Enhanced stimuli-responsive phase change gels through pickling-enabled ion permeation

Abstract

Supercooled phase change materials (PCMs) are desirable for thermal regulation in energy storage, electronics thermal management, and biomedical applications owing to their reversible phase change transitions and exceptional thermal energy storage capacity. Among these materials, hydrated salt phase change gels stand out due to their tunable mechanical properties and controllable solidification dynamics but are constricted by the trade-off between the low enthalpy and limited mechanical strength – a consequence of salt-ion-induced disruption of network crosslinking. Inspired by the ancient art of vegetable pickling, we develop stimuli-responsive phase change gels (PCGs) that simultaneously achieve ultrahigh stiffness and enthalpy through a novel ion permeation strategy. By leveraging the competitive hydration effect between salt ions and the polymer network—akin to the cucumber pickling mechanism—we engineer PCGs with unprecedented mechanical strength (up to 2120 MPa, the highest reported value for such materials) while maintaining exceptional thermal energy storage capacity (210 kJ kg⁻¹). The supercooling behavior of hydrated salts enables a haptic/temperature-triggered reversible transition between soft and rigid states. These PCGs exhibit outstanding deicing performance at low temperatures and excellent thermal management potential by effectively controlling the phase transition among melting, supercooling, and crystallization states. Furthermore, this approach is robust, cost-effective, and easily scalable, opening new possibilities for further thermal management.

---

New concepts

We demonstrate a biomimetic ion-permeation strategy for fabricating hydrated salt phase change gels that simultaneously achieve record mechanical strength and thermal energy storage. We report unprecedented findings that sodium acetate ions induce a salting-out effect, reorganizing polymer networks through competitive hydration and crystallization, bypassing the trade-off between mechanical robustness and thermal capacity in conventional chemically crosslinked materials. We elucidate the dynamic interplay between polymer chains, ions, and water molecules, revealing how ion-mediated dehydration enhances intermolecular hydrogen bonding and crystallinity. We engineer hydrated salt phase change gels with Young's modulus of 2120 MPa—surpassing existing hydrated salt-based phase change gels by 1–4 orders of magnitude—while maintaining a high enthalpy of 210 kJ kg⁻¹. We also achieve programmable phase transitions across a broad temperature range, enabling on-demand heat release and instantaneous mechanical stiffening via controlled crystallization. Our finding establishes a paradigm for designing multifunctional materials that integrate structural integrity with thermal responsiveness, addressing critical limitations in energy storage systems.

Introduction

Phase change materials (PCMs) have gained significant attention for their ability to store and release energy through reversible phase transitions, finding applications in diverse fields, including thermal energy storage,^1,2 deicing system,³ biomedical engineering,⁴ and mechanical actuators.^5,6 These materials demonstrate remarkable utility in maintaining physiological temperatures⁷ and enabling controlled drug release⁸ due to their high thermal storage capacity and precise phase transition temperatures. However, conventional PCMs are fundamentally limited by slow thermal response rates and narrow operational temperature windows,⁹ which restrict their use in applications requiring rapid thermal actuation or seasonal energy storage.

Supercooled PCMs present a promising alternative, maintaining a metastable liquid state below their melting point until externally triggered to release stored latent heat.^10,11 Among these, hydrated salt-based systems are particularly noteworthy for their exceptional latent heat storage capacity and stimulus-responsive behavior. Nevertheless, critical challenges persist, including phase separation and leakage issues that hinder practical implementation.^12–14 Current encapsulation strategies, such as core–shell structures or porous matrix confinement,^14,15 only partially address these limitations while introducing new complications like poor matrix-salt compatibility and suppressed supercooling effects.

Hydrogels have recently gained prominence as ideal host matrices for hydrated salt phase change materials, owing to their abundant hydrophilic functional groups,^16,17 which ensure high compatibility with hydrated salts while minimizing the effects of supercooling. These structural features enable exceptional compatibility with hydrated salts while effectively preserving supercooling characteristics. Recent advances have demonstrated that chemically crosslinked hydrated salt phase change gels (PCGs), fabricated through in situ polymerization, can achieve enhanced shape stability and reliable supercooling performance.^18–22 However, this approach presents several significant challenges: firstly, it is restricted to chemically crosslinked hydrogel systems, requiring precise control of multiple reagents and reaction conditions. Secondly, potential contamination from residual toxic crosslinking agents and unreacted monomers raises safety issues for practical applications. Thirdly, the presence of hydrated salts in non-aqueous systems substantially diminishes crosslinking effectiveness.

