Ruihang
Wen‡
,
Xiaobing
Chen‡
,
Gaocan
Qi
*,
Wenbin
Li
and
Zhihao
Yuan
*
School of Materials Science and Engineering, Tianjin Key Lab of Photoelectric Materials & Devices, and Key Laboratory of Display Materials and Photoelectric Devices (Ministry of Education), Tianjin University of Technology, Tianjin 300384, P. R. China. E-mail: gaocanqi@tjut.edu.cn; zhyuan@tjut.edu.cn
First published on 6th September 2024
In mechanism, surface superhydrophobicity is often thought to be the result of the trapped air within the grooves of a superhydrophobic surface, leading to a composite solid–liquid–air interface. However, the mechanism cannot reasonably reflect why the rough surfaces of intrinsically hydrophilic materials are capable of showing hydrophilicity or superhydrophobicity. In this work, several typical rough-textured inorganic materials (i.e. metal oxide, sulfide, selenide and halide) endowed with intrinsic hydrophilicity are taken as examples to reveal the superhydrophobic origin of intrinsically hydrophilic materials. The wettability of these rough-textured surfaces is usually hydrophilic when dried in N2, while it is hydrophobic or superhydrophobic when dried in O2. This distinct difference in wettability is closely related to anion vacancies of anions such as oxide, sulfide and halide ions. From the generation of H2O2 in water droplets on the rough-textured surfaces, it is found that the H2O2 yield increases with an increase in their hydrophobicity and decreases with an increase in their hydrophilicity, indicating an evident dependence of their surperhydrophobicity on the absorption of O22− on their surfaces. DFT calculation shows that introducing Va can give a higher adsorption-energy for molecular oxygen species (especially O22−) than N2 and H2O. Hence, we propose that the presence of abundant Va on intrinsically hydrophilic inorganic materials can result in the formation of preferential O22− adsorption layer in their grooves, which endows these inorganic materials with superhydrophobicity. Our results can provide further insight into the origin of superhydrophobicity in intrinsically hydrophilic materials and guide the design of water-proof functional surfaces.
Generally, superhydrophobicity can be obtained either by preparing micro-/nano-scale roughness structures on hydrophobic materials or by modifying a rough-textured surface with organic chemical coating. In this regard, many theoretical hypotheses and models have been proposed to interpret natural and artificial superhydrophobic phenomena. Among them, the Wenzel and Cassie–Baxter theoretical models, which propose a basic hypothesis that the wettability of solid surfaces is determined both by surface roughness and surface energy, have been widely adopted. According to the models, superhydrophobic behaviors are attributed to the ability to trap air within the grooves of a rough surface, leading to a composite solid–liquid–air interface under water droplets (i.e. a trapping-air layer mechanism). In recent years, many advances have been made in the construction of water-proof functional surfaces using intrinsically hydrophilic inorganic materials such as metal oxide, sulfide, selenide and halide, which have been proven to have wide applications in electro- or photo-controllable water permeation, superhydrophobic antibacterial materials, water-resistant electronic devices, and self-cleaning solar cells.11–15 Although some efforts have been devoted to interpret their superhydrophobic behaviors over recent years, such as hollow structure16 and re-entrant-textured structure,17,18 the trapping-air layer mechanism cannot reasonably explain how the rough surfaces of these intrinsically hydrophilic materials are superhydrophobic. For example, why is the trapping-air layer on their rough surfaces stable under water, given that the intrinsic hydrophilicity tends to facilitate the formation of a water adsorption layer on rough surfaces rather than a trapping air layer? Why do their rough-textured surfaces with superhydrophobicity generally showcase a phenomenon of high viscosity “pegging” to water droplets? And why can their rough-textured surfaces with the same roughness exhibit either superhydrophobicity or superhydrophilicity?19–21 Therefore, shedding new light on the origin of superhydrophobicity in intrinsic hydrophilic materials would be of great significance to further understand the superhydrophobic phenomenon as well as to design novel superhydrophobic materials for future applications. In many reports, the micro-/nano-textured inorganic materials are endowed with abundant anion vacancies (Va)22–25 (i.e., oxygen-, sulfide- and halogen-ion ones) both on the surface and in the bulk, and the Va can play a pivotal role in tailoring their functionalities, such as electrical conductivity,26 photocurrent,27 gas sensitivity,28 pH detection29 and catalytic and photocatalytic performance.30 Singh et al.31 reported that the wettability of some nano-textured metal oxides strongly depended on the oxygen-related defects. Yan and Zhang et al.32,33 found that the introduction of oxygen vacancies can promote the adsorption of O2 on catalytic materials and offer additional benefits for OER.
