Construction of hydrophobic surfaces on polyvinylidene fluoride film and cotton fabric using perfluorohexane functionalized carbon nanotubes and graphene oxide

Abstract

Hydrophobic surfaces have great potential in applications in oil–water separation, super water/oil repellents, friction reducing, etc. Hydrophobic performance has been extensively investigated in view of smart textile development. Oxidized multi-walled carbon nanotubes (CNTs) and graphene oxide (GO) were grafted with perfluoro-1-iodohexane, and 10.24 and 17.65 at% fluorine contents of these functional products were obtained, respectively. The surface chemistry of the functionalized CNTs and GO were characterized by X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy. Hydrophobic textiles were achieved by treating the functionalized CNTs and GO using a common dip-dry method. The functionalized CNTs and GO were also applied to polyvinylidene fluoride filter paper by the vacuum filtration method to form hydrophobic films. The morphologies of these surfaces were characterized by field emission scanning electron microscopy. The functional cotton fabrics showed hydrophobicity with water droplet contact angles (CAs) of 149.1°and 154.4°, respectively. The produced films showed hydrophobicity with CAs of 108° and 151°, respectively. The difference of the CA was attributed to the diversity of both the structure and the chemical composition. In future study, multifunctional materials could be created on the basis of the hydrophobic surfaces reported in this paper by combining them with other functional components, which has great potential in applications in the smart textiles.

Keywords

perfluorohexane carbon nanotubes graphene oxide hydrophobic film cotton fabric

Carbon-based nanomaterials, such as carbon nanotubes (CNTs) and graphene, are widely exploited in both the scientific and engineering communities, because of their many potential applications due to their extraordinary physical and chemical properties.^1–4 For the construction of hydrophobic surfaces, many studies have been carried out on various applications such as oil–water separation,⁵ super water/oil repellents,⁶ friction reducing,⁷ tunable wettability,⁸ etc.

It is well recognized that the wetting behavior of material surfaces is the common effect of the chemical composition and the microscopic geometry of the surfaces. Surfaces fabricated by carbon materials have been explored on the strength of their intrinsic nanosize benefit.^6,9–11 To lower the surface energy, chemical functionalization of CNTs and graphene has become a focus of special interest. Perfluorinated compounds^11–13 and long-chain alkyl silane^7,14 are the two main substances used to modify carbon nanomaterials utilizing chemical or physical approaches to lower the surface energy.

Various methods have been proposed for constructing hydrophobic surface with CNTs and graphene, such as aerogel,¹² film,⁸ foam,¹⁵ gel,¹¹ and textile coating.^10,16 Cotton is one of the most commonly used textile materials with perfect water absorbability, which is a drawback in some application areas. Many studies have been carried out on improving cotton fabric hydrophobicity using these carbon-based materials in recent years. Georoakilas et al.¹³ reported the functionalization of CNTs by perfluoroalkylsilanes as well their implication on the wetting properties of cotton fabric and found that the contact angle (CA) of a water droplet of 170° and a roll off angle less than 5° leave the textile completely dry. Shateri-Khalilabad and Yazdanshenas¹⁶ deposited graphene oxide (GO) on cotton fabric by the dip-pad-dry method followed by reduction with ascorbic acid to yield a fabric with a layer of graphene and the fabric was then reacted with methyltrichlorosilane to form polymethylsiloxane nanofilaments on the fibers surface to produce superhydrophobic electroconductive graphene-cotton with a CA of 163°. Various techniques are available for the construction of a hydrophobic surface on cotton fabrics, including gas treatments,^17–19 such as infrared, ultraviolet (UV), high-energy electron beam and low pressure plasma; and wet modifications,^20–22 such as the sol-gel process, layer-by-layer deposition, coating of polymers, and in-situ introduction of nanomaterials. Compared with gas treatments, wet modification of textiles is more widely used for hydrophobic surface construction because the production process is simple and will not affect bulk properties.

