Discoloration and Surface Changes in Spruce Wood after Accelerated Aging

Spruce wood is widely used in outdoor applications, but its susceptibility to degradation under exposure to sunlight and moisture is a major concern. This study investigates the impact of accelerated aging on spruce wood’s surface chemistry, microstructure, geometry, and discoloration. The study was performed in two outdoor aging modes: dry and wet. The accelerated aging effects were evident in the changes in spruce wood structure, as well as in the other studied properties. During aging, it developed significant discoloration. Under simulated rainless outdoor conditions (dry mode), spruce wood gradually became dark brown. Under conditions involving rain (wet mode), the discoloration was qualitatively different from the dry mode. FTIR spectroscopy showed that during the accelerated aging of wood, lignin was mainly degraded, especially in the early stages of the process. A linear correlation was found between the changes in lignin and the color changes in the wood. There was an increase in carbonyl groups in the dry mode, which contributed to the color change and was also influenced by changes in extractives. The wet mode caused the leaching out of carbonyl groups. The observed decrease in cellulose crystallinity, together with the degradation of hydrophobic lignin, may result in the increased hydrophilicity of photodegraded wood. For both modes, there were different changes in the wood micro- and macrostructure, reflected in the surface morphology. The roughness increased during the aging process in both modes. The slightest changes in the roughness parameters were identified in the grain direction in the dry mode; the most evident was that the roughness parameters increased perpendicular to the grain in the wet mode. The demonstrated mechanism backing up the aging-related changes to the spruce wood structure and the relations unveiled between these changes and the changes in the spruce wood surface properties can provide an issue point for seeking ways how to mitigate the negative effects of the environmental factors the wood is exposed to.


Introduction
In outdoor environments, the surfaces of almost all materials are affected by various radiation types, moisture, temperature, and emissions, which act in interactions.This induces the gradual degradation of the exposed surfaces.Wood is not an exception.Wood, a natural composite material, exhibits unique aging behavior compared with other materials [1][2][3][4].Wood has a characteristically complex surface morphology, backed up by its heterogeneous anatomical structure at the macro-, micro-, submicro-, and nanolevels [5,6].In addition, wood material is hygrophilous, porous, and anisotropic, which is important for its degradation during the aging process.Diversely shaped and assembled anatomical elements of wood represent a heterogeneous porous system.Besides this, species-dependent differences in pore shape, size, mutual linking, and arrangement are negative effects of environmental agents.This study observed and evaluated the influence of radiation on the accelerated aging process and the coupled influence of radiation in the interaction with water on changes in spruce wood surface's chemical composition, structure, geometry, and discoloration.

Experimental Material
The degradation phenomena generated during the accelerated aging process were studied on radial and tangential surfaces of spruce wood specimens, each 100 mm × 50 mm × 15 mm (length × width × thickness) in size-see Figure 1.All the specimens were exposed to a relative air humidity of 65% and a temperature of 20 • C for one month to attain a wood moisture content of 12%.Before placing the specimens into the xenotest chamber, their relevant surface properties and body parameters (roughness and waviness profiles, chemical properties, and color) were measured.These values served as comparative standards for evaluating the changes in the studied properties measured during aging.Changes in wood surface structure and properties associated with accelerated aging were observed at intervals after 100, 200, 400, and 600 aging hours.For each aging mode, 15 radial and 15 tangential specimens were used.
Polymers 2024, 16, x FOR PEER REVIEW 3 of 19 in making constructions intended for outdoor conditions, and, as such, it is exposed to the negative effects of environmental agents.This study observed and evaluated the influence of radiation on the accelerated aging process and the coupled influence of radiation in the interaction with water on changes in spruce wood surface's chemical composition, structure, geometry, and discoloration.

Experimental Material
The degradation phenomena generated during the accelerated aging process were studied on radial and tangential surfaces of spruce wood specimens, each 100 mm × 50 mm × 15 mm (length × width × thickness) in size-see Figure 1.All the specimens were exposed to a relative air humidity of 65% and a temperature of 20 °C for one month to attain a wood moisture content of 12%.Before placing the specimens into the xenotest chamber, their relevant surface properties and body parameters (roughness and waviness profiles, chemical properties, and color) were measured.These values served as comparative standards for evaluating the changes in the studied properties measured during aging.Changes in wood surface structure and properties associated with accelerated aging were observed at intervals after 100, 200, 400, and 600 aging hours.For each aging mode, 15 radial and 15 tangential specimens were used.

