Effects and Modification Mechanisms of Different Plasma Treatments on the Surface Wettability of Different Woods

Abstract

Plasma treatment of wood surfaces has shown significant effects, but different excitation methods used for different species of wood generally result in varied characteristics of wood surfaces. Secondly, plasma modification greatly enhances the absorption of liquids by wood, but the relationship between liquid absorption and surface wettability is rarely studied. Limited detailed investigation of the modification effects and mechanisms has hindered the large-scale applications of plasma treatment in the wood industry. In this study, two typical plasmas, radio frequency (RF) plasma and gliding arc discharge (GAD) plasma, were employed to treat three species of wood: poplar, black walnut, and sapele. By focusing on changes in the contact angle of the wood surface, an exponential equation fitting method is used to determine the measurement time for contact angles. The research identified that factors contributing to the decrease in contact angle after plasma modification include not only the increase in surface energy but also liquid absorption. SEM and XPS analyses demonstrate that plasma etching accelerated liquid absorption by modifying the surface topography, while the increase in surface energy was due to the addition of oxygen-containing groups. High-valence C=O and O-C=O groups serve as indicators of plasma-induced surface chemical reactions. RF modification primarily features surface etching, whereas GAD significantly increases the active surface groups. Thus, different plasmas, due to their distinct excitation modes, produce diverse modification effects on wood. Considering the various physical and chemical properties of plasma-modified wood surfaces, recommendations for adhesive use on plasma-modified wood are provided.

1. Introduction

Surface modification is one of the key technologies in wood engineering and processing. Traditional methods such as acid–base treatment, silane coupling agents, acrylate treatment, and enzyme treatment, while effective, often damage the wood and rely on the use of organic solvents and chemical reagents [1,2,3]. Hence, there is a substantial demand in both academia and industry for more environmentally friendly and efficient wood surface modification techniques.

Cold plasma has garnered widespread attention as a non-destructive surface treatment method that does not rely on chemical reagents. Although it can only modify the surface of materials and the effects are time-sensitive, its significant surface modification capabilities have led researchers in the field of wood science to extensively explore the wettability and related adhesive properties of plasma-modified wood. Since Bialski et al. first applied plasma for wood surface modification in 1975 [4], more than 50 years of research have ensued. However, this plasma remains at the laboratory stage, far from industrial application. The bottleneck lies in the instability of the modification effects, with different researchers often obtaining entirely contradictory results. Focusing on the most concerning aspects of plasma-modified wood, such as wettability and adhesive performance, many researchers have used RF and dielectric barrier discharge plasma (DBD) under low and atmospheric pressure to treat beech, oak, spruce, and pine veneers. Plasma treatment increases the surface energy of wood, causing the contact angle to decrease with longer treatment times [5,6,7]. However, some studies reported the opposite findings, where the contact angle with liquids remained unchanged or even increased after similar plasma treatment [8,9,10,11]. Traditional adhesive theory posits that plasma treatment improves the wettability of adhesives on wood, and plasma etching enhances the mechanical interlocking effect, thereby improving adhesive performance. Some studies support this hypothesis [5,12,13,14], but many researchers have also confirmed that this conclusion is partial [6,9,10,11]. The reasons for these issues stem from the complexity and variability of plasma excitation methods and wood surfaces. Different plasma excitation methods, atmospheres, equipment parameters, wood species, and sampling positions, even the position within the plasma “flame”, lead to vastly different properties in treated wood samples [15,16,17,18,19]. Researchers generally interpret the macroscopic wetting properties and bonding performance of wood adhesion based on certain microscopic characteristics, such as changes in surface morphology and surface functional groups, following plasma treatment. However, this approach may lead to biased conclusions and mechanisms. Researchers often drew conclusions based on the performance changes of a sole wood species treated with a single type of plasma. This lack of cross-comparison among various plasmas and different wood types inevitably resulted in shallow mechanism research and contradictory conclusions.

To understand the effects of different plasma treatments on wood surfaces, it is crucial to compare the properties of woods modified by different plasmas. Various property changes should be considered as an interconnected whole. This will help to summarize a set of rules that can explain and predict the effects of plasma modification on wood. Then, the applications (such as wood bonding) of plasma treatment could be guided by the clarified mechanism. In the field of wood research, the types of plasma commonly used to study wood modification include RF, dielectric barrier discharge plasma (DBD), GAD, and microwave plasma (MWP). Among these, there are plasmas generated under both atmospheric and low-pressure conditions. Therefore, this study employs GAD and RF plasma, primarily treating poplar, supplemented by black walnut and sapele, to characterize the properties of modified wood. GAD, a cold plasma generated by arc discharge, has a strong ability to induce chemical reactions and is widely used in advanced oxidation reactions for degrading wastewater and waste [20,21,22,23,24]. RF, a typical plasma generated under low pressure, has higher electron energy than other excitation forms and is widely used in metal etching and coating [25,26,27,28]. Poplar is a commonly used fast-growing species in the wood industry, while black walnut and sapele, with their high density and slow liquid penetration, serve as supplements. This study uses the contact angle of modified wood as an entry point, systematically investigating the relationships between contact angle (wettability) and the surface absorption properties, surface morphology, and surface chemical environment of modified wood. How these changes affect the adhesion performance of urea–formaldehyde resin was also examined. We found that different types of plasma resulted in significantly different modification effects. RF can etch the wood surface more, while GAD can greatly increase the number of surface groups. It is not safe to simply relate contact angle change to surface energy change caused by changes in surface chemical groups. The contact angle changes should also be attributed to the liquid absorption property of modified wood. The rapid absorption of liquids by plasma-modified wood leads to inaccuracies in contact angle measurements, which can be corrected using exponential equation fitting methods. The work also provides suggestions for the use of adhesives based on different plasma modification effects.

