Directional anisotropy of wood surface wettability: morphological dynamics of droplet behavior observed from different directions

2.1 Materials

Douglas fir heartwood wood was selected as the experimental material. Samples were taken from a height of 3 m above the tree base, targeting the 18th to 22nd growth rings outward from the pith. Defect-free specimens were cut into blocks with dimensions of 20 × 20 × 20 mm (longitudinal × radial × tangential, L × R × T). The tangential surfaces of the specimens were sectioned using a microtome (Leica RM 2245, Wetzlar, Germany) to expose EW, TW or LW, as shown in Figure 1a. After sectioning, the surfaces were further refined by sanding with 600-grit abrasive paper. The specimens were then air-dried under ambient conditions (65 % relative humidity, 20 °C) over one week, reaching in an air-dried density of approximately 500 kg/m³. Deionized water and diiodomethane (analytical grade, 99 %) were used as liquids to assess the wettability of wood, as well as the surface free energy.

Figure 1:

Schematic diagram of droplet wetting on EW, TW, and LW regions (a), and the experimental setup for measuring contact angles in the parallel (ǁ) and perpendicular (⊥) directions on wood samples (b).

2.2 Dynamic wettability test

The wettability of tangential surfaces of Douglas fir wood was evaluated using a contact angle measurement system (FCA2000A, Shanghai Aifeisi, China) via the sessile drop method. Droplet of varying volume (9, 12 or 15 μL) was placed onto EW, TW, or LW tangential surfaces. During testing, wood specimens were secured on the platform, and droplets were precisely deposited using a microsyringe, as shown in Figure 1b.

To capture the morphological change of the droplets, two filming system were used. The first camera recorded the droplet spreading along the wood grain (longitudinal direction), and the corresponding contact angle was denoted as θ^ǁ. The second camera captured the droplet spreading perpendicular to the grain (tangential direction), and the corresponding contact angle was denoted as θ^⟂, as illustrated in Figure 1b.

The experiment was repeated 5–10 times for each droplet size and wood tissue type. The morphology and contact angle of the droplets were observed and recorded at 2-s intervals over a period of 60 s. Due to the anisotropic nature of wood, droplet base diameters varied along different grain directions. The recorded images were analyzed using ImageJ software (https://imagej.nih.gov/ij/) to measure the base diameter (d) of and the height (h) of the droplet, as shown in Figure 2. The rate of change in droplet base diameter over time (k _t ) was calculated using Equations (1) and (2):

where d t ǁ and d t ⟂ are the droplet base diameter in the direction parallel and perpendicular to the wood grain at time t (s), respectively. d 0 ǁ and d 0 ⟂ represent the initial base diameters in those respective directions.

Figure 2:

Top-view and side-view schematics of the variation in the droplet base diameter during spreading.

The droplet volume was calculated based on the droplet’s base diameter and height. To account for the anisotropic nature of the wood, both directions of base diameters were considered in the volume calculation. The average diameter ( d ‾ ), as described in Equation (3) was used for this purpose (Chen et al. 2005). During the wetting process, the volume of the droplet (V) on the tangential surfaces at different times t was determined using Equation (4):

The permeability (V_Pt) of the droplets was calculated according to the following equation:

where V₀ is the initial volume of the droplet (9, 12 or 15 μL, depending on the experiment).

2.3 Surface free energy calculation

The surface free energy was calculated using the geometric mean method (OWKES method) (Owens and Wendt 2003). This approach requires two reference liquids to establish a system of equations that allow for the determination of both the polar and dispersive components of the surface energy. By substituting the surface free energy γ_L (mN/m), polar component γ L p (mN/m), and the dispersive component γ L d (mN/m) of water and diiodomethane, along with the initial contact angle ( θ 0 ǁ or θ 0 ⟂ ) into Equation (6) or (7), the polar and dispersive components of the sample in the parallel or perpendicular direction can be calculated. The total surface free energy γ_s (mN/m) in each direction is then calculated by summing the dispersive and polar components, as expressed in Equations (8) and (9):

where γ S d (mN/m) is the dispersive component of the solid surface, γ s p (mN/m) is the polar component of the solid surface, and the superscripts “ǁ” and “⊥” denote the parallel and perpendicular directions, respectively.

2.4 Anatomical characterization

A digital microscope (VHX-7000, Keyence, Japan) was used to capture cross-sectional images of the samples at a magnification of 300. The measurements included the cell width (d₁) and lumen width (d₂) of tracheid cells from EW, TW or LW in both the R and T directions. The obtained data were then incorporated into Equation (10) to calculate the ratio of cell wall (CWR), providing a quantitative assessment of the structural characteristics of different growth regions. Additionally, at a magnification of 400, the surface roughness (R_a) of the tangential section was measured to further evaluate its microstructural properties.

