A comparative study on the comprehensive properties of natural microfibers isolated from the large-clustered bamboos in the southwest of China using steam explosion

ABSTRACT To develop sustainable functional fibers and expand their novel applications, the large-clustered dendrocalamus sinicus (DS) and the dendrocalamus giganteus (DG) planted in the southwest of China were effectively isolated by steam explosion (SE). The fine and uniform bamboo microfibers of DS and DG corresponding to the smallest average widths of 18.24 μm and 17.16 μm were obtained, respectively. The relative content of cellulose in two bamboo species had a marked increase after SE but a decrease for hemicellulose and lignin without any introduction of toxic chemical reagents. The SE treatment improved the thermal stability, the crystallinity, and the surface hydrophilicity of bamboo samples with their morphologies varying from rod-shaped strips to fibrous filaments. The degrees of crystallinity for DS and DG increased from 57.63% and 57.53% to 73.67% and 74.10%, respectively. The thermal stability, mechanical properties, and hydrophobicity of bamboo microfibers derived from DS were superior to those of DG, which showed a higher maximum decomposition temperature (5.24°C), tensile strength (181 MPa), elongation at break (1.1%), and water contact angle (7.8°).


Introduction
Synthetic fibers with complex processing technology, high energy consumption and cost as well as potential pollution are used in various engineering fields, such as construction, aviation, bridge, transportation, and sports . To alleviate the worsening global problem of energy shortage and environmental pollution, the development of high value-added sustainable biomass products is paid more attention in recent years . Bamboo is one of the most important and the fastest-growing non-woody biomass resources in the world (Bian et al. 2020;Dai et al. 2021). The bamboo species in China occupies more than one-third of the global bamboo species . Natural bamboo resources have been widely used in handcrafts, paper and textile industry, construction, and furniture because of its high productivity, easy propagation, and excellent mechanical properties (Muhammad et al. 2019;. Recently, natural bamboo fibers have been utilized in the fields of fiber-reinforced polymer composites, adsorption, and storage-energy functional materials due to their inherent advantages of lightweight and high strength, renewability and biodegradability, abundant availability as well as naturally high content of cellulose fibers compared to other biomass resources (Chin et al. 2020;Gupta 2020;Habibi and Lu 2014;Tanpichai et al. 2021;Zequine et al. 2016;Zhang et al. 2019).
Many methods have been used to isolate the bamboo fibers from natural bamboo resources . Steam explosion (SE) is an environmental-friendly and low energy consuming method which combines physicochemical and mechanical shearing processes without the use of toxic chemicals (Yu et al. 2022). SE can effectively break down biomass recalcitrance by degrading the chemical components, defibrillating the fibers via auto-hydrolysis at high temperature and humidity, and explosively decompressing the material via rapid pressure drop, which greatly improves the specific surface area and porosity of biomass (Sarker et al. 2021;. It is effective to isolate various biomass resources including pineapple leaves (Song et al. 2021), wheat straw (Han et al. 2010), rice straw (Boonterm et al. 2016), sugarcane bagasse (Phinichka and Kaenthong 2018), Hibiscus sabdariffa var. altissima (Karakoti et al. 2020), and bamboo .
The bamboo microfibers of bambusa beecheyama with the average width of 16.7 ± 4.8 μm was successfully prepared using SE. The average width of bamboo microfibers obtained by SE can be further reduced to 7.5 ± 3.2 μm when the alkaline treatment and bleaching process were introduced (Tanpichai, Boonmahitthisud, and Witayakran 2019). Boonterm et al. comparatively studied the difference of isolated fibers by chemical treatment and SE. The latter can break down the natural fibers more effectively with the smaller average diameter (110-260 μm) and the larger aspect ratio , also appeared the higher fiber yielding and remaining contents of lignin and hemicellulose in comparison with the former (Boonterm et al. 2016).
Dendrocalamus sinicus (DS), as one of the largest (in size) bamboo species, and the similar dendrocalamus giganteus (DG), are widely distributed in Yunnan province, the southwest of China. The DS is a special large-clustered bamboo with a stalk height of 30 m and diameter of 30 cm (Dong, Zhang, and Yang 2012) and has been termed "the king of bamboo." To address the increasing interests of novel natural bamboo fiber functional materials and expand its application across various fields, this study aims to develop a facile, low cost, and environmental-friendly way to prepare the bamboo microfibers from DS and DG via SE. The effects of three key parameters (temperature, residence time and pressure) of the SE process on the micro-morphology and distribution of diameters of bamboo microfibers were investigated. The optimal parameters of SE for two large-clustered bamboos were selected by orthogonal test. The variations on the main chemical composition, micro-morphology, crystallinity, thermal stability, surface wettability, and mechanical properties of bamboo microfibers after SE were comprehensively characterized.

