Engineering transverse cell deformation of bamboo by controlling localized moisture content

Abstract

Bamboo’s native structure, defined by the vertical growth pattern of its vascular bundles and parenchyma cell tissue, limits its application in advanced engineering materials. Here we show an innovative method that controls localized moisture content to shape natural bamboo into a versatile three-dimensional (3D) structural product. Different temperatures along the transverse direction of bamboo were used to induce directional water transport within the bamboo, so that the distribution of internal stress was shifted from the bamboo surface to the inner layers. The internal stress shifting enabled the control of the transverse deformation. After densification, a 3D-molded bamboo product was obtained that retained the natural heterogeneous structure. The molded bamboo had a high specific strength of 740.58 MPa·kg⁻¹·m³ and impact resistance of 2033.29 J/m, surpassing most renewable and nonrenewable engineering materials. The life cycle assessment revealed that replacing metals and polymers in structural materials with 3D-molded bamboo significantly reduces carbon emissions. Our proposed “localized moisture gradient-driven uneven drying” strategy represents a sustainable path in transforming natural bamboo into high-performance engineering materials.

Bamboo’s native structure, defined by the vertical growth pattern of its vascular bundles and parenchyma cell tissue, limits its application in advanced engineering materials. Here the authors show a method that controls localized moisture content to shape natural bamboo into a versatile three-dimensional structural product.

Introduction

Enhancing the structural capabilities of engineering materials that possess sustainable and renewable sources presents significant challenges across various industries^1–5. Recent advances in forestry engineering have been concentrated on unlocking the structural potential of natural materials, such as wood and bamboo^6–11, for nontraditional applications such as nanoenergy^12–15, cell wall engineering^7,16–19, and addressing the challenges of material strength and toughness^20–22. The competition between the use of wood and bamboo has been ongoing for thousands of years, especially given the presence of alternatives such as metals, engineering plastics, and ceramics.

There are more than 1500 species of bamboo distributed primarily across Asia, the Americas, and Africa (Fig. 1a, b), which illustrates bamboo’s extensive adaptability and accessibility^23,24. Bamboo matures in only 3–5 years, a duration significantly shorter than the 20–60-year growth cycle of traditional wood, underscoring bamboo’s potential for mass production at low cost and its advantages in terms of biomass, carbon storage, growth rate, and annual yield (Fig. 1c, d). However, the cylindrical structure of bamboo limits its use in advanced engineering structures (Supplementary Fig. 1). By changing the inherent structure of bamboo created by its vertical growth pattern, one can not only enhance its functionality as an engineering material but also facilitate profitable carbon peaking and carbon neutral initiatives.

Fig. 1. The global impact of three-dimensional bamboo molding processing.

a The global supply and demand pattern of bamboo products, with the EU, the United States, and Japan being the main importing countries, and China and the EU being the main exporting countries. b Bamboo resources in China are mainly distributed in Fujian, Jiangxi, Hunan, Zhejiang, Sichuan, Guangdong, Guangxi, and Anhui provinces. c Biomass and carbon storage of major countries with global bamboo forests. d Comparison of growth rate and annual production between bamboo and fast-growing wood. e An overview of the uneven drying process of a bamboo model, showing the process from cell deformation to macroscopic geometric deformation based on local moisture content changes caused by temperature difference. f Large-scale production of molded bamboo from natural bamboo.

With variable success, investigators have used different methods of forming bamboo into a flat shape^25,26. Traditional bamboo processing involves cutting the round bamboo into small pieces and then bonding these pieces into laminated composites; however, the utilization rate of this method is usually only around 35% (excluding curved reconstituted bamboo)²⁷. Flattening techniques can improve the utilization rate, for instance, by directly flattening curved bamboo or by hot-pressing bamboo after hydrothermal softening^25,28,29. “Slit flattening” takes this one step further, using saturated steam at 160–180 °C to soften the bamboo before flattening it, effectively mitigating some of the physical and mechanical defects in bamboo^30,31. This flattening process overcomes the obstacles of transverse processing–namely, the difficulty of splitting bamboo without cracks–and can increase the utilization rate to 55–87%^21,32, and is commonly employed in bamboo composite flooring, veneers, and engineered furniture³³. Reconstituted bamboo can reach usage of more than 80%, although reconstitution necessitates breaking down the bamboo tissue and bonding it with glue³⁴. It is worth noting that previous studies have significantly enhanced the mechanical properties of bamboo through densification or delignification^20,21, but these methods have not resolved the inherent conflict between structural integrity and plasticity. In contrast, the “water shock” process⁷, effective for shaping wood, is unsuitable for bamboo (Supplementary Figs. 2a–d), posing a challenge to enhancing bamboo resource use and its transformation into engineering material.

