Dual Action of Phase Separation and Mechanical Locking Enabled Low‐Value Waste Wood Into High‐Performance Structural Phase‐Change‐Induced Self‐Healing Materials

ABSTRACT

Wood is used as a high‐performance structural material in advanced buildings owing to its high mechanical strength, sustainability, and environmental friendliness. However, to overcome the intrinsic size limitation of natural wood, developing green adhesives with high strength, self‐healing, recyclability, and low cost to replace polluting petroleum‐based products has become the key objective for the next stage in wood adhesives research. Inspired by the multiscale structure of gecko toe pads, this work develops a biomimetic self‐healing adhesive featuring dual mechanisms of phase separation and mechanical interlocking. By introducing sodium acetate trihydrate (SAT) into a poly(vinyl alcohol) (PVA)/wood powder (W) adhesive, phase separation occurs to promote the formation of a dense hydrogen‐bonding network and a mechanically interlocked microstructure, which makes the green, sustainable PVA/W/SAT adhesive with high shear strength (5 MPa). Benefiting from SAT's excellent phase‐change capability, the material achieves phase‐change‐induced self‐healing at crack interfaces. The combination of superior mechanical performance, self‐healing functionality, and environmental friendliness makes the PVA/W/SAT adhesive a promising candidate for transforming waste wood into high‐performance planks, significantly reducing the carbon footprint of structural materials.

A biomimetic, phase‐change‐induced self‐healing adhesive with dense hydrogen‐bonded networks and mechanically interlocked microstructures is developed through SAT‐induced phase separation inspired by gecko toe pads. The PVA/wood/SAT adhesive exhibits a shear strength of 5 MPa and crack‐interface, enabling phase‐change‐induced self‐healing by SAT's phase change, while remaining green and biodegradable for upgrading waste wood into high‐performance planks with a reduced carbon footprint.

1. Introduction

As a representative biomass material, wood has been widely utilized in the field of engineering materials due to its excellent mechanical strength, wide sources, and environmental friendliness [1, 2, 3, 4, 5]. However, natural wood makes it necessary to use adhesives during processing to achieve large‐scale recombination. The environmental footprint and mechanical properties of adhesives directly determine the life cycle performance of wooden components. Most commercial wood adhesives, particularly formaldehyde‐based resins like urea‐formaldehyde, melamine‐formaldehyde, and phenol‐formaldehyde [6, 7, 8], are derived from petrochemical feedstocks. Although they offer strong bonding performance, their production and use raise considerable environmental and health concerns due to the emission of volatile organic compounds and the non‐biodegradable, non‐renewable nature of the polymers [9]. These adhesives currently rely on an irreversible cross‐linked network after curing, which limits their application in recyclable and self‐healing fields, decreasing service life and recyclability to facilitate end‐of‐life material recovery. Therefore, developing green adhesives with high bonding strength and self‐healing performance constitutes an important goal for the next stage in the field of wood adhesives research [10, 11].

Currently, research on green adhesives predominantly focuses on bio‐derived or benign components, such as lignin [12, 13, 14], cellulose [15, 16, 17], soy [18, 19], and starch [20, 21, 22], as alternatives to petroleum‐based adhesives. These biomass materials are abundant, renewable, and possess unique chemical structures that offer potential to modulate adhesive performance. Gong et al. [23] obtained formaldehyde‐free particleboards with excellent mechanical properties and sustainability by in situ softening lignin and self‐crosslinking through hot pressing. Zhou et al. [24] fabricated a strong and tough bio‐based adhesive using soybean protein as the matrix and lignin‑polyurea (HBPU‐grafted oxidized lignin) as the framework with antibacterial properties, formaldehyde‑free characteristics, and low‑cost advantages. In addition, the biomimetic stiff‐vein/soft‐membrane hierarchical design has been applied to bio‐based wood adhesive systems, in which functionalized hexagonal boron nitride serves as the stiff vein and acts synergistically with a soy‐protein soft matrix, markedly improving wet shear strength, water resistance, flame retardancy, and antifungal performance, thereby demonstrating the generality of stiff–soft synergy and crack‐deflection/energy‐dissipation mechanisms [25]. Despite significant advancements in the field of green adhesives, there are several drawbacks, including low adhesive strength and poor self‐healing ability, limiting their broad industrial application. A particular technical hurdle has been the simultaneous integration of robust bonding strength with dynamic, reversible properties that enable self‐healing, a feature scarcely found in existing adhesive systems.

