A synergistically designed strain-insensitive conductive hydrogel with humidity-adaptivity supporting sustained functional durability

Abstract

While employing hydrogels as conductive interconnects in flexible electronics, two intrinsic challenges require comprehensive solutions: performance degradation resulting from water evaporation and functional instability caused by strain-induced resistance increase. Thus, we integrate hygroscopic lithium bromide (LiBr) directly into hydrogel polymerization and exploit its multifaceted characteristics to develop a conductive hydrogel with humidity-adaptive water retention and strain-insensitivity. The synthesis is guided by the idea of “less is more”, characteristics of acrylamide monomers (Aam), the silane coupling agent, and LiBr are synergistically leveraged to architect a hierarchical network comprising a primary backbone and dual-dynamic crosslinking. The prepared hydrogel possesses concomitant properties of ionic conductivity, softness, stretchability, and anti-freezing. Moreover, ionic and electronic hybrid conductivity (∼0.21 S cm⁻¹), an ultralow gauge factor (∼0.29) within a work strain range of 150% and electrical hysteresis (∼0.19%) are imparted via the incorporation of conductive additives. Cyclic tensile strain (10 000 cycles), prolonged exposure under fluctuated humidity conditions over 6 months (20%–50% fluctuated relative humidity), and low-temperature storage (−56 °C) were conducted to verify the sustained functional maintenance. The conductive hydrogel was practically qualified as a conductive interconnect for power supply and signal transmission. The results unfold a promising prospect of the conductive hydrogel with long-term reliability as a conductive interconnect.

---

New concepts

This paper proposes a synergistically designed conductive hydrogel following the principle of “less is more” in network construction. The physical and chemical properties of three components (acrylamide, the silane coupling agent, and lithium bromide) are fully exploited to prepare an ionic conductive hydrogel with multiple properties of softness, stretchability, low mechanical hysteresis, long-term water retention, and anti-freezing. Subsequently, by introducing conductive additives (PEDOT:PSS and MWCNTs), the hydrogel is endowed with excellent strain-insensitive conductivity and ultra-low electrical hysteresis. Through systematic investigations into mechanical and electrical performance under extreme conditions, sustained functional durability is verified. Compared with existing research, this synergistical design harnesses intrinsic material attributes and synergistic interplay between them and maximizes diverse hydrogel properties with minimal material. Among these, LiBr specifically plays critical roles in hydrogel network formation, humidity adaptivity, and conductivity. Ultimately, the hydrogel design proposed in this study reduces fabrication complexity while overcoming persistent functional decay issues in practical applications caused by environmental variations, enabling stable conductive interconnects and signal transmission in flexible electronics scenarios. Simultaneously, it underscores the significance and potential of rational material characteristics exploitation.

Introduction

Flexible electronics, typified by foldable mobile phones and wearable health monitoring devices, pose the challenge of developing flexible conductive interconnects. Hydrogels with adjustable plasticity have been broadly investigated and applied as a promising candidate.^1–4 However, the application of hydrogels has been consistently constrained to their environmental resilience.^5,6 Hydrogels are inherently constrained in terms of long-term durability due to unavoidable performance degradation derived from water evaporation; such limitation hinders their practical implementation dramatically.⁷ Suppression of water evaporation generally includes use of water-impermeable coatings and water substitution via immersion in hygroscopic solutions (e.g., CaCl, LiCl or LiBr).^8–11 Yet, it is imperative to consider the compatibility at the interface between the hydrogel and the coating, especially for circumstances requiring multiple stretching operations. Although chemical bonds at the interface are prevalently generated for interfacial adhesion, the extra processing adds complexity to synthesis.^12,13 Regarding the method of hygroscopic solution substitution, the effect caused by the concentration gradient difference on the structural stability of the hydrogel network is inevitable, and the substituted solution fundamentally manifests as a discrete system from the original hydrogel network, which is devoid of internal coupling mechanisms for compatibility.

Traditional hydrogels with a high gauge factor (GF) are critical for strain sensing applications. GF values are distributed across a wide range from fractions of unity to several thousands, depending on the hydrogel composition, fabrication method, and morphological structures.¹⁴ Paradoxically, high conductivity and low GF values are essential to maintain stable electrical connection. In parallel, low electrical hysteresis is equally crucial for signal consistency under deformation.¹⁵ Both intrinsically conductive polymers and those externally incorporated with conductive additives invariably incur substantial resistance increases during deformation due to the elongation of polymer chains and the separation of conductive additives, the accompanying electrical hysteresis is also unacceptable.

In this work, with an aim to address the long-term durability issue caused by water evaporation and obtain a conductive hydrogel with a low GF and electrical hysteresis, we synthesize a conductive hydrogel based on hygroscopic LiBr aqueous solution, which provides both humidity-adaptive water dynamics under the conditions of fluctuating relative humidity (RH) for sustained water retention and a physically collapsed hydrogel network conducive to strain-insensitive conductivity. The GF value is as low as 0.29, which falls below the commonly reported minimum range for conductive hydrogels, and is comparable to that of the reported strain-insensitive conductive hydrogels.¹⁶ A “less is more” concept is adopted for synergistic hydrogel engineering, where functionalities of Aam, the silane coupling agent 3-(trimethoxysilyl) propyl methacrylate (TMSPMA), and especially LiBr, are “squeezed” to play multi-roles for effective network construction. The prepared hydrogel is thereby originally equipped with properties of stretchability, humidity-adaptivity, ionic conductivity and anti-freezing. In summary, the synergistically engineered conductive hydrogel proposed in this work enables long-term functional durability and a stable conductive interconnect for applications in flexible electronics.