2. Materials and Methods

2.1. Materials

Urea (U, analytical reagent, AR), formaldehyde (F, 37%, wt%), sodium hydroxide (NaOH, AR), and formic acid (AR) were bought from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). The diiodomethane (AR) was purchased from Macklin Biochemical Co., Ltd. (Shanghai, China).

The poplar (Populus L. spp.), black walnut (Juglans nigra L. spp.), and sapele (Entandrophragma cylindricum Sprague) veneers were purchased from a local hardware market. This image is shown in Figure S1. The veneers were approximately 2.0 mm thick and measured 1200 × 800 mm, produced using the rotary cutting process. Defect-free and impurity-free areas of the wood veneers were selected for use, and a high-speed air gun was employed to remove surface wood chips and dust. The samples were dried in an oven to a moisture content of about 3% for storage and equilibrated to laboratory conditions for 48 h before use.

2.2. Synthesis of Urea-Formaldehyde Resin

Urea–formaldehyde resin was synthesized using a conventional process. The 37% (wt%) formaldehyde solution was adjusted to a pH of 9.0–9.5 using 20% (wt%) NaOH, then heated to 80 °C in a three-neck flask. The first portion of urea (U₁) was added, adjusting the F/U₁ molar ratio to 2.0 and the pH to 9.0–9.5. After stirring for 30 min, the pH of the mixture was adjusted to 4.5–5.0, and the temperature increased to 90 °C, continuing the reaction until the desired degree of polycondensation was achieved. At this point, the pH was adjusted to 9.0–9.5, and the temperature lowered to 60 °C. The second portion of urea (U₂) was added to achieve an F/(U₁ + U₂) ratio of 1.0, stirred for another 30 min, and the pH was adjusted to 7.5–8.0. The final product obtained was UF-1.0.

2.3. Plasma Treatment of Wood Surface with Plasma

The gliding arc discharge (GAD) plasma was performed using a CTD-2000 F plasma (Suman Plasma Co., Ltd., Nanjing, China). The schematic diagram is shown in Figure S2. The GAD plasma used air at atmospheric pressure (86–106 KPa). Its rated power was 1000 W. The GAD was equipped with a couple of blade electrodes whose narrowest distance was 3 mm, and its plasma nozzle was rectangular with sides of 70 mm×20 mm, a temperature of 110~130 °C, and the gas flow rate was 30 L/min. The treated poplar veneer sample is required to be 20 mm away from the nozzle, and the treated time was 10 s. The treated veneers were used within 2 h for bonding performance tests, SEM, and XPS characterizations.

The RF plasma used in this study is the HD-1B plasma (Shitai Plasma Co., Ltd., Suzhou, China). This schematic diagram is shown in Figure S3. The discharge type is capacitively coupled RF plasma. The specifications are as follows. The reaction chamber dimensions were 300 mm, height 500 mm. The working vacuum value was 10–100 Pa, RF power supply frequency was 13.56 MHz. The working rated power was 100 W. The reaction chamber’s working temperature was 60–80 °C. The wood to be treated was placed in the center of the reaction chamber. In this study, the plasma treatment time was 10 s, using O₂ as the atmosphere. The modified samples must be experimentally processed or characterized within 2 h of plasma treatment.

2.4. SEM Characterization of Wood Surface

The morphology of the poplar veneer samples before and after plasma irradiation was characterized by a VEGA3 scanning electron microscopy (SEM) (TESCAN Co., Ltd., Brno, Czech Republic). Semi-quantitative elemental analysis was performed using the EDS probe integrated with the instrument. Before characterization, the sample surfaces were coated with an Au film to enhance conductivity.

2.5. Characterization of Wood Surface with XPS

The instrument used for X-ray photoelectron spectroscopy (XPS) is the PHI 5000 Versa Probe II spectrometer (Physical Electronics, Ismaning, Germany). The samples were characterized using an Al–Kα source (1486.6 eV) in a low-pressure environment of 5 × 10⁻⁷ Pa. An X-ray beam with a spot size of 200 μm size was used at 15 kV. Spectrum fitting and quantitative analysis of the elemental bonding states were performed using Multipak 9.3 software from PHI.

2.6. Tests of Bonding Performance of Wood

Three-layer plywood was made using UF-1.0, with veneer dimensions of 400 mm × 130 mm × 2 mm. The plywood was hot-pressed for 5 min at 1.0 MPa and 120°C, with an adhesive applied to both sides of the core layer. The resulting plywood was conditioned for 24 h at (20 ± 2) °C and relative humidity of 50 ± 5%. The plywood samples were tested for dry bond strength according to China National Standard GB/T 17657–2013 (Test Methods of Evaluating the Physical and Chemical Properties of Wood-Based Panels and Surface-Decorated Wood-Based Panels) [29]. The testing machine used was the WDS-50KN machine (Shimadzu, Kyoto, Japan), and the lap shearing strength was calculated using the following formula (1):

S (MPa) = N (N)/A (mm²)

(1)

N is the maximum force, A is the glued area, and S is the Lap shearing strength.

