Shape-memory polymers (SMPs) are a class of smart materials capable of undergoing large deformations, fixing into a temporary shape, and subsequently recovering their original, permanent shape in response to an external stimulus such as heat, light, electricity, or chemical exposure.[1] This shape-memory effect (SME) is enabled by a dual-domain molecular architecture consisting of permanent netpoints that maintain the original structure and temporary switching segments that allow deformation and fixation, typically through phase transitions like glass transition (T_g) or melting (T_m).[2] Discovered in the mid-20th century with foundational polymer principles outlined by Paul J. Flory in 1953, SMP research has accelerated since the 1980s, particularly with polyurethane-based systems commercialized in the 1990s for applications requiring adaptive responses.[3]Key properties of SMPs include their lightweight nature, low cost, and high elasticity, often exhibiting recoverable strains exceeding 200%—far surpassing those of shape-memory alloys—along with tunable transition temperatures and biocompatibility in certain formulations.[1] These materials demonstrate excellent shape fixity (typically >95%) and recovery ratios (up to 98%), enabling repeated cycles of deformation and restoration without significant degradation, though performance depends on crosslinking density and stimulus intensity.[4] Additionally, SMPs offer versatility in processing, such as through extrusion, molding, or 3D printing, and can be engineered for biodegradability or multifunctionality, like self-healing, making them environmentally adaptable compared to rigid metallic alternatives.[2]SMPs are categorized by their composition and response mechanisms, including thermally activated types (e.g., semi-crystalline polyurethanes), light-responsive variants incorporating photothermal fillers like carbon nanotubes, and multi-stimuli systems that combine responses for complex behaviors such as one-way, two-way, or triple-shape memory.[5] Chemically, they encompass thermosets (covalently crosslinked for stability) and thermoplastics (physically crosslinked for recyclability), with emerging blends like polylactic acid/polyurethane composites enhancing mechanical strength and recovery.[4] Liquid crystalline elastomers represent a specialized subclass, leveraging ordered mesophases for precise actuation under lower stimuli thresholds.[1]Notable applications of SMPs span biomedical fields, such as self-expanding stents, drug delivery systems, and tissue scaffolds that adapt to body temperature; aerospace engineering, for deployable structures like solar arrays and morphing wings; and consumer products, including smart textiles and 4D-printed actuators for soft robotics.[2] In recent advancements, nanocomposites with magnetic nanoparticles enable remote actuation via fields, broadening uses in minimally invasive surgery and adaptive optics, while ongoing research focuses on sustainability through bio-based feedstocks to address environmental concerns.[5]
Introduction
Definition and History
Shape-memory polymers (SMPs) are a class of smart, stimuli-responsive materials capable of exhibiting the shape-memory effect (SME), whereby they recover their permanent shape from a temporary deformed state upon application of an external stimulus, most commonly heat.[6] This dual-shape capability arises from their molecular architecture, which includes netpoints that serve as permanent shape fixers—typically chemical crosslinks, crystalline domains, or interpenetrating networks—and switching segments that enable temporary shape fixation and recovery through reversible transitions, such as glass or melting points.[6][7] Unlike shape-memory alloys, SMPs offer advantages like lightweight construction, low cost, large recoverable deformations (up to 400%), and tunable properties, making them suitable for applications in actuators, biomedical devices, and adaptive structures.[8]The origins of SMPs trace back to a 1941 United States patent by Vernon and Vernon, who described heat-shrinkable properties in methacrylic ester resins for dental applications, marking the first documented observation of shape recovery in polymers.[6][9] In the 1960s, heat-shrinkable polyethylenes, such as films and tubings, were developed and commercialized for practical uses, laying further groundwork for polymer memory effects.[6] Systematic development of synthetic SMPs began in the 1980s in Japan, where researchers including Shunichi Hayashi and colleagues pioneered polyurethane-based formulations that demonstrated reliable thermo-induced SME, with key contributions from studies on segmented polyurethanes exhibiting high recovery rates.[6][10] These efforts built on the earlier patents but focused on advancing controllable shape programming for practical use.[11]Commercialization accelerated in the 1990s, with companies like Mitsubishi Heavy Industries introducing polyurethane SMPs for industrial applications such as smart textiles and seals, and mnemoScience GmbH launching biomedical-grade variants, including self-tightening sutures, marking the transition from research to market viability.[6][12] The 2000s saw expansion to multi-stimuli responsive SMPs, incorporating triggers like light, electricity, and pH alongside heat, as exemplified by nanocomposite integrations for enhanced actuation in aerospace and robotics.[6][13] Entering the 2020s, research has emphasized bio-based and sustainable SMPs, utilizing renewable feedstocks like plant oils and vanillin to address environmental concerns while maintaining performance.[6][7]
Fundamental Principles
Shape-memory polymers (SMPs) exhibit a distinctive two-stage process known as the shape-memory effect (SME), which enables them to switch between a permanent shape and a temporary shape in response to an external stimulus. In the programming stage, the polymer is deformed at a temperature above its transition temperature (T_trans), allowing the material to be reshaped into a temporary configuration; subsequent cooling below T_trans fixes this deformed state through mechanisms such as vitrification of glassy domains or crystallization, effectively locking the temporary shape while the permanent shape is preserved by stable network structures.[14] This fixation relies on the temporary suppression of chain mobility, creating a metastable state that can be maintained without external constraints.The recovery stage occurs when the stimulus is reapplied, raising the temperature above T_trans to soften the switching segments, thereby releasing the stored deformation and allowing the polymer to revert to its original permanent shape through elastic recovery driven by the netpoints. At the molecular level, SMPs consist of a polymer network architecture featuring permanent netpoints—such as covalent crosslinks or physical interactions in hard segments—that anchor the permanent shape and provide structural integrity. In contrast, temporary switching segments, often comprising glassy or crystalline domains, enable the fixation and release of the temporary shape by undergoing reversible transitions in mobility.[14] This dual-component design ensures that the SME is a reversible process, with the cycle repeatable over multiple iterations.The general cycle of the SME can be described as follows: the polymer begins in its original permanent shape, is then deformed and cooled to adopt a temporary shape, and upon restimulation, recovers to the permanent shape, completing the loop from original → deformed → temporary → recovered. A fundamental prerequisite for the SME in all SMPs is entropy elasticity, which serves as the driving force for recovery; during deformation, polymer chains adopt higher-energy, lower-entropy conformations, and upon release of the stimulus, the chains spontaneously recoil to their more probable, lower-energy state to maximize entropy, thereby restoring the permanent shape.[14] This entropic mechanism distinguishes SMPs from other responsive materials and underpins their ability to store and release macroscopic deformations efficiently.