In the previous work,34 we focused on the correlation between the wetting behavior and the oxygen vacancies for intrinsically hydrophilic metal oxides and found that the presence of oxygen vacancies makes the roughness on their surfaces more capable of adsorbing O2 than N2 and H2O, which resulted in an oxygen-preferred adsorption layer in the grooves (namely, O2-adsorption layer mechanism) rather than the trapping-air layer as commonly thought. The O2-adsorption layer mechanism can not only explain the origin of superhydrophobicity and the “pegging” phenomenon for intrinsically hydrophilic oxide materials but also conduct the design of excellent superhydrophobicity with low adhesion to the water droplet. But though our research has advanced the understanding of the superhydrophobic mechanism for intrinsic hydrophilic metal oxides, some fundamental questions remain unclear when the mechanism of the O2-adsorption layer is further extended to the intrinsic hydrophilic inorganic materials: Can the presence of Va on their micro-/nano-textured surfaces cause a preferential O2-adsorption rather than the air-trapped layer, as oxygen vacancies? How do the Va with positive charge adsorb electrically neutral O2 molecules to form a stable O2-adsorbed layer within the grooves of their rough surfaces under water-droplet? Recent researches have proved that the harvesting of H2O2 in the water droplet on the intrinsic hydrophilic inorganic materials strongly depends on their superhydrophobicity, which suggests that their superhydrophobicity can be closely related to the absorption of O22− on their rough-textured surfaces. If so, what is the role of adsorbed O22− ions on their superhydrophobic behaviors?
In this work, we take a variety of representative hydrophilic inorganic materials including sulfide, selenide, oxide and halide (Table S1†) as model materials to reveal their superhydrophobic origin. These inorganic materials were prepared with different topographic structures, such as nanorods, nanoplates, nanoparticles, microrods and microparticles, and the drying and water contact angle (WCA) test were performed in different atmospheres, namely, air, nitrogen (N2) and oxygen (O2). Surface anion defects and related O2 adsorption were investigated with X-ray photoelectron spectroscopy, Raman signals and spontaneous H2O2 generation. Combining the adsorption energies of H2O, N2 and O2 on ideal and defected crystal plane calculated by density functional theory (DFT), we proposed a new understanding for the origin of superhydrophobicity of intrinsically hydrophilic inorganic materials based on O22−-dominant O2-adsorption layer.
To investigate the effect of oxygen on the surface state, we studied the surface defects and related oxygen adsorption on ZnS nanorods and AgCl nanorods, as representative specimens, via X-ray photoelectron spectroscopy and Raman spectroscopy. Fig. 2a and c shows the S 2p and Cl 2p XPS spectra of ZnS nanorods and AgCl nanoparticles, respectively. The asymmetric S 2p spectrum can be deconvoluted into four peaks centered at 161.2 eV, 162.2 eV, 162.4 eV and 163.4 eV. The binding energies (BE) of 161.2 eV and 162.4 eV were derived from S 2p3/2 and S 2p1/2 of lattice S atoms in the ZnS crystal, while 162.2 eV and 163.4 eV were considered to be derived from S atoms in the vicinity of S vacancies (Vs).38,39 The obvious shoulder XPS peaks at 162.2 eV and 163.4 eV indicate the presence of abundant Vs defects on the surface of ZnS nanostructures, which has been confirmed by Jaewon et al.39,40 For Cl 2p, two peaks were observed at binding energies of 197.7 eV and 199.4 eV, which moved to a lower energy compared with BE of 198.6 eV and 200.2 eV of ideal lattice Cl in the AgCl crystal.41 This movement of Cl 2p peaks has been widely reported at reduced AgCl surface with Cl removal, namely Cl vacancy (VCl).25,42–44 Accompanied by the emergence of Vs and VCl defects, the O 1s peaks were both observed in the XPS spectra of ZnS nanorods (Fig. 2b) and AgCl nanoparticles (Fig. 2d). These results demonstrate that plenty of Va defects exist on the surface of inorganic micro/nano-structures, which might induce the adsorption of oxygen species from air.