In nature, biological systems have evolved sophisticated strategies to achieve reversible yet strong adhesion. A prime example is the gecko, whose toe pads enable remarkable adhesion and detachment through a combination of van der Waals forces and a sophisticated multiscale architecture of setae and spatulae [26]. This structure facilitates massive surface contact and mechanical interlocking with substrate asperities. The gecko's ability to repeatedly adhere without leaving a residue provides a powerful design principle for synthetic adhesives, emphasizing the synergy between chemical interactions and physical microstructure [27]. Inspired by this mechanism, we hypothesized that a biomimetic adhesive featuring a dual‐interaction network combining dense physical bonding (e.g., hydrogen bonding) and mechanical interlocking at the micro‐scale could achieve both high strength and dynamic functionality.

Herein, we develop a bionic phase‐change‐induced self‐healing wood adhesive inspired by the gecko's adhesive strategy (Figure 1a,b), in which the salting‐out effect of the hydrated inorganic salt sodium acetate trihydrate (SAT) on a poly(vinyl alcohol) (PVA)/wood powder(W) blend to induce phase separation, which intensifies interchain hydrogen bonding and results in a dense and stable hydrogen‐bond network. Moreover, the transformation of SAT from the melting state to the crystalline state during phase change forms a mechanical interlocking structure and greatly enhances the bonding strength between the adhesive and the substrate (Figure 1c–f and Figure S1) [28]. This further enhances bonding strength to damaged wood, thereby achieving high performance (Figure 1g). Meanwhile, due to the excellent phase‐change capability of SAT, through the self‐healing performance induced by phase transformation, the material can regain a strong bonding strength after being damaged. Furthermore, the addition of wood powder enhances the cohesion of PVA/SAT, fills voids in damaged wood, and improves the mechanical strength of the repaired wood. These superior mechanical and phase‐change‐induced self‐healing capabilities, as well as inherent environmental friendliness, show broad prospects for efficiently converting residual or waste wood into high‐performance structural materials.

FIGURE 1.

The design strategy of bionic wood adhesive. (a) The gecko adheres to the wall. (b) The schematic illustration of the adhesion of Gecko toes’ setae by adapting to rough surfaces (increased contact area). (c) The schematic illustration of the adhesion and cohesion mechanisms of PVA/W/SAT. (d–f) Polarizing micrographs showing the crystallization process from PVA/W to PVA/W/Melting SAT and then to PVA/W/Crystal SAT. (g) Low‐value waste wood transforming into high‐performance wood by PVA/W/SAT adhesive. (h) Radar plots comparing the performance of PVA/W/SAT and different adhesives, with the PVA/W/SAT normalized to the maximum value of each characteristic.

2. Results and Discussion

2.1. Interfacial Adhesion Mechanisms of PVA/W/SAT Composite Adhesives

To transform low‐value waste wood into high‐performance structural materials, PVA/W was prepared by mixing a 10 wt% PVA solution with 300 mesh wood powder at a mass ratio of 4:1. During bonding, an adhesive‐layer thickness of 200–300 µm was maintained. Lap joints were assembled with the substrate at 5 kPa pressure. The assembly material was then immersed in molten SAT at 90°C for 24 h. After being removed, the samples were allowed to crystallize upon cooling to room temperature, and PVA/W/SAT was obtained. Compared the tensile strength and work of fracture of PVA, PVA/SAT, and PVA/W/SAT, the addition of wood powder significantly enhances the shear strength of PVA/W/SAT, which is attributed to reduced volume shrinkage (Figure S2) during adhesive bonding and enhanced interfacial adhesion [29]. As a particulate filler, wood powder increases the solid content (Figure S3) and cohesion of PVA/SAT, leading to a 6.5‐fold improvement in tensile strength and a 16.3‐fold increase in work of fracture (Figures 2a and Figure S4). Finally, by comparing the shear strength of adhesives with varying PVA concentrations and wood flour contents, the formulation with 10% PVA and 20% wood flour is selected for subsequent studies due to its optimal overall performance (Figure S5). Notably, regardless of variations in parameters such as the wood powder particle size (60–300 mesh) or the source of raw materials (shavings, sawdust, and wood chips), the wood powder filler consistently and stably enhances the shear strength in the PVA/SAT adhesive system (Figure S6). These results validate the universality of wood‐powder‐reinforced PVA. They provide a basis for transforming low‐value wood powder into a biomimetic wood adhesive.

FIGURE 2.

Mechanical properties and molecular interaction mechanisms of the wood‐based adhesives. (a) Tensile strength and Work of fracture of PVA, PVA/SAT, and PVA/W/SAT. (b) The shear stress versus displacement curves for PVA/W, PVA/SAT, PVA/W/SAT. (c) The SEM image of the adhesive‐substrate interface. (d) XRD of PVA/W before and after immersion in SAT. (e) FTIR spectra of PVA/W and PVA/W/SAT. (f) The evolution of the characteristic peak of the hydroxyl group in FTIR spectra with time for an as‐prepared hydrogel soaked in SAT. (g) LF NMR spectra of PVA/W before and after SAT incorporation. (h) Low‐field (LF) NMR spectra of different hydrogels. (i) The distributions of the three water species in the different hydrogels were determined by integrating the corresponding characteristic peak areas in the LF curves. (j) Molecular dynamics (MD) simulations depicting cellulose and PVA molecules' interactions in water or SAT environments. (k) Total count of hydrogen bonds between cellulose and PVA, and between PVA molecules in water and SAT environments from MD simulations.