2.7. Contact Angle Measurement Instruments and Methods

The contact angle was measured using the JC2000D1 (Zhongchen Digital Technology Equipment Co., Ltd., Shanghai, China). This instrument can continuously measure the contact angle evolution over time. The contact angle measurement precision is 0.01°. The contact angles on the surface were measured using distilled water, diiodomethane, and UF as test reagents. All tests selected the tight face of the veneer as the measurement surface. For each veneer, five drops of liquid were tested, and each test item used the same four veneers, repeating the measurement 20 times to obtain the average value. For this study, different equilibrium times for the droplets were selected. For untreated wood samples, the measurement times were 20–25 s after dropping for distilled water and 10–15 s for diiodomethane. For GAD-treated samples, the measurement times were 10–15 s for distilled water and 8–12 s for diiodomethane. For RF-treated samples, distilled water could not be measured, and the measurement time for diiodomethane was 6–9 s. The theoretical basis for selecting the droplet equilibrium time can be found in the section “Equation Fitting Method for Measuring Wood Surface Contact Angles”.

2.8. Calculating Method of Wood Surface Energy

In this study, the OWRK method was used to calculate the surface energy of wood. This method uses the linear calculation formula (2) [30].

$${\displaystyle \frac{{\gamma }_L\left(1+\mathit{cos}\theta \right)}{2\sqrt{{\gamma }_L^d}}}=\sqrt{{\gamma }_S^d}+\sqrt{{\displaystyle \frac{{\gamma }_L^p}{{\gamma }_L^d}}}\sqrt{{\gamma }_S^p}$$

(2)

${\gamma }_L$ is the liquid surface energy, ${\gamma }_L^d$ is the dispersive component of the liquid surface energy, ${\gamma }_L^p$ is the polar component of the liquid surface energy, ${\gamma }_S^d$ is the dispersive component of the solid surface energy, ${\gamma }_S^p$ is the polar component of the solid surface energy, $\theta$ is the equilibrium contact angle. In this study, water and diiodomethane were used as test liquids, with their surface energy components as follows: water (${\gamma }_L^d=21.8 \mathrm{J}·{\mathrm{m}}^{-2}$ , ${\gamma }_L^p=51 \mathrm{J}·{\mathrm{m}}^{-2}$), diiodomethane (${\gamma }_L^d=50.8 \mathrm{J}·{\mathrm{m}}^{-2}$ , ${\gamma }_L^p=0 \mathrm{J}·{\mathrm{m}}^{-2}$) [30].

2.9. Liquid Absorption Mass and Liquid Absorption Rate of Wood Surface

The liquid absorption mass was measured using a METTLER TOLEDO electronic balance, model ME1002E/02, with a precision of 0.01 g. The liquid absorption rate was recorded using an electronic stopwatch with a precision of 0.01 s. To measure the absorption mass, the atmospheric pressure impregnation method was used. Veneers were cut to 30 mm × 10 mm × 2 mm, with the long side parallel to the wood grain. The wood was dried to absolute dryness before impregnation, then fully immersed in the test liquid for 1 min. Using tweezers, the wood was removed and suspended in the air for 15 min to allow surface-adhered liquid to naturally drip off. The mass before and after impregnation was measured using the balance, and the absorption mass ratio was calculated using the Formula (3):

$$A={\displaystyle \frac{m_1-m_0}{m_0}}\%$$

(3)

A is the liquid absorption mass percentage, m₁ is the mass after impregnation, and m₀ is the mass before impregnation. Each group of samples was measured 10 times to obtain the average value.

For measuring the liquid absorption rate, defect-free wood veneer surfaces were selected as the test surfaces. A 5 μL drop of the test liquid was placed on the wood surface, marking the initial time. The change in the shape of the droplet was observed under the contact angle measurement instruments. The end time was recorded when the droplet was completely absorbed. For liquids like diiodomethane, which maintain their droplet shape for a long time (over 10 min) after reaching a certain absorption amount (judged by the contact angle remaining unchanged), the time from the initial placement to the moment the contact angle became constant was taken as the end time. This method for measuring liquid absorption mass and absorption rate references a series of studies [31,32,33,34].

3. Results

3.1. Equation Fitting Method for Measuring Wood Surface Contact Angles

Measuring the equilibrium contact angle is an important method for evaluating the wettability of solid surfaces. However, wood has strong hygroscopicity, causing the test liquid to be absorbed by wood. Therefore, most research literature uses the contact angle after the liquid has been stationary for a few seconds to several tens of seconds as the equilibrium contact angle [35]. However, after plasma modification, the hygroscopicity of the wood surface is further enhanced, making contact angle measurement inaccurate. This study used equation fitting to accurately determine the contact angle measurement time [36,37,38].

