Strong Adhesives Mediated by Dynamic Phase‐Locking

ABSTRACT

Nature has long inspired the design of reversible smart adhesives; however, achieving both high adhesion strength and switchability remains a significant challenge, particularly for emerging applications (e.g., soft robotics and wearable electronics). Herein, we introduce switchable phase‐locking‐mediated adhesives (SPAs) that leverage high‐density hydrogen bonds to deliver ultrahigh adhesion (up to 15 MPa). The underlying mechanism involves dynamic phase‐locking mediation that harmonizes interfacial adhesion with bulk cohesion through interfacial mechanical locking and strain‐induced phase separation. This strategy optimizes performance across the complete adhesion lifecycle, including spreading, adhering, and debonding via temperature/force‐mediated phase‐locking. Through detailed molecular analysis of the SPA system, we uncover the mechanistic basis of ultrahigh adhesion and establish design guidelines applicable to future smart adhesive development.

Here, we introduce a facile supramolecular strategy to fabricate switchable phase‐locking‐mediated adhesives (SPAs) with strong adhesion. Understanding the core mechanism behind the high adhesion strength of our SPAs will serve as a crucial guide for the molecular crafting of innovative adhesive materials. We expect our research to foster the advancement and broader application of switchable adhesives across numerous burgeoning fields.

1. Introduction

High‐performance adhesives possessing both high adhesion strength and on‐demand switchability of adhesion are greatly needed in various burgeoning fields, such as soft robotics, artificial skin, and wearable electronics [1, 2, 3, 4]. Nature has long inspired researchers to fabricate these adhesives. For example, snails adhere to surfaces by secreting mucus: they first wet and conform to different substrates, then harden upon dehydration to form a mechanical lock, thereby providing strong adhesion, and can be softened with moisture for easy release [5, 6, 7]. Fundamental understanding of this mechanism has inspired the design of smart adhesives with reversible phase transition behaviors, such as shape‐memory polymers (SMPs) [8, 9, 10]. SMPs can reversibly switch from a fixed permanent shape to a programmed temporary shape under external stimulus, wherein mechanical interlocking can be dynamically established between SMP adhesives and substrates [11, 12]. However, the interfacial adhesion energy of SMP adhesives, dependent on rubber‐to‐glass transition shape locking, is much lower than the cohesive energy, frequently causing interfacial adhesion failure due to this mismatch. Therefore, it remains challenging to fabricate switchable adhesives with high adhesion strength (e.g., >10 MPa) via SMPs, possibly due to the difficulty in designing motifs capable of simultaneously providing both high bulk cohesion and strong interfacial interactions.

Supramolecular polymeric materials based on noncovalent interactions represent a promising platform for the design and construction of switchable adhesives, as these dynamic and reversible interactions can generate versatile structures and unique polymeric topologies through supramolecular assembly [13, 14, 15, 16, 17, 18]. Abundant supramolecular interactions can significantly increase interfacial energy between adhesive materials and substrates [19, 20, 21, 22, 23]. Meanwhile, dissociation of these supramolecular interactions can dissipate energy, thereby endowing adhesive materials with high toughness and enhanced bulk cohesion energy [24, 25, 26]. Owing to these combined effects, adhesion strength can be greatly improved via supramolecular design. Recently, a variety of supramolecular interactions, such as hydrogen bonds and host‐guest interactions, have been utilized to design switchable adhesives [27, 28, 29, 30]. Nevertheless, few switchable adhesives with high adhesion strength (e.g., >10 MPa) have been achieved through supramolecular strategies. This limitation likely arises from the challenge in designing supramolecular motifs and in elucidating the mechanism responsible for enabling the high adhesion strength of switchable adhesives.