Figure 2b illustrates the stress‐displacement curves for the pure PVA, PVA/W, PVA/SAT, and PVA/W/SAT adhesives. Upon incorporation of SAT, both PVA/SAT and PVA/W/SAT exhibit higher shear strength than pure PVA and PVA/W. The PVA/W/SAT achieves an interfacial shear strength of 5 MPa due to the dual action of SAT‐induced phase separation and crystallization mechanical locking. The pronounced increase is attributed to the introduction of SAT as a poor solvent, which disrupts the weak hydrogen‐bonding between PVA chains and water. This shift promotes the formation of intramolecular and intermolecular hydrogen bonds among hydroxyl groups on the PVA chains [30]. The SEM in Figure 2c shows that PVA/W/SAT forms a well‐defined mechanically interlocked structure with the wood substrate and achieves robust interfacial adhesion. Moreover, the hydroxyl groups of the PVA chains and the wood powder combine to form stronger hydrogen bonds. This robust hydrogen‐bond interaction significantly enhances the intrinsic strength of the material, leading to substrate failure rather than cohesive failure in the PVA/W/SAT adhesive (Figure S7).

Following SAT treatment, the x‐ray diffraction (XRD) pattern of the PVA/W adhesive shows notable changes (Figure 2d). The incorporation of melting SAT markedly enhances the intensity of the crystallization peak of PVA/W at 19.8°, which corresponds to the typical (101) crystal plane of PVA. This represents a significant deviation from the crystallization behavior observed without SAT, suggesting increased phase separation and crystalline content in PVA [31, 32, 33]. The characteristic diffraction peak of PVA at 19.8° in PVA/W is less pronounced than that in PVA/SAT(Figure S8), suggesting that the PVA crystalline structure is not well maintained in PVA/W, which is attributed to the good miscibility between wood powder and PVA [34]. Meanwhile, a new diffraction peak 2θ = 26.5° is observed in PVA/W, but disappears in PVA/W/SAT. This phenomenon is likely attributed to the introduction of SAT, which facilitates interactions between the cellulose in the wood powder and PVA through hydrogen bonding. These interactions disrupt the original ordered aggregates and crystalline structure of PVA/W and lead to the reconstruction of the crystalline structure between PVA and wood powder in the PVA/W/SAT adhesive. Additionally, we have compared the XRD of PVA/W under different immersion temperatures, times, and SAT concentrations. It is evident that with increasing temperature, time, and concentration, the characteristic diffraction peak of PVA at 19.8°has increased. This further proves that the salt‑out effect induced by SAT enhances intermolecular interactions between PVA chains Figure S9.

Moreover, as shown in Figure 2e and Figure S10a, the addition of melting SAT results in a shift of the ─OH stretching vibration peak at 3300 cm⁻¹ in PVA/W to a lower wavenumber, accompanied by a noticeable decrease in peak intensity. This indicates that SAT reduced water content and facilitates the formation of stronger hydrogen bonds within PVA, thereby reducing the detectability of ─OH groups. Similarly, the characteristic peak of the ─OH group moves to lower wavenumbers with the increase of soaking time (Figure 2f), enhancing the claim that the SSAT‐induced salting‐out effect favors the intermolecular interactions among PVA chains. A comparison of the FT–IR spectra of PVA/W/SAT before and after the melting and crystallization of SAT reveals that there are no significant changes in the characteristic peaks after crystallization. This shift indicates that dynamic, reversible hydrogen bonding is effectively “locked.” The melting‐crystallization transformation of SAT is a critical factor regulating hydrogen‐bond interactions, enabling dynamic and reversible bond transformations [35]. This process not only plays a pivotal role in modulating internal bonding dynamics but also provides phase‐change‐induced repairability for the PVA/W/SAT adhesive, as further illustrated in Figure S10b. Additionally, we have compared FTIR spectra of PVA/W under different immersion temperatures, times, and SAT concentrations. It is evident that with increasing temperature, time, and concentration, the hydroxyl peak of PVA/W/SAT shifts to a lower wavenumber. This further proves that the salt‑out effect induced by SAT enhances intermolecular interactions between PVA chains Figure S11. LF‐NMR reveals that the bound water‐related relaxation time (T2) in the PVA/W/SAT hydrogel shifts toward shorter times, indicating the formation of a strong hydrogen bonding network (Figure 2g) [36]. This observation is further supported by a comparison of the LF‐NMR spectra of the different hydrogels (Figure 2h). When the PVA hydrogel is treated with cellulose with salt assistance, the resulting PVA/W/SAT hydrogel reduces more free water and yields a large amount of immediate and bound water. This is primarily attributed to a dense H‐bonded crosslinking network formed between PVA and cellulose chains, which confines water mobility and thereby promotes the formation of more immediate and bound water (Figure 2i) [37]. To elucidate the mechanism underlying the morphology transition of the PVA/wood (composed primarily of cellulose, hemicellulose, and lignin) composite, we conducted molecular dynamics (MD) simulations to simulate the kinetic behavior of cellulose and PVA molecules under different environments (Figure 2j). A cellulose fibril was placed at the center of the simulation box, with PVA molecules randomly dispersed around it. After 50 ns of simulation, the cellulose curled, and PVA molecules were partially adsorbed on the cellulose surface, with some chains spontaneously curling. Moreover, the number of hydrogen bonds within cellulose and within PVA, as well as among PVA molecules, increased significantly (Figure 2k). These additional hydrogen bonds contribute to enhanced mechanical strength, thereby offering potential for efficient adhesion.

