Effect of CO₂ laser processing parameters and grain direction on color change and surface roughness of pine and beech wood

Karamanoglu, M. (2026). "Effect of CO₂ laser processing parameters and grain direction on color change and surface roughness of pine and beech wood," BioResources 21(4), 10276–10293.

Abstract

This study was conducted to determine the effects of CO₂ laser scanning speed, laser processing power, and fiber direction on color change and surface roughness in Scots pine (Pinus sylvestris L.) and Eastern beech (Fagus orientalis Lipsky) wood. To this end, laser processing was performed at four scanning speeds (200, 300, 400, and 500 mm/s) and five power levels (12%, 13%, 14%, 15%, and 16%) in both parallel and perpendicular directions to the grain. The surface roughness and total color changes observed in the samples were determined in accordance with TS EN ISO 21920-2 and ASTM D2244-25 standards. As a result, while Scots pine samples exhibited higher roughness values than Eastern beech, the total color change was higher in Eastern beech samples. A low scanning speed and high laser power increased both roughness and color change. Processing parallel to the grain increased roughness, while processing perpendicular to the grain resulted in a higher color change.

Download PDF

Full Article

Effect of CO₂ Laser Processing Parameters and Grain Direction on Color Change and Surface Roughness of Pine and Beech Wood

Mehmet Karamanoğlu

DOI: 10.15376/biores.21.4.10276-10293

Keywords: CO₂ laser processing; Pine and beech wood; Scanning speed; Laser power; Grain direction; Color change; Surface roughness

Contact information: Department of Materials and Materials Processing, Tosya Vocational School, Kastamonu University, Kastamonu, Turkey;

*Corresponding author: [email protected]

Graphical Abstract

INTRODUCTION

Wood is widely used as an important engineering material in furniture, decorative products, interior fittings, and industrial design applications because of its renewability, workability, aesthetic appeal, and versatility. However, the anisotropic structure of wood is one of the key factors that make it difficult to control the surface quality achieved during processing. Depending on the wood species, varying anatomical characteristics – such as density, annual ring structure, the ratio of springwood to summerwood, cell wall thickness, and fiber direction – directly affect machining quality and the post-machining surface treatments such as staining, paint/varnish applications, etc. (Rezaei et al. 2022; Açık and Tutuş 2023). Therefore, the results obtained from laser processing on wood surfaces must be evaluated not only in terms of technological parameters but also in conjunction with the wood’s species-specific characteristics, anatomical structure, moisture content, and surface structure. Indeed, it has been reported that CO₂ laser cutting/processing performance in different wood species is related to wood type, cutting direction, density, and anatomical structure; furthermore, in species with distinct annual ring patterns, such as pine, differences between spring and summer wood can more significantly affect surface morphology (Guo et al. 2021; Ružiak et al. 2025).

Laser processing offers advantages over traditional machining methods, including non-contact operation, high precision, low tool wear, the ability to process complex patterns, compatibility with automation, and low mechanical stress. Due to these characteristics, laser technology is increasingly being used for cutting, engraving, marking, and surface modification of wood and wood-based materials. One of the main reasons for the widespread use of CO₂ lasers in wood processing is that the 10.6 µm wavelength is highly absorbed by lignocellulosic surfaces, thereby enabling controlled energy transfer during cutting, engraving, and surface modification processes (Martínez-Conde et al. 2017; Ding et al. 2026).

In CO₂ laser systems, absorption of beam energy by the wood surface causes a rapid increase in surface temperature; as a result, vaporization, partial carbonization, chemical decomposition, and changes in surface morphology occur. This process manifests as a darkening of the surface color, an increase in the total color difference, and changes in surface roughness (Li et al. 2020; Kubovský et al. 2020; Li et al. 2021; Hasan et al. 2021; Li et al. 2022; Kúdela et al. 2023; Açık and Tutuş 2024). However, the heat-affected zone from the laser process, along with effects such as carbonization, microcracks, and irregular surface burning, can make it difficult to control surface quality. Therefore, laser parameters must be carefully selected based on the material type and the desired surface properties (Corleto et al. 2025; Ding et al. 2026).

Among the most critical process parameters in laser surface modification are laser power and scanning speed. In general, an increase in laser power can lead to a more pronounced color change and greater surface degradation by increasing the amount of energy transferred to the wood surface. Conversely, an increase in scanning speed can limit the thermal effect and the resulting surface changes because it reduces the duration the beam remains on the surface. In addition, parameters such as focal position, raster density, scanning interval, and line density also alter the energy density transferred to the surface and can affect both color change and surface roughness. The color change observed on laser-treated wood surfaces is generally considered the visible result of chemical transformations induced by thermal degradation. The basic components of wood – cellulose, hemicellulose, and lignin – undergo structural changes due to the localized thermal effect created by the laser beam.

In particular, the transformation of chromophore groups in lignin and hemicellulose structures results in a darkening of the surface color and an increase in the total color difference. Similar trends in many studies have been explained as being the effects of laser power and scanning speed on color change. In general, increasing laser power increases the energy transferred per unit area, resulting in a more pronounced darkening of the wood surface. In contrast, an increase in scanning speed reduces the duration of the beam’s contact with the surface, thereby decreasing the intensity of the thermal effect and potentially limiting the color change (Gurău et al. 2017; Gurău and Petru 2018; Li et al. 2020; Li et al. 2021; Kúdela et al. 2022; Açık and Tutuş 2024; Gochev and Vitchev 2024; Savchenko and Kibirkštis 2025). Studies conducted on plywood and various other wood species also demonstrate that CO₂ laser engraving modes can produce controlled tonal variations in surface color, and that these color changes may be related to laser power, speed, focal position, and energy density (Gochev and Vitchev 2022; Chernykh et al. 2024).

