Effects of Pressurized Superheated Steam Treatment on Dimensional Stability and Its Mechanisms in Surface-Compressed Wood

: Shape stability is one of the most important properties of surface-compressed wood used as a substitute for other energy-intensive adhesives, concrete, and metals. This study evaluated the dimensional stability, surface wettability, chemical structure, cellulose crystalline structure, and microstructure of surface-compressed wood. The surface-compressed wood was then treated with pressurized superheated steam. The equilibrium moisture content, thickness swelling ratio, and wettability of the wood decreased by 20.39%, 30.63% (moisture absorption), 40.51% (water absorption), and 86.95% after pressurized superheated steam treatment, respectively. In the pressurized steam environment, hemicelluloses were signiﬁcantly degraded, signiﬁcantly reducing the strong hygroscopic groups, particularly hydroxyl groups. The crystallinity and crystal width of cellulose in the compressed wood also increased by 8.02% and 37.61%, respectively, after pressurized superheated steam treatment, corresponding to dimensional stability. Dimensional stability, namely the shape ﬁxation of the surface-compressed wood, is a complex mechanism, including the hydrophobization of cell walls, the formation of cross-linkages, the reformation of microﬁbril chains, microstructural changes, and the relaxation of inner stresses, which reduced or even eliminated the recovery. This study demonstrates that pressurized steam treatment can effectively enhance dimensional stability in surface-compressed wood, which contributes to the substantial use of surface-compressed wood in the building and construction industries. We will further explore the relationship and mechanism between superheated steam pressure, treatment time, and dimensional stability.


Introduction
Wood compression is an eco-friendly wood modification technology based on hydrothermo-mechanical treatment [1]. This technology can significantly improve the physical and mechanical properties of low-density wood [2][3][4][5]. Traditional wood compression commonly refers to overall wood compression. However, the whole wood blocks require pre-softening by steam treatment, boiling, and high-frequency or microwave heating, followed by compression to densification, resulting in a considerable loss of wood volume (as high as 50%) and obvious shape instability [6,7]. Surface compression of solid wood, which can effectively reduce volume loss, uses hydro-thermo-mechanical treatment at temperatures above 160 • C to confer high density on the wood surface [8,9].
However, compressed wood without any treatment of shape fixation is highly sensitive to moisture or liquid water, and compressive deformation recovers easily. For the stabilization of the properties and quality of compressed wood, universal methods include resin impregnation, cross-linking reaction, heat treatment, superhydrophobic treatment, and so on [3,[10][11][12]. Owing to the advantages of green, environmental protection and simple operation, heat treatment is the most widely used commercially to stabilize the dimensions of compressed wood. The treatment temperature, time, and pressure are the main process parameters that affect the properties of heat-treated wood. Stamm (1956) is one of the first scholars to study the influence of pressure conditions on heat treatment, and he believed that wood undergoes more significant thermal degradation under closed conditions [13]. In one study, the dimensions of compressed wood were permanently fixed at 180 • C for 20 h or 200 • C for 5 h in atmospheric conditions [14]. Of note, the compressed wood was treated with saturated steam at 165 • C for 30 min, 180 • C for 8 min, or 200 • C for 2 min; the set recovery was lower than 2.0% for boiling in water [15][16][17]. Therefore, the treatment time for permanent fixation of compressed wood can be markedly shortened by the pressurized steam. In a previous study, we investigated surface-compressed wood fixation via treatment with atmospheric and pressurized superheated steam. Relative to that of surface-compressed wood after atmospheric steam treatment, the set-recovery for moisture and water absorption was reduced by about 63.8% and 48.9%, respectively, after pressurized superheated steam treatment. The hygroscopicity of the wood treated with pressurized steam was lower than that of the wood treated with atmospheric steam, so the dimensional stability was improved [18].
For decades, the fixation mechanism of compressive deformation under heat treatment or hydrothermal treatment has drawn research interest. Currently, most studies indicate that the main recovery reason for the compressive deformation of compressed wood was that part of the compressive stress was stored by the elastic cellulose molecular chain during compression [17,19] and then released during the resoftening of viscoelastic hemicellulose and lignin [1,20]. To explore the fixation mechanism of compressive deformation, we studied the chemical reactions (hydrolysis, degradation, cross-linking, and esterification reactions) of wood components (cellulose, hemicellulose, and lignin) and the changes in the wood cell wall microstructure [1,7,[21][22][23][24]. However, the fixation mechanism of compressive deformation is not to be illustrated, particularly under pressurized superheated steam. In the current study, heat treatment with atmospheric and 0.30 MPa pressurized superheated steam at 180 • C was applied to stabilize the dimension of the surface-compressed wood. The effects of pressurized superheated steam on dimensional stability were investigated. Moreover, to explain the shape fixation mechanism of the surface-compressed wood with pressurized superheated steam treatment, Fourier transformed infrared, X-ray diffraction, and scanning electric microscopy were tested to investigate the changes in wood components (cellulose, hemicellulose, and lignin), crystalline structure of the cellulose, and morphological structure of the surface-compressed wood.

