Preparation of hierarchical porous Zn-salt particles and their superhydrophobic performance

Accepted Manuscript Title: Preparation of hierarchical porous Zn-salt particles and their superhydrophobic performance Author: Dahai Gao Mengqiu Jia PII: DOI: Reference:

S0169-4332(15)02366-1 http://dx.doi.org/doi:10.1016/j.apsusc.2015.09.235 APSUSC 31449

To appear in:

APSUSC

Received date: Revised date: Accepted date:

15-2-2015 18-9-2015 28-9-2015

Please cite this article as: D. Gao, M. Jia, Preparation of hierarchical porous Zn-salt particles and their superhydrophobic performance, Applied Surface Science (2015), http://dx.doi.org/10.1016/j.apsusc.2015.09.235 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

 

Preparation of hierarchical porous Zn-salt particles and their superhydrophobic performance Dahai Gao, Mengqiu Jia*

cr

*Corresponding author. Tel.: +86 010-64438715

ip t

Institute of Materials Science, Beijing University of Chemical Technology, Beijing, 100029

us

E-mail address: [email protected] (M. Jia) Abstract

an

Superhydrophobic surfaces arranged by hierarchical porous particles were prepared using modified hydrothermal routes under the effect of sodium citrate. Two particle samples were

M

generated in the medium of hexamethylenetetramine (P1) and urea (P2), respectively. X-ray diffraction, scanning electron microscope, and transmission electron microscope were adopted for

d

the investigation, and results revealed that the P1 and P2 particles are porous microspheres with

Ac ce pt e

crosslinked extremely thin (10 nm to 30 nm) sheet crystals composed of Zn5(OH)8Ac2·2H2O and Zn5(CO3)2(OH)6, respectively. The prepared particles were treated with a fluoroethylene vinyl ether derivative and studied using Fourier transform infrared spectroscopy and energy-dispersive X-ray spectrometer. Results showed that the hierarchical surfaces of these particles were combined with low-wettable fluorocarbon layers. Moreover, the fabricated surface composed of the prepared hierarchical particles displayed considerably high contact angles, indicating great superhydrophobicity for the products. The wetting behavior of the particles was analyzed with a theoretical wetting model in comparison with that of chestnut-like ZnO products obtained through a conventional hydrothermal route. Correspondingly, this study provided evidence that high roughness surface plays a great role in superhydrophobic behavior.

Key Words Superhydrophobic

material,

Hydrothermal

process,

Hierarchical

structure,

Surface

 

Page 1 of 21

 

modification 1.

Introduction The elaborate structures in natural biological organisms effectively stimulate researchers to

ip t

develop new advanced functional materials. In fact, the technology for fabricating superhydrophobic materials has become a vital research field in recent years, for which natural

cr

materials such as lotus leaf [1] and water strider legs [2] provide great inspirations. The superhydrophobic surfaces with a water contact angle (CA) larger than 150° show emerging

us

applications in self-cleaning and antifouling [3–7], solar panels [8], and fluid flow reduction [9–11]. Several reports have indicated that low-energy surface is essential for forming the

an

superhydrophobic performance, but surface roughness plays a more vital role for enhancing it [12–15]. Hierarchical architectures are widely used in fabricating superhydrophobic materials

M

because they provide maximum roughness to material surfaces [5, 16, 17]. Many synthesis routes are applied to form hierarchical architectures, followed by surface treatments, to transform the surface into a low-energy composition to satisfy both factors of

d

superhydrophobicity. Among these synthesis processes, the facile and low energy-cost routes for

Ac ce pt e

establishing a dual-scale roughness draw considerable attention in terms of fabricating superhydrophobic materials. During the last few decades, the growth of metal oxide crystals in liquid phase, particularly the hydrothermal process induced by surfactant, has become an effective method for developing hierarchical structures. In fact, numerous studies have claimed that various metal oxides could be used to establish hierarchical architecture [18–21]. Among these compounds, hierarchical ZnO has been widely investigated into and adopted [22–24] because of its low cost and environmentally friendly properties. This advanced material not only shows valuable properties in sensor application [25–29], selective adsorption [30], and catalysis [31, 32], but it is also a good candidate in superhydrophobic application. Their simple surface treatment with low-wettable substances (especially with fluorine-containing molecules) can be easily conducted on inorganic particles [33–36]. A few studies have explored the superhydrophobic surfaces synthesized from the growth of ZnO crystals in liquid phase [37, 38]. However, these crystals usually form columnar nanorods  

Page 2 of 21

 

that could assemble and form chestnut or flowerlike spherical particles, whose morphology is limited by the hexagonal crystal type of ZnO [24, 27, 39–41]. This structure has a relatively fixed crystal size, which makes it inconvenient to control the diameter and distribution density of the prepared nanorods. Therefore, the lack of dimension adjustment obviously cannot help enhance

ip t

the superhydrophobic performance. To overcome this defect, the conventional hydrothermal process should be modified to produce an entirely different set of hierarchical particles composed of another type of crystals.

cr

In some studies, the products of hierarchical particles were prepared from Zn(II) precursors

us

using the modified hydrothermal routes. The obtained particles assembled by nanosheet crystals showed a flowerlike, porous structure. Sodium citrate was adopted as the structure indicator,

an

mixed with hexamethylenetetramine [42–44] and urea [45–47] leading to an ordered growth during the hydrothermal reaction. The morphology of the products exhibited sufficiently high

M

roughness because the crystals of the modified hydrothermal route products varied from those of the columnar type of ZnO. Nonetheless, these prepared particles have not been applied in the superhydrophobicity field. Sun et al. [48] prepared another type of particles using block aqueous

d

polymer P123 as the structure indicator under solvothermal condition and surprisingly

Ac ce pt e

demonstrated the superhydrophobic application. However, this preparation process seems complicated, time-consuming and low-yield.

