Physical and Chemical Changes in Hydrothermally Modified Wood

In this study, the simulation of artificial conditions, containing various chemical additives similar to a bog environment, was hypothesised as a relatively short way to develop a more sustainable wood product with better properties. This study aimed to determine how hydrothermal treatment with different iron additives with and without tannic acid, performed at 100 ◦C for 75 h, changed the mass and density of wood of different tree species (Pinus sylvestris L., Picea abies (L.) H. Karst., Pseudotsuga menziesii (Mirb.) Franco, Juglans regia L., and Acer platanoides L.); total phenolic compounds in biobased solvents after the hydrothermal processing andiron concentration in different wood layers. We also aimed to determine the wood chemical composition using Fourier-transform infrared technique. The study results showed that the softwood samples were more susceptible to hydrothermal modification than the hardwood samples, indicated by minorly different intensities on infrared spectra. The FTIR results showed bands of hardwood and softwood samples which along with the spectral characterisation of the wood constituents, which can further be utilized for creating a larger FTIR spectral database of wood samples for their identification. The factors with the most significant effect on the hydrothermal modification were tannic acid and inorganic Fe3O2 salt, used as an additive. The results indicated the suitability of inorganic Fe3O2 salt and tannic acid combination applied in this type of modification, because a higher concentration of iron was distributed on the inner and intermediate sublayers of wood samples. The hydrothermal modification approach to anchor tannic acid–Fe macromolecules in wood shown in this study has promising capabilities and may attract interest in the field of wood protection against weathering.


Introduction
Wood, chemically described as a polymeric composite of cellulose, hemicellulose, and lignin, is an economically important renewable resource widely used as an environmentally friendly material with properties suitable for industry and the private sector. Despite good natural wood properties, the requirements for wood or wood-based construction products are growing. The biodegradability, dimensional nonstability under varying moisture contents, and degradability by ultraviolet (UV) light often limit the use of natural wood and its products. Under outdoor exposure conditions, wood undergoes the photooxidation or photochemical degradation process: the colour changes, followed by the loss of fibres, and surface erosion is often observed [1][2][3]. The wood properties change due to internal chemical reactions, and wood strength is lost as it biologically degrades due to oxidation, hydrolysis, and dehydration reactions [4,5].
Different wood modification processes, such as thermal modification at elevated temperatures, chemical acetylation with acetic anhydride, furfurylation with furfural concentration in different wood layers; and (4) to analyse wood chemical composition using Fourier-transform infrared (FT-IR) technique.

Materials and Methods
The wood samples obtained from Scots pine (Pinus sylvestris), Norway spruce (Picea abies), Douglas fir (Pseudotsuga menziesii), walnut (Juglans regia) and Norway maple (Acer platanoides) were used to determine the influence of different hydrothermal treatments with additives Fe 2 O 3 or FeCl 3 with and without tannic acid. The sets of 20 wood samples of 20 × 20 × 30 mm dimensions from each tree species were exposed to the following treatments: H 2 (control). It was assumed that the selected additives and their concentrations were the lowest possible concentrations to cause the least impact on the environment, therefore, these solvents were assessed as biobased solvents. The wood samples were immersed into the biobased solvents and boiled at a constant temperature of 100 • C in a laboratory pot for 75 h (hereafter, hydrothermal treatment). More specific details about the experiment were provided in Aleinikovas et al. (2021) [26].
For the assessment of the concentrations of extractible phenolics, the biobased solvents with Fe 2 O 3 or FeCl 3 and with or without tannic acid after the processing of each tree species were sampled into individual laboratory test tubes and transported to the laboratory. The chemical study was focused to compare the concentrations of extractible phenolics from pine, spruce, fir, walnut, and maple. Total phenolic compounds in the biobased solvents after processing were semiquantitatively evaluated with a Folin-Ciocalteu reagent [27]. According to this procedure, the commercial reagent was diluted with distilled water at a ratio of 1:10. The reaction mixture was prepared by mixing 150 µL of tested solvent after thermal treatment, 750 µL of Folin-Ciocalteu reagent solution, and 600 µL of 7.5 percent sodium carbonate solution. The absorption was read after 60 min at λ = 765 nm. The samples were prepared in triplicate for each analysis and the mean value of absorbance was obtained. The same procedure was repeated for the standard solution of gallic acid, and the calibration line was construed (y = 10.8386x + 0.036, R 2 = 0.993). The TPC was expressed in mg gallic acid equivalents (GAE) per 1 kg of tested wood species after processing.