From a structural standpoint, bamboo exhibits pronounced anisotropy in the radial direction. The vascular bundles are denser and smaller on the external “green” side, whereas they are sparser but larger on the internal “yellow” side (Supplementary Fig. 1). This structural difference leads to non-uniform drying and shrinkage behavior; the radial shrinkage of bamboo is often greater than its tangential shrinkage, and the external surface typically shrinks more significantly³⁵. In the initial stage of high-humidity drying, moisture evaporates simultaneously from the cell walls and lumens; when the moisture content drops to about 40%, large-scale shrinkage has not yet occurred, but cracks are likely to appear. As it continues to drop below around 40%, radial shrinkage becomes dominant, and the final moisture content generally stabilizes at about 5%³⁶. Removing the bamboo’s outer (green) layer, inner (yellow) layer, and nodes can speed up drying; however, bamboo with intact nodes tends to crack at higher moisture content (29.4–47.5%), while node-free bamboo cracks at a lower moisture content (6.9–23.3%). Various pretreatment methods (e.g., boiling, carbonization) also affect subsequent drying performance; properly treated bamboo can retain 35–50% moisture content and is less prone to cracking or warping under convective drying, whereas hydrothermal treatment can dissolve certain cellular contents, resulting in greater overall shrinkage^27,37. Considering the significant influence of bamboo microstructure and drying conditions on drying performance, extensive research has focused on selecting and optimizing drying processes and parameters (e.g., air drying, kiln drying, microwave drying, hot pressing, and far-infrared drying). These efforts aim to efficiently remove moisture while minimizing drying stress, ultimately ensuring dimensional stability, adhesive bonding strength, and resistance to decay and mildew in bamboo products³⁸.

The processing of bamboo products to achieve dimensional stability requires uniform drying to reduce moisture content and enhance durability^20,39,40. Variations in temperature, steam pressure, and moisture content during drying can cause differential moisture migration and uneven drying stresses. These stresses may result in cell wall shrinkage and warping. By continuously adjusting temperature and humidity, we can control these internal stresses and achieve uniform drying. Theoretically, precise control of temperature and moisture distribution enables targeted manipulation of drying stresses, microscopic cell wall contraction, and macroscopic bamboo deformation.

Based on the aforesaid uniform drying theory, we present an innovative uneven drying process to mold the bamboo into a three-dimensional structure. We initiate deformation in the local cell walls of the bamboo by exploiting the internal stress differences caused by uneven drying, which then propagates from both sides (Fig. 1e). The molding process is implemented by using temperature to control moisture movement at designated points of the bamboo. The moisture-oriented evaporation and migration enable the control of bamboo deformation, which produces the desired rearrangement of bamboo’s transverse cell structure. As a result, the deformed bamboo is molded into a final product having enhanced mechanical properties (specific strength of 740.58 MPa·kg⁻¹·m³ and impact resistance of 2033.29 J/m). In this report, we outline the engineering process for manipulating moisture content in bamboo’s transverse layers, which relates to the internal stress alterations and cell deformation necessary for three-dimensional molding. We have also applied our “uneven drying” technique to other materials with similar structures, such as poplar wood, larix gmelinii, and mifei bamboo, further demonstrating the versatility and effectiveness of our method (Supplementary Fig. 2e-f). We further demonstrate large-scale fabrication of molded bamboo products (Fig. 1f), suggesting their potential for industrial scale production. Our proposed “localized moisture gradient-driven uneven drying” strategy to transforming bamboo into a sustainable building material paves a new way for using bamboo resource and for promoting environmental sustainability.

