Coassembly of hybrid microscale biomatter for robust, water-processable, and sustainable bioplastics

Abstract

Unlike conventional methods that typically involve extracting biopolymers/monomers from biomass using lots of hazardous chemicals and high energy, the direct utilization of biological matter (biomatter) without extraction offers a more sustainable alternative for bioplastic production. However, it often suffers from insufficient mechanical performances or limited processabilities. Herein, we proposed a hybrid microscale biomatter coassembly strategy that leverages the interactions between the inherent microarchitectures of waste cotton fiber and pollen particles. With minimal preprocessing, they form a castable slurry that can spontaneously organize into a dense fiber-laminate bioplastic network, exhibiting high mechanical properties (52.22 megapascals and 2.24 gigapascals) without using toxic organic chemicals or heavy machinery. The resulting bioplastic features controlled hydration-induced microstructural disassembly/reassembly, enabling water-based processability into complex, dynamic architectural systems. In addition, it demonstrates good biodegradability, closed-loop recyclability, and satisfactory environmental benefits, outperforming most common plastics. This study provides an instant nature-derived paradigm for bioplastics’ sustainable production, processing, and recycling, offering a promising solution for facilitating eco-friendly advanced applications.

Create high-performance, eco-friendly bioplastics without using toxic chemicals through hybrid upcycling of underutilized biomass.

INTRODUCTION

Plastics have played a crucial role in the advancement of society owing to their excellent mechanical properties and processabilities, leading to their application in fields ranging from basic packaging materials to sophisticated functional devices. However, plastics sourced from fossil fuels are emblematic of our reliance on nonrenewable resources and pose a formidable environmental challenge (1). With more than 450 million tons of plastics being produced annually worldwide, the resulting energy-intensive production processes, heavy release of greenhouse gases, low recyclability, and environmental persistence are alarming (2). Consequently, there is an urgent requirement to promote the development of renewable biomass resources and eco-friendly production/use/recycling strategies that are integral to achieving sustainability and fostering a circular bioeconomy (3–5). Biomass-derived polymers (e.g., cellulose, chitin, and starch) and monomers (e.g., ethylene and lactic acid) have been increasingly used to produce bioplastics (6). However, their intricate and energy-intensive extraction and refinement processes, biomaterial wastage, and the required use of toxic chemicals remain unavoidable (7). In addition, processing such materials into functional shapes usually involves harsh conditions, such as high temperatures. From a life cycle perspective, bioplastics often exhibit more negative environmental impacts than their petrochemical counterparts, offsetting the sustainability advantages associated with exploiting biomass resources (8).

A sustainable circular bioeconomy emphasizes maximizing the reuse and upcycling of biomass resources through eco-friendly approaches to reduce waste. Aligning with this paradigm, researchers have introduced an emerging in situ conversion strategy that focuses on the direct and integrative valorization of biological organisms or matter (biomatter), circumventing harsh extraction procedures and reducing biowaste (9, 10). In general, biomatter-based bioplastics are developed through two main approaches: “top-down” and “bottom-up.” Continual efforts have been made to create bioplastics via the “top-down” modification of macroscale biomatter, such as wood or bamboo (11–14). Despite recent advances in creating strong bioplastics by harnessing natural hierarchical structures, macroscale biomatter exhibits limited flexibility and processing capability during complex formation, particularly when compared to micro/nanoscale biopolymers/monomers. For example, challenges include toxic chemical requirements for delignification and a lack of effective ways to reprocess into delicate small-scale structures. As an alternative, microscale biomatter (e.g., lignocellulosic fiber, bacteria, fungi, and algae) can serve as instant biobuilding blocks for “bottom-up” self-assembly into bioplastics of various shapes through molds (15–21). However, insufficient interfacial bonding between biomatter units often results in a weak mechanical structure, unless supplemented with additional binders or mechanically assisted compression. These inherent limitations of “top-down” and “bottom-up” bioplastics, as comparatively summarized in table S1, pose challenges for their multifunctional and multiscenario applications. Hence, it is highly desirable to explore the feasibility of engineering readily available waste and low-value biomatter into bioplastics that have multimodal scalable processability and superior mechanical properties to pursue an eco-friendly and sustainable life cycle across the preparation, processing, and recycling processes.

In addition to being a renewable material resource, biomatter provides effective strategies for achieving exceptional material properties (e.g., static and dynamic adaptabilities) through the assembly of hybrid-scale components and architectures (22, 23). For example, nacre is composed of a “brick and mortar” architecture, wherein the inorganic “bricks” and the organic “mortar” interact tightly with one another, resulting in high strength and toughness (24). As another example, sea cucumbers evolve dermal tissues of variable stiffnesses and shapes by the entanglement of high-aspect-ratio collagen fibrils with a fibrillin matrix (25).

Thus, inspired by nature, we herein propose a hybrid microscale biomatter coassembly platform for the eco-friendly production of mechanically tough and easily shapeable bioplastics. To demonstrate this approach, we select inherently fibrous waste cotton and particulate pollen shells, both abundant but low in value, as the raw materials. Following minimal preprocessing, the two biomatter specimens combine and form a hybrid slurry that is directly cast to create a biomatter-hybrid (BH) bioplastic in an energy-saving and scalable manner without the requirement for toxic chemicals and hot pressing (Fig. 1A). They are intertwined and naturally build a highly dense fibrous-lamellar structure, straightforwardly leading to a high mechanical performance superior to that of most conventional plastics. Benefiting from its water-mediated reversible disassembly-reassembly transition in microstructure and micromorphology, we can use only water as a temporary plasticizer to mold the bioplastic repeatedly and even to paste separated pieces of different geometries and stiffnesses, allowing for the construction of complex and dynamic three-dimensional (3D) architectures, opening up opportunities for advanced multifunctional applications in a greener way. Compared to conventional plastics, the BH bioplastic demonstrated much less environmental impact, good biodegradability, and extended recyclability and reusability. This proposed hybrid biomatter coassembly approach suggests the potential for strategically upcycling biowaste and replacing conventional plastics with advanced and sustainable multifunctionality.