Accelerated Aging
The accelerated wood aging was simulated in a xenotest chamber Q-SUN Xe-3-HS (Q-Lab Europe, Ltd., Bolton, UK).The experimental material was arranged equidistantly across the xenotest chamber.To ensure equal radiation intensity and heat for all the specimens, the specimens were regularly shifted according to the recommended schedule.
The aging conditions in the xenotest chamber followed the standard ASTM G 155 [30].This standard is a fundamental one in determining the conditions for accelerated aging for non-metallic materials with the aid of a xenon discharge tube.Two modes for outdoor conditions were chosen: "dry mode" and "wet mode".The first mode simulated outdoor conditions in the case of when wood is exposed to UV radiation but protected from rainwater; the second simulated the conditions when wood is exposed to both factors, UV and rain.
The radiation intensity was 0.35 W•m -2 , with a radiation wavelength of 340 nm, following the standard.This value corresponds to the mean annual value for the temperate zone.The temperature, controlled on a black panel, corresponded to the maximum temperature on the panel surface.In both modes, one accelerated aging cycle consisted of two steps, covering 120 min altogether (Tables 1 and 2).For each mode, the total duration of the aging process was 600 h, which was equivalent to 300 cycles.

Accelerated Aging
The accelerated wood aging was simulated in a xenotest chamber Q-SUN Xe-3-HS (Q-Lab Europe, Ltd., Bolton, UK).The experimental material was arranged equidistantly across the xenotest chamber.To ensure equal radiation intensity and heat for all the specimens, the specimens were regularly shifted according to the recommended schedule.
The aging conditions in the xenotest chamber followed the standard ASTM G 155 [30].This standard is a fundamental one in determining the conditions for accelerated aging for non-metallic materials with the aid of a xenon discharge tube.Two modes for outdoor conditions were chosen: "dry mode" and "wet mode".The first mode simulated outdoor conditions in the case of when wood is exposed to UV radiation but protected from rainwater; the second simulated the conditions when wood is exposed to both factors, UV and rain.
The radiation intensity was 0.35 W•m -2 , with a radiation wavelength of 340 nm, following the standard.This value corresponds to the mean annual value for the temperate zone.The temperature, controlled on a black panel, corresponded to the maximum temperature on the panel surface.In both modes, one accelerated aging cycle consisted of two steps, covering 120 min altogether (Tables 1 and 2).For each mode, the total duration of the aging process was 600 h, which was equivalent to 300 cycles.

Fourier Transform Infrared (FTIR) Spectroscopy
FTIR spectra of the wood surface were recorded on a Nicolet iS10 FT-IR spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) equipped with Smart iTR using an Attenuated Total Reflectance (ATR) sampling accessory with a diamond crystal (Thermo Fisher Scientific).The spectra were recorded in an absorbance mode (A) from 4000 to 650 cm −1 with a spectral resolution of 4 cm −1 , and 32 scans were used.Measurements were made on four replicates per sample.The peak at 1030 cm −1 (assigned to CO stretching) was used to normalize the obtained infrared spectra.This band shows little changes during wood photodegradation in a xenotest chamber [17].Calculations for changes due to aging were made using the ratio of absorbances A 0 /A t , where A 0 is the absorbance of the corresponding peak of the reference sample, and A t is the absorbance of the peak at time t.

Color Measurement
The colorimetric values of the coordinates L*, a*, and b* on the referential and engraved specimens were measured with a spectrophotometer Spectro-guide 45/0 gloss (BYK-GARDNER GmbH, Geretsried, Germany).The measurements were taken at ten spots per specimen.The color differences ∆L*, ∆a*, and ∆b* under different irradiation conditions and the total color difference ∆E were determined according to the following equations: Note: the "2 − n" means the color value after wood surface irradiation, and the index "1" denotes the referential value measured on the wood surface before the aging process.
For each mode, the color was measured at five spots on each specimen, so the total number of measurements for each surface treatment mode was 75.

Evaluation of Wood Surface Morphology
The surface morphology of the wood irradiated at the xenotest was evaluated at specified time intervals based on the roughness and waviness parameters: R a (mean arithmetic deviation) and R z (the maximum peak height plus the maximum depression depth within the cut-off, or a sampling length) and Wa (mean arithmetic deviation) and Wt (the maximum peak height plus the maximum depression depth within the entire evaluation length).The roughness and waviness profiles were recorded with the aid of a profilometer, Surfcom 130A (Carl Zeiss, Oberkochen, Germany), consisting of a measuring unit and an evaluation unit.
Roughness was measured on radial and tangential surfaces, parallel to the grain course and perpendicular to the grain course, at five different spots.The entire traversing length consisted of a pre-length segment, five sample length segments lr (cut-off λc), and a post-length segment lp.The basic lengths were chosen from the interval 0.025−8 mm based on the preliminary measured values of the roughness parameters R a and R z .
The degradation of the wood surface structure after having finished the aging process was studied with the aid of a SEM microscope.

Evaluation of the Results
The results were processed in the program STATISTICA version 10.0 (Dell Statistica, Prague, Czech Republic).The basic statistical characteristics (arithmetical mean, standard deviation) were determined using descriptive statistics.The impacts of the studied factors on the given properties were evaluated with the multi-way variance analysis (MANOVA) and Duncan's test.