Another important effect of laser processing on surface quality is the change in surface roughness. In laser processing, surface roughness is important not only for the product’s aesthetic appearance but also for subsequent processes. In painting, varnishing, coating, and bonding applications, surface roughness is related to wettability, mechanical adhesion, and surface free energy. Excessively carbonized or irregularly degraded surfaces can, in some cases, negatively affect coating quality and adhesion. The R_a* and R_z* parameters are among the most used indicators for evaluating the morphological condition of processed wood surfaces. Since the early and late woods regions on the surface exhibit different densities, cell structures, and thermal degradation behaviors during laser processing, roughness values may vary depending on the wood species and fiber direction (Kúdela et al. 2020; Kúdela et al. 2022; Li et al. 2022; Tonietto et al. 2022; Kúdela et al. 2023; Kúdela et al. 2024). It is emphasized that surface roughness should be considered not only in terms of average deviation values but also in conjunction with peak-to-valley differences and the distribution of surface irregularities. Therefore, the joint evaluation of the R_a* and R_z* parameters is important for explaining both the general roughness trend and more pronounced profile irregularities on laser-processed wood surfaces (Tonietto et al. 2022; Adamčík et al. 2023).

Fiber direction is one of the primary variables that determines surface quality during laser processing. At the same time, fiber direction is an important variable that affects the structure of the heat-affected zone formed after laser processing. Due to the anisotropic structure of wood, the direction of the laser beam – whether parallel or perpendicular to the grain – can have different effects on energy absorption, cell wall degradation, carbonization, and surface geometry (Rezaei et al. 2022; Adamčík et al. 2023; Adamčík et al. 2025; Corleto et al. 2025). Fiber direction is considered not only as a variable limited to parallel and perpendicular directions but also as a fundamental anatomical factor that alters surface quality along different fiber angles (Gurău et al. 2021; Gurău et al. 2022; Gurău et al. 2024).

A significant portion of the literature on CO₂ laser wood processing focuses on a single wood species, a limited set of processing parameters, or a specific surface property. In contrast, the combined evaluation of laser power, processing speed, fiber orientation, and wood species is important for a more comprehensive explanation of both surface quality and processing results. Indeed, Doğan et al. (2024) reported that laser power and processing speed must be jointly optimized in thermally modified beech veneers to achieve optimal surface quality and processing efficiency; Ružiak et al. (2025), on the other hand, noted that jointly modeling laser power, cutting speed, and anatomical cutting direction across different wood species is effective in predicting processing quality. However, the effects of interactions among laser power, scanning speed, and fiber orientation across different wood species on both color change and surface roughness have not been adequately addressed within a single experimental design. By comparing wood species with varying anatomical characteristics under identical laser processing conditions and jointly evaluating R_a*, R_z*, and ΔE*, this study contributes to a more comprehensive understanding of the relationship between processing parameters and material properties in CO₂ laser wood processing.

The purpose of this study was to determine the effects of laser scanning speed, processing power, and fiber direction on the total color change and surface roughness of pine and beech wood surfaces. To this end, laser treatments were performed at scanning speeds of 200, 300, 400, and 500 mm/s and processing power levels of 12%, 13%, 14%, 15%, and 16%, both parallel and perpendicular to the grain. The total color change on the post-processing surfaces was evaluated using the ΔE* value, while surface roughness was assessed using the R_a* and R_z* parameters. The goal was thus to reveal the comparative effects of the variables – wood species, laser scanning speed, processing power, and grain direction – on the surface quality of the wood material.

EXPERIMENTAL

Materials

In this study, Scots pine (Pinus sylvestris L.) and Eastern beech (Fagus orientalis Lipsky) wood were selected as test materials because of their widespread use. The wood samples were obtained from the market as lumber. The lumber used in the study was randomly selected in accordance with ASTM D7787/D7787M-13 (2022) standards from mixed heartwood and sapwood sections; the samples used were from straight-grained, knot-free, crack-free, and free of color and density variations, with annual rings randomly distributed across the surfaces. The samples were cut into 350 × 100 × 15 mm blanks, with the grain direction parallel to the length axis. The samples were then conditioned in a climate chamber (TS 642 ISO 554, 1997) at a temperature of 20±2 ^oC and relative humidity of 65±5% until their weight stabilized (TS ISO 13061-1, 2021). After conditioning, the specimens were cut to net dimensions of 330 × 85 × 10 mm. They were then sanded with 80-, 100-, and 220-grit sandpaper. Eight specimens of each wood species were prepared in the specified dimensions (Fig. 1).

Fig. 1. Experimental test samples

Methods

The laser processing was performed using the Çelikmak CO₂ laser cutting/engraving system (Çelikmak, Konya, Turkey). An EFR brand F4 model CO₂ laser tube was used in the laser processing (Beijing EFR Laser S&T Co., Ltd., Beijing, China). The device has a nominal laser output power of 100 W and a peak output power of 120 W; the laser beam wavelength is 10.6 µm. For each wood type, laser processing was performed at four different scanning speeds (200, 300, 400, and 500 mm/s), at five different laser processing power levels (12%, 13%, 14%, 15%, and 16%) based on the nominal 100 W laser tube power set via the device software, and in two different fiber directions. Preliminary tests showed that while adequate surface modification could not be achieved at low laser power levels, excessive darkening and carbonization occurred at higher power levels. Therefore, laser power levels were set between 12% and 16% to ensure controlled surface modification. The laser processing patterns, target areas, and laser parameters were prepared using RD Works V8.01.11 software (Shenzhen RuiDa Technology Co., Ltd., Shenzhen, China) and transferred to the device. The focal length for laser processing was set to 50.8 mm and kept constant throughout all experiments. The distance between the laser head and the sample surface was set to 23.6 mm. The treatments were applied in a single pass and kept constant throughout the scanning process (76.2 raster density / 10.30 line spacing / 5 mm resolution). For each treatment combination, four 25 × 25 mm square areas were treated, resulting in a total of 320 laser-treated areas (Fig. 1). Laser treatments were performed under identical environmental conditions, and treatment parameters were held constant throughout the experiment. Auxiliary air was used, and the air pressure was set to 3 bar.

