The improvement of the lithium cycling efficiency obtained by substituting an aluminum substrate for a lithium substrate is explained by means of morphological and kinetic studies of the electrodes in the molar solution LiClO4‐normalpropylene carbonate . SEM observations show that the insertion rate of the lithium deposit into aluminum can be sufficiently high so as to avoid dendritic growth; ESCA analysis reveals that propylene carbonate leads to a chemical formation of a polymeric membrane on the electrodes, which is less important on the aluminum than on the lithium substrate. Electrochemical impedance measurements associated with polarization curve data point out that most of the surface is active, giving rise to an exchange current density of about 17 mA/cm2; these studies also demonstrate that diffusion processes in the passivating layer and in the bulk of the electrode are responsible for the limited lithium cycling performances with the aluminum substrate.

at short and long times of electrolysis, respectively, and the Pb system in C104-solutions follows these relationships. 9,~~ The anodic current due solely to the metal dissolution process (i.e., 2Fkl when Cb(pb + +) --0) can be obtained by extrapolation to t --> 0 from an i vs. t '/~ plot. 9 The correct values obtained in this way differ widely from the steady-state measurements. When the measurements are performed in the way described in this paper, where natural convection conditions are established, the physical meaning of the observed currents can be obtained using the diffusion layer concept. This leads towhere 8 is the diffusion layer thickness. The equivalent equation derived from Eq.[5] Zor long times of electrolysis isBoth Eq.[6] and [7] lead to a similar current-potential relationship which does not give direct information on the anodic process. When kl/k-1 >> Cb(pb + +), a situation which occurs in the potential range where an apparent good Tafel behavior was observed; the "Tafel" slope measured in the steady state is simply an indication of the reversible behavior of this system. Hence, no kinetic information on the anodic behavior can be obtained from steady-state experiments, as is wrongly assumed in the paper under discussion.The term kl/k-1 represents in this case the surface concentration of Pb + + ions, as defined by the Nernst equation. This is the reason why in Fig. 2 of the paper under discussion, the 30 mV slope (really a Nerstian slope for a 2 electron process) is observed for more anodic potentials as the bulk Pb + + ion concentration in solution is increased. It is also our opinion that due to the incorrect way in which the anodic experiment has been performed, an incorrect order with respect to the C104-ion has been found. There is a finite order with respect to C104 -3, as is discussed elsewhere. 10For the above reasons, the results presented in Table I of the paper and the mechanism proposed have to be considered dubious. M. Garreau, J. Th6venin, and D. Warin:z8We entirely agree with Brummer and co-workers when they consider that the determination of lithium cycling efficiency must be investigated, using not a foreign but a dense lithium substrate despite the inherent difficulties of this technique according to the determinant influence of the substrate on the properties of the electrodeposited lithium, x4 However, when faced with the same difficulties as those encountered by the authors, we have studied the cycling behavior of the lithium Facultad de Cleneias Qulmicas, Unlversldad Nacional de Cordoba, 5000 Cordoba, Argentina. A. S. Gioda, M. C. Giordano, and V. A. Macagno, This Journal, 125, 889 (1978). ~Groupe de Recherche du CNRS "Physique des Liquides et Electrochimie," assoeid'a l'Universite Pierre et Marie Curie, 75230 Paris Cedex 05, Francej, ~' M. Garreau, J. Thelvenin, and D. Warin, C.R. Acad. Sci. Paris, 286C, 545 (1978).electrode in a different way. The results we have obtained lead us to interpretations somewhat different from those given in the paper under consideration. ...

No abstract

No abstract