Preparation of Wood Specimens
Twenty-five-year-old white poplar (Populus tomentosa) trees, with diameters from 25 to 35 cm at breast height and air-dried density of 0.44 g·cm −3 , were harvested from Shandong Province, China. After the timber was kiln-dried to a moisture content (MC) of 12%, specimens measuring 400 mm (L) × 100 mm (T) × 25 mm (R) were prepared. Six replicates were performed for each treatment.

Surface Compression
The surface compression parameters used in this study were chosen based on a previous study [25]. The cross and radial sections of the specimens were sealed with paraffin. Then, the specimens were soaked in distilled water for 1 h with an average MC of 17%. The wet specimens were preheated at 180 • C for 50 s. Then, surface compression was conducted in the radial direction with a pressure of 6.00 MPa at 180 • C until the compression process was complete. Compressed specimens were kept under pressure for 30 min. The specimen was ultimately taken out until the temperature was reduced to 60 • C.

Heat Treatment of Surface-Compressed Wood
The compressed wood specimens were divided into three groups before post-treatment: surface-compressed wood without post-treatment (SCW), surface-compressed wood with atmospheric-steam heat treatment at 180 • C for 2 h (SCW + AHT), and surface-compressed wood with pressurized-steam (0.30 MPa) heat treatment at 180 • C for 2 h (SCW + PHT).
Heat treatment was performed in three stages with a sealed heat treatment tank (Xingnandrying 0938, Nanjing, China). In the first stage, high-temperature drying (maximum temperature: 130 • C) to approximately 0% MC was conducted before heat treatment. The specimens then underwent heat treatment with atmospheric steam or 0.30 MPa pressurized steam at the recommended temperature of 180 • C for 2 h. In the treatment with pressurized steam, the pressure gradually increased to 0.30 MPa at 180 • C. Subsequently, the lumbers were taken out from the tank until the temperature decreased to 25 • C.

Moisture Content
Samples measuring 20 mm (L) × 20 mm (T) × 20 mm (R) were taken from surfacecompressed wood. The samples were then divided into five layers in the radial direction and then placed in a balanced environment with a temperature of 20 • C and relative humidity of 65% until the mean mass of the samples was constant. The samples were oven-dried at 103 • C until the mean mass was unchanged.
where MC i is the moisture content for layer i; M ei is the mass for layer i after balanced treatment with a temperature of 20 • C and relative humidity of 65%; and M oi is the mass for layer i after oven-drying.

Determination of Dimensional Stability
To evaluate the effect of pressurized-steam treatment on the dimensional stability of SCW, we tested both SCW + AHT and SCW + PHT. Cubic specimens measuring 20 mm (L) × 20 mm (T) × 20 mm (R) were cut from the SCW and SCW + AHT. Dimensional stability arose from exposure to high humidity (RH = 90%, Temp = 40 • C) and immersion in water was tested at a specified time according to the standards GB/T 1934. 2-2009 (2009) and our preview studies [25,26].

Surface Wettability
The sessile drop method was used to measure the surface contact angle of water, diiodomethane, and formamide on the SCW, SCW + AHT, and SCW + PHT specimens, measuring 20 mm (L) × 20 mm (T) × 20 mm (R). The dimensions were measured at room temperature (20 • C) and an RH of 40%-50% using a contact angle tester (JC2000D, Shanghai, China). After the drop was deposited on the surface, the contact angle of each liquid was estimated for every 80 ms until 20 s.