Surface roughness measurements of the samples were performed in accordance with TS EN ISO 21920-2 (2022) using an Accretech Handysurf E-35 B (Tokyo, Seimitsu) stylus-scanning surface roughness measuring instrument. Roughness measurements were taken perpendicular to the fibers. In this study, the average surface roughness (R_a*) and ten-point average surface roughness (R_z*) values of the samples were measured. Measurements were performed at room temperature, ensuring that the samples and the instrument were parallel to the horizontal plane, away from sources of vibration and noise, with a stylus diameter of 5 μm, a measurement speed of 0.5 mm/s, a scan length of 12.5 mm, and a cut-off wavelength of λ_c = 2.5 mm. To ensure measurement accuracy, the instrument was recalibrated every 50 measurements.

Color measurements were performed in accordance with the principles specified in ASTM D2244-25 (2025). Color changes were determined using a Konica Minolta CM-2300d instrument, which measures in accordance with the CIEL*a*b* color system. The instrument was calibrated to the D65 light source and 10 °. According to the CIEL*a*b* system, the total color difference (∆E*) was determined using Eq. 1 (Karamanoğlu 2020).

(1)

The total color change (∆E*) values of the samples were evaluated relative to the control groups. The subscript 1 in ΔL*, Δa*, and Δb* (ΔL* = L*₂ – L*₁,Δa* = a*₂ – a*₁, Δb* = b*₂ – b*₁) refers to the untreated control samples, whereas subscript 2 refers to the samples subjected to laser processing.

All statistical analyses were performed in R (R Core Team 2025, version 4.5.1). A four-way factorial analysis of variance (ANOVA) was conducted to determine the individual and interactive effects of wood type (WT; two levels), laser scanning speed (LSS; four levels), laser processing power (LPP; five levels), and fiber direction (FD; two levels) on average surface roughness (R_a), ten-point average surface roughness (R_z), and total color change (ΔE). Each treatment combination was replicated 8 times (n = 8), yielding 640 observations per response variable. The full factorial model, including all two-way, three-way, and four-way interactions, was fitted using the “aov()” function of the base “stats” package. Prior to ANOVA, the normality of model residuals was examined using the Shapiro-Wilk test and Q-Q plots, and the homogeneity of variances was assessed using Levene’s test. Because the design was large and fully balanced (n = 8 per cell), ANOVA was considered robust to moderate departures from these assumptions. For significant factors, Duncan’s multiple range test and letter grouping were performed with the “agricolae” package (Duncan. Test) at p < 0.05. Significance levels were expressed as p < 0.05 (*), p < 0.01 (**), p < 0.001(***), and non-significant (ns). Significant two-way interactions were visualized using violin plots overlaid with box plots and individual observations. Pearson’s and Spearman’s correlation coefficients among R_a*, R_z*, and ΔE* were calculated using the base “stats” package (cor.test), both overall and within wood-type and fiber-direction subgroups. Data handling and plotting were performed with the “readxl”, “dplyr”, and “ggplot2” packages. The untreated control samples were measured for reference purposes and were not included in the ANOVA model.

RESULTS AND DISCUSSION

In this study, the changes induced by CO₂ laser processing parameters on the surfaces of pine and beech wood were evaluated in terms of average surface roughness (R_a*), ten-point average surface roughness (R_z*), and total color change (ΔE*). The results of the analysis of variance showed that the main factors – wood type (WT), laser scanning speed (LSS), laser processing power (LPP), and fiber direction (FD) – had a statistically highly significant effect on all three response variables (p < 0.001) (Table 1). These results indicate that surface quality after CO₂ laser treatment was governed by the combined effects of processing parameters and the wood’s heterogeneous, anisotropic structure. The anatomical properties of wood can influence the surface properties resulting from laser processing. However, differences in heartwood/sapwood, radial/tangential surfaces, and earlywood/latewood were not evaluated as independent variables in this study. Therefore, the obtained R_a*, R_z*, and ΔE* results were evaluated only within the scope of the wood type, laser scanning speed, laser processing power, and fiber direction examined in the study. The literature also emphasizes that surface quality during laser wood processing is shaped by the combined effects of factors such as laser power, scanning/cutting speed, focal position, raster density, fiber direction, moisture content, and wood type (Martínez-Conde et al. 2017; Kubovský et al. 2020; Ružiak et al. 2025; Ilçe and Çiftçi 2025; Ding et al. 2026).

According to the ANOVA results, R_a* emerged as the parameter most sensitive to the factors examined and to their interactions. For R_a*, all main factors – WT, LSS, LPP, and FD – were significant at p < 0.001. In addition, the two-way, three-way, and four-way interactions of all factors were found to be significant. In contrast, while all main effects were significant for R_z* values, the other interactions were found to be more limited. For ΔE*, all main factors are also significant. However, the three-way and four-way interactions involving WT × LPP were not significant. This structure indicates that in laser processing, R_a* responds to general microtopographic changes on the surface, R_z* responds primarily to peak-to-valley irregularities, and ΔE* responds more regularly to energy input and chemical/colorimetric transformations on the surface.

Table 1. Effect of Wood Type, Laser Scanning Speed, Laser Processing Power, and Fiber Direction on Average Surface Roughness, 10-Point Average Surface Roughness, and Total Color Change

The value of R_a* showed significant differences after CO₂ laser processing, depending on the wood type, scanning speed, processing power, and fiber direction. According to Duncan’s test results, the average R_a* value was 14.3 µm for pine and 7.17 µm for beech, indicating greater microtopographic roughness on pine surfaces. This difference was also evident in Fig. 2A, particularly at 200 mm/s, where pine exhibited a broader distribution and higher outlier values, suggesting stronger ablation and more heterogeneous surface degradation under low-speed conditions.