Correspondingly, researchers aim to obtain hierarchical particles containing different types of

crystals from Zn(II) precursors using the modified hydrothermal process to establish their superhydrophobic applications. In this study, the hierarchical particles were formed assisted by a soft template (i.e., sodium citrate). The modified hydrothermal process could induce a new type of sheet crystal with extremely low thickness and high porosity under certain alkali compounds (e.g., hexamethylenetetramine and urea), thereby generating high roughness. Two alkali media would induce significant differences in the micro- and nanoscale dimensions of the hierarchical particles. Subsequently, the surface modification by fluorocarbon polymer derivative could guarantee that the rough surface would be covered by the low-energy layer. These steps could guarantee that the surfaces constituted by such hierarchical particles produce considerably high CAs. This study thereby provides a facile and low-cost strategy for constituting effective superhydrophobic  

Page 3 of 21

 

surfaces.

2

Experimental

2.1 Materials

ip t

Zinc acetate dihydrate (ZnAc2·2H2O, 99%) and isocyanate-propyl-triethoxysilane (IPTES, 96%) were purchased from J&K Chemical Co., Ltd. Hexamethylenetetramine (C6H12N4, 99%)

cr

and urea (CO(NH2)2, 99%) were provided by Xilong Chemical Co., Ltd. Sodium citrate

us

(Na3C6H5O7·2H2O) was purchased from Beijing Chemical Works, and triethylamine (Et3N) was obtained from Tianjin Fu Chen Chemical Reagents Factory. Xylene and ethanol were provided by Beijing Yili Fine Chemical Co., Ltd. All chemicals listed above were used without further

an

purification. Fluoroethylene vinyl ether (FEVE) polymer (GK570, F content = 35 wt.% to 40 wt.%, number-average molecular weight = 14 000, hydroxyl value = 63.8 mgKOH/g, acid value =

M

4.1 mgKOH/g, solid content = 65 wt.%) was supplied by Daikin Fluorochemicals (China) Co.,

d

Ltd.

Ac ce pt e

2.2 Preparation of hierarchical particles

Sample P1: The mixture of ZnAc2·2H2O (6.036 g, 27.5 mmol), C6H12N4 (3.855 g, 27.5

mmol), and Na3C6H5O7·2H2O (0.809 g, 27.5 mmol) was dissolved in 250.0 g of deionized water. The reaction mixture was maintained at 20 °C to 25 °C for 15 min to 30 min under vigorous stirring. Consequently, the mixture was transferred into a Teflon-lined autoclave and heated from ambient temperature to 95 °C for 5 h at a fixed temperature with increasing rate. The reaction mixture in the autoclave was then maintained at 95 °C for another 4 h. After autoclave cooling, white products were separated through vacuum filtration and washed ethanol. The products were dried at 80 °C for 6 h to 8 h under vacuum, yielding 1.43 g of white powder. Sample P2: The mixture of ZnAc2·2H2O (6.585 g, 30.0mmol), CO(NH2)2 (3.603 g, 60.0mmol), and Na3C6H5O7·2H2O (0.809 g, 27.5 mmol) was dissolved in 250.0 g of deionized water. The preparation for this mixture was the same as that for P1, but it yielded 2.03 g of white powder.  

Page 4 of 21

 

“Chestnut-like” ZnO hierarchical particles have been reported in previous research, which is cited in this paper as reference [51].

2.3 Import of FEVE derivative

ip t

FEVE (1.0 g) was injected into a three-neck round bottom flask with a condenser and magnetic stirring and diluted by 20 mL of xylene. Et3N (2.0 mL) was then added to the solution as

cr

a catalyst, followed by IPTES (281 μL, 1.14 mmol) at ambient temperature. The reactive mixture

was subsequently refluxed at 140 °C for 12 h. This process produced a dark yellow solution

us

(FEVE-IPTES).

The prescribed solution was extracted and placed in a three-neck round bottom flask with a

an

condenser and magnetic stirring. Consequently, the solution was diluted by xylene, so the total amount of solvent could reach 5.0 mL. The products formed through the hydrothermal route were

M

added into the solution, generating a turbid dispersion. The mixture was refluxed at 140 °C for 8 h. After the reaction process, the products were washed with additional xylene (40 mL) to deprive

d

the unreacted starting materials. After standing for some time until the particles deposition, the clear liquid in upper level was poured out. The remaining dispersion (~2 mL) was immediately

Ac ce pt e

drop-cast onto glass slides and dried naturally for 24 h to yield a superhydrophobic surface.

2.4 Characterization

X-ray diffraction (XRD) was performed on a D/Max 2500 V/PC diffractometer

(RigakuCor-poration, Japan) and Cu Kα targets (λ = 0.154 nm) at a scanning rate of 0.0202 s–1. The functional groups of specimen were tested by Fourier transform infrared (FTIR) and measured on a Bruker Vector 22 infrared spectrophotometer. The samples were prepared on a KBr pellet and scanned at a wave number ranging from 4000 cm–1 to 400 cm–1, with a resolution of 2 cm–1, at 25 °C. The morphology and micro-nanostructures of the samples were analyzed using a field emission scanning electron microscopy (SEM, Hitachi S-4700) and a transmission electron microscope (TEM, Hitachi H800). Moreover, the surface chemical composition of the samples was examined with an energy-dispersive X-ray spectrometer (EDS) on SEM. The

 

Page 5 of 21

 

