Introduction

Weathering is a phenomenon that occurs slowly over time, which causes deterioration in both aspect and properties of materials, due to irreversible changes in their chemical properties1. The main factors of weathering process are light/UV radiation, temperature and oxygen, causing specific chemical degradation processes, including photolysis, thermolysis and oxidation. Understanding of the weathering process contributes to degradation behavior and affects the use of materials.

As a kind of biomass material such as bamboo is good light absorber due to special chemical characters of chromophores including phenolic hydroxyl groups, aromatic skeleton, double bonds and carbonyl groups2. Especially, the color transformation is resulted from the chromophoric structures (aromatic compounds)3. The bamboo scrimber is a novel bamboo-based composite material that is produced by the directional recombination of oriented bamboo fiber mats and phenolic resin, which overcomes the limitations of bamboo. When the bamboo scrimber is applied for outdoor use, the weather causes the most damage on its surface. In the early stage, photodegradation can cause surface yellowing and darkening, increased surface roughness, and appearance of small cracks, thereby compromise its esthetic value4. In addition, rainwater can cause erosion that further accelerates the photodegradation process5, where cracks gradually expand into the interior of bamboo materials. Light is continuously reflected and refracted within the cracks, which initiates additional photodegradation on the newly exposed surfaces inside the materials. The accelerated artificial weathering is often conducted in the study of bamboo-based composite, such conditions cannot fully and accurately reflect the true outdoor weather resistance of the composite materials6. However, some previous studies mainly focused on the influence of the preparation process on the surface photostability of bamboo-based composites. Dyeing treatment can improve the mass and thickness loss rate of bamboo fiber composite during the natural weathering process, but it has a negative effect on its color stability7. Surface finishing also helps to improve the photostability of bamboo-based composites. The acrylic-based coating containing 5 wt% UV absorbers can effectively improve the photostability of the bamboo scrimber substrate and delay the photodegradation of the coating itself8. Although our team has previously investigated the changes in surface color and mechanical properties of the bamboo-based composite after two years of natural weathering9, surface chemical analysis and microstructural analysis were not yet conducted more than two years observation. Therefore, a comprehensive evaluation of chemical change and surface properties during long-term exposure to natural weathering will provide important implication into outdoor applications.

This study aimed to investigate the natural weathering effect of bamboo scrimber in terms of the chemical changes of surface. To determine the consequence of surface changes during six years, the relationship between color and chemical composition during degradation process was established to obtain a better understanding of the mechanism of natural weathering in outdoor bamboo scrimber. Moreover, the experimental samples were placed in the different natural regions, and variety of samples with long-term measurement were compared and analyzed by method of FTIR, XPS, SEM. The results would be given a useful support for improve the performance of bamboo scrimber, which could be applied in different natural environment.

Results and discussion

Color change

The occurrence of the color change on the surface of the bamboo scrimber samples is related to an increase the number of chromophores during outdoor exposure time (Fig. 1b). The total color change of the samples as an exposure time is given in Fig. 1d. The color changes (ΔE*) were increased from 0 to + 18.19 after two years of exposure to outdoor and then maintained in a steady state. The tangential surface of bamboo scrimber was bright and had a color between red and yellow (Fig. 1d). It was reported that the quinone formed by the photodegradation of lignin is the main chromogenic group that causes discoloration of wood materials, including α-quinone and β-quinone (Fig. 1c), which displays yellow and red10. The weathering measurement for two years led to obvious darkening and a gray appearance. The bamboo scrimber did not continue to darken after four years. This was attributed to the effects of moisture and fungal growth. Specifically, most of the soluble lignin photodegradation products on the surface of samples were washed away by rain11, thereby leaving relatively intact crystalline cellulose12 that was not susceptible to UV light. Further, fungal erosion under humid conditions can lead to a gray-white surface of lignocellulosic materials13. Significantly, the variety of both values obtained in the south region was higher than that in the north region. The humidity and rainfall content that was beneficial for fungal growth in the south was higher than that in the north region, thus the degradation of lignin was occurred with fungal action simultaneously.