2.2. Adhesion Properties of PVA/W/SAT Adhesive

As a hydrated inorganic salt with phase transformation properties, SAT not only induces phase separation in PVA/W through a salting‐out effect but also forms a mechanically interlocking structure [38, 39] before and after the phase change, thereby significantly enhancing bonding capacity. The PVA/W/SAT adhesive exhibits excellent adhesion performance, and the adhered wood sheets withstand shear forces over 80 kg without failure under an overlapped bonded area of only ≈4 cm² (Figure 3a). This remarkable strength originates from the fluidity of the PVA/W adhesive in its melting state, which allows it to penetrate and fill voids on the rough substrate surface. Subsequent crystallization solidifies the microstructure and locks the interactions between the PVA/W/SAT adhesive and the substrate. Consequently, the transition from the soft PVA/W to the rigid PVA/W/SAT adhesive is utilized to enhance adhesion properties (Figure 3b) [40, 41].

FIGURE 3.

Mechanical interlocking and adhesion properties. (a) The PVA/W/SAT adhered woods with a shear bond (adhesive area: 4 cm²) enduring an 80 kg adult male. (b) Surface adaptation of soft PVA/W (increased contact area) and mechanical locking of rigid PVA/W/SAT adhesive after crystallization (enhances mechanical properties). (c) The shear strength and (d) shear stress curves of PVA/W/SAT before and after phase transition. (e) The shear strength of PVA/W/SAT under varying SAT treatment times and concentrations. (f) Adhesion strength and (g) shear stress displacement curves of PVA/W/Melting SAT adhesive for different wood substrates. (i) Adhesion strength and (j) shear stress displacement curves of PVA/W/Crystal SAT adhesive for different wood substrates. Simulation of the shear adhesion test for (h) PVA/W/Melting SAT and (k) PVA/W/Crystal SAT.

To further assess the influence of the SAT phase transition on adhesion performance, we compare the shear strength of PVA/W/Melting SAT and PVA/W/Crystal SAT. As shown in Figure 3c,d, the shear strength of PVA/W/SAT increases 53.9‐fold from 0.09 to 4.95 MPa after SAT crystallization. This is attributed to SAT‐induced phase separation and crystallization, which lock the PVA/W adhesive and result in a denser and stronger interfacial connection between the PVA/W/SAT adhesive and the wood substrate. The shear performance of PVA/W/SAT is significantly influenced by both the extent of SAT‐driven phase separation within PVA/W and the mechanical interlocking provided by SAT during its melt‐to‐crystal transition. In addition, as SAT treatment time and concentration increase, the degree of induced phase separation significantly increases, resulting in an increase in the shear strength of the PVA/W/SAT adhesive from 0.13 to 4.62 MPa after SAT crystallization (Figure 3e). The maximum shear strength is achieved and remains stable after 24 h treatment.

