Preparation of mechanical abrasion and corrosion resistant bulk highly hydrophobic material based on 3-D wood template

Abstract

Bulk highly hydrophobic wood (BH-wood) was successfully prepared by grafting long-chain alkyl groups onto wood cell walls via the ester linkage. The resulting wood showed lower surface free energy and favorably high hydrophobicity compared to non-treated wood. The microstructure and chemical composition of the control and treated wood were characterized by field emission scanning electron microscopy (FE-SEM), solid state ¹³C NMR, X-ray diffraction (XRD) analysis, and Fourier transform infrared (FT-IR) spectroscopy. The hydrophobic property of the wood was characterized according to contact angle (CA) measurements. The mechanical and chemical durability of the BH-wood were also evaluated. The results suggested that the BH-wood had low surface free energy microstructures extending throughout its whole volume, and possessed excellent mechanical abrasion and corrosion resistance. The self-cleaning property was also significantly improved in the BH-wood compared to the control.

---

Introduction

Superhydrophobic surfaces with water contact angle (CA) greater than 150° have garnered increasing research attention, as they possess not only excellent water-repellence but also self-cleaning, anti-icing, and anti-corrosive properties; of course, these qualities have substantial potential in terms of a wide array of applications.¹ Based on classical theories (i.e., the Wenzel and Cassie–Baxter models),^2,3 the surface roughness represented by various periodically or randomly distributed micro/nanostructures, as well as the surface chemical composition, are crucial in governing the surface wetting behavior. According to this basic principle,⁴ researchers have developed a number of methods to successfully construct superhydrophobic surfaces including chemical vapor deposition,⁵ chemical etching,⁶ sol–gel technique,⁷ solution-immersion,⁸ spin coating,⁹ and laser fabrication.¹⁰ Unfortunately, several of these fabrication techniques are too costly and time-consuming for practical implementation. More importantly, the persistent susceptibility to mechanical abrasion and chemical corrosion severely limit the use of superhydrophobic surfaces in practical applications. Mechanical abrasion can destroy the microscopic roughness structure and hydrophobic layer on the surfaces that are essential for superhydrophobicity.^11,12 Similarly, any chemical reagent penetrating into the textured surface is difficult to remove and thus negates the superhydrophobic behavior of the affected surfaces.¹³ Indeed, mechanical damage and chemical corrosion problems have been well established as the main barrier to the real-world application of superhydrophobic surfaces.¹⁴ Treatment techniques that can mitigate or prevent these problems are thus in quite urgent demand.

Bulk superhydrophobic strategies are in regards to the long-term durability of superhydrophobic properties, i.e., solving the problems of mechanical abrasion and chemical corrosion.¹⁵ There has been relatively little research conducted on optimizing the bulk process.^16,17 Furthermore, even a material with the lowest possible surface energy (6.7 mJ m⁻² for a surface with regularly aligned, closest-hexagonal-packed –CF₃ groups) has a water CA of only around 120°.¹⁸ Sufficient surface roughness is necessary to obtain sufficient hydrophobicity, so selecting or fabricating a 3-D template with appropriate coarseness is essential for preparing bulk superhydrophobic materials.

Over the past several years, a variety of substrate materials have been tested as superhydrophobic surfaces including aluminum,¹⁹ copper,²⁰ zinc,²¹ steel,²² glass,²³ silicon wafer,²⁴ oxides,²⁵ fabric,²⁶ papers,²⁷ and polymers.²⁸ Unlike those materials, wood is an environmentally friendly, healthy, and aesthetically pleasing biopolymer that is intrinsically heterogeneous due to its anatomical structure and porosity, with primary coarseness at the micro-/nanoscale.²⁹ In addition, the components of wood (cellulose, hemicelluloses, and lignin) are rich in hydroxyl groups that can react with hydrophobic groups to decrease the wood surface free energy resulting in superhydrophobic properties. In short, wood is a rather ideal template for preparing bulk superhydrophobic materials. On the other hand, the superhydrophobic wood surfaces with abrasion and corrosion resistant and self-cleaning properties are most important in application level.