In practical measurement, the liquid on the wood surface behaves as a time-dependent piecewise function. In the first stage, the liquid reaches equilibrium due to the surface tensions of the solid and liquid, with the net external force minimizing the surface energy, thus requiring a short time. In this stage, only a small amount of liquid is absorbed by the wood, having minimal impact on the contact angle. The main factor affecting contact angle is the deformation of the droplet due to attractive forces between the solid and liquid, which represents the wettability of the wood surface. The second stage involves the diffusion and absorption process, where the absorption of the liquid by the wood becomes the main factor affecting the contact angle. The decrease in the contact angle is due to the reduced total amount of liquid on the wood surface, which no longer conforms to the definition of the equilibrium contact angle. Therefore, defining the contact angle after a long stationary period as the equilibrium contact angle in some studies is contentious. The greatest difference between the two stages in the angle change function is the decay coefficient k. Therefore, this work expresses the wood contact angle as a piecewise function.

$$\theta ={\theta }_0+A{e}^{-k_1{t}_1},0<t_1<t_e$$

(4)

$$\theta ={\theta }_e{e}^{-k_1(t_2-t_{\mathrm{e}})},t_e\le t_2<t_3$$

(5)

$$\theta =B{e}^{-k_2(t_3-t_2)},t_2<t_3$$

(6)

In this function, t_e is the time at which the equilibrium contact angle is reached, t₂ is the time when liquid absorption and diffusion, and t₃ is the time after the liquid reaches steady absorption when the absorption rate changes. Equations (4) and (5) form a continuous exponential curve, while Equation (6) forms another exponential curve. The domain of whole exponential curves is t ∈ (0,∞), but the connection point at t₃ between the two exponential curves is discontinuous; the decay coefficients are k₁ and k₂ respectively. Equation (4) represents the first stage of the liquid-wetting wood surface, Equation (5) represents the second stage, where the wood steadily absorbs liquid after reaching the equilibrium contact angle, and Equation (6) represents the stage of accelerated liquid absorption or reaching a new equilibrium. In this paper, the stage of accelerated liquid absorption in Equation (6) is not the focus of the study and is, therefore, not addressed.

Upon substituting the experimental data to fit the exponential function, it was found that t_e is a time interval. In Figure 1, the function fitting curves for poplar, black walnut, and sapele wood with distilled water and diiodomethane are presented. The R-square values for all six graphs are closer to 1, indicating good fitting results. From the figures, it can be observed that the rate of change in the contact angle of distilled water in the 20–25 s time interval is significantly different from that in the 0–20 s interval, and the contact angle tends to stabilize after 25 s. Therefore, the average contact angle measured multiple times in the 20–25 s interval can be considered the equilibrium contact angle for distilled water. Similarly, the measurement time interval for diiodomethane, as shown in the figures, is 10–15 s. In subsequent measurements, the contact angles of all plasma-treated wood samples were measured at the time intervals using this method, as detailed in Section 2.7 of the experiments. Then, the measurements were averaged to obtain the equilibrium contact angle.

3.2. Effect of Plasma Treatment on Wood Wettability

The equilibrium contact angles of poplar were measured with distilled water and diiodomethane before and after plasma treatment.

Table 1. Contact angle and surface energy of poplar before and after plasma treatment.

Plasma Treatment Method	Contact Angle (°)		Surface Energy (Jm−2)
	Water	Diiodomethane	${\mathit{\gamma }}_{\mathit{S}}^{\mathit{d}}$	${\mathit{\gamma }}_{\mathit{S}}^{\mathit{p}}$	${\mathit{\gamma }}_{\mathit{S}}$
Ref.	81.03 ± 7.28	45.81 ±3.91	36.57	3.75	40.32
GAD	32.87 ± 4.17	20.65 ± 2.16	47.58	23.69	71.27
RF	0	16.03 ± 2.07	48.84	31.63	80.47

shows the values. The equilibrium contact angle of untreated wood is around 80° for distilled water and 45° for diiodomethane, indicating that the tested poplar wood itself cannot be well wetted by water or is relatively hydrophobic.

After plasma treatment, the highly polar distilled water (polarity parameter 10.2) became wettable on the wood surface, and the completely non-polar diiodomethane (polarity parameter 2.8) also became wettable on the wood surface, which is confusing. Table S1 lists the contact angle data for different woods before and after plasma treatment. The phenomenon of both polar and non-polar liquids becoming wettable was observed from three types of wood modified by two types of plasma. According to previous research, the reason for this phenomenon is that plasma treatment increases the surface energy of the wood, significantly enhancing both polar and non-polar surface energy [5,13,39]. Detailed data are provided in Table 1 and Table S2.

Calculations based on measured contact angles and OWRK theory suggest that after RF plasma treatment, the total surface energy of poplar doubled, with polar and dispersive surface energy increased by 800% and 33%, respectively. GAD-treated poplar showed a smaller increase in surface energy than RF-treated poplar, but the total surface energy, polar surface energy, and dispersive surface energy increased by 76%, 532%, and 30%, respectively, compared to untreated samples. The increase in surface energy resulted in reduced contact angles for liquids on plasma-treated wood, causing the test liquids to spread more evenly over the solid surface, as shown in Figure 2 for two test liquids on poplar at time t_e. Black walnut and sapele exhibited similar behavior to poplar, as shown in Table S2, consistent with the results in most of the literature [40,41]. For homogeneous, smooth, and non-porous surfaces, changes in contact angle can correspond to changes in surface energy and reflect changes in wettability. However, for porous materials like wood, directly using surface energy theory to relate to wettability and concluding that the activation capability of GAD is weaker than RF are likely incorrect.