Drawing inspiration from traditional cucumber pickling techniques that enhance texture and preservation,²³ herein we introduce a novel pickling-inspired ion permeation method to synthesize form-stable PCGs with exceptional mechanical strength and high latent heat. Unlike traditional capillary-driven immersion methods which rely on unidirectional solution absorption into porous matrices, our approach mimics the bidirectional water-ion exchange observed in natural vegetable pickling, during which salt redistribution occurs in a hierarchical, dynamic manner, reconfiguring the polymer network during permeation. This distinction enables the development of PCGs with remarkable properties: a supercooling degree of up to 70 °C, stable supercooled state at 10 °C, record-high latent heat enthalpy of 210 kJ kg⁻¹, and exceptional mechanical tunability of approximately 6 MPa in the supercooled state and 2120 MPa upon crystallization. Leveraging these substantial advantages, the supercooled PCGs enable rapid, large-scale deicing and thermal management through triggered crystallization-induced heating.

Results and discussion

Fabrication of PCG

Pickled cucumbers are globally cherished for their distinctive crispness, slightly chewy texture, and extended shelf life. These desirable characteristics arise from the pickling process, in which fresh cucumbers undergo prolonged immersion in a high-concentration sodium chloride brine, ranging from several days to months. During this period, a critical structural transformation occurs: the initially loose cell walls of the cucumber reorganize into a denser, more compact architecture²⁴ (Fig. 1a), resulting in a marked increase in toughness and firmness.^25,26 This textural reinforcement is driven by the unique composition of the cucumber cell wall, which consists of a cellulose–hemicellulose–pectin matrix rich in hydrophilic functional groups (–OH and –COOH).²⁷ The pickling process initiates an osmotic-driven ion exchange, where Na⁺ and Cl⁻ ions penetrate the cellular matrix while water molecules diffuse outward along the concentration gradient. As a result, the cell wall undergoes dehydration and compaction, leading to enhanced mechanical strength. This structural densification not only improves the cucumber's resilience but also contributes to its prolonged preservation.

Fig. 1 Biomimetic synthesis and characterization of high-strength SAT phase change gels. (a) and (b) Schematic illustration of the pickling-inspired fabrication process for sodium acetate trihydrate (SAT) PCGs, demonstrating the structural evolution from (a) fresh cucumber cell walls to pickled architecture and (b) corresponding hydrogel-to-PCG transformation. (c) Representative stress–strain curves quantifying the enhanced mechanical performance of salt-loaded PCGs compared to pristine hydrogel (PCG-0). (d) Toughness analysis demonstrating the toughness of PCGs with different salt contents. (e) Thermal characterization showing melting enthalpy (ΔH_m) and melting temperature (T_m) variations with increasing salt content (R). (f) Cross-sectional SEM micrographs showing microstructural evolution from porous hydrogel (PCG-0) to crystal-filled networks (PCG-1.52). (g) Elemental mapping of PCG-1.52 confirming homogeneous distribution of SAT crystals (C, O, Na) throughout the polymer matrix.

Inspired by the natural toughening mechanism of pickled cucumbers, we developed phase-change gels (PCGs) through a biomimetic strategy that replicates the osmotic-driven structural reinforcement observed in plant cell walls. Specifically, we selected a polyvinyl alcohol/sodium alginate (PVA/Alg) hydrogel as the precursor material due to its structural and chemical resemblance to the cellulose–hemicellulose–pectin matrix in cucumbers—particularly the abundance of –OH and –COOH groups, which enable similar ion-exchange and dehydration-driven densification. The PVA/Alg hydrogel matrix was fabricated via freeze–thaw crosslinking, a technique that promotes the formation of crystalline domains within the PVA network,²⁸ mimicking the ordered reinforcement of cucumber cell walls during pickling. Sodium alginate was incorporated to fine-tune network porosity and enhance hydrated salt absorption through hydrogen bonding, thereby optimizing the osmotic-driven phase transition (see Supplementary Methods and Supplementary Note 1, ESI†). By emulating the natural pickling process, this approach leverages brine-induced structural compaction to achieve mechanically robust, functionally adaptive gels. The fabrication time can also be reduced by decreasing the freeze–thaw cycles for PVA crosslinking, without compromising the material's performance (Supplementary Note 2, ESI†).

To replicate the structural reinforcement mechanism of pickled cucumbers, the prepared hydrogels were immersed in sodium acetate (SA) solutions of varying concentrations for 48 hours (Fig. 1b). This biomimetic “pickling” process facilitated the diffusion of Na⁺ and Ac⁻ ions into the polymer network, triggering in situ crystallization of sodium acetate trihydrate (SAT) and yielding phase-change gels (denoted as PCG-R, where R represents the SA/H₂O mass ratio; see Methods). Mechanical testing revealed a dramatic enhancement in material performance due to this ion-permeation strategy. The optimized formulation, PCG-0.6, achieved a tensile strength (σ) of 2.81 MPa—a fourfold increase over the untreated hydrogel (PCG-0, σ = 0.64 MPa). Similarly, its toughness (U) surged to 3.3 MJ m⁻³, exceeding the baseline (PCG-0, U = 0.6 MJ m⁻³) by more than fivefold (Fig. 1c and d). Further increases in salt concentration led to a slight decrease and eventual plateau in toughness (∼2 MJ m⁻³). This reduction is likely due to phase separation from the supersaturated solution (R > 0.6), which results in the formation of uniformly distributed SAT crystals within the network. The crystals can mechanically disrupt or fracture the surrounding polymer matrix under deformation, compromising overall strength and toughness. These results unequivocally demonstrate that salt-mediated structural densification, inspired by natural pickling, can profoundly enhance the mechanical robustness of synthetic hydrogels.