Herein, we propose a chemical design of switchable phase‐locking‐mediated adhesives (SPAs) with high adhesion strength based on a high density of hydrogen bonds. The key principle of design is to select monomers containing an adequate number of hydrogen‐bonding motifs, which can facilitate tunable phase‐locking behavior and dynamic microphase‐separation under strain, simultaneously obtaining a balance between interfacial adhesion and cohesion (Figure 1a). In this strategy, SPAs can first transition to a rubbery state and rapidly spread on various substrates through low‐temperature‐induced dissociation of intermolecular hydrogen bonds (Figure 1b). Then, SPAs form a physical interlocking architecture after cooling under mild force, which can amplify the interfacial adhesion strength, resembling a “mortise‐and‐tenon joint” structure. Meanwhile, newly formed hydrogen bonds between SPAs and substrates can promote interfacial energy dissipation, achieving chemical enhancement of adhesion strength. In addition, strain‐induced microphase‐separation forms due to high‐density intermolecular hydrogen bonds, which could significantly enhance the cohesive energy of the polymer during detachment. The phase‐locking‐mediated strategy optimizes performance across the complete adhesion lifecycle (spreading, adhering, and debonding), achieving approximately 15 MPa of lap‐shear adhesion strength with switchability in our SPAs. Drawing from this concept, our SPAs are presented as a model system to foster a general understanding of the molecular mechanisms underlying the ultrahigh adhesion strength of switchable adhesives.

FIGURE 1.

(a) Chemical design and preparation process of strong and switchable polymeric adhesives. (b) Schematic diagrams and molecular mechanism of dynamic phase‐locking mediated strategy combining interfacial mechanical locking and strain‐induced phase separation through high‐density hydrogen bonds.

2. Results and Discussion

2.1. Synthesis and Properties

To validate our molecular design, we have synthesized 2‐[[(butylamino)carbonyl]oxy]ethyl acrylate (BEA) as the soft monomer (Figure S1), wherein the long alkyl chain imparts a low glass transition temperature to its homopolymer while providing sufficient hydrogen‐bonding capability. Acrylic acid (AA) was carefully selected as the hard monomer owing to its excellent compatibility with the BEA segments. Specifically, the SPAs‐1–4 were copolymerized using a constant molar ratio of AA (set as 1) and varying the content of BEA from 0.75 to 1.5 via UV‐initiated free‐radical polymerization (TPO as photoinitiator). Notably, no macroscopic phase separation was observed across all compositions after copolymerizing these two monomers, resulting in high optical transparency (93% transmittance in the visible region) (Figure 2a).

FIGURE 2.

(a) Transmittance curves of the SPAs (film thickness = 0.2 mm; the inset shows a photo of the SPA‐2). Photo taken by F. Dai. (b) Stress−strain curves of SPAs. (c) The Young's modulus and toughness of SPAs. (d) DSC curves of SPAs. (e) Time‐temperature superposition rheological and tan 𝛿 curves of the SPA‐2 at the reference temperature of 20°C. (f) Adhesion strengths of SPAs on Al₂O₃ ceramic substrates at 20°C. (g) Pull‐off adhesion strengths of SPA‐0 and SPA‐2 on glass at 20°C. (h) Comparison of the adhesion strength of SPA‐2 with other previously reported switchable polymeric adhesives in the lap‐shear testing model [22, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47]. (i) Macroscopic adhesion of SPA‐2 at 20°C on glass (20 kg weight, adhesion area of about 3 cm²).

We further investigated the mechanical behavior of SPAs via stress–strain measurements (Figure 2b,c). The curves exhibited distinct mechanical characteristics depending on the BEA/AA molar ratio: the Young's modulus and tensile strength significantly decreased with increasing BEA content, while ductility increased. Notably, the toughness reached a maximum at a molar ratio of 1: 1 (SPA‐2), and further increasing the BEA content led to a decline in toughness. Specifically, this trend aligns closely with the shear adhesion strength measured for SPAs, indicating a correlated evolution of bulk mechanical and adhesive properties (Figure 2f).

As anticipated, the longer alkyl chains of BEA led to a significant reduction in T _g, which was confirmed by differential scanning calorimetry (DSC) (Figure 2d; Figure S3). Notably, we selected a formulation (molar ratio of BEA: AA = 1: 1, SPA‐2) to further study, as polymers with a glass transition temperature (T _g) slightly below room temperature exhibit high chain friction and energy dissipation. This design choice not only enhances mechanical performance but also facilitates an accessible adhesion switching window in subsequent bonding‐debonding cycles [31, 32].

Furthermore, using the time–temperature superposition (TTS) principle with a reference temperature of 20°C (Figure 2e) [33], we obtained rheological master curves for SPA‐2. These results revealed a transition from the glassy state to the rubber state with increasing temperature or decreasing frequency. Based on the Eyring model, an activation energy of 115 kJ·mol⁻¹ was derived (Figure S4) [34], which can be interpreted as the energy barrier for the dissociation of high‐density hydrogen bonds during segmental yielding. According to these preliminary results, we inferred that SPAs with good mechanical properties can function as smart adhesives with a low‐temperature switching window.