Wood Discoloration Induced by Accelerated Aging Process
The accelerated wood aging process induced considerable surface discoloration in spruce test specimens.This was true for both the dry and the wet modes.The results of the three-way variation analysis confirmed significant influences for all three factors investigated (aging mode, aging time, and irradiated surface type) and their interactions on spruce wood discoloration.The basic statistical characteristics for the individual color coordination values are shown in Table 3.Before aging, the average values of lightness L* in the individual spruce sample sets ranged from 84 to 88 (Table 3).These values were in good accordance with the values reported for spruce wood in [31,32].
During the accelerated aging process simulating outdoor conditions without rain (dry mode), lightness significantly decreased with extending irradiation time.During aging, the values of the color coordinates a* and b* also varied (Table 3).In all cases, the a* and b* values were positive, which means that the a* values fitted the red zone, and the b* values ranged within the yellow zone, equally in good accordance with the above-cited authors.With the progressing aging time, the coordinate a* was getting more and more saturated.The coordinate b* exhibited significantly increasing saturation as far as 200 aging hours; then, no significant variation could be observed.
The color variation generated by the effects of solar radiation simulated in the aging process could be expressed through the color coordinate differences ∆L*, ∆a*, and ∆b*; see Figure 2. The variations in color coordinates were also reflected in the total color difference ∆E*, with ∆E* > 12, which corresponds to the degree six of the six-degree discoloration scale [33], thus representing a different color hue compared with the original one.With further advances in aging time, the total color difference values increased, too, far beyond the value of 12. Before aging, the average values of lightness L* in the individual spruce sample sets ranged from 84 to 88 (Table 3).These values were in good accordance with the values reported for spruce wood in [31,32].
During the accelerated aging process simulating outdoor conditions without rain (dry mode), lightness significantly decreased with extending irradiation time.During aging, the values of the color coordinates a* and b* also varied (Table 3).In all cases, the a* and b* values were positive, which means that the a* values fitted the red zone, and the b* values ranged within the yellow zone, equally in good accordance with the above-cited authors.With the progressing aging time, the coordinate a* was getting more and more saturated.The coordinate b* exhibited significantly increasing saturation as far as 200 aging hours; then, no significant variation could be observed.
The color variation generated by the effects of solar radiation simulated in the aging process could be expressed through the color coordinate differences ∆L*, ∆a*, and ∆b*; see Figure 2. The variations in color coordinates were also reflected in the total color difference ∆E*, with ∆E* > 12, which corresponds to the degree six of the six-degree discoloration scale [33], thus representing a different color hue compared with the original one.With further advances in aging time, the total color difference values increased, too, far beyond the value of 12.The aging time-related color variations on the tangential and the radial surfaces were very similar, both in terms of quality and quantity.Despite the fact that statistically significant differences were confirmed between these two surface types, from a practical standpoint, they are unimportant.We suppose that these different color patterns are mainly due to the diverse heterogeneous structures across the radial and tangential surfaces.Spruce surface color variation during accelerated aging simulating outdoor conditions is also documented in Figure 3. Spruce wood belongs to light-colored wood species.In these species, the photodegradation causes more intensive darkening [9,10,13].The aging time-related color variations on the tangential and the radial surfaces were very similar, both in terms of quality and quantity.Despite the fact that statistically significant differences were confirmed between these two surface types, from a practical standpoint, they are unimportant.We suppose that these different color patterns are mainly due to the diverse heterogeneous structures across the radial and tangential surfaces.Spruce surface color variation during accelerated aging simulating outdoor conditions is also documented in Figure 3. Spruce wood belongs to light-colored wood species.In these species, the photodegradation causes more intensive darkening [9,10,13].During the accelerated aging process simulating outdoor conditions in the presence of rain (wet mode), the color variation was qualitatively different from the one associated with the dry mode.The lightness exhibited a conspicuous reduction after the first aging hours, with the coordinates a* and b* growing in saturation in the red and yellow areas, respectively (Table 3).This resulted in coloring the spruce wood into darker brown.After 100 aging hours, the ∆E* values were >12 (Figure 4), corresponding again to the degree six of the six-degree scale [33].After this moment, the interaction between radiation and water caused the wood surface color to fade step by step.After 600 aging hours, the lightness values were close to the original ones.The values of a* and b* decreased, too, shifting to achromatic colors.In this way, the wood obtained a brown-grey hue, the so-called patina.As the total color difference values ∆E* were close to 10, the final discoloration was found considerable thus far (degree five of a six-degree scale).Also, in this case, the aging timedependent differences between the radial and the tangential surfaces were very similar in their quality and quantity and insignificant in terms of practical use.The color variation is also documented in Figure 5.During the accelerated aging process simulating outdoor conditions in the presence of rain (wet mode), the color variation was qualitatively different from the one associated with the dry mode.The lightness exhibited a conspicuous reduction after the first aging hours, with the coordinates a* and b* growing in saturation in the red and yellow areas, respectively (Table 3).This resulted in coloring the spruce wood into darker brown.After 100 aging hours, the ∆E* values were >12 (Figure 4), corresponding again to the degree six of the six-degree scale [33].After this moment, the interaction between radiation and water caused the wood surface color to fade step by step.After 600 aging hours, the lightness values were close to the original ones.The values of a* and b* decreased, too, shifting to achromatic colors.In this way, the wood obtained a brown-grey hue, the so-called patina.As the total color difference values ∆E* were close to 10, the final discoloration was found considerable thus far (degree five of a six-degree scale).Also, in this case, the aging time-dependent differences between the radial and the tangential surfaces were very similar in their quality and quantity and insignificant in terms of practical use.The color variation is also documented in Figure 5.
At relevant time intervals in the accelerated aging process, average values of color coordinates were measured experimentally on spruce wood surfaces.These values were then used as the input data for Adobe Photoshop to perform color simulation (Figures 3 and 5-model).The figures illustrate good accordance between the factual specimens and the model obtained using the simulation.During the accelerated aging process simulating outdoor conditions in the presence of rain (wet mode), the color variation was qualitatively different from the one associated with the dry mode.The lightness exhibited a conspicuous reduction after the first aging hours, with the coordinates a* and b* growing in saturation in the red and yellow areas, respectively (Table 3).This resulted in coloring the spruce wood into darker brown.After 100 aging hours, the ∆E* values were >12 (Figure 4), corresponding again to the degree six of the six-degree scale [33].After this moment, the interaction between radiation and water caused the wood surface color to fade step by step.After 600 aging hours, the lightness values were close to the original ones.The values of a* and b* decreased, too, shifting to achromatic colors.In this way, the wood obtained a brown-grey hue, the so-called patina.As the total color difference values ∆E* were close to 10, the final discoloration was found considerable thus far (degree five of a six-degree scale).Also, in this case, the aging timedependent differences between the radial and the tangential surfaces were very similar in their quality and quantity and insignificant in terms of practical use.The color variation is also documented in Figure 5.At relevant time intervals in the accelerated aging process, average values of color coordinates were measured experimentally on spruce wood surfaces.These values were then used as the input data for Adobe Photoshop to perform color simulation (Figures 3  and 5-model).The figures illustrate good accordance between the factual specimens and the model obtained using the simulation.

