Exploiting Spatial Ionic Dynamics in Solid-State Organic Electrochemical Transistors for Multi-Tactile Sensing and Processing

Abstract

The human nervous system inspires the next generation of sensory and communication systems for robotics, human-machine interfaces (HMIs), biomedical applications, and artificial intelligence. Neuromorphic approaches address processing challenges; however, the vast number of sensors and their large-scale distribution complicate analog data manipulation. Conventional digital multiplexers are limited by complex circuit architecture and high supply voltage. Large sensory arrays further complicate wiring. An 'in-electrolyte computing' platform is presented by integrating organic electrochemical transistors (OECTs) with a solid-state polymer electrolyte. These devices use synapse-like signal transport and spatially dependent bulk ionic doping, achieving over 400 times modulation in channel conductance, allowing discrimination of locally random-access events without peripheral circuitry or address assignment. It demonstrates information processing from 12 tactile sensors with a single OECT output, showing clear advantages in circuit simplicity over existing all-electronic, all-digital implementations. This self-multiplexer platform offers exciting prospects for circuit-free integration with sensory arrays for high-quality, large-volume analog signal processing.

Keywords: in‐electrolyte computing; ion modulation; organic electrochemical transistor; self‐multiplexer platform; solid‐state; tactile sensors.

Figure 1

Working principle of the proposed ʻin‐electrolyte computingʼ platform. a) Comparison of conventional multiplexing schemes and our synapse‐like analog multiplexer. b) Schematic of a 5‐gate self‐multiplexing SSOECT, including the materials and chemical structures of the electrolyte polymer matrix, ionic liquid, and semiconductor used in this study. The distances between gates 1 and 5 and the channel are 1, 3, 6, 10, and 15 mm, respectively. The solid electrolyte covers the entire channel and gate area, isolated by a 1.4 µm‐thick parylene layer. Solid electrolytes containing 37.5 wt.% ionic liquid and transistor channels with W/L = 100/10 µm were used for electrical characterizations unless otherwise noted. The thickness of the spin‐coated semiconductor film is 81 nm, and all gates have a size of 600 µm × 600 µm. c) Transfer performance of the transistor for each gate with varying spatial dynamics, shown in linear and semi‐log scales. d) Shift of Vth with varying spatial dynamics of gates due to changes in V_eff. Threshold voltages were extracted from the intercept of the x‐axis. e) Transient behavior of multiterminal SSOECT measured with a rectangular pulse of 20 ms/−2 V pulse width/amplitude. f) Impedance spectrum amplitude (|Z| vs frequency) for varying CE‐WE distances. Measured data are shown with scatters, while the fitted curves using the equivalent circuit are shown with solid lines. g) Calculated bulk resistance and capacitance values of solid electrolytes at varying CE‐WE distances. h) Dynamic shifting of absorption spectra within 5 s under a −3 V gate bias on CE₁ and CE₇. i) Dynamic shift of absorption intensity at 432, 772, and 1200 nm at varying CE‐WE distances.

Figure 2

Investigation of parameters (ionic liquid concentration, channel dimension, channel material, and measurement protocol) influencing the speed and modulation ratio of the multiplexer. a) Calculated bulk resistance and capacitance values of solid electrolytes containing 12.5, 25, 37.5, and 50 wt.% ionic liquid at a 2 mm gate‐semiconductor distance. b) Extracted I_d for five gates from transfer curves under −1 V applied V_g with varying electrolyte concentrations. c) Output behavior of the local and remote gates under −1.4 V applied V_g for 5 min. d) Extracted I_d for five gates from transient curves under varying AC measurement protocols (PW/Amplitude = 0.5s/−0.6 V, 0.1s/−1 V, 0.05s/−1.2 V, and 0.02s/−2 V).

Figure 3

Integration of our analog multiplexer with the pressure sensor. Solid electrolytes with 37.5 wt.% ionic liquid were used. a) Schematic of the micro‐patterned pressure sensor and its integration with the analog multiplexer. b) Output behavior of the pressure sensor from 0.14 to 30.56 kPa under a 0.5 V read voltage. c) Sensitivity of the pressure sensor, extracted by deriving the slope in linear sensing regions. d) Transfer/transient behavior of the integrated system from 0.02 to 21.64 kPa. The pressure sensor is connected to G₁ with −1.5 V/−0.5 V V_g/V_d applied. e) Transient response of gates with varying spatial dynamics under 21.64 kPa. f,g) Comparison of sensitivities in the low and high‐pressure regions when the pressure sensor is connected to G₁ and G₅. OECTs with 37.5 wt.% solid electrolytes for and 25%.

Figure 4

Application of the SSOECT‐based multiplexer for array tactile signal detection. The distances between gates 1 and 12 and the channel are 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, and 23 mm, respectively. Solid electrolytes with a 37.5 wt.% ionic liquid ratio was used. a) Schematic representation of the integrated system. b) Transient behavior of the 12‐gate SSOECT measured with a 300 ms pulse and −1.2 V amplitude. c) Cumulative probability of channel conductance over 100 cycles for all gates. d) Modulation ratio in drain current versus changes in distance. e) Current output of the multiplexer when stimulated with an identical tactile sensor. f) Current output of the multiplexer when pressing single or multiple tactile sensors. g) Pattern recognition achieved by the integrated system through the output current sequence from multiple sampling cycles.