In this study, we explored a simple and moderate method of fabricating bulk highly hydrophobic wood (BH-wood) with excellent stability and durability against mechanical abrasion and chemical corrosion. BH-wood was prepared by taking advantage of its primary coarseness at the micro-/nanoscale; octadecyl groups were grafted onto wood cells via the ester linkage. The fabrication mechanism is illustrated in Fig. 1. The long-chain alkyl was chemically bonded onto cell walls through reaction between the wood hydroxyl groups and stearoyl chloride acyl chloride groups, thus forming a stable bulk highly hydrophobic structure. The fabrication mechanism, mechanical and chemical durability, and physical properties of the BH-wood were then investigated at length, as discussed below.

Fig. 1 BH-wood fabrication procedure.

Experimental

Materials

Defect-free and straight-grained sapwood of radiata pine (Pinups radiata D. Don) was obtained from a local wood product manufacturer (Shandong, China). Stearoyl chloride (≥95%) was purchased from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). Anhydrous toluene, ethanol, and acetone were purchased from Beijing Chemical Works (China).

Preparation of bulk highly hydrophobic wood samples

The wood samples were cut into 20 × 20 × 10 mm³ (radial × tangential × longitudinal) blocks, then Soxhlet-extracted with a mixture of toluene/ethanol/acetone (4 : 1 :1  v/v/v) for 12 h, oven-dried at 103 ± 2 °C until reaching a constant weight, and re-measured to determine their weights and dimensions.

Stearoyl chloride was dissolved in anhydrous toluene to prepare solutions at concentrations of 15%. Dried wood samples were immersed into the solution using a vacuum chamber (ca. 0.095 MPa, 30 min) and held for 24 h at ambient temperature. All samples were kept in an oven at 103 ± 2 °C for 12 h, then washed with toluene several times to remove any unreacted stearoyl chloride. The treated wood samples were dried at 103 ± 2 °C for 24 h and their oven dried-weights and dimensions were re-measured.

Characterization

Weight percent gain (WPG) and hydroxyl substitution degree. The weight percent gain (WPG) of the modified samples was calculated according to the following formula:

WPG (%) = (W₂ − W₁)/W₁ × 100 (1)

where W₁ and W₂ are the oven-dried weight of the sample before and after modification, respectively.

The degree of hydroxyl substitution (in mmol g⁻¹ oven-dry wood) was calculated as follows:

OH groups substituted = [(W₂ − W₁)/W₁]/M_W × 1000 (2)

where M_W is the molecular weight of the grafting group.

Field emission scanning electron microscopy (FE-SEM). The morphological features of wood samples were analyzed by field emission scanning electron microscopy (FE-SEM). The interior portions of the cross planes were exposed by slicing them with a rotary microtome, mounting them on conductive adhesives, sputter-coating them with gold, and observing them under a Hitachi SU8010 instrument (Japan) using a voltage of 5 kV.

¹³C CP/MAS solid-state NMR. All ¹³C solid-state NMR spectra were performed on a Bruker AVANCE III 400WB spectrometer operating at 9.4 T with 100.62 MHz frequency. Samples were packed into 4 mm zirconia rotors for observation. The contact time was 1 ms, the recycle delay was 3 s (adjusted to three times the 1H T₁ values) and a ramp-contact and spinal64 decoupling pulse program were utilized; 1480–106 064 scans were run for all samples at rotor spinning rate of 14 kHz.

X-ray diffraction (XRD) analysis. The crystallinity of the treated and untreated wood samples was evaluated with an XRD 6000 diffractometer (Shimadzu, Japan). The apparatus parameters were as follows: Cu-Kα radiation with graphite monochromator, 40 kV voltage, 40 mA electric current, and 2θ scan range of 5–40° with scanning speed of 2° min⁻¹.

Fourier transform infrared (FTIR) spectroscopy. The wood samples were milled to a 200-mesh particle size and embedded into potassium bromide (KBr) pellets at a weight ratio of 1 : 70. The pellets were then analyzed with an FTIR device (Nicolet 6700 Thermo Scientific, USA) ranging from 4000 to 400 cm⁻¹ at a 4 cm⁻¹ resolution for 32 scans.