Fourier Transformed Infrared (FTIR)
FTIR specimens were collected from the compressed layer(s) of surface-compressed wood. Wood particles were first obtained by cutting the compressed layer(s) and then ground into particles to pass a 200-mesh sieve, where the diameter of the hole was 0.075 mm. After drying at 103 • C for 4 h, the powder was blended with potassium bromide, further ground manually with an agate mortar, and made into a test tablet. FTIR tests were conducted on a Nicolet iS10 Fourier transform infrared spectrometer (Thermo Nicolet, Waltham, MA, USA). The spectra were collected for an accumulation of 64 scans with a resolution of 1 cm −1 between 4000 and 400 cm −1 . The software OPUS was used for baseline correction. The absorption at 1424 cm −1 , primarily due to the CH 2 scissor motion in cellulose, was used for spectral normalization. This absorption band was assumed to be essentially unaltered by steam treatment [7,27].

X-ray Diffraction
Laminated specimens were cut from the compressed layer of the SCW, SCW + AHT, and SCW + PHT, and then ground into particles to pass a 80-mesh sieve, where the diameter of the hole was 0.18 mm. To analyze the crystalline structure, the specimens of the SCW, SCW + AHT, and SCW + PHT were scanned with an X-ray diffractometer (XPERTPRO-30X) under the following conditions: Cu Kα radiation, λ = 0.154 nm; monochromator voltage = 40 kV; electric current = 40 mA, and diffractogram ranges of 2θ = 5 •~4 0 • with a scan rate of 5 • /min. The crystallinity index (CrI) and crystal dimension were calculated as follows: where CrI is the relative crystallinity (%), I 002 is the maximum intensity of the lattice diffraction angle of 002, I am is the amorphous scattering intensity, D is the dimension of the crystalline region (nm), k is the diffraction constant (0.89), λ is the incident wavelength (0.154), β is the diffraction peak half wide (radian), and θ is the diffraction angle ( • ).

Morphological Structure
The control, SCW, SCW + AHT, and SCW + PHT were scanned by scanning electric microscopy (SEM) to observe the changes in the morphological structure. Firstly, the specimens were cut from the transverse section of the control, SCW, SCW + AHT, and SCW + PHT. Then, they were oven-dried at 60 • C for 4 h. The cutting surfaces were sputtercoated with gold and scanned using the S4800 SEM (Hitachi, Tokyo, Japan). The SEM photograph magnification of the surface-compressed wood was 1000×.

Statistical Evaluation
One-way analysis of variance (ANOVA) was employed to analyze the statistical significance of the dimensional stability and crystalline structure at a 95% confidence interval of probability.

Moisture Content
The equilibrium moisture content (EMC) of the surface-compressed wood decreased from 10.74% to 10.39% after treatment with atmospheric steam and to 8.55% after treatment with pressurized steam (Table 1). ANOVA revealed that both atmospheric-steam heat treatment (p < 0.05) and pressurized-steam heat treatment (p < 0.05) had a significant influence on the EMC. The moisture content of the third layer was higher than that of the first and fifth layer. Steam heat treatment, particularly pressurized steam treatment, contributed to the hydrolysis reaction for hemicellulose or cross-linking reaction for lignin [21], leading to a significant reduction in hydroxyl or carbonyl groups with strong hygroscopic properties [28]. The chemical composition and chemical structure of compressed wood changed markedly because of the aggregation of acidic volatiles and the catalytic action of hydronic ions in a high-humidity environment [29]. Note: Standard deviation in parentheses. SCW, surface-compressed wood without post-treatment; SCW + AHT, surface-compressed wood with atmospheric steam heat treatment; SCW + PHT, surface-compressed wood with pressurized steam heat treatment. EMC, the equilibrium moisture content.

Moisture Absorption
After pressurized steam treatment, the moisture absorption-induced thickness swelling and volume swelling of the surface-compressed wood specimens were significantly decreased ( Figure 1). Compared with atmospheric steam processing, the thickness swelling rate decreased by 14.15% and the volume swelling ratio decreased by 14.14%, indicating that steam pressure effectively stabilized the dimensions of surface-compressed wood. ANOVA showed that both thickness swelling (p < 0.05) and volume swelling (p < 0.05) were influenced by pressurized steam treatment. Pressurized superheated steam accelerated the degradation of the hemicellulose, reducing the number of hygroscopic groups and enhancing the dimensional stability of the surface-compressed wood [26,27].
wood changed markedly because of the aggregation of acidic volatiles and the catal action of hydronic ions in a high-humidity environment [29]. Note: Standard deviation in parentheses. SCW, surface-compressed wood without post-treatm SCW + AHT, surface-compressed wood with atmospheric steam heat treatment; SCW + PHT, face-compressed wood with pressurized steam heat treatment. EMC, the equilibrium moisture c tent.