The fluorocarbons mostly employed are derivatives based on Perfluorooctane Sulfonate (PFOS) and Perfluorooctanoic Acid (PFOA) that are known to be bio-persistent.^23–25 Use of them within the European Union is subject to authorisation under the REACH Regulation. Hexa-perfluorinated compounds have attracted more attention in both academic and commercial contexts. As the basis of further study aiming at environment-friendly multifunctional nanomaterial development, perfluoro-1-iodohexane was chosen in this study to modify hydroxylated multi-walled CNTs (MWCNTs) and ethanolamine grafted GO, which would strike a balance between the hydrophobic function and the environmental benign demand. The perfluoro-1-iodohexane attached carbon nanomaterials were first successfully synthesized and fabricated on cotton fabrics by repeating dip-dry procedures. For comparison purposes, films on polyvinylidene fluoride (PVDF) filter papers were also fabricated by the vacuum filtration method. Water CAs of the fabrics and the films were recorded. Cheap cotton fabrics with porous stretchable structures functionalized by CNT–F and GO–F could be used in the design and fabrication of superhydrophobic materials, such as microfluidic devices, traffic indicators, drag-reducing coatings, or other functional micro/nano devices. In a further study, multifunctional materials could be created on the basis of the hydrophobic surfaces reported in this paper by combining with other functional components, which has great potential in applications in flexible electronics.

Experimental details

Materials

Cotton fabric (80^S/2 × 80^S/2, 120 × 80 after desizing, scouring, and bleaching) was used without further purification. MWCNTs (main range of outer diameter: 8–15 nm, length: 0.5–2 µm) produced according to a regular Chemical Vapor Deposition (CVD) method were purchased from XFNANO Materials TECH Co., Ltd, Nanjing, China. Graphite powder (burned residue ≤0.15%, granularity (≤30 µm) ≥ 95%) and all other chemicals were purchased from Sinopharm Chemical Reagent Co., Ltd, China. All of the above-mentioned materials and Deionized (DI) water were directly used without further purification.

Preparation of perfluorohexane-functionalized CNTs (CNT–F)

MWCNTs were oxidized by a mixture of concentrated sulfuric acid and nitric acid (v:v = 1:1) at 140℃ for 1 h to introduce hydroxyl-dominated oxygen-containing groups onto the surfaces.²⁶ A total of 500 mg oxidized MWCNTs (CNT-OH) were stirred with 20 mL N,N-Dimethylformamide DMF in a 50 mL three-necked flask flushed with nitrogen. A total of 0.55 g NaH was added into this suspension at 0℃ and stirred for another 30 min. After adding 2.5 mL perfluoro-1-iodohexane dropwise, the reaction was continued at 90℃ for 24 h under sonication with a probe tip apparatus (1 s on, 5 s off). The product was then vacuum-filtered using 0.22 µm pore size polytetrafluoroethylene (PTFE) filter paper, and the residue on the filter paper was washed extensively with DMF, ethanol, and water successively. The sample (CNT–F) was then dried in a vacuum oven at 50℃ for 96 h.

Preparation of perfluorohexane-functionalized graphene oxide (GO–F)

GO was synthesized from natural graphite by oxidation with H₂SO₄/KMnO₄ according to the modified Hummers’ method. A total of 2 g graphite powder, 1 g sodium nitrate, and 46 mL concentrate sulfuric acids were mixed in a 1 L beaker that had been cooled to 0–5℃ in an ice-bath. While maintaining vigorous agitation, 6 g of potassium permanganate was added to the suspension in 1 h, and the mixture was stirred for another 1 h at 0–5℃. The temperature of the suspension was brought to 45℃ and maintained for 1 h after the ice-bath was removed. After that, 120 mL water was added dropwise and stirred for 30 min. The suspension was then further diluted by 280 mL water and treated with 20 mL hydrogen peroxide. Centrifugation was proceeded to remove the liquor and wash the yellowish-brown deposit with 2 L 1 M hydrochloric acid and 9 L water. The reaction of GO and ethanolamine was performed according to the literature.²⁷ A total of 1 g GO and 1.3 g ethanolamine were dispersed in 300 mL water at pH 1–2 adjusting by aqueous 1 M HCl. Next the mixture was stirred for 48 h at room temperature. Subsequently, the suspension was washed several times with water by centrifuging and dried in a vacuum oven overnight to get GO-OH. Then 400 mg GO-OH was stirred with 20 mL DMF in a 50 mL three-necked flask flushed with nitrogen. A total of 1 g NaH was added into this suspension at 0℃ and stirred for another 30 min. After adding 5 mL perfluoro-1-iodohexane dropwise, the reaction was continued at 90℃ for 24 h under sonication with a probe tip apparatus (1 s on, 5 s off). The product was washed extensively with DMF, ethanol, and water by centrifuging successively. The sample (GO–F) was then dried in a vacuum oven at 50℃ for 96 h.