Beyond mechanical reinforcement, the fabricated PCGs demonstrated pronounced phase-change functionality. Comprehensive characterization via differential scanning calorimetry (DSC), X-ray diffraction (XRD), and Raman spectroscopy revealed the emergence of SAT crystallization at R = 0.6 (Fig. S8 and S9, ESI†). Intriguingly, PCG-0.6 exhibited modified thermal properties relative to pure SAT, showing both diminished enthalpy and a depressed melting point (30.6 °C vs. 58.6 °C; Fig. 1e, Fig. S10 and Table S2, ESI†). This suppression likely stems from constrained nucleation kinetics, wherein limited ion availability impedes the development of stable crystalline nuclei.¹⁸ As salt loading increased (R > 0.6), the thermal behavior converged toward pure SAT characteristics, with PCG-1.5 ultimately achieving an enthalpy of 210.3 kJ kg⁻¹ and a melting temperature approaching the theoretical maximum (Fig. 1e and Table S2, ESI†).

Microstructural characterization through scanning electron microscopy (SEM) unequivocally demonstrated the formation and evolution of SAT crystals within the gel matrix (Fig. 1f). At R = 0.6, the crystals displayed heterogeneous spatial distribution, indicative of partial salt–water binding and incomplete crystallization. Progressive increases in salt loading led to more uniform pore occupation and a striking decrease in porosity from 53.9% (PCG-0) to merely 9.7% (PCG-1.52) (Fig. S11, ESI†). This structural transformation reached optimal organization in PCG-1.52, where the sodium alginate-to-water molar ratio (∼1 : 3) precisely matched SAT's stoichiometry, facilitating complete crystallization and thorough network infiltration. Energy-dispersive X-ray spectroscopy (EDS) elemental mapping further corroborated these findings, revealing homogeneous distributions of carbon, oxygen, and sodium atoms throughout the composite, thereby confirming the uniform integration of SAT crystals within the polymer scaffold (Fig. 1g).

Stimuli-responsive phase transition

The engineered PCGs demonstrate remarkable tri-state phase transition capability, exhibiting fully reversible transformations between crystalline, molten, and metastable supercooled phases. As illustrated in Fig. 2a, the embedded SAT crystals melt upon heating above their transition temperature, while subsequent cooling produces a stable supercooled liquid state due to the high activation energy required for spontaneous recrystallization.²⁹ This characteristic thermal hysteresis is quantitatively confirmed through dynamic mechanical analysis, which reveals an abrupt reduction in the storage modulus (E′) at the melting point during heating, while maintaining structural integrity during cooling (Fig. S12, ESI†). The supercooled state exhibits exceptional stability yet remains highly responsive to external nucleation triggers. Controlled crystallization can be reliably initiated through either of two mechanisms: (1) introduction of seed crystals or (2) mechanical surface contact, both of which effectively lower the energy barrier for heterogeneous nucleation (Fig. 2a and Video S1, ESI†). This stimuli-responsive behavior enables precise, on-demand phase switching between the three distinct material states, highlighting the system's potential for programmable phase-change applications.¹¹

Fig. 2 Controlled phase transition behavior of PCGs. (a) Schematic illustration of the reversible phase transitions between crystalline, molten, and supercooled states in PCGs. (b) T-History curve showing the supercooling stability of PCG-1.52 at 10 °C and subsequent triggered crystallization with rapid heat release (7.5 °C min⁻¹). Inset photographs demonstrate the dramatic soft-to-stiff transition (scale bars: 1 cm). (c)–(f), T Comparative mechanical properties in different states: tensile stress (c) and (d) and Young's modulus (e) and (f) for supercooled (c) and (e) versus crystallized (d) and (f) states. (g) Direct comparison of bending modulus and strength between PCG-1.2 and conventional PAM-1.2 composite.²⁰ (h) Performance benchmarking showing PCG-1.52's superior Young's modulus (2120 MPa) and enthalpy (210.3 kJ kg⁻¹) relative to state-of-the-art salt hydrate PCGs: SSD/PAM/CA,³⁰ SSD/CA/PAAS,³¹ SAT/KPAM,³² PAM/SSD/MXene,² and PAM/CA/SSD/MA³³ (complete data in Table S4, ESI†).

