Surface Changes of Selected Hardwoods Due to Weather Conditions

The study is focused on the surface changes of five hardwoods (oak, black locust, poplar, alder and maple) that were exposed to natural weathering for 24 months in the climatic conditions of Central Europe. Colour, roughness, visual and chemical changes of exposed surface structures were examined. The lowest total colour changes (∆E*) were found for oak (23.77), the highest being recorded for maple (34.19). Roughness differences after 24-month exposure (∆Ra) showed minimal changes in poplar wood (9.41); the highest changes in roughness were found on the surface of alder (22.18). The presence of mould and blue stains was found on the surface of maple, alder and poplar. Chemical changes were characterized by lignin and hemicelluloses degradation. Decreases of both methoxy and carbonyl groups, cleavage of bonds in lignin and hemicelluloses, oxidation reaction and formation of new chromophores were observed. In the initial phases of the degradation process, the discoloration was related to chemical changes; in the longer period, the greying due to settling of dust particles and action of mould influenced the wood colour. The data were confirmed by confocal laser scanning microscopy. The obtained results revealed degradation processes of tested hardwood surfaces exposed to external environmental factors.


Introduction
One of the major contributions to securing sustainable development is the use of renewable natural materials, which undoubtedly include wood.In an effort to reduce the ecological burden, the surface of wooden structures is often left untreated.In addition to the traditional interior design elements, the use of non-treated wood is expanding even further into the exterior.Contemporary trends even prefer untreated wood that turns grey after exposure to weather under aboveground conditions over non-durable wood with applied surface coatings [1], since wood with higher natural durability [2] allows outdoor utilization without any deterioration risk [3].However, wood is susceptible to environmental degradation similar to other natural materials [4]-it changes surface properties due to weather and can even be attacked by biotic agents if the basic design principles are not respected.Weathering is defined as the slow decomposition of materials subjected to weather factors [5], causing mostly unwanted and premature product failures [6].The main weathering factors are solar radiation and water (moisture) acting synergistically with temperature, wind, airborne pollutants, biological agents, and acid rain, etc. [6][7][8].
The main organic components in wood, lignin, hemicelluloses and cellulose, react to weathering.Lignin is the wood component most susceptible to photodegradation caused mainly by ultraviolet (UV) Forests 2018, 9, 557 2 of 15 light [9][10][11].This aromatic biopolymer strongly absorbs UV radiation [12].The initial decrease in lignin content is accompanied by generation of carbonyl groups, whereas degradation of cellulose is indicated by a loss of weight and reduction in the degree of polymerization.Extractives, such as terpenes, terpenoids, phenols, lignans, tannins, flavonoids, etc. [10,13], responsible for wood colour, odour and natural durability against biodegradation [4], are also affected by weathering.Photodegraded products are leached out of the wood surface by water [1].The changes caused by weathering develop rapidly and cannot be fully avoided, as presented in previous work focused on weathering of untreated wood [14,15].The rate of degradation is usually related to wood species [12], intensity of light and light wavelengths [16], period of irradiation [12,17,18] and climatic factors occurring during exposure [19].
Weathering affects only the surface layers of wood [20].Atmospheric degradation is then manifested by the change of colour reflecting the chemical changes [21,22], followed by the formation of cracks and increased roughness of the samples [23,24].Untreated wood specimens exhibit higher colour changes in a shorter time than treated specimens [25].Some wood species turn yellow or brown; eventually, they turn grey due to growth of fungi and moulds [26] and by the dust particles in air which penetrate the porous structure of wood [19].These changes occur to a depth of only 0.05-2.5 mm [11,23].The depth has a relatively high correlation with the total colour difference value [15].The next parameter used to analyse the weathering effects is surface roughness.During natural weathering, increasing roughness has been reported [15,[27][28][29].The increase is caused by unequal erosion of the surface in the late wood and early wood with thin walled cells and lower density [30].As the result, the early wood is degraded faster [31], mainly in the case of softwoods with lower density [15].The wood also reacts to moisture and water from ambient atmosphere [32].Its surface wettability significantly increases during weathering [27].Some species reached full wettability after 1 year of exposure [33].Chemical changes in wood during weathering are often investigated by Fourier Transform Infrared Spectroscopy (FTIR) analysis [10,14,34,35].According to Pandey [35], the total colour difference was related to the lignin degradation and increase of carbonyl groups.Hardwoods underwent a faster degradation than softwoods because the syringyl structure in hardwoods degraded faster than guaiacyl structures in softwoods.Liu et al. [36] and Reinprecht et al. [37] studied surface changes of 10 tropical woods during weathering.According to the results, colour changes might be caused by leaching of extractives, oxidation and degradation of lignin; however, the degree of change was different for every wood species.In the study of Gupta et al. [14], FTIR was used for prediction of wooden cladding service life.There are efforts to minimize the use of harmful chemicals [38]; therefore, it is important to understand the degradation processes in untreated wood during weathering.The service life determination of wooden components is a crucial research constraint that needs a thorough investigation of material properties before the prediction is done [14,39].
The objective of this paper is to characterize hardwood' overall performance under specific exterior conditions of Central Europe (surface, visual and chemical parameters) using different scientific methods, considering the increasing application of untreated wood outdoors.The results can lead to deepening knowledge in this area, to comparison of naturally durable and less durable species and to proper and more frequent use of untreated hardwoods in applications directly exposed to weathering.During weathering, we assume that gradual lignin decomposition and changes in the hemicellulose complex are manifested by pronounced colour changes and increased surface roughness of the wood surface.These findings are supported by microscopic analysis.This study follows and expands the findings of the previously published study of Oberhofnerová et al. [15], which presents the results of colour and roughness changes after one year of natural weathering.