Material Properties
Thermal and Mechanical Properties
Shape-memory polymers (SMPs) exhibit temperature-dependent thermal properties that are fundamental to their functionality, with the glass transitiontemperature (Tg) serving as the primary switching mechanism in amorphous systems. For thermal SMPs, Tg typically ranges from 20°C to 100°C, allowing the switching segments to soften and enable deformation into a temporary shape when heated above this threshold, while cooling below Tg fixes the shape through vitrification.[15] In semi-crystalline SMPs, the meltingtemperature (Tm) plays a complementary role in shape fixing, where crystallization upon cooling below Tm provides a stable temporary configuration, as exemplified by polycaprolactone-based SMPs with Tm around 46°C.[16]Mechanically, SMPs demonstrate a pronounced contrast between their glassy and rubbery states, characterized by Young's modulus values that are significantly higher below Tg (often in the GPa range, e.g., approximately 750 MPa at room temperature for certain polyurethane formulations) and much lower above Tg (typically 1-10 MPa for polyurethane formulations), resulting in modulus ratios of 100:1 to 1000:1 that facilitate large deformations. Strain recovery capabilities reach up to 400%, reflecting the material's ability to revert to its original shape upon reheating, with stress-strain hysteresis indicating energydissipation during cycles.[17] Viscoelastic behavior is inherent, manifesting as time-dependent deformation and recovery rates influenced by chain mobility and crosslink density, which can be modulated through fillers to enhance damping properties.[18]SMPs generally possess a low density of approximately 1 g/cm³, offering advantages in weight-sensitive applications compared to metallic shape-memory alloys.[15] Many formulations, particularly polyurethane-based ones, exhibit good biocompatibility, supporting their use in biomedical contexts such as implants and drug delivery systems.[16]
Shape Memory Effect Metrics
The shape memory effect (SME) in polymers is evaluated using standardized metrics that quantify the ability to fix a temporary shape and recover the permanent one, providing insights into material performance and reliability. Key parameters include the shape fixity ratio, shape recovery ratio, recovery time, recovery force, and cycle stability, typically assessed through thermomechanical cyclic testing protocols that involve controlled deformation, cooling, unloading, and heating steps. These protocols, often adapted from dynamic mechanical analysis (DMA) methods, ensure reproducible measurement of SME under varying temperatures and strains, though no dedicated ASTM standard exists for shape memory polymers; general guidelines like ASTM D638 for tensile properties are sometimes incorporated for baseline mechanical characterization.[19][20]The shape fixity ratio (R_f) assesses the stability of the temporary shape following deformation above the transition temperature, cooling, and unloading. It is defined as R_f = \frac{\varepsilon_u}{\varepsilon_m} \times 100\%, where \varepsilon_u is the fixed strain after cooling and unloading, and \varepsilon_m is the maximum applied strain. This metric indicates how effectively the polymer retains the deformed configuration against elastic recoil, with values exceeding 95% signifying robust temporary shape stability in well-designed systems.[19]The shape recovery ratio (R_r) quantifies the extent to which the polymer returns to its original permanent shape upon stimulation, such as heating. It is calculated as R_r = \frac{\varepsilon_u - \varepsilon_f}{\varepsilon_u} \times 100\%, where \varepsilon_f is the final strain after recovery. High R_r values, often above 90%, reflect strong entropic elasticity driving the shape reversion, influenced briefly by the glass transition temperature (T_g) that sets the recovery threshold.[19]Recovery time, the duration required for complete shape reversion, and recovery force, the stress generated during actuation, are critical for practical applications and depend on external factors like heating rate and applied stress during programming. Faster heating rates reduce recovery time by accelerating chain mobility, while higher programming stresses increase recovery force up to several MPa, enabling applications in actuators. For instance, recovery times can range from seconds to minutes at rates of 1–10°C/min, with forces scaling linearly with initial deformation under constant stress conditions.[21][22]Cycle stability evaluates the durability of the SME over repeated programming and recovery cycles, measuring fatigue resistance through consistent R_f and R_r retention. Durable shape memory polymers maintain performance beyond 100 cycles, with advanced formulations like polyimides demonstrating stability over 1000 bending cycles without significant degradation in fixity or recovery ratios. This metric highlights the role of crosslinking density in preventing chain slippage and viscous flow accumulation during extended use.[23]
Advanced Memory Behaviors
Advanced memory behaviors in shape-memory polymers (SMPs) extend beyond the conventional binary shape memory effect, enabling more complex, programmable responses that mimic biological adaptability and facilitate sophisticated applications. These behaviors leverage multiple phase transitions or structural designs to achieve sequential or reversible shape changes, building on the fundamental principles of temporary shape fixation and permanent shape recovery.[24]Triple-shape memory represents a key advancement, allowing polymers to transition sequentially from an original shape (A) through two temporary shapes (B and C) upon controlled stimulation, typically thermal. This is achieved via materials with two distinct transition temperatures, such as separate glass transition (Tg) and melting (Tm) points, enabling a two-step programming process where the polymer is deformed and fixed at different temperatures before recovery in reverse order. The effect was first demonstrated in 2006 using a copolymer network of poly(ε-caprolactone) dimethacrylate and poly(ε-caprolactone) with crystallizable switching segments, achieving near-complete recovery ratios for both steps.[24] Subsequent developments have refined these systems for higher efficiency, with recovery strains exceeding 200% in optimized formulations.[25]Multi-shape memory effects further expand this capability, permitting polymers to store and recover three or more temporary shapes in a single cycle, often up to quadruple or quintuple shapes. These are realized through gradient structures with spatially varying transition temperatures or multiple distinct domains within the polymer network, allowing programmed sequential actuation by gradually increasing the stimulus intensity. For instance, a perfluorosulfonic acid ionomer exhibited quadruple-shape memory by exploiting multiple ionic relaxation modes, with shape fixity ratios above 95% across transitions. Similarly, semi-interpenetrating networks of poly(methyl methacrylate and poly(ethylene glycol) demonstrated quintuple-shape memory via broadened glass transitions combined with crystalline segments, enabling five distinct shape recoveries with minimal hysteresis.