Finishing procedure of the cotton fabric

A total of 200 mg of CNT–F or GO–F particles were sonicated in 10 mL water with a probe tip apparatus to form a homogeneous suspension. The cotton fabric (2 cm × 8 cm) was dipped in the suspension for 1 min, removed, and dried at 60℃. The dip-dry process was repeated three times and the final hydrophobic cotton fabric was obtained.

Characterization

Structure

The chemical composition and the structure of the samples were analyzed by the X-ray photoelectron spectroscopy (XPS, XSAM800, Kratos, UK) and Fourier transform infrared spectroscopy (FT-IR). XPS was performed with a monochromatic Al K_α X-ray source (1486.6 eV photons), operated at 180 W (12 kV and 15 mA) and a pressure of 2 × 10⁻⁷Pa. The survey spectra and detailed C1s spectra were obtained at a photoelectron take-off angle (α, with respect to the sample surface) of 20°. FT-IR spectra were recorded on a FT-IR spectrometer (Varian 040-IR, America) using KBr crystal in the infrared region 4000–500 cm^–1.

Morphology

The morphology was examined by field emission scanning electron microscopy (FE-SEM, S-4800, Hitachi Ltd, Japan). All samples for the measurements were prepared by fixing 5 mm × 5 mm sections to the support directly.

Water droplet CA measurement

The water droplet CA measurement was carried out at ambient temperature on a DSA30 CA system (Kruss, Germany) using a 5 µL water droplet. The reported CA is obtained by averaging the value measured on more than three different positions.

Thermal analysis

The thermal stability of the nanomaterials and fabric samples was examined by thermogravimetric analyzer (TGA, TG209 F1, NETZSCH, Germany) under a nitrogen flow. In each typical experiment, about 5 mg of samples was placed in a clean platinum pan and heated from room temperature to 800℃ at a rate of 10℃/min.

Results and discussion

Many uses for the high-performance carbon nanomaterial–polymer composites depend on an efficient loading of nanomaterials to the polymer matrix, which requires solution phase processing and manipulations to achieve homogeneous dispersions and strong interfacial bonding. CNTs can be well dispersed in common solvents by covalent or non-covalent modification. Oxidation by concentrated acid was the most useful method to obtain oxidized CNTs. For producing functional CNTs–polymer composites, the most common covalent approaches, such as etherification, amidation, and esterification, have been used for linking functional moieties onto oxidized CNTs. Perfluoro-1-iodohexane could react with oxidized CNTs by ether linkage if only massive hydroxyl groups were on the walls of MWCNTs. An appropriate oxidation process was chosen for this purpose.

Chemically modified GO has been studied to manufacture different functional graphene–polymer composites. The structure of GO has been the subject of debate for many years. There are multiple routes for preparing GO causing this ambiguity. It is generally agreed that hydroxyl and epoxy groups are two major oxygen-containing groups that mainly exist in the basal plane and with other moieties located at the edges, such as carboxyl and carbonyl groups. GO contains a range of reactive oxygen-containing groups, which render it a good candidate for chemical functionalization. Ethanolamine was attached onto the basal plane by the ring-opening reaction of epoxy groups to get GO–OH for increasing the amount of –OH.