Material Preparation
The experiment was conducted using three samples for each weathering period, all of them with dimensions 375 × 78 × 20 mm (Longitudinal × Tangential × Radial), from five hardwood species Forests 2018, 9, 557 3 of 15 (black locust, oak, maple, alder and poplar wood).A total of 75 samples were used for the experiment.Only the heartwood was tested.Before exposure, the test samples were sanded with sandpaper grit of 120 (P120).The specification of used wood species is given in Table 1.
Table 1.Specification of used hardwoods.Natural durability against fungi is given in the range of 1-5, where 1 signifies the best durability.Wood density was determined at 12% wood moisture according to ČSN 49 0108 [40].

Natural Weathering
The natural weathering test was performed at Suchdol, Prague, Czech Republic (50 • 07 49.68" N, 14 • 22 13.87" E, 285 m altitude) and lasted 24 months (from 15 December 2014 to 20 December 2016 for four periods-3, 6, 12 and 24 months).The overview of weather conditions is given in Table 2.The samples were exposed outdoors, at 45 • inclination angle, facing south and placed approximately 1 m above the ground according to [41].The specimens were stabilized at 20 ± 2 • C and 65% relative humidity (RH) before the measurements for 7 days.The colour, roughness, FTIR, and confocal microscopy were measured first on the surface before exposure (measurements were repeated after each weathering period).

Colour Measurement
The colour of the wood surface was measured using a spectrophotometer CM-600d (Konica Minolta, Osaka, Japan).A standard illuminant D65, viewing angle of 10 • and SCI (Specular Component Included), was used.The colour values were measured at 8 given positions of each sample (Figure 1) and expressed in the Lab system [43].This colorimetric system comprises three independent axes perpendicular to each other.The L* axis determines the lightness (0-black, 100-white), the a* axis determines the ratio between red (positive) and green (negative) and finally the b* axis represents the ratio between yellow (positive) and blue (negative).Total colour change ∆E* was calculated using the Equation (1) [44]: where ∆L*, ∆a*, ∆b* are differences in individual axes (difference between the value measured after and before exposure).

Roughness Measurement
Surface roughness was assessed with a stylus profilometer Talysurf Form Intra (Taylor-Hobson, Leicester, UK).The measurements were made according to standards [45,46] on the surface of each sample at the same positions, Figure 1.The Ra (arithmetical mean profile deviation) was measured in the perpendicular direction to the fibres.These measurements were considered representative surface-roughness indicators.The tracing length was 15 mm and cut-off lengths were 2.5 mm.