The high R_a* values observed in pine samples can be attributed to the annual ring structure of pine wood and the difference in density between early and late wood. Guo et al. (2021) reported that structural and density differences in the early and late wood regions during CO₂ laser cutting of pine wood create irregular morphology on the cross-sectional surface and that the laser effect does not progress homogeneously due to these anatomical differences. Similarly, Gurău et al. (2024) emphasized that species differences and the ablative behavior of early and late wood determine surface undulation and roughness on laser-cut wood surfaces. In this context, the higher R_a* values obtained in pine samples can be explained by the non-uniform absorption of laser energy along the annual ring.

As for the R_z* values, the average R_z* value for pine samples was determined to be 75.9 µm, while that for beech samples was 47.4 µm. The R_z* panel of the WT × LSS graph (Fig. 2B) shows that pine samples exhibit a wider distribution and higher extreme values, particularly at speeds of 200 mm/s and 400 mm/s. This pattern indicates that laser-induced surface irregularities in pine wood are higher not only in terms of average roughness but also in terms of peak-to-valley profile differences. Guo et al. (2021) noted that during CO₂ laser cutting of pine wood, a V-shaped kerf structure, carbonization, scoring, and irregular surface morphologies can be observed, and that differences between early and late wood create irregularities on the cut surface.

The total color change (ΔE*) exhibited behavior distinct from that of the surface roughness parameters. While R_a* and R_z* values were higher in the pine samples, ΔE* values were higher in the beech samples. According to the results of Duncan’s test, the mean ΔE* value was 16.0 for beech and 11.2 for pine. This indicates that changes in surface morphology and color do not necessarily progress in the same direction as a result of the laser treatment. In other words, while pine surfaces exhibited higher roughness, beech surfaces showed a more pronounced colorimetric change. In the ΔE* panel of the WT × LSS graph (Fig. 2C), beech samples exhibit higher ΔE* values, particularly at speeds of 200 and 300 mm/s, while pine samples generally exhibit lower ΔE* values. The pronounced color change in beech wood may be related to laser-induced thermal and chemical transformations. Kúdela et al. (2024) reported that following CO₂ laser processing of beech wood, the L* value decreased significantly, while ΔE* values increased, and that this color change was related to the thermal decomposition of carbonyl groups in the lignin and hemicellulose structures. Gurău et al. (2017) demonstrated that the total color difference in beech wood increased as the scanning speed decreased and the laser power increased, and that strong correlations could be established between ΔE* and surface roughness. The higher ΔE* values obtained in beech samples in this study indicate that the beech surface exhibits a more pronounced colorimetric response under the CO₂ laser beam.

Fig. 2. Parameters interactions of WT × LSS: R_a* (A), R_z* (B), and ΔE* (C)

The laser scanning speed has been shown to have a significant reducing effect on R_a*. The average R_a* value decreased from 17.6 µm at 200 mm/s to 10.3 µm at 300 mm/s, 8.07 µm at 400 mm/s, and 7.04 µm at 500 mm/s. According to Duncan’s test, all speed levels differed significantly from one another. The R_a* panel of the LSS × FD graph (Fig. 3A) visually supports this trend. The highest R_a* values were observed at 200 mm/s in both fiber directions, and the roughness values decreased gradually with increasing speed. This can be explained by the laser beam remaining in contact with the surface for longer at low scanning speeds, thereby increasing the thermal energy transferred to the surface. Gurău et al. (2017) investigated the effects of CO₂ laser power and scanning speed on surface roughness and color change in beech wood. They reported that roughness increased with laser power and decreased with scanning speed. Gurău and Petru (2018) similarly demonstrated that R_a* and other roughness parameters in maple wood are sensitive to increases in laser power and decrease with increases in scanning speed. Li et al. (2022) noted that R_a* and R_z* values in poplar wood became higher with increasing laser power but decreased with increasing feed rate and pass spacing. Therefore, the low-speed-high R_a* relationship observed in this study can be interpreted as a general surface degradation behavior resulting from the increased energy density transferred to the surface during CO₂ laser processing.

The effect of laser scanning speed on R_z* showed a decreasing trend parallel to that of R_a*. The highest R_z* was obtained at 200 mm/s, at 88.40 µm. This value decreased to 62.64 µm at 300 mm/s, to 48.24 µm at 400 mm/s, and to 47.28 µm at 500 mm/s. However, according to Duncan’s test results, 400 and 500 mm/s fell into the same statistical group. This result indicates that the effect of increasing speed on R_z* is particularly pronounced in the 200 to 400 mm/s range; beyond 400 mm/s, there is only a limited improvement in the peak-to-valley differences of the surface profile. In the R_z* panel of the LSS × FD graph (Fig. 3B), higher R_z* values are observed in both fiber directions at 200 mm/s, and these values decrease as speed increases. Li et al. (2022) reported that in poplar wood, R_a* and R_z* values increased with laser power but decreased with feed rate. Adamčík et al. (2023) reported that surface roughness after CO₂ laser treatment of beech, oak, and spruce wood varied with laser power and feed rate. Adamčík et al. (2025), on the other hand, demonstrated that R_a* and R_z* values in beech wood are sensitive to CO₂ laser power-speed combinations and that a smoother surface can be achieved with appropriate parameter selection.

The effect of the scanning speed on ΔE* was quite pronounced. The highest ΔE* value of 19.2 was obtained at 200 mm/s. As the scanning speed was increased to 300, 400, and 500 mm/s, the ΔE* values decreased to 14.19, 11.22, and 9.73, respectively. According to Duncan’s test results, all speed levels differed significantly from one another. The ΔE* panel of the LSS × FD graph (Fig. 3C) shows that the color change decreased with increasing speed in both fiber directions. This can be explained by the fact that at low speeds, the laser beam remains on the surface longer, resulting in greater energy input, which in turn increases the color change. Li et al. (2020) reported that laser power increases ΔL* and ΔE* in moso bamboo, whereas feed rate and scan width reduce them. Li et al. (2021) noted that laser power, feed rate, and focus position significantly affect the area of color change in poplar wood. Açık and Tutuş (2024) demonstrated that laser engraving power increases the total color change in different wood types, while laser engraving speed has a reducing effect. Gurău et al. (2017) also found that ΔE* in beech increases particularly at low scanning speeds and decreases sharply as the speed increases.