For the assessment of mass and volume changes of wood samples, all wood samples were oven-dried to constant weight before and after the treatment processing, and then the mass and volume of the samples were measured.
For the assessment of Fe concentration in wood, the treated wood samples were removed from the biobased solvents and prepared following this scheme: each sample was cut into three subsamples/layers, representing the outer, intermediate, and inner wood layers with the thickness of 2 mm per layer. Total Fe concentration was analysed by inductively coupled plasma optical emission spectroscopy according to the LST EN ISO 16968:2015.
The aromatic structure intensities and functional group identification of untreated (control) and treated wood samples were studied using Fourier-transform infrared (FT-IR) technique. FT-IR spectroscopy has often been used to investigate wood chemical composition. This method requires a simple sample preparation, is nondestructive, and uses a small sample amount. The FT-IR analyses were separately performed on the wood samples of each tested wood species. The FT-IR measurements with attenuated total reflection technique (ATR) were recorded using a Bruker Vertex 70 v (Bruker Daltonic, Bremen, Germany) spectrometer equipped with a diamond crystal ATR cell as a sample holder. The wood samples were sliced by microtome. The FT-IR spectra of wood samples were acquired in the range of 400-3800 cm −1 with a resolution of 4 cm −1 The instrument chamber was evacuated below 350 Pa during spectra acquisition to eliminate interference from CO 2 and water vapor bands. The resulting spectra were an average of 300 scans (100 scans at three random points). The spectra were normalized, and baseline corrections were made.
To determine the significant differences of wood parameters, the content of total phenolic compounds, and Fe concentration in different wood layers between different treatments, ANOVA and Tukey's Studentized Range (HSD) test were used. The different letters next to the mean values show statistically significant differences at p < 0.05 between the treatments. The obtained data were analysed using the statistical package SAS 9.4 (SAS Institute Inc., Wake County, NC, USA).

Changes of Wood Mass, Volume and Density after Treatment
The hydrothermal processing itself, when the control treatment without additives was used, caused wood mass loss by 4.2-5.3 percent for pine, spruce, and maple, and by 9.2-10.4 percent for fir and walnut wood (Table 1). When wood samples were treated with Fe 2 O 3 , the mass of wood samples decreased slightly by 2-4 percent for all species, except spruce wood, which mass did not change. When wood samples were treated with the pure additive Fe 2 O 3 , the mass of wood samples decreased by 3.5-5.8 percent for all species, except fir wood, which decreased by 15.0 percent. The treatment with the additives Fe 2 O 3 and tannic acid responded to the relatively slight mass loss for spruce, fir, maple, and walnut. The highest loss of mass was fixed in the wood treated with the pure additive FeCl 3 . The difference from the untreated wood of the same species ranged from 13.7-15.6 percent for pine, spruce, fir, and walnut to 20.7 percent for maple.
The thermal processing and different treatments caused the different responses of wood volume (Table 1). After the control treatment, wood volume decreased by 1.8-2.6 percent for fir and maple, 7.7-10.1 percent for pine, spruce, and walnut. Pure Fe 2 O 3 induced wood volume reduction from 2.6-3.8 percent for fir, maple, and walnut to 7.6-8.4 percent for pine and spruce. In this case, the addition of tannic acid caused higher wood volume reduction only for maple wood. The thermal processing with the pure additive FeCl 3 reduced wood volume by 9.2-11.4 percent for all species, except fir, in which volume decreased only by 3.8 percent. When tannic acid was added together with FeCl 3, the wood volume decrease was slightly reduced. The obtained changes in wood density were different both between different treatments and between tree species (Table 1). The control treatment caused the increase in pine and spruce wood density, the decrease in fir wood density, and had no impact on maple and walnut wood density. Similar changes in wood density were obtained after the treatment with the additive Fe 2 O 3 , showing the increased wood density of pine and spruce. The thermal treatment with the additive FeCl 3 decreased the wood density of all tree species. The highest changes of 10.5-12.1 percent from untreated wood were obtained for the fir and maple wood.
