Comparison of Multiple Degrees of Electrode Surface Roughness: Impedance, Charge Storage Capacity, Biofouling and Biocompatibility

Devices called neural electrodes are generally used to record and / or stimulate neural activity whilst extracellularly interfacing with neurons. The ultimate goal of the field is to develop neural electrodes that can continuously record and / or stimulate neural activity for long periods of time (10+ years). Electrodes require low electrode impedance to enable good signal to noise ratio and a high capacity for chemically stable injection of charge (CSC) for stimulating neural activity. Electrode function typically deteriorates after a few months in-vivo due two factors: biofouling and local cellular response (glial scarring). Biofouling is theorised to be an accumulation of multiple proteins at the electrode surface. However this has only been backed up experimentally by single protein models in literature. Electrodes need to be able to combat the effects of biofouling and glial scarring. This study uses nanometre scale roughened gold (Au) electrodes. Electrode roughening has previously been shown to lower impedance, increase CSC and reduce glial response and the effects of biofouling. We compared multiple degrees of roughness with the aim of finding the optimal degree for improving impedance, CSC, biofouling and cellular response. We found that surface roughening increased impedance and only increased CSC for two only degrees of roughness. To find the optimal degree of roughness across conditions, we suspect a larger range of roughness may be necessary to lower impedance with our fabrication technique. This study is the first to use a multiprotein biofouling model. Contrary to literature, we found that incubation with protein decreased impedance, likely due to protein-protein interactions not accounted for in single protein models. Biocompatibility was improved for two degrees of Au roughness. Roughening of SU8, a polymer used to surround the electrodes, decreased biocompatibility. We also used artificial cerebral spinal fluid (aCSF) as an electrolyte, which is more chemically similar to in-vivo than commonly used phosphate buffered saline solution (PBS). The use aCSF as a medium was significant as the measures in aCSF were different from that in PBS.