Shape memory polymers, or SMPs, are a class of smart materials that can be programmed into a temporary shape and then recover their permanent, original form when exposed to an external stimulus such as heat, light, or an electrical field. The underlying mechanism involves three steps: the material is deformed above its transition temperature, the deformed shape is fixed by cooling or light exposure, and the original shape is restored upon reapplication of the stimulus. Switching segments within the polymer allow molecular mobility during deformation, while fixed domains created by chemical crosslinking or crystalline structures store the permanent shape. Despite their versatility, neat SMPs suffer from well-known drawbacks, including low thermal conductivity, modest mechanical strength, and a lack of electrical conductivity, all of which limit their usefulness in demanding smart-device applications where uniform heating and fast recovery are essential.

To overcome these limitations, researchers have traditionally turned to synthetic nanofillers such as carbon nanotubes, graphene, and metal nanoparticles. These additives can dramatically improve conductivity, mechanical strength, and recovery rates, but they carry significant environmental and health baggage. They are derived from non-renewable sources, can be potentially toxic, are expensive to produce at scale, and often disperse poorly within polymer matrices, leading to weak interfacial bonding. As global sustainability goals tighten and life cycle assessment becomes a standard metric in product development, the field has been actively searching for greener alternatives. Bio-derived nanofillers such as cellulose nanocrystals, lignin nanoparticles, chitin nanofibers, and starch-derived crystals offer biodegradability, low toxicity, and carbon neutrality, and previous studies have shown they can deliver mechanical and actuation performance comparable to or better than their synthetic counterparts.

In the new study, two biodegradable polymer matrices were used: polycaprolactone, chosen for its low melting point of roughly 60 degrees Celsius and proven shape memory behavior, and polylactic acid, used to evaluate compatibility with lignin fillers. The cellulose nanocrystals were isolated from cellulose pulp through acid hydrolysis with 64 percent sulfuric acid, a process that selectively degrades amorphous regions and leaves behind rigid, rod-like particles measuring 100 to 300 nanometers in length and 5 to 20 nanometers in diameter. The crystals were then dialyzed, centrifuged, and freeze-dried, and their surfaces were modified using TEMPO-mediated oxidation to introduce carboxyl groups that improve aqueous dispersibility and promote hydrogen bonding with the polymer matrix. Lignin nanofibers received a silane treatment with 3-aminopropyltriethoxysilane in an ethanol-water solution, which enhanced their compatibility with the PLA matrix and strengthened interfacial bonding.

The composites were fabricated using three different processing routes: solution casting, in which polymer and dispersed nanofillers were cast into molds and dried; in-situ polymerization, in which silanized lignin was added directly to the lactide monomer during ring-opening polymerization to promote grafting and dispersion; and melt mixing in a twin-screw extruder at 80 to 100 degrees Celsius followed by compression molding. Filler loadings from 1 to 10 percent by weight were investigated, with optimal performance found in the 3 to 5 percent range, where uniform dispersion and mechanical integrity were best balanced. Scanning electron microscopy of fracture surfaces confirmed homogeneous dispersion with no large agglomerates, and a percolated nanonetwork was observed to form at loadings of 3 percent and above, aiding stress transfer and toughening throughout the material.

The mechanical and thermal results were substantial. Dynamic mechanical analysis revealed an increase in storage modulus of up to approximately 25 percent, alongside a shift in viscoelastic transition behavior that confirmed effective filler-matrix interactions. At 3 percent filler loading, Young’s modulus rose by about 20 percent and tensile strength by roughly 10 percent, although elongation at break showed a modest decline, a familiar trade-off in reinforced polymers. Differential scanning calorimetry recorded a modest increase in glass transition temperature of 2 to 3 degrees Celsius and a slight rise in melting temperature with CNC inclusion. Thermogravimetric analysis showed that the onset of thermal degradation was delayed by up to 30 degrees Celsius, confirming that the integrated fillers improved thermal stability, an important consideration for materials that must survive repeated heating cycles.