[26]The two-way shape memory effect (2W-SME) introduces reversibility, enabling cyclic transitions between two shapes without external stress or reprogramming after initial training. This is primarily driven by mechanisms such as oriented crystallization in semi-crystalline polymers, where cooling induces elongation due to aligned crystal growth, and heating causes contraction via melting, or by incorporating functional particles like magnetic nanoparticles that enable field-induced alignment and recovery. Early demonstrations relied on constant stress, but stress-free variants emerged in the 2010s using liquid crystalline elastomers or aligned domains. Advances in the 2020s have enhanced actuator performance, with composites achieving reversible strains up to 20-30% over hundreds of cycles, facilitated by magnetic particles for remote control and integration into soft robotics.[27][28]Programmable and sequential actuation has gained prominence in the 2020s through integration with 4D printing, where SMP structures are designed to undergo time-dependent, multi-step deformations in response to uniform stimuli. This involves printing gradients in composition or crosslinking density to create domains with varying activation thresholds, allowing complex folding or deployment sequences. For example, multi-material 4D-printed SMP nanocomposites have demonstrated selective actuation of up to four sequential shapes, with recovery speeds tunable from seconds to minutes, enabling applications in adaptive deployable structures.[29] These developments leverage additive manufacturing to embed programmability directly into the geometry, achieving shape changes that evolve over time without additional mechanical inputs.[30] As of 2025, shape memory elastomers have demonstrated rapid shape recovery in under 15 seconds, with shape fixity ratios of 88% and recovery ratios of 98%, expanding applications in dynamic systems.[2]Integration of self-healing with advanced memory behaviors enhances durability, using the shape memory effect to assist in crack closure and subsequent healing. In these multifunctional systems, shape recovery drives deformed regions together to close fissures, promoting contact between healing agents or dynamic bonds for repair. Recent advances feature SME-assisted closure in polymers with dynamic covalent networks, enabling healing of cracks up to several millimeters wide with recovery of over 90% mechanical strength after multiple cycles. This synergy has been demonstrated in biocompatible formulations, where thermal activation simultaneously triggers memory recovery and self-healing, advancing toward robust, long-lifetime materials.[31][32]
Synthesis and Structure
Physically Crosslinked SMPs
Physically crosslinked shape memory polymers (SMPs) rely on reversible non-covalent interactions, such as phase separation in block copolymers or supramolecular associations like hydrogen bonding and ionic bonds, to establish netpoints that preserve the permanent shape while permitting thermoplastic reprocessing. These physical netpoints contrast with permanent covalent bonds by allowing the material to be reshaped multiple times without degradation, facilitating applications requiring adaptability. The shape memory effect in these systems involves deforming the polymer above its transitiontemperature (T_trans), cooling to fix the temporary shape via physical stabilization, and recovering the original form upon reheating.[19][1]A key subclass comprises linear block copolymers, notably ABA triblock architectures, where the hard A blocks aggregate into phase-separated domains serving as physical netpoints, and the soft B block functions as the molecular switch responsive to thermal stimuli. For instance, polystyrene-b-poly(ethylene oxide)-b-polystyrene (PS-b-PEO-b-PS) exemplifies this structure, with polystyrene domains providing mechanicalintegrity and polyethyleneoxide enabling reversible softening. Polyurethane-based block copolymers also feature prominently, utilizing similar phasesegregation for dual-shape functionality.[19][33]Beyond block copolymers, semi-crystalline thermoplastics like polyolefins exhibit physical crosslinking through crystallite formation, which acts as temporary netpoints to lock in deformed shapes upon cooling below the melting temperature (T_m). In these polymers, the crystalline domains provide elasticity recovery above T_m, while amorphous regions contribute to deformability. Poly(ε-caprolactone) (PCL) networks represent an early example of semi-crystalline thermoplastic SMPs, developed in the early 2000s based on foundational SMP research from the 1980s, harnessing PCL's T_m of 55–60 °C for switching.[19][34]The thermoplastic nature of physically crosslinked SMPs confers distinct advantages, including recyclability through melt reprocessing and compatibility with standard fabrication techniques like extrusion and injection molding, which enhance scalability for industrial use.[19][1]
Chemically Crosslinked SMPs
Chemically crosslinked shape memory polymers (SMPs) rely on irreversible covalent bonds as netpoints to establish and maintain the permanent shape, providing enhanced structural stability compared to physically crosslinked systems. These covalent crosslinks form a thermoset network that resists deformation under stress, enabling high-fidelity shape recovery. Synthesis typically involves free-radical polymerization of multifunctional monomers, such as methacrylates or acrylates, initiated by thermal or photoinitiators to create uniform networks. Alternatively, click chemistry reactions, including thiol-ene additions, offer precise control over crosslinking density and are increasingly used for tailored architectures. Recent bio-based syntheses, such as those incorporating itaconic acid-derived monomers, further promote sustainability in thermoset networks.[35][36][37][38]A key class of chemically crosslinked SMPs is segmented polyurethanes, featuring alternating hard and soft segments where the hard segments—often derived from diisocyanates—serve as covalent netpoints for shape fixation, while soft segments, such as polyols, provide the entropic elasticity for temporary deformation. These materials exhibit phase separation that enhances the shape memory effect, with the hard domains anchoring the network and soft domains undergoing glass or melting transitions. The first commercial crosslinked polyurethane SMPs were developed by Mitsubishi Heavy Industries in the early 1990s, marking a milestone in practical applications for aerospace and automotive components.[39][40][41]Poly(ethylene oxide) (PEO)-based crosslinked SMPs, often synthesized by photo-crosslinking PEO macromonomers with diacrylate end-groups, form hydrophilic networks ideal for biomedical uses like stents and drug-eluting implants due to their biocompatibility and water responsiveness. These networks achieve high gel contents (90-94%) and demonstrate two-way shape memory with reversible strains up to 20% at body temperatures around 43-50°C, facilitating minimally invasive deployment.