To investigate the effect of surface roughness and chemical composition on the mechanical interlocking strength of SAT, we conduct shear tests on three typical types of wood (hardwood, softwood, and bamboo) and various non‐wood substrates (ceramic, steel, PMMA, and glass). The bonding strength of PVA/W/SAT adhesives on different wood substrates in the melting SAT state is evaluated through shear tests (Figure 3f,g). The application of melting SAT results in a moderate improvement in shear strength. This enhancement is attributed to hydrogen‐bond interactions between the PVA chains and the hydroxyl groups on the wood substrate surface, facilitated by SAT‐induced phase separation (Figure S12a). In contrast, the shear strength of PVA/W/SAT in the crystalline state is much higher than that in the melting state (Figure 3i,j), clearly demonstrating the adhesion enhancement provided by the mechanical interlocking of crystallized SAT. Meanwhile, the failure mode is substrate failure rather than adhesive failure, which confirms the strong bonding capability of PVA/W/SAT (Figure S12b). Consistent trends are observed across ceramic, steel, PMMA, and glass substrates before and after SAT crystallization, further indicating that the mechanical interlocking effect induced by SAT crystallization significantly enhances the adhesion performance of PVA/W/SAT (Figure S13). Interestingly, after lap‐shear testing, all non‐wood substrates remain intact and exhibit adhesive failure both before and after SAT crystallization. However, in wood substrates, cohesive failure occurs after SAT crystallization, whereas failure at the interface is observed before crystallization (Figure S12c,d). This difference is likely due to the higher surface roughness of wood, which allows stronger mechanical interlocking with PVA/W/SAT after crystallization [42, 43]. The ceramic substrates with rough surfaces demonstrated higher shear properties, which further supported this conclusion. PVA/W/Melting SAT exhibits larger displacement before failure compared with its crystalline counterpart (Figure 3g,j), which is attributed to the lower modulus of the soft PVA/W/Melting SAT than the rigid PVA/W/Crystalline SAT adhesive. Soft adhesives are prone to stress concentration during lap‐shear testing, leading to earlier bond failure. The same results are observed in the simulations (Figure 3h,k): there is negligible deformation in the PVA/W/Crystalline SAT adhesive under high shear stress, in contrast to significant deformation in the melting state under the same load. Furthermore, the wood sheets adhered with PVA/W/SAT demonstrate excellent durability, sustain repeated shear cycles with a stress range of 200 to −100 kPa without failure (Figure S14a), and withstand dynamic loading for over 3000 cycles without adhesion failure (Figure S14b). The work of debonding was determined by integrating the area under the shear force–displacement curves (Figure S15). Comparative analysis reveals an increase in the work of debonding after SAT crystallization, indicating that the crystallization process enhances the material's toughness.

2.3. Phase‐Change‐Induced Self‐Healing and Low Environmental Impacts Strategy

Hydrated inorganic salts exhibit excellent phase‐change capacity, enabling materials to undergo reversible crystallization and melting transitions and endowing them with self‐healing capability akin to metal welding [44, 45]. Moreover, unlike metals that melt at extremely high temperatures, hydrated inorganic salt crystals melt at moderate temperatures. As a favorable phase change material, SAT facilitates damage repair through recrystallization at crack interfaces (Figure 4a). Therefore, by melting in SAT and then recrystallizing it, the original bonding performance is effectively restored after damage. The DSC cycling test confirms that SAT retains good phase change performance after multiple cycles (Figure 4b). This process results in a remarkable 58.6‐fold increase in shear strength after self‐healing repair (Figure 4c,d). In addition, on account of the reversibility of the melting and crystallization phase transition, the adhesion strength of PVA/W/SAT after 100 self‐healing cycles still remained at a relatively high (Figure 4e). The self‐healing efficiencies remained unattenuated along with the self‐healing cycles (Figure 4f). The SEM micrographs obtained after phase‐change‐induced healing indicate that PVA/W/SAT retains a defined mechanically interlocked microstructure (Figure S16). By applying dynamic shear loads from 200 to −100 kPa to samples that undergo 100 self‐healing cycles, PVA/W/SAT withstands 3000 cycles without failure, confirming excellent cyclic shear stability after self‐healing (Figure 4g). After the self‐healing cycles, the sample still lifts a 25 kg bucket, demonstrating excellent phase‐change‐induced self‐healing (Figure 4h). In addition, flammability tests were conducted on PVA/W/SAT. As shown in Figure S17 and Note S1, the prepared PVA/W/SAT exhibits a certain level of flame‐retardant performance.

FIGURE 4.

Phase change‐induced self‐healing. (a) Differential scanning calorimetry (DSC) crystallization and melting curves of PVA/W/SAT and PVA/W. (b) Differential scanning calorimetry (DSC) crystallization and melting curves of PVA/W/SAT before and after undergoing 100 heating‐cooling cycles. (c) Photographs of the PVA/W/SAT after self‐healing. (d) The shear stress curves of PVA/W/SAT before and after phase change‐induced self‐healing. (e) The adhesion strength of PVA/W/SAT before and after self‐healing. (f) Self‐healing efficiencies during the repeated self‐healing cycles. (g) Shear stress during the repeated loading‐unloading cycles. (h) Photographs of the self‐healing adhesive lift a load of 25 kg after Self‐healing cycles. (i) Comparative environmental impact assessment of PVA/W/SAT and conventional adhesives, including epoxy resin and Polyurethane.