Brunauer–Emmett–Teller (BET) nitrogen adsorption-desorption measurements were conducted on the ASAP 2460 Surface Area and Porosity Analyzer to obtain the specific surface area of the samples. The CA and CA hysteresis (CAH) measurements were used to exhibit the hydrophobicity of the specimen and performed on a JGW-360A instrument (Chengde, China) at 25 °C with 5 μL

Results and Discussion

cr

3

ip t

to 10 μL droplets of ultrapure water.

us

3.1 Preparation of hierarchical particles

Scheme 1 illustrates the preparation process of the hierarchical superhydrophobic surface constituted by porous particles. During this preparation, the growth of crystals under the sodium

an

citrate-induced hydrothermal reaction produced spherical particles assembled by crosslinked sheet crystals. The samples obtained from this modified hydrothermal process (i.e., P1 and P2)

d

representation.

M

displayed similar composition and surface topology, and was thereby depicted by the identical

Ac ce pt e

To satisfy superhydrophobility, the inorganic surface should be replaced by a low-energy substance through the surface treatment. Several researchers have used small molecules containing fluorine to convert the surface wettability. But most of these compounds have a major disadvantage of high cost. In comparison, a polymeric agent (i.e., FEVE resin) with the advantage of low cost and wide offering could therefore sufficiently engender low wettability to the material surfaces because of abundant fluorine atoms and some active –OH groups in its backbones. Therefore, we used the FEVE derivative [49] given by the following functional group transformation:

FE V E

O H + O C N (C H 2 ) 3 S i(O E t) 3 ⎯ ⎯ → FE V E

O O C -N H (C H 2 ) 3 S i(O E t) 3

This –Si(OEt)3 capped derivative significantly shows a great trend to react with the surface hydroxyl of porous particles to offer connections between the hierarchical particles and the fluorocarbon segments. The great repulsive force of fluorine atom results in low-energy

 

Page 6 of 21

 

performance. In that case, the products could exhibit superhydrophobic performance because of their high roughness and low-energy surface composition.

3.2 Morphology and Performance

ip t

Apart from chemical composition, a suitable stereo shape is relevant to wetting performance because the roughness of particles is one of the crucial factors that affect the superhydrophobic

cr

surface. As such, the morphology of the typical products (i.e., P1 and P2) of the modified

hydrothermal route was thoroughly examined in this study. Figs. 1 and 2 demonstrate the

us

morphologies of P1 and P2, respectively. These figures particularly show that the SEM images of both samples reveal regular sphere particles in microscale for the product. The SEM images

an

illustrate as well the porous surface constituted by the nanoscale crosslinked sheet crystals. The prepared particles have a diameter of 3 μm to 6 μm, in which the small sheet crystals for P1 are

M

only 10 nm to 20 nm thick. The amplified sheet structure illustrated in Fig. 1C is seemed distorted because of the high-energy electron beam, which does not occur at low magnification. This

d

observation sufficiently demonstrates the thickness scale of the nanosheets. Therefore, the product exhibits an obvious hierarchical structure and a considerably high porosity. The TEM images (Fig.

Ac ce pt e

1D-E) prove that the particles are generated as hollow spheres because the central part of one particle seems lighter than its outer parts. A few broken particles in SEM images also support this architecture. Furthermore, the sheet crystals are distributed around the deepest part representing the hollow cores of the particles. In this case, the hierarchical structure offers the prepared particles great roughness, which becomes the foundation for the formation of the superhydrophobic surface. Similar morphology is observed in Fig. 2, which displays the microscope photographs of the P2 particles. These particles also clearly exhibit spherical shape and sheet crystals. Nonetheless, the dimensions of P2 vary from those of P1. The diameters of P2 particles often surpass 10 μm, and the thickness of their sheet crystals could reach up to 30 nm. The TEM images also show that the P2 particles have translucent, acute sheets and nearly dark central parts (Figs. 2D and 2E), indicating electron beam cannot traverse the center of these particles. This phenomenon suggests that the P2 samples probably have solid inner cores.

 

Page 7 of 21

 

However, the observation of sample P2 was usually disturbed by an electric discharge, which could have been caused by the charge accumulation on the sharp edges around the sheet crystals. These different dimension and shape features probably enhanced the superhydrophobic performance of P2 because of the increase in roughness.

ip t

The XRD patterns of the samples were detected to verify the chemical composition of the hierarchical particles. Fig. 3A depicts that the pattern of P1 shows a two series of diffraction peaks. The Miller indices for the first series were marked above the peaks. These narrow and strong

cr

peaks should be assigned to a highly crystallized ZnO because of the consistence of this chemical

us

compound with the standard card JCPDS No.36-145. By contrast, the peaks marked below the pattern, for which the distribution seems wide, are in agreement with Zn5(OH)8Ac2·2H2O particles

an

as reported by Xia et al. [43]. This result suggests that the generated Zn5(OH)8Ac2·2H2O crystals have a low crystallite size. Scherrer’s formulation to compute crystal size is expressed as (1)

M

D = 0.89λ/βcosθ

where λ is the wavelength of the X-rays (i.e., for CuKα, λ = 1.54 nm), β is the full width at half maximum (FWHM, rad), and θ is the diffraction angle corresponding to the selected peak. The

d

preceding formula was used to obtain the average crystal size of Zn5(OH)8Ac2·2H2O (i.e., 14 nm)

