Mixed-conducting particulate composites for soft electronics

See allHide authors and affiliations

Science Advances  24 Apr 2020:
Vol. 6, no. 17, eaaz6767
DOI: 10.1126/sciadv.aaz6767
  • Fig. 1 Control of size and density of mixed-conducting particles in an ion conducting scaffolding matrix enables various regimes of electronic operation.

    (A) MCP operates as an anisotropic conducting film (orange), transistor (blue), transistor with independent gating (green), diode (red), or resistor (black) depending on the particle size (α) and mean free path (λ) within the material (as represented by the schematic, upper right). These properties are defined with respect to distance between horizontally adjacent (and, for the anisotropic film, corresponding vertical) electrodes (lower right). The gray rectangles represent substrate and the yellow rectangles represent the electrical terminals. (B) Physical process steps to create conducting polymer particles: A solution of conducting polymer PEDOT:PSS with solvent additives is prepared (top). The solution is evaporated to form a highly conducting film. The film is broken into large particles and crushed into microparticles using a ball mill. The size of microparticles is reduced via sonication (bottom). Scale bar, 50 μm. (C) Conducting polymer particles are filtered to bounded sizes using sieves of distinct pore diameters (left to right, lower and upper bounds are 1 and 10 μm, 10 and 20 μm, 20 and 30 μm, and 30 and 40 μm, respectively). Scale bar, 100 μm.

  • Fig. 2 MCP forms uniform films with controllable particle size and conductivity.

    (A) Micrograph of two conformable arrays bonded together by MCP; arrow indicates the bonding area. Scale bar, 500 μm. Photo credit: Patricia Jastrzebska-Perfect, Columbia University. (B) Optical micrographs of blade-coated MCP films. Concentration [% (w/v)] of conducting polymer particles in nonconducting medium of CS, sorbitol, and glycerol is 6.1 (blue dot) or 18 (red dot). Distinct particles are visible in the nonconducting matrix (top right). Scale bar, 100 μm. (C) Pixel intensity distribution (where black = 0) of MCP; concentration [% (w/v)] of PEDOT:PSS particles in blade-coated solution is 6.1 (blue) or 18 (red). Data are represented as mean darkness (n = 6) ± SEM. Note that a higher concentration of particles produces a darker and wider distribution of pixel intensity. a.u., arbitrary units. (D) Optically measured distributions conducting polymer particle size in MCP (from left to right, using 1 and 10 μm, 10 and 20 μm, 20 and 30 μm, and 30 and 40 μm sieves). (E) Vertical (left axis, open circles) resistance of MCP is proportional to conducting polymer particle density in CS (particle size = 40 μm). Inset shows measurement configuration of vertical and horizontal resistances. O.L., overload.

  • Fig. 3 MCP creates high-performance anisotropic films, independently addressable transistors, and diodes by varying conducting polymer particle size and density.

    (A) MCP operating as an anisotropic conductive film. λ is smaller than electrode spacing. Scaffolding polymer matrix (light yellow), conducting polymer particles (dark blue), Au contacts (gold), and substrate (gray) are visible. (B) Adjacent MCP-bonded electrodes have consistently low vertical resistance (left axis) and high horizontal resistance (right axis) across a wide range of horizontal electrode pitch. (C) MCP operating as an independently gated transistor. λ is longer than d1 but shorter than d2. Mobile ions (green), and Au contacts forming the source (S), drain (D), and gate (G) are visible. (D) Output characteristics of MCP-based transistor operating in depletion mode (L = 250 μm, W = 5 mm, particle size = 30 μm), with VG varying from 0 to 0.6 V, in increments of 0.1 V (top to bottom). (E) NOR gate generated using un-patterned MCP-based transistors. Input signals are applied at gates 1 and 2 (G1 and G2, respectively). (F) Temporal response of MCP-based NOR gate (L = 100 μm, W = 500 μm, particle size = 30 μm) with gate pulse amplitudes of 0.5 V. (G) MCP operating as a diode. α is substantially less than d1, but λ is approximately equal to d1. (H) Output characteristic of MCP-based diode [same dimensions as (D), particle size = 10 μm].

  • Fig. 4 MCP creates an anisotropic interface for high–spatiotemporal resolution electrophysiologic signal transmission.

    (A) Photograph of conformable, flexible, and rigid neural probes with electronic circuits that can be bonded together using MCP to acquire neurophysiological signals in vivo. Scale bar, 5 mm. BGA, ball grid array. Photo credit: Dion Khodagholy, Columbia University. (B) High gamma oscillations are differentiable across electrodes of an MCP-bonded array placed on cortical surface of a freely moving rat [unfiltered local field potential (LFP) traces (black) and corresponding filtered traces (red, 60 to 100 Hz)]. Scale bar, 40 ms. (C) Trigger-averaged gamma band power is spatially confined across array placed on cortical surface of a freely moving rat. Scale bar, 1 mm. The white dashed rectangle indicates the electrodes that generated the traces in (B). (D) MCP-bonded flexible probe inserted into rat hippocampus permits recording of characteristic ripple oscillations in dorsal CA1 [sample wide-band traces (black, 0.1 to 20 kHz) superimposed on a heat map highlighting the instantaneous power in the ripple band (100 to 150 Hz); scale bar, 50 ms] as well as individual action potential waveforms (burst firing of putative pyramidal cell, white trace, zoomed in from location denoted by white star, 0.1 to 20 kHz; scale bar, 5 ms). Color bar same as (C). (E) Intraoperative photograph showing conformable MCP-bonded neural probe on the surface of human cortex with associated amplifier circuits. Scale bar, 10 mm. Photo credit: Dion Khodagholy, Columbia University. REF, reference; GND, ground. (F) Sample wide-band LFP (0.1 to 1250 Hz) acquired during intraoperative human recording demonstrating spatially diverse activity patterns acquired by MCP-bonded neural probe. Scale bar, 100 ms. (G) Spectrogram of neural data acquired by MCP-bonded neural probe revealing transition from anesthetized to awake state intraoperatively. Scale bar, 10 s. Color bar same as (C).

  • Fig. 5 MCP creates an anisotropic interface for high spatiotemporal biopotential sensing.

    (A) Micrograph of a high-density, conformable EMG array adhered to the wrist of a human participant using MCP (top left; scale bar, 10 mm). Cross-sectional schematic (top right) comparing gel (upper) and MCP interfaces (lower) between skin and electronics. Sample traces of MCP-acquired EMG (black) and ECG (red) signals are shown with their corresponding recording site on biceps and the wrist. Scale bar, 1 s; 4 mV. Photo credit: George Spyropoulos, Columbia University. (B) Spectrograms of EMG signals acquired using MCP-adhered conformable array placed over the wrist (bottom left schematic) reveal distinct patterns during voluntary flexion of each finger. Scale bar, 80 ms. (C) MCP-adhered conformable array permits noninvasive recording of independent nerve action potentials. Scale bar, 100 ms; 4 mV. (D) Current source density heat map of a sample nerve action potential from (C) as visualized across adjacent electrodes reveals the source localization and propagation. Scale bar, 5 ms. Color bar same as (B).

Supplementary Materials

  • Supplementary Materials

    Mixed-conducting particulate composites for soft electronics

    Patricia Jastrzebska-Perfect, George D. Spyropoulos, Claudia Cea, Zifang Zhao, Onni J. Rauhala, Ashwin Viswanathan, Sameer A. Sheth, Jennifer N. Gelinas, Dion Khodagholy

    Download Supplement

    The PDF file includes:

    • Figs. S1 to S14
    • Legend for movie S1

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article