Schematic descriptions of the modification processes of CNTs and GO are shown in Schemes 1 and 2, respectively. Perfluoro-1-iodohexane can engage in nucleophilic substitution with the hydroxyl group to form ether linkage catalyzed by the alkaline NaH, which proceeds with an S_RN1 mechanism.^28,29 After the functionalization, there are –COOH and –C=O groups still remaining at the edge of the GO³⁰ or the end of CNT–OH.²⁶ Hydrogen bonding interactions may form between those carbon nanomaterials and the polymer matrix, and could also improve the compatibility with water.
Scheme 1.
Functionalization of carbon nanotubes (CNTs) with perfluorohexane.

Scheme 2.
Functionalization of graphene oxide (GO) with perfluorohexane.

XPS analysis was performed on CNT-OH, CNT–F, GO, and GO–F samples and the spectra with the C1s high-resolution spectra are shown in Figure 1. The corresponding atomic sensitivity factors (ASFs) together with the individual peak areas have been taken into account to quantify the atomic concentration. The ASF values provided here for C 1 s, N 1 s, O 1 s, and F 1 s are 0.25, 0.42, 0.66, and 1.00, respectively. The intensity (I) of a photoelectron peak from the homogeneous solid referred to will usually be taken as the integrated area under the peak. Employing the ASF, the relative atomic concentration of i element (C_i) was calculated using equation (1)³¹
$C_{i} = \frac{I_{i} / {ASF}_{i}}{\sum_{j} I_{j} / {ASF}_{j}}$
(1)

Figure 1.
(a) X-ray photoelectron spectroscopy (XPS) survey spectra of carbon nanomaterials before and after functionalization and C 1s high-resolution XPS analysis of (b) CNT-OH; (c) CNT–F; (d) GO; (e) GO–F.

The calculated atom concentrations of F on the CNT–F and GO–F particles are 10.24% and 17.65% (Table 1), respectively, indicating a strong surface enrichment of fluorine. Further information could be inferred by curve fitting of the C 1 s spectra of the particles, whose results are listed in Table 2. All of the C 1 s high-resolution spectra exhibit the main C–C peak at around 284.6 eV. Curve-fitting studies carried out on the spectra of CNT–OH and GO particles reveal the appearance of oxygen-containing groups, including –OH, –COOH, and epoxy groups. For the CNT–F sample, the C1s spectrum represents the –CF₂ and –CF₃ in the perfluorohexane at the binding energy of 290.2 and 292.0 eV, respectively.³² The binding energy peak at 286 eV is attributed to C–O–C linkage³³ between CNTs and perfluorohexane chains. The typical peak at 288.5 eV indicates the existence of residue –COOH on the CNT–F surface.³⁴ For the GO–F sample, characteristic peaks of –CF₂, –CF₃, C–O–C, and C–N can also be found, which indicates that perfluorohexane was attached onto the GO surface successfully. The slight changes of the above-mentioned peaks compared to CNT–F were mainly caused by the different adjacent atoms. There are still many –COOH both on the surface of CNT–F and GO–F particles, which could increase their affinity with the matrix or the dissolution property in polar solvents.
Table 1.
Element contents of CNT–OH, CNT–F, GO, and GO–F particles read by X-ray photoelectron spectra and the corresponding analysis data.

Sample C (am%) O (am%) N (am%) F (am%) C/O C_f ratio^a

CNT–OH 85.43 14.57 – – 5.86 –

CNT–F 80.57 9.19 – 10.24 8.77 1

GO 73.88 26.12 – – 2.42 –

GO–F 68.09 10.21 4.05 17.65 6.67 2.3

a
Number of perfluorohexane chains per 100 carbon atoms on the CNT–F and GO–F particles.

CNT: carbon nanotube; GO: graphene oxide.

Table 2.
Binding energy and assignment of the contributions to the main X-ray photoelectron spectroscopy (XPS) peaks for CNT-OH, CNT–F, GO, and GO–F particles.