Ac ce pt e

from the diffraction peak at 2θ = 59.1°. This measurement obviously corresponds to the thickness of the sheets. However, compared with this crystal size, the size of ZnO crystals has a significantly far smaller dimension (i.e., up to 58 nm) calculated from the peak at 2θ = 31.8° using the same formula. The other diffraction peaks attributed to ZnO crystals can also lead to similar dimensions. In particular, the size of these diffraction peaks surprisingly surpasses the thickness of the sheet crystals measured from the SEM images. Only the core region of the particles could contain ZnO crystals, whereas the sheet crystals are almost composed of pure Zn5(OH)8Ac2·2H2O. Fig. 3B shows the XRD pattern of the P2 particles. This pattern confirms that the chemical composition of P2 is Zn5(CO3)2(OH)6, which is consistent with the report of Chen and Wang [45]. The crooked baseline in this pattern verifies the low degree of crystallinity. Using Scherrer’s formulation, we determined that the peak (200) of P2 produced a smaller crystal size (6 nm) than that of P1. This finding suggests that this type of particle is formed by a large amorphous region and some extremely tiny crystals. Such a large difference shown by P2 particles could lead to a change in  

Page 8 of 21

 

superhydrophobic performance. Although significant differences exist in the structure of both samples, their spherical shapes constituted by sheet crystals are similar. Therefore, these samples have nearly the same formation process. C6H12N4 and CO(NH2)2 are generally regarded as alkali, which yields OH–. At the beginning of the hydrothermal process, the seeds participate from Zn(OH)42– after the Zn(II)

ip t

precursor is converted into Zn(OH) 42– and is induced by OH-–. Surfactant addition (which is

referred to as sodium citrate in this study) is considered an essential condition that contributes to

cr

the growth of sheet crystals on those seeds through the soft-template effect. We have ever found

us

that the hydrothermal process only produce a few amount of hexagonal crystals of ZnO without the assistance of sodium citrate. All the sheet crystals then combine with one another to form a

an

porous framework during the period of crystals growth. For sample P1, the central parts of the cores could partially dissolve and transform back into Zn(II) ions and could recrystallize into sheet

M

crystals at the outer layer, thereby producing hollow particles with a hierarchical architecture. As mentioned in the preceding paragraphs, the sheet crystals have low thickness, producing a considerably great specific surface area for the particles. Fig. 4 shows the results of the BET

d

nitrogen adsorption-desorption analysis with the data of pure chestnut-like ZnO particles obtained

Ac ce pt e

by previous work [51] as comparison. The curve shapes of the three samples indicate that their adsorption performance belongs to type II adsorption, in which the adsorption transforms from mono-layer into multi-layer with the increase of P/P0. However, using the modified hydrothermal route, the samples have a great specific area, which is largely far more than the level of the chestnut-like ZnO particles obtained through the conventional hydrothermal approach. The significant hysteresis cycle essentially indicates the porosity among the produced particles. On the contrary, the results of the chestnut-like ZnO particles show almost no hysteresis because of their limited specific area of cluster structure assembled by nanorods with a relatively large dimension. Compared with P2, P1 has a smaller dimension and more tiny thickness of sheets; thus, its higher specific area becomes an observable fact.

3.3 Import of FEVE derivative

 

Page 9 of 21

 

Hierarchical products need further surface treatment to satisfy their low-energy chemical composition. The commonly used agents for this treatment are fluorocarbon or long-chain aliphatic compounds, which are of small-molecule substance. Compared with the pure hydrocarbon molecule, fluorocarbon compounds show more stability during application.

ip t

As previously mentioned, we applied FEVE derivative as a polymeric modifying agent. FEVE resin is composed of aliphatic structure containing fluorine atoms and active –OHs in side

chains. Using an –NCO functionalized molecule IPTES to condensation to yield urethane bonds,

us

molecules containing fluorine could modify an inorganic particle. 

cr

we functionalized the polymer backbones with the –Si(OEt)3 groups. Consequently, the polymeric

The FTIR spectra demonstrate the effect of the surface treatment. Fig. 5 shows the FTIR

an

spectra of ZnO/Zn5(OH)8Ac2·2H2O (P1) and Zn5(CO3)2(OH)6 (P2) particles before and after the surface modification using the FEVE derivative. Compared with the given spectra, the addition of

M

FEVE-IPTES imported new peaks at 2927 cm–1 for the modified samples, which could be attributed to the –CH2– imported by the IPTES molecule. In addition, the condensation reaction also generated –CO– bonds, corresponding to the adsorption at 1732 cm–1. The evidence of

d

Si-O-Si bonds shows their peak at 1099 cm–1. The relative intensity of –OHs adsorption in the

Ac ce pt e

region 3000 cm–1 to 3600 cm–1 weakens after the surface treatment. This condition could well explain the functional group transformation mentioned in Section 3. 1. Other than the FTIR results, the EDS pattern shown in Fig. 6 also provides evidence on the

effects of the surface treatment. For the untreated hierarchical particle samples, only the signals of Zn, O, and C were detected. Besides, the evidence of the F and Si atoms contained in the structure of FEVE-IPTES could be observed in the EDS pattern of the surface-treated samples (Figs. 6B and 6D). Table 1 lists the percentage composition of the surface elements corresponding to Fig. 6A-D. In particular, this table indicates that the O/Zn atom ratios of the untreated samples are 2.8, which approximately correspond to the inherent composition of ZnO/Zn5(OH)8Ac2·2H2O (P1) and Zn5(CO3)2(OH)6 (P2). However, for the surface-treated samples, this polymer modifier forms a monolayer of low-wettable molecules with a thickness of 10–10 m. Given the thickness of the sheet crystals (10 nm to 20 nm) and the detection range (～1 μm3) of EDS, the imported amount of F and Si atoms could only take up a small amount of surface composition, while the original  