Materials
The bamboo culms of DS and DG were harvested in 2019 from Yunnan Province, the southwest of China. They were air-dried to equilibrium moisture content (~15%) and cut into small strips with a size of 60 mm × 10 mm × 10 mm.

Orthogonal experiments of SE
To investigate the optimal key parameters of SE and to obtain bamboo fibers with the smallest width, three factors were selected as experimental variables. X: soaking time of bamboo strips, h; Y: pressure of the SE process, MPa; Z: residence time, s. Each factor had three levels, as listed in Table 1.
Specific experiments were performed on a SE instrument (QBS-80, Gentle Bioenergy Co. Ltd, China). After soaking, 25 g bamboo strips were placed into a 0.4 L SE chamber, and saturated steam was flowed into the chamber. The fibers were defibrillated due to the explosive decompression caused by the immediate pressure reduction.

Optical microscopy analysis
The width and distribution of bamboo microfibers after SE were examined using a LED light microscope (DM2000, Leica, Germany) equipped with a camera (DMC 4500, Leica, Germany).

Scanning electron microscope (SEM) analysis
The micro-morphological observations for the bamboos before and after SE were performed on an SEM (Zeiss Sigma VP, Germany) with a 5 kV acceleration voltage.

Fourier transform infrared spectroscopy (FTIR) analysis
FTIR spectra of the bamboos before and after SE were collected using a Fourier transform infrared spectrometer (Nicolet IS5, Thermo Fisher Corp., USA). Data was collected in the wave number range of 4000-400 cm −1 .

Chemical composition analysis
The variations in the major chemical components of the bamboos before and after SE were investigated according to the method described in the "Determination of structural carbohydrates and lignin in biomass" issued by the National Renewable Energy Laboratory (NREL).

X-ray diffraction (XRD) analysis
XRD measurements of the bamboos before and after SE were carried out on a diffractometer (Ultima IV, Rigaku, Japan) using Cu-kα as the X-ray source (λ = 0.15406 nm) and a 40 kV accelerating voltage. The degree of crystallinity for cellulose can be calculated by formula (1).
Where I 002 is the peak intensity of the (002) lattice diffraction at 2θ = 22.6°, which represents both the crystalline and amorphous region material, and I am is the diffraction intensity of amorphous fraction at 2θ = 18°.

Thermo-gravimetric (TG) analysis
The thermal stability of the bamboos before and after SE was evaluated by a thermogravimetric instrument (TG209-F1, Netzsch, Germany). About 5 mg specimens were heated to 800°C under nitrogen atmosphere at a heating rate of 10°C/min.

Water contact angle (WCA) measurement
WCA measurements of the bamboos before and after SE were carried out on OCA20 contact angle instrument (DataPhysics Instruments, Germany). The moisture contents of bamboo microfibers were controlled by conditioning treatment for one or two weeks prior to the WCA testing. About 10 g bamboo microfibers were placed in flat Teflon sheet and compressed them using a 500 g weight for 12 h in a desiccator. Four microliters of distilled water were dropped onto the surface of the specimens. The average WCA values were calculated from data at three different positions on each sample.

Test of mechanical properties
The electronic mechanical instrument (5583, Instron, USA) was used to test the tensile strength and elongation at break of bamboo microfibers referring to the ASTM D3379-75 standard. The samples with about length of 50 mm were placed between two grips with a gauge length of 20 mm and were tested at a crosshead speed of 1 mm/min.