Collectively, these findings demonstrate that a phase‑change‑enabled self‑healing approach can repeatedly restore interfacial integrity under moderate conditions and endure cyclic loading, underscoring application‑relevant reliability and durability. Nevertheless, translation to engineering practice must be evaluated beyond mechanical and healing metrics, incorporating life‑cycle considerations such as energy demand, emissions, and resource use. To evaluate the environmental impact of the PVA/W/SAT adhesive, we conduct a comprehensive cradle‐to‐grave life cycle assessment (LCA) (Figure 4i) and compare the results with those of commercially available epoxy resin and polyurethane (as detailed in Tables S1 and S2). LCA calculates environmental impacts per kilogram of material. It involves in‐depth research on a wide variety of environmental impact categories, including 11 distinct areas: abiotic depletion (ADP), abiotic depletion (Fossil Fuels) (ADF), acidification (AP), eutrophication (EP), Fresh water aquatic ecotox (FAETP), Global Warming (GWP), human toxicity (HTP), Marine aquatic ecotoxicity (MAETP), ozone Layer depletion (ODP), photochemical oxidation (PCO), Terrestrial ecotoxicity (TAETP), including the impacts on fossil fuel consumption, global warming potential, ecosystems, environmental issues, and human toxicity. By comparing these 11 categories, the PVA/W/SAT adhesive demonstrates a lower environmental impact than the two commercial adhesives. These results confirm that PVA/W/SAT is a sustainable and eco‐friendly adhesive with strong potential for broad application across multiple fields.

2.4. Fabrication and Repair of High‐Performance Isotropic Plates Enabled by PVA/W/SAT Composite Adhesives

Due to the dual action of phase separation and mechanical interlocking in the PVA/W/SAT adhesive, we fabricate high‐performance structural planks by sandwiching the adhesive between layers of natural wood (Figure 5a). By superimposing natural wood in different directions (longitudinal (L) and tangential (T)) with uniform interlayer incorporation of the PVA/W adhesive, followed by immersion in melting SAT, X‐Y‐X structural materials with excellent mechanical properties are obtained. By comparing the bonding performance with that of other biomass wood adhesives and fossil adhesives, the shear strength of the PVA/W/SAT adhesive is significantly higher than that of existing alternatives (Figure 5b, and Table S2). To investigate the mechanical behavior of the prepared high‐performance structural materials, tensile tests on natural wood and two different structural materials (X‐Y‐X structure and X‐X‐X structure; Figure S18) in the longitudinal (L) and tangential (T) directions (Figure 5c–f) are conducted. Because of the anisotropy of natural wood, natural wood shows different tensile properties (38.03 MPa in the L direction and 22.34 MPa in the T direction). The X‐X‐X structure, composed of wood layers bonded in the same orientation, retains similar anisotropy with slightly improved strength in both directions compared with natural wood. In contrast, the X‐Y‐X structural materials exhibit tensile properties significantly superior to natural wood and X‐X‐X structural materials in both the L and T directions (86.68 ± 5.81 and 80.40 ± 6.46 MPa, respectively). This enhancement in mechanical properties is attributed to the multidirectional stacking strategy, which enables effective dissipation of tensile forces and endows the plank with more isotropic mechanical behavior. Similar trends are observed in flexural performance (Figure 5g–j), where the bending strength of the X‐Y‐X structural materials in both the L and T directions (59.63 ± 7.29 and 54.99 ± 0.61 MPa, respectively) is superior to that of the X‐X‐X structural materials and natural wood. The X‐Y‐X structural materials not only meet the requirements for high mechanical strength essential in practical applications but also greatly enhance directional stability, thereby broadening their potential use in advanced structural applications.

FIGURE 5.

Tensile and flexural properties of structure wood in different directions (longitudinal (L) and tangential (T)) compared to natural wood. (a) Illustration and photograph of the preparation of the X‐Y‐X structure wood. (b) The shear strength of PVA/W/SAT compared with other adhesives (including Soy adhesive, Starch adhesive, Lignin adhesive, Petroleum adhesive). (c,e) Illustration of the structure of wood along two perpendicular tensile directions. (d,f)Tensile stress–strain profiles for X‐Y‐X structure wood along directions L and T. (g,i) Illustration of the structure wood along two perpendicular flexural directions. (h, j) Flexural stress–strain profiles of the X‐Y‐X structure wood along directions L and T.