Moisture Absorption
After pressurized steam treatment, the moisture absorption-induced thickness sw ing and volume swelling of the surface-compressed wood specimens were significan decreased ( Figure 1). Compared with atmospheric steam processing, the thickness sw ing rate decreased by 14.15% and the volume swelling ratio decreased by 14.14%, indi ing that steam pressure effectively stabilized the dimensions of surface-compressed wo ANOVA showed that both thickness swelling (p < 0.05) and volume swelling (p < 0 were influenced by pressurized steam treatment. Pressurized superheated steam acce ated the degradation of the hemicellulose, reducing the number of hygroscopic gro and enhancing the dimensional stability of the surface-compressed wood [26,27].

Water Absorption
Similar to moisture absorption, the water absorption of the SCW + AHT and SCW PHT was affected distinctly by the steam pressure. Water absorption measurement in cated that thickness swelling and volume swelling occurred rapidly within the first 6 h immersion, slowing down until the 12 h time point of immersion. Extending the period immersion in water led to the narrow thickness and volume swelling ( Figure 2). Xian al. also found similar results about the water absorption of compressed wood [26]. Mo over, the thickness swelling ratio and volume swelling ratio of the SCW + PHT redu to 11.42% and 18.83%, respectively, by immersion in water. Furthermore, the ASE val of the SCW + AHT and SCW + PHT were 23.81% and 33.58%, respectively. ANOVA a indicated that steam pressure significantly affected ASE (p < 0.05).

Water Absorption
Similar to moisture absorption, the water absorption of the SCW + AHT and SCW + PHT was affected distinctly by the steam pressure. Water absorption measurement indicated that thickness swelling and volume swelling occurred rapidly within the first 6 h of immersion, slowing down until the 12 h time point of immersion. Extending the period of immersion in water led to the narrow thickness and volume swelling ( Figure 2). Xiang et al. also found similar results about the water absorption of compressed wood [26]. Moreover, the thickness swelling ratio and volume swelling ratio of the SCW + PHT reduced to 11.42% and 18.83%, respectively, by immersion in water. Furthermore, the ASE values of the SCW + AHT and SCW + PHT were 23.81% and 33.58%, respectively. ANOVA also indicated that steam pressure significantly affected ASE (p < 0.05).

Surface Wettability
The contact angle of distilled water on the wood surface decreased gradually o time ( Figure 3). Relative to that of SCW, the initial contact angle of distilled water on surface of compressed wood increased by 11.61% and 21.43% after atmospheric and p surized steam treatment, respectively. After distilled water penetrated and was spread the wood surface for 20 s, the contact angle markedly decreased. The reduction rates w 58.34% for SCW, 39.64% for SCW + AHT, and 35.73% for SCW + PHT. The decay rat the contact angle per unit time of distilled water was significantly higher on SCW t that of SCW + PHT (p < 0.05) [30,31]. Roughness and EMC were negatively correlated w the contact angle of distilled water on the wood surfaces [32]. After steam heat treatm the roughness and EMC of the wood surfaces were markedly reduced [21,33]; simila the contact angle increased and the penetration spreading rate markedly decreased, i cating that pressurized steam treatment decreased the surface wettability of compres wood and improved its hydrophobic property.

Surface Wettability
The contact angle of distilled water on the wood surface decreased gradually over time (Figure 3). Relative to that of SCW, the initial contact angle of distilled water on the surface of compressed wood increased by 11.61% and 21.43% after atmospheric and pressurized steam treatment, respectively. After distilled water penetrated and was spread on the wood surface for 20 s, the contact angle markedly decreased. The reduction rates were 58.34% for SCW, 39.64% for SCW + AHT, and 35.73% for SCW + PHT. The decay rate of the contact angle per unit time of distilled water was significantly higher on SCW than that of SCW + PHT (p < 0.05) [30,31]. Roughness and EMC were negatively correlated with the contact angle of distilled water on the wood surfaces [32]. After steam heat treatment, the roughness and EMC of the wood surfaces were markedly reduced [21,33]; similarly, the contact angle increased and the penetration spreading rate markedly decreased, indicating that pressurized steam treatment decreased the surface wettability of compressed wood and improved its hydrophobic property.