The interaction between Va-enriched inorganic materials and ambient atmospheres, mainly O2 and N2, was studied by Raman spectroscopy analysis, which is shown in Fig. 2e and f. The Raman peaks at 280 cm−1, 335 cm−1 and 350 cm−1 in Fig. 2e were attributed to the first-order phonon modes of wurtzite ZnS nanorods,45 and the Raman signals at 95 cm−1, 143 cm−1, 233 cm−1, 344 cm−1 and 410 cm−1 in Fig. 2f belonged to the face-centered cubic AgCl.46 In comparison with the Raman spectra obtained in N2, several obvious new peaks centered at 840 cm−1, 862 cm−1 were found in the O2-dried ZnS sample, while the peaks at 938 cm−1 were found in the O2-dried AgCl sample. As reported in the present work, the peaks located at 840 cm−1 and 862 cm−1 were assigned to the O–O stretching of the adsorbed peroxide species (O22−) around Zn2+ with different degrees of defect aggregation.34,47 On the defective AgCl surface, the Raman peak centered at 938 cm−1 was attributed to the stretching of O22−.46,47 In addition, in the Raman spectra of the O2-dried samples, a weak peak centered at 1548 cm−1 and a broad peak ranging from 1050 cm−1 to 1170 cm−1 were observed in the case of ZnS, while a weak peak centered at 1550 cm−1 and a broad peak spanning from 1020 cm−1 to 1180 cm−1 were observed in the case of the AgCl sample. The Raman peaks at 1548 cm−1 or 1550 cm−1 generally corresponded to the O–O stretching vibrations of adsorbed O2 molecules, with a slight difference in the wavenumbers attributed to the varying adsorption environments.48 The two broad peaks might be attributed to the stretching and bending vibrations of a small amount of O2− around Zn2+ and Ag+, respectively.47 These Raman results indicated that inorganic materials with Va-defective surface favored the adsorption of O2 species rather than N2.
Recently, Zare et al. reported that hydrogen peroxide can be generated in water droplets on the surface of hydrophobic materials, and their works show that the yield of hydrogen peroxide is related to the WCAs of water droplets.49–51 In order to clarify the relationship between H2O2 yield and the hydrophobic properties of material surfaces, we recorded the yield of H2O2 on inorganic materials with different wettability. Fig. 3a, b, and c show the microscopic images and H2O2 yields of condensed water on ZnS nanorods, AgCl nanoparticles, and ZnO nanowires (their morphologies and XRD data are shown in ESI Fig. S3†), respectively, after being dried in different atmospheres. It can be seen that the condensed water spreads over the samples dried in N2 but beads up on the samples dried in air. The hydrophilicity of these inorganics in N2 and the hydrophobicity of the same samples in air are consistent with the results shown in Fig. 1. To quantify the concentration of H2O2 generated in water droplets on their surfaces, we measured the absorption spectrum of aqueous potassium titanium oxalate (PTO, K2TiO(C2O4)2·H2O) that had reacted with the condensed water. According to the calibration plot of the absorbance at 400 nm (see ESI S4†), the H2O2 concentrations in the condensed water on hydrophilic ZnS nanorods, AgCl nanoparticles and ZnO nanowires, which were dried in N2, are all approximately 0 μM L−1 (gray lines). On the other hand, for the corresponding air-dried hydrophobic samples, the H2O2 concentrations were ∼71.15 μM L−1, ∼52.05 μM L−1, and ∼63.24 μM L−1 (blue lines), respectively. Furthermore, O2 and H2 heating procedures were performed on the ZnO nanowires to adjust its wettability in air. The insets of water droplets in Fig. 3c3 show that H2 heat-treatment results in an enhanced superhydrophobicity with a CA of ∼162° compared to the unheated samples (∼150°), but the O2 heat-treatment induces a wettability transition from superhydrophobicity to hydrophilicity (CA ≈ 75°). For the H2-annealed sample, the H2O2 concentration in the condensed water droplet increases to 77.38 μM L−1 (purple line in Fig. 3c3) compared to the untreated sample (63.24 μM L−1). On the contrary, the H2O2 concentration decreases to 49.47 μM L−1 (orange line in Fig. 3c3) in the water droplet on the O2-annealed sample. These results demonstrate that the yield of H2O2 on these inorganic materials increases with the enhancement of their hydrophobicity, while it decreases with the reduction of their hydrophobicity. Zhang et al. have reported that O22− is an important intermediate in the formation of H2O2 during 2e− oxygen reduction reaction (ORR).49,52,53 Also, we have confirmed that O2 heat-treatment is capable of reducing Vo defects on the surface of ZnO nanomaterials, whereas H2 heat-treatment increases their occurrence.34 Therefore, it can be inferred that the hydrophobicity of inorganic materials is related to the amount of O22− adsorbed on the Va defects.