The mass loss of hydrothermally treated wood is often found due to wood thermal degradation [28,29]. Different studies showed contradictory data obtained from different wood treatment experiments: the data varied due to wood species, initial moisture contents, and treatment intensities. Tenorio and Moya [28] studied mass loss along with heat treatment and found its dependence on wood species, specifics of the drying process, and treatment intensity. Chaouch et al. [30] found that hardwoods (Fagus sylvatica L., Populus nigra L., Fraxinus excelsior L.) were more sensitive to thermal degradation than softwoods (Pinus sylvestris L., Abies pectinate Lam.), as demonstrated by the higher mass losses recorded for the same treatment duration. Only small mass losses were observed due to vaporization of volatile extractives and of bound water absorbed on the wood fibres until 160 • C and thermal degradation began at higher temperatures [30]. Earlier findings also indicated that wood species of lower densities generally are more resistant to thermal treatment than those of higher density, and this was explained by the intrinsic thermal properties of each wood species.
Norway maple Acer platanoides L.
* Note: the plus and minus signs, + and −,given next to the values indicate the increase and the decrease in the value.

Chemical Changes of Solutions, Used for Hydrothermal Wood Treatments
The biobased solvents, obtained after the thermal treatment using the different additives (Fe 2 O 3 ; FeCl 3, and tannic acid) and the control without additives, were chemically analysed for the solvent pH and the content of the total phenolic compound. The pH values of control wood biobased solvents were in the range of 3.05-3.84, depending on wood species (data not shown). The pH values of biobased solvents with FeCl 3 and Fe 2 O 3 additives after hydrothermal treatment decreased around 5.8 percent and 9.2 percent with respect to the control sample of FeCl 3 . The reduction in pH values indicated that changes arose from the strong acidity and the solid-solid reaction and formation of acidic functional groups on the surface of the wood. Acidic conditions may act as a depolymerization catalyst that further increases decomposition causing loss of mass, density, and mechanical properties. This result can be confirmed from the mass loss data shown in Table 1, where the mass of pine wood samples decreased 15.6 percent, spruce wood 3.7 percent, fir 15.5 percent, maple 20.7 percent, and walnut 13.9 percent, indicating significant reactions and changes in carbon skeleton structures from the transformation of organic matter in aqueous phase via temperature and acidic hydrolysis [31]. In addition, the components such as proteins or polysaccharides also began to hydrolyse at a hydrothermal temperature higher than 100 • C, which also affected changes in pH values of analysed biobased solvents. The solvents with Fe 2 O 3 additive showed different results with respect to the control sample. In this case, the pH values of the solvents decreased by 0.5-2.6 percent. The obtained results showed a similar trend to the mass loss data (see Table 1), where mass loss ranged between 4.1 and 5.8 percent, (except for fir wood, where mass decreased 15.5 percent). A similar tendency was observed in the solvents obtained after wood treatment with tannic acid.