Fig. 1. Preparation of the BH bioplastic.

(A) Schematic diagram showing the fabrication, water processability, recyclability, and degradability of the BH bioplastic. (B) Large amount of waste cotton fiber discarded from factories and (C) resulting fiber suspension prepared after homogenization pretreatment. (D) Photographic images of the bee pollen particle and (E) resultant pollen particle suspension prepared after the soapmaking-like method. (F) Casting of the homogeneous fiber/particle slurry. (G) Large scale of the BH bioplastic prepared using the blade casting method. (H) Radar plots comprehensively comparing the performance of the BH bioplastic with a representative petroplastic (i.e., PET) and a representative bioplastic (i.e., PLA).

RESULTS

Preparation of the BH bioplastic

In this study, two underused biomatter specimens were used to prepare the BH bioplastic, namely leftover waste cotton and low-value pollen shells. Waste cotton left over from the spinning and weaving processes in the cotton industry accounts for 15 to 30% of the total harvest, which conventionally ends up in landfills or is openly dumped in nature or the ocean (Fig. 1B) (26, 27). Before using this material, a homogenizer was used to break fiber (Fig. 1C). In addition, pollen is an abundant and low-cost by-product of beekeeping, with an annual reserve exceeding 30 million metric tons worldwide (28) (Fig. 1D). However, only the nutrient-rich components inside the pollen were previously deemed valuable, whereas the outer shell was discarded. Herein, by adopting a simple soapmaking-like method (29), water, ethanol, and sodium hydroxide were used to remove dust and surface lipids and to soften the hard shell to obtain a pollen shell suspension (Fig. 1E), thereby enabling its valorization. Subsequently, the prepared suspension stocks of cotton fiber and pollen microgels suspensions were mixed and filtered to create a fiber/particle slurry, which could be easily cast onto substrates without size limitations (Fig. 1F). After the evaporation of water, a piece of bioplastic was obtained; notably, this process was also applicable on a large scale (up to 200 cm by 50 cm by 0.1 cm; Fig. 1G), producing a product with excellent flexibility and foldability (fig. S1). Profiting from the complementary nature of the two biomatter specimens during hybrid coassembly, the obtained BH bioplastic was endowed with a high strength but a low density of 0.57 g cm⁻³ (fig. S2). In addition, it was found to be amenable to sustainable water-assisted processing and recycling. The processes used to prepare the BH slurry and the resulting bioplastic are both energy efficient and resource saving without requiring the use of toxic organic chemicals (Fig. 1H and table S2). These characteristics render this material a potentially sustainable and practical alternative to traditional petroplastics and bioplastics.

Structural and chemical interactions between the cotton fiber and the pollen shells

The slurrying property, which results from in situ preprocessing, determines the castability of this hybrid biomatter system for large-scale production. Initially, the raw pollen was compared before and after defatting. More specifically, after removing the surface lipids and dust (Fig. 2, A and B), the particle size of pollen changed from 27 to 30 μm (fig. S3, A to C, E, and F). The defatted pollen exhibited an intact round architecture, even in the dry environment of the scanning electron microscope. In addition, it demonstrated a tendency to settle quickly in water, resulting in an unstable, sand-like dispersion (Fig. 2C), which reflects the ease of separation of the defatted pollen particles from water. After alkali preprocessing, the pollen shells became hollow, swelled to 45 μm (fig. S3, D and G), and displayed a semisolid microgel (pollen microgel) aggregation state, which collapsed into a thin cake shape when dried (Fig. 2C). Fourier transform infrared spectroscopy (FTIR) showed that fatty acids or lipids (C═O stretching) at 1744 cm⁻¹ (30) and amide I (C═O stretching) and amide II (N─H bending and C─N stretching) in proteins at 1630 and 1550 cm⁻¹ (17, 31, 32) disappeared from the pollen after defatting and alkali preprocessing, indicating that most of the lipids and proteins were effectively removed. The emerging strong vibrations that appeared at 1581 and 1705 cm⁻¹ are assigned to C═C of sporopollenin (33, 34) and C═O stretching of carboxylic acids (31, 35), respectively, suggesting the exposure of the sporopollenin and the carboxylic groups in the pollen microgels (Fig. 2D and table S3). That is, after alkali treatment, the cytoplasm is completely removed, while the pollen shells partially degrade and expose more carboxyl groups (17), which softens the pollen particles and enhances their surface hydrophilicity. Consequently, the pollen particles absorb water, swell, and then crowd together at high concentrations, leading to a higher viscosity and stability of the pollen microgels. Compared to the raw cotton fiber, the homogenized fiber maintained the diameter of the raw fiber (~10 to 20 μm), shortened the length, and generated some tiny fiber (Fig. 2, E and F, and fig. S5, A to F), with no obvious change in chemical composition (fig. S5G). Therefore, after thorough mixing in a dilute state followed by concentration, the pollen microgels and fiber were able to form a stable slurry with a solid content of 1.5% (Fig. 2G) with high zeta potential (fig. S6) and high viscosity (Fig. 2H).