Surface Wettability
The contact angle of distilled water on the wood surface decreased gradual time ( Figure 3). Relative to that of SCW, the initial contact angle of distilled water surface of compressed wood increased by 11.61% and 21.43% after atmospheric an surized steam treatment, respectively. After distilled water penetrated and was sp the wood surface for 20 s, the contact angle markedly decreased. The reduction rat 58.34% for SCW, 39.64% for SCW + AHT, and 35.73% for SCW + PHT. The decay the contact angle per unit time of distilled water was significantly higher on SC that of SCW + PHT (p < 0.05) [30,31]. Roughness and EMC were negatively correlat the contact angle of distilled water on the wood surfaces [32]. After steam heat tre the roughness and EMC of the wood surfaces were markedly reduced [21,33]; sim the contact angle increased and the penetration spreading rate markedly decrease cating that pressurized steam treatment decreased the surface wettability of com wood and improved its hydrophobic property.

Chemical Structure
The chemical structure of the wood components is essential for a further unde

Chemical Structure
The chemical structure of the wood components is essential for a further understanding of the fixation mechanisms of SCW + PHT. Figure 4 presents the FTIR spectra of SCW, SCW + AHT, and SCW + PHT in the wavenumber range of 800-3800 cm −1 . Absorbance at 3600-3200 cm −1 is attributed to -OH stretching vibration, which includes the free hydroxyl groups that affect the dimensional stability and recovery of the compressed wood [34]. After pressurized steam treatment, the intensity of the characteristic peaks at 3350 cm −1 was significantly reduced, indicating that the wood cell wall became physically dehydrated because of high-temperature steam treatment. After dehydration, the free hydroxyl generated between cellulose molecular chains might have undergone etherification, significantly reducing the number of free hydroxyl groups [35,36].
FOR PEER REVIEW 7 of 13 3600-3200 cm −1 is attributed to -OH stretching vibration, which includes the free hydroxyl groups that affect the dimensional stability and recovery of the compressed wood [34]. After pressurized steam treatment, the intensity of the characteristic peaks at 3350 cm −1 was significantly reduced, indicating that the wood cell wall became physically dehydrated because of high-temperature steam treatment. After dehydration, the free hydroxyl generated between cellulose molecular chains might have undergone etherification, significantly reducing the number of free hydroxyl groups [35,36]. The band at 1457 cm −1 is attributed to CH2 symmetric bending on the xylan ring [37]. SCW + AHT and SCW + PHT at the absorbance of 1457 cm −1 exhibited fewer changes and lower relative intensities (by 2.17% and 7.62%, respectively), compared with SCW. Therefore, the xylan backbone was not affected by the steam treatment and the major degradation of the xylan was side group splitting [7,38]. The absorbance values at 1735 and 1235 cm −1 represent the stretching vibration of C=O in the O=C-OH group of the glucuronic acid unit and O-C in the acetyl unit in xylan, respectively [7,39,40]. Relative to that of SCW, the relative intensity of SCW + PHT at 1735 cm −1 decreased by 27.59%, and that at 1235 cm −1 decreased by 15.88%. All of the chemical changes could be attributed to the more porous structures presumably formed in wood cell walls after compression. Such microcracks would be conducive to steam penetration, resulting in the increased degradation rate of hemicellulose structures. Meanwhile, in a pressurized steam treatment environment, the accumulation of acidic degradation products and the increase in active hydronium ions significantly promoted deacetylation [28]. The increased degradation of hemicellulose structures reduced the water molecule adsorption points, leading to an improvement in the dimensional stability.
The band at 1595 cm −1 ascribed to the vibration in the aromatic ring of lignin plus The band at 1457 cm −1 is attributed to CH 2 symmetric bending on the xylan ring [37]. SCW + AHT and SCW + PHT at the absorbance of 1457 cm −1 exhibited fewer changes and lower relative intensities (by 2.17% and 7.62%, respectively), compared with SCW. Therefore, the xylan backbone was not affected by the steam treatment and the major degradation of the xylan was side group splitting [7,38]. The absorbance values at 1735 and 1235 cm −1 represent the stretching vibration of C=O in the O=C-OH group of the glucuronic acid unit and O-C in the acetyl unit in xylan, respectively [7,39,40]. Relative to that of SCW, the relative intensity of SCW + PHT at 1735 cm −1 decreased by 27.59%, and that at 1235 cm −1 decreased by 15.88%. All of the chemical changes could be attributed to the more porous structures presumably formed in wood cell walls after compression. Such micro-cracks would be conducive to steam penetration, resulting in the increased degradation rate of hemicellulose structures. Meanwhile, in a pressurized steam treatment environment, the accumulation of acidic degradation products and the increase in active hydronium ions significantly promoted deacetylation [28]. The increased degradation of hemicellulose structures reduced the water molecule adsorption points, leading to an improvement in the dimensional stability. The band at 1595 cm −1 ascribed to the vibration in the aromatic ring of lignin plus C=O stretching [41,42] showed a larger decrease: 12.87% for SCW + AHT and 25.81% for SCW + PHT. Notably, the intensity of the absorption band at 1508 cm −1 ascribed to aromatic skeletal vibration changed slightly: 3.10% for SCW + AHT and 5.73% for SCW + PHT. Relative intensity at 1595 cm −1 changed obviously and intensity at 1505 cm −1 only changed slightly, demonstrating that the C=O group linked to the aromatic skeleton of lignin was lost [7]. This observation indicates that cross-linkings were formed between aromatic units in the lignin.