Fig. 3. Parameters interactions of LSS × FD: R_a* (A), R_z* (B), and ΔE* (C)

Laser processing power also significantly increased the R_a* values. The R_a value, which was 6.69 µm at 12% power, increased with increasing power to 9.23 µm at 13%, 10.88 µm at 14%, 12.81 µm at 15%, and 14.14 µm at 16%. According to Duncan’s test, all power levels are statistically different from one another in terms of R_a*. In the R_a* panel of the LPP × FD graph (Fig. 4A), R_a* values became higher in both fiber directions with increasing power, with a more pronounced change in the PAF direction. This result indicates that as laser power increases, the amount of energy transferred to the surface rises, leading to greater cell wall degradation, carbonization, and surface irregularity. Kúdela et al. (2022) reported that CO₂ laser irradiation parameters affect surface roughness in oak wood, and that mass loss and roughness increase as the total irradiation dose transferred to the surface increases. Kúdela et al. (2023) noted that as the irradiation dose increased in spruce wood, R_a* and R_z* values also increased, and that this was related to irregular mass loss resulting from density differences in spring-summer wood. Gochev and Vitchev (2024), on the other hand, demonstrated that the laser focus position and power density significantly affected the R_a* and R_z* values in plywood, and that laser-treated surfaces were rougher than untreated surfaces. These studies support the finding that R_a* values increase with higher power.

Fig. 4. Parameters interactions of LPP × FD: R_a* (A), R_z* (B), and ΔE* (C)

In terms of fiber direction, the average R_a* value was found to be 12.6 µm in the PAF direction and 8.9 µm in the PEF direction. This difference is statistically significant. When the LSS × FD (Fig. 3A) and LPP × FD (Fig. 4A) graphs are evaluated together, the difference in fiber direction became particularly more pronounced under low scanning speed and high-power conditions. In the R_a* panel of the LSS × FD graph, it is evident that at 200 mm/s, the values in the PAF direction were higher and more widely distributed compared to those in the PEF direction. Similarly, in the LPP × FD graph, the PAF direction reached higher R_a* values at 15% and 16% power levels. This finding demonstrates that fiber direction is a determining factor in the surface distribution of laser energy and the degradation pattern in wood tissue.

Açık and Tutuş (2023) noted that density, anatomical structure, and cutting direction are decisive factors in the CNC laser machining performance of different wood types. Gurău et al. (2021) demonstrated that in laser and CNC machining of spruce wood at different fiber angles, the cutting angle is a key factor affecting surface quality. That surface quality was negatively affected at specific angles such as 15° and 60°. Gurău et al. (2022) reported that the fiber angle in maple affects the surface morphology of laser-processed surfaces, noting that the laser can produce a melting/smoothing effect, particularly in summerwood regions. Gurău et al. (2024), on the other hand, noted that the CO₂ laser reveals different anatomical details depending on the grain direction in beech and maple and that deeper ablation can occur in springwood regions. Accordingly, the higher R_a* values obtained in the PAF direction indicate that a more pronounced and irregular microtopographic structure forms on the surface, depending on the fiber direction.

The increase in laser processing power also increased the R_z* values. While R_z* was 43.0 µm at 12% power, it was 54.2 and 61.4 µm at 13% and 14% power, respectively. At 15% and 16% power levels, R_z* rose to 73.6 and 76.0 µm, respectively. According to the Duncan’s test, the 13% to14% range fell into the same group, while the 15% to 16% range fell into a higher group. The R_z* panel of the LPP × FD graph (Fig. 4B) shows that higher R_z* values form in the PAF direction, particularly at high power levels. This result indicates that an increase in laser power accentuates deep peak-to-valley irregularities on the surface, but that R_z* can exhibit a similar morphological response within certain power ranges. Kúdela et al. (2020) noted that laser power and raster density affect roughness and waviness parameters in beech wood, and that surface roughness increases under high-power and high-raster-density conditions. Kúdela et al. (2023) reported that R_a* and R_z* values in spruce wood increase with irradiation dose and mass loss, and that the increase can be much more pronounced, particularly in measurements perpendicular to the grain. Kúdela et al. (2022), on the other hand, demonstrated that the amount of laser-transferred energy in oak wood jointly affects surface morphology, color change, and roughness. These studies support the finding that R_z* increases with increasing power.

Fiber direction had a significant effect on R_z*. The average R_z* value was determined to be 72.2 µm in the PAF direction and 51.1 µm in the PEF direction. This result indicates that the maximum irregularities in the surface profile are higher in the PAF direction. When the R_z* panels of the LSS × FD (Fig. 3B) and LPP × FD (Fig. 4B) graphs are evaluated together, it is observed that the difference in fiber direction becomes more pronounced at low speeds and high powers. This indicates that the fiber direction affects not only the average surface roughness but also the deep peak-to-valley character of the profile. Rezaei et al. (2022) reported that surface quality in beech wood following CO₂ laser cutting varied with cutting speed, gas pressure, focal position, and wood moisture content, and that laser-cut surfaces exhibited different morphological characteristics compared to those cut by conventional methods. Corleto et al. (2025) investigated the effects of the focal point, cutting speed, and gas pressure on color change and surface roughness in beech wood and demonstrated that the focal position is significant for surface quality. Gochev and Vitchev (2024), on the other hand, noted that a change in the focal position alters the power density of the laser beam, which can lead to an increase in R_a* and R_z* values. These results indicate that the differences in R_z* caused by fiber direction and power-speed combinations are related to the laser beam’s energy density on the wood surface and the wood’s anatomical orientation.