ATR-FTIR Analysis
FTIR spectra of weathered wood surfaces 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 attached to a diamond crystal.The spectra were acquired by accumulating 64 scans at a spectral resolution of 4 cm −1 in an absorbance mode from 4000 to 400 cm −1 standardised using the baseline method and analysed using OMNIC 8.0 software (Thermo Fisher Scientific, Waltham, MA, USA).Measurements were performed on four replicates per sample.The spectral range 1800-800 cm −1 was evaluated.Each specimen was measured on the tangential face.

Confocal Laser Scanning Microscopy
To evaluate the surface degradation of hardwoods, microscopic structural changes of wood surfaces were studied using a confocal laser scanning microscope Lext Ols 4100 (Olympus, Tokyo, Japan) before and after each exposure of samples to natural weathering.

Statistical Evaluation
The statistical evaluation was done in Statistica 12 software (Statsoft, Palo Alto, CA, USA) and MS Excel 2013 (Microsoft, Redmond, WA, USA) using mean values, standard deviations, analysis of variance (ANOVA), and Tukey´s N HSD multiple comparison test at α = 0.05 significance level.

Results
The effect of wood species and period of exposure and their interactions on the measured properties (colour parameters and roughness) after weathering were evaluated as statistically significant (p < 0.05)-see Table 3.For further evaluation, Tukey's HSD test was used to compare specific differences between wood species.

Colour Changes
The colour parameters (L*, a*, b*) expressed as colour differences (∆L*, ∆a*, ∆b*) measured on the surface of investigated hardwoods (Tables 4-6) and corresponding colour differences (∆E*) are presented (Table 7).Total colour difference ΔE* increased over the duration of external exposure for

Roughness Measurement
Surface roughness was assessed with a stylus profilometer Talysurf Form Intra (Taylor-Hobson, Leicester, UK).The measurements were made according to standards [45,46] on the surface of each sample at the same positions, Figure 1.The R a (arithmetical mean profile deviation) was measured in the perpendicular direction to the fibres.These measurements were considered representative surface-roughness indicators.The tracing length was 15 mm and cut-off lengths were 2.5 mm.

ATR-FTIR Analysis
FTIR spectra of weathered wood surfaces 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 attached to a diamond crystal.The spectra were acquired by accumulating 64 scans at a spectral resolution of 4 cm −1 in an absorbance mode from 4000 to 400 cm −1 standardised using the baseline method and analysed using OMNIC 8.0 software (Thermo Fisher Scientific, Waltham, MA, USA).Measurements were performed on four replicates per sample.The spectral range 1800-800 cm −1 was evaluated.Each specimen was measured on the tangential face.

Confocal Laser Scanning Microscopy
To evaluate the surface degradation of hardwoods, microscopic structural changes of wood surfaces were studied using a confocal laser scanning microscope Lext Ols 4100 (Olympus, Tokyo, Japan) before and after each exposure of samples to natural weathering.

Statistical Evaluation
The statistical evaluation was done in Statistica 12 software (Statsoft, Palo Alto, CA, USA) and MS Excel 2013 (Microsoft, Redmond, WA, USA) using mean values, standard deviations, analysis of variance (ANOVA), and Tukey´s N HSD multiple comparison test at α = 0.05 significance level.

Results
The effect of wood species and period of exposure and their interactions on the measured properties (colour parameters and roughness) after weathering were evaluated as statistically significant (p < 0.05)-see Table 3.For further evaluation, Tukey's HSD test was used to compare specific differences between wood species.

Colour Changes
The colour parameters (L*, a*, b*) expressed as colour differences (∆L*, ∆a*, ∆b*) measured on the surface of investigated hardwoods (Tables 4-6) and corresponding colour differences (∆E*) are presented (Table 7).Total colour difference ∆E* increased over the duration of external exposure for all the hardwoods [25,28,47].The highest colour difference after two years of weathering was observed for maple (∆E* = 34.2) and the lowest for oak wood (∆E* = 23.8).