FTIR Spectroscopy
The color changes observed for the two accelerated aging modes were caused by changes in the wood surface chemistry.FTIR spectra showed that UV radiation alone or in combination with water leaching causes significant changes on the surface of spruce wood (Figures 6−9).The fingerprint region (1180-1900 cm −1 ) is shown in this study for clarity.

FTIR Spectroscopy
The color changes observed for the two accelerated aging modes were caused by changes in the wood surface chemistry.FTIR spectra showed that UV radiation alone or in combination with water leaching causes significant changes on the surface of spruce wood (Figures 6-9).The fingerprint region (1180-1900 cm −1 ) is shown in this study for clarity.At relevant time intervals in the accelerated aging process, average values of color coordinates were measured experimentally on spruce wood surfaces.These values were then used as the input data for Adobe Photoshop to perform color simulation (Figures 3  and 5-model).The figures illustrate good accordance between the factual specimens and the model obtained using the simulation.

FTIR Spectroscopy
The color changes observed for the two accelerated aging modes were caused by changes in the wood surface chemistry.FTIR spectra showed that UV radiation alone or in combination with water leaching causes significant changes on the surface of spruce wood (Figures 6−9).The fingerprint region (1180-1900 cm −1 ) is shown in this study for clarity.The most sensitive molecule is the lignin macromolecule because it is a good absorber of ultraviolet light.The photons of UV radiation can break the aromatic bonds of lignin.The produced free radicals react with oxygen to form carbonyl groups [34][35][36].The characteristic peak for guaiacyl lignin (1508 cm −1 ) [37] decreased by approximately 98% of the original absorbance, with a greater decrease observed in the wet mode, where it completely disappeared.A sharp decrease occurred during the first 100 h of irradiation, and further accelerating aging had little effect on lignin changes.Similar results had been obtained by [38], who found a decrease in the peak at 1510 cm −1 to 60% of the original absorbance value when the spruce wood was irradiated with UV light.In addition to the decrease in the peak at 1508 cm −1 , reduced absorbance was also observed around 1470 cm −1 (aromatic deformation of the C-H) and at 1264 cm −1 (guaiacyl ring breathing), which agrees with the results obtained with UV irradiation and the water leaching of Scots pine and spruce wood [39][40][41].Several authors found a strong decrease in the absorbance of lignin in wood during the photodegradation process associated with the formation of carbonyl groups [38,41,42].
Changes in lignin were, therefore, accompanied by the formation of new carbonyl groups (1653 cm −1 − C=O conjugated and 1730 cm −1 − C=O unconjugated), which caused changes in the color of the wood.The formation of carbonyl groups indicates photoinduced oxidation of the wood surface.The cleavage of β-O-4 bonds in lignin leads to the formation of different types of chromophores, in particular quinones [38,43].The growth of non-conjugated carbonyl groups was more pronounced in the early stages of irradiation.In contrast to the dry mode, where there was an increase in carbonyl groups (1653 cm −1 − C=O conjugated and 1730 cm −1 − C=O non-conjugated), in the wet mode of accelerated aging, we observed a decrease in both types of carbonyl groups due to the extraction of the formed carbonyl compounds with water, which is in line with the observation of [44,45].Color change follows a similar pattern to lignin degradation, with the greatest changes occurring in the early stages of irradiation and being greater in the T direction than in the R direction.We found a linear correlation between lignin content and color changes with high correlation coefficients (from 0.9653 to 0.9893) (Figure 10) and a non-linear correlation between carbonyl groups at 1735 cm −1 , which was also observed by Müller et al. [38].These authors explain the linear dependence with the formation of quinones and the non-linear dependence with the formation of aliphatic carbonyl bonds.The photo-oxidation and oxidation of cellulose and hemicelluloses lead to the formation of the aldehyde and ketone groups on carbon atoms C2 and C3, which may contribute to the increased absorbance at wavelengths of 1653 and 1730 cm −1 [38], but the rate of change in cellulose is much lower than in lignin [13].However, apart from lignin, the largest color changes during photodegradation can be caused by extractives.However, the content of extractives in wood is so small that, according to some authors [14,46], it is not possible to monitor their chemical changes using the FTIR method.On the other hand, Pandey [42], comparing unextracted and extracted UV-irradiated Acacia auriculaeformis wood, found that the photodegradation of polyphenolic extractives contributes to color changes.The different composition and content of extractives in individual woody plants can, therefore, influence color changes in the photodegradation process of the wood in different ways, which is consistent with several works [9,10,42].Correlations between color changes and changes in the mechanical properties of wood have been found in several works.According to our research, this correlation applies mainly to the woody plants of the temperate zone, and even then, only if these woody plants do not have significant amounts and composition of extractives [47].The ratio of absorbances A1370/A1508 can be used to calculate the wood crystallinity index (CI) [48].The peak at 1370 cm −1 is characteristic of cellulose, and that at 1508 cm −1 is characteristic of lignin; in our experiments, a linear increase with high correlation coefficients (r = 0.92-0.98)occurred due to the faster degradation of amorphous lignin compared with mostly crystalline cellulose, and greater changes occurred in the wet mode compared The ratio of absorbances A1370/A1508 can be used to calculate the wood crystallinity index (CI) [48].The peak at 1370 cm −1 is characteristic of cellulose, and that at 1508 cm −1 is characteristic of lignin; in our experiments, a linear increase with high correlation coefficients (r = 0.92-0.98)occurred due to the faster degradation of amorphous lignin compared with mostly crystalline cellulose, and greater changes occurred in the wet mode compared with the dry mode (Tables 4 and 5).Similar changes were also observed during the natural weathering of spruce wood [39] and the accelerated aging of pine wood [48].FTIR spectroscopy can be used to study changes in cellulose, particularly changes in its crystallinity.The absorbance at 1420-1430 cm −1 is caused by skeletal vibrations associated with C-H planar deformations of cellulose and is attributed to its crystalline part.Bands at 1372-1375 cm −1 (CH deformation vibrations in cellulose and hemicelluloses) and 1160-1165 cm −1 (C-O-C asymmetric valence vibrations in cellulose and hemicelluloses) are characteristic of carbohydrates.The band at 898 cm −1 is caused by CH deformations in cellulose and its amorphous part [49][50][51][52].
The A1420/A898 ratio was proposed by Nelson and O'Connor [53] as an empirical crystallinity index (LOI-lateral order index), and the A1372/A2900 ratio as a total crystallinity index (TCI).Regarding the mobility of cellulose chains, hydrogen bond intensity (HBI) is closely related to the crystallinity of cellulose, as well as to the amount of bound water [49].The band at 3400 cm −1 is due to −OH valence vibrations and at 1320 cm −1 to CH 2 rocking vibrations in cellulose, and the A3400/A1320 ratio is used to study HBI in different wood species [51,54,55].
Accelerated aging causes a similar progression of LOI values, namely a decrease in the first stage of action, which is consistent with the results obtained in the photodegradation of deciduous and coniferous trees [56,57].The decrease in TCI values accelerates in the later stages of photodegradation.The decrease in crystallinity increases the amount of amorphous cellulose, which leads to an increase in the hydrophilicity of aged wood.Changes in HBI values indicate the reorganization of hydrogen bonds or changes inbound water in cellulose.

Spruce Wood Surface Geometry and Morphology
The four-way variance analysis processing confirmed significant impacts for all four evaluated factors (aging mode, aging time, anatomical direction, and irradiated surface), as well as their interactions on the roughness parameters discussed.
The original spruce wood surface pre-treated with milling exhibited the lowest roughness values.As the unevenness of these surfaces overpassed 500 nm, they were considered surfaces with rough unevenness [58], and this is typical for wood.In most cases, significant differences in roughness between the radial and the tangential surfaces were not identified.Occasionally, significant differences in roughness parameters were confirmed; these, however, were not important from a practical viewpoint.Significantly lower roughness values that were measured in the grain direction than perpendicular to the grain were due to the orientation of the cell wall elements.The roughness data we obtained for the milled surface were in good accordance with the results reported in [59][60][61].
The advancing aging process caused roughness to increase in both aging modes (Figure 11).This figure demonstrates that the smallest changes in roughness parameters were observed in the grain direction in the dry mode, and the highest increase in roughness parameters was found perpendicular to the grain in the wet mode.In this case, 600 aging hours resulted in more than a three-fold increase in the two roughness parameters.The aging-related changes in the roughness parameters R a and R z were practically identical on the radial and tangential surfaces.
Changes in HBI values indicate the reorganization of hydrogen bonds or changes inbound water in cellulose.