Page 10 of 21

 

elements still largely dominate such region. Nevertheless, the results on the surface composition after the surface treatment show that the Zn/O (atom ratio) increases to 3.9 and 3.0 for P1 and P2, respectively, because the FEVE derivative imports abundant oxygen atoms onto the inorganic surface. Thus, the data for the element composition derived from the EDS patterns well support

ip t

the process of polymer molecules grafting. Therefore, the FTIR and EDS results clarify that the low-energy fluorocarbon polymeric layers were imported onto the hierarchical particles by the

cr

FEVE derivative.

us

3.4 Wetting Properties

The combination of the surface with the fluorocarbon polymer layer is simple to be

an

performed. However, the amount of the FEVE-IPTES used in this study influences the superhydrophobic performance of both the hierarchical particles. We applied an increasing amount

M

of FEVE-IPTES over the surface treatment. The effect of this undertaking is depicted in Fig. 7. The two line graphs in this figure illustrate that the CAs increase with respect to low FEVE

d

derivative amount, but both decrease when the FEVE/porous particles (w/w) become 1.0. In particular, the CAs reach a maximum value of 157±3° and 160°±3°, respectively, exhibiting an

Ac ce pt e

advanced superhydrophobic performance.

Fig. 7 also illustrates the evolution trend of the CAH, which displays an excellent minimum

value less than 8°, but could be more than 20° when insufficient and excess FEVE derivative is used, respectively. Figs 7A and 7B display that the minimum CAHs could be obtained at FEVE/particles (w/w) is 0.6 and 0.2, respectively. These results reflect a common feature that the superhydrophobic behavior strongly depends on the dose used for the surface treatment. Insufficient surface treatment (i.e., dose lower than horizontal axis of minimum CAH) could evidently leave the partial hydrophilic inorganic surface. However, for another case, the excess treatment (i.e., dose larger than horizontal axis of minimum CAH) yields several fluoro-containing molecular chains, but it produces residual functional groups, which induce a change in the surface chemical composition. Because the residual groups could be associated with water drop, the wetting surface exhibits more adhesive force to water, and could cause the surface partly convert

 

Page 11 of 21

 

into hydrophilicity. These effects incur the surface deterioration of the prepared superhydrophobic materials. The distinct of results displayed in two line graphs should also be noticed. The minimum points on both graphs may be attributed to different specific area of P1 and P2. P2 has a lower specific surface area and could realize optimal performance using less dose for surface

ip t

treatment. Correspondingly, the morphology and wetting state of the chestnut-like ZnO particles obtained through the conventional hydrothermal route were compared with those of the two

cr

hierarchical porous particle products understudy. This comparison is clearly shown in Fig. 8,

us

which reveals the great influence of the micro- and nanoshapes on the superhydrophobic performance. The porous particles show higher CAs than the chestnut-like ZnO particles. Given

an

that hierarchical morphology is important, it should be further illustrated theoretically. To our knowledge, two essential models have been applied to describe the wetting state (i.e., Wenzel and

M

Cassie–Baxter model). Compared with the Wenzel model, the Cassie–Baxter model is more effective when an airpocket is formed between the solid and liquid phases. The insertion of airpockets reduces the adhesion effect, thereby considerably reducing the CA hysteresis. The CA

d

for a wetting state conforming to the Cassie–Baxter model is given by the following equation:

Ac ce pt e

cos θ = f S L R f c os θ 0 + f S L − 1

（2）

where θ0 is the CA of the slide materials, θ is the apparent CA, and Rf is the surface

roughness of the solid–liquid interface. The solid–liquid fraction fSL represents the solid–liquid interface ratio in the whole wetting area. The sheet crystals and the great area of pores contribute to the production of high CAs. The observed morphology of P1 and P2 has sheet crystals of small thickness in the sphere particles (10 nm to 30 nm). If the liquid drop could only contact solid surface at the outer edge of the sheet crystals, then an extremely low fSL value could be obtained. Contrary to the nanorod clusters in the chestnut-like particles, the diameter of rods reaches 100 nm, inducing large fSL and relatively low CAs. The CA and CAH variation of P2 was determined to be advantageous than that of P1. This observation indicates that the micro- and nanostructural units may properly match each other to reduce the practical solid–liquid interface. The durability of the prepared superhydrophobic surface should be considered, even for the

 

Page 12 of 21

 

prepared sample closest to the ideal wetting state, because it is related to the practical service life. To test these properties, the CA and CAH evolution with respect to immersion time was recorded (Fig. 9). The CA values tend to decrease gradually with the increase of immersion time, whereas the CAH values increase with the immersion processing. The same trends have been noted by

ip t

certain studies on superhydrophobicity [36, 50]. The prepared surface could undergo gradual penetration during water immersion because the fluorocarbon layer cannot cover the whole hierarchical surface. The strong exclusion of fluorocarbon chains and the existence of airpockets

cr

restrain the direct contact of water and hierarchical surface, and they prevent water from

us

penetrating into the internal parts of the hierarchical structure. When wetting and adsorption occur in certain defects, the nonreversible hydration of the local region would gradually increase. This

an

penetrating process comes into a relative balance for both the prepared samples within about four days. After this immersion time, the line charts show a significant turning point, that is, only small

M

changes in CA and CAH could be observed. In general, the superhydrophobicity of the prepared samples do not significantly deteriorate. In sum, the hierarchical particles obtained through the hydrothermal route only carry hydrophilic surface, and the surface treatment causes

4.

Ac ce pt e

d

water-repellency to this hierarchical structure.