The total polyphenol content of biobased solvent samples was analysed by the Folin-Ciocalteu method ( Figure 1). The measured exchanges of total phenolic compounds (TPC) have different behaviour depending on the wood species studied. In general, the increased content of TPC in the solvents was observed in all tested samples with different additives with respect to the control sample. The content of TPC in the solvents with FeCl 3 and Fe 2 O 3 additives depended only on wood species. Differences were observed of analysed solvents with Fe 2 O 3 additive, where TPC concentration did not increase significantly with respect to the control sample. Unfortunately, we could not find any similar results to compare with the data published previously. However, these results indicate the hypothesis that solvent-soluble compounds during thermal processing can cause an accumulation of the discoloured compounds on the surface due to migration from the inner wood when the solvent evaporates. This reaction is usually caused by a chemical interaction between iron ions and wood compounds, such as hydrolysable tannins, which form nonextractable compounds and increase the stain and durability of wood [32]. Meanwhile, the TPC concentration in the solvents with FeCl 3 additive increased about 1.5-2.3-fold with respect to the control samples. This indicates solubilization of the phenolic compounds found in wood, depending on pH media. Temperature and acidic medium might encourage hydrolysis associated with the release of certain phenolic compounds that are not initially soluble in water (for example, condensed tannin) or partially hydrolysed lignin fragments. Furthermore, the hydrolysis may also depend on wood species as in the case of phenolic compounds [33]. The kinetics of the reactions depends on hydrolysis as well as the mass transfer of the water to the molecules and the diffusion of the molecules from the wood to the solution [34]. The combined influence of FeCl 3 and Fe 2 O 3 and tannic acid showed that determined TPC contents were similar, except for the solvents obtained from pine and fir samples, where the TPC extraction contents significantly increased in the solvents with Fe 2 O 3 and tannic acid additives. The concentrations of TPC were 0.8-1.3-fold higher than in biobased solvents with FeCl 3 and tannic acid additive, compared with FeCl 3 without additives. In the case of spruce, maple, and walnut, a minor variation was observed between the tested biobased solvents with Fe 2 O 3 and Fe 2 O 3 and tannic acid additive, although this was not significant when observed on the experimental field. Usually, water, iron, and phenolic substances enrich the colour of the wood. When iron dissolves in water, iron ions are produced since wood is an acidic substance, and together with iron ions undergoes partial hydrolysis in an exothermic reaction. The strongly acidic solution may accelerate ionization by oxygen, in this case, iron ions react with hydroxyl groups in phenolic compounds in wood and promote a dark colour and durability of the wood [32].

Fe Concentration in Treated Wood
Tannins, extracted from wood material, have good compatibility and may form covalent bonds with hydroxyl groups on the surface of wood fibres. In this study, we used tannic acid, FeCl3 and Fe2O3 in an aqueous solution to improve some properties of wood. It is expected that this simple, green technology can improve the functionality of wood by attracting iron ions from water to some extent. The concentration of iron in the wood after different treatments showed large variability between the treatments and wood species (Table 2). In this section, the emphasis was put on the analysis of the modification effect of the iron complex on the wood surface, where a sublayers analysis was performed every 2 mm. Compared to the control samples, the content of Fe ions increased 3-15-fold in the intermediate layer, and 2-8-fold in the inner layer, depending on the different tree species. After hydrothermal treatment with FeCl3 and Fe2O3 additives, the content of iron on the outer surface of the wood increased significantly compared to the control. For instance, on the outer layer of pine wood the concentration of iron increased 77-and 48-fold, maple wood 42-and 12-fold, and walnut wood 29.6-and 15.8-fold, respectively. In general, the obtained results showed that a higher concentration of iron was accumulated on the surface when wood samples were treated with the Fe2O3 additive. According to   [35] and Zhang et al. (2012) [36], thermal treatment could enhance the magnetic properties of Fe2O3 and positive charges on surface hydroxyl groups, which were composed by the protonation and deprotonation. New active sites could be formed through electrostatic interaction with positive