3.1. Chemical Composition
As a result of TM, silver birch wood underwent ML due to the breakdown of structural wood components, namely xylan and acetyl groups, resulting in increased extractive content (
Table 3). For birch wood, the ML following TM varied from 5.9 to 12%. After TM at regimes B/160/120/4 and B/170/60/4, birch wood had the greatest average ML (10.1–12.0%). Following TM at 160 °C for 60–90 min, the lowest average ML was recorded (regime B/160/60/4, B/160/90/3, and B/160/90/4). A higher ML after TM was obtained by raising initial pressure, provided that all other process parameters stayed the same. With the exception of regime B/160/90/4, which displayed statistically significant ML differences from B/160/120/4 and B/170/60/4, the ML values reported by TM displayed broad error margins. The inorganic part (ashes) of untreated and all TM wood specimens was 0.2–0.3%; hence, the data are not given. The most thermally unstable wood cell wall polymers, hemicelluloses, are amorphous polysaccharides that break down to produce most of the ML. Their breakdown frequently starts with the cleavage of acetyl groups from hemicellulose (xylan) side groups, which produces acetic acid and catalyses further breakdown [
29]. Our earlier research found that TM promotes shrinkage in the tangential, radial, and total volume of birch wood. The reduction in the tangential direction of TM birch wood was higher (4.4–6.4%) than in the radial (2.9–4.7%), leading to volumetric reductions in the whole board ranging from 7.1 to 10.2% [
5].
The chemical components in silver birch wood changed due to TM in a nitrogen atmosphere. The most significant alterations were seen in the content of acetone extractives, xylan, and acetyl groups, while glucan and lignin showed less significance. The percentage of acetone-soluble extractives in TM wood increased 2–6 times (4.0–12.6%) when compared to untreated birch wood (1.9%). A similar pattern was seen for birch (
Betula spp.), aspen (
Populus tremula), and grey alder (
Alnus incana) following TM in saturated steam under pressure at 160 °C for 3 h and 170 °C for 1 h. It resulted in increased content of acetone-soluble extractives by a ratio of 2–7 [
19]. In most research studies, TM of wood resulted in higher levels of extractives. However, the opposite impact has also been seen [
8]. Surprisingly, the maximum amount of extractives (12.6%) was recovered from wood TM in regime B/160/120/4, while regime B/170/60/4 produced 9.6%. As a result, the most extractives were recovered from the samples with the highest ML after TM. The extractive content climbed as the T
max time increased. The initial pressure increase exhibited no apparent trend.
The lignin content increased significantly after treatments B/150/120/5, B/160/60/4, B/170/30/4, and B/170/60/4, but declined for the remainder. The relative content of lignin remained similar regardless of the TM parameters utilized. Because cellulose degrades exclusively in amorphous regions, TM wood contains a greater proportion of crystalline cellulose. The preferential breakdown of amorphous polysaccharides leads to a proportionate rise in lignin content. Despite increased lignin concentration, it still experiences chemical changes during thermal modification, such as de-polymerization and re-polymerization [
30]. Furfural and hydroxymethylfurfural are examples of hemicellulose breakdown products that can react with lignin to enhance the concentration of lignin [
31].
The relative amount of glucan after TM in nitrogen increased by 0.6 to 4.4%, with regimes B/160/120/4 and B/170/60/4 producing the highest content (43.7–44.2%). The relative increase in glucan content was due to sample ML and acetone extractives being washed out of the wood structure. The content of arabinan was down from 0.7% for untreated wood to 0.1–0.3% for TM birch, but mannan content was 0.9–1.3%. Between all TM regimes, significant changes or correlations were not observed. Birch wood modified under the B/160/120/4 and B/170/60/4 regimes showed the greatest reduction in xylan content (15.5–16.2%). Both treatments resulted in identical acetyl group cleavage, with amounts as low as 2.4%. Also, regime B/160/90/4 resulted in substantial xylan destruction of 18.9%, whereas acetyl group cleavage occurred at the same level as in other treatments (3.2–4.2%). The degradation of hemicellulose is dependent on its composition; arabinoxylan, predominant in hardwoods, breaks down more quickly than galactoglucomannan, present in softwoods [
32]. Compared to main-chain sugars like xylose, mannose, and glucose, side-chain sugars like arabinose and galactose are more labile [
33]. Short rotation teak (SRT) and long rotation teak (LRT) hemicellulose content decreased from 26.3 to 10.7% for SRT and 27.6 to 9.6% for LRT after heat treatment at 220 °C for 20 h under nitrogen atmosphere, while the relative content of cellulose increased from 41.1 to 46.0% for SRT and 39.9 to 48.1% for LRT. For SRT and LRT, the relative lignin content grew from 32.6 to 43.3% and from 32.5 to 42.3%, respectively. After heat treatment, the total extractive contents of SRT and LRT fell by 3.8% and 30.4%, respectively. It was due to the evaporation of volatile compounds [
8].