CA and surface free energy measurement. We utilized the same apparatus and method for CA measurements on our wood sample cross sections as those used in a previous study.³⁰ We took six independent CA measurements for the test liquid to account for the non-homogeneity of wood. Three liquids were used for CA measurements: distilled water, formamide, and diiodomethane. The specifications of their surface tension and components are shown in Table 1.³⁰ The geometric mean equation (OWRK method) and the acid–based approach (vOCG method) based on Young's equation γ_S = γ_L cos θ + γ_SL were used to evaluate the surface free energy, where γ_S is the surface tension of the solid (S), γ_L is the surface tension of the liquid (L), γ_SL is the surface tension of the solid–liquid interface, and θ is the contact angle between S and L.

Table 1 Surface tension and components of test liquids

Liquid	Surface free energy (mJ m^−2)
	γL	γ^LWL (γ^dL)	γ^ABL (γ^pL)	γ^+L	γ^−L
Distilled water	72.8	21.8	51.0	25.5	25.5
Formamide	58.0	39.0	19.0	2.28	39.6
Diiodomethane	50.8	50.8	0	0	0

---

The OWRK method³¹ uses the following equation:

where γ^d_S and γ^p_S are the dispersion and polar components in the surface free energy of a solid (mJ m⁻²), respectively, and γ^d_L and γ^p_L are the dispersion and polar components in the surface free energy of a liquid (mJ m⁻²), respectively.

The vOCG method³¹ uses the following equation:

Instead of describing the polar component (hydrogen bond component) as γ^p, we describe it here as γ^AB, where AB refers to the acid–base interactions. The non-polar (dispersion) term previously described as γ^d is γ^LW here, where LW represents the London–van der Waals forces. The surface free energy can thus be described as γ = γ^LW + γ^AB. Because the polar term was redefined to take into account the acid–base interactions, the component γ^AB is a combination of contributions from electron donors (γ⁻) and electron acceptors (γ⁺). The sum of the acid–base components can be redefined accordingly as .

Mechanical abrasion test. As schematically depicted in Fig. 7a, under a load of 500 g weight (12.5 kPa), the sample surface was rubbed against sandpaper (80 mesh) and moved for 25 cm along at a speed of 3 cm s⁻¹. This process is one abrasion cycle; ten cycles were conducted in total. CAs were measured after each abrasion cycle. The BH-wood samples parallel to the cross section were cut, sanded (80 mesh sandpaper), and washed with ethanol, then the CA of the cutting surface as a function of time was measured.

Chemical durability test. The BH-wood samples were immersed into HCl solution (pH = 2), NaOH solution (pH = 12), and individual organic solvents for 24 h, then the dynamic CAs were measured to determine the samples' chemical durability.

Self-cleaning property. Water was poured with a syringe onto the surface contaminated by carbon-powder to compare the self-cleaning properties of the control and BH-wood samples.

Repellency towards contaminated water. Pristine and BH-wood samples were immersed into water dyed with methylene blue and then withdrawn to compare their ability to repel contaminated water.

Finger-wipe test. We conducted finger-wipe tests by pressing the BH-wood with one finger, then testing its water repellency by dropping blue-dyed water droplets immediately onto the surface.

Mirror-like phenomenon test. The BH-wood sample was immersed into water to visualize mirror-like phenomena.

Scratch-resistance test. We conducted knife-scratch tests by scratching the BH-wood sample repeatedly with a sharp razor blade. Blue-dyed water droplets were then dropped onto the scratched surface to test its water repellency.

Ultrasonic washing test. We examined the washing durability of the BH-wood by submerging the samples in a volume of distilled water six times that of the samples themselves for 1 h under ultrasonication (40 kHz frequency, 100 W). The samples were collected at certain intervals and dried at 103 ± 2 °C for 3 h followed by CA measurements.

UV radiation test. The BH-wood samples were placed in a QUV accelerated weathering tester (QUV/Spray, Q-Lab Co., USA) at 50 °C for a week. The UV lamp power was 40 W and radiation wavelength was 340 nm, and the distance from samples to UV lamps was about 10 cm.