The R_z* results also indicate that, in studies on laser wood processing, not only the average roughness but also larger-scale irregularities in the surface profile must be considered. Gurău et al. (2021) recommended the combined evaluation of roughness, waviness, and primary profile parameters to understand wood surface quality. Tonietto et al. (2022) emphasized that average parameters, such as R_a*, may not fully capture the characteristics of the valley and contact areas on heterogeneous surfaces. In this regard, the combined evaluation of R_a* and R_z* is important for explaining both the average and peak-to-valley characteristics of laser-induced surface degradation.

Laser processing power had an increasing effect on ΔE*. The ΔE* value, which was 11.6 at 12% power, increased to 13.1 at 13%, 13.6 at 14%, 14.4 at 15%, and 15.1 at 16%. According to Duncan’s groupings, 13% and 14% were in the same group, while 15% and 16% were in the highest group. The ΔE* panel of the LPP × FD graph (Fig. 4C) shows that color change increased in both fiber directions as power increased. This result is consistent with the fact that as laser power increases, the energy transferred to the surface rises, leading to stronger thermal discoloration. Gochev and Vitchev (2022) reported that as laser power increases in plywood, carbonization and color saturation increase, while lighter tones form at lower powers. Chernykh et al. (2024) noted that color changes resulting from laser engraving across different wood species can be correlated with energy density and process parameters. Kúdela et al. (2024) demonstrated that even at low laser power and raster density, distinct color differences can occur in beech wood, and that as energy input increases, the surface color shifts toward darker brown tones. Hasan et al. (2021) noted that as laser power increases on redwood, both engraving depth and carbonization increase, while laser-material interaction decreases with increasing speed.

In terms of fiber direction, the average ΔE* was 14.6 in the PEF direction and 12.6 in the PAF direction. Accordingly, the PEF direction was more effective at inducing color change. In the ΔE* panel of the LPP × FD graph (Fig. 4C), higher ΔE* values occurred in the PEF direction as power was increased. In the LSS × FD graph (Fig. 3C), however, ΔE* values decreased in both fiber directions as the scanning speed increased; nevertheless, the PEF direction tended to produce greater color change at low speeds. This finding indicates that fiber direction affects roughness and color shift through different mechanisms. While mechanical/morphological irregularities increased in the PAF direction, the color shift became more pronounced in the PEF direction. This variation indicates that laser-wood interaction was related not only to surface topography but also to surface chemistry and light absorption. Ding et al. (2026) noted that photothermal processes in laser-wood interactions can lead to complex outcomes due to the wood’s chemical composition, anisotropy, and heterogeneity. Gurău et al. (2024) demonstrated that the anatomical details and chemical changes revealed by the laser can vary depending on the fiber direction in beech and maple. Kúdela et al. (2024) emphasized that color changes are particularly associated with changes in the chromophores of lignin and hemicellulose structures. Therefore, the higher ΔE* in the PEF direction indicates that different thermal discoloration and chromophore transformations may occur on the surface depending on the fiber direction.

Overall, different results were observed for surface roughness and color change on pine and beech wood samples following CO₂ laser treatment. Pine exhibited higher roughness values in terms of R_a* and R_z*, while beech showed a more pronounced color change in terms of ΔE*. A low scanning speed and high laser power increased both surface roughness and color change. While R_a* and R_z* values increased in the PAF direction, ΔE* values increased in the PEF direction. These results indicate that the selection of optimal parameters for CO₂ laser wood surface treatment should not be made in isolation with the sole aim of achieving lower roughness or greater color change, but rather by taking into account both the wood species type and the fiber direction. Although the laser’s non-contact, precise nature and suitability for digital manufacturing provide significant advantages, effects such as carbonization, color darkening, heat-affected zones, and surface irregularities must be controlled (Kúdela et al. 2022; Gochev and Vitchev 2024; Ding et al. 2026).

To further clarify the relationships among the measured responses, Pearson and Spearman correlation coefficients were calculated for R_a*, R_z*, and ΔE* (Table 2). The two roughness parameters were strongly and consistently correlated (Spearman’s ρ = 0.96, p < 0.001), confirming the internal reliability of the roughness measurements. The pronounced difference between the Pearson (r = 0.72) and Spearman (ρ = 0.96) coefficients for the R_a* to R_z* pair indicates a monotonic but non-linear relationship, in which R_z* increases disproportionately faster than R_a* under high-energy processing conditions, most likely because isolated deep craters and carbonized cavities influence the peak-to-valley parameter (R_z*) more strongly than the arithmetic mean deviation (R_a*). In contrast, ΔE* exhibited only weak linear correlations with R_a* (r = 0.24, p < 0.001) and R_z* (r = 0.16, p < 0.001) at the overall level. This weak association indicates that surface roughness and color change are governed largely by distinct mechanisms during CO₂ laser processing: material ablation primarily determines surface morphology, whereas the thermal-chemical degradation of wood constituents and the associated formation of chromophores primarily determine color change. This interpretation is consistent with the opposite behavior of the fiber-direction factor, in which processing parallel to the grain increased roughness, whereas processing perpendicular to the grain increased color change. When the analysis was restricted to individual wood species, the R_a* to ΔE* correlation strengthened to a moderate level (Scots pine: r = 0.57; Eastern beech: r = 0.54; both p < 0.001), indicating that although the two responses originate from a common energy input, they respond at different rates depending on the wood structure.