We believed that the breakdown of structural components would improve the hydrophobicity of TM wood. Our previous studies showed that regime B/170/60/4 had the best anti-swelling efficiency (ASE) (63% after the fifth cycle) compared to regimes B/160/120/4 and B/160/90/4, which had ASE of 45 and 33%, respectively [
5]. Chemical composition changes did not allow for the prediction of wood water-related properties following TM in nitrogen. Practical testing of TM wood utilizing a variety of procedures is required to fully characterize the resulting material.
The TM of Scots pine wood also caused changes in chemical components (
Table 4). Pine wood showed a lower ML (3.9–9.0%) after TM than did birch wood (5.9–12.0%). At 180 °C (regimes P/180/30/5 and P/180/60/5) and 170 °C with maximum T
max (regimes P/170/120/4 and P/170/120/6), the highest ML of pine wood (7.6–9.0%) was obtained. At 160 °C (regimes P/160/120/5 and P/160/180/5), the lowest ML was recorded. In our earlier research, TM pine wood showed greater loss in the tangential direction (2.9–4.5%) than in the radial direction (1.5–2.5%), with volumetric changes ranging from 4.4% to 6.9% [
5]. The inorganic part (ashes) of untreated and all TM wood specimens was 0.2–0.3%; hence, the data are not given.
Scots pine TM in a nitrogen environment enhanced the quantity of acetone extractives, lignin, glucan, and mannan, but xylan and acetyl groups were reduced. The proportion of acetone-soluble extractives in TM wood increased to 1.8–4.7% as compared to untreated pine wood (0.4%). However, the total amount was considerably lower than that of TM birch wood. The maximum amount of extractives (4.7%) was recovered from wood TM in regime P/180/60/5, while the 30 min treatment also revealed a significant extractive content (3.3%). Although regimes P/170/120/4 and P/170/120/6 had relatively high ML, their extractive contents were only 2.9 and 2.7%, respectively. The lignin content increased after all treatments except P/160/120/5; however, the rise was not considerable and did not follow a clear trend depending on the TM parameters utilized.
The relative amount of glucan following TM in nitrogen increased for all treatments, with the exception of regime P/170/60/4, which showed no significant increase. The regimes P/170/90/4, P/170/120/4, and P/170/120/6 produced the highest yields (45.8–46.6%). Mannan content increased throughout all treatments except P/170/120/4 and P/180/30/5. The xylan amount was lowered from 6.3% in untreated wood to 5.0–5.9% in TM wood. The regimes with the lowest xylan content were P/170/120/4 and P/180/60/5. Untreated pine wood has a significantly lower acetyl group concentration (1.6%) than birch wood (4.7%). Regimes P/170/120/4 and P/180/60/5 likewise had the lowest acetyl group level, 1.0–1.1%. The concentration of galactan and arabinan decreased from 1.5 and 1.4% in untreated pine wood to 0.9–1.3% and 0.1–0.5% in TM pine, respectively. Significant differences between TM regimes were not identified; hence, these data are not included.
Our earlier research found that the ASE of regime P/180/60/5 was the greatest (51% after the fifth cycle), indicating the most significant structural alterations following TM. Regimes P/170/120/4, P/170/90/6, P/170/120/6, and P/180/30/5 had slightly lower ASE (46 to 48%) [
5]. The association shows that TM pine samples with the highest ML and acetone extractives content and the lowest xylan and acetyl group content had the highest ASE values. As a result of the thermal degradation of chemical components, particularly xylan, pine wood’s water-related qualities improved.
Given that the concentration of xylan, arabinan, galactan, and acetyl groups in TM pine wood dropped by 3–4%, it is reasonable to conclude that ML is mostly a result of heat breakdown of natural pine resin. This is further supported by the comparatively low acetone-soluble extractives concentration in TM pine wood, which is analogous to the previously described compound reduction.