Width distribution of bamboo fibers
The main process chart for the SE of bamboo is illustrated in Figure 1. The appearance images of bamboo microfibers for two species are shown in (Figures 2-4) It is observed that all samples were successfully isolated into bamboo microfibers by SE with various colors and width sizes. The average width values of two bamboo microfibers were measured using an LED light microscope. The differences on the average width and distribution of the bamboo microfibers as well as the light or dark yellow color were probably attributed to the major chemical composition and anatomical structure of two bamboo species. The average width and distribution of bamboo microfibers after SE for DS and DG are presented in (Figures 3-5), respectively. As shown in Figure 3, the average widths for the various orthogonal groups of DS were different. The significance factors on DS microfibers in SE experiments were ranked as Y (pressure) > Z (residence time) > X (soaking time) according to the analysis of the orthogonal experiment results. The largest and smallest average widths for DS were DS-G and DS-B, respectively, at 141.87 and 46.06 μm. Similarly, the significance factors on DG microfibers in SE experiments were ranked as X (soaking time) > Y (pressure) > Z (residence time). The largest and smallest average widths for DG were DG-C and DG-I, respectively, corresponding to the 196.70 and 99.86 μm in Figure 5. Generally, the bamboo microfibers of DS were thinner than those of DG. To obtain the finest bamboo microfibers, the optimal parameters for DS and DG were X 2 Y 3 Z 3 (24 h, 2.3 MPa, 350 s) and X 2 Y 3 Z 1 (24 h, 2.3 MPa, 250 s), respectively.

SEM analysis
The micro-morphology of the bamboo strips and isolated optimum bamboo microfibers were further examined by SEM. The cross-section of the bamboo tissue structures showed many vascular bundles consisting of hollow vessels surrounded by fibrous sclerenchyma cells without damage before SE (Dixon and Gibson 2014) (Figures 6a-d). The intact cell-wall structures including the thick-walled cell and the parenchyma for two bamboo species also had no significant morphological differences observed in longitudinal section (Figures 6b-e). Long bamboo microfibers can be effectively isolated when subjected to rapid depressurization during SE, as seen in (Figures 6c-f). Residual hemicellulose and lignin that act as binders were not found on the surface, which was due to the washing process by deionized water prior to SEM observation.

FTIR analysis
The FTIR spectra of two bamboo species before and after SE are shown in (Figure 7). All bamboo samples including strips and microfibers displayed the similar curves, implying that they had the similar chemical components. The broad adsorption bands in the range of 3700 to 3000 cm −1 are attributed to the stretching vibrations of -OH (Khawas and Deka 2016). The adsorption band at 1738 cm −1 is associated with acetyl and uronic ester groups in hemicellulose or the ester linkage of carboxylic groups of ferulic and p-coumaric acid in lignin or hemicellulose . The adsorption bands located at 1512, 1378, and 1254 cm −1 are assigned to the aromatic skeletal vibration of lignin. A small band at 2920 cm −1 is related to the stretching vibration of C-H in cellulose and hemicellulose (Phinichka and Kaenthong 2018). The band at 1040 cm −1 corresponds to the C-O-C stretching vibration in the pyranose ring and the glycosidic bonds of cellulose (Xu et al. 2022). The intensity of the adsorption bands at 3435, 2929 and 1040 cm −1 , 1378, and 1254 cm −1 were higher when two bamboo strips were subjected to SE, inferring the cellulose and lignin had the higher contents. The SE process consisted of a high-temperature cooking phase and the thermal decompression stage. The oligomeric and monomeric sugars were formed with partially soluble and decomposed small molecular substances like furfural due to the auto-hydrolysis of hemicellulose (Sui and Chen 2016). Ester bonds in carbohydrates and lignin were broken down during SE (Chen, Li, and Li 2014), which led to the melting and recondensation of lignin possibly covering the surface of bamboo microfibers.