Cellulose Crystalline Structure
The X-ray diffraction patterns of SCW, SCW + AHT, and SCW + PHT are presented in Figure 5. In the figure, the apparent diffraction peaks at 15.60 • , 22.16 • , and 34.42 • are associated with their corresponding (101), (002), and (040) crystallographic planes, respectively [43]. The curves are similar in peak shape and peak position, but vary in the intensity of the diffraction peak. These observations suggest that the unit cell structure of wood cellulose in the crystalline region remained the same, but the relative crystallinity and crystal plane size of cellulose changed.

Cellulose Crystalline Structure
The X-ray diffraction patterns of SCW, SCW + AHT, and SCW + PHT are presen in Figure 5. In the figure, the apparent diffraction peaks at 15.60°, 22.16°, and 34.42° associated with their corresponding (101), (002), and (040) crystallographic planes, res tively [43]. The curves are similar in peak shape and peak position, but vary in the in sity of the diffraction peak. These observations suggest that the unit cell structure of w cellulose in the crystalline region remained the same, but the relative crystallinity crystal plane size of cellulose changed. The crystallinity index (CrI) of SCW + AHT was 55.97%, reflecting an increas 7.99%, and that of SCW + PHT was 58.07%, reflecting an increase of 12.58% relative to CrI of SCW ( Figure 6). Guo et al. [23] and Chen et al. [19] also found that CrI increa significantly after saturated steam treatment at 160 °C. One-way ANOVA indicated the post-treatment conditions (p < 0.05) significantly affected the crystallinity of w with the significance level set to 5%. The increase in the relative crystallinity of c pressed wood after steam treatment may be attributed to the ordered orientation or c tallization of the amorphous region or the amorphous region of cellulose in the wood The crystallinity index (CrI) of SCW + AHT was 55.97%, reflecting an increase of 7.99%, and that of SCW + PHT was 58.07%, reflecting an increase of 12.58% relative to the CrI of SCW ( Figure 6). Guo et al. [23] and Chen et al. [19] also found that CrI increased significantly after saturated steam treatment at 160 • C. One-way ANOVA indicated that the post-treatment conditions (p < 0.05) significantly affected the crystallinity of wood, with the significance level set to 5%. The increase in the relative crystallinity of compressed wood after steam treatment may be attributed to the ordered orientation or crystallization of the amorphous region or the amorphous region of cellulose in the wood cell walls [44]. significantly after saturated steam treatment at 160 °C. One-way ANOVA indicated that the post-treatment conditions (p < 0.05) significantly affected the crystallinity of wood, with the significance level set to 5%. The increase in the relative crystallinity of compressed wood after steam treatment may be attributed to the ordered orientation or crystallization of the amorphous region or the amorphous region of cellulose in the wood cell walls [44]. Further, the average width and length of the crystalline region were calculated using the diffraction peak half-width of the (002) and (400) planes. The crystalline region of SCW + AHT had an average width of 33.46 Å and length of 205.56 Å. These values reflect increases of 18.15% and 1.17%, respectively, relative to those of SCW. Moreover, SCW + PHT had an average width of 38.97 Å and an average length of 210.78 Å. These values reflect increases of 37.61% and 3.74%, respectively, relative to those of SCW. The change in the crystallite region along the directions of cellulose (002) and (040) is generally attributed to the rearrangement of cellulose molecular chains in some adjacent microfilaments. Kuribayashi et al. [45] reported that superheated steam treatment clearly affects the crystalline structure of wood cellulose. Pressurized superheated steam treatment increased the crystallite dimension, which also impeded the penetration of water molecules into the cell wall, decreasing its hygroscopicity [29]. Figure 7 presents the cross-sectional SEM images of the control specimen, SCW, SCW + AHT, and SCW + PHT. After surface compression treatment, P. tomentosa exhibited a significant decrease, a change in the shape of the vessels from oval to narrow and flat, and buckling in the wood fiber cells. Notably, the wood cell walls showed tiny cracks and a slightly stripped intercellular layer because of the compression treatment, significantly increasing the number of mesopores and micropores [19,24]. After steam treatment, the surface roughness and the number of transverse cracks increased (Figure 7a,b). These microcracks yielded more open spaces in the wood, further facilitating the diffusion and transfer of high-temperature superheated steam, which may lead to the increased degradation of chemical components [46]. increasing the number of mesopores and micropores [19,24]. After steam treatment, the surface roughness and the number of transverse cracks increased (Figure 7a,b). These microcracks yielded more open spaces in the wood, further facilitating the diffusion and transfer of high-temperature superheated steam, which may lead to the increased degradation of chemical components [46].