Fig. 1: The manufacturing process, weathering process and the total color difference.
Fig. 1: The manufacturing process, weathering process and the total color difference.The alternative text for this image may have been generated using AI.
Full size image

a Manufacturing process of bamboo scrimber. b Weathering process of bamboo scrimber. c The main approaches for forming colored substances (α-quinone and β-quinone). d ΔE* results for 6 years between two regions: (●) the north and (■) the south.

FTIR spectra analysis

The FTIR spectroscopy is a very useful method for evaluating the surface of chemical components, and the chemical changes in bamboo materials induced by natural weathering especially solar radiation. The most representative FTIR bands were observed within a spectral range of 1800–800 cm−1, as summarized in Table 114. Figure 2-spectrum a and b show the FTIR spectrosciot of bamboo scrimber samples placed in the north and south region for six years weathering period, respectively. The characteristic peak of cellulose at 897 cm−1 was used as the internal standard peak, and the ratio of the intensity of the characteristic peak of lignin and the intensity of the carbonyl absorption peak to the internal standard peak was calculated as the relative peak intensity, so as to quantitatively compare the degradation of lignin and carbonyl group during the weathering period. With the weathering time prolongation, a positive signal at 1506 cm−1 which can be assigned to the partial decomposition of lignin. Since the absorption peak at 1506 cm−1 was arising from aromatic skeletal vibration of benzene ring is characteristics for lignin, the significant decrease in the intensity of that peak indicated degradation of lignin during the ageing process2. The absorption peak at 1232 cm−1 was lower than the characteristic absorption peak at 834 cm−1, which proved that the guaiac-based phenylpropane unit (G) was more sensitive to weathering than the syringyl phenylpropane unit (S). A comparative study on the degradation behavior of bamboo and two wood species demonstrated that bamboo was less affected by degradation due to the higher density of bamboo and the higher amount of syringyl units present in bamboo lignin15. Moreover, the intensity of the band at 1720 cm−1 corresponding to C=O on the sample surface gradually decreased. Absorbed UV light in bamboo reacts with the phenolic hydroxyl group of lignin to form free radicals, which are subsequently transformed into unsaturated carbonyl compounds (o- and p-quinonoid structures) via demethylation or side chain cleavage16. These low-molecular-weight soluble lignin degradation fragments are leached and washed from the weathered bamboo surface by rain11,17 which is the main cause of the decrease in C=O band intensity.

Table 1 Characteristic bands between 1800 and 800 cm−1 in the FTIR spectra of the bamboo scrimber.
Full size table
Fig. 2: FTIR results of bamboo scrimber during 6 years.
Fig. 2: FTIR results of bamboo scrimber during 6 years.The alternative text for this image may have been generated using AI.
Full size image

a FTIR spectra of scrimber samples weathered in the north region. b FTIR spectra of scrimber samples weathered in the south region. c Behavior of band at 1721 cm−1 against carbohydrate at 1368 cm−1. d Behavior of band at 1506 cm−1 against carbohydrate at 1368 cm−1. e Change of crystallinity (I1423/I897). f Variation of ratio of 1721 cm−1, 1506 cm−1, and 1423 cm−1 calculated by the data of 0 and 6th year. g Relative intensity of absorption bands samples weathered in the south region and h the correlation between carbonyl and lignin in the south region.

The intensity of the absorption bands related to lignin at 1595, 1506, 1232, and 834 cm−1 decreased by varying degrees during weathering. The decrease of the above four absorption peaks demonstrated that the lignin on the surface of bamboo scrimber had undergone severe degradation reaction during the weathering period, which further verified the change trend of chemical composition in the whole exposure cycle. Further, the lignin absorption band at 1506 cm−1 was attributed to the aromatic skeletal vibration of the benzene ring, where the decrease in intensity indicated severe lignin degradation during weathering. Among various weathering factors, solar radiation accounted for 80% to 95% of the degradation due to the photooxidation of lignin11. Short-wavelength UV light in solar radiation has sufficient energy to cleave many of the critical functional groups and linkages in these compounds. Thus, lignin is the main component affected by photodegradation15.