The phase transition behavior exhibits exceptional thermal and mechanical responsiveness, characterized by significant latent heat release during crystallization accompanied by pronounced mechanical stiffening. Notably, the system maintains outstanding supercooling stability with a supercooling degree (ΔT_s) of approximately 70 °C, approaching that of pure SAT (78 °C) (Fig. S13, ESI†). We attribute this near-perfect supercooling to the presence of Alg in the PVA network, which provides controlled nucleation sites. This is supported by the observation that Alg-free samples exhibit even greater supercooling (ΔT_s up to 82 °C) than pure SAT (Fig. S7, ESI†), suggesting that the polymer matrix effectively regulates nucleation kinetics. Additionally, the supercooling degree can be fine-tuned by optimizing processing parameters, such as heating temperature (Fig. S14, ESI†). As demonstrated by the T-history profile in Fig. 2b, PCG-1.52 showcases both remarkable supercooling persistence (>50 hours at −10 °C) and rapid thermal responsiveness. Upon triggering, the system achieves an impressive heating rate of 7.5 °C min⁻¹, peaking at 43.5 °C. This combination of long-term metastability and rapid energy release highlights the material's potential for thermal energy storage and management applications.

The contrast of mechanical properties between supercooled and crystallized states exhibits strong salt-loading dependence, becoming progressively more pronounced with increasing R values (Fig. 2c–f). While PCG-0.6 maintains similar mechanical properties in both states, PCG-1.52 undergoes remarkable property enhancement upon crystallization, achieving a 4-fold increase in tensile strength and an extraordinary 380-fold amplification in Young's modulus (E). This striking mechanical transition is vividly illustrated in Fig. 2b, where a 2 mm PCG-1.52 film transitions from being highly pliable in the supercooled state to being capable of supporting a 1 kg load without detectable deformation when crystallized. These results demonstrate that the crystallization process simultaneously enhances both the thermal energy storage capacity and mechanical rigidity of the PCGs, establishing a clear structure–property relationship between SAT crystallization and material performance.

Benchmark testing demonstrates the exceptional performance of our biomimetic PCGs compared to conventional phase-change materials.²⁰ The PCG-1.2 formulation exhibits superior mechanical properties, delivering 5-fold greater flexural strength and 12-fold higher elastic modulus than standard polyacrylamide (PAM)-SAT composites under identical test conditions (Fig. 2g and Fig. S15, ESI†). Impact resistance evaluations reveal outstanding durability, with our PCGs maintaining structural integrity under repeated hammer strikes while conventional composites suffered complete collapse (Video S2, ESI†). The optimized PCG-1.52 sets new performance standards, achieving an unprecedented Young's modulus of 2.12 GPa – surpassing existing phase-change materials by 1–4 orders of magnitude (Fig. 2h). Remarkably, this mechanical enhancement coexists with exceptional thermal performance (ΔH = 210.3 kJ kg⁻¹, η = 79.6%), exceeding all previously reported values for comparable systems^{2,18–20,30–35} (Table S4, ESI†).

Strengthening mechanism

The exceptional mechanical reinforcement observed in these gels stems primarily from a salting-out effect,³⁶ whereby elevated SA concentrations progressively diminish polymer solubility. In our system, SA ions effectively disrupt the hydrogen-bonding network between water molecules and the PVA/Alg hydrogel matrix.³⁷ Through systematic investigation of the ternary interactions among polymer chains, SA ions, and water molecules, we determined that the Alg mechanical contribution was minimal due to its low concentration (Fig. S6, ESI†). As illustrated in Fig. 3a, the PVA architecture comprises hydrophobic carbon backbones decorated with hydrophilic hydroxyl groups.³⁸ The freeze–thaw process promotes the formation of crystalline domains through interchain hydrogen bonding between these –OH groups, establishing durable crosslinks that resist dissolution.³⁹ These crystalline regions alternate with hydrated amorphous domains where –OH groups maintain hydrogen bonds with water molecules. During SA immersion, the penetrating Na⁺ and Ac⁻ ions preferentially hydrate, effectively dehydrating the polymer network and thereby increasing the crosslink density through enhanced polymer–polymer interactions. This increase is supported by the enhanced intensity of the characteristic C–C stretching vibration peak around 800 cm⁻¹ in the Raman spectra (Fig. 3b) and the diffraction peak of crystalline PVA at 2θ = 19.3° in the XRD spectra⁴⁰ (Fig. S9, ESI†).

Fig. 3 Mechanism of strength enhancement in PVA networks through pickling-inspired processing. (a) Schematic illustration of microstructural evolution and molecular interactions during the pickling process: (i) formation of crystalline domains in PVA through freeze–thaw cycling, (ii) hydrated amorphous regions maintaining flexibility, (iii) penetration of Na⁺ and Ac⁻ ions into the polymer network, (iv) ion hydration displacing bound water molecules, (v) enhanced polymer chain aggregation, and (vi) final reinforced network with SAT crystallization. (b) Raman spectra comparing PCG-1.52, PCG-0, and pure SAT, highlighting the intensified C–C stretching vibration (∼800 cm⁻¹) indicating increased polymer crystallinity. (c) FT-IR spectra (2815–2990 cm⁻¹ region) showing the red-shifted C–H stretching vibration, suggesting enhanced hydrophobic interactions with increasing salt content. (d) Differential thermogravimetric (DTG) analysis demonstrating the improved thermal stability of PCG-1.52 compared to both pristine hydrogel (PCG-0) and pure SAT, evidenced by slower water loss kinetics and higher decomposition temperature.