Fig. 2. Structural and chemical interactions between the fiber and the microgels.

(A to C) SEM of raw pollen (A), defatted pollen particles (B), and pollen microgels (C). The insets showed the raw materials and the states of the corresponding aqueous dispersions. (D) ATR-FTIR of the raw pollen, defatted pollen, and pollen microgels. (E and F) SEM images of the raw cotton (E) and cotton fiber after homogenization preprocessing (F). (G) Optical microscopic image of BH slurry. (H) Rheological properties of the fiber suspension, pollen microgel suspension, and BH slurry with a solid content of 1.5%. (I) Schematic diagram demonstrating self-assembly of the cotton fiber and the pollen microgels into the BH bioplastic upon the evaporation of water. (J to L) Cross-sectional SEM images of the fiber/particle slurry at the initial (J), middle (K), and final (L) stages of the water evaporation process. (M) LSCM image of the fibrous lamellar network within the BH bioplastic. (N and O) XRD spectra (N) and ATR-FTIR (O) of the pure fiber sample, pure pollen sample, and BH bioplastic. a.u., arbitrary units.

BH bioplastic was fabricated by directly casting a BH slurry, in which the minimally preprocessed waste biomatter gradually coassembled and bound together to form a dense structure during water evaporation, as depicted in Fig. 2I. Under air-drying conditions, the BH slurry with 1.5% solid content required 42 hours to dry completely to obtain the BH bioplastics, whose thickness could be adjusted (fig. S7). It was found that vacuuming and heating can speed up the preparation process, but it will cause structural defects, affecting the mechanical properties (fig. S8). In addition, we explored the rheological properties and drying time of BH slurry with different solid contents and fiber contents. It was observed that higher solid contents or lower fiber contents resulted in higher viscosity, while lower solid contents increased drying time (fig. S9). To better understand these interactions, the microstructural evolution was observed by scanning electron microscopy (SEM) during the different stages of drying. In the initial slurry state, the pollen microgels were observed as deflated spheres (because of the dehydration required for SEM) with spiky protrusions on the surface, and the fiber and microgels were found to be well dispersed (Fig. 2J and fig. S10A). As the water evaporated, the pollen microgels were gradually compressed parallel to the evaporation interface, where the fibers were interspersed (Fig. 2K and fig. S10B). After drying, the pollen microgels collapsed completely into lamellar structures, which were tightly wrapped and stacked with fibers (Fig. 2L and fig. S10C). Laser scanning confocal microscopy (LSCM) observations showed that the fiber (red) was interlaced to form a 3D network skeleton filled with pollen (green) as a continuous matrix (Fig. 2M). The absence of fibrils led to cracks in the pure pollen bioplastic when the thickness increased to 100 μm, highlighting the important role of the fiber network in maintaining an intact film by reducing evaporation-induced stresses during drying (fig. S11) (10).

Subsequently, pure fiber and pollen samples were prepared to investigate the chemical interactions between the cotton fiber and pollen lamellae within the BH bioplastic. The x-ray diffraction (XRD) patterns of BH bioplastic showed characteristic peaks corresponding to fiber and pollen, including those of the cellulose I crystals within the fiber (2θ = 14.6°, 16.6°, and 22.6°) (12, 36), and a peak corresponding to the carbonaceous organism with an amorphous pollen structure (2θ = 20.0°) (37) (Fig. 2N). In addition, the FTIR spectrum of the BH bioplastic (Fig. 2O) exhibited strong absorbance bands at 1057, 1028, and 897 cm⁻¹, which were attributed to the C─O stretching, C─O─C pyranose ring skeletal vibration, and β-glycosidic linkages of cellulose in the cotton fiber (17, 38–40). In addition, the peaks at 1705 and 1581 cm⁻¹ were assigned to C═O stretching of carboxylic acids and C═C vibration of sporopollenin in pollen. These results confirm the successful mixing and structural preservation of the two biomatter specimens. Compared with the pure fiber and pollen samples, the absorbance peak assigned to the O─H stretching vibration of hydroxyl groups shifted from 3334 and 3347 cm⁻¹ to 3328 cm⁻¹ in the BH bioplastic. This red shift indicates a change in the hydrogen bonding environment, which is associated with hydrogen bonding interactions between the fiber and pollen microgels (17, 41).

Mechanical strength of the BH bioplastic

The pure fiber sample displayed a mesh-like structure with many large interstices owing to insufficient physical entanglement, which resulted in poor mechanical stability (Fig. 3, A and B, and fig. S12A) (14). The pure pollen sample, formed by closely stacked pollen lamellae, exhibited a relatively flat surface and highly dense cross-sectional structure (Fig. 3, C and D, and fig. S12B), suggesting a high density of hydrogen bonds between adjacent pollen lamellae. However, the pollen sample lacks extensive, long-range physical connections spanning hundreds of micrometers. In contrast, the hybrid sample capitalized on the natural structural advantages of the biomatter interwoven into a stable fibrous-lamellar hybrid network (Fig. 3, E and F, and fig. S12C). In this structure, the fiber provided connectivity as an interlaced skeleton, while the pollen lamellae, functioning as a continuous matrix, fill and bond the fiber network. The BH bioplastic had a transparency of 65.47% at 600 nm (fig. S13A) and a haze of 91.19% (fig. S13B). It was demonstrated that the refractive index of pure pollen at 600 nm is 1.50 (fig. S13C), similar to that of cotton fiber (42). Therefore, the relatively high transparency may be due to the close packing of the pollen matrix and cotton fibers and their matched refractive index, which reduces the scattering and refraction of light. In addition, the obtained BH bioplastic exhibited a roughness of 7.94 μm, which can be tuned by different methods (fig. S14), suggesting its broad potential for multifunctionality.

