results in a significant reduction in electrochemically active surface area. Polarization analysis indicates that the addition of Ru can influence the rate-determining step or mechanism by which oxygen is evolved. In a PEM water electrolysis cell, small additions of Sn do not significantly reduce the operating performance, however larger additions cause a performance loss due to a reduction in active surface area and increased ohmic resistance. When a pure IrO2 anode is used, a cell voltage is 1.61 V at 1 A cm−2 and 90°C.

The electrochemical impedance of metal hydride electrodes

Abstract The electrochemical impedance responses for different laboratory type metal hydride electrodes were successfully modeled and fitted to experimental data for AB 5 type hydrogen storage alloys as well as one MgNi type electrode. The models fitted the experimental data remarkably well. Several AC equivalent circuits have been proposed in the literature. The experimental data, however, could not always be satisfactorily approximated. The approximation model presented here exhibits smooth fit to the experimental results for all frequencies in the whole range from 10 kHz to 0.1 mHz. Equivalent circuits, explaining the experimental impedances in a wide frequency range for electrodes of hydride forming materials mixed with copper powder, were obtained. Both charge transfer and spherical diffusion of hydrogen in the particles are important sub processes that govern the total rate of the electrochemical hydrogen absorption/desorption reaction. To approximate the experimental data, equations describing the current distribution in porous electrodes were needed. Indications of one or more parallel reduction/oxidation processes competing with the electrochemical hydrogen absorption/desorption reaction were observed. The impedance analysis was found to be an efficient method for characterizing metal hydride electrodes in situ.

Electrochemical Oxidation of CO on Pt and Ni Point Electrodes in Contact with an Yttria-Stabilized Zirconia Electrolyte II. Steady-State and Impedance Measurements

Electrochemical measurements were performed with Pt and Ni electrodes forming an approximate (macroscopic) point contact with a solid electrolyte, (Y 2 O 3 ) 0.08 (ZrO 2 ) 0.92 , for mixtures of CO-CO 2 in the temperature range 827 to 950°C and 775 to 914°C, respectively. Steady-state current and impedance spectra vs. overpotential under the different experimental conditions were compared with the theoretical models derived in Part I of this paper. A low-frequency inductive loop in the fourth quadrant of the impedance spectra appearing under certain experimental conditions is consistent with a model for a reaction mechanism involving two adsorbed intermediates. At high anodic overpotentials, the data for Ni electrodes are consistent with the formation of an insulating layer of nickel oxide (NiO) in the electrode/electrolyte interface. Irrespective of the experimental conditions, no effects of diffusion limitations were seen in the impedance spectra. The electrode metal plays a significant role in the overall reaction. For Ni, the results indicate a lower specific elecirocatalytic activity for CO-CO 2 than for the H 2 -H 2 O reaction.

Determination of diffusion coefficients of depositing ions in molten chlorides by transient electrochemical techniques

The electrochemical reduction of Pb(II), Zn(II) and Mg(II) ions at glassy carbon and tungsten electrodes in molten KCl–LiCl eutectic was studied by linear sweep voltammetry (LSV), convolutive potential sweep voltammetry (CPSV), chronopotentiometry and chronoamperometry. In the case of lead deposition, the initial nucleation stage was found to influence the shape of the voltammetric current peak. CPSV was consequently believed to be a more reliable method than LSV for the calculation of the diffusion coefficient of the Pb(II) ion. Similar complications were not observed for zinc and magnesium since the nucleation overpotentials for these metals were significantly lower than for lead. On the other hand, lithium codeposition made it difficult to interpret the zinc and especially the magnesium convolution voltammograms. Chronopotentiometry yielded practically identical results for DPb(II) as the voltammetric techniques. However, due to a substantial residual current, the Sand equation was not obeyed for the Zn(II) and Mg(II) ions. Determination of diffusion coefficients from single potentiostatic current transients and the Cottrell equation was not found to be a very reliable method. Empirical expressions for the temperature dependence of D in the range 400–500 °C were calculated for all three ions.

Oxoacidity reactions in equimolar molten CaCl2-NaCl mixture at 575 °C

Abstract The behaviour of an yttria-stabilized zirconia electrode in the CaCl2-NaCl melt has been investigated at 575 °C. The response of the electrode is Nernstian and permits its use in the titration of Mg(II) with oxide ions. The solubility product of MgO was also determined. We have calculated the equilibrium constant of the carbonate ion dissociation as well as those corresponding to the systems HCl H 2 O and Cl 2 O 2 , finding that molten CaCl2-NaCl is more acidic than the LiCl-KCl eutectic melt and the BaCl2-CaCl2-NaCl (23.5:24.5:52 mol%) melt. On the other hand, the mixture can be purified with respect to oxide ions by gaseous reagents such as HC1 and Cl2.