3.2. Mechanical Strength
The density of birch wood decreased as a result of TM (
Table 5); this was caused by the degradation of wood’s structural components, primarily xylan and acetyl groups (
Table 3). The density of air-dried, untreated silver birch wood was 652 ± 18 kg×m
−3. Density was lowered across all TM samples. As compared to native wood for regimes B/160/120/4, B/170/30/4, and B/170/60/4, a significant drop was detected in the density of TM birch, which also had the highest ML after TM (
Table 2), indicating a relationship between these parameters. The MOR of untreated birch was 124 MPa, which was lowered by 15–42% for all TM specimens. Regimes B/160/120/4, B/170/30/6, and B/170/60/4 produced the lowest MOR values (72–85 MPa). The decrease in MOR was primarily due to the breakdown of hemicelluloses during TM. Hemicelluloses are depolymerized into oligomers and monomers using hydrolysis processes. This involves the breakdown of the main-chain components, mannose, glucose, and xylose, after the cleavage of the side-chain elements, arabinose and galactose. Furfural and hydroxyl-methyl-furfural are produced by dehydrating the related pentoses and hexoses, respectively [
20]. Our findings (
Table 3) supported this, as they showed significant xylan degradation and acetyl group cleavage. Following TM under regimens B/150/120/5 and B/160/90/4, the MOR was reduced. TM at 150 °C generated small changes in birch wood’s chemical composition and had a negligible effect on MOR.
The untreated birch MOE was 13,300 MPa, and the TM process resulted in an increase in the average MOE values of birch wood based on the treatment parameters. Only regime B/160/60/4 showed an insignificant reduction. The highest MOE (14,700–15,100 MPa) was observed for specimens following TM under regimes B/150/120/5, B/160/90/4, and B/170/30/6. However, the MOE values had substantial error margins, and the differences between all TM regimes were insignificant.
The MOR of beech wood (with an initial density of 880 kg×m
−3) dropped from 166 N×mm
−2 to 110, 90, and 80 N×mm
−2, respectively, after 1.5, 2.5, and 3.5 h of TM at 185 °C under a nitrogen flow at 10 bar pressure. After 2.5 h of TM at 185 °C in nitrogen, the MOR and MOE of beech wood (with an initial density of 700 kg×m
−3) decreased from 110 to 70 N×mm
−2 and 12,000 to 11,000 N×mm
−2, respectively. After 2.5 h of TM at 195 °C, the MOR and MOE of poplar wood (with initial density 410 kg×m
−3) decreased from 57 to 53 N×mm
−2 and 7600 to 7500 N×mm
−2, respectively. After 2.5 h of TM at 195 °C, the MOR of birch wood (with a lower initial density of 560 kg×m
−3) decreased from 110 to 100, but the MOE increased from 13,000 to 14,000 N×mm
−2 [
6]. MOR decreased after TM in nitrogen, whereas MOE values varied. However, in our study, the TM of birch wood in nitrogen led to a greater reduction in MOR.
The tangential surface Brinell hardness of untreated and TM birch wood was substantially higher than that of the radial surface. TM birch wood’s tangential surface had a greater Brinell hardness than that of untreated wood, with the exception of regimes B/160/120/4 and B/170/60/4. It is difficult to explain why the Brinell hardness of the tangential surface improved after TM under regimes B/160/60/5, B/160/90/3, and B/160/90/4. SEM pictures indicated that silver birch (
Betula pendula) wood morphological elements (libriform, vessels, rays, and yearly rings) had a significant decrease in size after 1 h of TM in saturated steam at 160 °C. The linear lumen sizes decreased more in the radial direction (2.9%) than in the tangential direction (0.5%). After being treated at 180 °C, the wood’s morphological structure started to disintegrate [
34]. Perhaps the radial shrinkage of birch wood at 160 °C compacted the tangential surface and generated a modest rise in Brinell hardness. The microstructural alterations of birch wood caused by TM in saturated steam at 160 °C for 1 h were investigated using micro X-ray computed tomography. Above all studied wood species, birch displayed 16% ML and the largest volume loss (19%) after TM; however, porosity after TM was reduced from 34% to 29% [
35]. As a result, the destruction of chemical components and microstructural changes following the TM process did not always result in a simultaneous drop in mechanical strength in all anatomical directions of the wood.
Radial surface Brinell hardness was comparable between untreated and most TM wood specimens, taking error limits into account, with the exception of regimes B/160/120/4, B/170/30/4, and B/170/60/4. These treatments yielded the lowest xylan and acetyl group amounts (
Table 3). TM in nitrogen at T
max 160 °C and maximum time at T
max (120 min), as well as treatments at 170 °C (30 and 60 min), resulted in the most extensive thermal degradation of birch wood components.
As a result, MOR and Brinell hardness were lowered significantly. The Brinell hardness of European beech (
Fagus sylvatica L.) wood was lowered from 44.8 to 39.9 N×mm
−2 after 6 h of TM in a nitrogen environment at 190 °C. However, the specific anatomical surface evaluated for indentation was not mentioned [
21]. Beech wood density (681 ± 20 kg×m
−3) utilized in that investigation was equivalent to that of silver birch wood; however, it recorded more than 2 fold higher Brinell hardness than birch wood. The anatomical structure of beech wood is comparable to that of birch wood; hence, it is unclear why such disparate results were achieved.