Results and discussion

WPG and hydroxyl substitution degree

The WPG of BH-wood samples increased by about 12.25% after modification. The number of substituted hydroxyl groups was 0.460 mmol g⁻¹, indicating that a large amount of hydroxyl groups had bonded with stearyl chloride.

Microstructure and chemical composition

The cross-section morphologies of the control and BH-wood were characterized by FE-SEM, as shown in Fig. 2. The natural porous structure of the control sample was readily observable. For the BH-wood, the long-chain alkyl was more likely anchored in the cell walls by chemical reactions, and the wood microstructure was bulked by this cell wall modification. According to Rowell,³² the HCl released as a by-product of chloride treatment causes degradation of the wood; however, there was no such change observed in our BH-wood samples, indicating that the modification process had no significant impact on the wood tissue microstructure.

Fig. 2 FE-SEM observations of (a) untreated wood and (b) BH-wood.

The FTIR spectra of the control and BH-wood samples are shown in Fig. 3. The prominent band at 3370 cm⁻¹ in control samples, assigned to –OH group stretching vibration, showed a slight decrease after the modification indicating that a portion of the –OH groups had been replaced.^33,34 The absorption band at 2900 cm⁻¹ in control wood was assigned to C–H stretching vibration, which was provided by the wood components.³⁵ Two strong adsorption peaks at 2917 cm⁻¹ and 2850 cm⁻¹ appeared in the BH-wood samples that can be attributed to –CH₃ asymmetrical stretching vibrations and –CH₂ symmetrical stretching vibrations from long alkyl chains.^36,37 In addition, the characteristic peak at 1738 cm⁻¹ shifted to 1704 cm⁻¹ with higher intensity due to the appearance of new carbonyl groups.³⁴ These changes indicate that long alkyl chains were successfully attached to the wood cell walls via ester linkage.

Fig. 3 FTIR spectra of control and BH-wood samples.

The ¹³C CP/MAS spectra of control and treated wood samples are shown in Fig. 4. The control wood spectra showed signals at 105, 88, and 65 ppm due to the ordered cellulose C1, C4, and C6 carbons, respectively; the signals between 72 and 74 ppm were assigned to the C2, C3, and C5 carbons of cellulose. The signals at 83 and 62 ppm can be attributed to the amorphous cellulose C4 and C6 carbons as well as the less-ordered cellulose chains of the crystallite surfaces.^38,39 The peaks at 21 ppm and 172 ppm were assigned to methyl C and carboxylic C in acetoxy groups of hemicelluloses, respectively.^40,41 The lignin OMe groups give a signal at 55 ppm. After stearoylation, there was a shift at 14 ppm assigned to –CH₃ groups and a shift at 33 ppm assigned to –CH₂– groups of the stearoyl chains.⁴² Peaks that resonated between 18 and 37 ppm are typical of carbons of fatty aliphatic chains.⁴³ These results indicate that long-chain alkyl was successfully grafted onto the wood substrate, in accordance with the FTIR analysis.

Fig. 4 ¹³C CP/MAS solid-state NMR spectra of (a) control and (b) BH-wood.

In addition, the C6 and C2/C3/C5 signal for stearoyl-modified wood split the shift in two. The peak corresponding to C4 in the highly ordered cellulose became stronger due to differences in the crystallinity, which likely occurred because the form of the paracrystal segments changed after long-chain alkyl was grafted onto the wood fibers. This result is in accordance with the XRD test.

The crystalline structures of control and treated samples were characterized by XRD, as discussed above (Fig. 5). For the untreated wood, the diffraction peaks at diffraction angles of 15.1°, 22.0°, and 34.5° can be assigned to the crystal planes (101), (002), and (040) of cellulose, respectively.⁴⁴ Compared to the control wood, the intensity of these peaks became stronger and the crystallinity index remarkably increased in the BH-wood, likely because the long-chain alkyl grafting onto wood cell walls caused increased number of paracrystal segments. It is also possible that the lignin partial degradation caused by hydrochloric acid release resulted in the increased crystallinity index.

Fig. 5 XRD curves for control and BH-wood samples.