Suggested Mechanism of Dimensional Stability
The set recovery of compressed wood can be reduced or even eliminated by the hydrophobization of cell walls, the formation of cross-linkages between lignin, the reformation of microfibril chains, and the relaxation of inner stresses during post-treatment. Figure 8 presents a spring-dashpot model of the mechanism underlying the dimensional stability of surface-compressed wood. After surface compression treatment, the compressive strain was temporarily retained by the plastic deformation of amorphous matrix polymers, whereas elastic energy was stored in the hydrophobic and elastic fibers in the cell walls [19].
The first mechanism is hydrophobization (Figure 8c), which prevents moisture sorption of the amorphous matrix polymers and succeeding dimension changes. SCW + PHT showed a reduction in EMC from 10.74% to 8.55%, and a decrease in moisture/water absorption-induced thickness swelling of 30.63% and 40.51%, respectively. In the pressurized steam environment, hemicelluloses undergo significant deacetylation and hydrolysis degradation reactions under the catalysis of acidic and active water and ions. Thus, the strong hygroscopic groups (particularly hydroxyl groups) were significantly reduced (Figure 8d), hence the hydrophobic effect of the cell wall and stress relaxation. drophobization of cell walls, the formation of cross-linkages between lignin, the reformation of microfibril chains, and the relaxation of inner stresses during post-treatment. Figure 8 presents a spring-dashpot model of the mechanism underlying the dimensional stability of surface-compressed wood. After surface compression treatment, the compressive strain was temporarily retained by the plastic deformation of amorphous matrix polymers, whereas elastic energy was stored in the hydrophobic and elastic fibers in the cell walls [19]. The first mechanism is hydrophobization (Figure 8c), which prevents moisture sorption of the amorphous matrix polymers and succeeding dimension changes. SCW + PHT showed a reduction in EMC from 10.74% to 8.55%, and a decrease in moisture/water absorption-induced thickness swelling of 30.63% and 40.51%, respectively. In the pressurized steam environment, hemicelluloses undergo significant deacetylation and hydrolysis degradation reactions under the catalysis of acidic and active water and ions. Thus, the strong hygroscopic groups (particularly hydroxyl groups) were significantly reduced (Figure 8d), hence the hydrophobic effect of the cell wall and stress relaxation.
The second mechanism is cross-linking between lignin, which is suggested to mechanically restrict shape fixation [14,24]. Owing to moisture absorption, the volume swelling of SCW + PHT was effectively reduced from 20.79% to 12.37% (RH, 90%; temperature, 40 °C). In a pressurized steam environment, the yield of acetic acid accelerated lignin demethylation, providing more chemical reaction sites for cross-linking reactions. The stable structure formed by cross-linking between lignin molecules limits the set recovery of elastic microfibers. The second mechanism is cross-linking between lignin, which is suggested to mechanically restrict shape fixation [14,24]. Owing to moisture absorption, the volume swelling of SCW + PHT was effectively reduced from 20.79% to 12.37% (RH, 90%; temperature, 40 • C). In a pressurized steam environment, the yield of acetic acid accelerated lignin demethylation, providing more chemical reaction sites for cross-linking reactions. The stable structure formed by cross-linking between lignin molecules limits the set recovery of elastic microfibers.
The third mechanism refers to the crystallinity and crystal dimensions of the cellulose, which are also attributed to the fixation of compressive strain. Notably, the crystallinity of SCW + PHT increased by 8.02% and the crystal width increased by 37.61% relative to those of SCW. Crystalline cellulose is inaccessible to water molecules and thermal stability under dry conditions, but it will be rearranged after high-temperature steam treatment [19]. Therefore, the rearrangement in cellulosic microfibrils was proposed to clarify the increase in the crystallinity and crystal dimensions of cellulose [23,45,47].
After surface compression, some microcracks were found in the buckled cell walls. Microcracks promote the diffusion and transfer of high-temperature steam in wood, as well as accelerate the degradation of hemicellulose and the cross-linking reaction of lignin. Meanwhile, microfracture within and/or between microfibrils may be treated by steaming, allowing the rearrangement of cellulose [19,23,45]. During the process, the elastic energy reserved in the microfibrils is released, and the shape fixation of surface-compressed wood is perpetual by the rearranged fibers. Meanwhile, multiple mechanical and chemical changes in the cell wall occurred. Hence, it is a complex mechanism that includes the hydrophobization of cell walls, the formation of cross-linkages, the reformation of microfibril chains, microstructural changes, and the relaxation of the inner stresses. These processes reduced or even eliminated set recovery.