2.5 Significance analysis

Data were analyzed using SPSS version 27.0 (SPSS Inc, Chicago, IL, USA) statistical software to analyze the effects of tissue type, droplet volume and observed direction on contact angle, permeability as well as surface free energy by Duncan’s multiple comparison test (p = 0.05).

3.1 Morphology of droplet during wettability

Figure 3a shows a top-down view of the morphological changes of 9 μL droplets during the wetting process on the tangential surfaces of EW, TW and LW. Due to the limited imaging resolution of the equipment, the “0 s” frame in Figure 3a does not perfectly represent the initial moment of droplet contact with the wood surface. It can only be considered as an approximation of the initial state, as the droplet was already slightly deformed and not perfectly circular at that time point. As wetting time increased, the droplets showed obvious directional spreading especially along the longitudinal direction of EW and TW surfaces, in contrast to the more uniform spreading observed on LW surface. This anisotropic spreading behavior was recorded by the side view images of the droplets captured by the contact angle measurement (Figure 3b). These images illustrated the great spreading along the grain direction (parallel) compared to the perpendicular direction, behaving as larger base diameter in the parallel direction (d^ǁ) than in the perpendicular direction (d^⟂). This phenomenon was particularly noticeable in EW and TW, where droplets spread more effectively along the longitudinal direction. This can be attributed to the tube-like structure of tracheid. When the water was deposited on the wood surface, it would be spreading more efficiently through the continuous tracheid lumens along the longitudinal direction. In contrast, in the perpendicular direction, water diffused through cell wall and lumen subsequentially, as well as through pits between tracheids and ray cells. The resistance to spreading was lower in the longitudinal direction, where water followed a more direct path through the continuous lumens (Siau 2012). The differences between tissue types further explained the observed spreading behavior. Due to the larger lumens and thinner cell walls of EW tracheid cells, droplets in EW surface spread more rapidly along the longitudinal direction, leading to more pronounced anisotropic spreading.

Figure 3:

Top-view (a) and side-view (b) images illustrating the spreading behavior of droplets on EW, TW, and LW regions over time (0 s, 6 s, 10 s, 30 s, and 60 s). Time-dependent changes in droplet base diameter rates in the parallel (ǁ) and perpendicular (⟂) directions for EW (c), TW (d), and LW (e).

Figure 3c–e shows the change rate of base diameter in both parallel and perpendicular directions relative to the wood as a function of wetting time. In these figures, the change rate of the base diameters initially increased rapidly within the first 20 s, then stabilized over the subsequent 40 s. Consistently, for all tissue types and droplet volumes, the change rate of the base diameter in the parallel direction was greater than that in the perpendicular direction. This observation aligned with the results presented in Figure 3a.

The greatest increase in base diameter was observed in EW (Figure 3c), followed by TW (Figure 3d), and the smallest increase was found in LW (Figure 3e), irrespective of the observation direction or droplet volume. As mentioned previously, the larger cell lumens in EW and TW facilitated easier penetration of droplets, resulting in faster and broader spreading. In contrast, the thick cell wall of LW hindered water penetration, leading to smaller changes in droplet diameter (Panshin and Zeeuw 1981).

When droplet volume increased, no clear pattern emerged in the change of the droplet diameter across different tissue types, nor in the differences between the perpendicular and parallel directions. This likely reflected the structural differences among the tissue types (EW, TW and LW), which led to complex interactions between the wood and water during the wetting process.

3.2 Time-dependent contact angle

Figure 4a–c illustrate the changes in contact angle over wetting time for droplets of different volumes (9, 12 and 15 μL) on different tissue types (EW, TW and LW), measured in both parallel and perpendicular directions relative to the wood grain. Regardless of droplet volume, contact angle on EW and TW showed more significant reductions than on LW. The dynamic contact angles were lower on EW surface compared to LW surface, likely due to variations in surface roughness and differences in chemical composition (Scheikl and Dunky 1998). At all wetting times, the contact angle in the parallel direction (θ^ǁ) was consistently smaller than that in the perpendicular direction (θ^⟂), irrespective of tissue type or droplet volume. Chen et al. (Chen et al. 2022) pointed out that changes in contact angle represented a trade-off between surface spreading and penetration. Water spread more effectively through continuous tracheid lumens, explained the smaller contact angles observed in the parallel direction.

Figure 4:

Variation of droplet contact angles over time in the parallel (ǁ) and perpendicular (⟂) directions for different droplet volumes: 9 (a), 12 (b) and 15 μL (c) on EW, TW, and LW.

The difference in contact angles θ^ǁ and θ^⟂ after 60 s of wetting was calculated (Figure 5). Statistical analysis revealed that droplet volume significantly affected the contact angle difference, except for the 9 and 12 μL droplets on LW surface. As droplet volume increased, the contact angle difference diminished. Larger droplets enhanced surface tension in both the parallel and perpendicular directions, behaving as less anisotropic spreading behavior (Vafaei and Podowski 2005). These findings highlighted the importance of considering the measurement direction when evaluating the wettability of wood and its products. A contact angle difference of around 10° could lead to over- or underestimating wood wettability, depending on the observed direction.