Chemical composition analysis
The chemical composition variations of bamboo samples before and after SE is illustrated in Figure 8. Before SE treatment, the bamboo strips for two species showed moderate contents of hemicellulose and lignin and a high content of cellulose. The cellulose, hemicellulose, and lignin for the DS control and DG control were 51.95%, 17.41%, and 30.64% and 55.76%, 18.57%, and 25.67%, respectively. According to previous reports, the general compositions for most lignocellulosic biomasses are 25-45 wt% cellulose, 20-40 wt% hemicellulose, and 10-25 wt% lignin, respectively (Nanda et al. 2013), while for bamboo is typically 42-47 wt% cellulose, 22-24 wt% hemicellulose, and 23-31 wt% lignin He et al. 2014). This implied that DS and DG were two excellent raw materials for the development of novel fibrous materials. The relative content for cellulose showed a marked increase when the bamboo strips were treated by SE, from 51.95% to 65.17% for DS and 55.76% to 65.71% for DG. However, the hemicellulose content sharply decreased from 17.41% to 10.05% for DS, and from 18.57% to 11.37% for DG, respectively. There was a slight reduction in lignin content corresponding to   a decrease from 30.64% to 24.78% for DS and 25.67% to 22.92% for DG, respectively. During the SE process, hemicellulose was hydrolyzed by acetic acid derived from acetyl groups, lignin was depolymerized and recondensed, and the amorphous cellulose was disrupted (Yu et al. 2022). The fibrous structures were formed by the combined effects of water flushing, volume expansion, and tissue isolation, resulting in varying chemical components and redistribution of partial components. The similar variations on the contents of cellulose, hemicellulose, and lignin have been reported for pineapple leaf fibers and rice straw fibers (Chen et al. 2011;. This implies that the hemicellulose and lignin in various biomass resources could be partially eliminated by an eco-friendly steam explosion way instead of the chemical treatment using toxic solvents, which can effectively reduce the cost for the preparation of biomass fibers.   Figure 9 shows the thermogravimetric (TG) and derivative thermogravimetric (DTG) curves for DS and DG before and after SE at the optimized parameters. Generally, there were three stages of thermal decomposition for two bamboo specimens. The first stage corresponding to a 3-4% mass loss was likely attributed to residual moisture evaporation. The main thermal decomposition was in the range of 220-450°C due to the cleavage of glycosidic linkages in cellulose, the rearrangement and cracking of cellulose, and the decarboxylation and decarbonylation of residual hemicellulose and partial lignin. The last stage over a broad temperature range of 450-800°C was mainly associated with lignin decomposition over several oxygenated functional groups and chain scission (Wang et al. 2019(Wang et al. , 2019. The maximum decomposition temperatures (T max ) corresponding to a maximum thermal weight loss rate of the bamboo samples are displayed in Figure 9b. There were remarkably elevated T max values for DS and DG by SE treatment compared to their control samples. The specific temperature increased from 322.94°C and 337.87°C to 363.30°C and 358.06°C, respectively, corresponding to increases of 12.50% and 5.98%, respectively. This can be attributed to the fact that the conversion and removal of the hemicellulose and the relatively higher cellulose contents with a high molecular weight and an ordered arrangement delayed the thermal decomposition of bamboo microfibers Vänskä et al. 2016), which is well consistent with the major chemical composition analysis. Figure 10 shows the XRD profiles of bamboo samples before and after SE. Two characteristic peaks corresponding to 2θ = 15.8° and 22.2° are related to the (110) and (002) crystalline planes, respectively. The peaks for all bamboo samples before and after SE were verified as diffraction peaks of the cellulose I structure (Reis et al. 2020), which indicated that SE treatment was unable to change the crystalline structure of cellulose. This is consistent with previous reports for other biomasses. After treatment, the sharper shape of the peak at 22.2° was due to the increasing crystallinity with the efficient removal of noncellulosic polysaccharides and the solvation of amorphous zones (Kaushik and Singh 2011).

XRD analysis
As reported by a previous study, this effect was similar to the acid treatment (Cherian et al. 2010). Crystallinity is a critical parameter for evaluating the effect of SE. The degrees of crystallinity for DS and DG were 57.63% and 57.53%, respectively. Whereas they increased to 73.67% and 74.10% after SE treatment. The increased crystallinity was due to the high-temperature steam restructuring the amorphous and paracrystalline cellulose regions. The strains that arise from the crystallization phase in biosynthesis of native cellulose as well as the cell wall formation by cellulose interacting with hemicellulose and lignin were effectively released (Cherian et al. 2008;Kaushik and Singh 2011). In addition, the weak peaks at 9.5° and 11.2° may correspond to the crystalline cellulose exposed on the surface of bamboo fibers after SE treatment (Xu et al. 2018). The similar increasing trend for the crystallinity of steam exploded pineapple leaves also has been reported . Nevertheless, the steam explosion with a longer time and more treatment cycles may lead to the decrease of the crystallinity of microfibers due to the decomposition of cellulose.