Conclusions
This study evaluated the effect of pressurized superheated steam treatment on the EMC, contact angle, chemical structure, cellulose crystalline structure, and microstructure of surface-compressed wood. The shape fixation of surfaced-compressed wood was successful after post-treatment with 0.3 MPa superheated steam at 180 • C for 2 h. The EMC of SCW + PHT decreased from 10.74% to 8.55%; in addition, the moisture/water absorptioninduced thickness swelling decreased by 30.63% and 40.51%, respectively. Moreover, the volume swelling decreased and the contact angle increased significantly. On the basis of the changes in chemical structure, crystalline structure, and microstructure of SCW + PHT, three kinds of mechanisms underlying the dimensional stability of pressurized superheated steam were speculated. In the pressurized steam environment, hemicelluloses undergo significant deacetylation and hydrolysis degradation. The stable structure was formed by cross-linking between lignin, and the rearrangement of cellulose microfibrils increased the crystallinity and crystal dimensions of cellulose. Microfracture further promoted the diffusion and transfer of high-temperature steam in wood, as well as accelerated the degradation of hemicellulose, lignin cross-linking, and the rearrangement of cellulose. These results demonstrate that pressurized superheated steam treatment can potentially improve the dimensional stability of surface-compressed wood in industrial production. Of course, this paper also has some shortcomings. The experimental treatment time was only 2 h, and the temperature was only 180 • C, which could not provide the optimal treatment conditions for the experiment. Therefore, we will further explore the relationship and mechanism between superheated steam pressure, treatment time, temperature, and dimensional stability. Funding: This research was funded by the Nature Science Foundation of China, formation mechanism and controllability of wood sandwich compression by hydro-thermal control (Grant No. 31670557) and permanent fixation of wood plastic deformation by a pressurized heating medium containing superheated steam, and its mechanism & applicability (Grant No. 32071690).

Institutional Review Board Statement:
Not applicable for studies not involving humans or animals.

Informed Consent Statement:
Not applicable for studies not involving humans or animals.

Data Availability Statement:
The data used to support the findings of this study are available from the corresponding author upon request.