[28]Thermoset chemically crosslinked SMPs provide advantages such as superior chemical resistance—manifesting as swelling rather than dissolution in solvents—and high recovery forces approaching 100%, attributed to the robust covalent architecture that sustains mechanical integrity. However, their irreversible nature leads to non-reprocessability, requiring specialized molding during initial fabrication and limiting recyclability, in contrast to physically crosslinked SMPs that permit re-melting and reshaping.[42][43]In the 2020s, bio-based variants have gained prominence for sustainability, utilizing plant-oil-derived crosslinkers like castor oil multifunctional polyols in polyurethane networks via thiol-ene reactions, achieving shape fixity and recovery rates of 99% and 98%, respectively, at strains up to 200% while reducing petrochemical dependence.[44]
Nanocomposites and Blends
Shape-memory polymer (SMP) nanocomposites incorporate nanoparticles such as carbon nanotubes (CNTs), graphene, cellulose nanocrystals (CNCs), and magnetic particles like Fe₃O₄ to enhance the base polymer network's performance. These fillers, typically added at low loadings (e.g., 1-5 wt.%), act as physical crosslinkers or reinforcements, improving mechanical strength and enabling multifunctionality. For instance, incorporating 5 wt.% Fe₃O₄ nanoparticles into polyurethane-based SMPs increases shape fixity from 82% to 90% by promoting additional crosslinking during deformation, while also allowing magnetic actuation for remote control.[45] Carbon-based fillers like multi-walled CNTs (MWCNTs) at 0.5-3 wt.% impart electrical conductivity, enabling electroactive shape recovery with up to 95% efficiency at low voltages (e.g., 40 V), and piezoresistive properties for self-sensing applications.[45][46] Recent advances in the 2020s, including entangled nanoparticle aggregates, have further boosted robustness, with graphene oxide (GO) hybrids achieving 98% shape recovery and 99% fixity in under 7 seconds under near-infrared (NIR) stimulation due to photothermal effects.[45][47]Processing of SMP nanocomposites primarily involves melt blending for uniform dispersion, in situ polymerization to chemically anchor fillers to the polymer chains, or solution mixing followed by casting, though challenges like agglomeration persist and are mitigated by surface functionalization (e.g., silane coating on silica nanoparticles).[45] In shape memory elastomers reinforced with 10 wt.% silica nanoparticles, tensile strength and modulus increase by approximately 30-50%, while maintaining recovery ratios above 90%, demonstrating enhanced load-bearing capacity without sacrificing the core shape memory effect. Multifunctional benefits extend to self-healing, as seen in polypyrrole (PPy)-doped blends where post-healing recovery supports loads up to 3846 times the material's weight.[45] These hybrids outperform pure SMPs in demanding environments, with 2025 reviews highlighting their role in scalable 4D printing for dynamic structures.[45]SMP blends, formed by mixing two or more polymers such as thermoplastic SMPs with elastomers or plastics, allow tunable glass transition temperatures (Tg) and improved processability through phase-separated morphologies. Common examples include polylactic acid (PLA)/thermoplastic polyurethane (TPU) blends, where increasing TPU content raises the shape-setting temperature by 14% and reduces recovery time by 12%, enabling customization for biomedical or aerospace uses.[5] Compatibilizers like maleic anhydride-grafted PLA (PLA-g-MA) at 2-5 wt.% enhance phase stability in immiscible systems, reducing domain sizes from micrometers to nanometers and boosting interfacial adhesion, which elevates impact strength to 63 J/g in polypropylene (PP)/TPU blends.[5] These blends combine the rigidity of one polymer with the elasticity of another, yielding up to 20% higher tensile modulus (e.g., 57.4 MPa in TPU/PCL vs. 48.4 MPa for pure TPU) while preserving shape recovery above 85%.[5]Melt blending and in situ polymerization are preferred processing routes for SMP blends, offering cost-effective routes to hybrid materials, though dispersion uniformity requires compatibilizers to prevent phase separation-induced weaknesses. In PLA/poly(ε-caprolactone) (PCL) blends, adding polyethylene glycol (PEG) tunes Tg downward for lower-temperature actuation, achieving 86% recovery and 95% fixation when reinforced with 6 wt.% CNTs.[5] Such formulations provide mechanical enhancements without relying on covalent crosslinking alone, addressing limitations in pure systems and supporting applications in soft robotics.[5]
Mechanisms of Shape Memory
Thermally Induced Effect
The thermally induced shape memory effect (SME) in shape-memory polymers (SMPs) is activated by external heat, which triggers a reversible change between a temporary and a permanent shape. When an SMP is heated above its transition temperature—typically the glass transition temperature (Tg) for amorphous polymers or the melting temperature (Tm) for semicrystalline ones—the switching segments soften, allowing polymer chains to regain mobility. This enables entropy-driven recoiling, where the chains spontaneously return to their more disordered, higher-entropy state, recovering the original permanent shape. Upon cooling below the transition temperature, the temporary shape is fixed through immobilization of the polymer chains in the switching domains, locking the deformed configuration in place.[48]Programming the temporary shape involves deforming the SMP while it is in the softened state above the transition temperature and then cooling it under stress to fix the new configuration. Common methods include free recovery, where the material is allowed to revert unconstrained upon reheating, measuring shape recovery ratios often exceeding 90% in well-designed systems; constrained recovery, which applies constant strain during heating to quantify recoverystress, typically reaching several MPa; and cold drawing, where deformation occurs at temperatures below Tg to induce oriented chain alignment for enhanced fixation. These approaches rely on the polymer's dual-domain structure, with netpoints maintaining the permanent shape and switching segments enabling temporary deformation.[49]Efficiency of the thermally induced SME depends on several factors, including uniform heating to ensure consistent chain mobilization across the material volume, as gradients can lead to incomplete recovery. Environmental conditions also play a role; for instance, water uptake in hydrophilic SMPs can depress Tg by up to 30°C with less than 5 wt% absorption, accelerating response but potentially compromising stability in moist environments.[48]A key limitation of thermal triggering is the slow response time in bulk samples, often taking minutes due to heat diffusion limitations, which restricts applications requiring rapid actuation. This can be addressed by incorporating embedded heaters, such as conductive fillers or micro-heaters, to enable localized and faster heating rates.[8]Historically, thermal-based actuation has dominated SMP applications, accounting for the majority of early commercial uses such as heat-shrink tubing and foams, due to its simplicity and reliability.[50]
