- Email: [email protected]
metal hydride batteries through copper-coating
Surface and Coatings Technology 167 (2003) 263–268
Improved performance of a metal hydride electrode for nickelymetal hydride batteries through copper-coating F. Feng, D.O. Northwood* Mechanical, Automotive and Materials Engineering, University of Windsor, Windsor, Ont., Canada N9B 3P4
Abstract Four important properties (exchange current density, apparent activation energy, high-rate dischargeability and discharge potential) were studied for a LaNi4.7 Al0.3 metal hydride (MH) electrode fabricated using alloy powders coated with copper. A half-cell set-up was employed with a 6 M KOH electrolyte solution. The results are compared with those for a similar electrode without the Cu-coating. The exchange current density and apparent activation energy for the MH electrode reaction have been evaluated using the potentiodynamic method. The Cu-coated powder electrode showed improved performance, i.e., a higher exchange current density, larger high-rate dischargeability and lower discharge potential, compared to the electrode made with uncoated alloy powder. Thus, Cu-coating ensures both stable discharge performance and high specific power of a NiyMH battery. This improvement may be due to the following two reasons: (i) corrosion or oxidation protection of alloy powders; (ii) decrease in contact resistance. There is no significant difference between the two apparent activation energies for the electrode reactions for the electrodes with and without the Cu-coating at the same hydrogen concentration. 䊚 2002 Elsevier Science B.V. All rights reserved. Keywords: Metal hydride electrode; Exchange current density; NiyMH battery
1. Introduction Nickelymetal hydride (NiyMH) rechargeable batteries have been attracting considerable attention because of their inherent advantages such as high energy density and non-polluting nature w1,2x. The performance of a NiyMH battery closely depends on the characteristics of the negative MH electrode. Fundamental properties, such as maximum discharge capacity, cycle life time, exchange current density, apparent activation energy, high-rate dischargeability and discharge overpotential, can be used to evaluate the electrochemical performance of MH electrodes. A lower exchange current density leads to a higher overpotential. A larger overpotential leads to a decrease in usable capacity and an increase in anode corrosion, and thus results in a further decrease in cycle life. Iwakura et al. w3–5x have found that the high-rate dischargeability increases asymptotically with increasing exchange current density. The magnitude of the exchange current density is mainly determined by the *Corresponding author. Tel.: q1-519-253-3000 Ext. 4785; fax: q 1-519-973-7007. E-mail address: [email protected] (D.O. Northwood).
structure of the electrodes and the composition of the hydrogen-absorbing alloys w6x, but also by the charge transfer process at the interface between the MH electrode and the electrolyte. The exchange current density can also be used to derive an apparent activation energy, which is a useful intrinsic parameter for evaluating the electrochemical properties of MH electrodes. Significant improvements in performance have been realized by microencapsulation of the hydrogen storage alloy powder in various kinds of electroless coatings, e.g., Cu, Ni–P and Ni–B w4,7–9x. Microencapsulation provides complete coverage of the intermetallic alloy powder to prevent corrosion or oxidation. Metals such as Cu and Ni are softer and less brittle than the intermetallic alloy, and thus provide both the necessary contact between alloy powders, and good electrical and thermal conduction w7x. Microencapsulation has been shown to be effective in improving the cycle lifetime and in generating a high dischargeability w8,9x. The steady-state performance of a LaNi4.7Al0.3 MH electrode with a direct Cu addition, or a NiqCu addition, has been investigated in our previous work w10,11x. The exchange current density at 0 8C of the LaNi4.7Al0.3 electrode with a Cu addition is approxi-
0257-8972/03/$ - see front matter 䊚 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0257-8972(02)00917-9
264
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
mately 150 mA gy1, but it is only approximately 30 mA gy1 for the electrode without a Cu addition w10x. The direct addition of Cu powders to the MH alloy particles is a simple and inexpensive process. However, the performance of the MH alloy particles can be further enhanced by microencapsulation (i.e. copper-coating). The objectives of the current study are to investigate the steady-state performance (i.e. exchange current density and apparent activation energy) and non-steadystate performance (i.e. high-rate dischargeability and discharge potential) of both Cu-coated and uncoated LaNi4.7Al0.3 MH electrodes and identify the reason(s) for any performance improvement. The present study involved the determination of the electrocatalytic characteristics (exchange current density and apparent activation energy) for the discharge process of the LaNi4.7Al0.3 electrodes at temperatures ranging from 273 to 318 K. 2. Experimental The LaNi4.7Al0.3 alloy was obtained from Ergenics (Hy-stor 207). It was first mechanically pulverized to a particle size of approximately 40–60 mm before plating. The copper plating is done by immersing the alloy powder in an aqueous acid bath containing only CuSO4 (0.16 g mly1) and H2SO4 (pH 4–5). The coating is easy to apply without the need for any pretreatment of air-exposed alloys. The negative electrode with the Cu-coated powders was made by mixing the alloy powder (the weight of the alloy powder with Cu-coating was 70.00 mg and the uncoated LaNi4.7Al0.3 alloy was 63.70 mg) with nickel powder (63.70 mg). Thus the weight ratio of active material to nickel powder is 1:1. A polytetrafluoroethylene dispersion (4 wt.