Using surface energy to explain the enhanced wettability of wood by plasma treatment presents a paradox: the OWRK equation calculates surface energy based on contact angles and then uses the calculated surface energy to explain changes in contact angles, which is circular reasoning. The reality is that the major factor causing the drastic reduction in contact angle after plasma treatment is the absorption of liquid by the wood. Liquid absorption by wood results in contact angles smaller than those determined by surface energy, making the calculated surface energy appear larger. Although this work corrects the equilibrium contact angle measurement time t_e to eliminate the influence of liquid absorption, plasma treatment accelerates liquid absorption by wood, amplifying the measurement deviation in contact angles and making contact angle characterization of wettability for plasma-treated wood inaccurate.

This study used two indicators to characterize the liquid absorption properties of wood. First, the absorbed liquid mass of wood samples of the same size within a fixed time (1 min).

Table 2. Absorption mass and absorption rate of poplar before and after plasma treatment.

Plasma Treatment Method	Absorption Mass		Absorption Time (S, 5 μL Each Time) and Contact Angle
	H2O	CH2I2	H2O			CH2I2
			${\mathit{\theta }}_0(°)$	${\mathit{\theta }}_{\mathit{t}}(°)$	Time	${\mathit{\theta }}_0(°)$	${\mathit{\theta }}_{\mathit{t}}(°)$	Time
Ref.	30.1 ± 1.7%	15.7 ± 0.8%	83.07 ± 5.45	0	339 ± 27	45.29 ± 3.87	23.25 ± 1.46	155 ± 23
GAD	37.0 ± 0.9%	22.8 ± 1.1%	35.7 ± 3.98	0	122 ± 21	23.85 ± 2.65	12.39 ± 1.02	128 ± 12
RF	50.2 ± 2.6%	47.1 ± 1.0%	7 ± 1.22	0	3 ± 1.2	17.2 ± 2.19	6.26 ± 0.44	56 ± 14

shows the absorption mass of poplar for distilled water and diiodomethane, absorption mass for black walnut and sapele are shown in Table S3. Second, the absorption rate is characterized by the time required for a droplet of the same volume (5 μL) to reach a stable state on the wood surface, and the contact angle was measured to describe the droplet’s shape, providing additional information on the droplet’s behavior. Table 2, Tables S4 and S5 list the times and contact angle data for distilled water and diiodomethane droplets to reach saturation absorption in all samples.

Table 2, Tables S4 and S5 show that GAD-treated wood samples have higher absorption mass and faster absorption rate, with absorption mass increasing by about 15%–30% compared to untreated wood samples, and absorption times halved. RF-treated wood samples showed much higher absorption mass and rate than both untreated and GAD-treated samples, indicating that RF has a greater impact on liquid absorption than GAD, partially disproving the notion that GAD is less active than RF; GAD’s effect on contact angle is smaller than RF’s, primarily due to the stronger absorbency of RF-treated samples. RF-treated wood samples absorbed distilled water very quickly, with all wood species absorbing water droplets within 5 s, reducing the contact angle to 0°. The absorption mass of distilled water in poplar increased by 66% compared to untreated wood, while sapele showed a threefold increase. The absorption characteristics of diiodomethane in Table 2 show a similar trend, but the rate increase is less pronounced than with distilled water. This is because RF treatment significantly increased surface energy, especially polar surface energy. RF-treated samples quickly absorbed water and diiodomethane, which makes the calculated surface energy higher than the actual state; this is because, if no liquid were absorbed, the contact angle changes used as the basis for surface energy calculations would not be so small. Although contact angle measurements are widely used in wood research to characterize surface energy and wettability [15,39], the above analysis shows that this method cannot accurately distinguish whether a reduction in contact angle is due to increased surface energy or liquid absorption. Therefore, if absorption is involved, the calculated surface energy is overestimated, particularly in cases where plasma treatment increases both the absorption rate and mass of liquid. Using contact angles to calculate surface energy under such conditions introduces significant errors. This also highlights the necessity of using the exponential equation in Section 3.1 for selecting the contact angle measurement time when the liquid is rapidly absorbed by the wood.

3.3. Effect of Wood Morphology Changes on Absorption Characteristics

The micro-scale effects of plasma wood modification are primarily reflected in the etching of the wood’s microstructure and changes in surface groups. These two microscopic changes both affect the surface wettability and absorption characteristics of wood for liquids, but the changes in physical morphology have a greater impact on transport properties such as absorption amount and absorption rate [7,42,43]. Figure 3 shows SEM images of poplar samples. The surface of the untreated wood sample is smooth, with clear pore boundaries and intact structures without any coverings. No cracks or gaps were observed in multiple images of untreated wood samples.