Results

Design and manufacturing of three-dimensional molded bamboo

As depicted in Fig. 1e and Supplementary Fig. 3, we achieved a 360° transverse rotation of the bamboo model by using the described technology. This structural transformation involved three steps: partial delignification, uneven drying, and compression molding. Delignification optimized the cell wall structure to provide space for transverse cell deformation driven by drying stresses (Supplementary Figs. 4–6). This is because increased cell wall gaps provide greater freedom for transverse deformation and reduce the inter-wall bindings, which facilitates cell deformation under drying stress^20,41,42. To achieve controllable transverse deformation of bamboo, as shown in Fig. 2a, b, we regulated the local asymmetric temperatures on both sides of the bamboo. This uneven drying strategy produced precise heat transfer, enabling design of moisture migration paths to control the occurrence and development of drying stress, which in turn promoted cell deformation. That is, the key to control bamboo cell deformation is to control local water content. In our typical experiment (Supplementary Fig. 7), a bamboo stem with a height of 5 cm, a diameter of 8 cm, and a wall thickness of 0.9 cm was cut in half. Directional evaporation and migration of water was controlled by setting a 50 °C temperature difference between the two sides of the bamboo. After 600 min, the bamboo expanded into a flat shape.

Fig. 2. Principle of uneven drying process.

a, b The process diagram of bamboo being flattened when ΔT = 50 °C. c An uneven drying model. d The model was divided into different regions, along the X-axis (bamboo thickness direction) into I, II, III, IV, and V Regions, and in the YZ plane into sections 1, 2, 3, 4, 5, 6, 7. e–h Bamboo moisture distribution during uneven drying at ΔT = 50 °C. i–m Internal stress distribution of bamboo during uneven drying at ΔT = 50 °C, with color mapping corresponding to the moisture distribution shown in (d). n–q Processing bamboo into 90°, 180°, 270°, 360° virtual image display.

To analyze the transition from cell deformation to macroscopic deformation, we constructed an analytical model for uneven drying of delignified bamboo at a temperature difference of 50 °C, as illustrated in Fig. 2c, d. The model was divided into five regions along the thickness direction and into seven regions perpendicular to the thickness direction. The distributions of moisture and internal stress were quantified across these regions during the uneven drying process (Fig. 2e–m and Supplementary Fig. 8). The principle of uneven drying involves free water in the bamboo migrating via the large capillary system to the evaporation interface, driven by capillary tension and water vapor pressure from heating, and subsequently evaporating into the air. Supplementary Fig. 9 shows that after lignin removal, the transverse direction of the bamboo facilitates easier transfer of free water. Under the influence of heat, the bound water in bamboo was absorbed by the microcapillary system, transported, and evaporated⁴⁰. Consequently, driven by gradients of water content, temperature, and water vapor pressure, the bound water in bamboo continuously migrated and diffused from the bamboo interior to the evaporation interface, as both liquid and vapor. When the moisture content of the middle side (4 section) of the bamboo surface (Region I) close to the high temperature dropped below the fiber saturation point (FSP, ~30% moisture content), the bamboo began to shrink. However, because the moisture content of each inner layer was higher than the FSP, the surface layer was subject to compression due to restraint, whereas the inner layer was subjected to tension at the same time. As the drying progressed, the internal tension in bamboo gradually shifted to compression as moisture moved from regions of high to low moisture content. For instance, tension in Region II transitioned to compression after 300 min of drying (Fig. 2j), stress in Region III shifted after 450 min (Fig. 2k), and stresses in Regions IV and V adjusted after 600 min (Fig. 2l, m). This stress direction change induced the transverse deformation of the bamboo. Larger temperature differences (ΔT = 100, 150, 200, 250 °C, Supplementary Fig. 7) required less time for flattening bamboo.

Finally, a compression molding was applied when the bamboo was at the FSP to obtain the molded bamboo. The FSP molding process was designed to maintain an orderly arrangement of bamboo cellulose during densification, ensuring that the molded bamboo preserved its natural heterogeneous structure, as shown in Supplementary Figs. 10 and 11. Therefore, by designing the drying process for different layers of bamboo, various shape processing of the transverse organization of bamboo can be obtained as depicted in Fig. 2n–q, for bamboo molded into “90°, 180°, 270°, 360°, and N, E, F, U” (Supplementary Fig. 12), respectively.