Fig. 3. Mechanical strength of the BH bioplastic.

(A to F) SEM images of the top view [(A), (C), and (E)] and the cross section [(B), (D), and (F)] of the pure fiber bioplastic [(A) and (B)], pure pollen bioplastic [(C) and (D)], and BH bioplastic [(E) and (F)]. The insets show the transparencies of corresponding samples [(A), (C), and (E)]. (G) Tensile stress-strain curves of the pure fiber sample, pure pollen sample, and BH bioplastic. (H) Tensile strengths and Young’s modulus of the pure fiber sample, pure pollen sample, and BH bioplastic. (I) Comparison of the mechanical performance of the BH bioplastic with widely used thermoplastic and degradable plastics. (J to L) Fracture surfaces of the pure fiber sample (J), pure pollen sample (K), and BH bioplastic (L).

Subsequently, it was demonstrated that different hybrid ratios of the fiber and pollen generate different levels of mechanical performance. More specifically, when the fiber content was increased from 0 to 30%, the strength and modulus of the BH bioplastic increased; however, these properties decreased when the fiber content was increased further to 40 and 50% (fig. S15). The highest tensile strength (52.22 MPa) and Young’s modulus (2.24 GPa) with a large strain of 6.15% were obtained for the sample containing 30% fiber; these values represented improvements in both the tensile strength and Young’s modulus compared to the single biomatter–based samples (Fig. 3, G and H). Notably, the BH bioplastic exhibits mechanical properties that are comparable or superior to those of widely used thermoplastics and biodegradable plastics (Fig. 3I) (43, 44). Subsequently, SEM was used to investigate the fracture behaviors of the different specimens during tensile tests. For the pure fiber samples, the weak interaction forces between the fibers caused them to be pulled out at different lengths (Fig. 3J and fig. S16A). The pollen sample showed a flat extended crack path almost parallel to the stress, exhibiting brittle fracture (Fig. 3K and fig. S16B). In contrast, the BH bioplastic displayed a relatively straight tensile fracture surface owing to the presence of the macroscopic microfiber network wherein the pollen lamellae were attached to the microfibers through high-density hydrogen bonding (Fig. 3L and fig. S16C).

Eco-friendly shapeability through water-assisted molding and pasting

During the life cycles of most commercial plastics, processing into the desired shapes is an essential step that typically involves energy-intensive sophisticated equipment operating at high temperatures. In this study, water alone was used to mold the BH bioplastic by soaking, fixing, and drying at 25°C via a process that was driven by the rearrangement of the hybrid structural network (Fig. 4A). Changes in the microstructure were examined by SEM during hydration and dehydration to investigate the water-molding process. It was found that the hybrid network structure exhibited restricted swelling and partial disassembly upon exposure to water. Consequently, the wet BH bioplastic had sufficient elastic deformation spaces and an appropriate wet mechanical strength (fig. S17) (45) to accommodate the compressive and tensile stresses during molding. After drying, this deformed hybrid network reassembled compactly and adaptively, reforming into a dense structure with a fixed shape (Fig. 4B and fig. S18). Using a dynamic vapor sorption approach to reveal the water-assisted molding properties of the BH bioplastic (fig. S19), it was deduced that the large number of primary adsorption sites and the difficulty in forming water clusters facilitated water transmission, thereby achieving water-molding and fast shape fixation (44, 46).

Fig. 4. Eco-friendly shapeability via water-assisted molding and pasting.

(A) Schematic diagram showing the transformation of a BH bioplastic strip into a helical shape via water-assisted molding. (B) Photographic and corresponding cross-sectional SEM images of the BH bioplastic during the water-molding process [(B), I to IV]. (C) Schematic and mechanistic diagrams of the water-assisted pasting process. (D) Cross-sectional and enlarged SEM images of the water-pasted BH bioplastic. (E and F) Top-view SEM (E) and 3D profiles (F) of the BH bioplastic surfaces in the dry (I) and wet (II) states. (G) Photographic images of the BH bioplastic following water pasting. (H) Photographic images showing the ability of the water-pasted bioplastic to lift a 1000-g weight. (I) Tensile stress-strain curves of the original and pasted BH bioplastics.

Neither the pure fiber nor the pure pollen samples could be molded using the water-assisted method because of their faster water absorption rate and higher water absorption capacities (fiber, 492%; pollen, 353%; BH bioplastic, 166%) (fig. S20, A and B). These characteristics led to a loose microstructure in the wet state (fig. S21), which weakened its mechanical properties, thereby hindering its water processability. The fiber sample is superhydrophilic, as demonstrated by the rapid absorption of a water drop, while the pollen sample, although slightly hydrophobic (fig. S20C), experienced marked volume changes upon soaking and drying, which caused it to detach from the mold and fail to obtain the intended shape (fig. S22). In contrast, the dense fibrous-lamellar network structure in the BH bioplastic brings stronger hydrogen bonds and enhances the structural integrity, which restricts the hydrophilicity and water penetration pathways, thereby reducing water absorption.