WCA analysis
WCA measurements for loose bamboo microfibers were carried out using the sessile drop method on the compressed flat surface of fibers. As presented in Figure 11, these images of water droplets clearly displayed the variations in surface wettability of various bamboo specimens. It can be noticed that there was an obvious difference in the control samples of DS and DG. The former had a higher water contact angle value (85.3°) than that of the latter (77.5°), corresponding to weaker hydrophilicity. However, the hydrophilicity greatly increased when the bamboo strips were treated with SE and showed WCAs decreased to 64.8° and 65.3°, respectively. The bamboos composed of the major chemical components (cellulose, hemicellulose, and lignin) have an inherent recalcitrance with the natural self-assembling structure, in which the cellulose is wrapped by lignin and hemicellulose. The softening, partial degradation, redistribution, and recondensation were occurred for the lignin during the SE process, leading to more hydrophilic cellulose being exposed on the surface of microfibers (Spiridon 2020). Additionally, impurities, such as silicon, in bamboo were reduced after treatment (Boonterm et al. 2016). These results demonstrate that the SE treatment had a positive effect on increasing the hydrophilic properties of the sample surface. This is consistent with the above-mentioned chemical composition and FTIR analysis.

Analysis of mechanical properties
The mechanical properties of bamboo fibers after SE are presented in Figure 12. It is known that the cellulose fibers in bamboo are made up of many cellulose molecules in an extended chain conformation, resulting in the outstanding mechanical strength. The mechanical properties of bamboo microfibers generally are determined by the chemical composition, the microfibrillar angles and the special layered structure of fiber cell walls (Zhang, Wu, and Qiu 2010). The stress-strain behavior under constant tensile loading appeared the linear elastic action before fracture in Figure 12a. The microfibers after SE derived from DS displayed a higher tensile strength (662 MPa) and elongation at break (6.2%) than that derived from DG in Figure 12b. The physical microfibrillar angle and the chemical composition of the bamboo microfibers are the major factor that affecting their mechanical properties (Mohanty, Misra, and Drzal 2005). It can be deduced that the relatively high cellulose and lignin contents but a relatively low hemicellulose content were conductive to the elevation of the mechanical properties of natural fibers (Park et al. 2006). Additionally, the weak interfacial bonding between the broad and narrow lamellae also facilitated the slipping of microfiber cell structures during tension, which is consistent with previous reports (Yu et al. 2014).

Conclusion
The fine and uniform bamboo microfibers derived from the DS and the DG planted in southwest of China were successfully isolated by SE. The smallest average widths of DS and DG corresponding to 18.24 μm and 17.16 μm can be obtained by the optimum process parameters. The relative contents of cellulose in two bamboo species had a marked increase after SE but a decrease for hemicellulose and lignin without any introduction of toxic chemical reagents. The SE treatment improved the thermal stability, the crystallinity, and the surface hydrophilicity of bamboo samples with their morphologies varying from rod-shaped strips to fibrous filaments. The degrees of crystallinity for DS and DG increase from 57.63% and 57.53% to 73.67% and 74.10%, respectively. The DS microfibers had a higher thermal stability, mechanical performance and water contact angle than the DG microfibers. Generally, the microfibers with good mechanical properties from two bamboo species are excellent candidates for novel natural fiber composite functional materials in the future.

Highlights
• The natural microfibers derived the largest bamboo species are firstly obtained by SE.
• The smallest average widths of DS and DG are 18.24 and 17.16 μm, respectively. • The contents of cellulose increase but hemicellulose and lignin decrease after SE.
• The thermal stability, crystallinity, and hydrophilicity of microfibers are improved. • DS has a higher T max by 5.24°C, tensile strength by 181 MPa, and elongation at break by 1.1% than DG.