We used scanning electron microscopy (SEM) to examine the integrity of transverse bamboo cells during drying (Supplementary Fig. 13). Supplementary Fig. 14 shows that we subjected bamboo samples to uneven drying under natural sunlight, controlling their transformation from cylindrical to flat states.

Mechanical properties of molded bamboo

Considering the specific shape requirements of engineering materials^43,44, such as honeycombs being denser than cylinders (Supplementary Fig. 15), after FSP compression molding, we processed bamboo samples into triangles, squares, pentagons, and hexagons (Fig. 3a–d). The molded bamboo retained its heterogeneous structure and improved the bonding strength between bamboo macro-fibers by numerous hydrogen bonds and Van der Waals forces, thereby further enhancing its mechanical properties. The tensile strength of molded bamboo improved from 193.7 MPa for natural bamboo to 875.4 MPa, bending strength from 173.6 MPa to 375.3 MPa, compression strength from 73.7 MPa to 102.7 MPa, and impact resistance from 988.1 J/m to 2033.3 J/m (Fig. 3e–k). And the density was 1209–1315 kg/m³ (Fig. 3l, and Supplementary Table 1).

Fig. 3. Mechanical properties of molded bamboo.

a–d Triangular, quadrilateral, pentagon, and honeycomb molded bamboo. e–f Tensile properties of natural and molded bamboo. g, h Compression properties of natural and molded bamboo. i, j Flexural properties of natural and molded bamboo. k Impact properties of natural and molded bamboo. l Density of natural and molded bamboo. m Stiffness vs. density of molded bamboo compared with other materials. n Strength vs. density of molded bamboo compared with other materials. Data in (e, g, i, k, and l) are reported as their means ± SDs from n = 5 and n = 5 independent samples, respectively. Source data are provided as a Source Data file.

In structural design, emphasis is placed on the strength, modulus, density, and other parameters of the material^45,46, with having a light weight as an important factor. For instance, as shown in Fig. 3m, Carbon Fiber Reinforced Plastics (CFRP) in composite materials represent a particularly attractive class of materials because of properties like longitudinal wave speed of 10⁴m/s, which contribute to the growing use of CFRP in aerospace. Additionally, ceramics (Fig. 3n) offer strength per unit weight that is equivalent to metals and extremely strong stiffness per unit weight, but brittleness and toughness limit use of ceramics as structural material. In contrast to polymers, which have low densities, metals often have very high densities and high prices. For molded bamboo, its specific stiffness approaches typical values for advanced materials, the maximum specific strength reached 740.58 MPa kg⁻¹ m³. Importantly, it is considerably lower in price compared to metals, polymers, and other engineering materials (Supplementary Fig. 16, and Supplementary Tables 2–5).

Additionally, in the construction industry, effective moisture resistance is a primary prerequisite for bamboo materials⁴⁷. Consequently, the dimensional stability of molded bamboo necessitates investigation, as illustrated in Supplementary Figs. 17, 18. We conducted compression rebound tests on molded bamboo and heat-treated bamboo, with and without a polyurethane coating, at a relative humidity of 85%. Coated heat-treated bamboo exhibited a rebound rate of up to 45.6%, whereas coated molded bamboo demonstrated good dimensional stability with a rebound rate of 1.20% (Supplementary Fig. 17). The molded bamboo was further tested in water immersion experiments, and the results showed that the coated molded bamboo was stable for 24 h (Supplementary Fig. 18).

SEM was used to study the cell wall microstructural changes from the initial state to the compressed state and then to the rebound state. The extraction of lignin from the molded bamboo led to the depletion of a significant amount of hemicellulose, which led to the loss of the cell wall expansion function (Supplementary Figs. 17f–m) and good dimensional stability. These findings indicate that rapidly-growing bamboo can be processed into a deformable structural material possessing both strength and toughness, making it highly suitable for applications such as lightweight construction.