Simultaneously, two separate BH bioplastics were pasted together using only water (Fig. 4C). Similar to the molding process, the water-wetted parts of the bioplastics were partially disassembled and then reassembled adaptively and conformally along the joint interface, with mutual embedding being observed as the water evaporated (Fig. 4D and fig. S23). Notably, the surface changed from relatively smooth to wrinkled after water was applied (Fig. 4E), resulting in an increase in surface fluctuation and roughness (Fig. 4F), which therefore enhanced the interface contact between the bioplastic pieces. As a result, the water-pasted BH bioplastic exhibited an excellent tensile strength, similar to those of the original bioplastic, allowing a weight of 1000 g to be lifted (Fig. 4, G and H). The mechanical properties remained high after pasting (Fig. 4I), indicating that water-assisted pasting effectively retained the native material properties and potentially enabled a broader range of geometrical and functional values for bioplastics. In addition, the pH of water does not affect the water-assisted processability (fig. S24).

Multidimensional and multifunctional applications

With the above-described water-assisted processability in mind, it was considered that various geometries could easily be designed for the BH bioplastic, allowing complex shape customization to be achieved under ambient conditions. To demonstrate this capability, simple shape elements were molded. Using repeated soaking and drying, the same BH bioplastic strip was reversibly processed into different shapes, ranging from a semicircle to an acute triangle as the mold bending curvature was increased (Fig. 5A). The mechanical strength of the BH bioplastic did not weaken but rather improved slightly after multiple water-molding cycles (Fig. 5B). This improvement was attributed to the repeated opening and closing of its structural network, forcing the fibers and pollen lamellae to rearrange and stack more tightly together. These factors contribute to the mechanical stability of the BH bioplastic during processing, which is particularly important for the generation of complex self-supporting architectures. For example, bioplastics exhibiting tower-like 3D shapes can be generated by cutting parallel to the edges of a square, lifting along the center, and drying at room temperature, thereby providing a sustainable plastic processing method for shapes and structures with a wide range of designs (Fig. 5C). In addition, water molding is a universal method for processing bioplastics of different thicknesses (fig. S25).

Fig. 5. Multidimensional and multifunctional applications of the BH bioplastic.

(A) Reversible water molding for single-element shapes. (B) Tensile stress-strain curves of the BH bioplastic after cycled water molding into different double-element shapes. (C) Construction of complex self-supporting tower structures. (D) Photographic images of various 3D structures (i.e., Möbius strip, bow, and flower) prepared by water pasting. (E) Complex biomimetic D. helix reproduced by combining water molding and water pasting. (F) Tensile stress-strain curves of the BH bioplastic under different RH values. (G) Photographic images showing the fabrication of a square load-bearing structure and its mechanical strength upon the application of a 1000-g weight at RH levels of 30, 60, and 90%. (H) Schematic representation of a bistable structure that can switch between two states under longitudinal loads. (I and K) Experimental photographic images (I) and finite element simulations (K) of the compression process of the bistable structure. (J) Experimental and simulated force-displacement curves of the compression process.

Water pasting, which complements the water-molding approach, can transform flat bioplastics into intricate 3D shapes, such as Möbius bands, bows, and flowers. This is achieved by connecting specific locations of one or multiple twisted or bent plastic strips, which cannot be realized using only bending/folding–based molding processes (Fig. 5D). Furthermore, water pasting can be used to join different water-molded bioplastic parts with more complex geometries, as validated by mimicking nature. More specifically, in the current study, the bioplastics were molded into petals of different shapes, and subsequently, the centers of the petals were pasted together, successfully replicating the complex morphology of Dendrobium helix (Fig. 5E).

The convenient transformation and combinability of multiple geometries, along with the observed exceptional mechanical properties, endow the BH bioplastic with advanced functional applications that go beyond the simple packaging and container functions reported for other bioplastics. Before discussing the value of the prepared bioplastic, its performance stability in different temperature and humidity environments should be considered. Thus, thermogravimetric analysis was used to demonstrate that the BH bioplastic exhibited good thermal stability (fig. S26). Using dynamic mechanical analysis, it was further confirmed that the loss modulus of the BH bioplastic is almost constant regardless of temperature change, at a T_g of 173°C, which is higher than those of polycarbonate (150°C) and polylactic acid (PLA; 60°C) (47), demonstrating its good resistance to high temperatures (fig. S27). More intuitively, they exhibited negligible thermal shrinkage or collapse at 250°C over a period of 10 min (fig. S28). Furthermore, the BH bioplastic retained a certain degree of water processability when heated and can also be thermally processed like thermoplastics (fig. S29).

Tensile tests at different relative humidity (RH) levels (i.e., 30, 60, and 90%) showed that there was a decrease in mechanical properties when RH increased (Fig. 5F), attributed to the increased water content of the BH bioplastic (fig. S30). However, even at 90% RH, the tensile strength value (~37 MPa) was still comparable to those of common plastics. In addition, it was shown that the BH bioplastic molded into a helical coil shape was able to endure four cycles of humidity changes between 25 and 75% (fig. S31A). The helical coil BH bioplastic can retain its shape for up to 6 months in a natural environment (RH from 30 to 70%) (fig. S31B), with virtually no change in microstructure, mechanical properties, or chemical composition (fig. S31, C to F). To further illustrate the adaptability of this material, a square load-bearing structure was designed on the basis of the use of sequential water molding and water pasting (Fig. 5G). The structured BH bioplastic, weighing only 0.1 g, was able to withstand a 1000-g weight at RH values of 30, 60, and 90%, in addition to withstanding a 500-g weight of hot water at 100°C (fig. S32), suggesting its potential value in applications requiring a lightweight and strong nature.

