Extracellular voltage noise probes the interface between retina and silicon chip
Click to view
(Color) Schematic of the interface between neural tissue (retina) and a silicon-based capacitive sensor. The electronic current between source and drain of the silicon-based field effect transistor is modulated by the extracellular voltage changes within the interfaced neural tissue. An inert oxide between sensor and neural tissue prevents electric or ionic charge transfer. The rabbit retina is used here as an example neural tissue. The retina is interfaced in an epiretinal configuration. Major retinal cell classes are shown. The neural tissue is separated from the sensor oxide by an electrolyte-filled cleft (blue). The capacitive sensor (diameter of top contact 7.4 μm), the cleft (0.05–1 μm) and neural tissue (vertical extension 100 μm) are not drawn to scale.
Click to view
(Color) Frequency-independent retinal voltage noise allows for evaluation of tissue resistance in an appropriate frequency range. (a) Overall power spectral density (PSD, SV ) of voltage noise caused by active neural tissue (retina) with recording sensor (black trace). The PSD of the bare sensor is recorded separately (gray trace). The net PSD (green trace) of the interfaced active retina is obtained by subtracting the PSD of the bare sensor from the overall PSD. It is frequency-independent in the range of 10 - 100 kHz. (b) Overall PSD of voltage noise caused by silenced retina with recording sensor (brown trace). Neuronal action potentials are inhibited using TTX. The net PSD (green) due to the interfaced silenced retina is obtained by subtracting the PSD of the bare sensor (gray). It is frequency-independent in the range of 0.1-100 kHz. (c) Overall PSD of the active retina (black trace) as displayed in (a). The PSD of the bare sensor is replaced by a theoretical 1/f spectrum of the sensor (gray trace) that is matched to the overall PSD at 0.1 kHz. The net tissue PSD (orange trace) is obtained as the difference between the measured PSD and the 1/f spectrum. In the frequency range of 10-100 kHz it is similar to the PSD of the silenced retina (green). The arrows mark the suggested frequency range for probing the tissue contact. The constant PSD (dashed line) is used to calculate an effective resistance Reff ∼ 0.9 MΩ. All traces presented in (a)–(c) were obtained from the same retina.
Click to view
(Color) Estimated maps of the retina-chip interface. (a) Experimental PSD at low frequencies represented as a map of the effective resistance Reff (x,y) that is defined by the Nyquist relation. (b) Interpretation of the effective resistance in terms of a model of the retina-chip contact as a map of the tissue conductivity σtissue (x,y) > 0.8 mS/cm assuming a perfect contact with a distance dcleft = 0, and as a map of the distance dcleft (x,y) > 0 assuming a constant tissue conductivity σtissue = 0.8 mS/cm.
Article metrics loading...
Full text loading...