Natural wood often contains various defects that arise during its growth, such as knots, cracks, and physical damage, which severely compromise mechanical properties. To simulate these natural defects during wood growth, point damage is inflicted on the surface of natural wood. Then, the damaged regions are filled and bonded using PVA/W/SAT, resulting in a repaired high‐performance wood product (Figure 6a). The tensile strength properties of natural wood, damaged wood, and PVA/W/SAT repaired wood reveal that the repaired wood is comparable to natural wood and significantly higher than damaged wood, demonstrating effective mechanical restoration (Figures 6b and Figure S19). Scanning electron microscopy (SEM) images of the wood before and after repair show that the PVA/W/SAT adhesive forms obvious mechanical interlocking with the wood structure, fully infiltrates the defective regions, and shows no visible delamination (Figure 6c,d). Micro‐computed tomography (CT) further verifies that the PVA/W/SAT achieves complete filling at the damaged site of the wood (Figure S20). This effective repair is attributed to the excellent fluidity of PVA/W, which enables close conformity to the rough surface of the wood damage and the formation of a high‐strength interlocking structure after SAT crystallization. After tensile testing, failure does not occur at the interface between the PVA/W/SAT adhesive and the wood (Figure 6e). The PVA/W/SAT adhesive exhibits excellent compatibility with the repaired wood, and the tensile strength at the repaired place is significantly higher than that of other sections of the wood.

FIGURE 6.

Repair of low‐value damaged wood and the mechanical properties of repaired wood. (a) Photograph of damaged wood transformed into repaired wood by PVA/W/SAT. (b) Tensile stress curves of Damaged wood and PVA/W/SAT repaired wood in directions L and T. (c,d) SEM images of damaged wood and the repaired wood. (e) Photograph of the broken parts of the repaired wood after the tensile test. (f) The damaged wood was repaired by PAV/W/SAT and superimposed into X‐Y‐X structural wood. (g,h) The tensile and flexural strengths of X‐X‐X and X‐Y‐X structured wood, repaired by different materials (including PVA, PVA/SAT, PVA/W/SAT), were compared in different directions (longitudinal and tangential) with natural and damaged wood.

To further enhance the mechanical properties of damaged wood, a multilayer X‐Y‐X structure is constructed using PVA/W/SAT repaired wood (Figure 6f). To investigate the mechanical behavior of the repaired wood after high‐performance structural transformation, tensile and flexural tests are conducted in both the L and T directions on the following materials: undamaged natural wood; wood with point damage; and repaired wood assembled into X‐X‐X and X‐Y‐X structures using PVA, PVA/SAT, and PVA/W/SAT (Figure 6g,h). Due to the absence of wood powder as a reinforcing particulate filler, the mechanical strength of PVA and PVA/SAT‐repaired wood is greatly limited. The tensile and flexural strengths of PVA and PVA/SAT repaired wood are slightly higher than those of damaged wood, but remain significantly lower than those of natural wood, and the repair effect is not pronounced. Nonetheless, the incorporation of wood powder not only enhances the solid content and viscosity of PVA but also reduces volume shrinkage during bonding, acting as a solid filler and promoting stronger interfacial integration between the PVA/W/SAT adhesive and wood. Consequently, the mechanical strength of PVA/W/SAT repaired wood is superior to that of wood repaired without wood powder. The damaged wood repaired by the PVA/W/SAT X‐Y‐X structure shows mechanical properties superior to natural wood in both the L and T directions. Additionally, similar repair effects are observed in isotropic planks such as particleboard (Figure S21), corroborating the general applicability of this approach. The PVA/W/SAT X‐Y‐X structure exhibits exceptional mechanical properties, demonstrating great potential for transforming damaged wood into high‐performance structural materials.

3. Conclusion

This study designed a biomimetic phase‐change‐induced self‐healing adhesive that leverages SAT as a poor solvent to induce phase separation in a PVA/W, upon phase‐transition crystallization, to generate a stable mechanical interlocking structure. This dual mechanism enables efficient bonding to diverse substrates, achieving a shear strength of up to 5 MPa. Owing to the excellent reversible crystallization–melting phase transition of SAT, changing the temperature induces recrystallization at damaged interfaces to form new mechanical interlocks, thereby promoting self‐healing. Even after 100 damage‐healing cycle tests, samples can be repeatedly re‐bonded through simple temperature control, while maintaining excellent adhesion strength, which significantly extends service life. Notably, X‐Y‐X structural panels fabricated from PVA/W/SAT exhibit superior mechanical performance distinct from the anisotropic behavior of natural wood, showing comparable strength in the longitudinal (L) and transverse (T) directions (near‐isotropic), with a tensile strength of approximately 80 MPa and a flexural strength of approximately 60 MPa, both markedly higher than those of natural wood. It can also efficiently repair damaged wood, making the mechanical properties of the repaired plank superior to those of natural wood. Moreover, a comprehensive life‐cycle assessment indicates a lower environmental footprint for PVA/W/SAT. These attributes not only address the critical need for high‐performance green adhesives but also signify a substantial step toward reducing the carbon footprint of structural materials.