2.2. Interfacial Phase‐Locking Enhances Adhesion Strength

After obtaining the SPAs, we proceed to investigate their phase‐locking properties. As shown in Movie S1, the 3D‐printing flower based on SPA‐2 can rapidly switch its shape through thermal stimuli due to the low T_g value of SPA‐2 (Figure S5). The T_g value of SPA‐2 is slightly below room temperature, which provides a low‐temperature switching window for phase‐locking behavior, as discussed above.

To systematically evaluate the effect of phase‐locking behavior on adhesion, both smooth and rough glass substrates were employed. For adhesives without phase‐locking properties, although their intermolecular interactions at the atomic scale are inherently strong, macroscopic objects often fail to achieve robust bonding (SPA‐0). In fact, adhesion strength typically decreases as the substrate surface becomes rougher (Figure S6), a phenomenon known as the “adhesion paradox” [35, 36]. To overcome this paradox, maximizing the effective contact area between the adhesive and the substrate through interfacial phase‐locking‐enabled conformal contact can be an effective strategy [37].

For example, lap‐shear tests were conducted on both rough and smooth glass substrates to evaluate the interfacial adhesion performance of SPA‐2 (Figure S7), wherein SPA‐2 exhibited a pronounced enhancement in adhesion strength on rough glass substrates. In contrast, the adhesion performance of SPA‐0 (without phase‐locking properties) primarily relies on the supramolecular network formed through non‐covalent interactions, resulting in relatively low adhesion (Figure S6). For SPA‐2, its interfaces contained microscopic voids in the glassy state due to incomplete conformal contact, analogous to mismatched mortise‐and‐tenon joints in woodworking (Figure S8). Upon heating above its T _g value, SPA‐2 transitions to a low‐modulus state, which enables it to fully spread across both substrate surfaces under moderate pressure, analogous to inserting a soft tenon into a mortise. Subsequent cooling restores the glassy state, resulting in mechanical interlocking that mimics a firmly joined mortise–tenon structure. This phase‐locking mechanism ensures optimal interfacial contact and significantly enhances adhesion (Figures S9 and S10), providing an effective strategy to overcome the classical adhesion paradox.

In addition, we examined the adhesive behavior on rough Al₂O₃ ceramic substrates (Figure 2f). SPA‐2 exhibited the highest adhesion strength (15.36 MPa), which is consistent with the toughness trend observed earlier (Figure 2b,c). We further conducted lap‐shear tests on various substrates (Figure S11). Adhesion strengths of SPA‐2 on zirconia, stainless steel, copper, glass, polyethylene (PE), and polytetrafluoroethylene (PTFE) substrates were 10.10 MPa, 8.80 MPa, 8.30 MPa, 7.51 MPa, 0.30 MPa, and 0.18 MPa, respectively (Figure S12). The highest shear adhesion strengths of different SPA samples were obtained on Al₂O₃ at room temperature. These results demonstrate the robust and broadly applicable high adhesion performance of SPAs across diverse surfaces.

To testify the effect of phase‐locking behavior on adhesion, pull‐off tests have also been carried out for SPA‐0 and SPA‐2 under the optimal preload we have determined, respectively (Figure 2g; Figure S2). Compared with SPA‐0, SPA‐2 with phase‐locking properties can greatly improve adhesion strength from 0.62 MPa to 3.25 MPa in these pull‐off tests. Notably, SPA‐2 achieved a pull‐off strength of 3.25 ± 0.35 MPa and a lap‐shear strength of 15.36 ± 1.03 MPa, surpassing nearly all previously reported switchable adhesive materials (Figure 2h; Figure S13 and Table S1) [22, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47]. Meanwhile, SPA‐2 can effectively reduce its pull‐off adhesion strength to 0.20 MPa at 60°C, wherein the switching ratio can be calculated as about 16, simultaneously exhibiting a low‐temperature switchability (Figure S14).

To further demonstrate practical adhesion performance, a weight‐lifting test was conducted by using glass substrates bonded with SPA‐2. An adhesive area of approximately 3 cm² successfully supported a 20 kg weight, underscoring the exceptional load‐bearing capacity of our SPA‐2 (Figure 2i; Movie S2). In addition, repeatability and practical reusability of SPA‐2 have been investigated by both lap‐shear and pull‐off tests (Figure S15).