Hydrophobicity and surface free energy

Fig. 6 illustrates the changes of CAs over time. The CAs on different surfaces of the untreated wood samples decreased dramatically in a short time. By contrast, the CAs on the cross section of BH-wood samples remained larger than 150°, while on the radial section and tangential section remained about 140° and showed no obvious changes in 120 s, indicating excellent hydrophobic properties. However, the CAs on longitudinal surfaces were only about 140°, which were lower than CAs on the transverse surfaces. This was determined by the wood peculiarity due to its unique anatomical structure and porosity. The longitudinal surface of wood displayed a micro-grooved structure, and the transverse surface showed a honeycomb-like appearance. According to the model of Cassie and Baxter, when the water droplets dropped in the longitudinal surfaces, since air in the micro-grooves can escape in both directions, the water droplets filled the micro-grooves, causing smaller CAs. By contrast, when the water droplets dropped in the transverse surface, the air trapped in the larger pores and cannot escape. As a result, the water droplet was sitting above a shallow air layer and cannot fill the cavities or spread over the surface, so the CAs was around 152°.^45,46 In addition, the dynamic sliding angles on the transverse surfaces were measured to be ranged from 13° to 38°, which was higher than the superhydrophobicity of 10°. This was likely due to the larger size pores (the diameter about from 10 μm to 50 μm) on the transverse surface. Although the roughness surface contributed by these pores with different size and shape can support the spherical water droplets, the adhesion between the transverse surface of wood and water droplets was slightly higher.

Fig. 6 Contact angle as a function of time for control wood and BH-wood on difference surface, and the profile of water droplets on the different surfaces of BH-wood.

The initial CAs of the test liquids are shown in Table 2. Due to the small standard deviations and the dramatically reduced CAs of untreated wood, it was effective and appropriate to use the initial CA to replace the equilibrium CA to calculate the surface free energy. The surface free energy results calculated by OWRK and vOCG methods are provided in Tables 3 and 4, respectively. The total surface free energy of the untreated wood samples was 45.34 mJ m⁻² (as-calculated by OWRK) which is in accordance with values presented in the literature ranging from 40 mJ m⁻² to 60 mJ m⁻².^47,48 The dispersion components remarkably decreased while the polar components slightly increased after treatment, and as a result, the total surface free energy decreased to 17.00 mJ m⁻².

Table 2 Contact angle of untreated and treated wood samples

Sample	θ (degrees)
	Water	Formamide	Diiodomethane
Reference	93.24(3.64)	68.35(6.10)	27.65(3.40)
BH-wood	153.65(1.81)	136.07(2.68)	88.24(3.16)

---

Table 3 Surface free energy of treated and control wood samples (OWRK method)

Samples	γS	Surface energy components (mJ m^−2)
		γ^dS	γ^pS
Reference	45.34	45.16	0.17
BH-wood	17.00	13.49	3.50

---

Table 4 Surface free energy of treated and control wood samples (vOCG method)

Samples	γS	Surface energy components (mJ m^−2)
		γ^LWS	γ^ABS	γ^+S	γ^−S
Reference	41.98	41.98	0	0	0.49
BH-wood	13.42	13.42	0	0	0

---

According to the vOCG results, the surface free energy reduced from 41.98 mJ m⁻² to 13.42 mJ m⁻² after treatment. The highly hydrophobicity of the BH-wood samples can be explained by the reduced surface free energy combined with the material's inherent coarseness.

Mechanical abrasion and chemical durability

The mechanical stability of superhydrophobic surfaces is crucial for their use in practical applications. As described above, we assessed the mechanical durability of our highly hydrophobic bulk material samples with scratch tests. The water CAs after each abrasion cycle are shown in Fig. 7b. The BH-wood retained highly hydrophobicity, with CAs above 150°, after being scratched repeatedly.

Fig. 7 (a) Sandpaper abrasion test; (b) CAs as a function of number of abrasion cycles for BH-wood surfaces; (c) spherical water droplets rested steadily on BH-wood surface and cutting surface; (d) CAs as a function of time for BH-wood cutting surface.