Fundamental principles and properties

The industrial standard operating concentration of 55 wt% is selected as the concentration of LiBr aqueous solution, as it achieves the optimal balance between ensuring high hygroscopicity and avoiding crystallization. In Fig. 1a, Aam and TMSPMA are directly dissolved in the solvent of 55 wt% LiBr aqueous solution, assembling P(Aam-co-TMSPMA) chains as primary network units. Subsequently, hydrolysed silanol groups from TMSPMA deprotonate into silanolate ions (Si–O⁻) under alkaline conditions and then condense to siloxane Si–O–Si covalent networks.¹⁷ Dynamic ionic interactions are concurrently generated between unreacted Si–O⁻ and Li⁺. Additionally, polar carbonyl (C=O) groups originating from the amide (–CONH₂) groups of Aam form ion–dipole interactions with Li⁺, offering an extra supplementary dynamic crosslinking.^18–20 The hierarchal network of the P(Aam-co-TMSPMA)/Li⁺ hydrogel (PAT/Li⁺ HG) achieves both rigidities dominated by Si–O–Si covalent bonds and stretchability via ionic and ion–dipole interactions. The dynamic interactions exhibit reversible rupture-reconfiguration under tensile strain, enabling stretchability via energy dissipation during polymer chain slippage.²¹ Simultaneously, charge transport of Li⁺ generates pathways for ionic conductivity.²² Meanwhile, LiBr leads to strong hydration, which serves as the core mechanism for humidity-adaptivity and strain-insensitivity. Specifically, the hydration not only guarantees a slow water evaporation rate but also dynamically maintains water equilibrium across various RH values. As demonstrated in Fig. 1b, evaporation dominates the water dynamics once the hydrogel is exposed to low RH conditions. This process lasts until the LiBr concentration climbs to a critical point and subsequently accelerates water absorption. Conversely, the water dynamics transition to absorption dominance until water equilibrium re-establishes as RH rises to a relatively high level. The dynamic process ensures sustained water retention without additional treatment, inherently resolving the core durability challenge. Besides, the hydration of LiBr also gathers water molecules leading to a collapsed internal network during polymerization, benefiting low GF and electrical hysteresis. In addition, the resistance increment induced by the separation of conductive additives is mitigated until the collapsed network fully extends under stretching, contributing to the strain-insensitivity. Hitherto, a conductive hydrogel network achieving maximized realization of multiple hydrogel features with minimal components is established. Low-temperature phase stability and hydration of LiBr provide anti-freezing capability. The electrical conductivity and stretchability of the prepared conductive hydrogel were preserved for LED illumination without fracture or delamination under −56 °C conditions (Fig. 1b).

Fig. 1 (a) Synergistic design of the hydrogel network and mechanical performance. (b) Illustration of the humidity-adaptivity mechanism and anti-freezing. (c) Application of MP-PAT/Li⁺ CH in conductive interconnects and sensing. (d) Weight changes at various RH values referring to humidity-adaptivity. (e) Nyquist plots of PAT/Li⁺ HG and MP-PAT/Li⁺ CH at original, stretched and recovered states. (f) Performance comparison of MP-PAT/Li⁺ CH with other hydrogels.^{15,21,23–36}

Successively, multi-walled carbon nanotubes (MWCNTs) and poly (3,4-ethylenedioxythiophene): polystyrene sulfonate (PEDOT:PSS) are incorporated into the polymerization of PAT/Li⁺ HG as conductive additives. The MWCNT/PEDOT:PSS reinforced PAT/Li⁺ HG, denoted as MP-PAT/Li⁺ CH, enhances the conductivity by establishing electronic conductive pathways. The interlayer shear relaxation dynamics of MWCNTs allow sliding-mediated reorientation rather than fracture, while PEDOT:PSS penetrates MWCNT interstices to form a strain-tolerant percolation network.³⁷ The inherent ductility of PEDOT:PSS compensates for localized resistive surges brought about by MWCNT detachment through the π–π interaction between MWCNTs and PEDOT:PSS; therefore continuous charge transport pathways are guaranteed under mechanical deformation.³⁸ These mechanisms complementarily reinforce the strain-insensitive capability of MP-PAT/Li⁺ CH beyond the collapsed structure. As shown in Fig. 1a, PAT/Li⁺ HG and MP-PAT/Li⁺ CH stretched to 400% reveal the contribution of dual-dynamic crosslinking. Also, knotted, twisted and pressed MP-PAT/LI⁺ CH demonstrate mechanical toughness. The exceptional sustained durability and strain-insensitive conductivity of MP-PAT/Li⁺ CH significantly improve functional longevity in real-world scenarios, such as serving as stretchable and reliable conductive interconnects for power supply and signal transmission in flexible electronics. Fig. 1c illustrates the potential of the proposed hydrogel in electrical interconnects and sensing. An electroluminescent (EL) patch utilizing MP-PAT/Li⁺ CH as the conductive interconnect was fabricated and illuminated; it remained functional under bending (Fig. S1 and Movie S1). Notably, near-linear electromechanical response and low electrical hysteresis were verified by adopting an MP-PAT/Li⁺ CH as a human joint motion sensor. When integrated on a finger joint, the MP-PAT/Li⁺ CH shows consistent resistance responses to various bending angles (Fig. 1c).

We also measured the weight changes of the MP-PAT/Li⁺ CH under different humidity conditions to verify the humidity-adaptive water dynamics (Fig. 1d). Specifically, the MP-PAT/Li⁺ CH was pre-equilibrated at 30% RH for over 7 days and then exposed to 15% RH, 30% RH, 60% RH and finally 30% RH conditions for 12-hour each, sequentially. Experimental observations revealed that the weight fluctuations occurred in desiccating (15% RH) and humid (60% RH) environments, and the weight recovered to the original equilibrium state upon returning to moderate humidity (30% RH), mediated by LiBr concentration-regulated water dynamics. The recorded weight changes under extreme humidity conditions remained remarkably low at only 1.64% (15% RH) and 1.65% (60% RH), respectively.