Hence, these results suggest that interfacial phase‐locking is established at rough interfaces through heating and cooling cycles. Our SPAs can overcome the classical adhesion paradox on rough substrates via phase‐locking‐enabled architecture, thereby demonstrating the enhancement of adhesion performance arising from phase‐locking behavior.

2.3. Dynamic Phase Separation

To explore the reason why our SPAs possess ultrahigh adhesion strength, we proceeded to investigate the balance between interfacial adhesion and cohesion in this section. For SPA‐2 locked on rough surfaces, the applied load is more uniformly distributed across the interfacial contact area during detachment, which leads to a gradual increase in adhesion strength as the apparent contact area grows (Figure 3a; Figure S16) [37]. In contrast, stress is highly concentrated at the crack edge for rubbery adhesives, resulting in low adhesion strength. Hence, the cohesive energy of SPAs plays a critical role in determining the overall adhesion strength, in addition to the interfacial energy enhanced by phase locking.

FIGURE 3.

(a) Schematic comparing adhesive interfaces in the rubbery state and the phase‐locking state during peeling. (b) Schematic diagram of a SPA‐based sample for 90° peel testing, with a woven cotton fabric directly embedded as a stiff backing during polymerization. (c) 90° peeling force–displacement curves of the SPA‐0 and SPA‐2 adhered to a smooth glass substrate (peeling speed: 10 mm· min⁻¹). (d) The interfacial toughness values of the SPA‐0 and SPA‐2. (e) Residue remaining on the substrate after peeling reveals the transition of the adhesive failure mode from interfacial debonding in SPA‐0 to cohesive failure in SPA‐2. The sample shown was 5 mm in width. (f,g) Measured peeling forces and the corresponding interfacial toughness values of the SPA‐2 adhered to a rough glass substrate at different peeling speeds. (h) Photos of SPA‐2 being stretched to 40% and 280% strains at the strain rate of 0.05 s⁻¹. The sample shown was 5 mm in width. (i) Stress–strain curves of SPA‐2 at different strain rates. (j,k) AFM phase images of SPA‐2 before and after stretching under 400% strain.

Then, we carried out 90° peeling tests to investigate cohesive behavior, wherein cotton‐backed samples were fabricated as shown in Figure 3b. Both SPA‐0 and SPA‐2 were examined on a glass substrate by heating with a hot‐air gun. These samples were cured for 6 h prior to testing, and these results are shown in Figure 3c,d. For SPA‐2, the measured interfacial toughness reached 2100 J·m^‒2, which is substantially higher than that of rubbery SPA‐0. More importantly, neither clean detachment nor complete cohesive failure was observed for SPA‐2 after peeling (Figure 3e), implying a balanced contribution of interfacial adhesion and bulk cohesion. For SPA‐0, despite much lower toughness than that of SPA‐2 (Figure S17), no adhesive residue was left on the substrate after peeling, demonstrating low interfacial adhesion. These results indicate that SPA‐2 exhibits both high cohesion and interfacial adhesion, which are strengthened by intermolecular interactions within the polymer network and abundant physical interactions with the substrate, respectively. This combination facilitates effective stress transfer from interfacial debonding to bulk tearing during adhesion failure.

Furthermore, the contribution of intermolecular interactions to cohesion was evaluated by testing the interfacial toughness of SPA‐2 at different peeling rates (Figure 3f,g). As the rate increased from 2 to 10 mm·min^‒1, the toughness rose from 1600 to 2100 J·m^‒2. This rate‐dependent behavior can be attributed to strain‐induced dynamic nanophase separation (Figure 3h; Figure S18 and Movie S3). Specifically, SPA‐2 remained optically transparent at low strain rates, whereas higher rates induced a light‐scattering bluish tint, indicating the formation of nanoscale phases under rapid deformation. The adhesion strength declined when the peeling speed was further increased to 20 mm·min^{− 1}, wherein glassy behavior restricts effective energy dissipation by strongly immobilizing polymer chains. The stress‐strain curves of SPA‐2 can clearly reflect its glassy behavior under high strain rates (Figure 3i; Figure S19). Collectively, dynamic nanophase separation can effectively tune the adhesion performance of our SPAs via enhancing cohesion.

To further probe the nanophase separation in our SPA‐2, we explored its hierarchical heterogeneous network using an atomic force microscope (AFM). As shown in Figure 3j,k, upon stretching to 400% strain, SPA‐2 exhibited more pronounced phase separation aligned along the stretching direction compared to the undeformed state, highlighting the strain‐induced formation of nanoscale phase domains.