Conclusion

Special hierarchical sphere particles constituted by nanosheet crystals are prepared through

the modified hydrothermal route. A typical product has a huge specific area, which is far more than that of the chestnut-like ZnO particles based on the conventional hydrothermal process. The surface treatment with the FEVE derivative is facile and generates a fluorocarbon molecule layer on the prepared particles, leading to a low-energy composition on the hierarchical rough surface. This combination generates high CAs close to 160°. These results indicate that the modified hydrothermal route decreases the dimension of crystals, further reducing the solid–liquid interface area. In this case, the high superhydrophobic performance could be guaranteed.

Acknowledgements

 

Page 13 of 21

 

The authors acknowledge the support provided by the Beijing Municipal Natural Science Foundation funded project (2152024) and the China Postdoctoral Science Foundation funded project (2014M550593). The authors are also grateful to the Fundamental Research Funds for the Central Universities (ZY1406) for their assistance.

ip t

References

cr

[1] B. Bhushan, Y.C. Jung, Micro- and nanoscale characterization of hydrophobic and hydrophilic leaf surfaces, Nanotechnology, 17 (2006) 2758.

us

[2] X. Gao, L. Jiang, Biophysics: water-repellent legs of water striders, Nature, 432 (2004) 36-36. [3] P.F. Rios, H. Dodiuk, S. Kenig, S. McCarthy, A. Dotan, Durable ultra-hydrophobic surfaces for

an

self-cleaning applications, Polymer Advanced Technology, 19 (2008) 1684-1691. [4] B. Bhushan, Y.C. Jung, K. Koch, Micro-, nano- and hierarchical structures for

M

superhydrophobicity, self-cleaning and low adhesion, Philosophical Transactions of the Royal Society of London, A, 367 (2009) 1631-1672.

d

[5] Y.C. Jung, B. Bhushan, Mechanically durable carbon nanotube-composite hierarchical structures with superhydrophobicity, self-cleaning, and low-drag, ACS Nano, 3 (2009) 4155-4163.

Ac ce pt e

[6] T. Kamegawa, Y. Shimizu, H. Yamashita, Superhydrophobic surfaces with photocatalytic self-cleaning properties by nanocomposite coating of TiO2 and polytetrafluoroethylene, Advanced

Materials, 24 (2012) 3697-3700.

[7] Y.C. Jung, B. Bhushan, Wetting behavior of water and oil droplets in three-phase interfaces for hydrophobicity/philicity and oleophobicity/philicity, Langmuir, 25 (2009) 14165-14173.

[8] H.C. Barshilia, S. John, V. Mahajan, Nanometric multi-scale rough, transparent and anti-reflective ZnO superhydrophobic coatings on high temperature solar absorber surfaces, Solar Energy Materials and Solar Cells, 107 (2012) 219-224.

[9] B. Su, M. Li, Q. Lu, Toward understanding whether superhydrophobic surfaces can really decrease fluidic friction drag, Langmuir, 26 (2009) 6048-6052. [10] Z. Ming, L. Jian, W. Chunxia, Z. Xiaokang, C. Lan, Fluid drag reduction on superhydrophobic surfaces coated with carbon nanotube forests (CNTs), Soft Matter, 9 (2011) 4391-4396.

 

Page 14 of 21

 

[11] M. Cheng, M. Song, H. Dong, F. Shi, Surface adhesive forces: a metric describing the drag-reducing effects of superhydrophobic coatings, Small, 11 (2014) 1665-1671. [12] E. Celia, T. Darmanin, E. Taffin de Givenchy, S. Amigoni, F. Guittard, Recent advances in designing superhydrophobic surfaces, Journal of Colloid Interface Science, 402 (2013) 1-18.

ip t

[13] M.S. Mozumder, H. Zhang, J. Zhu, Mimicking lotus leaf:development of micro-nanostructured biomimetic superhydrophobic polymeric surfaces by ultrafine powder coating technology, Macromolecular Materials Engineering, 296 (2011) 929-936.

cr

[14] P. Ragesh, V. Anand Ganesh, S.V. Nair, A.S. Nair, A review on self-cleaning and multifunctional

us

materials', Journal of Materials Chemistry A, (2014).

[15] Y. Yan, N. Gao, W. Barthlott, Mimicking natural superhydrophobic surfaces and grasping the

Colloid Interface Science, 169 (2011) 80-105.

an

wetting process: A review on recent progress in preparing superhydrophobic surfaces, Advanced

M

[16] P. Brown, E. Talbot, T. Wood, C. Bain, J. Badyal, Superhydrophobic hierarchical honeycomb surfaces, Langmuir, 28 (2012) 13712-13719.

[17] M. Raza, E. Kooij, S. van, H. Zandvliet, B. Poelsema, A colloidal route to fabricate hierarchical

d

sticky and non-sticky substrates, Journal Colloid Interface Science, 385 (2012) 73-80.

Ac ce pt e

[18] X. Zhou, S. Chen, D. Zhang, X. Guo, W. Ding, Y. Chen, Microsphere organization of nanorods directed by PEG linear polymer, Langmuir, 22 (2006) 1383-1387.

[19] J. Liu, X. Huang, Y. Li, Z. Li, Q. Chi, G. Li, Formation of hierarchical CuO microcabbages as stable bionic superhydrophobic materials via a room-temperature solution-immersion process, Solid State Science, 10 (2008) 1568-1576.

[20] Y. Wang, B. Li, T. Liu, C. Xu, Z. Ge, Controllable fabrication of superhydrophobic TiO2 coating with improved transparency and thermostability, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 441 (2014) 298-305.