Electrochemical impedance spectroscopy (EIS) measurements were conducted to discern the difference in charge transport characteristics between PAT/Li⁺ HG and MP-PAT/Li⁺ CH at three distinct states: original, stretched (150% strain), and recovered. The corresponding Nyquist plots are presented in Fig. 1e. An increased bulk resistance is observed under stretching, accompanied by concurrent elevation of interfacial resistance and enhanced polarization. In contrast to the 268% increase in bulk resistance of PAT/Li⁺ HG, the bulk resistance of MP-PAT/Li⁺ CH rises by 110% in the stretched state. The lower increase indirectly reflects the contribution of MWCNT and PEDOT:PSS to the low GF. In the recovered state, reduction in bulk resistances relative to their initial values reached 80% for PAT/Li⁺ HG and 94% for MP-PAT/Li⁺ CH. These substantial decreases prove that conductive channels are optimized during the stretching process. The MP-PAT/Li⁺ CH manifests higher initial bulk resistance, attributed to blockage effects caused by the conductive additives. However, the bulk resistance declines to a lower value in the recovered state due to structural reorganization and emergent electronic pathways. Meanwhile, interfacial charge accumulation and polarization intensity in the MP-PAT/Li⁺ CH are mitigated, confirming the establishment of a hybrid ionic-electronic conductive network.

The performance of the MP-PAT/Li⁺ CH in terms of GF, mechanical hysteresis (MH), electrical hysteresis (EH) and water loss (WL), compared to other recently presented hydrogels, is illustrated in Fig. 1f.^{15,21,23–36} The weight-based water loss is quantitively defined as the average water loss ratio per hour, normalized to account for the varied aging durations reported in different studies. Detailed comparison with other hydrogels is provided in Table S1. Comparison of the performance clearly manifests that the conductive hydrogel developed in this work achieves high electrical conductivity, low mechanical hysteresis, exceptionally minimal gauge factor (GF) and electrical hysteresis through straightforward synergistic engineering. Essentially, humidity-adaptivity significantly promotes sustained durability. The combined attributes position this hydrogel as a promising candidate for reliable flexible conductive interconnects requiring sustained functional stability.

Experimental section

Materials

Lithium bromide aqueous solution (LiBr, 55 wt% in H₂O) (Aladdin, L108933), acrylamide (Aam) (Aladdin, A108470), 3-(trimethoxysilyl) propyl methacrylate (TMSPMA) (Aladdin, S111153), ammonium persulfate (APS) (Aladdin, A112447), sodium pyrosulfite (SPS) (Macklin, S818096), NaOH (Aladdin, S683695), poly(3,4-ethylenedioxythiophene):polystyrene sulfonate solid particles (PEDOT:PSS) (Macklin, P917189), multi-walled carbon nanotubes (MWCNTs) (XFNANO, 100234), and phenylbis(2,4,6-trimethylbenzoyl)phosphine oxide (i-819) (Aladdin, P138333) were used.

Preparation and characterization methods

Fig. 2a demonstrates the diversified synthesis processes of the proposed hydrogels. The transparent and ionic conductive PAT/Li⁺ HG is produced via straightforward fabrication: 4 μL of TMSPMA are added into 1 mL of 55 wt% LiBr aqueous solution with 2 mol L⁻¹ Aam, and then, 15 μL of 10% (w/w) NaOH aqueous solution are added to adjust pH, followed by 60 s vortex stirring. Afterward, 8 μL of 10% (w/v) APS aqueous solution as an initiator and 3 μL of 10% (w/v) SPS aqueous solution as an accelerator are subsequently added, followed by 5 s vortex mixing. The PAT/Li⁺ HG is obtained through pouring the solution into 40 mm × 20 mm × 1 mm Teflon mold for 1-hour ambient curing. The synthesis of the MP-PAT/Li⁺ CH necessitates supplementary processing during precursor preparation: MWCNT (2mg mL⁻¹) and PEDOT:PSS solid-state particulates (1.5% (w/v)) are dispersed via intensive mechanical stirring for 72 hours at 1500 rpm. Subsequent processes are performed identically as previously described, with equal amounts of TMSPMA, NaOH, APS and SPS added under the same conditions to complete gelation. Furthermore, 10 μL of photo-initiator of i-819 saturated in ethanol is utilized to substitute APS and SPS, facilitating compatibility with direct ink writing (DIW) (Fig. S2). The straightforward fabrication workflow for various hydrogel forms not only reduces complexity but also ensures adaptability across diverse application scenarios.

Fig. 2 (a) Synthesis processes of PAT/Li⁺ HG, MP-PAT/Li⁺ CH and direct ink writing (DIW). (b) Schematic mechanisms of the MP-PAT/Li⁺ CH. (c) The swelling ratio of mass of the MP-PAT/Li⁺ CH immersed in LiBr aqueous solution with different concentrations. (d) The DSC curve of the MP-PAT/Li⁺ CH. (e) Demonstration of the stretchability of PAT HG. (f)–(i) Examination of the MP-PAT/Li⁺ CH in terms of viscosity, transparency, stretchability, conductivity and DIW.