BH-wood samples parallel to the cross section were cut and sanded with sandpaper (80 mesh), then washed with ethanol. Water droplets dropped onto these surfaces still displayed a spherical shape on the cutting surface, as shown in Fig. 7c. The CAs of the cutting surface as a function of time are shown in Fig. 7d, where the CAs remained above 150° for 3 min, indicating excellent highly hydrophobic properties and demonstrating that the low-surface-energy microstructures extended throughout the BH-wood's whole volume. These observations suggest that the highly hydrophobic bulk material we fabricated was mechanically robust and would be able to withstand practical application.

As discussed above, we investigated the chemical stability of the BH-wood samples by conducting multiple CA measurements. Fig. 8a–f show the dynamic CAs of the BH-wood samples after they were soaked in different chemical reagents (hydrochloric acid solution of pH 2, sodium hydroxide solution of pH 12, n-hexane, toluene, acetone, and DMF, respectively) for 24 h. The CAs all remained above 150° for 2 min, indicating excellent acid-, alkali-, and organic solvent resistance properties.

Fig. 8 Contact angle as a function of time for BH-wood samples soaked into different chemical solutions for 24 h. (a) HCl solution (pH = 2); (b) NaOH solution (pH = 12); (c) n-hexane; (d) toluene; (e) acetone; (f) DMF.

As shown in Fig. 9a and b, blue-dyed water droplets immediately spread across the untreated wood surface in dramatic contrast to the spherical water droplets resting steadily on the BH-wood surface. The self-cleaning property of the BH-wood surface was characterized thought a dirt-removal test in which water was dropped onto the wood surface after contaminating it with carbon powder. As shown in Fig. 9c and e, when dropped onto the slightly tilted surface, the water droplets merged with the carbon powder and stuck to the untreated wood surface, whereas the water droplets readily rolled off the BH-wood surface taking the dirt with them (Fig. 9d and f) and leaving a dry and clean surface.

Fig. 9 Optical photographs of dyed water droplets on (a) control wood surface; (b) BH-wood surfaces; (c, e) snapshots of the self-cleaning process on untreated wood surfaces; (d, f) snapshots of the self-cleaning process on BH-wood surfaces.

The highly hydrophobic performance of BH-wood after ultrasonic washing (40 kHz, 100 W) was also evaluated, as described above. Fig. 10 shows the changes in CA as a function of ultrasonic washing time for the BH-wood. The CAs of the surface remained almost constant around 150° after ultrasonic washing for a duration of 60 min, indicating that the BH-wood had sufficiently durable highly hydrophobic properties to withstand ultrasonication.

Fig. 10 CAs as a function of time of ultrasonic washing in water for BH-wood surfaces.

The stability of the highly hydrophobicity under UV light was carry out,⁴⁹ and the results are shown in Fig. 11. The CAs of the surface remained about 150° after the UV radiation for a duration of 168 h, indicating that the BH-wood has excellent stable hydrophobicity to withstand the UV radiation.

Fig. 11 CAs as a function of exposure time in UV radiation for the BH-wood surfaces.

As shown in Fig. 12a, repellency towards dyed water can also demonstrate self-cleaning ability – we observed both by immersing the wood samples in methylene-blue-dyed water. Once withdrawn from the water, the BH-wood remained clean with no trace of contamination; the control wood, conversely, was considerably polluted.

Fig. 12 (a) Repellency towards dyed water; (b) finger-wipe test; (c) scratch-resistance test; (d) mirror-like phenomenon; (e) common household liquids on the BH-wood surface.

Fig. 12b shows the finger-wipe test, the CA is about 148° and the results of which indicated that the BH-wood surface can sustain mechanical damage as well as oil and salt contaminations caused by human contact.

Fig. 12c shows the scratch-resistance test, the CA is about 151°, in which the BH-wood surface retained excellent water-repellency even after being damaged repeatedly with a razor blade.

As shown in Fig. 12d, the BH-wood surface exhibited mirror-like phenomena under water when observed at an oblique angle; this is a signature of trapped air and the establishment of composite solid–liquid–air interfaces² which effectively prevent surface wetting under water.