Roughness Changes
The average roughness values (R a ) of unweathered (control) and weathered samples expressed as roughness differences after and before each exposure period (∆R a ) are showed in Table 8.The following figures (Figures 2-6) show the 3D microstructure of the surface of all examined wood samples after 2 years of weathering obtained by confocal laser scanning microscopy.The degree of surface erosion, determining the wood roughness change, is represented by a colour range ranging from purple to red.

Visual Evaluation
Visual evaluation of samples was performed using confocal laser scanning microscopy of the wood sample surface before and during exposure (Figure 7).The detailed analysis of surface changes during weathering of wood species was performed by confocal laser microscope (Figures 8-12).

Visual Evaluation
Visual evaluation of samples was performed using confocal laser scanning microscopy of the wood sample surface before and during exposure (Figure 7).The detailed analysis of surface changes during weathering of wood species was performed by confocal laser microscope (Figures 8-12).

Visual Evaluation
Visual evaluation of samples was performed using confocal laser scanning microscopy of the wood sample surface before and during exposure (Figure 7).The detailed analysis of surface changes during weathering of wood species was performed by confocal laser microscope (Figures 8-12).

Changes in the FTIR Spectra
Chemical changes of wood surface components were evaluated by ATR-FTIR analysis.They are expressed as the differential FTIR spectra (difference in absorbance after and before of each exposure period, denoted as AD3 (Absorbances Difference after 3 month of weathering), AD6-after 6 months and so on, Figures 13-17).The positive values refer to the increase, negative values to the decrease in the absorbance.The absorbance of bands at 1730 cm −1 (C=O vibration of non-conjugated carbonyl groups in hemicelluloses), 1600 cm −1 and 1505 cm −1 (C=C skeletal vibrations in aromatic structure of lignin), 1460 cm −1 (C−H bending in CH 3 group in lignin), 1235 cm −1 (C−O stretching in lignin and xylan) and 1267 cm −1 (C−O stretching in lignin structure) decreased during natural weathering.These results indicate that the structure of the lignin polymer and hemicelluloses was degraded to a significant extent.

Changes in the FTIR Spectra
Chemical changes of wood surface components were evaluated by ATR-FTIR analysis.They are expressed as the differential FTIR spectra (difference in absorbance after and before of each exposure period, denoted as AD3 (Absorbances Difference after 3 month of weathering), AD6-after 6 months and so on, Figures 13-17).The positive values refer to the increase, negative values to the decrease in the absorbance.The absorbance of bands at 1730 cm −1 (C=O vibration of non-conjugated carbonyl groups in hemicelluloses), 1600 cm −1 and 1505 cm −1 (C=C skeletal vibrations in aromatic structure of lignin), 1460 cm −1 (C−H bending in CH3 group in lignin), 1235 cm −1 (C−O stretching in lignin and xylan) and 1267 cm −1 (C−O stretching in lignin structure) decreased during natural weathering.These results indicate that the structure of the lignin polymer and hemicelluloses was degraded to a significant extent.

Changes in the FTIR Spectra
Chemical changes of wood surface components were evaluated by ATR-FTIR analysis.They are expressed as the differential FTIR spectra (difference in absorbance after and before of each exposure period, denoted as AD3 (Absorbances Difference after 3 month of weathering), AD6-after 6 months and so on, Figures 13-17).The positive values refer to the increase, negative values to the decrease in the absorbance.The absorbance of bands at 1730 cm −1 (C=O vibration of non-conjugated carbonyl groups in hemicelluloses), 1600 cm −1 and 1505 cm −1 (C=C skeletal vibrations in aromatic structure of lignin), 1460 cm −1 (C−H bending in CH3 group in lignin), 1235 cm −1 (C−O stretching in lignin and xylan) and 1267 cm −1 (C−O stretching in lignin structure) decreased during natural weathering.These results indicate that the structure of the lignin polymer and hemicelluloses was degraded to a significant extent.

