944
Effects of Chloride on the Oxygen Reduction Reaction
EXPERIMENTAL
Measurements were performed in a conventional all-glass, three-compartment electrochemical cell with a rotating Pt ring- Pt disk electrode, Pt|Pt RRDE, (Pine Instruments, (A = 0.164 cm2) using a bipotentiostat (Pine Instruments). A Pt foil and a reversible hydrogen electrode in the same solution (RHE) were used as counter and reference electrodes, respectively. Oxygen reduction polarization curves were recorded in oxygen saturated 0.1 M HClO4.
RESULTS AND DISCUSSION
Shown in the top panel, Fig.1 are polarization curves recorded at a scan rate, ν = 20 mV/s for a Pt disk electrode in oxygen saturated 0.1 M HClO4in the absence (black curve) and in the presence of 20 μM (red curve) and 100 μM NaCl (blue curve). The corresponding currents, measured with the ring polarized at a potential positive enough to detect hydrogen peroxide oxidation under diffusion limited conditions (ca. 1.2 V) are shown in the bottom panel in this figure. As evident from these results, the presence of chloride in the electrolyte leads not only to a significant shift in the onset of oxygen reduction, but also to the generation of hydrogen peroxide. Particularly interesting is the peak-like feature observed in the ring current as a function of the disk potential. Correlations between the chloride coverage and the ORR rate were sought by implementing a method in which the potential was stepped from 0.05 V, a value at which there is no chloride adsorption, to various more positive potentials where chloride adsorption is known to ensue in solutions devoid of oxygen. The results of these experiments, shown in Panel A, Fig. 2, indicate that for certain potentials, the current attains a limiting value associated with the diffusion controlled adsorption of chloride and later decays down to zero signaling that the equilibrium coverage at that potential was reached. Additional insight into this phenomenon was obtained from theoretical simulations performed within the COMSOL platform by coupling the convective diffusion equation with the surface dynamics for the adsorption|desorption process. This strategy yielded transient curves (see Panel B, Fig. 2) which strongly resemble those found experimentally. Efforts are now underway to explore whether a more detailed analysis of the curves will afford reliable values of the kinetic parameters involved.
ACKNOWLEDGEMENTS
This work was supported by a grant from NSF.
REFERENCES
1. J. Zhang, Frontiers in Energy, 5, 137, 2011.
2. Wang, Z.-L.; Xu, D.; Xu, J.-J.; Zhang, X.-B., Chem. Soc. Rev. 2014.
3. N. M. Markovic, H. A. Gasteiger, B. N. Grgur, and P. N. Ross, J. Electroanal.Chem., 467, 157, 1999.
4. T. J. Schmidt, U. A. Paulus, H. A. Gasteiger, and R. J. Behm, J. Electroanal. Chem., 508, 41, 2001.
5. N. M. Markovic, R. R. Adzic, B. D. Cahan, and E. B. Yeager, J. Electroanal.Chem., 377, 249, 1994.
Figure Captions:
Fig. 1. Polarization curves recorded at a scan rate, ν = 20 mV/s for a Pt disk electrode in oxygen saturated 0.1 M HClO4 in the absence (black) and in the presence of 20 μM (red) or 100 μM (blue) NaCl. The corresponding ring currents recorded when the ring was polarized at Ering= 1.2 V are shown in the lower panel.
Fig. 2. Panel A: Chronoamperometric curves recorded in 0.1 M HClO4 with 20μM NaCl at a rotation rate of ɷ= 400 rpm starting from Eini=0.05 V to the various final potentials as indicated in the legend. Panel B: Simulated currents for kads/kdes =106/M, 900 rpm, C0=.1mM, Γmax=2x10-9mol/cm2 and for various values for the adsorption rate (legend). Limiting current is consistent with Levich equation.