charges and attract iron ions. Slightly different results were obtained from wood samples treated with FeCl3 and Fe2O3 with tannic acid. The concentration of iron on the outer wood surface was lower compared to the wood species treated without tannic acid additive. For instance, the amount of iron ions on the layer of spruce wood treated with Fe2O3 was 2.2-fold higher than that treated with Fe2O3 with tannic acid additive. However, the concentration of iron ions in the intermediate and inner layers were 2.6-and 2.2-fold higher than wood treated with Fe2O3 with tannic acid additive, respectively. The same tendencies were observed for other studied wood species. It showed interesting interactions by forming complexes of tannic acid and metal ions. Unfortunately, we could not find any similar results to compare with the data published previously. However, according to Ma et al. (2020) [37], a tannic acid and metal complex was able to generate rough hierarchical structures on the surface of the wood. A study by Huang et al. (2020) [38] showed that ferrous sulphate reacted with tannic acid to form a complex of ferrous-tannic acid. This complex was unstable and was easy to oxidize to form ferric tannate composite, which resulted in the coexistence of Fe 2+ and Fe 3+ in the

Fe Concentration in Treated Wood
Tannins, extracted from wood material, have good compatibility and may form covalent bonds with hydroxyl groups on the surface of wood fibres. In this study, we used tannic acid, FeCl 3 and Fe 2 O 3 in an aqueous solution to improve some properties of wood. It is expected that this simple, green technology can improve the functionality of wood by attracting iron ions from water to some extent. The concentration of iron in the wood after different treatments showed large variability between the treatments and wood species ( Table 2). In this section, the emphasis was put on the analysis of the modification effect of the iron complex on the wood surface, where a sublayers analysis was performed every 2 mm. Compared to the control samples, the content of Fe ions increased 3-15-fold in the intermediate layer, and 2-8-fold in the inner layer, depending on the different tree species. After hydrothermal treatment with FeCl 3 and Fe 2 O 3 additives, the content of iron on the outer surface of the wood increased significantly compared to the control. For instance, on the outer layer of pine wood the concentration of iron increased 77-and 48-fold, maple wood 42-and 12-fold, and walnut wood 29.6-and 15.8-fold, respectively. In general, the obtained results showed that a higher concentration of iron was accumulated on the surface when wood samples were treated with the Fe 2 O 3 additive. According to   [35] and Zhang et al. (2012) [36], thermal treatment could enhance the magnetic properties of Fe 2 O 3 and positive charges on surface hydroxyl groups, which were composed by the protonation and deprotonation. New active sites could be formed through electrostatic interaction with positive charges and attract iron ions. Slightly different results were obtained from wood samples treated with FeCl 3 and Fe 2 O 3 with tannic acid. The concentration of iron on the outer wood surface was lower compared to the wood species treated without tannic acid additive. For instance, the amount of iron ions on the layer of spruce wood treated with Fe 2 O 3 was 2.2-fold higher than that treated with Fe 2 O 3 with tannic acid additive. However, the concentration of iron ions in the intermediate and inner layers were 2.6-and 2.2-fold higher than wood treated with Fe 2 O 3 with tannic acid additive, respectively. The same tendencies were observed for other studied wood species. It showed interesting interactions by forming complexes of tannic acid and metal ions. Unfortunately, we could not find any similar results to compare with the data published previously. However, according to Ma et al. (2020) [37], a tannic acid and metal complex was able to generate rough hierarchical structures on the surface of the wood. A study by Huang et al. (2020) [38] showed that ferrous sulphate reacted with tannic acid to form a complex of ferrous-tannic acid. This complex was unstable and was easy to oxidize to form ferric tannate composite, which resulted in the coexistence of Fe 2+ and Fe 3+ in the composites immobilized on wood fibres. Tannic acid-Fe complexes may adhere to the surface of wood fibres via strong covalent bonds [38]. There, abundant phenolic hydroxyl groups on the surface of tannic acid make the tannin-Fe complex on the surface of the modified material more hydrophilic. Therefore, it is possible that Fe 2+ ions may be absorbed into intermediate and inner wood layers due to the chelation properties of tannic acid, and the ability to form the tannin-Fe complex.