The density of pine wood after TM decreased (
Table 6). It was a consequence of the ML caused by the degradation and evaporation of natural resins and destruction of the structural components of wood, mainly xylan and acetyl groups (
Table 4). The density of air-dried untreated Scots pine wood was 581 ± 22 kg×m
−3, which decreased following all TM treatments. However, the variations were not statistically significant. There was no correlation between ML and density for TM pine wood, as was the case with birch wood.
The MOR of untreated Scots pine was 98 MPa, which was lowered by 2–32% among all TM specimens. The MOR values for P/170/120/4, P/170/90/6, P/170/120/6, P/180/30/5, and P/180/60/5 were significantly lower (67–79 MPa) than those for untreated pine wood. The average MOR values following TM at 170 °C with both initial pressures (4 and 6 bar) tended to decrease as the time at Tmax increased. However, the changes between these regimes were insignificant.
The MOE of untreated Scots pine was 12,400 MPa, and MOE values decreased when the TM process was applied. Only the regimes P/160/180/5 and P/170/60/4 showed insignificant improvement. Specimens following TM at 170 °C and time at Tmax 120 min with both beginning pressures (4 and 6 bar) had the lowest MOE (10,000–11,200 MPa). However, all MOE values were in a similar range, and the variations were not significant.
Pine wood’s MOR and MOE decreased from 110 to 80 N×mm
−2 and 16,000 to 13,000 N×mm
−2 after 3 h of TM at 185 °C in nitrogen [
6]. Scots pine’s MOR dropped from 88.7 to 85.9 N×mm
−2 following TM at 165–185 °C and T
max 0–90 min, whereas its MOE increased from 9660 to 10,660 N×mm
−2. Radiata pine and Norway spruce exhibited similar tendencies in terms of MOR and MOE after TM [
20]. Overall, our findings are similar to those of previously mentioned studies, and TM in nitrogen in a closed pressured process causes loss of MOR for Scots pine wood. However, prior investigations have found inconsistent MOE changes following TM. Still, it should be noted that the error limitations in these tests were wide, and the rise in MOE following TM was not significant.
The Brinell hardness of the radial surface of untreated and TM pine wood was much higher than that of the tangential surface, and the pattern was opposite when compared to silver birch. The Brinell hardness of the tangential surface of Scots pine was 13.8 N×mm−2 before treatment. For the regimes P/170/60/4, P/170/90/4, P/170/120/4, and P/170/120/6, it was slightly lower after TM. The Brinell harness was altered by the rest of the treatments, although not significantly. TM Scots pine wood’s radial surface Brinell hardness increased, particularly under regimes P/170/90/4, P/170/120/4, P/170/60/6, and P/170/90/6.
Using micro X-ray computed tomography, the microstructure of pine wood was investigated after TM in saturated steam at 160 °C for 1 h. Scots pine wood demonstrated 17% volume loss and 15% ML, but the porosity (28%) following TM remained constant [
35]. In our previous research employing TM of pine wood, the tangential direction (2.9–4.5%) showed a higher reduction than the radial direction (1.5–2.5%) [
5]. It is likely that greater tangential direction shrinkage after TM leads to microstructure rearrangement in a denser structure that raises the radial surface’s hardness.
The Brinell hardness of Scots pine (
Pinus sylvestris L.) decreased from 40.7 to 37 N×mm
−2 after 6 h of TM in nitrogen atmosphere at 190 °C, but the changes were not statistically significant. The anatomical surface that was evaluated for indentation was not mentioned; however, the density (595 ± 7 kg×m
−3) was quite similar to that of the pine wood we utilized for our research [
21]. Since the approach employed by the authors of that study was the same, it is impossible to explain how they arrived at such values. It is confusing since they quoted a different study in which the Brinell hardness of pine wood (heartwood and sapwood) was measured following a saturated steam treatment. The hardness of sapwood increased from 11.6 MPa to 12.0 MPa, whereas that of heartwood decreased from 11.4 MPa to 10.7 MPa after 3 h of TM at 150 °C. Increasing the TM temperature to 180 °C decreased wood hardness to approximately 9 MPa in both tested zones [
36]. These values are comparable to our findings. However, after TM in nitrogen under pressure, pine wood Brinell hardness remained constant across the tangential surface while increasing significantly for the radial surface. The Brinell hardness parallel to the grain of Scots pine after TM at 165–185 °C and time at T
max 0–90 min was obviously increased from 36 to 53.2 N, whereas the hardness perpendicular to the grain increased slightly from 17.5 to 18.4 N after TM [
20]. In that investigation, the difference between grain directions was doubled for untreated wood and tripled after TM. The reasons for the variations in grain orientations and Brinell hardness improvements following TM were not fully explained.