After GAD treatment, the most noticeable etching phenomenon is the lamellar peeling of the wood’s surface structure. This peeling phenomenon is clearly visible in the GAD-8K figure with the yellow circle in Figure 3. None of the GAD-treated samples showed uniform wood cracks. The average energy of active particles in GAD is only about 2 eV, which is insufficient to cause significant etching marks through impact, but its ability to induce oxidation reactions far exceeds that of other plasma discharge types, as discussed in numerous studies [21,44]. Combined with subsequent EDS and XPS surface analyses, it can be inferred that active particles in GAD cause oxidation reactions on the wood surface, leading to the preferential decomposition of lignin, which forms O-C=O, O-C-O, and C=O bonds through oxidation. This indicates that the peeling phenomenon in GAD is, in fact, the result of lignin decomposition on the wood surface, leaving these groups bonded to the wood surface, containing many oxygen-containing groups [11,15,16]. This surface peeling phenomenon is caused by the oxidation reaction initiated by active substances by GAD. Although it appears similar to etching on the surface, it can be referred to as “chemical etching”. Figure S4 shows similar etching marks on black walnut and sapele, indicating similar regular surface peeling phenomena.

After RF treatment, pyramidal particles appeared in the green circle of RF-8K in Figure 3, with particle sizes ranging from 2 to 2.5 μm. Such large particles cannot be products of oxidation reactions under the plasma treatment conditions in this study but are rather fragments caused by the impact of active particles from the plasma. Additionally, RF treatment caused cracks on the wood surface, as shown by the area in the red circle of RF-1K in Figure 3. The RF-treated sapele, shown in Figure S5, also exhibited widespread spherical and cubic particles. These phenomena are examples of RF’s “physical etching” of poplar. This is related to the emission mechanism of RF. Active particles generated in a low-pressure environment have a longer mean free path due to high vacuum, allowing greater acceleration and kinetic energy. According to the literature, RF electron temperatures typically range from 3 to 5 eV [45,46,47]. Such high-energy particles cause mechanical damage to the wood interface upon impact. Therefore, RF’s effect on wood is more related to physical etching, whereas GAD primarily changes the chemical environment of the wood interface through oxidation reactions. RF modification also involves oxidation reactions, with the adhesion of particles being oxidation products, too; the relevant evidence is discussed in detail in the subsequent EDS spectral section.

From the above plasma etching phenomena, it is inferred that GAD alters the surface material of wood through oxidation reactions. This change has a minimal impact on the absorption of liquid on the surface. The increased liquid absorption in GAD-treated wood is mainly due to the increase in polar surface energy due to added oxygen-containing groups. Consequently, GAD-treated wood absorbs more liquid than untreated wood, but without creating new transport pathways, the absorption mass only increases by about 20%. Differently, RF damages the wood surface, altering its morphology and connecting some pore structures in veneer, resulting in larger surface cracks and pores, creating new liquid storage spaces and transport channels. This effect is far stronger than the liquid absorption effect caused by increased polar surface energy. Therefore, RF-treated samples complete water absorption within seconds, with diiodomethane absorption mass increasing by 1–3 times and absorption rates significantly higher than those of untreated and GAD-treated samples.

3.4. EDS Analysis of Plasma-Treated Wood Surfaces

EDS characterization was performed on poplar samples, and Figure 4 shows the specific locations on the poplar subjected to EDS analysis, covering 7 positions on 3 samples. Since the untreated wood sample had a uniform surface, only 2 test points were selected.

From the data in

Table 3. C/O ratio table for each EDS characterization area on poplar.

Characterization Area	C (%)	O (%)	O/C
Ref.-1	53.53	41.29	0.8
Ref.-2	52.97	43.57	0.8
GAD-1	50.39	41.88	0.8
GAD-2	47.41	47.03	1.0
GAD-3	24.91	65.12	2.6
RF-1	34.82	54.60	1.6
RF-2	20.09	63.86	3.2
RF-3	53.93	37.29	0.7

, it can be seen that the oxygen-to-carbon atom ratio at the two test points on the untreated poplar sample is approximately O/C = 0.8. The number of carbon atoms on the surface of the untreated poplar is greater than the number of oxygen atoms, which is consistent with the subsequent XPS characterization, and the data at the two test points are basically consistent, indicating uniform surface material.

For the GAD-treated samples, Test Point 1 is located in an area without etching near the etched patches, with an O/C ratio of 0.8, similar to the untreated poplar sample’s test points. This suggests that no etching or oxidation reaction occurred at this point, and the carbon-to-oxygen ratio is consistent with that of the untreated sample. Test Point 2, as shown in the figure, is partially etched, with the O/C ratio roughly changing to 1:1. The increase in oxygen atoms indicates an increase in oxygen-containing groups, signifying that an oxidation reaction has occurred. Test Point 3, located in a fully etched area, shows a significant peeling of surface material, with an O/C ratio of 2.6. The reversal of the carbon-to-oxygen ratio indicates that the substance at this test point is a reaction product, as cellulose, hemicellulose, and lignin in wood cannot have such a carbon-to-oxygen ratio. The data from the three test points validate the analysis of the SEM images: GAD treatment results in chemical etching on the poplar surface, and this oxidation reaction is progressive, with varying degrees at different locations, implying that prolonged treatment time may deepen the reaction. Similar conclusions were drawn by Jamali, Galmiz, and Talviste [15,16,48].