Discussion
During the weathering process, colour parameter changes showed practically identical trends for all analysed wood samples.A decreased L* (Table 4, negative values of ∆L*) value was observed, similar to Turkoglu et al. [25].Darkening of the wood surface was more pronounced after twelve months of natural weathering.The value a* gradually decreased (negative trend in ∆a*), mainly after 12 months.The lowest values were found for oak and alder (Table 5).The decrease in the chromaticity coordinate b* (negative trend in ∆b*) was similar (Table 6).A sharp drop in the b* value (increase in b* difference) was observed in the first half of the 24-month cycle, with the remainder of the exposure declining only slightly.The reduction is caused by leaching of decomposed lignin and extractives from the surface [25,48].At the end of the two-year exposure, values a* and b* stabilized at a level which correspond to the grey colour.Undoubtedly, untreated wood exhibits higher colour changes than treated wood during natural weathering [1,25,27].The highest changes in colour parameters were observed in initial phases of weathering as in the studies of Lesar et al. [1] and Sharrat et al. [49].The more pronounced changes were noted during the first 12 months; then, the other changes were negligible due to the formation of the grey degraded layer which acts as a protective barrier and slows the weathering process [50].The greying caused by the action of weather and dust settling into the porous and degraded surface can be visibly observed after 6 months, especially for less durable

Discussion
During the weathering process, colour parameter changes showed practically identical trends for all analysed wood samples.A decreased L* (Table 4, negative values of ∆L*) value was observed, similar to Turkoglu et al. [25].Darkening of the wood surface was more pronounced after twelve months of natural weathering.The value a* gradually decreased (negative trend in ∆a*), mainly after 12 months.The lowest values were found for oak and alder (Table 5).The decrease in the chromaticity coordinate b* (negative trend in ∆b*) was similar (Table 6).A sharp drop in the b* value (increase in b* difference) was observed in the first half of the 24-month cycle, with the remainder of the exposure declining only slightly.The reduction is caused by leaching of decomposed lignin and extractives from the surface [25,48].At the end of the two-year exposure, values a* and b* stabilized at a level which correspond to the grey colour.Undoubtedly, untreated wood exhibits higher colour changes than treated wood during natural weathering [1,25,27].The highest changes in colour parameters were observed in initial phases of weathering as in the studies of Lesar et al. [1] and Sharrat et al. [49].The more pronounced changes were noted during the first 12 months; then, the other changes were negligible due to the formation of the grey degraded layer which acts as a protective barrier and slows the weathering process [50].The greying caused by the action of weather and dust settling into the porous and degraded surface can be visibly observed after 6 months, especially for less durable