Thermal analysis demonstrates a significant 53.6% enhancement in PVA crystallization enthalpy (increasing from 62.97 to 96.69 kJ kg⁻¹) as R progresses from 0 to 0.6 (Fig. S16 and Table S5, ESI†), with complementary FT-IR spectroscopy revealing the underlying crosslinking mechanism through two key observations: (1) increased carboxylate group ratios confirming greater SA incorporation and (2) red-shifted C–H stretching vibrations (Fig. 3c) indicative of strengthened hydrophobic interactions and diminished PVA–water hydrogen bonding – while maintaining unchanged core chemical structures (Fig. S17, ESI†), thereby excluding chemical reactions between SA ions and the gel network. These structural modifications originate from the competitive hydration behavior of Na⁺ and Ac⁻ ions, which form extensive hydration shells that displace water molecules from PVA chains (Fig. 3a(iv)), ultimately driving network densification through three synergistic effects: macroscopic shrinkage, mechanical reinforcement via enhanced polymer–polymer interactions, and concurrent SAT crystallization within the matrix (Fig. 3a(v)–(vi)).

Microstructural and thermal characterization collectively confirm the ion-mediated reinforcement mechanism, with differential thermogravimetric (DTG) analysis revealing that PCG-1.52 demonstrates markedly slower water loss kinetics and elevated decomposition temperatures compared to both the pristine hydrogel and pure SAT (Fig. 3d) – clear evidence of the profound influence exerted by ionic species and the polymer network on water molecule dynamics. These results establish that sodium acetate ions strengthen the PCGs through a dual-action mechanism: salting-out effects that reduce polymer–water affinity and competitive hydration that simultaneously enhances intermolecular polymer bonding while disrupting aqueous solvation shells. This synergistic action ultimately yields the observed mechanical reinforcement while maintaining the material's thermal energy storage capabilities.

Potential applications

To validate the practical utility of the superior properties of such PCGs, we assessed the potential for deicing applications through controlled phase-transition-induced heat release at low temperatures, as illustrated in Fig. 4a. Specifically, the supercooled PCG-1.52 sample (dimensions: 2 mm × 20 mm × 30 mm) was encapsulated in a polytetrafluoroethylene (PTFE) membrane and positioned on a platform with a 10° inclination. Due to the inherent hydrophobicity of the PTFE surface, water droplets deposited at room temperature spontaneously slid off, thereby maintaining a dry substrate (Video S3, ESI†). In contrast, when the temperature was reduced to −10 °C, the deposited droplets rapidly froze and adhered firmly to the substrate, simulating typical conditions of ice accretion that can lead to structural damage. To remove the frozen droplets, we induced a phase transition from the supercooled state to the crystalline state by applying localized mechanical stimulation via needle-tip triggering, which released a substantial amount of latent heat, as demonstrated by the infrared photography (Fig. 4b). Remarkably, within just 10 seconds, the interfacial ice layer melted, allowing the frozen droplet to detach autonomously from the surface, despite the ambient temperature remaining constant at −10 °C (Fig. 4c). This phase-transition behavior conclusively demonstrates the efficacy of high-latent-heat PCGs in active deicing applications. Furthermore, the PCG exhibited exceptional stability over 100 freeze–thaw cycles, with negligible degradation in latent heat capacity (Table S3, ESI†) and supercooling properties (Fig. S19, ESI†). These results underscore the robustness and long-term reliability of PCGs, making them promising candidates for real-world deicing applications.

Fig. 4 Deicing performance of PCGs. (a) Schematic representation of ice adhesion on surfaces and the active deicing mechanism enabled by controlled heating of superhydrophobic PCG. (b) Infrared thermal images revealing the spatial temperature distribution during phase transition, with localized heating reaching approximately 40 °C (scale bar: 1 cm). (c) Digital image documenting the complete ice removal process, showing sequential detachment of an ice droplet from the PCG surface (scale bar: 0.25 cm).

In addition to their potential for deicing, our PCGs also demonstrate promising applications in outdoor wearable thermal management and the stabilization of injured joints. A 3 mm-thick circular PCG-1.52 patch (Ø 100 mm) in its supercooled (flexible) state can conform to the curvature of the forearm. Upon mechanical triggering, the material undergoes rapid crystallization and releases heat, maintaining a skin-contact temperature above 36 °C for over 20 minutes (Fig. 5a and b). Following the heat release, the material transitions to a rigid state. This soft-to-stiff transition offers the potential to provide mechanical support for injured joints. We assessed the mechanical properties of PCG-1.52 after 50 and 100 melting/crystallization cycles, observing that its mechanical strength remains unchanged in the crystallized state, while the supercooled state exhibits even enhanced strength, increasing from 1.9 MPa at the first cycle to 3.4 MPa after 100 cycles (Fig. S20, ESI†). This strength improvement can be attributed to the intrinsic properties of PVA, as PVA hydrogels form crosslinked crystalline domains during freeze–thaw cycles, which leads to an increase in strength.⁴ Consequently, the controllable heat release and mechanical stability of the PCGs make them not only suitable for human thermal management but also ideal for stabilizing injured joints outdoors. Furthermore, we explore the thermal management potential for lithium-ion batteries, where the heat absorbed during the crystallization-to-melting transition effectively reduces the battery temperature by 24.3% (Supplementary Note 3, ESI†).