Figure 5:

Differences in contact angles between the parallel (ǁ) and perpendicular (⟂) directions for different tissue types and droplet volumes (9–15 μL). *p < 0.05, n.s.: no significant difference.

3.3 Penetration ratio

When a water droplet contacted the wood surface, both surface spreading and penetration occurred simultaneously. The wicking of the water into the wood through capillary action reduced the droplet volume above the wood surface (Wang et al. 2015). Previous studies (Rafsanjani et al. 2014; Xiao et al. 2024) calculated the penetration ratio assuming uniform spreading across the droplet’s radius. However, as demonstrated in the literatures (Shi and Gardner 2001) and in Section 3.1, the droplet spread non-uniformly, regardless of droplet volume or tissue type. Therefore, more accurate calculations of droplet volume should consider the variation in droplet radius during wetting (Comstock 1970; Wang et al. 1991). Figure 6a–c illustrated the time-dependent penetration ratio, considering both the parallel and perpendicular diameters ( d t ǁ and d t ⟂ ). The penetration ratio increased most rapidly during the first 20 s, followed by a slowdown, likely due to the diminishing driving force for droplet diffusion over time, with resistance within the liquid penetration pathways progressively becoming the dominant factor (Burridge et al. 2019). Throughout the wetting, the penetration ratio showed dependence on droplet volume and tissue type. For droplets of 9 μL (Figure 6a) or 12 μL (Figure 6b), there was no significant variation in penetration volume between EW and TW after 60 s. However, when droplet was 15 μL, the penetration after 60 s of wetting was 65.4 %, 48.4 % and 41.5 %, respectively for EW, TW and LW. Additionally, the time-dependent penetration ratio increased with larger droplet volumes. Since the grain direction must be considered when measuring the contact angle on wood surfaces (Autengruber et al. 2020; Shi and Gardner 2001), calculating the penetration ratio based on both the parallel and perpendicular diameters could provide a more accurate estimation.

Figure 6:

The variation of permeability over time for different droplet volumes: 9 (a), 12 (b) and 15 μL (c) on EW, TW, and LW.

3.4 Surface free energy

Figure 7 demonstrates the directional variation in surface free energy, with significantly higher values observed parallel to the grain than perpendicular to it (p < 0.05). This disparity likely originated from the preferential alignment of microstructural elements along the wood’s longitudinal axis, which enhanced molecular interactions at liquid-solid interfaces (Frybort et al. 2014). This difference was particularly evident in the parallel direction, where the values were 17–24 % higher than those in the perpendicular direction, underscoring the non-uniformity of their surfaces.

Figure 7:

Surface free energy in the parallel (ǁ) and perpendicular (⟂) directions for EW, TW, and LW. *p < 0.05.

Surface free energy parameter fundamentally characterizes a material’s thermodynamic propensity for interfacial contact with liquids (Pizzi and Mittal 2010). Among all the wood types, EW exhibited the highest surface free energy values, aligning with its dominant role in capillary-driven fluid transport. This elevated surface energy correlated with increased permeability and lower contact angles, as observed in Figures 4 and 6. Conversely, the low surface free energy values of LW reflected its limited liquid affinity.

3.5 Impact of anatomical structure on wettability

The ratio of cell walls in the radial and tangential directions for EW, TW, and LW was calculated (Figure 8a), and this ratio significantly influenced wood wettability. As expected, the cell wall ratio increased progressively from EW to LW, with these differences reflected in surface roughness in the tangential section. Figure 8b–d shows that EW surfaces exhibited the highest roughness (R_a = 78.3 μm), followed by TW (R_a = 60.8 μm) and LW (R_a = 30.8 μm). These roughness differences significantly affected capillary (F_c), adhesion (F_a) and resistance (F_r) forces during wetting (Tang et al. 2023). Figure 8e illustrates a schematic of droplet interaction with wood anatomical structures in tangential sections. The anisotropic cellular structure, with directional variations in the tangential and longitudinal directions, influenced these forces. In the tangential direction, droplets encountered multiple consecutive cell walls, resulting in higher F_r, which was greatest in LW and lowest in EW. This increased F_r and reduced F_c hindered effective wetting. In contrast, the longitudinal direction exhibited enhanced F_c, which were weakest in LW and strongest in EW, due to tracheid orientation, while F_r remained relatively weaker. Under gravitational effects, LW demonstrated the strongest F_a, restricting liquid spreading, whereas EW’s weaker F_a allowed for greater liquid extension (Cheng and Sun 2006; Engelund et al. 2012; Monteiro et al. 2021; Zhao et al. 2020).

Figure 8:

Cell wall ratio in the tangential (T) and radial (R) directions for EW, TW, and LW (a), roughness distribution of EW (b), TW (c), and LW (d) on the tangential section, and schematic of droplet interaction with wood considering different forces (e).