In order to determine ratio of lignin degradation and non-conjugated carbonyl absorption band at 1721 cm−1, lignin reference band at 1506 cm−1 and carbohydrate band at 1368 cm−1 was measured18,19. The proportion of carbonyl groups during natural weathering was calculated by taking into of intensity of carbonyl band at 1721 cm−1. The I1721/I1368 ratio represents the relative changes as a function of weathering time is plotted in Fig. 2c. The I1721/I1368 decreased to 70.5% of its original value for samples exposed in the north region, while it became 54.7% of its original value for samples weathered in the south region (Fig. 2e). The relative changes in the lignin/carbohydrate ratio (I1506/I1368) at different weathering periods for bamboo scrimber were displayed in Fig. 2d. The I1506/I1368 decreased to 41.3% of its original value after six years natural weathering in the north region, whereas it became 30.0% of its original value for samples at same weathering time in the south region (Fig. 2e). Overall, the I1721/I1368 and I1506/I1368 ratios of the samples decreased as weathering continued, further confirming the severe lignin degradation and loss of the degradation products.

The absorbance ratio (I1423/I897) represented the determination of crystallinity of cellulose19. It was observed that the I1423/I897 ratio decreased for bamboo scrimber samples during natural weathering as shown in Fig. 2f. The I1423/I897 ratio of samples placed in the south region showed reduction of 40.0%, and 35.2% observed in the north region (Fig. 3e). The results indicated that a decrease in crystallinity caused by natural weathering raised the amorphous portion of cellulose. This is in agreement with the previous study20, which the I1423/I897 ratio decreased in the samples during weathering because of the degradation of crystallized cellulose in the bamboo. This indicated that crystalline cellulose remained relatively intact during the initial stages of weathering. Fungal spores may have entered the bamboo surface along the photodegradation path to accelerate erosion12. Moreover, the linear relationship between I1506/I897 and I1721/I897 on the surface of bamboo scrimber was significant, indicating that carbonyl and lignin degradation were synchronized (Fig. 2g, h). A similar finding has been previously reported for the natural weathering of wood6.

Fig. 3: XPS survey spectra of bamboo scrimber exposed in different natural regions.
Fig. 3: XPS survey spectra of bamboo scrimber exposed in different natural regions.The alternative text for this image may have been generated using AI.
Full size image

A in the north region: C1s spectra and values of carbon peak component of samples within natural weathering (a) 0 year, (b) 2 years, (c) 4 years, and (d) 6 years; B in the south region: C1s spectra and values of carbon peak components of samples within natural weathering (a) 0 year, (b) 2 years, (c) 4 years, and (d) 6 years.

XPS analysis

XPS results provide complementary information about the change of surface chemical compositions, as illustrated in Figs. 3 and 4. The most prominent elements in the XPS spectra of bamboo scrimber in this study were C and O (H is not detected through XPS), which were attributed to cellulose, hemicellulose, lignin, and extractives. The C peaks were primarily assigned to the benzene ring, methylene ether bond, and hydroxymethyl groups of PF resin, and O peaks to the phenolic hydroxyl groups, hydroxymethyl groups, methylene quinone, and methylene ether bonds of PF resin.

Fig. 4: Difference in XPS spectral parameters during natural weathering based on the XPS survey spectra and C1s spectra.
Fig. 4: Difference in XPS spectral parameters during natural weathering based on the XPS survey spectra and C1s spectra.The alternative text for this image may have been generated using AI.
Full size image

a C total value. b O total value. c O/C ratio. d A/B ratio. e C1/C2 ratio. f Cox/Cunox ratio between two regions: (■) the north and (●) the south.