The shape memory performance itself was impressive and durable. At 5 percent CNC loading, the composites achieved a shape fixity ratio of approximately 97 percent and a shape recovery ratio of about 95 percent, meaning the material both held its programmed temporary shape almost perfectly and returned nearly completely to its original form when reheated. Repeated thermomechanical cycling over ten cycles produced only a slight degradation of less than 3 percent, indicating minimal fatigue. The authors attribute these improvements to increased stiffness and enhanced heat conduction across the matrix-filler interface, which allows the material to heat more uniformly and recover more reliably than neat polymer, which often suffers from uneven heating and delayed response.

The most striking results came from the energy harvesting measurements. The team built a triboelectric nanogenerator, or TENG, using the composite film as one tribo-layer, repeatedly contacting and separating it from a PTFE sheet under cyclic compression at 1 to 2 hertz while capturing voltage and current with a digital oscilloscope. Peak output voltage climbed from roughly 25 volts at 1 percent filler loading to approximately 80 volts at 5 percent loading under 25 percent strain, a more than threefold increase. The gain is attributed to enhanced surface roughness, improved dielectric properties, and a larger effective contact area induced by the CNC incorporation. Triboelectric systems generate charge through contact electrification and electrostatic induction, and the nanofiller network increases both the triboelectric polarity and the real contact area during each deformation cycle.

Crucially, the researchers demonstrated that the intrinsic shape recovery process of the polymer itself could serve as the internal mechanical driving force for electricity generation, removing the need for any external actuation. As the material deforms and recovers, internal stresses and friction at the nanofiller interface drive charge separation and polarization, generating current in connected circuits. Durability testing showed stable output over approximately 500 repetitive deformation cycles, with minor drops after 1000 cycles that were partially recoverable through shape memory reheating, meaning the material can effectively self-restore its electrical performance using the same thermal trigger that powers its actuation. The authors note that the electrical outputs are competitive with systems built on synthetic fillers, yet the composite remains biodegradable and sustainable.

The implications extend across several fast-growing fields. Wearable and flexible electronics demand compact, self-sustaining materials, and an SMP composite that harvests energy from body heat triggered recovery, or from motions like walking, breathing, or typing, offers a route to self-powered sensors and skin-like devices without batteries. In soft robotics and autonomous sensing systems, the same thermo-mechanical-electrical coupling cycle could power onboard electronics while the material performs its actuation function. The work also closes gaps the field has struggled with: it rigorously links filler morphology, dispersion quality, and interfacial bonding to both actuation behavior and energy output, and it evaluates shape recovery, mechanical strength, and electrical generation together in a unified framework. As smart systems increasingly require sustainable, high-performance materials, this demonstration suggests that the future of self-powered devices may be grown in forests and fields rather than synthesized in reactors, with compostable composites quietly converting the physics of shape recovery into usable electricity.

Subject of Research: Biodegradable shape memory polymer composites reinforced with bio-based nanofillers for triboelectric energy harvesting

Article Title: Multifunctional shape memory polymer composites reinforced with bio-based nanofillers for energy harvesting applications

Article References: Multifunctional shape memory polymer composites reinforced with bio-based nanofillers for energy harvesting applications. (n.d.). https://doi.org/10.1007/s44493-026-00006-5

Image Credits: AI Generated

DOI: 10.1007/s44493-026-00006-5

Keywords: shape memory polymers, cellulose nanocrystals, lignin nanofibers, triboelectric nanogenerator, energy harvesting, biodegradable composites, polycaprolactone, polylactic acid, wearable electronics, self-powered sensors, soft robotics, sustainable materials

The performance of a filtering facepiece respirator is judged by two competing quantities: filtration efficiency, the fraction of incident aerosol particles it captures, and pressure drop, the resistance the filter poses to airflow. The NIOSH N95 standard, comparable to FFP2 in Europe and KN95 in China, requires filtration efficiency above 95 percent for a specified sodium chloride challenge aerosol, along with pressure drops below 245.2 pascals for exhalation and 343.2 pascals for inhalation at a flow rate of 85 liters per minute. The tension between these requirements is captured by the quality factor, defined as the negative logarithm of penetration divided by pressure drop; a higher quality factor means a filter collects more particles while breathing stays easier. Conventional N95s achieve their numbers with meltblown polypropylene fibers one to ten micrometers in diameter, made effective by embedded electrostatic charges that boost particle capture beyond what the fiber geometry alone could deliver.