Sample XPS peak (eV) Assignment

CNT-OH 284.6 C–C

285.8 C–OH

287.6 O=C–O

289.5 COOH

CNT–F 284.6 C–C

286 C–O–C

288.5 O=C–O

290.2 –CF₂

292 –CF₃

GO 284.7 C–C

285.6 C–OH

286.7 Epoxy

287.8 O=C–O

288.9 COOH

GO–F 284.6 C–C

285.8 C–O–C

287.3 C–N

288.9 C=O

291.3 –CF₂

293 –CF₃

The FT-IR spectra (shown in Figure 2) can also give chemical structure information of the carbon nanomaterials. In the CNT–OH sample, the FT-IR spectra display stretching and bending bands of hydroxyl groups at 3420 and 1409 cm^–1, respectively. For the CNT–F, the peaks at 1557 and 1458 cm^–1 were associated with the skeleton vibration of CNT–F. The spectrum of GO showed the presence of bands located at 3450 and 1395 cm^–1, corresponding to the stretching and bending mode of hydroxyl groups over the basal plane; 1056 cm^–1 corresponding to the stretching mode of epoxy groups also over the basal plane; and 1733 cm^–1 corresponding to carbonyl groups located on the edges of the GO sheets (–COOH or C=O). For GO–F, the peaks at 1568 and 1456 cm^–1 were due to the skeleton vibration of GO. Compared with GO, the doublet peaks at 2920 and 2856 cm^–1 can be clearly assigned to C-H stretching vibrations of ethanolamine attached on GO. The peaks at 1202 cm^–1 (GO–F) and 1162 cm^–1 (CNT–F) can be attributed to C–F stretching vibration¹³ with slight wavenumber change mainly because of the different adjacent atoms. In addition, the peaks at 1143 cm^–1 (GO–F) and 1110 cm^–1 (CNT–F) were assigned to the formation of C–O–C.¹⁴ The presence of residue –OH groups can be detected at around 3400 cm^–1 in both spectra, which is comparable with the analysis from XPS.
Figure 2.
Fourier transform infrared spectra of CNT–OH, CNT–F, GO, and GO–F particles.

Results of the thermogravimetric analysis of the CNTs, graphite, CNT–F, and GO–F particles are shown in Figures 3 and 4. Because the organic functional moieties on carbon nanomaterials are thermally unstable, the TGA gives useful information about functionalized carbon nanomaterials.¹³ Pristine CNTs and graphite samples are mainly stable below 600℃. The oxygen-containing contents and grafted perfluorinated compounds are decomposed at the temperature below which pristine CNTs and graphite begin to lose weight. The CNT–F particles show a main and gradual thermal degradation step from 166.7℃ to 253.4℃, while the GO–F sample shows mass change from 181.1℃ to 320.8℃. As can be seen, weight losses between 100℃ and 600℃ are 2.2%, 48.1%, 4.3%, and 74.9% for CNTs, CNT–F, graphite, and GO–F, respectively. The greater mass changes of CNT–F and GO–F are attributed to the organic functional moieties. Based on TGA data, the weight content of the attached perfluorinated compounds on the GO–F are more than that on CNT–F, which has been confirmed by XPS data. In addition, another assumed reason for the lower weight loss of CNT–F than GO–F was that the number of carbon layers of MWCNTs was more than that of GO and the functional components were anchored on the outermost surface of the CNTs or GO, which means a higher percentage of the thermal stable sp² carbon atoms on CNT–F than GO–F.
Figure 3.
Thermogravimetric analysis curves of (a) raw multi-walled carbon nanotubes (CNTs); (b) graphite; (c) CNT–F; and (d) GO–F particles.

Figure 4.
Differential Thermal Gravity (DTG) curves of (a) raw multi-walled carbon nanotubes (CNTs); (b) graphite; (c) CNT–F; and (d) GO–F particles.