The total mechanism schematic of the MP-PAT/Li⁺ CH is illustrated in Fig. 2b. LiBr contributes to the primary properties of the MP-PAT/Li⁺ CH playing three roles: (1) conferring humidity-adaptive water dynamics via hygroscopic nature, (2) introducing hydration based structural collapse for low GF and electrical hysteresis, and (3) participating in dynamic crosslinking interactions with Aam and the silane coupling agent. Electrical conductivity and strain-insensitivity are substantially improved by means of incorporating and developing characteristics of MWCNT and PEDOT: PSS.

The swelling ratio was measured through immersing MP-PAT/Li⁺ HG in LiBr aqueous solutions at different concentrations to simulate RH gradients (Fig. 2c). Differential scanning calorimetry (DSC) thermograms at 10 °C min⁻¹ is illustrated in Fig. 2d; it reveals relatively consistent heat flow across the temperature range from −70 °C to 0 °C, with no detectable melting or crystallization peaks. This absence of phase transitions demonstrates the anti-freezing performance at a practical temperature of 56 °C. Fig. 2e illustrates the poor stretchability of the water-based P(Aam-co-TMSPMA) hydrogel (PAT HG), which corroborates the role of LiBr. The properties of the proposed hydrogel in terms of viscosity, transparency, mechanical and electrical performance, as well as DIW compatibility, are shown in Fig. 2f–i.

Results and discussion

Characterization

The attenuated total reflectance infrared (ATR-IR) spectra of PAT HG, PAT/Li⁺ HG and MP-PAT/Li⁺ CH are illustrated in Fig. 3a. The transmittance peaks at 817 cm⁻¹ and 1132 cm⁻¹ correspond to the symmetric and asymmetric stretching vibrations of the siloxane (Si–O–Si) group, and the peaks at 1433 cm⁻¹ and 1660 cm⁻¹ are attributed to –CH₂ vibration and C=O stretching vibration, which correspond to the polymerization of acrylamide.^39,40 The broad peak encompassing the range 3000–3700 cm⁻¹ indicates the stretching vibration of hydroxyl (–OH) groups.^41,42 The observed red shifts of Si–O–Si from 1132 cm⁻¹ to 1102 cm⁻¹ and C=O from 1660 cm⁻¹ to 1628 cm⁻¹ in Fig. 3b and c confirm the formation of ionic and ion–dipole interactions of Li⁺–O.^43–45 Moreover, apparent attenuation around 3200 cm⁻¹ demonstrates the consumption of –OH associated with Si–OH, while the residual peak around 3400 cm⁻¹ is primarily attributed to the absorbed water molecules within the polymeric matrix.⁴⁶

Fig. 3 (a) ATR-IR spectrum of water-based PAT HG, PAT/Li⁺ HG and MP-PAT/Li⁺ CH. (b) and (c) Shifts of Si–O–Si and C=O peaks caused by Li⁺–O interactions. (d) and (e) High-resolution XPS spectra of O 1s and Li 1s. (f) Raman spectrum of MP-PAT/Li⁺ CH. (g) and (h) SEM micrographs of MP-PAT/Li⁺ CH and the conductive hydrogel based on water. (i) XRD pattern of the MP-PAT/Li⁺ CH.

The high-resolution spectra of O 1s and Li 1s obtained via X-ray photoelectron spectroscopy (XPS) are presented in Fig. 3d and 2e, respectively. Deconvolution of the O 1s spectrum reveals two distinct components at 532 eV and 533.59 eV, assigned to the bridging oxygen in Si–O–Si and the terminal hydroxyl groups of Si–OH, respectively. A peak centered at 530.23 eV, accompanied by the Li 1s peak at 55.55 eV, is ascribed to the formation of Li⁺–O interactions involving both ionic and ion–dipole interactions.^47–49

The Raman spectrum illustrated in Fig. 3f contains main signature peaks at 1435 cm⁻¹ and 1509 cm⁻¹, representing the symmetrical and anti-symmetrical stretching vibrations of the aromatic C_α=C_β within PEDOT:PSS.⁵⁰ The D and G bands, representing structural defects and sp² hybridized carbon vibration, are observed at 1368 cm⁻¹ and 1593 cm⁻¹.⁵¹ The cross-section of internal structures at the fractured cross-section of MP-PAT/Li⁺ CH and the conductive hydrogel based on water without LiBr are visualized in Fig. 3g and h; an obvious entangled and collapsed internal structure induced by the LiBr hydration effect of the PAT/Li⁺ CH is observed. The X-ray diffraction (XRD) pattern is presented in Fig. 3i. The characteristic diffraction peak corresponding to the (002) planes of MWCNTs locates at 2θ = 25.5°, while a broad diffraction peak at 2θ of 28° indicates the existence of PEDOT:PSS and the amorphous structure of PAM.^52–54 Additionally, diffraction peaks at 2θ values of 38° and 53° are identified by LiBr hydration.⁵⁵