%) was added to the mixture as a binder. The mixture was filled into a porous nickel foam plate with a diameter of 1.0 cm, and was then pressed at a pressure of 500 MPa for 2 min into a sheet of 0.5 mm thickness. The results obtained for this electrode were compared with previous results for an electrode without Cu-coating as reported in Ref. w11x. A sintered glass apparatus with three compartments was used for the electrochemical chargeydischarge experiments. The negative electrode was placed in the central compartment and two nickel positive electrodes (Ni(OH)2 yNiOOH) were placed on either side. The hydride electrode and counter electrode were separated by a porous frit. The experimental apparatus was set in a water bath at controlled ("0.5 K) temperatures ranging from 273 to 318 K. The chargeydischarge and polarization tests were conducted using a Solartron 1285 Potentiostat with CORRWARE software. The emphasis of these chargeydischarge tests of the cells was on the electrochemical stability of the negative electrode. Thus, the capacity of the positive electrode was designed to
be sufficiently higher than that of the negative MH electrode to avoid cathode limitation. The electrolyte was a 6 M KOH aqueous solution. A HgyHgOy6 M KOH electrode was used as a reference electrode. A Luggin capillary tube, which connected to the reference electrode and working electrode, was placed close to the working electrode in order to minimize the ohmic drop across the electrolyte solution. After an activation treatment, which involved 8 chargeydischarge cycles, the MH electrode was charged at a constant current (60 mA gy1) until the hydrogen concentration reached its saturated value. The potential of the MH electrode was measured against the HgyHgO reference electrode. The linear polarization curves were obtained under potentiodynamic conditions at a scan rate of 1 mV sy1 after the open-circuit potential (i.e. equilibrium potential) was stabilized (i.e. a variation in the potential was less than 1 mV for 1 h). Then, the MH electrode was discharged for 1 h at a constant current (30 mA gy1), and the above procedure was repeated until the electrode was discharged to a potential of y0.6 V vs. the HgyHgO reference electrode. Based on the set-up of the Solartron 1285 Potentiostat, the value of the anodic polarization current is defined as a positive value. 3. Results and discussion 3.1. Determination of exchange current density In an alkaline aqueous solution, the hydrogen atoms produced at the surface of the MH alloy powder during the charge process are instantly adsorbed and then diffuse into the bulk of the MH alloy. The electrochemical reactions are therefore the charge transfer process at the interface between the MH alloy powder and the electrolyte. The electrochemical kinetics of the charge transfer process is determined by DC polarization methods and the slope of the linear polarization curve represents the electrode resistance, composed of the ohmic and polarization resistances. The diffusion resistance is usually negligible in polarization tests at scan rates of 1 mV sy1. The charge transfer process can be determined by using the exchange current density. At a small overpotential (-20 mV), the conventional Butler–Volmer equation can be rewritten as w12x jsI0
Fh RT
(1)
where h is the overpotential of the electrochemical reaction for a MH electrode (mV); j is the applied current density (mA gy1); I0 is the exchange current density (mA gy1); F is the Faraday constant (96 487 C moly1) and R is the gas constant (8.314 J Ky1 moly1). Therefore the exchange current density
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
265
3.2. Effect of copper coating on exchange current density
Fig. 1. Steady-state polarization curves at 25 8C for both Cu-coated and uncoated LaNi4.7Al0.3 electrodes at different hydrogen concentrations.
can be obtained from the slope of the j vs. h plot. Even though the value of I0 is obtained for the small overpotential region, it should be applicable at large overpotential regions during a non-equilibrium discharge process because it reflects the intrinsic characteristics of the electrode. Selected steady-state polarization curves obtained at 25 8C in the half-cell set-up are shown in Fig. 1 for MH electrodes with and without Cu-coating the powders. The slope of polarization curves for the electrode with Cu-coated powders is lower than that of the electrode made from uncoated powders. This means that the exchange current density of the electrode with Cucoated powders is higher than that of the electrode without the Cu-coating. The exchange current densities were estimated using Eq. (1) at different states of discharge for the electrodes with and without Cu-coating at different temperatures and the results are presented in Fig. 2. The linear polarization process mainly involves the hydrogen reduction and oxidation reactions at the surface layer of the alloy powder. The polarization process does not involve the hydrogen diffusion process in the interior of the MH alloy powder. Thus, the linear polarization is reversible for the hydrogen redox reactions. As the internal resistance (R) is inversely proportional to the exchange current density of an electrode, i.e., RshyIsRTy(FI0 ), the high exchange current density reflects not only the high reaction rate of the electrode, i.e., high rate chargeydischarge capability, but also the low degradation rate of the electrode performance.