FTIR Analysis of Wood Samples
Hydrothermal treatment with various additives was used to increase hardwood and softwood functionality. The FTIR spectra of obtained control and studied hardwood and softwood samples indicated the fingerprint regions (Figures 2-6).               Due to the complexity of the wood, many characteristic peaks in the fingerprint region of the spectra made it complicated to assign all spectral peaks to their constituents. Analysed samples showed infrared spectra with the same profile, whereas the positions of the main bands remained the same, and only minor intensities occurred with respect to different wood samples. Small shifts were observed as a function between the treated and untreated wood samples (Figures 2-6). The characteristic signal bands' assignments of the main peaks, observed in the fingerprint region, which ranged between 3500 and 400 cm −1 , were made on the basis of the literature data ( Table 3). The bands at O-H stretching band were at 3500-3000 cm −1 and did not significantly change during the thermal treatment. Notable observations include a broad band at 3300 and 2921 cm −1 , which were attributed to the hydroxyl groups in phenolic and aliphatic structures of the cellulose. The bands between 2924 and 2854 cm −1 , arising predominantly from CH stretching in aromatic methoxyl groups and in methyl and methylene groups of side chains [39], were particularly clear for the control samples of maple, spruce, and fir. The bands from 1400 to 1800 cm −1 were assigned to the aromatic nature of the structure. Compared the FTIR spectrum of the control samples with wood samples treated with FeCl3 and Fe2O3 and tannic acid additives, peaks of phenolic (O-H) deformation region were slightly more broadened than in the control samples. These changes occurred due to changes in Fe-tannic acid complexation. Complexes that are formed between Fe ions and phenolic groups with the presence of hydroxyl groups increase the stability of the complexes and form a stable metal-polyphenol complex coating which had a high affinity. A possible mechanism of ion change could be considered as Fe 3+/2+ attaching hydroxyl groups donating electrons to metal ions [40,41]. The consisted Fe-tannin complex strongly binds carbohydrate-binding proteins, which are usually rich in residues featuring aromatic side chains. Therefore, the affinity between the Fe-tannin complex and carbohydrates is attributed to Van der Waals force and hydrophobic interactions, suggesting that hydrophobic interactions play an important role in the formation of surface-bound units [42]. According to Grasel et al. (2015) [43] and Falcão et al. (2013) [44], the C-O vibrations of condensed tannins can be seen at 1375-1365; 1160-1155 and 1075-1065 cm −1 , while hydrolysable tannins at 1315-1290; 1200-1170 and 1035-1025 cm −1 . According to our data, hydrolysable tannins together with their Fe complexes can be seen according to C-O stretch vibrations at 1315-1105 cm −1 . Due to the complexity of the wood, many characteristic peaks in the fingerprint region of the spectra made it complicated to assign all spectral peaks to their constituents. Analysed samples showed infrared spectra with the same profile, whereas the positions of the main bands remained the same, and only minor intensities occurred with respect to different wood samples. Small shifts were observed as a function between the treated and untreated wood samples (Figures 2-6). The characteristic signal bands' assignments of the main peaks, observed in the fingerprint region, which ranged between 3500 and 400 cm −1 , were made on the basis of the literature data ( Table 3). The bands at O-H stretching band were at 3500-3000 cm −1 and did not significantly change during the thermal treatment. Notable observations include a broad band at 3300 and 2921 cm −1 , which were attributed to the hydroxyl groups in phenolic and aliphatic structures of the cellulose. The bands between 2924 and 2854 cm −1 , arising predominantly from CH stretching in aromatic methoxyl groups and in methyl and methylene groups of side chains [39], were particularly clear for the control samples of maple, spruce, and fir. The bands from 1400 to 1800 cm −1 were assigned to the aromatic nature of the structure. Compared the FTIR spectrum of the control samples with wood samples treated with FeCl 3 and Fe 2 O 3 and tannic acid additives, peaks of phenolic (O-H) deformation region were slightly more broadened than in the control samples. These changes occurred due to changes in Fe-tannic acid complexation. Complexes that are formed between Fe ions and phenolic groups with the presence of hydroxyl groups increase the stability of the complexes and form a stable metal-polyphenol complex coating which had a high affinity. A possible mechanism of ion change could be considered as Fe 3+/2+ attaching hydroxyl groups donating electrons to metal ions [40,41]. The consisted Fe-tannin complex strongly binds carbohydrate-binding proteins, which are usually rich in residues featuring aromatic side chains. Therefore, the affinity between the Fe-tannin complex and carbohydrates is attributed to Van der Waals force and hydrophobic interactions, suggesting that hydrophobic interactions play an important role in the formation of surface-bound units [42]. According to Grasel et al. (2015) [43] and Falcão et al. (2013) [44], the C-O vibrations of condensed tannins can be seen at 1375-1365; 1160-1155 and 1075-1065 cm −1 , while hydrolysable tannins at 1315-1290; 1200-1170 and 1035-1025 cm −1 . According to our data, hydrolysable tannins together with their Fe complexes can be seen according to C-O stretch vibrations at 1315-1105 cm −1 .