That electrostatic crutch, the authors note, is also the conventional respirator’s Achilles heel. Surface charges decay with time, heat, moisture and liquid exposure, which is why manufacturers recommend limited shelf lives and limited periods of use. During the pandemic, testing of pristine commercial N95 and KN95 respirators revealed wide variability in performance, often falling below specification, because charge application and retention during production and distribution were inconsistent. And because polypropylene is not environmentally degradable, every retired respirator adds to the accumulating tide of plastic waste. Electrospun nanofiber filters offer a way out: their fibers are roughly an order of magnitude smaller than meltblown fibers, and that alone provides a favorable tradeoff between capture efficiency and air resistance, enough to reach N95 targets purely through mechanical filtration mechanisms such as diffusion and interception.

The MIT team chose poly(ε-caprolactone), or PCL, as their polymer, a synthetic polyester long used in biomedical applications and generally considered biocompatible. Previous studies have shown PCL degrades faster in soil and compost than other degradable polyesters such as polylactic acid, polyhydroxybutyrate and polybutylene succinate, and it also breaks down in natural aquatic environments. Earlier work had produced electrospun PCL air filters, but none had been specifically designed or tested against NIOSH N95 performance targets. Equally important was the choice of solvent. The researchers electrospun their fibers from a one-to-one mixture of acetic acid and formic acid, both classified by the FDA as biologically benign Class 3 solvents and rated favorably by green-chemistry solvent guides, in contrast to the dimethylformamide and dichloromethane common in the electrospinning literature. Because formic acid contains water that can catalyze hydrolysis of the polymer, the team added it to the solution only minutes before spinning, a trick that kept the process stable for hours and could be adapted to continuous manufacturing with inline mixing.

By tuning the polymer concentration of the spinning solution from 6 to 14 percent by weight, the researchers produced smooth, mostly bead-free fiber mats with average diameters spanning 60 to 300 nanometers. They then measured filtration efficiency and pressure drop for each mat across a range of basis weights, controlled simply by spinning time. Plotting the negative logarithm of penetration against pressure drop revealed a striking regularity: each set of filters fell on a straight line through the origin, the slope of which is the quality factor. This graphical construction doubles as a design tool. Choosing the fiber diameter and solidity sets the slope of the line, while adjusting basis weight moves the design along it, and a filter meets N95 requirements whenever its line passes through the target region defined by the efficiency and pressure-drop limits. All but the thickest-fiber media, spun from 14 percent solutions, crossed the target region, and every set met the less stringent inhalation limit.

The data showed a steep rise in quality factor for fibers thinner than 100 nanometers, consistent with purely mechanical filtration. Smaller fibers also shifted the most penetrating particle size downward, from just under 80 nanometers for the largest fibers studied to around 60 nanometers for the smallest, and because count-based measurements weight the smallest, hardest-to-capture particles equally, media achieving 95 percent efficiency by count are expected to meet or exceed that threshold under the photometric methods used in official NIOSH testing. The design analysis carried a practical bonus: filters made of finer fibers not only required less material to hit the efficiency target, they also tolerated larger manufacturing deviations in basis weight before falling out of specification, a meaningful advantage for real production lines where fiber deposition varies across a web and over time.

Comparing measured pressure drops against classical theory exposed an intriguing gap in the literature. For larger fibers, the slip-flow-modified Kuwabara model, Pich’s adaptation for small Knudsen numbers, predicted air resistance accurately. But for the finest fibers, with average diameters near 60 nanometers and Knudsen numbers around 2.2, well beyond the conventional slip regime, the model increasingly overpredicted resistance, meaning the real filters breathed even more easily than theory suggested. The deviation followed an empirical power law in the Knudsen number, with a fitted exponent of about minus 1.27. The authors stress this relationship is an observation, not a theory, and note that no validated model yet exists for either pressure drop or filtration efficiency in this transition regime, despite the considerable practical value of sub-100-nanometer fibers.