The exchange current densities of the Cu-coated electrode are higher compared to the uncoated electrode at temperatures ranging from 273 to 318 K (Fig. 2). This indicates a better performance for the Cu-coated electrode. This lower polarization (i.e. higher exchange current density) for Cu-coated electrode may be due to the following two facts: (i) corrosion or oxidation protection of alloy powders; (ii) decrease in contact resistance (less ohmic polarization) brought about by the copper in the active layer and conducting grid. Microencapsulation by copper obviously provides complete coverage of the intermetallic alloy powder surface that helps prevent corrosion or oxidation of both the active material (LaNi4.7Al0.3 alloy) and the catalyst (Ni clusters). Also, the presence of copper grains enhances the effectiveness of the current collection processes, and further improves the charge transfer process on the alloy powder and electrolyte interface. Therefore it is reasonable to conclude that the microencapsulation by copper of the LaNi4.7Al0.3 alloy powders can act as a micro-current collector of the electrode alloy powder, and thus increase the exchange current density. Scanning electron microscopy analysis shows that the surface of the uncoated alloy particles are typically relatively smooth cleavage planes whereas the surface of the 9 wt.% Cu-coated alloy particles has an ‘orange-peel’ appearance (Fig. 3a and b). In Fig. 3a, the small particles on the smooth cleavage plane are the LaNi4.7Al0.3 alloy debris. Given the complete change in appearance of the surface of the alloy particles after Cu-
Fig. 2. Variation of exchange current density (I0) with temperature for both coated and uncoated LaNi4.7Al0.3 electrodes at different hydrogen concentrations.
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
266
coating, it is believed that copper has fully coated the alloy particles. 3.3. Apparent activation energy of Cu-coated material The apparent activation energy for dehydriding is calculated from the plot of exchange current density at constant hydrogen concentration vs. 1yT using the following equation w13x: w
≠ ln I0 z B1E ≠C F y DTG ~
x
EasyR
|
(2)
where Ea is the apparent activation energy (J molHy1). The unit of Ea used is J molHy1 because the product of the MH–KOH electrochemical reaction is one H atom. A plot of ln I0 vs. 1yT should produce a straight line assuming a constant activation energy, and the apparent activation energy is calculated from the slope of this line. The apparent activation energy is interpreted as the energy that the reactants must possess in order for the reaction to occur. The apparent activation energy decreases with increasing hydrogen concentration for both coated and uncoated MH electrodes: see Fig. 4. The decrease in the apparent activation energy at higher hydrogen concentrations can probably be attributed to a decrease in the activation for hydrogen adsorption. It can also be seen from Fig. 4 that there is no significant difference between the apparent activation energies for the coated and uncoated electrodes at the same hydrogen concentration. This indicates that microencapsulation by copper does not change the activation of the electrochemical reaction. In the other words, copper coating
Fig. 4. Apparent activation energy (Ea) vs. hydrogen concentration for the LaNi4.7Al0.3 hydride electrodes.
does not facilitate the electrochemical reaction for the MH electrode. 3.4. High-rate dischargeability of Cu-coated material Significant improvements in high-rate dischargeability of a MH electrode have been observed for a 9.0 wt.% copper coated LaNi4.7Al0.3 MH electrode. High-rate dischargeability is calculated from the ratio of high (200 mA gy1) to low (20 mA gy1) rate discharge capacity:
High-rate dischargeabilitys
Q200 Q20
(3)
Fig. 3. Scanning electron micrograph of the LaNi4.7 Al0.3 alloy particle surface without Cu-coating (a) and with 9.0 wt.% Cu-coating (b). In (a), the surface state of alloy particles is smooth cleavage and the small particles on the smooth cleavage plane are the LaNi4.7Al0.3 alloy debris. In (b), the surface state of alloy particles is changed and has an ‘orange-peel’ appearance after Cu-coating.
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
267
vs. HgyHgO at 20 mA gy1 discharge current density and y0.833 V vs. HgyHgO at 200 mA gy1 discharge current density. The discharge potentials of the Cucoated LaNi4.7Al0.3 electrode are approximately the same value (i.e. y0.930 V vs. HgyHgO) at both 20 and 200 mA gy1 discharge current densities. This indicates that Cu-coating ensures low discharge potential of a NiyMH battery at a relatively large discharge current density. Thus, Cu-coating ensures not only stable discharge performance but also high specific power of a NiyMH battery at a relatively large discharge current density, which is attributed to the lower resistance of the electrode after Cu-coating the particles. 4. Conclusions Fig. 5. High-rate dischargeability of both LaNi4.7Al0.3 alloys electrodes with and without 9.0 wt.% Cu-coating.