XPS survey spectra of bamboo scrimber samples obtained in different regions (Fig. 3A, B). C1s spectra were deconvoluted into four components (C1-C4) in order to obtain an understand of the functional groups. C1 (at 284.7 eV) mainly originates from lignin, fatty acids, and other extractives, C2 (at 285.6 eV) from cellulose and hemicellulose, C3(at 286.5 eV) from cellulose and hemicellulose, and C4 (at 288.3 eV) from hemicellulose and extractives. The C1 contribution decrease due to natural weathering for scrimber samples, from 69.6% to 50.1% in Fig. 3A-a (exposed in the north region) and 69.6% to 37.8% Fig. 3B-a (exposed in the south region), respectively. The intensity of C2 peak increase from 17.7% to 31.7% in Fig. 3A-b (exposed in the north region) and from 17.7% to 43.0% in Fig. 3B-b (exposed in the south region), respectively. The C1 contribution decreased while the C2 contribution increased within first two years significantly. It may be attributed to an increase of the C-O groups at the surface of carbohydrates and lignin by oxidation and hydrolysis reactions which happened during weathering process. It has been previously reported that the significant differences were demonstrated by the participation of oxygen and singlet oxygen in surface photooxidation reactions21, which led to the formation of products with more stable chemical bonds. Similar to the change trend of C2 contribution, an increase in the C3 peak was observed in Fig. 3A–c and B–c. This indicates that the percent contribution of a lower extend to carbonyl groups and the O-C-O linkage in cellulose and hemicellulose within exposure time. This also proved that the scrimer surface was poor in lignin and comparatively rice in cellulose and hemicellulose after natural weathering. The C4 peak representing a carbon atom linked to a carbonyl and non-carbonyl oxygen was about 5% which explained by a possible low content carboxylic group on the surface of bamboo scrimber. The decrease in C4 content (Fig. 3A–d, B–d) were in agreement with previous research which demonstrated the decrease may have been caused by the degradation of hemicellulose12 and the oxidation of extractives22. Further, the significant variation under natural weathering on the sample surface were displayed in the south region, comparing with samples exposed in the north region.

It has also been previously reported that the degradation of cellulosic materials and polymers can be detected by a change in the O/C atomic ratio23. The O/C ratio can be quantitatively determined using total peak area of O and C atoms, and the increase of O content and O/C ratio, and decrease of C content for different years were presented in Fig. 4a–c. The O/C ratio increased from 0.31 to 0.81 in the south region, and from 0.31 to 0.47 in the north region, respectively. These changes were related to a loss of surface lignin and increase in carbohydrate content which occurred by the oxidation and hydrolysis reactions. These findings were in agreement with a previous report by Huang et al.19. The ratio of aromatic carbon to aliphatic carbon (C1/C2) is related to the lignin content of a material based on the presence of aromatic carbon in lignin24. The C1/C2 value decreased by 2.36 and 3.06 in the north and south region respectively (Fig. 4e), which indicated that the degradation of lignin in the north region was less than that happened in the south region. The decrease in C1/C2 during weathering was consistent with the trend observed for the relative intensity of I1506/I897. This further confirmed that the photodegradation of lignin was just happened on the sample surface.

The degree of surface oxidation was quantified on the basis of the oxygen-to-unoxygenated carbon ratio (Cox/Cunox = (C2 + C3 + C4)/C1)25,26. The Cox/Cunox value of sample surface ranged from 0.44 to 1.00 in the north region, while increased from 0.44 to 1.64 in the south region, within six years, which indicated that the degree of oxidation on the sample surface was aggravated because of the generation of more hydroxyl, carbonyl, and carboxyl groups. These results were also in agreement with previous study which suggested that the oxidation and hydrolysis surface reactions take place on wood components during weathering and also removed the degradation of carbon-containing compounds and extractive27. The acidification of sample surface was quantified on the basis of the acid to base ratio (A/B = (C2 + C4)/(C1 + C3))28. The A/B value increased to 0.47 and 0.76 in the north and south region respectively (Fig. 4d), indicating that the alkaline of surface before weathering test had become to be acid significantly in the south region. Meanwhile, the increase of the relative cellulose content in the south region was higher compared with that in the north region. The surface of bamboo scrimber became more acidic during weathering, where the A/B ratio increased rapidly during the initial two years of weathering and stabilized in the following four years. This may have been related to the rapid photodegradation of lignin and increase in cellulose content on the surface of bamboo-based composites during the initial years of weathering29.