For the RF-treated samples, Test Point 1 is on a scattered particle, with an O/C ratio of 1.6, also showing a reversal of the carbon-to-oxygen ratio. However, the oxygen content in these particles is significantly lower than in Test Point 3 of the GAD samples, indicating fewer oxygen-containing groups. This suggests that oxidation reactions occurred during the high-speed particle impact but to a lesser extent than in the GAD-induced oxidation reaction. At Test Point 2, there is an accumulation of fallen particles, and SEM shows a certain degree of melting, with an O/C ratio of 3.2. Although both Test Points 1 and 2 consist of the same plasma-induced fallen particles, their O/C ratios differ significantly, indicating that the oxidation reaction induced by RF is not uniform. Test Point 3 shows no visible etching on the wood surface despite being very close to Points 1 and 2, with an O/C ratio still at 0.7, indicating no changes occurred at this point.

From the EDS characterization, it can be inferred that plasma-induced etching and oxidation reactions occur simultaneously. The etched areas show an increase in oxygen content, and the degree of etching is closely related to the oxygen content, indicating that plasma-induced etching has two aspects: “physical etching” from high-energy particle impacts and “chemical etching” from active particle-induced oxidation reactions. However, the modification effects induced by different plasma emission methods vary, emphasizing different aspects of modification processes.

3.5. XPS Analysis of Plasma-Treated Wood Surfaces and Its Relation to Wettability Changes

Figure 5 shows the XPS characterization of poplar samples, and

Table 4. XPS data of the O/C ratio and the components of the deconvoluted C1s spectra of the poplar after plasma treatment.

Plasma Treatment Method	O/C	Relative Content(%)
		C1(C-C) 284.75 eV	C2(C-O) 286.37 eV	C3(C=O) 287.48 eV	C4(O-C=O) 288.6 eV
Ref.	0.33	58.99	29.73	7.06	4.23
GAD	0.74	17.82	45.22	24.32	12.65
RF	0.42	57.42	27.68	9.79	5.12

provides the peak fitting data from XPS analysis. From the full spectrum of XPS, it can be seen that the oxygen content on the surface of poplar significantly increased after plasma treatment compared to untreated samples. The O/C ratio reversed, with carbon three times the oxygen in untreated poplar, while RF treatment increased oxygen by 30%, and GAD treatment increased it by 110%. This indicates that GAD induces oxidation reactions more effectively than RF. The XPS data align with the trends in the oxygen/carbon ratio observed in the preceding EDS analysis. The SEM, EDS, and XPS results collectively point to the fact that plasma treatment increases the proportion of oxygen-containing groups on the surface, with the extent of the increase varying depending on the type of wood and plasma. Similar conclusions have been drawn by Talviste, Novák, Král, and others, and these findings are corroborated by many researchers [13,15,39].

Research suggests that the increased oxygen is mostly bonded with carbon atoms in high-valence states [13,15]. In Table 4, for untreated poplar, C-C (284.75 eV) and C-O (286.37 eV) are the main bonding forms on the surface, primarily from hydroxyl groups in ether structures. After RF treatment, C-C (284.75 eV) and C-O (286.37 eV) remain the main bonding forms, with proportions similar to untreated wood. However, the proportion of C-O bonds decreases by 2.05%, while C=O (O-C-O) (287.48 eV) increases by 2.73% and O-C=O (288.6 eV) by 0.89%. O-C-O mainly exists in polysaccharides, while C=O is almost nonexistent in polysaccharides and is a carbonyl or aldehyde group from oxidation, and O=C-O is a carboxyl group, which is absent in natural wood; this implies that the more carboxyl groups (O=C-O) present, the more intense the oxidation reaction. Therefore, RF treatment does not significantly alter the chemical environment of the wood surface, consistent with the conclusions of Sven, Andrés, and others [17,18]. The likely reason is the high vacuum environment (20 Pa) in RF, where the gas molecule density is only one hundred-thousandth of normal atmospheric pressure, resulting in fewer active particles and, thus, a lower probability of oxidation reactions. Additionally, the macroscopic environment temperature of RF is around 60 °C, further reducing the likelihood of chemical reactions.

The C-C peak is most prominent in untreated poplar. After GAD treatment, the C1S peak of C-O becomes dominant. GAD treatment significantly increases the bonding forms of C=O (O-C-O) and O-C=O by 17.26% and 8.42%, respectively. Other plasma excitation forms increase these bonds by less than 5%. GAD modification results in a several-fold increase in surface oxygen-containing groups, especially high-valence groups. Given that the proportion of O-C=O is a measure of the extent of plasma-induced chemical reactions, it demonstrates the high chemical reactivity of GAD treatment. This is the fundamental difference between GAD and other plasmas. GAD’s high chemical reactivity is due to its atmospheric environment, with active particle densities reaching 10¹³~10¹⁵ cm⁻³, leading to a higher probability of oxidation reactions. GAD excitation is caused by arc discharge, a typical thermal plasma, which raises the gas temperature at the nozzle to about 320 °C, further enhancing the rate and likelihood of oxidation reactions. The suitability of GAD for advanced oxidation reactions has been demonstrated in research and industrial [22,23]. The strong oxidation-inducing capability of GAD is also confirmed by GAD-treated black walnut, as shown in Figure S6 and Table S6. The surface group changes in black walnut after GAD and RF treatment are consistent with those in poplar, further supporting the experiment for the above discussion.