where Q200 is the discharge capacity at a discharge current density of 200 mA gy1 and Q20 is the discharge capacity at a discharge current density of 20 mA gy1, respectively. Fig. 5 shows both high and low rate discharge curves for both the LaNi4.7Al0.3 electrode and Cu-coated LaNi4.7Al0.3 electrode, respectively. The high-rate dischargeabilities are 88.4% for the LaNi4.7Al0.3 electrode and 99.4% for Cu-coated LaNi4.7Al0.3 electrode (Table 1). The factors that improve high-rate dischargeability are the same that improve the exchange current density. 3.5. Discharge potential of Cu-coated material Discharge potential of a MH electrode is an important property, especially when NiyMH batteries are applied to electric vehicles (EVs) because discharge potential determines the specific power of EVs. A lower value (i.e. more negative) of discharge potential leads to a high specific power of the battery. It can also be seen from Fig. 5 that the discharge potential for the Cucoated LaNi4.7Al0.3 electrode is lower than that for the LaNi4.7Al0.3 electrode, especially at a large discharge current density (i.e. 200 mA gy1). The discharge potential of the uncoated LaNi4.7Al0.3 electrode is y0.914 V Table 1 High-rate dischargeability parameters of both LaNi4.7 Al0.3 and 9.0 wt.% Cu-coated LaNi4.7Al0.3 alloy electrodes Alloy electrode
Discharge current density (mA gy1)
Discharge time (h)
High-rate dischargeability (%)
LaNi4.7Al0.3
20 200
15.5 1.37
88.4
9.0 wt.% Cu-coated LaNi4.7Al0.3
20 200
16.1 1.6
99.4
Four important properties (exchange current density, apparent activation energy, high-rate dischargeability and discharge potential) of LaNi4.7Al0.3 MH electrodes with and without Cu-coating were investigated in a half-cell set-up. The principal experimental findings are as follows: (1) The electrode made with Cu-coated powders showed higher exchange current densities, larger highrate dischargeability and lower discharge potential of a NiyMH battery at a relatively large discharge current density compared to the electrode made with uncoated powders. Thus, Cu-coating ensures not only stable discharge performance but also high specific power of a NiyMH battery. It is suggested that this is due to the following two reasons: (i) corrosion or oxidation protection of alloy powders; (ii) decrease in contact resistance. (2) There is no significant difference in the apparent activation energies for the reactions of the electrodes with and without Cu-coating at the same hydrogen concentration. Thus microencapsulation by copper of the alloy powders does not change the activation of the electrochemical reaction. Acknowledgments The authors are grateful to Mr. G. Lin, Mr. J. Zhang and Prof. Z. Zhou (Institute of Hydrogen Storage Materials, Shanghai University, Shanghai, China 200072) for their help with the copper-coating process. Funding for this work is being provided by the Natural Science and Engineering Research Council of Canada through a Research Grant (A4391) awarded to Professor Derek O. Northwood. References w1x J.J.G. Willems, Philips J. Res. 39 (Suppl. 1) (1984) 3. w2x T. Sakai, T. Hazama, H. Miyamura, N. Kuriyama, A. Kato, H. Ishikawa, J. Less-Common Met. 172–174 (1991) 1175.
268
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
w3x C. Iwakura, M. Matsuoka, K. Asai, T. Kohno, J. Power Sources 38 (1992) 335. w4x M. Matsuoka, K. Asai, Y. Fukumoto, C. Iwakura, J. Alloys Comp. 192 (1993) 149. w5x C. Iwakura, Y. Fukumoto, M. Matsuoka, T. Kohno, K. Shinmou, J. Alloys Comp. 192 (1993) 152. w6x H. Kronberger, Int. J. Hydrogen Energ. 21 (1996) 577. w7x H.H. Law, B. Vyas, S.M. Zahurak, G.W. Kammlott, J. Electrochem. Soc. 143 (1996) 2596. w8x T. Sakai, H. Ishikawa, K. Oguro, C. Iwakura, H. Yoneyama, J. Electrochem. Soc. 134 (1987) 558.
w9x C. Iwakura, Y. Kajiya, H. Yoneyama, T. Sakai, K. Oguro, H. Ishikawa, J. Electrochem. Soc. 136 (1989) 1351. w10x J. Han, Ph.D. Dissertation, University of Windsor, Canada, 2000, pp. 283–289. w11x F. Feng, J. Han, M. Shen, M. Geng, Z. Zhou, D.O. Northwood, J. New Mater. Electrochem. Sys. 2 (1999) 45. w12x B.N. Popov, G. Zheng, R.E. White, J. Appl. Electrochem. 26 (1996) 603. w13x E. Ndzebet, O. Savadogo, Int. J. Hydrogen Energ. 20 (1995) 635.
