Electron transfer processes occurring on platinum neural stimulating electrodes: calculated charge-storage capacities are inaccessible during applied stimulation.

Abstract

Neural prostheses employing platinum electrodes are often constrained by a charge/charge-density parameter known as the Shannon limit. In examining the relationship between charge injection and observed tissue damage, the electrochemistry at the electrode-tissue interface should be considered. The charge-storage capacity (CSC) is often used as a predictor of how much charge an electrode can inject during stimulation, but calculating charge from a steady-state i-E curve (cyclic voltammogram) over the water window misrepresents how electrodes operate during stimulation. We aim to gain insight into why CSC predictions from classic i-E curves overestimate the amount of charge that can be injected during neural stimulation pulsing. In this study, we use a standard electrochemical technique to investigate how platinum electrochemistry depends on the potentials accessed by the electrode and on the electrolyte composition. The experiments indicate: (1) platinum electrodes must be subjected to a 'cleaning' procedure in order to expose the maximum number of surface platinum sites for hydrogen adsorption; (2) the 'cleaned' platinum surface will likely revert to an obstructed condition under typical neural stimulation conditions; (3) irreversible oxygen reduction may occur under neural stimulation conditions, so the consequences of this reaction should be considered; and (4) the presence of the chloride ion (Cl^-) or proteins (bovine serum albumin) inhibits oxide formation and alters H adsorption. These observations help explain why traditional CSC calculations overestimate the charge that can be injected during neural stimulation. The results underscore how careful electrochemical examination of the electrode-electrolyte interface can result in more accurate expectations of electrode performance during applied stimulation.