Comparing the physicochemical properties of wood treated by RF and GAD, RF treatment resulted in significant cracks but did not greatly change the surface chemical groups. In contrast, GAD modification significantly increased various oxygen-containing groups, and the introduction of highly reactive surface groups greatly enhanced the surface reactivity. The modification effects of different plasmas on wood are not consistent, and simply comparing a few parameters to measure plasma activity or infer its impact on the macroscopic properties of wood is incomplete. This is the reason why conclusions regarding plasma-treated wood are not uniform.

3.6. Analysis of Bonding Property of Plasma-Treated Wood

The effects of plasma modification result from the combined influence of various physicochemical properties; therefore, UF adhesive was used to bond poplar wood to test the bonding strength. Based on wood bonding theory [19,49,50], plasma treatment-induced surface etching is beneficial for forming adhesive nails in wood bonding, and the increased surface energy can enhance the adhesion of the infiltrated adhesive to the wood surface. These characteristics are favorable factors for improving bonding strength.

Table 5. Contact angle and absorption mass of plasma-modified poplar for UF.

Plasma Treatment Method	Contact Angle (°, UF-F/U = 1.0)	Absorption Mass
Ref	85.84 ± 6.42	16.6 ± 0.8%
GAD	37.15 ± 4.13	26.9 ± 1.1%
RF	24.59 ± 2.88	47.3 ± 1.0%

presents the contact angles and absorption characteristics of plasma-modified wood for UF adhesive. Figure 6 shows the shear strength data from 10 bonding experiments, where the bonding strength of plasma-treated poplar fluctuates around the average value of untreated poplar. Multiple experiments failed to conclude that bonding strength was consistently improved. The reason can be that RF etching produces fragments and grooves, while GAD only peels off the surface layer, leading to inconsistent liquid absorption by the wood. After GAD treatment, the increase in highly polar C=O and O-C=O groups far exceeds that of RF-treated samples several times. Additionally, different surface groups have significantly varying bonding properties with adhesives. For example, it is well-recognized that epoxy and isocyanate adhesives easily form chemical bonds with R-OH groups, while urea–formaldehyde resin mainly forms hydrogen bonds with oxygen-containing groups. However, the bonding effect caused by hydrogen bonds is much weaker than that of chemical bonds [19,49,51]. The excellent modification effects did not translate into superior bonding characteristics, further illustrating that plasma modification is a comprehensive reflection of multiple property changes, and wood bonding performance cannot be linearly connected to modification effects. Differences in modification results can affect adhesive properties to varying degrees, resulting in significant variations in bonding effects for plasma-treated wood. Therefore, it is extremely simplistic to conclude that plasma can enhance wood bonding based solely on its surface modification effects.

Can the properties of plasma-modified wood be used to improve bonding? This work proposes that the targeted use of adhesives according to the macro and microscopic properties of plasma-modified wood for targeted bonding is a feasible solution. For example, if the introduced new groups have the potential to form bonds with adhesive groups, matching the introduced group types, quantities, and bonding forms with sensitive adhesives can achieve enhanced bonding strength facilitated by plasma. In another study by the author [52], GAD-treated poplar was used to construct a surface rich in active groups (activated wood surface, AWS), which was combined with epoxidized soybean oil (ESO) and urea–formaldehyde resin (UF) to form an ESO–UF–AWS ternary co-curing system that exhibited significantly improved bonding strength. Another example is the use of etching effects to aid bonding performance. Bamboo, due to its unique structure, has weak liquid penetration and absorption, leading to poor bonding performance with conventional adhesives like UF. Using plasma to etch the bamboo surface not only increases surface roughness but also enhances penetration. This should also improve bonding characteristics, as Wu et al. confirmed [53]. In summary, using plasma to improve wood bonding performance must fully consider the plasma, wood bonding surface, and adhesive properties to achieve targeted bonding through good coordination among the three parties. This is an effective way to enhance bonding performance using plasma.

4. Conclusions

The effects of plasma modification on wood vary significantly depending on the type of plasma excitation. Even though plasma treatments generally increase the wettability of modified wood, the underlying mechanisms differ, which in turn affects the interaction between the modified surface and adhesives, influencing bonding performance. This study has drawn the following specific conclusions:

(1)

Due to differences in the number and energy of high-energy particles generated by plasma excitation, different types of plasma have varying effects on wood modification. RF plasma can significantly etch the wood surface, creating noticeable grooves and micro-particles. Differently, GAD plasma greatly increases the number of surface groups, especially oxygen-containing active groups, resulting in different surface properties.

(2)

Plasma modification can reduce the contact angle of liquids on the wood surface, but the reasons for this reduction differ. For RF, the primary reason for the decreased contact angle is increased liquid absorption, whereas for GAD, it is the increase in surface oxygen-containing groups.

(3)

Plasma modification causes liquids to be rapidly absorbed by the wood surface, making contact angle an inappropriate method for measuring wood surface wettability. For rapidly changing contact angles, a parameter equation using continuously measured values and determining the equilibrium contact angle measurement time t_e can be employed to obtain relatively accurate contact angle measurements.

(4)

Plasma modification of wood has the potential to improve bonding performance, but suitable adhesives must be matched to the specific characteristics of the plasma modification to achieve enhanced bonding strength.