[21] J. Wang, P.S. Lee, J. Ma, One-Pot Synthesis of hierarchically assembled tungsten oxide (hydrates) nano/microstructures by a crystal-seed-assisted hydrothermal process, Crystal Growth & Design, 9 (2009) 2293-2299. [22] X. Zhou, D. Zhang, Y. Zhu, Y. Shen, X. Guo, W. Ding, Y. Chen, Mechanistic investigations of peg-directed assembly of one-dimensional ZnO nanostructures, The Journal of Physical Chemistry  

Page 15 of 21

 

B, 110 (2006) 25734-25739. [23] L. Chen, Y. Chuang, Hydrothermal synthesis and characterization of hexagonal zinc oxide nanorods with a hexamethylenetetramine (HMTA) template-assisted at a low temperature, Mater. Lett., 68 (2012) 460-462.

ip t

[24] X. Su, H. Zhao, F. Xiao, J. Jian, J. Wang, Synthesis of flower-like 3D ZnO microstructures and their size-dependent ethanol sensing properties, Ceramics International, 38 (2012) 1643-1647.

[25] X. Luo, Z. Lou, L. Wang, X. Zheng, T. Zhang, Fabrication of flower-like ZnO

cr

nanosheet/nanorod-assembled hierarchical structure and enhanced performance in gas sensor,

us

New Journal of Chemistry, (2013).

[26] H. Zhao, X. Su, F. Xiao, J. Wang, J. Jian, Synthesis and gas sensor properties of flower-like 3D

an

ZnO microstructures, Materials Science and Engineering: B, 176 (2011) 611-615. [27] P. Rai, Y. Yu, Synthesis of floral assembly with single crystalline ZnO nanorods and its CO

M

sensing property, Sensors and Actuators B: Chemical, 161 (2012) 748-754. [28] X. Chen, X. Jing, J. Wang, J. Liu, D. Song, L. Liu, Self-assembly of ZnO nanoparticles into hollow microspheres via a facile solvothermal route and their application as gas sensor,

d

CrystEngComm, 15 (2013) 7243-7249.

Ac ce pt e

[29] H. Zhang, R. Wu, Z. Chen, G. Liu, Z. Zhang, Z. Jiao, Self-assembly fabrication of 3D flower-like ZnO hierarchical nanostructures and their gas sensing properties, CrystEngComm, 14 (2012) 1775-1782.

[30] K.V. Murali, A.J. Ragina, K.C. Preetha, K. Deepa, T.L. Remadevi, Influence of ammonia, lithium hydroxide, and hexamine on ZnO films synthesized by successive ionic layer adsorption and reaction technique, Journal of Materials Science, 48 (2012) 1852-1861.

[31] Y. Lai, M. Meng, Y. Yu, X. Wang, T. Ding, Photoluminescence and photocatalysis of the flower-like nano-ZnO photocatalysts prepared by a facile hydrothermal method with or without ultrasonic assistance, Applied Catalysis B: Environmental, 105 (2011) 335-345. [32] M. Wang, Y. Zhang, Y. Zhou, F. Yang, E.J. Kim, S.H. Hahn, S.G. Seong, Rapid room-temperature synthesis of nanosheet-assembled ZnO mesocrystals with excellent photocatalytic activity, CrystEngComm, 15 (2013) 754-763. [33] L. Xu, R.G. Karunakaran, J. Guo, S. Yang, Transparent, Superhydrophobic surfaces from one-step  

Page 16 of 21

 

spin coating of hydrophobic nanoparticles, Acs Applied Materials Interfaces, 4 (2012) 1118-1125. [34] H. Zhou, H. Wang, H. Niu, A. Gestos, X. Wang, T. Lin, Fluoroalkyl silane modified silicone rubber/nanoparticle composite: A super durable, robust superhydrophobic fabric coating, Advanced Materials, 24 (2012) 2409-2412.

ip t

[35] L. Gao, J. He, A facile dip-coating approach based on three silica sols to fabrication of broadband antireflective superhydrophobic coatings, Journal of Colloid Interface Science, 400 (2013) 24-30.

[36] N. Valipour Motlagh, F.C. Birjandi, J. Sargolzaei, N. Shahtahmassebi, Durable, superhydrophobic,

cr

superoleophobic and corrosion resistant coating on the stainless steel surface using a scalable

us

method, Applied Surface Science, 283 (2013) 636-647.

[37] J. Zhang, J. Zhang, A facile method for preparing a non-adhesive superhydrophobic ZnO nanorod

an

surface, Materials Letter, 93 (2013) 386-389.

[38] J. Wu, J. Xia, W. Lei, B. Wang, Superhydrophobic surface based on a coral-like hierarchical

M

structure of ZnO, PLoS One, 5 (2010) e14475.

[39] Y. Wang, M. Li, Hydrothermal synthesis of single-crystalline hexagonal prism ZnO nanorods, Materials Letter, 60 (2006) 266-269.

d

[40] Y. Chen, R. Yu, Q. Shi, J. Qin, F. Zheng, Hydrothermal synthesis of hexagonal ZnO clusters,

Ac ce pt e

Materials Letter, 61 (2007) 4438-4441.

[41] F. Wang, X. Qin, D. Zhu, Y. Meng, L. Yang, Y. Ming, PEG-assisted hydrothermal synthesis and photoluminescence of flower-like ZnO microstructures, Materials Letter, 117 (2014) 131-133.

[42] Z. Xia, J. Sha, Y. Fang, Y. Wan, Z. Wang, Y. Wang, Purposed Built ZnO/Zn5(OH)8Ac2·2H2O Architectures by Hydrothermal Synthesis, Crystal Growth & Design, 10 (2010) 2759-2765.