In addition to joining separate parts, water pasting can be used to build thick and stiff bioplastic by stacking multiple layers of thin and flexible bioplastics on top of one another. This enables the design of specific mechanical metamaterials by strategically assembling a modular BH bioplastic with different geometries and stiffnesses, ultimately leading to the construction of advanced functional systems. To test this hypothesis, we designed a bistable structure that can switch between two stable states under an external force (Fig. 5H). This structure consisted of thick parts (2000 μm) for support and thin parts (100 μm) to ensure snap-through instability and good energy absorption properties. Consequently, a thick bioplastic was fabricated by water-pasting 20 layers of the thin bioplastic (fig. S33A), and the resulting structure was found to be 55 times stiffer than the thin bioplastic (fig. S33, B and C), and the density was slightly increased to 0.63 g cm⁻³ (fig. S33D). After water-molding two thick bioplastics into T- and U-shapes, they were water pasted with two thin bioplastics to obtain a bistable structure (fig. S34). When an external pressure was applied, the thin bioplastic gradually bent and eventually transitioned to another stable flat state, triggering the closure of the bistable structure (Fig. 5I). This process was modeled and validated by using the finite element method (fig. S35 and movie S1). During compression, the force increased with an increase in displacement and then decreased to reach a negative value (Fig. 5J). This result indicated that the bistable structure has two stable states that can switch under longitudinal loads (Fig. 5K).

Environmental impact, recyclability, and biodegradability

Life cycle assessments (LCAs) were conducted to analyze the cradle-to-gate environmental impacts of the BH bioplastic and to compare these results with those of three commonly used conventional plastics, namely acrylonitrile butadiene styrene (ABS), polyvinylchloride (PVC), and polyethylene terephthalate (PET), and three biodegradable plastics, namely PLA, polybutylene succinate (PBS), and butylene adipate-co-terephthalate (PBAT). Figure S36 shows a flowchart of the continuous production and recycling of bioplastics after end of use, while the input and output for each step are presented in table S4. Considering the lower mass required for low-density and high-strength plastics in the same application, the functional unit was set to the environmental impact per cubic centimeter per megapascal to represent both the environmental impact and material performance (table S5). The prepared BH bioplastic demonstrated a competitive performance across most categories of environmental impacts, such as global warming impact, fossil resource scarcity, and particulate matter formation (Fig. 6A and table S6). In most cases, the environmental impact of the BH bioplastic is one to five times lower than that of other biobased plastics such as PLA, PBS, and PBAT. When compared to conventional fossil fuel–based plastics, the environmental impact of the BH bioplastic is close to the lower bond of their impact range, highlighting its strong potential as a sustainable, next-generation bioplastic.

Fig. 6. Environmental impact, recyclability, and biodegradability.

(A) LCA of BH bioplastic production compared to those of other conventional plastics (ABS, PET, and PVC) and biodegradable plastics (PBS, PBAT, and PLA). Note: CO₂ eq, carbon dioxide equivalent; PM_2.5, particulate matter 2.5. CFC-11 eq, CFC-11 (trichlorofluoromethane) equivalent; P, phosphorus; CTUe, comparative toxicity unit for ecotoxicity; CTUh, comparative toxicity unit for human health. (B) Remolding and recycling of the BH bioplastic. (C) Biodegradability testing of the BH bioplastic, PVC, and PET in the soil environment.

After service, the BH bioplastic can be easily remolded into different shapes by resoaking and remolding, and the end-of-life bioplastics can be crushed into small pieces (1 cm²) and stirred for 12 hours into a homogeneous suspension for recasting (Fig. 6B and fig. S37). These characteristics increase their utilization value while avoiding the use of toxic reagents or complicated recycling processes (48). Moreover, this hybrid biomatter coassembly strategy is not limited to cotton fiber and sunflower pollen but is also widely applicable to other fibers (i.e., coniferous and broadleaf fibers) (fig. S38) and other pollens (i.e., rape, lotus, and rose pollens) (fig. S39). To evaluate the degradation performances of some select examples, PET, PVC, and the BH bioplastic with the same dimensions were buried in a natural soil environment at a depth of 10 cm and observed over time. It was found that after 2 months of degradation, the mass of the BH bioplastic was lost by 20% (fig. S40A), and there was no obviously difference in FTIR except for the appearance of characteristic peaks of the soil (fig. S40B) (49), but SEM showed cracks (fig. S40C). The BH bioplastic gradually broke down and completely disappeared after 6 months (Fig. 6C). However, the PET plastic and PVC foam did not exhibit any degradation. Overall, these results demonstrated the excellent multifunctional potential, convenient recyclability, and long-term service performance potential of the BH bioplastic, which are key features in the material development of a modern, sustainable society.

DISCUSSION

In summary, an innovative coassembly strategy was presented to provide a notable avenue for leveraging the inherent properties of microscale low-value biomatter, with the aim of achieving economical and sustainable plastics that spontaneously form dense and strong fibrous-lamellar hybrid network structures. More significantly, it was possible to plastically and reversibly rearrange this hybrid network through structural opening and closing, triggered solely by hydration, thereby endowing the resulting BH bioplastic with sustainable water processability. It was possible to generate samples of various functional shapes by integrating units of different geometries and stiffnesses. The eco-friendly nature of the proposed BH bioplastic, in addition to its mechanical strength, multifunctional adaptability energy-saving production route, facile processing, and efficient recycling characteristics, underscores its potential as a sustainable alternative for reducing and engineering plastic pollution and promoting a circular bioeconomy. Overall, this study presents a microscale biomatter engineering paradigm for the design and construction of high-performance and low-environmental-impact bioproducts, which we believe may also extend to other combination types of plant and animal fibers (e.g., flax, sisal, silk, and wool) and biomicroparticles (e.g., fungal spores, bacteria, algae, and cell debris). As emerging and extraordinarily pure biomatter-assembled materials, BH bioplastics are expected to expand the realm of sustainable and eco-friendly fabrication of functional bioproducts.