Improved performance of a metal hydride electrode for nickelymetal hydride batteries through copper-coating F. Feng, D.O. Northwood* Mechanical, Automotive and Materials Engineering, University of Windsor, Windsor, Ont., Canada N9B 3P4
Abstract Four important properties (exchange current density, apparent activation energy, high-rate dischargeability and discharge potential) were studied for a LaNi4.7 Al0.3 metal hydride (MH) electrode fabricated using alloy powders coated with copper. A half-cell set-up was employed with a 6 M KOH electrolyte solution. The results are compared with those for a similar electrode without the Cu-coating. The exchange current density and apparent activation energy for the MH electrode reaction have been evaluated using the potentiodynamic method. The Cu-coated powder electrode showed improved performance, i.e., a higher exchange current density, larger high-rate dischargeability and lower discharge potential, compared to the electrode made with uncoated alloy powder. Thus, Cu-coating ensures both stable discharge performance and high specific power of a NiyMH battery. This improvement may be due to the following two reasons: (i) corrosion or oxidation protection of alloy powders; (ii) decrease in contact resistance. There is no significant difference between the two apparent activation energies for the electrode reactions for the electrodes with and without the Cu-coating at the same hydrogen concentration. 䊚 2002 Elsevier Science B.V. All rights reserved. Keywords: Metal hydride electrode; Exchange current density; NiyMH battery
1. Introduction Nickelymetal hydride (NiyMH) rechargeable batteries have been attracting considerable attention because of their inherent advantages such as high energy density and non-polluting nature w1,2x. The performance of a NiyMH battery closely depends on the characteristics of the negative MH electrode. Fundamental properties, such as maximum discharge capacity, cycle life time, exchange current density, apparent activation energy, high-rate dischargeability and discharge overpotential, can be used to evaluate the electrochemical performance of MH electrodes. A lower exchange current density leads to a higher overpotential. A larger overpotential leads to a decrease in usable capacity and an increase in anode corrosion, and thus results in a further decrease in cycle life. Iwakura et al. w3–5x have found that the high-rate dischargeability increases asymptotically with increasing exchange current density. The magnitude of the exchange current density is mainly determined by the *Corresponding author. Tel.: q1-519-253-3000 Ext. 4785; fax: q 1-519-973-7007. E-mail address: [email protected] (D.O. Northwood).
structure of the electrodes and the composition of the hydrogen-absorbing alloys w6x, but also by the charge transfer process at the interface between the MH electrode and the electrolyte. The exchange current density can also be used to derive an apparent activation energy, which is a useful intrinsic parameter for evaluating the electrochemical properties of MH electrodes. Significant improvements in performance have been realized by microencapsulation of the hydrogen storage alloy powder in various kinds of electroless coatings, e.g., Cu, Ni–P and Ni–B w4,7–9x. Microencapsulation provides complete coverage of the intermetallic alloy powder to prevent corrosion or oxidation. Metals such as Cu and Ni are softer and less brittle than the intermetallic alloy, and thus provide both the necessary contact between alloy powders, and good electrical and thermal conduction w7x. Microencapsulation has been shown to be effective in improving the cycle lifetime and in generating a high dischargeability w8,9x. The steady-state performance of a LaNi4.7Al0.3 MH electrode with a direct Cu addition, or a NiqCu addition, has been investigated in our previous work w10,11x. The exchange current density at 0 8C of the LaNi4.7Al0.3 electrode with a Cu addition is approxi-
0257-8972/03/$ - see front matter 䊚 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0257-8972(02)00917-9
264
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
mately 150 mA gy1, but it is only approximately 30 mA gy1 for the electrode without a Cu addition w10x. The direct addition of Cu powders to the MH alloy particles is a simple and inexpensive process. However, the performance of the MH alloy particles can be further enhanced by microencapsulation (i.e. copper-coating). The objectives of the current study are to investigate the steady-state performance (i.e. exchange current density and apparent activation energy) and non-steadystate performance (i.e. high-rate dischargeability and discharge potential) of both Cu-coated and uncoated LaNi4.7Al0.3 MH electrodes and identify the reason(s) for any performance improvement. The present study involved the determination of the electrocatalytic characteristics (exchange current density and apparent activation energy) for the discharge process of the LaNi4.7Al0.3 electrodes at temperatures ranging from 273 to 318 K. 2. Experimental The LaNi4.7Al0.3 alloy was obtained from Ergenics (Hy-stor 207). It was first mechanically pulverized to a particle size of approximately 40–60 mm before plating. The copper plating is done by immersing the alloy powder in an aqueous acid bath containing only CuSO4 (0.16 g mly1) and H2SO4 (pH 4–5). The coating is easy to apply without the need for any pretreatment of air-exposed alloys. The negative electrode with the Cu-coated powders was made by mixing the alloy powder (the weight of the alloy powder with Cu-coating was 70.00 mg and the uncoated LaNi4.7Al0.3 alloy was 63.70 mg) with nickel powder (63.70 mg). Thus the weight ratio of active material to nickel powder is 1:1. A polytetrafluoroethylene dispersion (4 wt.