Table 2. Pearson and Spearman Correlation Coefficients among Average Surface Roughness (R_a), 10-point Average Surface Roughness (R_z), and Total Color Change (ΔE*), Calculated Overall and within Wood Type and Fiber Direction Subgroups

CONCLUSIONS

Wood type, laser scanning speed, laser processing power, and fiber direction had statistically significant effects on R_a*, R_z*, and ΔE*. This result demonstrates that surface quality on laser-processed wood is determined not only by technological parameters but also by the wood’s anatomical properties.
Pine samples exhibited higher surface roughness values compared to beech samples. The average R_a* value was 14.3 µm for pine and 7.17 µm for beech; the average R_z* value was 75.9 µm for pine and 47.4 µm for beech. In contrast, the total color difference was found to be higher in the beech samples than in the pine samples. The ΔE* value was 16.02 for beech and 11.17 for pine. This indicates that responses to roughness and color differences differed between the two wood types.
Increasing the laser scanning speed reduced the surface roughness and total color change values. The highest R_a*, R_z*, and ΔE* values were obtained at a scanning speed of 200 mm/s; at higher scanning speeds, the surfaces exhibited lower roughness and more limited color change.
Increasing the laser processing power increased surface roughness and total color change. Overall, the highest R_a*, R_z*, and ΔE* values were obtained at laser power levels of 15% and 16%.
The fiber direction affected the examined surface properties in different ways. Processing parallel to the fibers resulted in higher R_a* and R_z* values, while processing perpendicular to the fibers yielded higher ΔE* values. These findings indicate that, to achieve low surface roughness and controlled color change during CO₂ laser surface treatment of pine and beech wood, the wood type, scanning speed, laser power, and fiber direction must be evaluated together.

ACKNOWLEDGMENTS

The author is grateful for the support of Kastamonu University Center for the Application and Research of Wood Culture.

REFERENCES CITED

Açık, C., and Tutuş, A. (2023). “Investigation of CNC laser processing performance of some wood species used in industrial product manufacturing,” Journal of the Faculty of Engineering and Architecture of Gazi University 38(1), 461-470. https://doi.org/10.17341/gazimmfd.986215

Açık, C., and Tutuş, A. (2024). “Investigation of the effect of laser modification on some wood color characteristics,” Journal of The Faculty of Engineering and Architecture of Gazi University 39(3), 1973-1981. https://doi.org/10.17341/gazimmfd.1233588

Adamčík, L., Igaz, R., Štefančin, L., Kubovský, I., and Kminiak, R. (2023). “Evaluation of the surface irregularities of the cross-section of the wood after CO₂ laser cutting,” Materials 16(22), article 7175. https://doi.org/10.3390/ma16227175

Adamčík, L., Kminiak, R., Kyselica, K., Igaz, R., and Kubovský, I. (2025). “Optimization of CO₂laser technological parameters and their impact on the surface quality of cut wood,” Forests 16(5), article 785. https://doi.org/10.3390/f16050785

ASTM D2244-25 (2025). “Standard practice for calculation of color tolerances and color differences from instrumentally measured color coordinates,” ASTM International, West Conshohocken, PA.

ASTM D7787/D7787M-13 (2022). “Standard practice for selecting wood substrates for weathering evaluations of architectural coatings,” ASTM International, West Conshohocken, PA.

Chernykh, M., Korepanova, A., Maksimova, E., Sevryugin, V., Gilfanov, M., and Štollmann, V. (2024). “Selecting color of mosaic pattern elements for laser engraving on wood,” Acta Facultatis Xylologiae Zvolen 66(2), 89-102. https://doi.org/10.17423/afx.2024.66.2.08

Corleto, R., Gaff, M., Sethy, A. K., Kelkar, B. U., Nemeth, R., Ditommaso, G., and Rezaei, F. (2025). “Effects of CO₂ laser parameters on quality of laser cut beech wood (Fagus sylvatica L.) surface and extent of heat affect zone,” Journal of the Indian Academy of Wood Science 22(1), 67-77. https://doi.org/10.1007/s13196-024-00361-2

Ding, Y., Dreimol, C. H., Rouèche, M., Stumpp, A., Holtz, R., Zhou, C., and Burgert, I. (2026). “Laser as a toolbox for wood processing and functionalization,” MRS Bulletin 1-16. https://doi.org/10.1557/s43577-026-01063-5

Doğan, G., Wascher, R., Avramidis, G., and Gerhard, C. (2024). “Using Taguchi optimization to increase machining quality and efficiency when CO₂ laser structuring thermally modified beech wood veneers,” Lasers in Engineering Old City Publishing, 58.

Gochev, Z., and Vitchev, P. (2022). “Colour modifications in plywood by different modes of CO₂ laser engraving,” Acta Facultatis Xylologiae Zvolen res Publica Slovaca 64(2), 77-86. https://doi.org/10.17423/afx.2022.64.2.08

Gochev, Z., and Vitchev, P. (2024). “The effect of the focal length position of the focusing lens on the discoloration roughness of plywood under different CO₂ laser engraving modes,” Acta Facultatis Xylologiae Zvolen 66(2), 75-88. https://doi.org/10.17423/afx.2024.66.2.07

Guo, X., Deng, M., Hu, Y., Wang, Y., and Ye, T. (2021). “Morphology, mechanism and kerf variation during CO₂ laser cutting pine wood,” Journal of Manufacturing Processes 68, 13-22. https://doi.org/10.1016/j.jmapro.2021.05.036

Gurău, L., Petru, A., Varodi, A., and Timar, M. C. (2017). “The influence of CO₂ laser beam power output and scanning speed on surface roughness and colour changes of beech (Fagus sylvatica),” BioResources 12(4), 7395-7412. https://doi.org/10.15376/biores.12.4.7395-7412

Gurău, L., and Petru, A. (2018). “The influence of CO₂ laser beam power output and scanning speed on surface quality of Norway maple (Acer platanoides),” BioResources 13(4), 8168-8183. https://doi.org/10.15376/biores.13.4.8168-8183

Gurău, L., Coşereanu, C., and Paiu, I. (2021). “Comparative surface quality of larch (Larix decidua Mill.) fretwork patterns cut through by CNC routing and by laser,” Applied Sciences 11(15), article 6875. https://doi.org/10.3390/app11156875