Chemistry and hydrogen-bond binding mechanism of molded bamboo

The superior mechanical properties of molded bamboo can be attributed to its densely laminated structure^4,48–50, as illustrated in Fig. 4a. Macroscopically, the process of uneven drying caused bamboo to shrink more, and the bamboo became less porous after densification⁴⁸. For example, SAXS results (Fig. 4b–d) demonstrated that Q values in the Guinier region of molded bamboo were significantly reduced compared with natural and delignified bamboo; thus, compression changed the large-scale structure of bamboo, potentially closing some pores or channels due to increased density¹⁴. Furthermore, 2D SAXS images revealed that the long, aligned cellulose microfibrils in the cell wall were well preserved^17,18, resulting in enhanced uniformity and orderliness within the molded bamboo’s internal structure. At the mesoscale, molded bamboo retained the heterogeneous structure of natural bamboo and formed a denser laminated structure after densification. At the micro and nanoscales, the removal of lignin exposed more surface area of the cellulose microfibrils^4,9,49, potentially enhancing fiber interactions and thereby increasing the density of hydrogen bonds. Figure 4e illustrates the evolution of cellulose, hemicellulose, and lignin contents during preparation. Additionally, Raman spectroscopy indicated a decrease in lignin at the 1600 cm⁻¹ peak and an increase in cellulose at the 2800 cm⁻¹ peak (Supplementary Fig. 5). Fourier transform infrared spectroscopy further confirmed the increased density of hydrogen bonds and the role of hydroxyl groups in the molded bamboo processing (Fig. 4f).

Fig. 4. Structure and composition of molded bamboo cellulose during preparation.

a Hierarchical structure of bamboo from macroscale to molecular scale. b–d The SAXS pattern of natural bamboo and molded bamboo. e Evolution of the cellulose, hemicellulose, and lignin content from natural bamboo to molded bamboo. f The FTIR spectra of natural bamboo to molded bamboo process. g The deconvoluted FTIR spectra of natural bamboo, delignified bamboo, and molded bamboo. h The XRD curves of natural and molded bamboo. i The ¹³C CP/MAS NMR spectra of natural and molded bamboo. Values in (e) represent their means ± SDs from n = 5 independent samples. Source data are provided as a Source Data file.

Throughout the transformation from natural bamboo to molded bamboo, there was a gradual increase in the quantity of free hydroxyl groups and at cellulose positions 2, 3, 6, (Figs. 4g, Supplementary Table 6). These orderly arranged bamboo macrofibers, primarily composed of cellulose, formed numerous hydrogen bonds with the bound water in the cell walls^21,51,52. The binding energy of cellulose/cellulose/water is reported to be 2.57 times that of cellulose/cellulose^53,54; the greater binding energy of cellulose/water provides the basis for the excellent mechanical properties of molded bamboo using water as an adhesive. The crystallinity of molded bamboo was higher than that of natural bamboo because of treatment with peroxyformic acid, a mild delignification method^55,56 (Fig. 4h). The ¹³C CP/MAS NMR spectra (Fig. 4i) revealed the effects of delignification, with signal reductions at 152–110 and 56.0 ppm indicating the partial loss of aromatic structure and methoxyl groups in lignin. Decreased peak intensities at 172.6 and 20.7 ppm, and between 60 and 105 ppm, suggested that hemicellulose was partially removed. Cellulose signals persisted, indicating preservation of the cellulose molecular structure during chemical treatment.

Environmental benefits of three-dimensional molded bamboo

In investigating the replacement of traditional metals and plastics with bamboo, we assessed the environmental benefits of molded bamboo to support the achievement of carbon neutrality and reduction targets. Using a cradle-to-gate life cycle assessment, we evaluated the carbon intensity and acidification potential of molded bamboo production and conducted normalized analysis of the process (Figs. 5a, b, and Supplementary Fig. 19, Supplementary Tables 7–11). The carbon emission from producing 1 kg of molded bamboo was 5.24 kg CO₂ eq., and sulfur dioxide emission was 0.0256 kg SO₂ eq. Compared with common building materials, this bamboo processing method effectively reduced energy and resource consumption, thereby lessening environmental burdens. Figure 5c,d and Supplementary Tables 12–17 illustrate the complete life cycle emissions of molded bamboo, from cradle to grave, accounting for its environmental impact during use and life die stages. Compared with recyclable materials such as alloy, steel, polyethylene, and polyphenylene sulfite, molded bamboo exhibited a lower environmental impact across nearly all categories, especially in terms of fossil resource scarcity, freshwater ecotoxicity, marine eutrophication, terrestrial acidification, and water consumption. This extensive low environmental impact would help to achieve global carbon peak and carbon neutrality. The extensive benefits of molded bamboo likely arise from its bio-based and renewable nature, necessitating less fossil energy to produce and minimizing ecosystem impacts while using water resources more efficiently^15,47,49.