Determination of chemical components

The results of chemical change were displayed in Fig. 5. Weathering of six years led to a trend of decrease in the lignin (Fig. 5c) and extractive content (Fig. 5d, e), and a trend of increase in α-cellulose (Fig. 5a) and holocellulose (Fig. 5b) of bamboo scrimber samples at both experimental sites. The chemical components performed significant difference between two experimental sites. The slight reduction of lignin content in the north region was lower than that in the south region, which showed 4.7% and 8.0%, respectively. The content of α-cellulose and holocellulose performed slight increase by 13.3% and 11.1% for samples exposed in the south region, respectively. However, the scrimber samples exposed in the north region showed the lower increase of 10.4% (α-cellulose) and 10.7% (holocellulose). The significant decreases of hot- and cold-water extractives were found in the south region that were 83.8% and 72.5%, whereas the decrease of both two extractive contents were 47.9% and 59.1% in the north region, respectively. For the first stage, the degradation was happened in lignin due to the absorption of ultraviolet, which acts in combination with moisture and temperature2. According to the data collection from different regions, the precipitation and temperature in the south region was higher than those in the north region, and hence the significance of lignin content had been found in the south region. Moreover, the decomposing of lignin on the surface of scrimber was scoured by rain, resulted in the increase of holocellulose content.

Fig. 5: Difference of chemical components between two regions during natural weathering.
Fig. 5: Difference of chemical components between two regions during natural weathering.The alternative text for this image may have been generated using AI.
Full size image

a α-cellulose content. b Holocellulose content. c Lignin content. d Hot-water extractives content. e Cold-water extractives content; the numbers followed by the same small or capital letters are not significantly different at 5% significance level using Duncan’s test.

Statistically, the lignin content decreased significantly during the first two years of weathering at both two regions, but remained relatively constant during the subsequent four years. However, the hot- and cold-water extractive contents dropped significantly during each successive measurement. Previous studies have shown that extractives play an important role in the photodegradation of wood22,30, while the highest photodegradation rate occurs on the bamboo-based composites surface and even exceeds that of lignin. The α-cellulose content increased significantly in the first four years, and the subsequent increase was not significant. Based on the cellulose and holocellulose contents, the hemicellulose content was also found to increase, especially during the first two years of weathering. This was attributed to the significant degradation of lignin and extractives.

Microstructure

Scanning electron micrographs of a typical cross section of bamboo scrimber surface revealing the microporous structure after weathering process are shown in Figs. 6 and 7. The natural structure of bamboo includes stiff fiber caps of the vascular bundles embedded within the soft parenchyma matrix, which imparts mechanical stability to the bamboo culms31. The degradation of the chemical components of bamboo, especially lignin, induced microstructural changes in the bamboo scrimber. It was revealed the formation of micro-checks in the secondary walls of the thick-walled fiber cells and parenchyma cells after the first two years of weathering (Figs. 6b, c and 7b, c). These structures gradually deteriorated as weathering continued. The micro-checks expanded further during the subsequent four years of weathering. These observations were similar to previously reported microstructural changes of bamboo after accelerated aging using a xenon light source32, which further demonstrated that UV light had a significant effect on the bamboo scrimber. In addition, substantial micro-checks were observed in the compound middle lamella of the fiber walls, which were connected via the cell corners. Further development and enlargement of the micro-checks were attributed to the photodegradation of lignin as well as cell wall contraction caused by moisture exposure19,33. This finally led to the formation of visual cracks on the transverse surfaces of scrimber samples by the end of the six-year weathering period (Figs. 6b, c and 7b, c). Lignin binds cellulose microfibrils in the various cell wall layers33 and is the most light-sensitive cell wall component15. The large variation in the lignification degree across the fiber wall led to a higher content of lignin in the compound middle lamella and cell corners31. This led to the extension of micro-checks as weathering continued.