Spruce Wood Surface Geometry and Morphology
The four-way variance analysis processing confirmed significant impacts for all four evaluated factors (aging mode, aging time, anatomical direction, and irradiated surface), as well as their interactions on the roughness parameters discussed.
The original spruce wood surface pre-treated with milling exhibited the lowest roughness values.As the unevenness of these surfaces overpassed 500 nm, they were considered surfaces with rough unevenness [58], and this is typical for wood.In most cases, significant differences in roughness between the radial and the tangential surfaces were not identified.Occasionally, significant differences in roughness parameters were confirmed; these, however, were not important from a practical viewpoint.Significantly lower roughness values that were measured in the grain direction than perpendicular to the grain were due to the orientation of the cell wall elements.The roughness data we obtained for the milled surface were in good accordance with the results reported in [59][60][61].
The advancing aging process caused roughness to increase in both aging modes (Figure 11).This figure demonstrates that the smallest changes in roughness parameters were observed in the grain direction in the dry mode, and the highest increase in roughness parameters was found perpendicular to the grain in the wet mode.In this case, 600 aging hours resulted in more than a three-fold increase in the two roughness parameters.The aging-related changes in the roughness parameters Ra and Rz were practically identical on the radial and tangential surfaces.Changes in HBI values indicate the reorganization of hydrogen bonds or changes inbound water in cellulose.

Spruce Wood Surface Geometry and Morphology
The four-way variance analysis processing confirmed significant impacts for all four evaluated factors (aging mode, aging time, anatomical direction, and irradiated surface), as well as their interactions on the roughness parameters discussed.
The original spruce wood surface pre-treated with milling exhibited the lowest roughness values.As the unevenness of these surfaces overpassed 500 nm, they were considered surfaces with rough unevenness [58], and this is typical for wood.In most cases, significant differences in roughness between the radial and the tangential surfaces were not identified.Occasionally, significant differences in roughness parameters were confirmed; these, however, were not important from a practical viewpoint.Significantly lower roughness values that were measured in the grain direction than perpendicular to the grain were due to the orientation of the cell wall elements.The roughness data we obtained for the milled surface were in good accordance with the results reported in [59][60][61].
The advancing aging process caused roughness to increase in both aging modes (Figure 11).This figure demonstrates that the smallest changes in roughness parameters were observed in the grain direction in the dry mode, and the highest increase in roughness parameters was found perpendicular to the grain in the wet mode.In this case, 600 aging hours resulted in more than a three-fold increase in the two roughness parameters.The aging-related changes in the roughness parameters Ra and Rz were practically identical on the radial and tangential surfaces.