MATERIALS AND METHODS

Materials and chemicals

Pollen particles (sunflower, canola, lotus, and rose) are from Shaanxi GTL Biotech Co., Ltd., China. Fibers (cotton, coniferous, and broadleaf) are processed waste from Allmed Medical Products Co., Ltd., China. Ethanol (C₂H₅OH) and sodium hydroxide (NaOH) were all of analytical grade and were purchased from Sinopharm Chemical Reagent Co., Ltd., China. ABS, PET, PVC, polyhydroxyalkanoates, PBS, polycaprolactone, PBAT, polyvinyl acetate, and PLA are from Alibaba (China) Co., Ltd., China. All aqueous solutions were prepared with ultrapure water (electrical resistivity, 18.2 megohms·cm).

Fabrication of pollen microgels

First, pollens of different species (sunflower, canola, lotus, and rape) were stirred with water (v:v = 1:2) for 2 hours at 50°C using an overhead electronic stirrer (DLAB OS20-S, China) and constant temperature magnetic stirrer (DF-101S, China). Then, the pollen suspension was passed through a nylon mesh (pore diameter, 100 μm) to remove any contaminating particulate matter. The filtered pollen was stirred in ethanol (v:v = 1:1) at reflux for 2 hours, and then the ethanol was removed by a vacuum filtration device (Joanlab VP-15L, China). The step was repeated twice, and then the pollen was air dried to obtain defatted pollen. Afterward, defatted pollen was mixed with NaOH solution (10 wt %) according to v:v = 1:2, stirred at 80°C for 2 hours by a constant temperature magnetic stirrer (DF-101S, China), and washed with NaOH/water solution through a nylon mesh (pore size, 20 μm) until the filtrate was clarified. Then, it was mixed with an equal amount of NaOH solution (10 wt %), and the reaction was continued for 12 hours at 80°C. Last, the pollen microgels with a solid content of 1% were obtained by rinsing with continuous water until they were neutral, after which the excess water was filtered through a nylon mesh (pore size, 20 μm).

Fabrication of fiber suspension

First, the discarded raw fiber materials (cotton, coniferous fiber, and broadleaf fiber) were hand selected to remove leaves, stems, and visible dust and impurities. Then, the fibers were crushed into small pieces and homogenized with a homogenizer (A25 digital homogenizer, China) at 10,000 rpm for 5 min to obtain a uniformly dispersed fiber suspension with a solid content of 1%.

Fabrication of the BH bioplastic

The pollen and fiber were mixed according to different solid content ratios (10 to 50 wt %). Then, the mixture was concentrated by filtration to make a fiber/particle slurry of 1.5% solid content for casting. The filtered water was recycled. The slurry was cast on a hydrophobic substrate by a coating test machine (CHTB-07, China), and the BH bioplastic was obtained after water evaporation at room temperature. All data analyses are based on the BH bioplastic with a 30% fiber content if not specified otherwise.

Characterization

Photographic images and movie were taken using a digital camera (Nikon, Japan). The basic optical morphology of fiber and pollen was observed using an inverted biomicroscope (Motic AE2000, Canada). The samples were sputter coated with gold and used for morphological observation by SEM (Zeiss Germany SEM 500, Germany) at 25°C and 35% humidity. The rheological properties of the samples were tested by a rheometer (TA Instruments DHR-2, US) using 60-mm parallel plates, and the test gap was set to 1.0 mm. The zeta potential was tested by a zeta potentiometer (ZetaSizer Nano-ZS, UK). The structure of fiber and pollen was observed dynamically and detected in real time by LSCM (Leica STELLARIS 5 SR, Germany). The fiber mixture was stained with methylene blue, and the pollen microgels were autofluorescent. The fiber, pollen, and BH bioplastic were placed on thin slides and imaged under laser excitation channels at 561 and 488 nm. The composition and molecular structure or morphology of the material were analyzed by XRD (SmartLab SE, Japan) and recorded at 40 kV and 40 mA using Cu-Kα radiation (λ = 0.154 nm) in the 2θ range of 10° to 40°. The chemical components were analyzed by FTIR (Nicolet170-SX Thermo Nicolet Ltd., US) in the wavenumber range of 4000 to 500 cm⁻¹. The transmittance and haze of the samples were measured by ultraviolet-visible spectroscopy (PerkinElmer Lambda 950, US) at 200 to 800 nm. The refractive index was determined by an elliptical polarization spectrometer (Horiba France SAS, France) at 200 to 800 nm. The water contact angle was measured by a contact angle meter (SL200KS, US). The thermostability was measured by thermogravimetry (Mettler Toledo TGA2, Japan) at temperatures ranging from 25° to 600°C at a heating rate of 10°C min⁻¹ in N₂. The 3D morphology and roughness of the samples were observed by the white light interference 3D surface profiler (NewView9000, US).

