FIGURE 5.

(A) Simplified HD-MEA geometry for COMSOL simulations: 36 planar electrodes at a pitch of 17.5 μm and featuring an area of 5 × 9 μm²; 4 reference electrodes were placed at the borders of the array. E1 and E2 denote the electrodes that were used for the simulations in panel B. prb1 and prb2 denote two probe locations, at the right and at the left side of electrode E1, that were used to compute the voltage and the electric-field values for the three configurations in panel B. The two probe locations were chosen so as to compare the effect of using a second array electrode. (B) The simulation was performed in voltage mode, and an amplitude of ±100 mV was used. The voltage and electric field distributions after stimulating with one electrode against a global reference electrode in solution are represented in panels (i) and (iv), those for a stimulation with two neighbored electrodes with synchronized waveforms of opposite signs are represented in panels (ii) and (v), and those for applying a biphasic waveform to one electrode, while the neighboring electrode was grounded are represented in panels (iii) and (vi). The electric field (EF norm) was calculated as $\sqrt{E_{\text{x}}^2+E_{\text{y}}^2+E_{\text{z}}^2}$. (C) Voltage (i, ii, iii) and electric-field (iv, v, vi) distributions at the probe locations in the respective configurations (i–vi) of panel B. The voltage drop, which is responsible for the AP initiation by electrical stimulation, is comparable in all the cases, except for the two-electrode stimulation (ii), where it is somewhat decreased. The electric field is stronger in the case of the two-electrode stimulation (v). (D) Stimulation with a biphasic voltage waveform of six different neurons using the three configurations simulated in COMSOL in panel B. There were differences in the voltage required to evoke activity in the different neurons (range between 60 and 120 mV), however, there were no major differences for using the three stimulation scenarios explained and displayed in panel B.