Thermodynamics of the Effect
The shape memory effect (SME) in polymers is fundamentally an entropy-driven process rooted in the thermodynamics of polymer chain conformations. When a shape memory polymer (SMP) is deformed above its transition temperature in the rubbery state, the molecular chains become extended and oriented, resulting in a significant decrease in conformational entropy (ΔS < 0), while the internal energy change (ΔH) remains relatively small due to minimal bond alterations. This stored entropic energy is captured by the Gibbs free energy equation:\Delta G = \Delta H - T \Delta Swhere ΔG is the Gibbs free energy change, ΔH is the enthalpy change, T is the absolute temperature, and ΔS is the entropy change. In the deformed state, the negative ΔS makes -TΔS positive, stabilizing the temporary shape upon cooling and fixation below the transition temperature. Upon reheating, the increasing T amplifies the -TΔS term, rendering ΔG negative and driving spontaneous recovery to the original, high-entropy configuration as the chains recoil to maximize disorder. This mechanism, analogous to the elasticity of crosslinked rubbers, underpins the macroscopic shape recovery in most thermally activated SMPs, with entropic contributions often accounting for over 90% of the driving force in glassy systems.[51][52]For SMPs relying on a glassy transition as the switching mechanism, the free energy equation for the phase switching between rubbery and glassy states can be derived from the thermodynamics of vitrification. Consider the switching domains, typically the soft segments above the glass transition temperature (T_g), which enable deformation. The process begins with deformation in the rubbery state, followed by cooling to induce vitrification, fixing the temporary shape via frozen chain conformations. The free energy change for vitrification, ΔG_vit, is:\Delta G_\text{vit} = \Delta H_\text{vit} - T \Delta S_\text{vit}where ΔH_vit is the enthalpy of vitrification (typically small, on the order of a few kJ/mol, representing the difference in intermolecular interactions between mobile rubbery and rigid glassy phases), and ΔS_vit is the configurational entropy loss during freezing of segmental motion (negative and dominant below T_g). To derive this, start from the equilibrium at T_g, where ΔG_vit = 0, so T_g = ΔH_vit / ΔS_vit. Below T_g, the glassy state is metastable with lower free energy due to the -TΔS_vit term outweighing ΔH_vit. Upon heating above T_g, devitrification occurs as chain mobility resumes, releasing the stored entropic strain and enabling recovery. This derivation assumes a first-order-like approximation for the second-order glass transition, incorporating the heat capacity jump (ΔC_p) via integration: ΔH_vit ≈ ∫ ΔC_p dT over the transition range, often yielding ΔH_vit ≈ 10-50 J/g for common amorphous SMPs like polyurethanes. The negative ΔS_vit in the fixed state thus provides the thermodynamic impetus for recoiling, with recovery efficiency tied to the sharpness of the transition.[53][54]In crystalline SMPs, such as those based on poly(ethylene oxide) (PEO) segments, phase transitions play a key role in forming physical netpoints that stabilize the permanent shape, with entropy changes during crystallization contributing to the overall SME thermodynamics. Upon cooling after deformation, partial crystallization of the switching segments occurs, reducing chain mobility and fixing the temporary shape through a decrease in entropy (ΔS_cryst < 0). For PEO-based SMPs, the entropy change associated with crystallization is approximately 20-30 J/mol·K, reflecting the ordering of helical chains into crystalline lamellae and the loss of rotational freedom. This ΔS_cryst, calculated as ΔH_fus / T_m (where ΔH_fus is the enthalpy of fusion, ~7-9 kJ/mol per repeating unit, and T_m ≈ 330-340 K), enhances shape fixity by creating temporary crosslinks, while melting upon heating releases this entropic constraint, facilitating recovery alongside the primary entropic driving force from deformation. Such crystalline contributions are particularly pronounced in block copolymers where PEO domains provide dual thermal transitions (T_g and T_m), amplifying the thermodynamic contrast between fixed and recovered states.[55][56]The recovery process also involves overcoming energy barriers associated with chain disentanglement and viscous flow, characterized by an activation energy (E_a) typically ranging from 100-300 kJ/mol in glassy SMPs. This E_a, derived from Arrhenius analysis of recovery kinetics (rate ∝ exp(-E_a / RT)), quantifies the barrier for segmental relaxation above T_g, influenced by crosslinking density and fillers; for example, in polylactide-based SMPs, E_a ≈ 186 kJ/mol, while polyurethane variants show 80-100 kJ/mol. Higher E_a values correlate with slower recovery but greater thermal stability of the temporary shape, balancing responsiveness and durability in applications.[57][58]To predict and optimize the SME under complex loading, finite element analysis (FEA) is employed to model thermo-mechanical coupling, integrating heat transfer, viscoelasticity, and phase transition kinetics. In FEA frameworks, the coupled equations solve for temperature-dependent modulus evolution (e.g., via Prony series for relaxation) alongside thermal diffusion (∂T/∂t = α ∇²T + Q, where α is diffusivity and Q is heat source), capturing how localized heating drives nonuniform recovery strains. User-defined material subroutines in software like ABAQUS implement these, enabling simulation of multi-axial deformation and entropy/enthalpy contributions, as validated for epoxy-based SMPs with recovery strains up to 200%. This approach reveals stress concentrations at phase boundaries and informs design for high-fidelity actuation.[59][60]
Stimuli-Responsive Variants