Fig. 5 Wearable thermal management and shape-fixing performance of PCGs. (a) Infrared thermography of the forearm before triggering, immediately after triggering, and 20 min later, demonstrating effective thermal delivery and retention. (b) Temperature profile showing rapid heat release upon triggering, maintaining a temperature above 36 °C for over 20 min.

Conclusions

In summary, inspired by the natural toughening mechanism of pickled cucumbers, we developed a bioinspired ion-permeation strategy to fabricate PCGs with simultaneously enhanced mechanical strength (achieving a record Young's modulus of ∼2.12 GPa and tensile stress of ∼8.79 MPa) and thermal energy storage capacity (∼210 kJ kg⁻¹), outperforming existing materials by 1–4 orders of magnitude through salt-induced network densification. These PCGs exhibit unique stimuli-responsive phase transitions across a broad temperature range (−10 to 57 °C), featuring instantaneous heat release (ΔT > 40 °C) and dramatic mechanical transformation (from ∼5.58 MPa in supercooled state to ∼2.12 GPa when crystallized), enabled by exceptional supercooling stability (ΔT_s ≈ 70 °C), with practical deicing performance at low temperatures and excellent thermal management potential by adjusting the states of melting, supercooling, and crystallization. We envision that our bio-inspired strategy will open new avenues for advanced applications spanning thermal energy storage systems, smart deicing technologies, and beyond, with its unique combination of mechanical robustness and precisely controllable phase-change functionality.

Methods

Materials

Polyvinyl alcohol (PVA) (molecular weight 146 000–186 000, 99+% hydrolyzed) was purchased from Sigma-Aldrich. Sodium alginate (Alg) was supplied by Fuchen Chemical Reagent Co., Ltd. Anhydrous sodium acetate (SA) and Sodium acetate trihydrate (SAT) were obtained from Macklin Reagent Co., Ltd.

Preparation of PCG

A series of SA solutions with varying SA content was prepared by dissolving SA powder in deionized water and heating it in an oven at 80 °C. The PVA/Alg hydrogels, prepared using the method described in the Supplementary Method (ESI†), were then immersed in the SA solutions for 48 hours. The mass of each substance added is listed in Table S6 (ESI†). The water content of the hydrogel was considered when preparing the SA solution to ensure the mass ratio of SA to H₂O (R) within the gel network was set at 0, 0.3, 0.6, 0.9, 1.2, and 1.52 when permeation equilibrium was achieved. Finally, the gels were removed from the SA solution, and the surface solution was gently wiped dry. The gels were designated as PCG-R.

Mechanical characterization

The hydrogels were prepared using dog-bone-shaped silicone molds with a width of 10 mm and a thickness of 4 mm. The PCGs were directly prepared using these dog-bone-shaped hydrogels. The width and thickness of each specimen were measured with a caliper. Force–displacement data were obtained using a universal testing machine (Jinan Sida, WDW-10E 10 kN) through tensile testing. Specimens were tested at a speed of 50 mm min⁻¹ until rupture. The stress–strain curves were derived by dividing the measured force by the initial cross-sectional area and the measured displacement by the initial gauge length. Toughness was determined by calculating the area under the stress–strain curves. Three specimens were tested for each condition.

Thermophysical properties characterization

Phase-change properties of PCGs were tested using a Differential Scanning Calorimeter (DSC, Netzsch, 214 Polyma). Samples were sealed in an alumina pan and determined in a nitrogen atmosphere. The heating program was as follows: (i) holding at 10 °C for 1 minute, (ii). heating to 80 °C at a rate of 5 °C min⁻¹, and (iii) holding at 80 °C for 1 minute.

The loading rate η of PCGs was calculated using the following formula (1):

η = ΔH_m,PCG/ΔH_m,SAT × 100% (1)

Here, ΔH_m,PCG represents the enthalpy of PCG, and ΔH_m,SAT is the enthalpy of pure SAT.

The thermal stability of the PCGs was analyzed using a thermogravimetric analyzer (TGA, Netzsch, 209F3). The samples were placed in an open alumina crucible and heated from 30 °C to 230 °C at a rate of 10 °C min⁻¹ in a nitrogen atmosphere.

Microscopic morphological characterization

The hydrogels and PCGs were freeze-dried for 48 hours to completely remove water. Their microscopic morphology was then examined using field emission scanning electron microscopy (SEM, Zeiss, Sigma 300) and Energy dispersive X-ray spectroscopy (EDS, Bruker, XFlash6I30) to observe the microstructure and elemental distribution.