Mechanical properties

Retention of mechanical stability under multiple tensile strains and prolonged aging is critical for long-term performance. The mechanical properties of the MP-PAT/Li⁺ CH are analysed under two distinct protocols: cyclic tensile strain up to 10 000 cycles and 4-week aging. The hydrogel was exposed directly to ambient indoor conditions (∼24 °C and a fluctuant low RH of ∼20%), as opposed to constant temperature and relatively high humidity environments. Under continuous tensile loading, the MP-PAT/Li⁺ CH exhibits a maximum tensile strain exceeding 500% in its as-fabricated state (Fig. 4a). And under aging conditions, an obvious decline appeared during the first week, followed by a decelerated degradation rate over the subsequent 3 weeks, eventually stabilizing at the strain of approximately 200% (Fig. 4b). The variation in stretchability stems from several contributing factors: condensation during continuous aging exacerbates structural rigidification, and Li⁺ enrichment near crosslinking nodes leads to the formation of ionic clusters that further restrict chain mobility; such processes cease once the equilibrium state is established over time. To verify the attenuation mechanism of stretchability, the spectra of ATR-IR and high-resolution Li 1s of the newly prepared and 6-month aged MP-PAT/Li⁺ HG are recorded. The enhanced intensity at 817 cm⁻¹ representing Si–O–Si symmetric stretching vibrations of the aged MP-PAT/Li⁺ HG validates continuous condensation of Si–OH groups after gelation (Fig. S3). For Li⁺ enrichment, the shift toward lower binding energy and noticeable peak splitting of Li 1s indicate the change of coordination environment of Li⁺ (Fig. S4). Young's modulus and toughness, as functions of cycle numbers and aging time, are depicted in Fig. 4c. With increasing cycle number, the Young's modulus gradually decreases while toughness increases. Conversely, prolonged aging elevates Young's modulus but diminishes toughness. In the case of cyclic tensile strain, the stretching-triggered rupture of dynamic bonds leads to stress concentration redistribution via energy dissipation, thereby promoting microstructural toughness and reducing stiffness. For prolonged aging, the phenomenon is explained by the dynamic LiBr induced water absorption and evaporation. At the initial water evaporation-governed stage, interchain interactions strengthen with water evaporation; the resultant chain aggregation over time restricts segmental mobility and increases Young's modulus. Besides, suppressed viscoelastic relaxation resulting from reduced water content causes embrittlement. However, the decline terminated as the concentration of LiBr reaches the critical point and then the water absorption-dominated process alleviates mechanical performance degradation, thereby sustaining it at a relatively consistent level. It is noteworthy that the maximum stress is below 30 kPa, the softness is driven by the dual dynamic crosslinked network and collapsed structure, and the interfacial mechanical mismatch is minimized when the hydrogel is functionalized as a conductive interconnect. Successive strain-recovery curves with 50% strain step were measured and illustrated in Fig. 4d, within each pair of adjacent hysteresis loops, the latter loop maintains relatively consistent hysteresis characteristics across the strain range of its predecessor, indicating strain-invariant viscoelasticity and non-damaging network reconfiguration. Such phenomenon is regarded as evidence for the contribution of the collapsed network. Identical tests were conducted under cyclic tensile strain and aging conditions with hysteresis profiles presented in Fig. 4e and f. Hysteresis changing curves converge to the value around 7% along with strain escalation up to 400% in the cyclic tensile strain test, validating fatigue-resistant network reconfiguration and strain-adaptive energy dissipation. As for the aging condition, the tests were conducted within the strain of 200% in accordance with the reduced stretchability. The hysteresis stabilized around its original value with less than 1% fluctuation. As a primary constituent affecting the mechanical property, effects of TMSPMA content and LiBr concentration were also evaluated, indicating the optimized performance of the selected formula (Fig. S5). Tan δ curves of the original and 4-week aged MP-PAT/Li⁺ CH were measured via angular frequency scanning (0.1–100 rad s⁻¹ at a strain of 1%), and the evidential elastic response dominated mechanical characteristic is shown in Fig. 4g.

Fig. 4 Mechanical property analysis of the MP-PAT/Li⁺ CH. (a) and (b) Stress–strain curves under cyclic tensile strain and 4-week aging. (c) Variation of Young's modulus and toughness with cycle number and aging time. (d) Strain-recovery curves under successive stepwise stretching. (e) and (f) Strain-dependent mechanical hysteresis under cyclic tensile strain and 4-week aging. (g) Tan δ (G′′/G′) of the original and 4-week aged MP-PAT/Li⁺ CH. (h) Shear strain sweep of G′′ and G′ of the original and 4-week aged MP-PAT/Li⁺ CH. (i) Weight variation with aging time for 6 months with photographs of the samples.

The Tan δ curve of the aged hydrogel substantiates the results of tensile tests. The rheological behaviour of the MP-PAT/Li⁺ CH under shear strain sweep (0.01 ≤ γ ≤ 1 at 2 rad s⁻¹) reveals the viscoelastic response. Within the linear viscoelastic regime, the near-constant storage modulus (G′) and loss modulus (G′′) signify a well-defined elastic network (Fig. 4h).

Fig. 4i illustrates the weight change of the MP-PAT/Li⁺ CH over time during the water retention test. The MP-PAT/Li⁺ CH retained 93% of its initial weight after exposure of 6 months to dynamically varied environments. The inset presents comparative visualization of pristine hydrogels alongside those subjected to 6-month aging, demonstrating that the macroscopic morphology persisted after prolonged exposure. This observation substantiates the exceptional water retention capability of the hydrogel. To systematically evaluate the influence of LiBr concentration on the water retention ability, weight changes of the MP-PAT/Li⁺ CH prepared in LiBr aqueous solution with differed concentrations were measured at various RH levels (Fig. S6).