Fig. 5. Life cycle analysis of molded bamboo production.

a The process flow diagram of the life cycle analysis system boundary. b Life cycle emissions from cradle-to-gate are compared with normalized data. c Environmental impacts per cm³/MPa based on the specific tensile strength for molded bamboo prepared by uneven drying process. d Environmental impacts per cm³/GPa based on the specific stiffness for molded bamboo prepared by uneven drying process, normalized to the higher impact material for each environmental impact category. (Note: Fine particulate matter formation—PMF, Fossil resource scarcity—FRS, Freshwater ecotoxicity—FWET, Freshwater eutrophication—FWE, Global warming—GW, Human carcinogenic toxicity—HCT, Human non-carcinogenic toxicity—HNCT, Ionizing radiation—IR, Land use—LU, Marine ecotoxicity—MET, Marine eutrophication—ME, Mineral resource scarcity—MRS, Ozone formation and Human health—OFHH, Ozone formation and Terrestrial ecosystems—OFTE, Stratospheric ozone depletion—SOD, Terrestrial acidification—TA, Terrestrial ecotoxicity—TET, Water consumption—WC, Alloy and Steel are in a 75% high recycling rate scenario, and PE and PPS are in a 30% recycling scenario).

Furthermore, when we considered the cost-effectiveness of molded bamboo (Supplementary Table 18), we constructed lightweight rigid beams to assess cost metrics. In addition to satisfying the criterion of light weight, molded bamboo’s economic competitiveness surpassed that of stone, wood, cast iron, and steel. For thin bamboo, processing methods are demonstrated in Supplementary Fig. 20, particularly rotating/cutting bamboo, which requires less pretreatment and drying time, providing favorable conditions for large-scale preparation of bamboo molds.

Discussion

Using an uneven drying technique, we show that natural bamboo can be molded into multi-functional structural materials by proper control of the moisture content of transverse bamboo layers. The control of moisture content in different layers shifts the distribution of internal stress, facilitating the molding of bamboo transverse cells. Delignification of bamboo promotes molding efficiency. The products molded at the FSP exhibited high specific strength (740.58 MPa kg⁻¹ m³) and impact resistance (2033.29 J/m). The uneven drying technique overcomes the challenges associated with bamboo’s lateral cells molding, enabling the bamboo to be used as products that require complex shapes. The life cycle evaluation proved that production and use of molded bamboo have relatively low emissions. The cost-effectiveness analysis and dimensional stability under high humidity suggested that molded bamboo has great potential in structural and nonstructural applications.

Methods

Descriptions of the methods are provided in the Supplementary Information.

Author contributions

T.B., J.Y. and J.Q.L. contributed equally to this work. T.B., J.Y. and J.Q.L. conceived and designed the study, performed the core experiments, analyzed the data, and drafted the initial manuscript. J.Z. supervised the life‑cycle assessment and provided critical input on data interpretation and manuscript revision. H.Y. and X.H. carried out the large‑scale molding experiments, conducted SAXS and solid‑state NMR measurements, and assisted with data analysis and manuscript editing. S.G. and Z.T. performed SEM imaging and mechanical testing and curated the resulting datasets. S.Q.S. and G.H. proposed the uneven‑drying concept, advised on materials selection and offered continuous feedback throughout the project. J.L. provided theoretical guidance on bamboo cell‑wall mechanics, supervised the experimental strategy and critically reviewed the manuscript. W.C. and D.W. secured funding, coordinated project resources, and contributed to manuscript editing. C.C. broadened the experimental scope, oversaw research progress, and, together with W.C., D.W. and G.H., finalized the manuscript. All authors discussed the results, commented on the manuscript and approved the final version.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Data availability

The data supporting the findings of this study are available within the article and its supplementary files. Any additional requests for information can be directed to, and will be fulfilled by, the corresponding authors. Source data are provided with this paper.

Competing interests

H.G.P., T.B., J.Y., J.Q.L., H.Y., X.W.H. and W.L.C. are inventors of the granted Chinese patent CN116494340B, which is related to the subject of this study. The authors declare no other competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-025-59453-3.