Robust skin-integrated conductive biogel for high-fidelity detection under mechanical stress

Abstract

Soft conductive gels are essential for epidermal electronics but often face challenges when interfacing with uneven surfaces or areas with extensive hair, especially under mechanical stress. In this study, we employed the concept of liquid-to-solid transformation to enhance integration at biointerfaces and designed an in-situ biogel capable of rapidly transitioning between liquid and solid states within 3 min via a temperature switch. The biogel features a semi-interpenetrating polymer network design and dual conduction pathways, resulting in high tensile strength (~1-3 MPa), a skin-compatible modulus (~0.3-1.1 MPa), strong skin adhesive strength (~1 MPa), and superior signal-to-noise ratio (SNR, ~30-40 dB). The biogel demonstrates significant performance in mechanically demanding environments, showing potential for accurately capturing outdoor exercise data, monitoring muscle recovery from sports-induced fatigue, and in vivo monitoring of cardiac physiological signals. The liquid-to-solid transformation concept, coupled with the design strategy for highly integrated and stable soft conductive materials, provides a basis for advancing conductive interface designs for high-fidelity signal acquisition.

Fig. 1. Design and characterization of the in situ biogel.

a Chemical composition of the in situ biogel. b Liquid-to-solid transformation concept and the application of in situ biogel. c Schematic diagram and optical images of the in situ rapid gelation process of the biogel. d Deformation capability of the in situ biogel on the skin surface. e Adhesion ability of the in situ biogel to the skin. f Optical image of a small-scale in situ biogel capable of bearing a significant weight, demonstrating its strength and toughness. g Comparison with recent high-performance soft gels and elastomers, evaluating six main aspects of performance relevant to soft materials applications. Detailed data is available in Supplementary Table 1.

Fig. 2. Rheology and mechanical performances of the in situ biogel.

a Rheology test of the eight types of in situ biogels with varying amounts of DES (0.10, 0.19, 0.38, 0.56, 0.74, 0.93, 1.13, and 1.32). b Bond energy calculation of ChCl with the hydroxyl group of glycerol and the carbonyl and hydroxyl groups of gelatin via DFT (where M1 and M2 refer to two sections of gelatin). c Strength and toughness analysis of the three types of in situ biogels. (n = 5 biologically independent samples). d Tensile modulus assessment of the three types of in situ biogels and skin. (n = 5 biologically independent samples). e Adhesive strength evaluation of the in situ biogel with porcine skin, determined through a shear test. (n = 3 biologically independent samples) f Adhesive strength measurement of the in situ biogel with electrodes. (n = 3 biologically independent samples) g Strength and adhesion comparison between in situ biogel and high-performance soft conductive materials, with detailed data available in Supplementary Table 2. h Adhesive strength of pre-formed and in situ biogels on porcine skin and electrode surfaces. (n = 3 biologically independent samples) i Adhesion comparison between in situ biogel and commercial biogel on the skin. j Optical images captured during the shear tests of the biogel with porcine skin. k The adhesion mechanism of the in situ biogel with the skin. Data are presented as mean ± s.d. unless otherwise specified.

Fig. 3. Electronical properties of the in situ biogel.

a EIS curve of three kinds of in situ biogels with different amounts of DES. b Nyquist plot derived from the EIS measurements. c Ionic and electrical resistances obtained from the EIS analysis. d Schematic diagram of the conductive formation mechanism of the in situ biogel. e Comparison between generated voltage and measured voltage of the in situ biogel and in situ biogel without PEDOT:PSS across frequencies ranging from 0.1 Hz to 1000 Hz. The heat map visualizes discrepancies between the generated and measured signals. f Root mean square error (RMSE) analysis for each waveform, providing quantification of differences between the generated and measured signals.

Fig. 4. Performances of the in situ biogel for ECG and sEMG signal detection.

a ECG detection using commercial biogel: (i) Recorded ECG Signals. (ii) Heatmap depicting mean and standard deviation, reflecting signal stability. (iii) Moving average analysis for signal stability assessment. b ECG detection after 24 h of commercial biogel application. c ECG detection using in situ biogel. d ECG detection after 48 h of in situ biogel application. e Optical images comparing ECG tests with commercial and in situ biogel. f SNR of ECG signals obtained using commercial biogel (0 h, 24 h) and in situ biogel (0 h, 48 h). g sEMG signals from the forearm during weightlifting, captured with both commercial and in situ biogels under two conditions: normal application and moistened by NaCl solution. (i) Noise magnification assessment. h Spectrograms recorded by commercial biogel and in situ biogel under different conditions. i SNR of the sEMG signals. j sEMG signals and spectrograms of the forearm while lifting various weights. k sEMG signal acquisition with a commercial gel pad during vigorous shaking, along with optical images of the pad’s condition after shaking. l sEMG signal acquisition with in situ biogel during vigorous shaking. m Changes in noise and SNR before and after vigorous shaking. (n = 5 biologically independent samples) Data are presented as mean ± s.d. unless otherwise specified.

Fig. 5. Outdoor exercise monitoring with in situ biogel.

a Wireless circuit design for sEMG signal acquisition. b sEMG signals of thigh and calf during walking. c Comparison of thigh and calf sEMG signals and spectrograms over time. d Linear regression analysis of the mean frequency of sEMG signals during walking. This decline of the mean frequencies indicates muscle fatigue. e sEMG signals of thigh and calf during the running. f sEMG signals of thigh and calf during squats. g sEMG signals of the calf during stretching.

Fig. 6. Muscle recovery monitoring post-sport fatigue with in situ biogel.

a Procedure for recovery monitoring using in situ biogel. Comparison of initial sEMG signals with those after fatigue. b, Initial sEMG signals and post-exercise signals for left and right calves. c Recovery signals at various time intervals: 10, 30, 60, 90, 120, and 180 mins, 24 h, and 48 h. d IMNF changes at various times, including before and after 45 min of cross-country running, as well as during post-fatigue recovery. (n = 3 biologically independent samples) e Recovery status from fatigue. Red: left calf, Blue: right calf. Data are presented as mean ± s.d. unless otherwise specified.

Fig. 7. Biocompatibility assessment of in situ biogel.

a Hemolytic test to evaluate blood compatibility using optical images and quantitative analysis of hemolysis ratios (n = 4 biologically independent samples, one-sided). b Live/dead dye assays. c Cell proliferation study assessing biocompatibility by culturing RS1 cells with the in situ biogel for 48 h and 72 h (n = 4 biologically independent samples, one-sided). d Histological analysis of skin and subcutaneous tissue after two weeks of applying biogel on the skin, stained with H&E. e Histological analysis of skin and subcutaneous tissue after two weeks of in situ biogel implantation, stained with H&E. f Histological analyses of rat organs (heart, liver, spleen, lung, and kidney) with and without in situ biogel implantation. Data are presented as mean ± s.d. unless otherwise specified. Differences between the two groups were analyzed using an unpaired Student’s t test. * denotes P  <  0.05, ** denotes P  <  0.01, and *** denotes P  <  0.001.

Fig. 8. Cardiac physiological measurements using the biogel patch.

a The biogel patch for detecting heart signals in a rat. b Schematic representation of four cardiac sensor channels on a heart model. c The cardiac patch adhered to the surface of a rat’s heart. d Photograph showing the biogel patch adhered to the cardiac surface for signal detection. e Image highlighting the ligation point in a myocardial infarction (MI) rat model. f, g Real-time monitoring of cardiac signals from healthy (f) and myocardial infarction (g) rat hearts. h Real-time heart rate and progression of cardiac signals in an agonal MI rat. i Cardiac signals are recorded at different stages.