[43] Z. Xia, Y. Wang, Y. Fang, Y. Wan, W. Xia, J. Sha, Understanding the origin of ferromagnetism in zno porous microspheres by systematic investigations of the thermal decomposition of Zn5(OH)8Ac2·2H2O to ZnO, Journal of Physical Chemistry C, 115 (2011) 14576-14582.

[44] Y. Fang, Z. Xia, F. Yu, J. Sha, Y. Wang, W. Zhou, Formation mechanism of hollow microspheres consisting of ZnO nanosheets, CrystEngComm, 14 (2012) 8615-8619. [45] X. Wang, W. Cai, G. Wang, C. Liang, Standing porous ZnO nanoplate-built hollow microspheres and kinetically controlled dissolution/crystal growth mechanism, Journal of Materials Research, (2012) 1-8.  

Page 17 of 21

 

[46] X. Wang, W. Cai, S. Liu, G. Wang, Z. Wu, H. Zhao, ZnO hollow microspheres with exposed porous nanosheets surface: Structurally enhanced adsorption towards heavy metal ions, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 422 (2013) 199-205. [47] S. Das, K. Dutta, A. Pramanik, Morphology control of ZnO with citrate. A time and concentration

ip t

dependent mechanistic insight, CrystEngComm, 15 (2013) 6349-6358. [48] Z. Sun, T. Liao, K. Liu, L. Jiang, J. Kim, S. Dou, Robust superhydrophobicity of hierarchical ZnO hollow microspheres fabricated by two-step self-assembly, Nano Research, 6 (2013) 726-735.

cr

[49] F. Piccinini, M. Levi, S. Turri, Photoactive sol–gel hybrid coatings from modified fluorocarbon

us

polymers and amorphous titania, Progress of Organic Coating, 76 (2013) 1265-1272.

[50] S. Yuan, S. Pehkonen, B. Liang, Y. Ting, K. Neoh, E. Kang, Superhydrophobic

an

fluoropolymer-modified copper surface via surface graft polymerisation for corrosion protection, Corrosion Science, 53 (2011) 2738-2747.

M

[51] Gao D, Jia M. Hierarchical zno particles grafting by fluorocarbon polymer derivative: Preparation

Ac ce pt e

d

and superhydrophobic behavior[J]. Appl. Surf. Sci., 343 (2015): 172-180.

 

Page 18 of 21

 

Figure and Table Captions

Scheme 1 Preparation of the hierarchical superhydrophobic surface constituted by the prepared

ip t

porous sphere particles depicted by the blue and gray colors; these particles were obtained via the

cr

modified hydrothermal process and treated by the FEVE derivative FEVE-IPTES, yielding a CA

us

of up to 160°

Fig. 1 Morphology and XRD pattern of the prepared hierarchical particles (P1) through the

an

modified hydrothermal process: (A–C) SEM images, (D, E) TEM images

Fig. 2 Morphology and XRD pattern of the prepared hierarchical particles (P2) through the

M

modified hydrothermal process: (A–C) SEM images, (D, E) TEM images

d

Fig. 3 XRD pattern of the prepared hierarchical particles: (A) ZnO/Zn5(OH)8Ac2·2H2O (P1); (B)

Ac ce pt e

Zn5 (CO3)2(OH)6 (P2)

Fig. 4 N2 adsorption and desorption isotherm of the hierarchical particles containing Zn(II): (A) “chestnut”

ZnO

particles

through

the

conventional

hydrothermal

route;

(B)

ZnO/Zn5(OH)8Ac2·2H2O particles (P1); (C) Zn5 (CO3)2(OH)6 particles (P2) Fig. 5 FTIR spectra of the prepared hierarchical particles: (a) untreated P1, (b) P1 treated with the FEVE derivative, (c) untreated P2, (d) P2 treated with the FEVE derivative Fig. 6 EDS patterns of the prepared hierarchical particles: (A) untreated P1, (B) P1 treated with the FEVE derivative, (C) untreated P2, (D) P2 treated with the FEVE derivative Fig. 7 Variations of CA and CAH of water droplets on the prepared superhydrophobic surface constituted by (A) P1 and (B) P2 treated by various amounts of FEVE derivative

 

Page 19 of 21

 

Fig. 8 Comparison of SEM images and water wetting state of the prepared superhydrophobic surfaces constituted by (A) P1, (B) P2, (C) “chestnut” ZnO particles Fig. 9 CA and CAH variations of the prepared products (A) P1, (B) P2 with the increase of water

ip t

immersion time

treatment

P2, untreated

   

32.35

Atom%

16.24

45.47

Weight%

37.22

35.35

Atom%

11.48

44.52

Weight%

52.43

Atom%

19.92

Weight% Atom%

F

us

47.21

Si

20.44

-

-

38.28

-

-

24.41

2.55

0.48

40.95

2.70

0.34

35.45

12.12

-

-

55.02

25.06

-

-

49.18

35.96

12.17

2.48

0.21

18.12

54.14

24.40

3.15

0.18

Ac ce pt e

P2, treated

Weight%

C

an

P1, treated

O

M

P1, untreated

Zn

d

Samples

cr

Table 1 Chemical composition of the prepared hierarchical particles before and after the surface

Highlight 

Hierarchical particles with high roughness were prepared by modified hydrothermal route.

The high roughness is provided by extremely low thickness of sheet crystals.

FEVE polymer derivative was used for surface treatment of hierarchical surface.

 

Page 20 of 21

 

The novel particles via surface treatment were firstly used as superhydrophobic materials.

The product properties were compared with multi-scale ZnO particles via conventional route.

 

ip t

 

Ac ce pt e

d

M

an

us

cr

 

 

Page 21 of 21