Spruce Wood Surface Geometry and Morphology
The four-way variance analysis processing confirmed significant impacts for all four evaluated factors (aging mode, aging time, anatomical direction, and irradiated surface), as well as their interactions on the roughness parameters discussed.
The original spruce wood surface pre-treated with milling exhibited the lowest roughness values.As the unevenness of these surfaces overpassed 500 nm, they were considered surfaces with rough unevenness [58], and this is typical for wood.In most cases, significant differences in roughness between the radial and the tangential surfaces were not identified.Occasionally, significant differences in roughness parameters were confirmed; these, however, were not important from a practical viewpoint.Significantly lower roughness values that were measured in the grain direction than perpendicular to the grain were due to the orientation of the cell wall elements.The roughness data we obtained for the milled surface were in good accordance with the results reported in [59][60][61].
The advancing aging process caused roughness to increase in both aging modes (Figure 11).This figure demonstrates that the smallest changes in roughness parameters were observed in the grain direction in the dry mode, and the highest increase in roughness parameters was found perpendicular to the grain in the wet mode.In this case, 600 aging hours resulted in more than a three-fold increase in the two roughness parameters.The aging-related changes in the roughness parameters Ra and Rz were practically identical on the radial and tangential surfaces.The solar radiation and heat simulated in the dry aging mode caused depolymerization, primarily of lignin, and multiplication of carbonyl groups, but without their washing out.Consequently, the wood morphology remained without substantial changes.On the other hand, the wet mode, comprising radiation, oxidation, and heat in interactions with water, caused a decline in the carbonyl groups because of their extraction with water.Neither hydrolyzing reactions may be excluded.The hydrolysis mainly affects the acetyl groups in hemicelluloses [25].All these phenomena resulted in more conspicuous wood surface degradation.Individual annual rings in spruce wood show characteristic differences in density between the early wood and the late wood, so the degradation primarily concerned early wood [62,63].Moli ński et al. [64] reported that the average density value for early spruce wood was about 300 kg•m −3 , while for the late wood, it was 750 kg•m −3 .This difference in density resulted in more extensive early wood erosion in the wood aging process and evident increases in roughness, measured mainly perpendicularly to the grain course.These differences in erosion extent between the early and late wood caused the forming of the so-called "plastic texture" (Figure 12).This is typical for coniferous wood species [63,65].
surface degradation.Individual annual rings in spruce wood show characteristic differences in density between the early wood and the late wood, so the degradation primarily concerned early wood [62,63].Moliński et al. [64] reported that the average density value for early spruce wood was about 300 kg•m −3 , while for the late wood, it was 750 kg•m −3 .This difference in density resulted in more extensive early wood erosion in the wood aging process and evident increases in roughness, measured mainly perpendicularly to the grain course.These differences in erosion extent between the early and late wood caused the forming of the so-called "plastic texture" (Figure 12).This is typical for coniferous wood species [63,65].In the dry mode, the wood moisture content was kept constant (4.5 ± 0.3%); accordingly, the dimensional changes were negligible.In the case of the wet mode, the average moisture content in the specimens subject to aging was significantly higher, ranging from 11% to 16%.Due to the steeper moisture gradient, the surface layers revealed restricted swelling or shrinkage.This probably enhanced the frequency occurrence of surface cracks observed in the case of the wet mode equally on the radial and the tangential surfaces (Figure 12).
Waviness was evaluated only perpendicularly to the grain.Also, for waviness, the results of the three-way variance analysis confirmed the important effects of all three studied factors (aging mode, aging time, and irradiated surface) and their interactions on the waviness parameters Wa and Wt.The waviness parameters varied with the aging time both in dry and wet modes (Figure 13).The two parameters significantly increased with prolonged aging time.Evidently, bigger changes were observed for the wet mode.While in the case of dry mode, the Wa values, after accomplishing the aging process, were found to be 2.5 to 3 times bigger than the original ones, in the case of wet mode, the corresponding increase was from 7.5-to 10-fold.Similar qualitative and quantitative changes during the aging process were also observed for the parameter Wt.Higher waviness was observed on radial surfaces.In the dry mode, the wood moisture content was kept constant (4.5 ± 0.3%); accordingly, the dimensional changes were negligible.In the case of the wet mode, the average moisture content in the specimens subject to aging was significantly higher, ranging from 11% to 16%.Due to the steeper moisture gradient, the surface layers revealed restricted swelling or shrinkage.This probably enhanced the frequency occurrence of surface cracks observed in the case of the wet mode equally on the radial and the tangential surfaces (Figure 12).