Discussion
During the weathering process, colour parameter changes showed practically identical trends for all analysed wood samples.A decreased L* (Table 4, negative values of ∆L*) value was observed, similar to Turkoglu et al. [25].Darkening of the wood surface was more pronounced after twelve months of natural weathering.The value a* gradually decreased (negative trend in ∆a*), mainly after 12 months.The lowest values were found for oak and alder (Table 5).The decrease in the chromaticity coordinate b* (negative trend in ∆b*) was similar (Table 6).A sharp drop in the b* value (increase in b* difference) was observed in the first half of the 24-month cycle, with the remainder of the exposure declining only slightly.The reduction is caused by leaching of decomposed lignin and extractives from the surface [25,48].At the end of the two-year exposure, values a* and b* stabilized at a level which correspond to the grey colour.Undoubtedly, untreated wood exhibits higher colour changes than treated wood during natural weathering [1,25,27].The highest changes in colour parameters were observed in initial phases of weathering as in the studies of Lesar et al. [1] and Sharrat et al. [49].The more pronounced changes were noted during the first 12 months; then, the other changes were negligible due to the formation of the grey degraded layer which acts as a protective barrier and slows the weathering process [50].The greying caused by the action of weather and dust settling into the porous and degraded surface can be visibly observed after 6 months, especially for less durable wood species such as maple and alder (Figure 7).After 12 months of weathering, the greying was obvious for all the tested hardwoods.The tearing of wood fibers was observed after 24 months of weathering (Figures 8-12).The presence of slight fungal growth was observed after a few months of weathering on poplar, alder and maple (Figures 10-12) as in the study of Mohebby and Saei [27].
As shown by the measured values, the surface roughness value of native wood samples increased with weathering time (Table 8 and Figures 2-6), as presented in other studies [27,28].However, the surface roughness of some wood samples (black locust, oak and alder) partly decreased with weathering between the 6th and 12th month period.Similar trends have also been observed by Turkoglu et al. [51].Increasing roughness is caused by lignin decomposition [26].Another effect can be the absorption of the thermal component of solar radiation.Turkulin et al. [52] reported that this radiation mostly degraded the middle lamella, which is between two cell walls and holds the cells together.This degradation increases the roughness of the wood surface [31].Kerber et al. [53] also reported that in addition to the leaching of lignin degraded by natural weathering reactions, the increase in the roughness of the wood is also related to sudden changes in humidity (absorption and desorption of the humidity) causing the presence of superficial cracks.Also, rain water helps to remove loosened fibres and particles produced during irradiation and to move leached lignin fragments to the top layer that is mainly composed of cellulose (which causes the grey colour of the wood and a rough texture [12]).Elevated temperature [54], high moisture content [52] and diluted acid [55] can increase the photodegradation rate.Additional factors contributing to weathering are superficial wetting and drying, generating surface stresses that can cause checking [7,56].The changes in surface structure after 2 years of weathering were apparent in laser scanning microscopy as well (Figures 2-6), where tearing of the wood fibres (Figures 8-12) and surface erosion, more obvious in less dense earlywood than in latewood [30], caused by abiotic factors, were observed (Figures 8-12).
The dominant factor responsible for wood degradation is lignin, which absorbs ultraviolet and visible radiation due to its chromophoric groups.However, the effect of visible light to surface degradation at later stages of weathering was observed [57][58][59].According to Norrstrom [60], lignin is responsible for absorbing 80% to 95% of the total UV light absorbed by wood, carbohydrates 5 to 20% and extractives about 2%.FTIR spectra of weathered wood (Figures 13-17) showed that the absorbance at 1730 cm −1 (C=O vibration of the non-conjugated carbonyl groups, stretching in xylan) decreased at higher exposure times.This is in accordance with findings of natural and artificial weathering of seven tropical woods [37].The decreasing trend of this band was probably caused by the condensation reactions of lignin as well as by hemicellulose deacetylation [61].Hemicelluloses are the least resistant to thermal treatment, and carboxylic acids (predominantly acetic acid) are formed by their decomposition.Carboxylic acids cause the depolymerization of cellulose and cleavage of bonds in lignin [62,63].
The band occurring at 1600 cm −1 is characteristic of aromatic compounds and is attributed to aromatic skeleton vibrations.In our case, a sustained decrease in absorbance was recorded.The absorbance at 1505 cm −1 (and C=C skeletal stretching vibrations in aromatic rings in lignin) showed a slight decrease.The samples did not show remarkable changes in this regard.A similar effect in the intensity was also observed after UV treatment of wood [8,64].Decreased absorption at that band is interpreted as lignin decay combined with the formation of new carbonyl groups, evidencing photo-induced oxidation of the wood surface.During the weathering test, the decrease of the band at 1462 cm −1 (C−H asymmetric bending in CH 3 group in lignin) revealed the loss of lignin [65].The peaks at 1235 cm −1 (C−O stretching in lignin and xylan) and 1267 cm −1 (C−O stretching and breathing of guaiacyl ring) are assigned to lignin [66,67].Described phenomena may indicate changes in lignin structures leading to a decrease of methoxy groups in lignin.
The degradation mechanism is complex with different paths leading to water soluble products and finally to chromophoric groups like carboxylic acids, quinines or hydroperoxides [68].These water-soluble compounds can be extracted from wood by rain during weathering processes, resulting in the decrease of carbonyl groups and the increase in surface roughness.

