Conductive and elastic bottlebrush elastomers for ultrasoft electronics

Abstract

Understanding biological systems and mimicking their functions require electronic tools that can interact with biological tissues with matched softness. These tools involve biointerfacing materials that should concurrently match the softness of biological tissue and exhibit suitable electrical conductivities for recording and reading bioelectronic signals. However, commonly employed intrinsically soft and stretchable materials usually contain solvents that limit stability for long-term use or possess low electronic conductivity. To date, an ultrasoft (i.e., Young's modulus <30 kPa), conductive, and solvent-free elastomer does not exist. Additionally, integrating such ultrasoft and conductive materials into electronic devices is poorly explored. This article reports a solvent-free, ultrasoft and conductive PDMS bottlebrush elastomer (BBE) composite with single-wall carbon nanotubes (SWCNTs) as conductive fillers. The conductive SWCNT/BBE with a filler concentration of 0.4 - 0.6 wt% reveals an ultralow Young's modulus (<11 kPa) and satisfactory conductivity (>2 S/m) as well as adhesion property. Furthermore, we fabricate ultrasoft electronics based on laser cutting and 3D printing of conductive and non-conductive BBEs and demonstrate their potential applications in wearable sensing, soft robotics, and electrophysiological recording.

Fig. 1. Design of ultrasoft and conductive BBE composites.

a Different applications require soft and/or stretchable materials with different levels of Young’s modulus (E). In comparison, conventional electronic materials (silicon and metals) applied on rigid electronics (e.g., the Utah array) have a much higher E than intrinsically stretchable materials. b The approximate Young’s modulus range of different biological tissues and commonly used intrinsically stretchable materials. Hydrogels and BBEs are two main materials that can match the softness of tissues. c Chemical structures of PDMS crosslinkers DMS-R22, monomers MCR-M11, and crosslinked BBEs. d A schematic showing the nanometer-scale composition of conductive PDMS BBE containing bottlebrushes and SWCNTs (dimensions are not to scale). e An integrated ultrasoft electronic made of the conductive SWCNT/BBE and non-conductive pure PDMS BBE.

Fig. 2. Characterizations of pure PDMS BBEs.

a Cycling tests of PDMS BBEs with different crosslinking ratios (molar ratio of MM:CL = 600:1, 900:1, and 1200:1) at the strain of 400%. The dash line represents the fitting curve obtained from the stress-strain model for unentangled polymer networks. The inset shows photographs of BBE under stretching. b The Young’s modulus of PDMS BBEs with different crosslinking ratios. c The comparison of Young’s modulus of PDMS BBEs obtained from the cyclic tests under different strain rates (0.035 s⁻¹, 0.070 s⁻¹, and 0.14 s⁻¹). d Long term cycling tests (1000 cycles at the strain of 400%) of the PDMS BBE with the crosslinking ratio (MM:CL) of 1200:1. e The Ashby-style plot of elastic range and Young’s modulus of different elastomers and PDMS BBE in this work. f The adhesive shear strength between the PDMS BBE (with the crosslinking ratio (MM:CL) of 1200:1) and different substrates, including PDMS Sylgard 184, Ecoflex 00-10, glass slide, copper sheet, and porcine skin. g Photographs of the adhesion test between the PDMS BBE and different substrates. The scale bar is 5 mm. Error bars denote the standard deviation of the measurements.

Fig. 3. Characterization of the conductive SWCNT/BBE.

a Cycling tests of PDMS BBEs with 0.2 wt%, 0.4 wt%, and 0.6 wt% SWCNT at the strain of 50%. 20 cycles were conducted, with the first half cycle represented by the dash line. b The Young’s modulus and c conductivity of PDMS BBEs with 0.2 wt%, 0.4 wt%, and 0.6 wt% SWCNT. d The Ashby-style plot of the conductivity and Young’s modulus of different conductive elastomers with metallic nano materials, carbon-based materials, conducting polymers, liquid metal, or MXene as conductive components. Hydrogels are not included as they contain water and are not solvent-free. Data points are from references labeled in Supplementary Fig. 12 and Supplementary Table 4. e Normalized change in resistance as a function of tensile strain. f Cyclic durability (2500 cycles) of the normalized change in resistance for BBE with 0.4 wt% CNT under cyclic loading to 100% strain. The inset shows a representative resistance change within the cyclic measurements. g Environment stability of the normalized change in resistance for SWCNT/BBE stored in HCl, H₂O₂ solutions or exposed to air. h The adhesive shear strength between SWCNT/BBE and different substrates. Error bars denote the standard deviation of the measurements.

Fig. 4. Laser-cut strain sensors and 3D-printed touch pad based on ultrasoft BBEs.

a The schematic of the fabrication process of laser-cut strain sensors. b Photograph of the laser-cut strain sensor adhered on the fingertip of a human hand. The strain sensor can easily attach on human body due to its self-adhesive property. c Overlaid photographs of a pneumatic actuator (inflated at different pressure levels) without sensor, with a SWCNT/BBE strain sensor, and with a SWCNT/Ecoflex strain sensor. The inflation pressure levels were kept the same for the three groups. The scale bar is 1 cm. d Bending angles of the pneumatic actuators under 12 kPa, 22 kPa, and 33 kPa of inflated pressure in c: without sensor (gray), with a SWCNT/BBE strain sensor (blue), and with a SWCNT/Ecoflex strain sensor (pink). e The schematic and photograph of a hornworm attached with the SWCNT/BBE sensor. f The crawling speed and g normalized length change of the hornworm with and without the SWCNT/BBE sensor. h The resistance change of the SWCNT/BBE sensor with the crawling of the hornworm. The insets show photographs of the hornworm at the two different states (left side photo: the hornworm body was stretched; right side photo: the hornworm kept still). i Expanded schematic of the printed touch pad. j The sensory response for placing a petal of the orchid on the touch pad. The scale bar is 5 mm. k Human–machine interaction through touching the robotic e-skin on the robot hand. Error bars denote the standard deviation of the measurements.

Fig. 5. 3D-printed SWCNT/BBE electrodes for ECG recording.

a Schematic of the ECG measurement. RA, LA, and RL represent the locations of right arm electrodes, left arm electrodes, and right leg electrodes, respectively. The three ECG electrodes were 3D printed with 0.6 wt% SWCNT/BBE. Dimensions are not to scale. b Bode plot of skin impedance measurements using SWCNT/BBE, copper sheet, and hydrogels as electrodes on the skin. The insets show a schematic of the measuring principle, and a photograph of 3D printed SWCNT/BBE electrodes. The scale bar is 2 cm. c Experimental data of ECG signals measured by commercial gel electrodes and SWCNT/BBE electrodes.