Waviness was evaluated only perpendicularly to the grain.Also, for waviness, the results of the three-way variance analysis confirmed the important effects of all three studied factors (aging mode, aging time, and irradiated surface) and their interactions on the waviness parameters Wa and Wt.The waviness parameters varied with the aging time both in dry and wet modes (Figure 13).The two parameters significantly increased with prolonged aging time.Evidently, bigger changes were observed for the wet mode.While in the case of dry mode, the Wa values, after accomplishing the aging process, were found to be 2.5 to 3 times bigger than the original ones, in the case of wet mode, the corresponding increase was from 7.5-to 10-fold.Similar qualitative and quantitative changes during the aging process were also observed for the parameter Wt.Higher waviness was observed on radial surfaces.The changes in the chemistry and morphology of the spruce wood surface during the aging process were also reflected through the changes in this microscopic wood structure.Before the aging, the cell elements in the milled spruce wood surfaces created a compact structure.The tracheids were cross-linked by means of undamaged middle lamellae.The inner surfaces of cross-cut tracheids exhibited intact bordered pits pair (Figure 14a,b).In the dry mode, after 600 aging hours, surficial cracks were observed; the bonds between the cell elements, however, were mostly preserved sufficiently compact (Figure 14c-f).The inner surfaces of early tracheids exhibited cracks in cell walls and in neighborhoods of bordered pits.These were, to a certain extent, distorted and lacking toruses.The character of cell walls destruction indicates their fragility aggravated due to radiation effects.On the other hand, the surfaces of intact, non-cut late wood tracheids manifested the oc- The changes in the chemistry and morphology of the spruce wood surface during the aging process were also reflected through the changes in this microscopic wood structure.Before the aging, the cell elements in the milled spruce wood surfaces created a compact structure.The tracheids were cross-linked by means of undamaged middle lamellae.The inner surfaces of cross-cut tracheids exhibited intact bordered pits pair (Figure 14a,b).In the dry mode, after 600 aging hours, surficial cracks were observed; the bonds between the cell elements, however, were mostly preserved sufficiently compact (Figure 14c-f).The inner surfaces of early tracheids exhibited cracks in cell walls and in neighborhoods of bordered pits.These were, to a certain extent, distorted and lacking toruses.The character of cell walls destruction indicates their fragility aggravated due to radiation effects.On the other hand, the surfaces of intact, non-cut late wood tracheids manifested the occurrence of melt mass, identified, based on our chemical analyses, as lignin and lignin derivatives (Figure 14e,f).A similar phenomenon was observed by Ko et al. [66] during beech wood pre-treatment with hot water.
More conspicuous changes in the structure were observed after having accomplished the aging in the wet mode (Figure 15).After 600 aging hours, the surfaces exhibited more abundant cracks (Figure 15a,b).From many spots, the compound middle lamellae (Figure 15d), consisting mainly of lignin, were eroded.The removal of the middle lamellae considerably debilitated the connections between the tracheids.The supplied liquid water detached the tracheids from the surface step by step, and the surface roughness was progressively pronounced.The bordered pits were degraded, too.The pores were extended, and the toruses were eliminated totally or partially as the result of margin degradation (Figure 15c).The surfaces did not exhibit melted lignin because, unlike in the dry mode, this substance was also washed away with water.A similar form of surface degradation has been documented for several tropical woods subject to natural and accelerated aging [67].More conspicuous changes in the structure were observed after having accomplis the aging in the wet mode (Figure 15).After 600 aging hours, the surfaces exhibited m abundant cracks (Figure 15a,b).From many spots, the compound middle lamellae (Fi 15d), consisting mainly of lignin, were eroded.The removal of the middle lamellae siderably debilitated the connections between the tracheids.The supplied liquid wate tached the tracheids from the surface step by step, and the surface roughness was prog sively pronounced.The bordered pits were degraded, too.The pores were extended, the toruses were eliminated totally or partially as the result of margin degradation (Fi 15c).The surfaces did not exhibit melted lignin because, unlike in the dry mode, this stance was also washed away with water.A similar form of surface degradation has documented for several tropical woods subject to natural and accelerated aging [67].