Fig. 6: SEM images of fiber cell in transverse surface of bamboo scrimber.
Fig. 6: SEM images of fiber cell in transverse surface of bamboo scrimber.The alternative text for this image may have been generated using AI.
Full size image

a Fiber cell of raw bamboo scrimber (0 year). b Fiber cell of weathered bamboo scrimber in the north region. c Fiber cell of weathered bamboo scrimber in the south region. The red arrows indicated the characteristics of fiber cell and parenchyma cell; the scale bars have been labeled inside the figures.

Fig. 7: SEM images of parenchyma cell in transverse surface of bamboo scrimber.
Fig. 7: SEM images of parenchyma cell in transverse surface of bamboo scrimber.The alternative text for this image may have been generated using AI.
Full size image

a Parenchyma cell of raw bamboo (0 year). b Parenchyma cell of weathered bamboo scrimber in the north region. c Parenchyma cell weathered in the south region. The small red arrows indicated the characteristics of fiber cell and parenchyma cell; the scale bars have been labeled inside the figures.

The micro-cracks appeared in the fiber cell were significantly different compared two natural regions (Figs. 6b, c and 7b, c). The samples weathered in the north regions just showed concentric cracks in the fiber cells, and the differentiation of parenchyma cells was not notable. However, the concentric cracks were emerged in the secondary walls of fiber cells on the transverse section after two years of natural weathering, and then a trend of flake separation was happened in the south region. After six years of weathering test, the axial cracks were distinctly observed in the space of fiber cells. The differentiation of parenchyma cells was glaringly obvious, with appearance of various cracks within the following four years. Further, the influence of natural weathering process in fiber cells was significant compared to the parenchyma cells. This is attributable to the manufacturing process of bamboo scrimber, which the resin penetrated into lumen and intercellular layer of parenchyma cells through the pits. Due to the dense arrangement of fiber cells, there was almost no resin being observed in the fiber cells. Otherwise, the lignification of fiber cells was higher than that of parenchyma cells, and hence fiber cells performed significant variation compared to parenchyma cells during natural weather period.

Consequently, the striking differences were found between the north and south region, and the natural degradation properties were obviously observed in the south region attributing to higher temperature and humidity. The clear understanding of natural weathering process was contributed to design and develop the manufacturing parameters of bamboo scrimber that could be suitable for diverse utilizations.

Methods

Raw materials

Four-year-old moso bamboo (Phyllostachys pubescens Mazel) was obtained from Zhejiang Province, China. Low-molecular-weight PF resin (solids content = 47.91%; viscosity = 35 cps; pH = 10–11) was supplied by Beijing Dynea Chemical Industry Co., Ltd.

Preparation of bamboo scrimber

Bamboo raw materials were sawn and split into bamboo-oriented fiber mat, then cut them with average length, width and thickness of 1000 mm × 100 mm × 2~5 mm. The obtained fiber mat was steeped in PF adhesive solution which was diluted to 15% solid content for 3 min, and then dried to the moisture content of 8~10% uniformly. The glued fiber mat was hot pressed with temperature of 150 °C, and the target density of 1.15 g cm−3 (Fig. 1a). The density of bamboo scrimber was determined by measuring its air-dry weight and volume after the composites were kept in a conditioning room with a relative humidity of 65 ± 3% at 25 °C ± 2 °C for 2 weeks. After measurement, the samples with the exact density of 1.15 g cm−3 were selected for evaluating properties. The 200 pieces of bamboo scrimber samples were cut for preparation of natural weathering test.