%) was added to the mixture as a binder. The mixture was filled into a porous nickel foam plate with a diameter of 1.0 cm, and was then pressed at a pressure of 500 MPa for 2 min into a sheet of 0.5 mm thickness. The results obtained for this electrode were compared with previous results for an electrode without Cu-coating as reported in Ref. w11x. A sintered glass apparatus with three compartments was used for the electrochemical chargeydischarge experiments. The negative electrode was placed in the central compartment and two nickel positive electrodes (Ni(OH)2 yNiOOH) were placed on either side. The hydride electrode and counter electrode were separated by a porous frit. The experimental apparatus was set in a water bath at controlled ("0.5 K) temperatures ranging from 273 to 318 K. The chargeydischarge and polarization tests were conducted using a Solartron 1285 Potentiostat with CORRWARE software. The emphasis of these chargeydischarge tests of the cells was on the electrochemical stability of the negative electrode. Thus, the capacity of the positive electrode was designed to
be sufficiently higher than that of the negative MH electrode to avoid cathode limitation. The electrolyte was a 6 M KOH aqueous solution. A HgyHgOy6 M KOH electrode was used as a reference electrode. A Luggin capillary tube, which connected to the reference electrode and working electrode, was placed close to the working electrode in order to minimize the ohmic drop across the electrolyte solution. After an activation treatment, which involved 8 chargeydischarge cycles, the MH electrode was charged at a constant current (60 mA gy1) until the hydrogen concentration reached its saturated value. The potential of the MH electrode was measured against the HgyHgO reference electrode. The linear polarization curves were obtained under potentiodynamic conditions at a scan rate of 1 mV sy1 after the open-circuit potential (i.e. equilibrium potential) was stabilized (i.e. a variation in the potential was less than 1 mV for 1 h). Then, the MH electrode was discharged for 1 h at a constant current (30 mA gy1), and the above procedure was repeated until the electrode was discharged to a potential of y0.6 V vs. the HgyHgO reference electrode. Based on the set-up of the Solartron 1285 Potentiostat, the value of the anodic polarization current is defined as a positive value. 3. Results and discussion 3.1. Determination of exchange current density In an alkaline aqueous solution, the hydrogen atoms produced at the surface of the MH alloy powder during the charge process are instantly adsorbed and then diffuse into the bulk of the MH alloy. The electrochemical reactions are therefore the charge transfer process at the interface between the MH alloy powder and the electrolyte. The electrochemical kinetics of the charge transfer process is determined by DC polarization methods and the slope of the linear polarization curve represents the electrode resistance, composed of the ohmic and polarization resistances. The diffusion resistance is usually negligible in polarization tests at scan rates of 1 mV sy1. The charge transfer process can be determined by using the exchange current density. At a small overpotential (-20 mV), the conventional Butler–Volmer equation can be rewritten as w12x jsI0
Fh RT
(1)
where h is the overpotential of the electrochemical reaction for a MH electrode (mV); j is the applied current density (mA gy1); I0 is the exchange current density (mA gy1); F is the Faraday constant (96 487 C moly1) and R is the gas constant (8.314 J Ky1 moly1). Therefore the exchange current density
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
265
3.2. Effect of copper coating on exchange current density
Fig. 1. Steady-state polarization curves at 25 8C for both Cu-coated and uncoated LaNi4.7Al0.3 electrodes at different hydrogen concentrations.
can be obtained from the slope of the j vs. h plot. Even though the value of I0 is obtained for the small overpotential region, it should be applicable at large overpotential regions during a non-equilibrium discharge process because it reflects the intrinsic characteristics of the electrode. Selected steady-state polarization curves obtained at 25 8C in the half-cell set-up are shown in Fig. 1 for MH electrodes with and without Cu-coating the powders. The slope of polarization curves for the electrode with Cu-coated powders is lower than that of the electrode made from uncoated powders. This means that the exchange current density of the electrode with Cucoated powders is higher than that of the electrode without the Cu-coating. The exchange current densities were estimated using Eq. (1) at different states of discharge for the electrodes with and without Cu-coating at different temperatures and the results are presented in Fig. 2. The linear polarization process mainly involves the hydrogen reduction and oxidation reactions at the surface layer of the alloy powder. The polarization process does not involve the hydrogen diffusion process in the interior of the MH alloy powder. Thus, the linear polarization is reversible for the hydrogen redox reactions. As the internal resistance (R) is inversely proportional to the exchange current density of an electrode, i.e., RshyIsRTy(FI0 ), the high exchange current density reflects not only the high reaction rate of the electrode, i.e., high rate chargeydischarge capability, but also the low degradation rate of the electrode performance.