Conclusions
The analysis of the experimental results has allowed us to draw the followin clusions.

•
Sunlight exposure was a major factor in spruce wood discoloration.As early a 100 irradiation hours, the total color difference values were ∆E* > 12, thus repr

Figure 1 .
Figure 1.Prepared test specimens of spruce wood.

Figure 1 .
Figure 1.Prepared test specimens of spruce wood.

Figure 2 .
Figure 2. Differences in color coordinates and the total color difference during accelerated aging (dry mode).

Figure 2 .
Figure 2. Differences in color coordinates and the total color difference during accelerated aging (dry mode).

Figure 3 .
Figure 3. Color variation in spruce wood specimens during accelerated aging (dry mode).

Figure 4 .
Figure 4. Differences in color coordinates and the total color difference during accelerated aging (wet mode).

Figure 3 .
Figure 3. Color variation in spruce wood specimens during accelerated aging (dry mode).

19 Figure 3 .
Figure 3. Color variation in spruce wood specimens during accelerated aging (dry mode).

Figure 4 .
Figure 4. Differences in color coordinates and the total color difference during accelerated aging (wet mode).Figure 4. Differences in color coordinates and the total color difference during accelerated aging (wet mode).

Figure 4 .
Figure 4. Differences in color coordinates and the total color difference during accelerated aging (wet mode).Figure 4. Differences in color coordinates and the total color difference during accelerated aging (wet mode).

Figure 5 .
Figure 5. Color variation in spruce wood specimens during accelerated aging (wet mode).

Figure 5 .
Figure 5. Color variation in spruce wood specimens during accelerated aging (wet mode).

Figure 5 .
Figure 5. Color variation in spruce wood specimens during accelerated aging (wet mode).

Figure 10 .
Figure 10.The dependence between lignin content and color changes.(A) dry mode, (B) wet mode.

Figure 11 .
Figure 11.Parameters Ra and Rz varying during the accelerated aging process.radial surface, tangential surface.

Figure 11 .
Figure 11.Parameters R a and R z varying during the accelerated aging process.

Figure 11 .
Figure 11.Parameters Ra and Rz varying during the accelerated aging process.radial surface, tangential surface.
Changes in HBI values indicate the reorganization of hydrogen bonds or changes inbound water in cellulose.

Figure 11 .
Figure 11.Parameters Ra and Rz varying during the accelerated aging process.radial surface, tangential surface.tangential surface.

Figure 12 .
Figure 12.Morphology of radial spruce wood surface after 600 accelerated aging hours (wet mode).

Figure 12 .
Figure 12.Morphology of radial spruce wood surface after 600 accelerated aging hours (wet mode).

Figure 13 .
Figure 13.Parameters Wa and Wt varying during accelerated aging process (perpendicular to the grain).

Figure 13 .
Figure 13.Parameters Wa and Wt varying during accelerated aging process (perpendicular to the grain).

Figure 14 .
Figure 14.Structure of milled radial surfaces of spruce wood.(a,b) Before aging.(c-f) Afte aging hours-dry mode.

Figure 15 .
Figure 15.Structure of milled radial surfaces of spruce wood (a-d) after 600 aging hours-wet mode.

Table 1 .
The aging parameters set according to the Standard ASTM G 155 "dry mode".

Table 2 .
The aging parameters set according to the Standard ASTM G 155 "wet mode".

Table 3 .
Basic statistical characteristics of color coordinate values L*, a*, and b* for spruce wood at different phases of accelerated aging (number of measurements for each sample set n = 75).