Hence, these results demonstrate that the adhesion performance of SPA‐2 can be readily tuned by strain‐induced dynamic nanophase separation. In fact, the nanophase separation phenomenon can be attributed to the high density of hydrogen bonds in SPA‐2, which is a key factor in constructing our ultrahigh‐adhesion‐strength SPAs. This phenomenon substantially enhances bulk cohesive energy and balances interfacial adhesion and cohesion, thereby enabling excellent comprehensive adhesion performance.

2.4. Molecular Mechanism for Ultrahigh Adhesion and Switchability

In this section, we proceeded to elucidate the molecular mechanism underlying the ultrahigh adhesion and switchability of our SPAs, which primarily arises from the synergy of dynamic phase‐locking and stress‐induced phase separation (Figure 4a).

FIGURE 4.

(a) Schematic of hydrogen bonding evolution: intramolecular bonds within the polymer and interfacial bonds at the adhesive‐substrate interface during heating and cooling. (b) 2D SAXS patterns of unstretched and rapidly stretched films (400%, 25°C, 0.015 s⁻¹). (c) 1D SAXS profiles of unstretched and rapidly stretched films, showing scattering contributions along the stretching (horizontal) and perpendicular (vertical) directions. (d) Temperature‐dependent low‐field ¹H NMR spectra. (e) Low‐field ¹HNMR T ₁−T ₂ map of SPA‐2 at 130°C. (f) Temperature‐variable FT‐IR spectra of SPA‐2 upon heating from 11 to 90°C (interval: 1°C). (g,h) 2DCOS synchronous and asynchronous spectra generated from (f). In 2DCOS spectra, red colors indicate positive intensities, while blue colors represent negative intensities.

First, we employed small‐angle X‐ray scattering (SAXS) to discern the heterogeneous polymeric network. As shown in Figure 4b,c, unstretched SPA‐2 exhibited a circular, symmetric scattering pattern, whereas the rapidly stretched sample exhibited a distinct, elongated ellipsoidal shape, suggesting the formation of phase‐separated hard domains aligned along the stretching direction [48, 49]. Scattering profiles (1D) were obtained by sectoral integration along the stretching (horizontal) and perpendicular directions, which were further compared with those of the unstretched sample (Figure 4c). Specifically, the scattering intensity in the low‐Q region increased from the unstretched sample to the stretched sample with perpendicular direction, with the highest intensity observed along the stretching direction. This indicates the development of a phase‐separated domain structure oriented along the stretching direction during rapid stretching, wherein the radius of gyration (R _g) is about 39 nm (Figure S20). In other words, the morphology of SPA‐2 evolves from a dense network into a cross‐linked, island‐like structure due to nanophase separation. Therefore, these hard nanodomains can function as geometric nanoconfinements, substantially enhancing the toughness of SPAs during deformation and significantly contributing to the cohesion as well as resultant adhesion strength [50, 51].

To get deeper insight into the molecular mechanism associated with phase‐locking and nanophase separation, low‐field ¹H NMR was employed to monitor changes in hydrogen atom mobility via spin–spin relaxation time (T ₂) [52]. Low‐field ¹H NMR can effectively probe hydrogen atom mobility through T ₂ measurements, where longer T₂ values correspond to higher mobility [53, 54]. For example, only a single T₂ peak was observed for SPA‐2 below its glass transition temperature, indicating restricted chain mobility (Figure 4d). Increasing the temperature to 40°C markedly enhanced the relaxation time, revealing partial dissociation of intermolecular hydrogen bonds and a transition of polymer chains from a constrained disordered state to a rubbery state. In this state, SPA‐2 can readily conform to and wet various substrates, while hydrogen bonds can reform at the interface to promote adhesion. Upon returning to room temperature, the polymeric chains become immobilized, fixing the macroscopic shape of SPA‐2 and mechanically locking the interface between SPA‐2 and substrate, thereby enhancing the resultant adhesion strength. Collectively, these results provide direct molecular‐level evidence for temperature‐dependent switching of adhesion mediated by phase‐locking behavior.