The exchange current densities of the Cu-coated electrode are higher compared to the uncoated electrode at temperatures ranging from 273 to 318 K (Fig. 2). This indicates a better performance for the Cu-coated electrode. This lower polarization (i.e. higher exchange current density) for Cu-coated electrode may be due to the following two facts: (i) corrosion or oxidation protection of alloy powders; (ii) decrease in contact resistance (less ohmic polarization) brought about by the copper in the active layer and conducting grid. Microencapsulation by copper obviously provides complete coverage of the intermetallic alloy powder surface that helps prevent corrosion or oxidation of both the active material (LaNi4.7Al0.3 alloy) and the catalyst (Ni clusters). Also, the presence of copper grains enhances the effectiveness of the current collection processes, and further improves the charge transfer process on the alloy powder and electrolyte interface. Therefore it is reasonable to conclude that the microencapsulation by copper of the LaNi4.7Al0.3 alloy powders can act as a micro-current collector of the electrode alloy powder, and thus increase the exchange current density. Scanning electron microscopy analysis shows that the surface of the uncoated alloy particles are typically relatively smooth cleavage planes whereas the surface of the 9 wt.% Cu-coated alloy particles has an ‘orange-peel’ appearance (Fig. 3a and b). In Fig. 3a, the small particles on the smooth cleavage plane are the LaNi4.7Al0.3 alloy debris. Given the complete change in appearance of the surface of the alloy particles after Cu-
Fig. 2. Variation of exchange current density (I0) with temperature for both coated and uncoated LaNi4.7Al0.3 electrodes at different hydrogen concentrations.
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
266
coating, it is believed that copper has fully coated the alloy particles. 3.3. Apparent activation energy of Cu-coated material The apparent activation energy for dehydriding is calculated from the plot of exchange current density at constant hydrogen concentration vs. 1yT using the following equation w13x: w
≠ ln I0 z B1E ≠C F y DTG ~
x
EasyR
|
(2)
where Ea is the apparent activation energy (J molHy1). The unit of Ea used is J molHy1 because the product of the MH–KOH electrochemical reaction is one H atom. A plot of ln I0 vs. 1yT should produce a straight line assuming a constant activation energy, and the apparent activation energy is calculated from the slope of this line. The apparent activation energy is interpreted as the energy that the reactants must possess in order for the reaction to occur. The apparent activation energy decreases with increasing hydrogen concentration for both coated and uncoated MH electrodes: see Fig. 4. The decrease in the apparent activation energy at higher hydrogen concentrations can probably be attributed to a decrease in the activation for hydrogen adsorption. It can also be seen from Fig. 4 that there is no significant difference between the apparent activation energies for the coated and uncoated electrodes at the same hydrogen concentration. This indicates that microencapsulation by copper does not change the activation of the electrochemical reaction. In the other words, copper coating
Fig. 4. Apparent activation energy (Ea) vs. hydrogen concentration for the LaNi4.7Al0.3 hydride electrodes.
does not facilitate the electrochemical reaction for the MH electrode. 3.4. High-rate dischargeability of Cu-coated material Significant improvements in high-rate dischargeability of a MH electrode have been observed for a 9.0 wt.% copper coated LaNi4.7Al0.3 MH electrode. High-rate dischargeability is calculated from the ratio of high (200 mA gy1) to low (20 mA gy1) rate discharge capacity:
High-rate dischargeabilitys
Q200 Q20
(3)
Fig. 3. Scanning electron micrograph of the LaNi4.7 Al0.3 alloy particle surface without Cu-coating (a) and with 9.0 wt.% Cu-coating (b). In (a), the surface state of alloy particles is smooth cleavage and the small particles on the smooth cleavage plane are the LaNi4.7Al0.3 alloy debris. In (b), the surface state of alloy particles is changed and has an ‘orange-peel’ appearance after Cu-coating.
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
267
vs. HgyHgO at 20 mA gy1 discharge current density and y0.833 V vs. HgyHgO at 200 mA gy1 discharge current density. The discharge potentials of the Cucoated LaNi4.7Al0.3 electrode are approximately the same value (i.e. y0.930 V vs. HgyHgO) at both 20 and 200 mA gy1 discharge current densities. This indicates that Cu-coating ensures low discharge potential of a NiyMH battery at a relatively large discharge current density. Thus, Cu-coating ensures not only stable discharge performance but also high specific power of a NiyMH battery at a relatively large discharge current density, which is attributed to the lower resistance of the electrode after Cu-coating the particles. 4. Conclusions Fig. 5. High-rate dischargeability of both LaNi4.7Al0.3 alloys electrodes with and without 9.0 wt.% Cu-coating.