To test the concept end to end, the team fabricated prototype duckbill-style respirators by electrospinning a PCL filtration layer, with average fiber diameter of 103 nanometers, directly onto a polylactic acid spunbond substrate, then laminating a second spunbond layer to form a spunbond-nanofiber-spunbond sandwich. Panels were laser-cut and ultrasonically welded along the edges. When tested on a TSI 8130 automated filter tester under NIOSH-specified conditions, all three prototypes exceeded 95 percent filtration efficiency and stayed below the inhalation pressure-drop limit, with two of the three also clearing the stricter exhalation limit. Because mechanical filters generally gain efficiency as particles load onto them, unlike electret filters that can lose their charge and fail under loading, the pristine prototypes likely represent a floor rather than a ceiling for performance, though rising pressure drop during loading means service-life studies remain an important next step.

The biodegradability promise was put to a quantitative test as well. Incubating the nanofiber mats in a lipase solution, the researchers found that degradation rate scaled inversely with fiber diameter, exactly what is expected when hydrolysis is confined to the fiber surface, and the smallest fibers converted completely to soluble products within roughly eight hours while bulk PCL pellets lost only 13 percent of their mass over twelve days under the same conditions. The authors estimate their accelerated enzyme test runs roughly five to ten times faster than degradation in compost or soil, and they caution that real-world breakdown involves complex microbial communities and variable conditions. The results nonetheless reinforce a single design principle with triple payoffs: fibers below 100 nanometers deliver higher filtration quality factors, need less material, forgive more manufacturing variability, and vanish faster after disposal. The remaining bottleneck, they note, is the spunbond support layers, whose fibers are an order of magnitude larger and dominate the mass of the finished respirator, and a systematic shelf-life study for the moisture-sensitive nanofiber media is still needed before biodegradable N95-class respirators can leave the laboratory for the factory floor.

The choice of polycaprolactone reflects decades of industrial familiarity. The polymer, first synthesized in the 1930s, is produced by ring-opening polymerization of ε-caprolactone and is processed commercially into packaging films, adhesives, and medical devices, meaning a respirator supply chain would not depend on novel synthesis routes. Its low melting point of roughly 60 degrees Celsius, however, could constrain sterilization methods and storage conditions, a consideration the authors’ planned shelf-life work will need to address.

The electrospinning setup itself remains close to industrial practice. The researchers used a multi-needle lab-scale spinner feeding solution at half a milliliter per hour onto a rotating drum collector under modest voltage, conditions compatible with the roll-to-roll nanofiber production lines already used for commercial filtration products. Depositing fibers directly onto a spunbond substrate, rather than onto a foil that requires a separate transfer step, further simplifies scaling.

Regulatory context matters as well. NIOSH certification of a biodegradable respirator would require testing beyond the filtration bench work reported here, including exhalation valve leakage where applicable and breathing-machine evaluations of complete facepieces. The prototype duckbill respirators, assembled by ultrasonic welding without adhesives, represent an early but concrete step toward that certification pathway.

Subject of Research: Biodegradable electrospun polycaprolactone nanofiber filter media designed to meet N95 respirator filtration and breathing-resistance standards

Article Title: Electrospun biodegradable polycaprolactone filter media for filtering facepiece respirators

Article References: Ewell, N., Fleishman, S., Mulherin, K., & Rutledge, G. C. (2026). Electrospun biodegradable polycaprolactone filter media for filtering facepiece respirators. Journal of Materials Science: Polymers, 1(1), Article 21. https://doi.org/10.1007/s44493-026-00021-6

Image Credits: AI Generated

DOI: 10.1007/s44493-026-00021-6

Keywords: electrospinning, polycaprolactone, nanofibers, N95 respirators, air filtration, biodegradable polymers, aerosol filtration, quality factor, pressure drop, enzymatic degradation, personal protective equipment, green solvents