The bands located at 1507 and 1593 cm −1 , which purely comprise aromatic skeletal vibrations of benzene rings, were recorded as the characteristic peaks of lignin. A vibration region ranged from 1150 to 1300 cm −1 stretching C-O vibrations, and more speciality recorded near 1230 cm −1 which referred to syringyl lignin. According to Saha et al. (2019) [45] in hardwoods usually, the guaiacyl-syringyl type of lignin is predominant, while in softwoods, guaiacyl type lignin is common. The peak at 1273 cm −1 was caused by guaiacyl lignin. Ramage et al. (2017) [46] and Sharma et al. (2020) [47] reported that softwood and hardwood differ in their chemical composition, with various forms of lignin components such as guaiacyl or syringyl units, hemicelluloses such as xylan, glucomannan, and mannose constituent and wood extractives. Our research partially substantiated this statement. Cellulose and hemicellulose [47,48] 1230 C-O-C symmetric stretching and OH plane deformation Syringyl lignin [45] 1273 C-C Guaiacyl lignin structure [45,53] The largest values of FT-IR spectra were assigned to absorption bands between 1050-1250 cm −1 , which gave bands C-O-C bridges in cellulose and hemicelluloses. Moreover, there was C-O bands (1026 and 1050 cm −1 ) stretching, usually attributed to cellulose and hemicellulose. These peaks were assigned to the C=O and COO asymmetric tension vibrations in the aromatic rings and aromatic skeletal and the C=O stretch vibration, respectively [48].
In general, during wood treatment, only the minor part of lignin was shifted. Lignin was affected during hydrothermal treatment enriched with inorganic/organic components, which is the probable reason for the increased intensity of this peak. FTIR spectra of this study may assume that used additives during hydrothermal treatment may have an influence on lignin shift, however, it can be concluded that increased temperature may affect examined wood during the hydrothermal process.

Conclusions
The simulation of artificial conditions, containing various chemical additives close to the bog environment, was hypothesized as a relatively short way to develop a more sustainable wood product. As the objective of this study was to test different wood modification treatments, searching for environmentally friendly technology, the impact of the phenolic substance-tannic acid-naturally found in wood, especially oak wood, was also tested. The potential of FTIR spectroscopy combined with other chemometric methods was evaluated to show the influence of additives such as Fe 2 O 3 and FeCl 3 in combination with tannic acid on wood samples. It was observed that the softwood samples were more susceptible to hydrothermal modification than the hardwood samples, indicated by minor different intensities on infrared spectra. FTIR results showed bands of hardwood and softwood samples, which along with the spectral characterization of the wood constituents, can further be utilized for creating a larger FTIR spectral database of wood samples for their identification. The factors with the most significant effect on the hydrothermal modification were tannic acid and inorganic Fe 3 O 2 salt used as the additive. The results indicated the suitability of the inorganic Fe 3 O 2 salt and tannic acid combination applied in this type of modification because the higher concentration of iron was distributed on the inner and intermediate sublayers of wood samples. The hydrothermal modification approach to anchor tannic acid-Fe macromolecules in wood shown in this study has promising capabilities and may attract interest in the field of wood protection against weathering.