where Q200 is the discharge capacity at a discharge current density of 200 mA gy1 and Q20 is the discharge capacity at a discharge current density of 20 mA gy1, respectively. Fig. 5 shows both high and low rate discharge curves for both the LaNi4.7Al0.3 electrode and Cu-coated LaNi4.7Al0.3 electrode, respectively. The high-rate dischargeabilities are 88.4% for the LaNi4.7Al0.3 electrode and 99.4% for Cu-coated LaNi4.7Al0.3 electrode (Table 1). The factors that improve high-rate dischargeability are the same that improve the exchange current density. 3.5. Discharge potential of Cu-coated material Discharge potential of a MH electrode is an important property, especially when NiyMH batteries are applied to electric vehicles (EVs) because discharge potential determines the specific power of EVs. A lower value (i.e. more negative) of discharge potential leads to a high specific power of the battery. It can also be seen from Fig. 5 that the discharge potential for the Cucoated LaNi4.7Al0.3 electrode is lower than that for the LaNi4.7Al0.3 electrode, especially at a large discharge current density (i.e. 200 mA gy1). The discharge potential of the uncoated LaNi4.7Al0.3 electrode is y0.914 V Table 1 High-rate dischargeability parameters of both LaNi4.7 Al0.3 and 9.0 wt.% Cu-coated LaNi4.7Al0.3 alloy electrodes Alloy electrode
Discharge current density (mA gy1)
Discharge time (h)
High-rate dischargeability (%)
LaNi4.7Al0.3
20 200
15.5 1.37
88.4
9.0 wt.% Cu-coated LaNi4.7Al0.3
20 200
16.1 1.6
99.4
Four important properties (exchange current density, apparent activation energy, high-rate dischargeability and discharge potential) of LaNi4.7Al0.3 MH electrodes with and without Cu-coating were investigated in a half-cell set-up. The principal experimental findings are as follows: (1) The electrode made with Cu-coated powders showed higher exchange current densities, larger highrate dischargeability and lower discharge potential of a NiyMH battery at a relatively large discharge current density compared to the electrode made with uncoated powders. Thus, Cu-coating ensures not only stable discharge performance but also high specific power of a NiyMH battery. It is suggested that this is due to the following two reasons: (i) corrosion or oxidation protection of alloy powders; (ii) decrease in contact resistance. (2) There is no significant difference in the apparent activation energies for the reactions of the electrodes with and without Cu-coating at the same hydrogen concentration. Thus microencapsulation by copper of the alloy powders does not change the activation of the electrochemical reaction. Acknowledgments The authors are grateful to Mr. G. Lin, Mr. J. Zhang and Prof. Z. Zhou (Institute of Hydrogen Storage Materials, Shanghai University, Shanghai, China 200072) for their help with the copper-coating process. Funding for this work is being provided by the Natural Science and Engineering Research Council of Canada through a Research Grant (A4391) awarded to Professor Derek O. Northwood. References w1x J.J.G. Willems, Philips J. Res. 39 (Suppl. 1) (1984) 3. w2x T. Sakai, T. Hazama, H. Miyamura, N. Kuriyama, A. Kato, H. Ishikawa, J. Less-Common Met. 172–174 (1991) 1175.
268
F. Feng, D.O. Northwood / Surface and Coatings Technology 167 (2003) 263–268
w3x C. Iwakura, M. Matsuoka, K. Asai, T. Kohno, J. Power Sources 38 (1992) 335. w4x M. Matsuoka, K. Asai, Y. Fukumoto, C. Iwakura, J. Alloys Comp. 192 (1993) 149. w5x C. Iwakura, Y. Fukumoto, M. Matsuoka, T. Kohno, K. Shinmou, J. Alloys Comp. 192 (1993) 152. w6x H. Kronberger, Int. J. Hydrogen Energ. 21 (1996) 577. w7x H.H. Law, B. Vyas, S.M. Zahurak, G.W. Kammlott, J. Electrochem. Soc. 143 (1996) 2596. w8x T. Sakai, H. Ishikawa, K. Oguro, C. Iwakura, H. Yoneyama, J. Electrochem. Soc. 134 (1987) 558.
w9x C. Iwakura, Y. Kajiya, H. Yoneyama, T. Sakai, K. Oguro, H. Ishikawa, J. Electrochem. Soc. 136 (1989) 1351. w10x J. Han, Ph.D. Dissertation, University of Windsor, Canada, 2000, pp. 283–289. w11x F. Feng, J. Han, M. Shen, M. Geng, Z. Zhou, D.O. Northwood, J. New Mater. Electrochem. Sys. 2 (1999) 45. w12x B.N. Popov, G. Zheng, R.E. White, J. Appl. Electrochem. 26 (1996) 603. w13x E. Ndzebet, O. Savadogo, Int. J. Hydrogen Energ. 20 (1995) 635.