Electrical properties

Considering the strain attenuation demonstrated by mechanical tests, the working range is set within 150% strain. Fig. 5a and b show electrical hysteresis loops varying with tensile strain cycles and prolonged aging. Electrical hysteresis remains invariant and stabilizes to ∼0.19%, demonstrating stable and low hysteresis under operational conditions (Fig. 5c). The I–V curves of the MP-PAT/Li⁺ CH were achieved via four-point probe measurements (Fig. 5d, e and Fig. S7); the inset diagram illustrates the corresponding conductivity variations. Behaviour of ohmic conductivity is confirmed by the near-linear electromechanical response. After 4-week aging, the conductivity reached a minimum value of ∼0.21 S cm⁻¹. The observed reduction in resistance-strain variation and enhanced conductivity arise from the conductive network optimization, driven by cyclic strain induced spatial redistribution of conductive additives and water dynamics during aging. The dependency of conductivity on LiBr concentration is shown in the SI (Fig. S8). The response time and release time at 1.3% ΔR/R are ∼80.67 ms and ∼75.05 ms, respectively. (Fig. 5f). The GF curves as a function of strain indicate near-linear trend and low values, especially within 150% strain (∼0.29). After 10 000 cycles, the gauge factor declines to ∼0.18, demonstrating enhanced strain-insensitive behaviour over multiple operations (Fig. 5g). Additionally, the near-linear resistance variation within 150% strain further corroborates the role of the collapsed network. In the low-strain regime, the inherent network undergoes a transition from a collapsed to a fully extended state. However, once the polymer chains become fully extended, further stretching induces the separation of conductive additives, leading to a rapid increase in resistance. The electrical response is examined to be stable at different strain and stretching frequencies (Fig. 5h and i). Resistance changes of the MP-PAT/Li⁺ CH at various RH levels (Fig. S9) indicate that the hydrogel possesses good environmental stability. Eventually, continuous cyclic tensile strain was conducted for over 900 minutes, exhibiting well-maintained resistance less than 4% drop (Fig. 5j).

Fig. 5 Electrical characteristics of the MP-PAT/Li⁺ CH. (a) and (b) Resistance-strain curves in the loading–unloading process under cyclic tensile strain and long-term aging. (c) Changes in electrical hysteresis with cycle number and aging time. Stress–strain curves under cyclic tensile strain and long-term aging. (d) and (e) I–V characteristics and conductivity variation under cyclic tensile strain and long-term aging. (f) Response and release time. (g) Gauge factor varies with strain, cycle number and aging. (h) and (i) Resistance change with differed strain and strain frequency. (j) Long-term stability test of resistance change.

Durability and applications

Sustained functional durability across diverse conductive interconnection applications is enabled by mechanical and electrical stability supported by humidity-adaptive water retention. Performance under cyclic tensile strain and prolonged aging is illustrated in Fig. 6a. A light-emitting diode (LED) was electrically connected by the MP-PAT/Li⁺ CH. The illumination remains uninterrupted and was slightly attenuated beyond 200% tensile strain, even after 10 000 times of stretch and 4-week aging (Fig. 6a). When functioning as conductive interconnects, the capability for frequency-modulated signalling transmission in control systems is equally critical alongside power supply. Thus, robust signal retention is crucial for conductive hydrogels. A programmable RGB LED (WS2812B) employs a pulse-width-modulated (PWM) serial protocol for control signals following a non-return-to-zero (NRZ) scheme. The integrated driver in each LED automatically latches a 24 bit signal while propagating subsequent data to the next unit via an internal shift register, enabling scalable daisy-chaining. The control principle of WS2812B requires conductive interconnects with low bulk resistivity, high temporal fidelity, and minimal electrochemical polarization for high-frequency and time-sensitive control signal transmission. We fabricated an LED array (1 × 3) embedded in a flexible silicone substrate, adopting the MP-PAT/Li⁺ CH as an interconnect for 5 V DC power supply and control signal transmission (Fig. 6a). Synchronized green illumination persisted under tensile strain and twisting; the maintained control signal via the MP-PAT/Li⁺ CH proves the feasibility of the MP-PAT/Li⁺ CH as a flexible circuit interconnect (Fig. S10). Furthermore, sequential cyclic illumination with different colours was demonstrated for the LED array under tensile strain (Movie S2).

Fig. 6 (a) Conductivity of the MP-PAT/Li⁺ CH after 10 000 cyclic tensile strain and 4-week aging and application in a stretchable LED array circuit. (b) Waveform variation of the original and stretched MP-PAT/Li⁺ CH spans from 10 kHz to 10 MHz and conductive interconnects for a heartbeat sensor. (c) Schematic circuit diagram and photographs of a robot hand control system. (d), (e) and (f) Mechanical and electrical performance of the MP-PAT/Li⁺ CH after 6-months of aging.

The frequency dependence of the MP-PAT/Li⁺ CH is analysed (Fig. 6b). The hydrogel exhibits stable square-wave transmission (frequency <10 kHz) in both original and 150% strained states. Distortion caused by stretching escalates significantly as the signal frequency reaches 1 MHz. The progressive waveform distortion, phase shift, and amplitude attenuation beyond 1 MHz stem from the dielectric relaxation arising from Li⁺ ion migration lag. However, signal consistency under deformation at frequencies up to 10kHz remains essential for flexible electronic signal transmission. A photoplethysmography-based pulse sensor applying the MP-PAT/Li⁺ CH for both power supply and sinusoidal cardiac signal transmission was demonstrated as well. Continuous heart rate monitoring during mechanical stretching confirmed stable interconnects supported by the MP-PAT/Li⁺ CH (Fig. 6b and Fig. S11).

With the proliferation of flexible intelligent wearable devices, conductive hydrogels are playing an increasingly critical role.⁵⁶ We utilized the MP-PAT/Li⁺ CH to serve as a strain sensor in robotic hand control, in addition to a high gauge factor; attention should also be paid to the linearity and consistency of the sensor during cyclic stretching to achieve precise control. Therefore, although the conductive hydrogel prepared in this study exhibits strain-insensitive characteristics, its absolute resistance change during stretching can still be effectively monitored. Thus, we fabricated a sensor array based on the MP-PAT/Li⁺ CH, enabling precise control of the robotic hand. As evidenced, the robotic fingers are accurately controlled attributed to the linear resistance change within 150% strain and low electrical hysteresis of the conductive hydrogel proposed in this work (Fig. 6c and Fig. S12, Movie S3).