As the temperature further increases to 90°C, two additional T ₂ components emerge. According to these results in Figures S21 and S22, the component with the shorter relaxation time corresponds to poly[2‐[[(butylamino)carbonyl]oxy]ethyl acrylate] (PBEA), while the other is attributed to oligomers. In the corresponding 2D T ₁–T ₂ plot at 130°C (where T ₁ is the spin–lattice relaxation time), all three components are clearly resolved (Figure 4e). For T ₁–T ₂ plots, a lower T ₁/T₂ ratio indicates higher molecular mobility, and the diagonal line represents a completely mobile liquid state [53]. SPA‐2 has a very high T ₁/T₂ ratio (149.0), which is significantly higher than PBEA (18.9) and oligomers (2.7). These results indicate that SPA‐2 forms a distinct, less mobile phase within the polymeric matrix.

To further investigate the molecular mechanism of SPA‐2 during adhesion switching, variable‐temperature transmission infrared spectroscopy was employed to discern dissociation of hydrogen bonds from 11 to 90°C (Figure 4f). Meanwhile, 2D correlation spectroscopy (2D‐COS) was obtained from Figure 4f to analyze the sequence of thermal response associated with changes in polymer interactions (Figure 4g,h). According to Noda's rules (Table S2) [55], the following thermal‐response sequence of different carbonyl species was identified (→ indicates earlier response): 1740 → 1697 → 1758 → 1705 → 1675 cm⁻¹, which corresponds to: v(COOH, free, PAA) → v(COOH, oligomeric H‐bonds, PAA) → v(C = O, free, PBEA) → v (COOH, dimer) → v(─NH─C═O, carboxylic acid–urethane H‐bonds). As the earliest‐responding disordered hydrogen bonds dissociate, more free dangling polymer chains are generated, which not only enhances interfacial adhesion but also promotes better surface conformality due to the low‐modulus state. Notably, these effects contribute to the high adhesion of SPAs prior to the shape‐locking process. Upon returning to lower temperatures, the high density of reassociated hydrogen bonds strengthens both interchain connectivity and energy‐dissipating capability of the polymeric networks. More importantly, these rapid and pronounced thermal‐response components also act as a driving force for stress‐induced phase separation, wherein hydrogen bonds cannot dissociate rapidly and instead self‐aggregate into nanoscale hard domains under strain. Therefore, the reversible dissociation and reassociation of these intermolecular interactions across temperature cycles enable both strong adhesion and switchable behavior in our SPAs.

Based on these results, we conclude that SPA‐2 exhibits enhanced chain mobility and an abundance of free carboxylic groups and dangling chains at elevated temperatures, which can facilitate interfacial spreading and strong adhesion. As the temperature decreases below the phase‐locking condition, the reformation of high‐density hydrogen bonds enhances the mechanical integrity of the polymer and ensures effective phase fixation. In this state, SPA‐2 exists as a highly homogeneous and compatible system. Meanwhile, the formation of nanoscale phase separation under applied stress reinforces the physical strength of the adhesive and thus enhances interfacial toughness. This unique combination of phase‐locking‐assisted behavior and strain‐induced nanophase separation gives rise to the unprecedented adhesion strength observed in our SPAs.

2.5. Applications

To demonstrate the high adhesion strength and switchable properties of SPAs, we designed a mushroom‐inspired adhesive micropillar array using 3D printing. This architecture leverages the error‐compensating effect of micro/nanostructures in gripping, enabled by their inherent adhesion‐switching capability [56, 57]. Conventional replica molding methods present several limitations (Figure 5a), including: (i) the need for prefabrication of single‐use molds; (ii) potential damage to soft or tacky materials during demolding; and (iii) geometric constraints for structures with larger bases than tips [58]. To overcome these limitations, we formulated a photocurable ink from the SPA precursor with Sudan I as a photoabsorber. Using digital light processing (DLP) 3D printing (Figure 5b), we endowed SPA surfaces with precisely defined, high‐aspect‐ratio micropillar arrays (Figure 5c; Figures S23 and S24). Compared with traditional molding methods, this approach offers superior controllability and reproducibility.

FIGURE 5.

(a) Conventional replica molding methods present several limitations. (b) Schematic diagram of DLP‐based 3D printing process for an adhesive surface with mushroom‐shaped micropillar arrays. (c) SEM images of mushroom‐shaped micropillar arrays. Scale bars are 400 um. (d) Compensation effect of mushroom‐shaped micropillar arrays on contact interface error and the enhancing effect of phase‐locking on adhesion switch. (e) Schematic diagram of the temperature‐control system. (f) Working process of pick‐and‐place gripper based on adhesive micropillar arrays.

