Flexible graphene/GO electrode for gel-free EEG

Objective. Developments in electroencephalography (EEG) technology have allowed the use of the brain-computer interface (BCI) outside dedicated labratories. In order to achieve long-term monitoring and detection of EEG signals for BCI application, dry electrodes with good signal quality and high bio compatibility are essential. In 2016, we proposed a flexible dry electrode made of silicone gel and Ag flakes, which showed good signal quality and mechanical robustness. However, the Ag components used in our previous design made the electrode too expensive for commercial adaptation. Approach. In this study, we developed an affordable dry electrode made of silicone gel, metal flakes and graphene/GO based on our previous design. Two types of electrodes with different graphene/GO proportions were produced to explore how the amount of graphene/GO affects the electrode. Main results. During our tests, the electrodes showed low impedance and had good signal correlation to conventional wet electrodes in both the time and frequency domains. The graphene/GO electrode also showed good signal quality in eyes-open EEG recording. We also found that the electrode with more graphene/GO had an uneven surface and worse signal quality. This suggests that adding too much graphene/GO may reduce the electrods' performance. Furthermore, we tested the proposed dry electrodes' capability in detecting steady state visually evoked potential. We found that the dry electrodes can reliably detect evoked potential changes even in the hairy occipital area. Significance. Our results showed that the proposed electrode has good signal quality and is ready for BCI applications.

Automated summary

The study developed an affordable dry electrode using silicone gel, metal flakes, and graphene/GO based on previous design. The electrodes showed low impedance and good signal correlation to wet electrodes, with graphene/GO electrodes demonstrating good signal quality in eyes-open EEG recording. However, excessive graphene/GO led to uneven surfaces and decreased performance, suggesting careful optimization is needed.