As illustrated in Fig. 6d–f, key mechanical and electrical properties of stretchability, low gauge factor and low hysteresis persist for at least 6 months. The declined mechanical hysteresis and GF to 3.45% and 0.08 are attributed to long-term water dynamics. Although prolonged aging leads to reduction in stiffness and roughness, the conductive hydrogel maintains property integrity within the functional working range. The results underscore exceptional mechanical and electrical durability derived from inherent humidity-adaptive properties. The sample remained functional from the initial 6-month assessment until manuscript submission (8 months total), implying that the proposed hydrogel enables extremely prolonged functional durability against dynamic RH condition.

Conclusions

The proposed conductive hydrogel, engineered via a synergistic design strategy, leverages intrinsic material characteristics to the largest degree for sustained functional durability and low electrical hysteresis. Maintenance of mechanical and electrical performances under harsh conditions of cyclic tensile strain and prolonged aging are demonstrated. The facile fabrication process and compatibility with DIW establish a robust foundation for implementing hydrogel-based materials in reliable flexible electronics.

Author contributions

Z. Li made the primary contribution to this work, with auxiliary works executed collaboratively by the co-authors.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the SI.

Detailed experimental setups, tabulated comparison with similar hydrogels, additional characterization data and experimental results, and demonstrations of the hydrogel application are included. See DOI: https://doi.org/10.1039/d5mh01139f

Acknowledgements

The authors acknowledge the financial support from the National Natural Science Foundation of China (Grant No. 62271176).