Shape-memory polymers (SMPs) exhibit responsiveness to a variety of non-thermal stimuli, enabling the shape memory effect (SME) through mechanisms distinct from heat-induced transitions, such as direct molecular reconfiguration or indirect energy conversion. These variants incorporate functional additives like nanoparticles or chromophores to trigger shape recovery remotely or precisely, offering advantages in applications requiring contactless actuation.[16]Light-induced SMPs primarily operate via photothermal conversion, where absorbers such as carbon black, dyes, or carbon nanotubes convert UV, visible, or near-infrared (NIR) light into localized heat, or through direct photoisomerization without significant temperature rise. For instance, azobenzene-containing polymers undergo reversible trans-cis isomerization under UV (e.g., 365 nm) and visible light (e.g., 430 nm), enabling shape changes in hydrogels with up to 99% recovery. NIR activation with gold nanorods or graphene oxide allows remote control, as demonstrated in epoxy vitrimers coated with polydopamine-modified multi-walled carbon nanotubes (MWCNTs), achieving self-healing alongside SME under 808 nm laser irradiation. These systems provide spatiotemporal precision, surpassing thermal methods in accessibility for biomedical uses.[16][61][62]Electro-active SMPs respond to electrical stimuli through Joule heating from conductive fillers like graphene, carbon nanotubes (CNTs), or boron nitride (BN), or via electrostatic forces causing dielectric deformation. Incorporation of 0.26 wt% graphene foam into polyurethane composites enhances conductivity to enable shape recovery in under 60 seconds at low voltages (e.g., 0.8 V/mm), with some systems achieving 100% recovery in 8 seconds using 15 wt% BN fillers. Graphene-enabled variants offer response times on the order of seconds, significantly faster than pure thermal SMPs, due to efficient resistive heating.[16][63][64]Magneto-responsive SMPs utilize magnetic nanoparticles, such as Fe₃O₄, to induce SME via alternating magnetic fields that generate eddy currents or Néel relaxation for localized heating. A polycaprolactone-based polyurethane (PCLU) nanocomposite with Fe₃O₄ achieves 97% shape recovery in 60 seconds under a 20 kHz field, enabling non-contact actuation suitable for in vivo applications. These systems allow precise energy delivery without invasive probes.[16]Chemo-responsive SMPs trigger SME through chemical interactions, such as pH changes or solvent-induced swelling that plasticize the polymer network and lower the glass transition temperature. Polyvinyl alcohol (PVA) hydrogels recover 99% of their shape in water at room temperature within 30 minutes via hydrogen bond disruption, while pH-sensitive polyurethane composites with cellulose nanocrystals respond to acidic or basic environments by altering chain mobility. Hydrogel-based variants excel in swelling-driven recovery for drug delivery.[16][65]Hybrid multi-stimuli systems combine triggers like light and magnetic fields or electro and chemo for enhanced control and two-way SME, where polymers cycle reversibly between shapes without reprogramming. Recent 2025 developments include electro-induced two-way SMPs using liquid crystalline networks under stress-free conditions, achieving bidirectional deformation via combined electric and orientational stimuli. Photoluminescence thermadapt two-way SMPs based on syndiotactic 1,2-polybutadiene with phosphors enable light-triggered reversible shape changes. These hybrids, such as PCL-TPU/MWCNT nanofiber blends responsive to both electric and NIR light, improve precision in complex environments.[16][66][67]Challenges in non-thermal variants include filler aggregation leading to uneven dispersion and reduced mechanical integrity, slow response in solvent-based chemo systems, and energy inefficiency in light or magnetic actuation due to heat dissipation. Biocompatibility remains critical for electro- and magneto-active fillers, necessitating surface modifications to minimize toxicity. Ongoing research focuses on optimizing nanofiller interfaces for scalable, durable performance.[16][62]
Comparisons with Other Materials
Versus Shape-Memory Alloys
Shape-memory polymers (SMPs) are organic materials primarily composed of long-chain macromolecules with cross-linked networks, resulting in densities typically ranging from 0.9 to 1.3 g/cm³, which makes them significantly lighter than shape-memory alloys (SMAs).[68] In contrast, SMAs, such as nickel-titanium (NiTi) alloys, are metallic compositions with densities around 6.45 g/cm³, leading to much heavier structures for equivalent volumes.[69] This density difference highlights SMPs' advantage in weight-sensitive applications, while SMAs provide greater structural robustness due to their metallic nature.[70]A key distinction in performance lies in recovery capabilities: SMPs can achieve recovery strains up to 400%, far exceeding the 8% typical for SMAs like NiTi.[71] However, SMAs generate much higher recovery stresses, often reaching 500 MPa or more, compared to the 5–100 MPa range for SMPs.[70][72] These trade-offs mean SMPs excel in scenarios requiring large deformations with minimal force, whereas SMAs are suited for high-load actuation.[71]SMPs offer greater versatility in stimuli responsiveness, activating via thermal, light, electrical, magnetic, chemical, or biological triggers, enabling multi-stimuli designs.[19] SMAs, primarily NiTi variants, respond mainly to thermal stimuli with austenite finish temperatures typically between 30°C and 100°C, limiting their adaptability without additional modifications.[19] This flexibility in SMPs allows for broader integration into diverse environments compared to the more constrained thermal activation of SMAs.[70]In terms of economics and fabrication, SMPs are more cost-effective, with production costs generally under $50/kg, and they support easy processing methods like injection molding, extrusion, and 3D printing.[70] SMAs, however, are considerably more expensive due to complex alloying and heat treatment requirements, though they can be machined similarly to conventional metals.[70] These factors make SMPs preferable for scalable, low-cost manufacturing, while SMAs demand specialized techniques that increase overall expense.[19]Regarding durability, SMPs demonstrate superior fatigue resistance in low-stress regimes, often maintaining shape fixity and recovery over more than 1,000 cycles in bending or low-strain tests.[23] SMAs, conversely, excel in high-force applications with cycle lives reaching millions under moderate strains, but they suffer degradation faster in extreme deformation scenarios.[73] This positions SMPs for repeated, gentle actuations and SMAs for robust, force-intensive operations.[70]
Versus Other Smart Polymers
Shape-memory polymers (SMPs) differ from hydrogels, another class of smart polymers, primarily in their actuation mechanisms and performance characteristics. SMPs achieve shape recovery through an entropy-driven process, where temporary deformations are fixed by physical crosslinks and recovered via the elastic restoration of polymer chains upon stimulus application, often in dry conditions.[74] In contrast, hydrogels typically respond via solvent-induced swelling and deswelling, where water absorption or expulsion drives volume changes and shape alterations.[75] SMPs generally enable higher recoverable strains, up to 800% in some thermal-responsive variants, making them suitable for large-deformation applications.[76] Hydrogels, however, often exhibit faster response times on the order of seconds due to rapid diffusion processes, though their maximum strains are typically lower, under 2% for ionic variants.