4. Experimental Section

4.1. Materials and Chemicals

Sodium acetate trihydrate was purchased from Macklin Biochemical Technology Co., Ltd. Poly (vinyl alcohol)‐1799 (PVA‐1799) particles with a hydrolysis degree of 98%–99% were procured from Shanghai Aladdin Biochemical Technology Co., Ltd. Poplar wood powder was provided by Hongyao Mineral Products Processing Co., Ltd. Unless otherwise noted, the reagents were used as received without further purification.

4.2. Preparation of PVA/W/SAT

Solutions containing 10 wt% PVA were formulated by dissolving PVA particles in deionized water, with the mixture subjected to intense stirring and heated to 95°C to ensure complete dissolution. Subsequently, the 10 wt% PVA solution was mechanically stirred and evenly mixed with 300 mesh natural wood powder in a ratio of 4:1 to obtain PVA/W precursor. The previously prepared PVA/W was immersed in the melting SAT for one day, until the melting SAT fully impregnated into the PVA/W to prepare PVA/W/SAT.

4.3. The Molecular Dynamic Simulations for the PVA/W/SAT Adhesive

In this study, we utilized Materials Studio to build models of CEL, PVA, and Acetate molecules. The Density Functional Theory calculations were performed using the Gaussian 16 software. The B3LYP functional was adopted for all calculations. For geometry optimization, a 6–31G(d, p) basis set was used for all atoms. Then Multiwfn was used to fit the restrained electrostatic potential (RESP) charge [46]. Other bond and non‐bond parameters were obtained via AuToFF software. All‐atom MD simulations were conducted using the GROMACS software package, version 2021.5 [47, 48, 49]. The OPLS‐AA force field [50, 51] was employed to describe the molecule. The SPCE force field was used for water molecules. The molecular force field consists of nonbonded and bonded interactions. The nonbonded interaction contains van der Waals and electrostatic interactions. Energy minimization was initially carried out for the system, employing the steepest descent method to address initial contact issues. Followed by a 50 ns simulation was then performed under the NPT ensemble to relax the structure and achieve system equilibrium. The pressure was maintained at p = 1.0 bar using a Berendsen barostat, and the temperature was controlled at 298 K using a velocity‐rescale thermostat with a coupling constant of τ = 0.1 ps. All hydrogen bonds were constrained using the LINCS algorithm [52]. Simulations were performed with a time step of 2 fs, and the neighbor list was updated every 10 steps. Periodic boundary conditions were applied in all three directions. PyMOL‐3.0.3 was used for visualization.

4.4. COMSOL Multiphysics

The adhesion properties of PVA/W/Melting SAT and PVA/W/Crystalline SAT were computationally simulated utilizing COMSOL Multiphysics software. The material properties were determined by experimental measurements. In the model, two surfaces were subjected to opposite tensile action, with all displacements and rotations constrained except for the displacement along the tensile direction.

4.5. Methodology for Life Cycle Analysis (LCA)

A cradle‐to‐gate LCA was carried out to evaluate the environmental repercussions associated with PVA/wood/SAT, juxtaposing it with conventional adhesives like Epoxy resin and Polyurethane, which are widely used across sectors such as construction, furniture assembly, and packaging. The LCA focuses on 1 kg of PVA/W/SAT, Epoxy resin, and Polyurethane as the first functional unit. The life cycle inventory (LCI) data incorporated primary data from PVA/W/SAT production experiments (Table S1) and secondary data sourced from the Ecoinvent 3.5 database. The life cycle impact assessment (LCIA) utilized the “CML 2001” methodology for its analysis. Table S2 shows the simulation result.

4.6. Characterization

The surface structures of PVA/W and PVA were examined via an MS60 optical microscope from China. Mechanical properties were assessed through mechanical testing on a UTM2503 universal testing machine by Suns, China. The crystalline structures were evaluated through x‐ray diffraction analysis using a Rigaku D/max 2200 VPC x‐ray diffractometer (XRD) from Japan. The analysis covered a range from 10° to 90°, with a scanning speed of 5° min⁻¹. Furthermore, to analyze the chemical composition, a Fourier–transform infrared (FTIR) spectrum was obtained using a Thermo Fisher Scientific Nicolet 6700 Fourier transform infrared spectrometer from the USA, covering a spectral range of 550–4000 cm⁻¹. The morphologies of the adhesive–substrate interface, the adhesive–substrate interface after phase transition induced self‐healing, and the wood both after point damage and after the repair using PVA/W/SAT composites were examined using a Hitachi TM3030 scanning electron microscope (SEM) from Japan. Meanwhile, BRUKER 3D XRM micro‐computed tomography (CT) characterization was conducted on the wood both before and after the repair process. The thermal performance of both PVA/W and PVA/W/SAT composites was analyzed using a Q20 differential scanning calorimetry (DSC) from the USA, conducting measurements across a temperature range from −30°C to 90°C at a heating rate of 5°C min⁻¹ under a nitrogen atmosphere.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma73927‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.