References

1. Y. Ye and F. Jiang, Nano Energy, 2022, 99, 107374.
2. T. Zhu, Y. Ni, G. M. Biesold, Y. Cheng, M. Ge, H. Li, J. Huang, Z. Lin and Y. Lai, Chem. Soc. Rev., 2023, 52, 473–509.
3. Y. Si, S. Chen, M. Li, S. Li, Y. Pei and X. Guo, Adv. Intell. Syst., 2022, 4, 2100046.
4. Z. Wang, X. Xu, K. Zhang, R. Tan, S. Zhang, Y. Su and J. Hu, Adv. Funct. Mater., 2025, 35(3), 2413478.
5. T. W. Sun, M. Venkatesan, Y. C. Hsu, J. Chandrasekar, W. C. Chen, J. S. Bénas, C. J. Cho, J. H. Lin, F. C. Liang, A. Y. Rwei and Others, Nano Energy, 2025, 133, 110416.
6. F. Han, S. Chen, F. Wang, M. Liu, J. Li, H. Liu, Y. Yang, H. Zhang, D. Liu, R. He and Others, Adv. Sci., 2025, 12(16), 2412726.
7. J. Y. Yu, S. E. Moon, J. H. Kim and S. M. Kang, Nano-Micro Lett., 2023, 15, 51.
8. S. N. Subraveti and S. R. Raghavan, ACS Appl. Mater. Interfaces, 2021, 13, 37645–37654.
9. X. Li, D. Lou, H. Wang, X. Sun, J. Li and Y. N. Liu, Adv. Funct. Mater., 2020, 30, 2007291.
10. F. Sun, X. Huang, X. Wang, H. Liu, Y. Wu, F. Du and Y. Zhang, Colloids Surf., A, 2021, 625, 126897.
11. Z. Wu, H. Ding, K. Tao, Y. Wei, X. Gui, W. Shi, X. Xie and J. Wu, ACS Appl. Mater. Interfaces, 2021, 13, 21854–21864.
12. P. Le Floch, X. Yao, Q. Liu, Z. Wang, G. Nian, Y. Sun, L. Jia and Z. Suo, ACS Appl. Mater. Interfaces, 2017, 9, 25542–25552.
13. Q. Liu, G. Nian, C. Yang, S. Qu and Z. Suo, Nat. Commun., 2018, 9, 846.
14. G. Li, C. Li, G. Li, D. Yu, Z. Song, H. Wang, X. Liu, H. Liu and W. Liu, Small, 2022, 18(5), 2101518.
15. S. Hong, T. Park, J. Lee, Y. Ji, J. Walsh, T. Yu, J. Y. Park, J. Lim, C. B. Alston and L. Solorio, et al. , ACS Sens., 2024, 9, 662–673.
16. Y. Y. Lee, H. Y. Kang, S. H. Gwon, G. M. Choi, S. M. Lim, J. Y. Sun and Y. C. Joo, Adv. Mater., 2016, 28, 1636–1643.
17. X. Li, C. Wu, J. Wu, R. Sun, B. Hou, C. Liu and M. Chen, ACS Appl. Mater. Interfaces, 2024, 16, 51699–51714.
18. M. Li, L. Chen, Y. Li, X. Dai, Z. Jin, Y. Zhang, W. Feng, L. T. Yan, Y. Cao and C. Wang, Nat. Commun., 2022, 13, 2279.
19. K. Wen, C. Xin, S. Guan, X. Wu, S. He, C. Xue, S. Liu, Y. Shen, L. Li and C. W. Nan, Adv. Mater., 2022, 34, 2202143.
20. D. Zhang, Z. Luo, H. Xu, Y. Guo, H. Chen, Y. Ye, J. An, J. Hui, Y. Shi and S. Yang, et al. , Adv. Funct. Mater., 2024, 34, 2409134.
21. L. Zhou, B. Zhao, J. Liang, F. Lu, W. Yang, J. Xu, J. Zheng, Y. Liu, R. Wang and Z. Liu, Mater. Horiz., 2024, 11, 3856–3866.
22. H. Song, J. Lee, M. Sagong, J. Jeon, Y. Han, J. Kim, H. G. Jung, J. S. Yu, J. Lee and I. D. Kim, Adv. Mater., 2024, 36, 2407381.
23. Y. Zhang, L. Wang, F. Li, Q. Huang, X. Zhou, J. Luo, C. Huang and K. Jiang, Ind. Crops Prod., 2024, 222, 119722.
24. W. Wu, J. Chang, Y. He, Z. Guo, S. Wang and J. Mao, Sustainable Mater. Technol., 2024, 42, e01126.
25. P. Cao, Y. Wang, J. Yang, S. Niu, X. Pan, W. Lu, L. Li, Y. Xu, J. Cui and G. W. Ho, et al. , Adv. Mater., 2024, 36, 2409632.
26. W. Shi, H. Li, J. Chen, Y. C. Ching, C. H. Chuah, C. Xu, M. Liu, J. Zhang, K. Y. Ching and Y. Liang, et al. , Adv. Sci., 2024, 11, 2404451.
27. R. Wang, C. Liu, Z. Li, Y. Li and X. Yu, ACS Sens., 2024, 9, 5322–5332.
28. W. Huang, X. Wang, F. Luo, X. Zhao, K. Chen and Y. Qin, ACS Appl. Mater. Interfaces, 2024, 16, 49834–49844.
29. Y. Lu and W. Zhong, Chem. Eng. J., 2024, 490, 151691.
30. S. Han, Y. Hu, J. Wei, S. Li, P. Yang, H. Mi, C. Liu and C. Shen, Adv. Funct. Mater., 2024, 34, 2401607.
31. Y. Zheng, J. Wang, T. Cui, J. Zhu and Z. Gui, J. Colloid Interface Sci., 2024, 653, 56–66.
32. D. Chen, H. Bai, H. Zhu, S. Zhang, W. Wang and W. Dong, Chem. Eng. J., 2024, 480, 148192.
33. S. Lu, Z. Ma, X. Huang, Y. Wu, Z. Wang, J. Liu, M. Ding, L. Qin and G. Dong, Chem. Eng. J., 2024, 496, 153704.
34. S. A. N. Entifar, N. A. E. Entifar, A. F. Wibowo, J. H. Kim, Y. S. Sembiring, J. M. W. Saputro, H. G. Kim, J. O. Kim, G. Xie and J. Oh, et al. , Chem. Eng. J., 2025, 509, 160971.
35. L. Zheng, D. Liu, J. Dai, Z. Hao, W. Wu and X. Yin, Chem. Eng. J., 2024, 500, 157289.
36. W. Wang, P. Guo, X. Liu, M. Chen, J. Li, Z. Hu, G. Li, Q. Chang, K. Shi and X. Wang, et al. , Adv. Funct. Mater., 2024, 34, 2316346.
37. D. Singh and A. Rawal, Carbon, 2022, 190, 299–311.
38. A. S. Pillai, A. Chandran and S. K. Peethambharan, Appl. Mater. Today, 2021, 23, 100987.
39. Z. Li, T. Zhao and W. Lv, Ind. Eng. Chem. Res., 2021, 60, 10193–10202.
40. Y. Jia, J. Chen, W. Liu and D. Yin, J. Polym. Res., 2019, 26, 1–9.
41. D. Ngo, H. Liu, Z. Chen, H. Kaya, T. J. Zimudzi, S. Gin, T. Mahadevan, J. Du and S. H. Kim, npj Mater. Degrad., 2020, 4, 1.
42. F. J. Mofrad, A. Ahmadpour and S. O. Movahed, J. Appl. Polym. Sci., 2024, 141, e55640.
43. R. Ellerbrock, M. Stein and J. Schaller, Sci. Rep., 2022, 12, 11708.
44. J. R. Rodriguez, B. Seo and B. M. Savoie, Batteries Supercaps, 2022, 5, e202100223.
45. C. Lim, J. H. Kim, Y. Chae, K. K. Lee, K. Kwak and M. Cho, Anal. Chem., 2021, 93, 12594–12601.
46. X. Tian and H. Tian, React. Chem. Eng., 2021, 6, 1002–1015.
47. Y. Zhang, L. Zhu, L. Chen, L. Liu and G. Ye, Rev. Adv. Mater. Sci., 2019, 58, 32–37.
48. A. Liu, G. Liu, C. Zhu, H. Zhu, E. Fortunato, R. Martins and F. Shan, Adv. Electron. Mater., 2016, 2, 1600140.
49. K. N. Wood and G. Teeter, ACS Appl. Energy Mater., 2018, 1, 4493–4504.
50. J. Li, J. Liu, C. Gao, J. Zhang and H. Sun, Int. J. Photoenergy, 2009, 650509.
51. A. D. Dobrzańska-Danikiewicz, W. Wolany, D. Łukowiec, K. Jurkiewicz and P. Niedziałkowski, Nanomater. Nanotechnol., 2017, 7, 1847980417707173.
52. N. Tanty, A. Patra, K. P. Maity and V. Prasad, Bull. Mater. Sci., 2019, 42, 1–7.
53. H. M. Ragab and A. Rajeh, J. Mater. Sci.: Mater. Electron., 2020, 31, 16780–16792.
54. F. P. Du, N. N. Cao, Y. F. Zhang, P. Fu, Y. G. Wu, Z. D. Lin, R. Shi, A. Amini and C. Cheng, Sci. Rep., 2018, 8, 6441.
55. E. Courbon, P. D’Ans, O. Skrylnyk and M. Frère, J. Energy Storage, 2020, 32, 101699.
56. C. Xue, Y. Zhao, Y. Liao and H. Zhang, et al. , Adv. Mater., 2025, 37, 2416275.

---