Structural characterization

Fourier transform infrared spectroscopy (FT-IR, Bruker, Tensor27) was employed to record FT-IR spectra of the Aerogel 8P2A (hydrogel after freeze-drying), SAT, and PCGs within the wave number range of 400–4000 cm⁻¹. Raman spectroscopy (HJY LabRAM Aramiswas) was employed to record Raman spectra of the SAT and PCGs within the wavenumber range of 70–4000 cm⁻¹.

Characterization of supercooling properties and stimuli-responsive characteristics

The supercooling degree and supercooling stability of SAT and PCGs were assessed using the step-cooling curve method. Precision K-type thermocouples with an accuracy of ±0.5 °C were inserted into samples weighing 2.5 g to measure their temperatures. Temperature data were collected using a data acquisition device (Agilent, 34970A). The experiments were conducted within a high-low temperature alternating humidity test chamber (SANWOOD, TEMI1500).

To measure the supercooling degree, samples were initially heated to 85 °C and held for 4 hours to ensure complete SAT melting, during which the melting temperature T_m was recorded. Subsequently, they were cooled to −40 °C, and their actual crystalization temperature T_c was recorded. The supercooling temperature ΔT_s was calculated using formula (2):

ΔT_s = T_m − T_c (2)

For evaluating supercooling stability, samples were heated to 85 °C and held for 4 hours to ensure complete SAT melting. Then, they were cooled to −10 °C and maintained at this temperature. Supercooling stability was confirmed when the phase change gel remained supercooled for over 48 hours below its melting temperature. After being supercooled for more than 48 hours, crystallization of the phase change gel was induced by puncturing it with a needle.

Application for deicing

A rectangular specimen of PCG-1.52, measuring 2 mm in thickness, was initially coated with a hydrophobic polytetrafluoroethylene (PTFE) membrane. Subsequently, it underwent multiple layers of hydrophobic coating (Glaco), each followed by drying in a 50 °C oven. This process was repeated three times to ensure uniform coverage. Following coating, the specimen was subjected to heating and cooling cycles to achieve a supercooled state. It was then transferred to a temperature-controlled chamber set at −10 °C. A droplet of water was introduced onto its surface and monitored until freezing ensued. Afterward, the specimen was inclined at a 10° angle to simulate freezing rain on a sloped surface. To induce crystallization, a fine needle was inserted into the material. The entire experimental procedure was captured and documented using a digital camera (SONY, 4K FDR-AX700) and a thermal camera (Fluke, Tis75+).

Author contributions

P. Y. and X. W. contributed equally to this work. Z. L. conceived the research. Z. L., X. W., and S. W. supervised the research. P. Y. carried out the experiment. P. Y. and X. W. collected data and analyzed the experimental results. P. Y. and X. W. wrote the manuscript and all the authors agreed on its final contents.

Data availability

All data is available in the main text or the ESI.†

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

We acknowledge support from the National Natural Science Foundation of China (No. 22278145), the National Key Research and Development Program of China (No. 2020YFA0210704), and the Dongguan Key Research and Development Program (No. 20231200300152).