Natural weathering test

Natural weathering test was performed on the tangential surface of the OBFRC samples. The samples were positioned facing south at an inclination of 45° to the horizontal axis and 75 cm from the ground. The samples were conducted at the different natural weathering conditions of north outdoor test site (39.8° N, 116.47° E) belongs to temperate monsoon climate and the south outdoor test site (23.17° N, 113.33° E) belongs to tropical monsoon climate, and were observed for six years from 10th January 2012 to 10th January 2018, respectively. The basic parameters of climate were shown in Table 2. The exposed samples were measured various properties periodically.

Table 2 The difference in basic parameters of climate between two natural regions.
Full size table

Color change

The tangential surface color of the bamboo scrimber samples was determined every year over the six-year weathering period using a chroma meter (CR-400, Konica Minolta, Japan) in accordance with the Commission Internationale de I’Eclairage (CIE) L*a*b* parameters, where L* is the lightness coordinate, a* is the red/green coordinate, and b* is the yellow/blue coordinate. The mean color data were determined from the six measurement points of three OBFRC samples. The total color difference (ΔE*) was calculated according to the following equation:

$$\Delta E^ \ast = \sqrt {\left( {L_{{{\mathrm{t}}}}^ \ast - L_{{{\mathrm{o}}}}^ \ast } \right)^2 + \left( {a_{{{\mathrm{t}}}}^ \ast - a_{{{\mathrm{o}}}}^ \ast } \right)^2 + \left( {b_{{{\mathrm{t}}}}^ \ast - b_{{{\mathrm{o}}}}^ \ast } \right)^2}$$
(1)

where subscripts o and t denote values before and after weathering for t years, respectively.

Chemical composition

The functional groups on the surface of the bamboo scrimber samples were determined through Fourier-transform infrared (FTIR) spectroscopy in the range 400–4000 cm−1 (Nicolet iS10, Thermo Scientific, USA) (64 scans, 4 cm−1 resolution). The air-dried samples powder mixed with KBr (mass ratio of 1:10) for measurements that were repeated at least five times.

The surface chemistry of the bamboo scrimber samples were prepared with a small chip of approximately 5 × 2 × 1 mm3 on the tangential surface. The treatment was evaluated based on X-ray photoelectron spectroscopy (XPS; AXIS Ultra, ESCALAB 250, Thermo Fisher Scientific) using a monochromatic Al Kα (hν = 1486.6 eV) X-ray source with a power 225 W. The resolution of the instrument for the source in hybrid lens mode was 0.48 eV for Ag 3d5/2. Survey scans spanning binding energies from 1100 eV to 0 eV were collected. High-resolution spectra were used to determine the types of oxygen-carbon bonds by curve fitting of C1s peak and deconvolution into four subpeaks (C1, C2, C3, and C4) with XPS spectral deconvolution (XPSPEAK software, Version 4.1). A Shirley base-line was used for background subtraction, and Gaussian (60%)-Lorentzian (40%) peaks were used for spectral deconvolution. The peak widths (FWHM) of fitted components were ranged from 1.1 to 1.2 eV for C1s.

The chemical components at a surface depth of 2 mm were quantified as a sum of gravimetrically determined acid-insoluble lignin, holocellulose, and α-cellulose34,35,36. Samples were not extracted prior to lignin determination. Delignified samples were treated with cyclohexane/acetone (9:1 ratio) at boiling temperature for 1 h and refluxed. The solution was filtered off and samples were washed with cyclohexane/acetone several times. The mass of extractives was weighed after drying. Briefly, the samples were measured every two years over the six-year weathering period.

Microstructure

Bamboo scrimber samples with dimension of 10 mm(R) × 10 mm (T) × 5 mm (L) were prepared, and smooth surfaces were obtained using a sliding microtome. The samples were coated with gold via a Gressington sputter coater (ULVAC G-50DA, Japan) and observed with a scanning electron microscopy (SEM; Hitachi SU-70). 10 kV Voltage was applied with secondary electron mode.

Statistical analysis

Differences in the analysis of chemical components were statistically examined with one-way ANOVA, and post-hoc Duncan’s tests using SPSS software (Version 21.0, IBM, New York, USA). A significance level of 5% was used. All the measurements were repeated five times.