As shown in Fig. 12e, we used four common household liquids (coffee, milk, fruit juice, and ink) to examine the surface repellency toward them on the BH-wood surface. Similar to the water droplets, the liquid droplets remained steadily spherical on the wood surface.

Conclusion

Bulk highly hydrophobic material was successfully prepared based on a 3-D wood template through grafting long-chain alkyl onto wood cell walls via ester linkage. The prepared wood exhibited excellent highly hydrophobic properties and reduction in surface free energy values by 62.5% compared to the control according to the OWRK method and 68.0% by the vOCG method. The BH-wood also showed outstanding stability and durability against mechanical abrasion and chemical corrosion, as well as favorable self-cleaning performance. Most importantly, the low-surface-energy microstructures extended throughout the BG-wood's whole volume, suggesting that the excellent highly hydrophobicity existed not only at the surface but also the interior.

Acknowledgements

This research was supported by “The Fundamental Research Funds for the Central Universities” (No. 2016ZCQ01), Special Fund for Forestry Research in the Public Interests (Project 201204702) and Key Laboratory of Bio-based Material Science and Technology (Northeast Forestry University), Ministry of Education (SWZCL 2016-11).

Notes and references

1. Y. Zhang, H. Xia, E. Kim and H. Sun, Soft Matter, 2012, 8, 11217–11231.
2. A. B. D. Cassie and S. Baxter, Trans. Faraday Soc., 1944, 40, 546–551.
3. R. N. Wenzel, Ind. Eng. Chem., 1936, 28, 988–994.
4. X. M. Li, D. Reinhoudt and M. Crego-Calama, Chem. Soc. Rev., 2007, 36, 1350–1368.
5. J. Yu, L. Qin, Y. Hao, S. Kuang, X. Bai, Y.-M. Chong, W. Zhang and E. Wang, ACS Nano, 2010, 4, 414–422.
6. B. Qian and Z. Shen, Langmuir, 2005, 21, 9007–9009.
7. D. Su, C. Huang, Y. Hu, Q. Jiang, L. Zhang and Y. Zhu, Appl. Surf. Sci., 2011, 258, 928–934.
8. C. Liu, S. Wang, J. Shi and C. Wang, Appl. Surf. Sci., 2011, 258, 761–765.
9. L. Xu, R. G. Karunakaran, J. Guo and S. Yang, ACS Appl. Mater. Interfaces, 2012, 4, 1118–1125.
10. C. Dong, Y. Gu, M. Zhong, L. Li, K. Sezer, M. Ma and W. Liu, J. Mater. Process. Technol., 2011, 211, 1234–1240.
11. Y. Xiu, Y. Liu, D. W. Hess and C. P. Wong, Nanotechnology, 2010, 21, 155705.
12. X. Zhu, Z. Zhang, X. Men, J. Yang, K. Wang, X. Xu, X. Zhou and Q. Xue, J. Mater. Chem., 2011, 21, 15793.
13. T. Verho, C. Bower, P. Andrew, S. Franssila, O. Ikkala and R. H. Ras, Adv. Mater., 2011, 23, 673–678.
14. P. Roach, N. J. Shirtcliffe and M. I. Newton, Soft Matter, 2008, 4, 224–240.
15. X. Zhu, Z. Zhang, G. Ren, J. Yang, K. Wang, X. Xu, X. Men and X. Zhou, J. Mater. Chem., 2012, 22, 20146.
16. S. Ramakrishna, K. S. Santhosh Kumar, D. Mathew and C. P. Reghunadhan Nair, Sci. Rep., 2015, 5 DOI:10.1038/srep18510.
17. K. S. S. Kumar, V. Kumar and C. P. R. Nair, J. Mater. Chem. A, 2014, 2, 15502.