References

1. G. Li, G. Hong, D. Dong, W. Song and X. Zhang, Adv. Mater., 2018, 30, 1801754.
2. X. Qi, T. Zhu, W. Hu, W. Jiang, J. Yang, Q. Lin and Y. Wang, Compos. Sci. Technol., 2023, 234, 109947.
3. M. Hou, Z. Jiang, W. Sun, Z. Chen, F. Chu and N.-C. Lai, Adv. Mater., 2024, 36, 2310312.
4. J.-J. Ye, L.-F. Li, R.-N. Hao, M. Gong, T. Wang, J. Song, Q.-H. Meng, N.-N. Zhao, F.-J. Xu, Y. Lvov, L.-Q. Zhang and J.-J. Xue, Bioact. Mater., 2023, 21, 284–298.
5. H. Zhang, J. Mai, S. Li, J. T. Althakafy, H. Liu, A. K. Alanazi, S. M. El-Bahy, X. Li, X. Wu, R. Wang, N. Wang and X. Mai, Adv. Compos. Hybrid Mater., 2022, 5, 2042–2050.
6. Y. Bao, J. Lyu, Z. Liu, Y. Ding and X. Zhang, ACS Nano, 2021, 15, 15180–15190.
7. H. Liu, F. Zhou, X. Shi, K. Sun, Y. Kou, P. Das, Y. Li, X. Zhang, S. Mateti, Y. Chen, Z.-S. Wu and Q. Shi, Nano-Micro Lett., 2023, 15, 29.
8. C. Zhu, D. Huo, Q. Chen, J. Xue, S. Shen and Y. Xia, Adv. Mater., 2017, 29, 1703702.
9. G. G. D. Han, H. Li and J. C. Grossman, Nat. Commun., 2017, 8, 1446.
10. S. A. Mohamed, F. A. Al-Sulaiman, N. I. Ibrahim, Md. H. Zahir, A. Al-Ahmed, R. Saidur, B. S. Yılbaş and A. Z. Sahin, Renewable Sustainable Energy Rev., 2017, 70, 1072–1089.
11. V. I. Kalikmanov, in Nucleation Theory, ed. V. I. Kalikmanov, Springer Netherlands, Dordrecht, 2013, pp. 253–276.
12. S. Xi, L. Wang, H. Xie and W. Yu, ACS Nano, 2022, 16, 3843–3851.
13. P. Lian, R. Yan, Z. Wu, Z. Wang, Y. Chen, L. Zhang and X. Sheng, Adv. Compos. Hybrid Mater., 2023, 6, 74.
14. Z. Zeng, D. Huang, L. Zhang, X. Sheng and Y. Chen, Adv. Compos. Hybrid Mater., 2023, 6, 80.
15. M. Graham, J. Smith, M. Bilton, E. Shchukina, A. A. Novikov, V. Vinokurov and D. G. Shchukin, ACS Nano, 2020, 14, 8894–8901.
16. L. Hu, P. L. Chee, S. Sugiarto, Y. Yu, C. Shi, R. Yan, Z. Yao, X. Shi, J. Zhi, D. Kai, H. Yu and W. Huang, Adv. Mater., 2023, 35, 2205326.
17. X. Xue, Y. Hu, S. Wang, X. Chen, Y. Jiang and J. Su, Bioact. Mater., 2022, 12, 327–339.
18. F. (Kuo) Yang, A. Cholewinski, L. Yu, G. Rivers and B. Zhao, Nat. Mater., 2019, 18, 874–882.
19. Y. Fang, X. Xiong, L. Yang, W. Yang, H. Wang, Q. Wu, Q. Liu and J. Cui, Adv. Funct. Mater., 2023, 33, 2301505.
20. Y. Fang, Z. Bai, L. Yang, Q. Liu, W. Xu, J. Wei, K. Yang, Q. Wang and J. Cui, Adv. Funct. Mater., 2024, 2314353.
21. Q. Liu, Y. Fang, X. Xiong, W. Xu and J. Cui, Angew. Chem., Int. Ed., 2024, 63, e202320095.
22. T. B. H. Schroeder and J. Aizenberg, Nat. Commun., 2022, 13, 259.
23. H.-M. Guo, Y. Zhao, M.-N. O. Yang, Z.-H. Yang and J.-H. Li, J. Hazard. Mater., 2020, 386, 121882.
24. K. M. Yoo, I. K. Hwang, G. Eog Jr and B. Moon, J. Food Sci., 2006, 71, C97–C101.
25. E. K. McMurtrie and S. D. Johanningsmeier, J. Food Qual., 2018, 2018, 1–13.
26. R. Buescher and C. Hamilton, J. Food Qual., 2000, 23, 429–441.
27. N. Sai Prasanna and J. Mitra, Carbohydr. Polym., 2020, 247, 116706.
28. X. Jiang, N. Xiang, H. Zhang, Y. Sun, Z. Lin and L. Hou, Carbohydr. Polym., 2018, 186, 377–383.
29. V. I. Kalikmanov, in Nucleation Theory, ed. V. I. Kalikmanov, Springer Netherlands, Dordrecht, 2013, pp. 17–41.
30. T. Wang, N. Wu, H. Li, Q. Lu and Y. Jiang, J. Appl. Polym. Sci., 2016, 133, app.43836.
31. Y. Luo, W. Yu, J. Qiao, X. Zhao, H. Wu, X. Sheng, Y. Chen and P. Lin, Chem. Eng. J., 2022, 440, 135632.
32. M. Song, L. Wang, F. Shao, H. Xie, H. Xu and W. Yu, Chem. Eng. J., 2023, 464, 142682.
33. T. Zhu, W. Jiang, X. Shi, D. Sun, J. Yang, X. Qi and Y. Wang, Composites, Part A, 2023, 170, 107526.
34. H. Jing, L. Xu, X. Wang, Y. Liu and J. Hao, J. Mater. Chem. A, 2021, 9, 19914–19921.
35. C. Yin, J. Lan, X. Wang, Y. Zhang, R. Ran and L.-Y. Shi, ACS Appl. Mater. Interfaces, 2021, 13, 21810–21821.
36. W. Wei, Adv. Sci., 2023, 10, 2302057.
37. Y. Marcus, Chem. Rev., 2009, 109, 1346–1370.
38. L. Serairi, C. Santillo, P. Basset, M. Lavorgna and G. Pace, Adv. Mater., 2024, 36, 2403366.
39. C. M. Hassan and N. A. Peppas, Macromolecules, 2000, 33, 2472–2479.
40. C.-C. Yang, S.-J. Chiu, K.-T. Lee, W.-C. Chien, C.-T. Lin and C.-A. Huang, J. Power Sources, 2008, 184, 44–51.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5mh00597c

‡ These authors contributed equally to this work.

---