A total of 2 mg of CNT–F or GO–F particle was dispersed in 5 mL water to form a homogeneous suspension (Figures 5(a) and 6(a)). These suspensions were then vacuum-filtered using 0.22 µm pore size PVDF filter paper to form thin films with a diameter of 1.5 cm (Figures 5(b) and 6(b)). CNTs and graphene have very low dispersibility in most solvents with intrinsic van der Waals forces. The as-perfluorinated carbon materials showed good dispersibility in water due to the residue –COOH and introduced polar –C–F bond on the surface. The amounts of CNT–F and GO–F loaded on cotton fabrics were 20.2 and 21.2 wt%, respectively. The white cotton fabrics after finishing (Figures 5(c) and 6(c)) have an observable color change due to the existence of black carbon nanomaterials.
Figure 5.
Photographs of (a) stable colloidal solution (24 h after sonication) of CNT–F in water; (b) CNT–F film; (c) cellulose fabric treated by CNT–F.

Figure 6.
Photographs of (a) stable colloidal solution (24 h after sonication) of GO–F in water; (b) GO–F film; (c) cellulose fabric treated by GO–F.

The wetting performance of the surfaces can be characterized by variety of techniques such as the CA, tilt angle, and spray tests. This work used the CA, which is the most widely used method due to its effectiveness, simplicity, and ease of testing. The surface morphologies and CAs of the hydrophobic films are shown in Figure 7. In the case of GO–F film, the CA was 151°. High surface roughness consisted of many separated GO–F domains with flake dimensions ranging from hundreds of nanometers to several micrometers. In the case of CNT–F film, the CA was only 108°. The reasons for the CA difference between CNT–F film and GO–F film can be assumed as follows: first of all, although the atom concentrations of F on the CNT–F and GO–F particles mentioned above are 10.24% and 17.65%, respectively, which means the content of –C–F bonds on CNT–F is lower than that on GO–F. However, it is not appropriate to make a conclusion of the GO–F particle constructing a lower surface energy film on PVDF paper only by the F content, because the number of carbon layers of MWCNTs are much more than that of GO, and the -OH groups, which can react with perfluoro-1-iodohexane, are anchored on the outermost surface of CNT–F. Therefore, even a higher density of the –CF group might have been introduced on the surface of CNT-OH. Because the etherification of perfluoro-1-iodohexane with oxidized CNTs under the adopted condition barely gave the carboxyl group consumption, there remained many hydrophilic groups on every layer of CNT–F ends. The much more hydrophilic groups mitigating the hydrophobicity of CNT–F product could be the real cause of the CA difference between CNT–F and GO–F films. Secondly, hollows could be easily seen from the SEM images of the film constructed by CNT–F particles (marked in Figure 7(a)). The CNT–F was dispersed in water utilizing ultrasonication and there were many bubbles in the dispersion of the black particles, which could not burst during the following vacuum filtration. Although the reason for bubbles is not clear, the capillary system may be caused by these bubbles and could compromise the hydrophobicity of the film surface. The last reason is that the CNT–F bundles aggregated in the vacuum filtration process tend to attach to the filter paper to form smooth nanosheets and could not tilt the edge to form nanoscale roughness like in GO–F nanosheets.
Figure 7.
Scanning electron microscopy images with dfferent magnifications of CNT–F film (a,c), GO–F film (b,d), and the corresponding contact angles (e,f).