At temperatures below the glass transition (T _g), phase‐locking enhances adhesion, whereas heating allows shape recovery of the micropillar tips driven by the phase‐locking effect and phase recovery (Figure 5d) [56]. The printed adhesive surface was integrated with a customized temperature‐control system comprising a copper heat sink, coolant channels, a fan, and a controlled heating unit, enabling rapid thermal cycling (Figure 5e; Figure S25). This system was assembled on a robotic arm to fabricate a versatile pick‐and‐place gripper (Figure S26). As shown in Figure 5f, the gripper can readily manipulate a glass three‐neck flask on demand (Movie S4). Specifically, the SPA‐based surface was heated to above T _g (50°C), allowing the micropillars to conform to the curved surface of the object. After positioning the gripper, a gentle preload was applied to ensure full interfacial contact. The adhesive micropillar array was cooled to activate phase‐locking and maximize adhesion strength (pressing and cooling for 247 s). Finally, the flask was transported to the target location and released by reheating triggers (heating for 41 s). More importantly, the 3D‐printed SPA‐based adhesive micropillar array exhibits strong and adaptable adhesion to diverse materials and non‐planar geometries (Figure S27), significantly outperforming conventional planar adhesive interfaces.

Additionally, our SPA‐2 exhibits a high damping factor (tan δ > 1) under a wide frequency range (Figure S28), high stiffness (Young's modulus = 723 MPa, Figure 2c), and strain‐rate‐hardening properties (Figure 3i). Combined with its switchable adhesion mentioned above, SPA‐2 can work as constrained damping materials in various fields, such as the shipbuilding industry [59]. The comprehensive performance can greatly expand its potential in many practical applications.

3. Conclusion

In summary, we have designed and prepared a series of switchable adhesives with strong adhesion by mediating dynamic phase‐locking based on a high density of hydrogen bonds. Owing to the dynamic interfacial mechanical locking and strain‐induced phase separation enabled by hydrogen bonds, a balance between cohesion and interfacial adhesion is achieved, simultaneously yielding high adhesion strength and excellent switchability. As a result, our SPAs overcome the long‐standing trade‐off between high adhesion strength and reversibility in smart adhesives, enabling both high‐strength adhesion and high reversibility for potential applications in various fields. To further support our findings, various macroscopic adhesion tests and a versatile pick‐and‐place gripper were conducted to demonstrate the high adhesion strength (up to 15 MPa) and practical application, respectively. In addition, SPAs possess excellent constrained damping performance, which can expand their applications. More importantly, we present our SPAs as a model system to gain a general understanding of the molecular mechanism behind the high adhesion strength and switchability. We anticipate that our work could facilitate the advancement and application of switchable adhesives in many burgeoning fields.

4. Experimental Section

4.1. Synthesis of 2‐[[(butylamino)carbonyl]oxy]ethyl Acrylate (BEA)

2‐Hydroxyethyl acrylate (100.0 mmol, 11.6 g) was mixed with n‐butyl isocyanate (101.0 mmol, 10.0 g). Then, 2 drops of catalyst (DBTDL) were added to the mixture. The solution was stirred at 85°C under N₂ atmosphere overnight. The crude product was purified by flash column chromatography (methanol/dichloromethane, 1:10 v/v) to afford BEA as a yellow oil (20.2 g, yield, 95%). The ¹H NMR spectrum of BEA is shown in Figure S1. ¹H NMR (CDCl₃, room temperature, 400 MHz) δ (ppm): 6.39−6.34 (m, 1H), 6.11−6.03 (m, 1H), 5.80−5.77 (m, 1H), 4.71 (s, 1H), 4.28−4.21 (m, 4H), 3.13−3.08 (m, 2H), 1.44−1.37 (m, 2H), 1.31−1.22 (m, 2H), 0.84 (t, J = 7.3 Hz, 3H).

4.2. Synthesis of SPAs

SPAs were synthesized by the radical copolymerization of AA and BEA. Typically, for the 1: 1 molar ratio of AA to BEA, 0.360 g of AA (5.0 mmol), 1.076 g of BEA (5.0 mmol), and 0.004 g of TPO (photo‐initiator, 0.25 mol% with respect to the total monomer) were mixed. The mixture solution was then injected into a rectangular mold with silicone rubber as the spacer. It was photopolymerized for 30 min to form the elastomer. The copolymer elastomer with other molar ratios of AA to BEA was synthesized with the same procedure by varying feed ratios.