Conclusions
The study revealed the overall degradation process of hardwoods caused by abiotic factors in the exterior using different experimental methods.Five hardwoods from the temperate climate zone with different natural durability (oak, black locust, poplar, alder and maple wood) were exposed to natural weathering for 24 months.Significant colour changes were noticeable after 3 months of weathering.During the first year of weathering, all the hardwoods were characterized by high colour changes.The decrease in lightness L* indicated gradual darkening of the samples, decreasing a* and b* values showed a more pronounced shift to the grey colour.During the second year of weathering, the colour changed only slightly.The highest colour difference after 2 years of weathering was recorded for maple wood, the lowest for oak.The roughness change R a had a similarly increasing trend (with the exception of poplar), but as opposed to the colour changes, the largest increase was observed during the second year of weathering, specifically for alder.ATR-FTIR analysis confirmed the assumption of degradation, in particular, in the structure of lignin and hemicelluloses of hardwoods.Decreases of both methoxy and carbonyl groups, cleavage of bonds in lignin and hemicelluloses, oxidation reaction and formation of new chromophores were observed during weathering.Based on the presented results, the use of more durable wood species (oak, black locust) in the exterior can be recommended (the lowest discolouration, relatively low roughness changes, no formation of moulds or fungi).Alder was also characterized by relatively good values of colour changes, but mould and degraded wood fibers were formed on its surface.
The significant impact of weathering factors on the quality and wood colour and roughness was demonstrated.The obtained findings regarding surface parameters (colour, roughness) were supported by visual performance, microscopic and chemical analysis.Presented results give some useful information about the surface degradation and related chemical changes of different hardwoods during exposure to weathering.The obtained results can increase the possible use of untreated hardwoods in applications directly exposed to weathering.

Figure 1 .
Figure 1.Points on the samples for measuring colour (dots) and roughness (lines).

Figure 1 .
Figure 1.Points on the samples for measuring colour (dots) and roughness (lines).

Forests 2018, 9 ,
x FOR PEER REVIEW 6 of 15 degree of surface erosion, determining the wood roughness change, is represented by a colour range ranging from purple to red.

Figure 3 .
Figure 3. 3D image of black locust wood surface before (left) and after 2 years of weathering (right)confocal laser scanning microscopy (area of 800 × 800 μm).

Forests 2018, 9 ,
x FOR PEER REVIEW 6 of 15 degree of surface erosion, determining the wood roughness change, is represented by a colour range ranging from purple to red.

Figure 3 .
Figure 3. 3D image of black locust wood surface before (left) and after 2 years of weathering (right)confocal laser scanning microscopy (area of 800 × 800 μm).

Forests 2018, 9 ,
x FOR PEER REVIEW 6 of 15 degree of surface erosion, determining the wood roughness change, is represented by a colour range ranging from purple to red.

Figure 3 .
Figure 3. 3D image of black locust wood surface before (left) and after 2 years of weathering (right)confocal laser scanning microscopy (area of 800 × 800 μm).

Figure 7 .
Figure 7. Confocal microscopy of samples during exposure to natural weathering (each scan corresponds to an area of 2.5 × 2.5 mm).

Figure 6 .
Figure 6.3D image of maple wood surface before (left) and after 2 years of weathering (right)-confocal laser scanning microscopy (area of 800 × 800 µm).

Figure 7 .
Figure 7. Confocal microscopy of samples during exposure to natural weathering (each scan corresponds to an area of 2.5 × 2.5 mm).Figure 7. Confocal microscopy of samples during exposure to natural weathering (each scan corresponds to an area of 2.5 × 2.5 mm).

Figure 7 .
Figure 7. Confocal microscopy of samples during exposure to natural weathering (each scan corresponds to an area of 2.5 × 2.5 mm).Figure 7. Confocal microscopy of samples during exposure to natural weathering (each scan corresponds to an area of 2.5 × 2.5 mm).

Figure 7 .
Figure 7. Confocal microscopy of samples during exposure to natural weathering (each scan corresponds to an area of 2.5 × 2.5 mm).