Static mechanical test

Mechanical testing was conducted according to the standard test method using an Instron 3343 universal test system with a 1-kN pressure measuring element. The samples with a size of 50 mm by 5 mm were loaded into the tester with a clamping distance of 20 mm and extended uniaxially at a constant rate of 5 mm min⁻¹ until rupture. The mechanical strengths in the dry state were tested at RH = 30% and a temperature of 25°C. The wet state is when the samples are thoroughly soaked in water for 12 hours. Different humidity states are the samples placed in a programmable constant temperature and humidity chamber (HT-3-2252, China) at different humidity levels (RH = 30, 60, and 90%) for 12 hours. Mechanical compression and three-point bending tests were carried out on a universal testing machine (SANS UTM6503, China) with a loading capacity of 100 N and a loading/unloading speed of 10 mm/min.

Dynamic vapor sorption

The dynamic water vapor adsorption behavior of the BH bioplastic was evaluated using a dynamic vapor sorption instrument (BSD-VVS, China). The BH bioplastic was sized at 10 mm by 10 mm. The measurement was performed at 20°C with the RH first reduced to 0% until an equilibrium mass change per minute (dm/dt) of <0.002% min⁻¹ was achieved for 10 min, and the initial dry mass was recorded. Subsequently, the RH was increased in 5% steps from 5 to 95% (adsorption) and then decreased to 0% (desorption) in the reverse order. In each step, the RH was kept constant until equilibrium was reached and the mass was tested. The water content was calculated for each RH on the basis of the equilibrated bioplastic mass.

Dynamic mechanical analysis

Mechanical properties were tested by a dynamic mechanical analyzer (TA Instruments Q800, US) with a sensor force of 150 N. The samples with dimensions of 30 mm by 5 mm were loaded into the test machine with a clamp distance of 8 mm, and the measurements were conducted with a contact force of 0.01 N and a frequency of 10 Hz. Humidity sweep tests were carried out on RH from 30 to 90%, with an RH increase rate of 2% min⁻¹ at 25°C. Temperature sweep tests measured temperatures from −10° to 200°C with a temperature increase rate of 5°C min⁻¹.

Density calculation

The samples with the same size (20 mm by 50 mm by 0.05 mm) were allowed to stand for at least 12 hours in an environment with RH 30% and 25°C and weighed using an analytical balance (Mettler Toledo, US) to a weight of M. Thicknesses and edge lengths were determined to calculate the volume of V.

$$\text{Density} \mathrm{\rho }=\frac{M}{V}$$ (1)

Water processing

First, the bioplastic is soaked in ordinary water (pH 7 and temperature of 25°C) for 10 min, then different shapes are fixed using a mold, and the bioplastic is dried naturally (≈1.5 hours) to obtain the designed shape successfully, which is water molding. By following the same soaking process, the joints of multiple wetted bioplastics are tightly fitted together and dried naturally, and the joints are water pasted successfully.

Water absorption calculation

The samples were cut into the same size, and the initial mass (M₀) was measured using an analytical balance. Then, samples were soaked in water for 10 min, and the mass (M) was measured again.

$$\text{Water absorption}=\frac{M-M_0}{M_0}\times 100\%$$ (2)

Simulation method of bistable structure deformation

The simulations on the monotonic compression deformation of the bistable structure were carried out using the finite element software (ABAQUS). The elastic modulus of the material was set to 2242 MPa, the yield stress was set to 26.7 MPa, the linear plastic hardening modulus was set to 604 MPa, and the Poisson’s ratio was set to 0.4. The eight-node linear hexahedral element (C3D8R) was adopted in the simulation. A static general analysis step was set with the Nlgeom switch turned on. When the deformation reaches a certain level, the connected rod will contact the rod at the bottom. Thus, the structure was set to contact, and the type of contact was surface-to-surface contact. Under the boundary condition, vertical displacement was applied to the top surface of the structure, and the bottom surface was fixed entirely. To measure the mechanical properties of the unit cell, the load-displacement curves were recorded and analyzed with Origin 2024.

Biodegradability test

The biodegradability of the BH bioplastic, PVC, and PET plastic was tested by cutting them into small pieces of 5 cm by 5 cm and burying them in natural soil at a depth of 10 cm with the temperature ranging from 10° to 30°C and pH of 8.30 ± 0.04 at a position of 30.31°N, 114.21°E. The samples were periodically removed from March to September 2023 for observation of biodegradability.

Recycling processes

The end-of-life BH bioplastic can be crushed into small pieces with an area of less than 1 cm² and then dispersed into a homogeneous suspension by stirring in water overnight (~12 hours) and again by natural water evaporation (~42 hours) for repreparation of the BH bioplastic.

Life cycle assessment (LCA)

The LCA analysis followed the ISO standard series 14040 and was conducted using ReCiPe 2016 based on Brightway2 and Activity Browser. The life cycle inventory data of the upstream production of chemicals, electricity, and water were collected from ecoinvent 3.7 and the literature. The functional unit was set to environmental impact per cubic centimeter per megapascal to represent both environmental impacts and functionality. The system boundaries included the acquisition of bee pollen and cotton residues, transportation, and BH bioplastic production. See the Supplementary Materials for details.

Statistical analysis

All experiments were performed in triplicate, and the results were presented as mean values ± SDs (n = 3).

Supplementary Materials

The PDF file includes:

Supplementary Text

Figs. S1 to S40

Tables S1 to S6

Legend for movie S1

References

Other Supplementary Material for this manuscript includes the following:

Movie S1