4.3. Lap‐Shear Tests

Lap‐shear tests were performed on a universal testing machine (UTM2103, Shenzhen Suns Technology) at 20 mm· min⁻¹. Binder clips were utilized to hold the lap joint specimens (~200 kPa) during heating (with a hair dryer for 30 s) and held by binder clips for 6 h at 20°C before testing. SS means stainless steel. The rough glass samples were polished with sandpaper (240‐mesh or 600‐mesh) for 10 min. Then, the rough glass substrates were adequately cleaned with tap water and dried naturally. 240‐mesh sandpaper treatment is used for the rough glass substrate unless otherwise specified.

4.4. Pull‐off Tests

Pull‐off tests were performed on a universal testing machine (UTM2103, Shenzhen Suns Technology) at 20 mm· min⁻¹. The pull‐off adhesion strength was measured using a custom setup with smooth glass slides mounted on a temperature‐controlled test platform and a mechanical sensor. Samples were punched into a wafer with a radius of 3 mm and placed on the platform. After heating with a hot‐air blower for 30 s, a preload of 5 N was applied according to Figure S2. The stage was then set to the target temperature and held for 6 h before the pull‐off tests.

4.5. 90° Peeling Test

A woven cotton fabric was used as the stiff backing for the 90° peeling test. During the process of sample preparation, the fabric was first placed on the bottom of a mold consisting of two transparent PET films separated by a 1 mm thick silicone spacer. The SPAs precursor was then poured into the mold and slowly infiltrated the fabric. Two glass plates were further used to fix the mold, and the precursor was polymerized under UV light for 30 min. Then, the obtained elastomer with the fabric backing was cut into strips and applied onto different substrates. Binder clips were utilized to hold the lap joint specimens (~200 kPa) during heating (with a hair dryer for 30 s) and held by binder clips for 6 h at 20°C before testing. The measured peeling force reached a plateau when the peeling process reached a steady state. The interfacial toughness was calculated as Γ = F_S/W, where F_S is the steady‐state peeling force, and W is the width of the sample.

4.6. Low‐Field NMR Measurements

¹H NMR probes were used to measure the T ₂ values of the elastomer. Temperature‐variable low‐field ¹H NMR spectra were collected from 0 to 130°C with an interval of 10°C. 2D low‐field NMR measurements were tested using T ₁/T ₂ correlation pulse sequence (IR‐CPMG).

4.7. Temperature‐Variable IR Measurement and 2D Correlation Spectroscopy (2DCOS)

To acquire temperature‐variable IR spectra, the sample was positioned between two ZnSe tablets. These samples were compressed into an ultrathin film for transmission IR measurements. For the acquisition of temperature‐variable IR spectra, two tablets were sealed with Parafilm and heated at a temperature gradient from 11°C to 90°C (1°C· min⁻¹). The acquired data were processed to obtain 2D correlation analysis with the 2D Shige ver.1.3 software (Shigeaki Morita, Kwansei Gakuin University, Japan, 2004–2005), and related results were visualized as contour maps. In these maps, the intensity of positive correlations is represented by red colors, whereas negative correlations are depicted in blue.

4.8. Small‐Angle X‐ray Scattering (SAXS)

The film was stretched horizontally, with its surface plane perpendicular to the X‐ray beam. The sample‐to‐detector distance was fixed at 1.7 m. The SAXS patterns were first normalized to an absolute scale, then azimuthally averaged to derive intensity profiles, and finally corrected by subtracting the empty background. For anisotropic samples (stretched films), 2D data were azimuthally averaged over 40° sectors (−20° to 20° and 70° to 110°) along the horizontal and vertical directions to isolate scattering contributions parallel and perpendicular to the stretching direction. 1D scattering data were fitted using the Beaucage model in Igor Pro software for size analysis.

4.9. Statistical Analysis

All adhesion tests were repeated independently with similar results at least three times to obtain the standard deviation (error bars).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: adma73976‐sup‐0001‐SuppMat.docx.

Supporting File 2: adma73976‐sup‐0002‐MovieS1.avi.

Supporting File 3: adma73976‐sup‐0003‐MovieS2.avi.

Supporting File 4: adma73976‐sup‐0004‐MovieS3.avi.

Supporting File 5: adma73976‐sup‐0005‐MovieS4.avi.

Data Availability Statement

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