[75][76]Compared to piezoelectric polymers such as polyvinylidene fluoride (PVDF), SMPs emphasize programmable shape recovery rather than instantaneous deformation. Piezoelectric polymers generate strain through voltage-induced electric fields, producing converse piezoelectric effects that cause direct mechanical deformation without inherent shape memory; they revert elastically upon voltage removal but do not "remember" programmed temporary shapes.[75] SMPs, by contrast, store and recover complex temporary shapes via phase transitions, enabling one-way or reversible actuation primarily for dynamic recovery tasks.[75] This positions SMPs as actuators for applications requiring large, controlled recoveries, such as deployable structures, while piezoelectric polymers excel in sensors and high-precision actuators due to their fast response times (up to kHz frequencies) and high stiffness, albeit with low strains under 1%.[75][76]In relation to self-healing polymers, SMPs prioritize shape fixation and recovery over damage repair. Self-healing polymers restore integrity through reversible chemical bonds or physical interactions, such as hydrogen bonding or dynamic covalent networks like Diels-Alder reactions, focusing on mending cracks or restoring mechanical properties without altering overall form.[31] SMPs, however, rely on distinct permanent and reversible phases to enable shape programming and stimulus-triggered recovery, with no inherent focus on bond reformation for healing.[31] Since the 2020s, multifunctional hybrids combining both properties have emerged, such as hydrogels integrating shape memory for structural reconfiguration and self-healing for durability, enhancing applications in tissue engineering.[31]SMPs offer distinct advantages in programmability and biocompatibility relative to many rigid smart polymers. Their ability to be programmed with multiple temporary shapes through sequential deformation and fixation allows for complex, multi-stage responses not readily achievable in stiffer materials like piezoelectric polymers.[75] Additionally, SMPs exhibit excellent biocompatibility, supporting cell viability and integration in biomedical contexts, surpassing the rigidity and potential cytotoxicity of alternatives like certain electroactive or piezoelectric variants.[77]A key limitation of SMPs is their slower actuation compared to electroactive polymers (EAPs). Thermal or light-triggered SMPs often require 10 seconds to several minutes for recovery due to heat transfer or phase transition kinetics, whereas EAPs, such as dielectric elastomers, achieve responses in milliseconds via direct electrostatic forces.[76][75] This makes SMPs less ideal for high-speed applications, though electroactive composites can mitigate this by enabling faster Joule heating.[78]
Applications
Biomedical Applications
Shape-memory polymers (SMPs) have found significant application in biomedical fields due to their ability to undergo large deformations and recover to predefined shapes in response to physiological stimuli, such as body temperature, enabling minimally invasive procedures. In cardiovascular interventions, SMP-based stents, particularly those made from polyurethane foams, facilitate self-expansion upon deployment, reducing the need for additional mechanical support and minimizing vessel trauma. For instance, the Igaki-Tamai biodegradable coronary stent, composed of poly-L-lactic acid (PLLA) with shape-memory properties, was clinically introduced in the early 2000s and expands upon heating to approximately 70°C (via a balloon catheter) to support arterial healing over 6-9 months before degrading.[79][80] Similarly, the TrelliX Embolic Coil System, utilizing porous polyurethane SMP foam, has been evaluated in clinical trials (NCT03988062) for cerebral aneurysm treatment, where the foam compresses for catheter delivery and expands to fill the aneurysm sac at 37°C, promoting thrombosis while allowing tissue integration.[77] The U.S. Food and Drug Administration (FDA) cleared Shape Memory Medical's IMPEDE Embolization Plug in 2018, a polyurethane SMP foam device for peripheral vascular occlusion, demonstrating biocompatibility and recovery ratios exceeding 90% in vivo.[81]In orthopedics, SMPs enhance suture and scaffold technologies by enabling shape deployment that improves fixation and reduces surgical invasiveness. Self-knotting sutures made from thermoplastic polyurethane (TPU)/polycaprolactone (PCL) blends contract upon heating to 37-40°C, tightening knots automatically and distributing tension evenly to promote wound healing without manual adjustment.[77] Bone scaffolds incorporating SMPs, such as poly(L-lactide-co-ε-caprolactone) (PLMC) in 8:2 or 9:1 ratios, provide temporary structural support for defect repair, expanding post-implantation to match bone contours and degrading over time to avoid secondary surgeries.[77] PCL/hydroxyapatite (HA) composite scaffolds loaded with bone morphogenetic protein-2 (BMP-2) further stimulate osteogenesis, with shape recovery facilitating cell infiltration and vascularization in rabbit calvarial models.[77] These properties stem from the polymers' net-point architecture, where permanent crosslinks maintain integrity during temporary deformation via hydrogen bonding.[82]For drug delivery, SMP microcapsules offer controlled release mechanisms triggered by the shape-memory effect (SME), ensuring precise payload deployment at physiological temperatures. PCL-based microcapsules, with diameters around 100-500 μm, encapsulate therapeutics and deform to a compact form for injection, then recover to release contents via diffusion or rupture at 37°C, achieving sustained release over days to weeks.[77] This combines biodegradation with SME, as demonstrated in studies where microcapsules loaded with model drugs like ibuprofen exhibited 80-95% recovery and tailored release profiles modulated by crosslink density. Such systems minimize burst release and enhance bioavailability, particularly for localized treatments.In tissue engineering, SMPs enable 4D scaffolds that adapt dynamically to biological cues, supporting cell growth and tissue morphogenesis. Poly(lactic acid)-PCL (PLA-PCL) scaffolds self-expand at body temperature to repair neural tube defects, providing a conformable matrix that integrates with surrounding tissue and promotes axonal regeneration in animal models.[77] Recent advances in 2025 highlight dimensional SME in 4D-printed SMP composites, where scaffolds undergo programmed multiaxial expansion to accommodate volumetric tissuegrowth, as seen in PCL-based structures with recovery strains up to 200%.[83] These scaffolds, often fabricated via additive manufacturing, exhibit biocompatibility with minimal inflammatory response, fostering extracellular matrix deposition.SMPs demonstrate strong biocompatibility, with many formulations approved for clinical use due to low cytotoxicity and tunable degradation rates aligning with healing timelines. FDA-cleared examples include polyurethane SMP foams for hemostatic applications, such as the IMPEDE-FX RapidFill device (IDE approved 2023), which expands to seal vascular sites and supports wound closure by promoting clot formation without adhesives.[84] These foams, with porosities of 90-98%, facilitate oxygen permeation and cell migration, as validated in biocompatibility assays per ISO 10993 standards.[85]