18. T. Nishino, M. Meguro, K. Nakamae, M. Matsushita and Y. Ueda, Langmuir, 1999, 15, 4321–4323.
19. H. Wang, D. Dai and X. Wu, Appl. Surf. Sci., 2008, 254, 5599–5601.
20. L. Kong, X. Chen, G. Yang, L. Yu and P. Zhang, Appl. Surf. Sci., 2008, 254, 7255–7258.
21. T. Ning, W. Xu and S. Lu, Appl. Surf. Sci., 2011, 258, 1359–1365.
22. X. Zhang, M. Honkanen, M. Järn, J. Peltonen, V. Pore, E. Levänen and T. Mäntylä, Appl. Surf. Sci., 2008, 254, 5129–5133.
23. S. A. Mahadik, M. S. Kavale, S. K. Mukherjee and A. V. Rao, Appl. Surf. Sci., 2010, 257, 333–339.
24. Y. He, C. Jiang, H. Yin, J. Chen and W. Yuan, J. Colloid Interface Sci., 2011, 364, 219–229.
25. S. S. Latthe and A. V. Rao, Surf. Coat. Technol., 2012, 207, 489–492.
26. X. Zhu, Z. Zhang, J. Yang, X. Xu, X. Men and X. Zhou, J. Colloid Interface Sci., 2012, 380, 182–186.
27. X. Huang, X. Wen, J. Cheng and Z. Yang, Appl. Surf. Sci., 2012, 258, 8739–8746.
28. J. Liu, X. Xiao, Y. Shi and C. Wan, Appl. Surf. Sci., 2014, 297, 33–39.
29. J. Li and J. Qiu, Sci. Silvae Sin., 2005, 41, 189.
30. Z. Qin, Q. Gao, S. Zhang and J. Li, BioResources, 2014, 9, 3088–3103.
31. C. J. van Oss, R. J. Good and M. K. Chaudhury, Langmuir, 1988, 4, 884–891.
32. R. M. Rowell, Handbook of wood chemistry and wood composites, CRC Press, New York, 2005, pp. 383–413.
33. Y. Dong, Y. Yan, S. Zhang and J. Li, BioResources, 2014, 9, 6028–6040.
34. B. D. Mattos, T. V. Lourençon, L. Serrano, J. Labidi and D. A. Gatto, Wood Sci. Technol., 2014, 49, 273–288.
35. Z. Gao, M. Ma, X. Zhai, M. Zhang, D. Zang and C. Wang, RSC Adv., 2015, 5, 63978–63984.
36. H. Chang, K. Tu, X. Wang and J. Liu, RSC Adv., 2015, 5, 30647–30653.
37. K. K. Pandey, J. Appl. Polym. Sci., 1999, 71, 1969–1975.
38. M. Jebrane and G. Sèbe, Carbohydr. Polym., 2008, 72, 657–663.
39. T.-Q. Yuan, L.-M. Zhang, F. Xu and R.-C. Sun, Ind. Crops Prod., 2013, 45, 36–43.
40. H. Wikberg and S. Liisamaunu, Carbohydr. Polym., 2004, 58, 461–466.
41. M. Alesiani, F. Proietti, S. Capuani, M. Paci, M. Fioravanti and B. Maraviglia, Appl. Magn. Reson., 2005, 29, 177–184.
42. M. Granström, M. K. née Pääkkö, H. Jin, E. Kolehmainen, I. Kilpeläinen and O. Ikkala, Polym. Chem., 2011, 2, 1789–1796.
43. S. Berlioz, S. M. Boisseau, Y. Nishiyama and L. Heux, Biomacromolecules, 2009, 10, 2144–2151.
44. Y. Dong, K. Wang, Y. Yan, S. Zhang and J. Li, Holzforschung, 2016, 70, 919–926.
45. X. Wang, Y. Chai and J. Liu, Holzforschung, 2013, 67, 667–672.
46. K. Tu, X. Wang, L. Kong, H. Chang and J. Liu, RSC Adv., 2016, 6, 701–707.
47. M. Gindl and S. Tschegg, Langmuir, 2002, 18, 3209–3212.
48. N. S. Little, T. E. McConnell, N. E. Irby, S. Q. Shi and J. J. Riggins, Wood Fiber Sci., 2013, 45, 206–214.
49. S. S. Latthe, P. Sudhagar, A. Devadoss, A. M. Kumar, S. Liu, C. Terashima, K. Nakataa and A. Fujishima, J. Mater. Chem. A, 2015, 3, 14263–14271.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19549k

---