Figure 8 .
Figure 8. Confocal laser scanning microscopy of oak wood before and after 2 years of weathering (detail).

Figure 8 .
Figure 8.Confocal laser scanning microscopy of oak wood before and after 2 years of weathering (detail).

Figure 10 .
Figure 10.Confocal laser scanning microscopy of poplar wood before and after 2 years of weathering (detail).

Figure 11 .
Figure 11.Confocal laser scanning microscopy of alder wood before and after 2 years of weathering (detail).

Figure 9 .
Figure 9. Confocal laser scanning microscopy of black locust wood before and after 2 years of weathering (detail).

Figure 10 .
Figure 10.Confocal laser scanning microscopy of poplar wood before and after 2 years of weathering (detail).

Figure 11 .
Figure 11.Confocal laser scanning microscopy of alder wood before and after 2 years of weathering (detail).

Figure 10 .
Figure 10.Confocal laser scanning microscopy of poplar wood before and after 2 years of weathering (detail).

Figure 10 .
Figure 10.Confocal laser scanning microscopy of poplar wood before and after 2 years of weathering (detail).

Figure 11 .
Figure 11.Confocal laser scanning microscopy of alder wood before and after 2 years of weathering (detail).

Figure 11 .
Figure 11.Confocal laser scanning microscopy of alder wood before and after 2 years of weathering (detail).

Figure 12 .
Figure 12.Confocal laser scanning microscopy of maple wood before and after 2 years of weathering (detail).

Figure 12 .
Figure 12.Confocal laser scanning microscopy of maple wood before and after 2 years of weathering (detail).

Figure 14 .
Figure 14.Differential FTIR spectra of black locust samples.

Figure 14 .
Figure 14.Differential FTIR spectra of black locust samples.

Figure 14 .
Figure 14.Differential FTIR spectra of black locust samples.

Figure 14 .
Figure 14.Differential FTIR spectra of black locust samples.

Figure 15 .
Figure 15.Differential FTIR spectra of poplar wood samples.Figure 15.Differential FTIR spectra of poplar wood samples.

Figure 15 .
Figure 15.Differential FTIR spectra of poplar wood samples.Figure 15.Differential FTIR spectra of poplar wood samples.

Figure 17 .
Figure 17.Differential FTIR spectra of maple wood samples.

Figure 17 .
Figure 17.Differential FTIR spectra of maple wood samples.

Figure 17 .
Figure 17.Differential FTIR spectra of maple wood samples.

Table 1 .
[40]ification of used hardwoods.Natural durability against fungi is given in the range of 1-5, where 1 signifies the best durability.Wood density was determined at 12% wood moisture according to ČSN 49 0108[40].

Table 1 .
[40]ification of used hardwoods.Natural durability against fungi is given in the range of 1-5, where 1 signifies the best durability.Wood density was determined at 12% wood moisture according to ČSN 49 0108[40].

Table 1 .
[40]ification of used hardwoods.Natural durability against fungi is given in the range of 1-5, where 1 signifies the best durability.Wood density was determined at 12% wood moisture according to ČSN 49 0108[40].

Table 1 .
[40]ification of used hardwoods.Natural durability against fungi is given in the range of 1-5, where 1 signifies the best durability.Wood density was determined at 12% wood moisture according to ČSN 49 0108[40].

Table 2 .
[42]her conditions during the natural weathering test.Values are based on data from a weather station[42].

Table 4 .
Exposure time-dependence of L* value (lightness differences ∆L* are in brackets).The data represent mean values.

Table 5 .
Exposure time-dependence of a* value (chromaticity differences ∆a* are in brackets).The data represent mean values.

Table 6 .
Exposure time-dependence of b* value (chromaticity differences ∆b* are in brackets).The data represent mean values.

Table 7 .
Exposure time-dependence of ∆E* value.The data represent mean values.

Table 8 .
Exposure time-dependence of arithmetical mean profile deviation R a (roughness differences ∆R a are in parentheses).