Gurău, L., Coșereanu, C., Timar, M. C., Lungu, A., and Condoroţeanu, C. D. (2022). “Comparative surface quality of maple (Acer pseudoplatanus) cut through by CNC routing and by CO₂ laser at different angles as related to the wood grain,” Coatings 12(12), article 1982. https://doi.org/10.3390/coatings12121982

Gurău, L., Angelescu, A. M., and Timar, M. C. (2024). “Surface morphology and chemical changes of maple and beech cut through by CO₂ laser under different angles relative to the wood grain,” Forests 15(10), article 1767. https://doi.org/10.3390/f15101767

Hasan, S. M., Hubeatir, K. A., and Abd, D. S. (2021). “Effect of CO₂ laser parameters on redwood engraving process complemented by Taguchi method,” Materials Today: Proceedings 42, 2566-2572. https://doi.org/10.1016/j.matpr.2020.12.580

Ilçe, A.C., and Çiftçi, S. (2025). “Optimization of diode laser cutting parameters for wood veneers using artificial neural networks,” Wood Material Science & Engineering 2025, article 2586259. https://doi.org/10.1080/17480272.2025.2586259

Karamanoğlu, M. (2020). Resistance Properties of Water-Based Varnishes Modified with Nanoparticles Against the Aging Effect of UV on Surface of Some Thermo-Wood Materials, Doctoral Thesis, Kastamonu University, Institute of Science, Department of Forest Industrial Engineering, Kastamonu, Turkey.

Kubovský, I., Krišťák, Ľ., Suja, J., Gajtanska, M., Igaz, R., Ružiak, I., and Réh, R. (2020). “Optimization of parameters for the cutting of wood-based materials by a CO₂ laser,” Applied Sciences 10(22), article 8113. https://doi.org/10.3390/app10228113

Kúdela, J., Kubovský, I., and Andrejko, M. (2020). “Surface properties of beech wood after CO2 laser engraving,” Coatings 10(1), article 77. https://doi.org/10.3390/coatings10010077

Kúdela, J., Kubovský, I., and Andrejko, M. (2022). “Influence of irradiation parameters on structure and properties of oak wood surface engraved with a CO₂ laser,” Materials 15(23), article 8384. https://doi.org/10.3390/ma15238384

Kúdela, J., Andrejko, M., and Kubovský, I. (2023). “The effect of CO₂ laser engraving on the surface structure and properties of spruce wood,” Coatings 13(12), article 2006. https://doi.org/10.3390/coatings13122006

Kúdela, J., Kubovský, I., and Andrejko, M. (2024). “Discoloration and chemical changes of beech wood after CO₂ laser engraving,” Forests 15(12), article 2211. https://doi.org/10.3390/f15122211

Li, R., Chen, J., and Wang, X. A. (2020). “Prediction of the color variation of moso bamboo during CO₂ laser thermal modification,” BioResources 15(3), 5049-5057. https://doi.org/10.15376/biores.15.3.5049-5057

Li, R., He, C., Chen, Y., and Wang, X. (2021). “Effects of laser parameters on the width of color change area of poplar wood surface during a single irradiation,” European Journal of Wood and Wood Products, 79(5), 1109-1116. https://doi.org/10.1007/s00107-021-01706-1

Li, R., He, C., Xu, W., and Wang, X. A. (2022). “Prediction of surface roughness of CO₂ laser modified poplar wood via response surface methodology,” Maderas. Ciencia y Tecnología 24. https://doi.org/10.4067/s0718-221×2022000100442

Martínez-Conde, A., Krenke, T., Frybort, S., and Müller, U. (2017). “Comparative analysis of CO₂ laser and conventional sawing for cutting of lumber and wood-based materials,” Wood Science and Technology 51(4), 943-966. https://doi.org/10.1007/s00226-017-0914-9

R Core Team. R (2025). A Language and Environment For Statistical Computing; R Foundation for Statistical Computing, Vienna, Austria, Available online: https://www.R-project.org/.

Rezaei, F., Wimmer, R., Gaff, M., Gusenbauer, C., Frömel-Frybort, S., Kumar Sethy, A., and Niemz, P. (2022). “Anatomical and morphological characteristics of beech wood after CO₂-laser cutting,” Wood Material Science and Engineering 17(6), 459-468. https://doi.org/10.1080/17480272.2022.2134820

Ružiak, I., Kristak, L., Adijans, I., Kubovský, I., Richvalská, J., Štefančin, L., and Todaro, L. (2025). “Enhancing CO₂ laser cutting efficiency for diverse wood species using artificial neural networks,” Forests 16(6), article 881. https://doi.org/10.3390/f16060881

Savchenko, O., and Kibirkštis, E. (2025). “Development and research of the laser engraving process in packaging manufacturing technology,” Mechanics 31(3), 237-242. https://doi.org/10.5755/j02.mech.41295

Tonietto, L., Arnold, D. C. M., de Oliveira, V. C., Menegotto, C. W., Grondona, A. E. B., Costa, C. A. D., and Gonzaga Jr, L. (2022). “Method for evaluating roughness and valley areas coefficients of surfaces acquired by laser scanner,” Scientific Reports 12(1), article 1486. https://doi.org/10.1038/s41598-022-04847-2

TS EN ISO 21920-2 (2022). “Geometrical product specifications (GPS) – Surface texture: Profile – Part 2: Terms, definitions and surface texture parameters,” Turkish Standards Institute, Ankara.

TS ISO 13061-1 (2021). “Physical and mechanical properties of wood – Test methods for small clear wood specimens – Part 1: Determination of moisture content for physical and mechanical tests,” Turkish Standards Institute, Ankara.

TS 642 ISO 554 (1997). “Standard atmospheres for conditioning and/or testing; Specifications,” Turkish Standards Institute, Ankara.

Article submitted: July 14, 2026; Peer review completed: August 15, 2026; Revised version received: August 20, 2026; Accepted: August 21, 2026; Published: September 2, 2026.

DOI: 10.15376/biores.21.4.10276-10293