Industrial and Engineering Applications
Shape-memory polymers (SMPs) have found significant utility in aerospace engineering, particularly for deployable structures that require compact storage and reliable activation in space environments. For instance, SMP composites are employed in origami-inspired booms and hinges for antennas and solar arrays, enabling self-deployment upon thermal stimulation without complex mechanical systems.[86] These materials offer a lightweight alternative to shape-memory alloys (SMAs), with carbon fiber-reinforced SMP composites achieving packing efficiencies up to 13% and deployment times as low as 10 seconds at temperatures around 150°C, while weighing significantly less—such as 0.164 kg for a sample versus over 12 kg for equivalent PVC structures.[86] This density advantage reduces launch costs and enhances payload capacity in satellites and habitats.[86]In the automotive sector, SMPs enable morphing components and self-repairing parts, improving safety and maintenanceefficiency. Bumpers and spoilers can be designed to deform on impact and recover their original shape via heat activation, potentially replacing heavier steel elements that constitute over 50% of vehicle composites.[87] For example, polyurethane-based SMPs programmed through deformation and cooling allow for post-accident repair by reheating, addressing the high incidence of vehicledamage—77% of U.S. drivers report at least one accident—while offering low-cost manufacturing and biocompatibility benefits over traditional materials.[87] Such applications leverage SMPs' tailorable glass transition temperatures for integration into existing assembly processes.[87]Textiles represent another key industrial domain for SMPs, where they impart smart functionality to fabrics for adaptive fitting and durability. SMP fibers and coatings enable garments to shrink or expand in response to body heat or moisture, providing a custom fit with minimal pressure, as seen in intimate apparel and sportswear.[88] Commercially, wrinkle-free clothing has been achieved through SMP-treated cotton, which resists creasing for hundreds of wash cycles without formaldehyde, exemplified by products like Mitsubishi's DiAPLEX fabrics that adjust water vapor permeability based on temperature for breathability.[88] Similarly, Toray's MemBrain® integrates SMPs for waterproof yet vapor-permeable outdoor gear, demonstrating industrial scalability since the early 2000s.[88]In packaging, SMPs facilitate tamper-evident seals that irreversibly change shape or color upon unauthorized access, enhancing security for industrial goods. Multilayered SMP films, developed through initiatives like the 2015 SBIR program, serve as bolt seals that deploy or fracture visibly when tampered with, providing a lightweight, stimuli-responsive alternative to metal locks.[89] These applications have scaled industrially since the 2010s, with SMP composites used in self-closing or shape-recovering packaging for food and logistics, where thermal or mechanical triggers ensure integrity detection.[90]The global SMP market, valued at approximately USD 700 million in 2025, is largely propelled by demand from engineering sectors like aerospace and automotive, reflecting their adoption for lightweight, adaptive components.[91]
Emerging and Multifunctional Applications
Shape-memory polymers (SMPs) have found innovative applications in photonics, where their ability to undergo reversible shape changes modulates optical properties such as refractive index, enabling dynamic control over light propagation. For instance, SMP-based optical switches exploit thermally induced deformations to alter the effective refractive index, switching between transparent and opaque states with response times on the order of seconds under moderate heating. This mechanism has been demonstrated in tunable photonic devices operating from visible to terahertz wavelengths, offering advantages over traditional electro-optic materials due to mechanical simplicity and low power consumption.[92] Additionally, SMPs facilitate the development of tunable lenses by deforming curved surfaces to adjust focal lengths, as seen in antireflective structures that reversibly switch reflectivity from 5% to over 90% upon shape recovery at ambient temperatures.[93]In brand protection, SMPs enhance anti-counterfeiting measures through holographic labels that deform under controlled heat, revealing hidden patterns or colors only upon authentication, a technology emerging in the 2010s. These labels, often based on polycaprolactone (PCL) networks with combined shape and color memory, store high-security information that is invisible at room temperature but activates via thermal stimuli around 50–60°C, making replication difficult without precise programming. Such features have been integrated into encryption labels for product authenticity verification, providing a tamper-evident layer beyond static holograms.[94]The integration of SMPs in 4D printing has revolutionized time-responsive structures for soft robotics, where printed components evolve their shape post-fabrication in response to environmental triggers. Programmable SMP inks, developed in the 2020s, enable multi-shape memory effects in robotic grippers and crawlers, achieving deformations up to 200% strain with recovery forces exceeding 1 MPa, suitable for untethered operations in confined spaces. Recent advancements include magnetically responsive 4D-printed SMPs for sustainable soft actuators, demonstrating crawling speeds of 20 cm/min in biomedical prototypes.[95][96]Multifunctional SMPs combine shape memory with self-healing capabilities through dynamic covalent bonds, such as Diels-Alder linkages in bio-based composites, allowing puncture repairs with over 90% tensile strength recovery after 24 hours at 80°C. Self-sensing variants incorporate embedded conductive fillers like reduced grapheneoxide, enabling real-time strain monitoring during actuation with gauge factors up to 50, as in ethylene-vinyl acetate composites for wearable devices. Two-way SMP actuators, leveraging crystalline-amorphous transitions, provide reversible deformations without external programming, cycling over 100 times with 150% bidirectional strain for applications in adaptive robotics.[97][98][99]Sustainability efforts in SMPs emphasize bio-based formulations derived from renewable sources like Eucommia ulmoides gum, reducing reliance on petroleum-derived polymers while maintaining shape fixity above 95%. These materials support eco-packaging solutions, such as deformable barriers that self-adjust to protect contents during transit, extending shelf life by 30% in modified atmosphere applications. In 2025, SMP nanocomposites with phase-change inclusions have advanced energy harvesting, converting mechanical deformations into electrical output with efficiencies reaching 15% in triboelectric generators for low-power sensors.[97][100][101]