DFT calculations were further performed to study the hydrophobicity of inorganic materials induced by O2 adsorption at Va. The effect of Va on the adsorption energies (Ea) of O2 species (O2, O2−, O22−), N2 and H2O molecules was calculated using wurtzite ZnS, rocksalt AgCl and wurtzite ZnO as representatives of sulfides, halides and oxides, respectively. Herein, the ZnS-(100) plane of wurtzite, AgCl-(100) plane of rocksalt and ZnO-(100) plane of wurtzite, which are predominantly exposed planes grown by common solution approaches, were selected as model unit cells for periodic slab.20,54,55 The defective surfaces with Va concentration of 1/6 monolayer were modeled by removing a sulfur atom, a chlorine atom or an oxygen atom at the top atomic layer from the corresponding ideal crystal surface. The most stable adsorption configurations for O22−, N2, and H2O molecules on ideal and defective ZnS-(100), AgCl-(100) and ZnO-(100) surfaces, as well as their adsorption energies (Ea), are given in Fig. 4 (the Ea differences of O2, O2− and O22− shown in ESI S5† manifested that O22− exhibited the most stable adsorption on the Va sites of the three materials). Fig. 4a1 and c1 show that the ideal ZnS-(100) and AgCl-(100) crystal surfaces exhibit strong adsorption capability for H2O, with an Ea of −0.584 eV and −0.712 eV respectively, but display very weak adsorption for both O22− and N2. This result is consistent with the wetting characteristics of intrinsically hydrophilic inorganic materials. When Va defects are present on ZnS-(100) and AgCl-(100) surfaces, the Ea for H2O at the defect sites are −0.410 eV and −0.516 eV, respectively, which are lower than those at ideal surfaces. The adsorption of H2O is suppressed at Va defects mainly due to the repulsion between positively charged H atoms and the electropositive metal atoms surrounding the Va sites. It is worth noting that the adsorption of O22− is significantly enhanced at the Va sites, with its Ea on VS, VCl, and Vo is much larger than that of H2O on both Va and ideal crystal planes as well as that of N2 (the Ea of H2O and N2 on ideal and defective ZnO-(100) plane has been reported in our previous work34). These calculated thermodynamic data demonstrated that the presence of Va led to a transition of the inorganic crystals from its original H2O affinity to O22− affinity. Furthermore, the O22−-adsorbed Va sites were also capable of adsorbing other O2 species, O2 and O2−; the Ea are listed in ESI S6.† Therefore, the presence of Va defects caused the inorganic materials to have stable O22−-predominant O2 molecular adsorption layer, which would prevent the wetting of water on the inorganic surfaces. This defect-induced hydrophobicity is consistent with the variation in the hydrophobicity of ZnO and ZnS nanomaterials caused by the tuning of vacancy concentration, as shown in Fig. 3c3 and Fig. S7.†
According to the above experimental results and theoretical calculations, we provided a new understanding toward the origin of superhydrophobicity in rough-textured inorganic materials, which is schematically shown in Fig. 5. (1) The abundant Va defects on the surface of rough-textured inorganic materials gave rise to the preferential adsorption of O22− from the air compared to N2 and H2O. This preferential O22− adsorption enabled the defective inorganic surface to anchor O22− from air, leading to the formation of an O22−-predominant O2 molecular adsorption layer on the micro-/nano-structures. The preferentially adsorbed O2 molecular layer hindered the contact of the water droplet on the inorganic surfaces and endowed the intrinsically hydrophilic inorganic materials with the property of hydrophobicity. (2) The adsorbed O2 layer was stable under water and hard to be replaced by H2O because of the larger adsorption energy of O22− than H2O, which ensured the stability of hydrophobicity. (3) The rough-textured structures led to an increase of the Va content per unit area and promoted the stability of O22−-predominant O2 molecular adsorption layer within the grooves. The hydrophobicity of inorganic materials was further enhanced to superhydrophobicity by the rough textured micro-/nano-structures. This understanding of superhydrophobicity focused on the critical role of Va-induced O22− adsorption, which furthered the formation of the O2 adsorption layer on the Va site rather than the air-trapped layer, as commonly believed, and provided beneficial development of the traditional superhydrophobicity mechanism.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4qi02059f |
‡ These authors contributed equally to this work. |
This journal is © the Partner Organisations 2024 |