*2.5. Characterization*

The transmission electron microscope (TEM) was carried out by a transmission electron microscope (JEM-1400, Tokyo, Japan) operating at 120 kV. The concentrations of CNTs, TOCNF-CNT, and TOCNF-CNT@PANI water suspensions were 0.05–0.1 wt%. The absorption spectra of CNTs, TOCNF-CNT, and TOCNF-CNT@PANI suspensions were measured at room temperature using an

ultraviolet-visible (UV-vis) spectrophotometer (Tu-1810, Purkinje Co., Beijing, China) at a wavelength of 200–1000 nm. Spectra were collected using water as a reference at a scan speed of 0.5 nm s<sup>−</sup>1. Infrared spectra were performed by Fourier transform infrared (FTIR) spectroscopy (Nicolet iS50, Thermo Fisher Scientific Inc., Madison, WI, USA) with attenuated total reflection (ATR) mode. The wavelength range was from 4000 to 500 cm−<sup>1</sup> at a resolution of 4 cm−1. X-ray diffraction (XRD) spectra were obtained by an X-ray diffractometer (Ultima IV, Rigaku, Japan) at 40 kV and 30 mA, and the angle range was 2θ = 5~40◦ at a scanning rate of 5◦ min<sup>−</sup>1. The microstructure of composite hydrogel was observed by a JSM-7600F scanning electron microscope (Nippon electronics Co., Ltd., Tokyo, Japan) at a voltage of 15 kV.

Uniaxial compression measurements were carried out through a universal mechanical testing machine (CMT4304, Shenzhen, China) on a cylindrical sample (diameter of 20 mm; the height of 5 mm) at a compression speed of 8 mm min−<sup>1</sup> at room temperature in air. The values of stress (σ) and strain (ε) were calculated from the force and deformation of the original size of samples. The compressive elastic modulus (*E*e) was calculated from the rake angle ratio of the linear part (ε < 20%) of the σ-ε curve. The specific stress (σs) was obtained by dividing the density (ρ). The energy absorption (*E*a) was the integral area of the part under the σ-ε curve.

Tensile stress-strain measurements were performed using a universal mechanical testing machine (CMT4304, Shenzhen, China) at a pulling rate of 20 mm min−1. The hydrogel samples of initial and healed (20 s) in a size of 30 <sup>×</sup> <sup>15</sup> <sup>×</sup> 5 mm<sup>3</sup> were nipped to the tensile machine during the testing process. All the tensile measurements were repeated for three times. The healing efficiencies (ηF) were calculated from the break stress (*F*healed) of the healed hydrogel divided that of the initial one (*F*original). The healing efficiencies (ηK) were defined from that the break strain (*K*healed) of the healed hydrogel sample divided that of the initial one (*K*original). (η<sup>F</sup> = *F*healed/*F*original × 100%, η<sup>K</sup> = *K*healed/*K*original × 100%).

The dynamic rheological properties, including dynamic frequency scanning, dynamic strain scanning, and continuous step strain, were tested by a Rheometer (HAAKE600, Waltham, MA, USA) with a plate diameter of 40 mm and a gap of 500 μm. The dynamic strain range was 0.01 to 100% with angular frequency (ω) of 1 Hz. The linear viscoelastic region (LVR) was decided by storage modulus (*G* ), and the *G* was independent of the strain in the LVR. In the following measurement of each sample, 1% strain (γ) was selected to maintain the dynamic oscillatory deformation within the LVR. In dynamic frequency scanning measurement, the relationship between the shear storage modulus (*G* ), loss modulus (*G*), and angular frequency (ω) were recorded at ω = 0.1–100 rad s<sup>−</sup>1, γ = 1%, and 25 ◦C. The complex modulus (*G*\*) was calculated by Equation (1).

$$G = \sqrt{G'^2 + G'^2} \tag{1}$$

Continuous step strain tests were conducted to study the recovery property of the hydrogels under the applied shear stress. A procedure to the program was as follows: 1% (800 s) →80% (800 s) → 1% (800 s) → 80% (800 s) → 1% (800 s), the *G* and *G* versus time were measured at ω = 1 Hz and 25 ◦C.

Conductivity tests of hydrogel electrodes. A square-shaped hydrogel with a size of 1 × 1 × 10 cm<sup>3</sup> was sandwiched by two pieces of platinum electrodes. The resistance (*R*) values of the hydrogel were decided by current-voltage (*I*–*V*) measurement using an electrochemical workstation (CHI 760E, Shanghai, China). The conductivity (σ) was achieved from Equation (2):

$$
\sigma = \frac{1 \times d}{R \times S} \tag{2}
$$

where σ was the conductivity (S m−1), *R* was the resistance (Ω), *d* was the length (m), and *S* was the cross-sectional area (m2) of the sample, respectively.

Electrochemical measurements of the hydrogel-based electrodes were performed on a three-electrode system using an electrochemical workstation (CHI 760E, Shanghai, China). The working electrode, the counter electrode, the reference electrode, and electrolyte were the

TOCNF-CNT@PANI/PVA-2 hydrogel (1.5 g), platinum plate electrode, mercury/mercury oxide (Hg/HgO) electrode, and 6 M KOH aqueous solution, respectively. The cyclic voltammetry (CV) test was carried out at scan rates of 40 mV s−<sup>1</sup> from <sup>−</sup>0.2–0.8 V, the galvanostatic charge-discharge (G-CD) test was performed over the voltage range of <sup>−</sup>0.2–0.8 V at a current density of 0.4 A g−1, and electrochemical impedance spectra (EIS) test was measured over the frequency range from 0.01 Hz to 100 kHz at open circuit potential (alternating current perturbation voltage was 5 mV). The specific capacitance (*C*s) values were calculated from the G-CD curves using Equation (3):

$$\mathbf{C}\_{s} = \frac{I\Delta t}{m\Delta V} \tag{3}$$

where *C*<sup>s</sup> represented the specific capacitance (F g−1), *I* represented the discharge current (A), Δ*t* represented the time of discharge (s), Δ*V* represented the voltage of discharge (V), and *m* represented the mass of active materials (g).
