Table 1.

Summary of desirable properties and currently used parameters of electrodes for intracortical recording.

Property	Value	Requirements
Materials	Recording sites/Interconnections: Pt, Ir, Pt-Ir, Au, Iridium oxide, Polysilicon, W, Al Substrate/core: Si, Glass, Metal wires, Alumina, Polyimide, LCP, Parylene-C, SU-8, Silk Coatings: SiO2, SixNy, Glass, S-isonel, Teflon, PEDOT, PSS, PPy, CNTs, PEG, Laminin, Silk	Safe to use and able to reliably perform conducting Resistive to attacks of body fluids and products of metabolism Reliable and hermetic chronically
Young's modulus of implant	Polymers: 1 × 106–1 × 109 Pa Silicon: 130–185 × 1 × 109 Pa Metals: >1 × 109 Pa	As close as possible to the elastic modulus of brain tissue (0.1–1 × 103 Pa) Allows for easy implantation without tissue dimpling Reduces movement-induced trauma
Average impedance range of electrode	<1 MΩ Typically: 20–150 kΩ (at 1 kHz)	Of low value to decrease noise Allows for recording of single unit activity Can be obtained with a large surface area (Cogan, 2008)
Dimensions of implant	Diameter: Preferably: <12 μm Typically ca 125 μm	As small as possible Allows decrease foreign body response effect Allows promote interface biocompatibility (Ludwig et al., 2011)
Power density of implant	<40–60 mW/cm2	Small to avoid heating up neural tissue more than 2°C (Wolf, 2008)
Signal to noise ratio	>5	As high as possible to appropriately differentiate shapes of spikes Not below 1.25, as it is then considered noise (Chapin and Nicolelis, 1999; Smith et al., 2002)
Recording site geometry	>50 μm	Allows to decrease impedance and improve recording selectivity As small as possible with high surface to area ratio. Larger for LFP recording (Nelson and Pouget, 2010)
Capacitance of electrode-tissue interface	150 pF–1.5 nF (Different depending on electrode area and surface roughness) (Harrison, 2008)
Number of penetrating shanks per implant	>1–100/implant	Many shanks give space for more recording sites Of mechanical properties allowing for easy implantation, good tissue integration, and minimal tissue displacement during penetration Of possibly minimum volume to avoid extensive trauma
Density of recording sites per penetrating shank	>1–1,000	As many as possible to allow increase in spatial representation of recorded signal and to monitor several single neurons (Scholvin et al., 2016)