Figure 8 shows typical SEM images and CAs of the cotton fabrics treated by CNT–F and GO–F, respectively, indicating that the hydrophobic particles have been deposited on the surface of cellulose fibers, forming rough surfaces of multiple scales, with CAs reaching 149° and 154.4° for CNT–F and GO–F applications, respectively. It also could be concluded that the CA values were greater on textiles than films. Bahners et al.³⁵ suggested the following three peculiar factors influencing the wetting behavior of a droplet on a textile surface. The first one is that, macroscopically, the cellulose fabric has a coarse, textured surface, which may have similar effects on the wetting behavior of a droplet discussed as the Wenzel or Cassie–Baxter cases. Secondly, as on a cylindrical substrate, the solid–liquid interface will always be smaller than the liquid–vapor interface, and this prohibits the droplets spreading totally. The cylindrical geometry of a synthetic fiber has the strange consequence that a droplet will not spread even along an ideally hydrophilic surface. The last reason is that the capillary system, even though in the hydrophobic textile, provokes the penetration of sessile droplets. In this case, the hydrophilic groups on the cellulose fiber were covered by the nanomaterials but the rough structure of the fabric was retained. Liu et al.’s research⁶ showed that the modified CNTs formed nanoscale surface roughness on the microscale cellulose fabric and this micro-nanoscale binary structure resulted in the formation of an artificial lotus leaf structure on cotton. In addition, an assignable reason is that a hydrogen bond may form between the –COOH on the terminal of multi-walled CNT–F and –OH on the cellulose macromolecules to reduce the hydrophilicity of CNT–F mentioned above. The lower surface energy inducted by the perfluoro compound is recovered completely.
Figure 8.
Scanning electron microscopy images with dfferent magnifications of cotton fabrics treated by CNT–F (a,c), GO–F (b,d), and the corresponding contact angles (e,f).

Textiles that are used for various stretchable electronic devices need to be resistant to washing and chemicals.³⁶ Figure 9 shows the CAs of cotton fabrics treated by GO–F and CNT–F after being exposed to water and 4 M HNO₃ for 30 min; barely any change was found. The strong binding of CNT–F or GO–F and the cotton fabrics may be due to two reasons³⁶:(1) large van der Waals forces and hydrogen bonding exist between the residue oxygen-containing groups such as –COOH on the CNT–F or GO–F and the –OH on the cellulose; (2) the flexibility of CNT–F or GO–F allow them to be conformally adhered to the surface of cotton fibers, which maximizes the surface contact area between carbon nanomaterials and textile fibers. However, the fabrics became hydrophilic when they exposed to 2 M KOH for 30 min, which might be caused by desorption of the GO–F and CNT–F in the alkali medium. Thermal analysis of the cotton fabrics before and after treatment by CNT–F and GO–F was performed and the curves are shown in Figures 10 and 11. The lower mass change and the nearly constant decomposition temperature ranges indicate that the thermal behavior of the modified cotton fabric remained unchanged. This could be a good starting point for further study by combining other functional components, such as flame retardancy, which makes it a promising multifunctional material for smart textiles.
Figure 9.
Contact angles (CAs) of GO–F/cotton and CNT–F/cotton before and after immersion in water and 4 M HNO₃.

Figure 10.
Thermogravimetric analysis curves of untreated cotton fabric and cotton fabrics finished by CNT–F and GO–F.

Figure 11.
Differential Thermal Gravity (DTG) curves of untreated cotton fabric and cotton fabrics finished by CNT–F and GO–F.

Conclusions

In summary, a novel strategy has been demonstrated to functionalize the hydroxylated MWCNTs and ethanol amine grafted GO with perfluorohexane to obtain organic–inorganic hybrid nanomaterials with low surface energy, which could construct superhydrophobic surfaces on PVDF papers or cotton fabrics. In consideration of the environment problem, it is a successful example of using a hexa-perfluorinated compound as the functional component of low surface energy in organic–inorganic hybrid nanomaterial preparation. The films constructed and the cotton fabric samples treated by these particles show hydrophobicity with CAs greater than 90°or, in the case of GO–F, even greater than 150°. The cotton fabrics were still hydrophobic after exposure to water and 4 M HNO₃, while they became hydrophilic when exposed to 2 M KOH. The phenomenon that the CA of CNT–F treated fabric was much greater than that of the film and the suggested reasons give a tip in the exploration of hydrophobic surface fabrication. Confirmation of the perfluorohexane functionalized carbon nanomaterials might lead to further studies aiming at multifunctional hybrid nanomaterial exploration.

Sample	C (am%)	O (am%)	N (am%)	F (am%)	C/O	C_f ratio^a
CNT–OH	85.43	14.57	–	–	5.86	–
CNT–F	80.57	9.19	–	10.24	8.77	1
GO	73.88	26.12	–	–	2.42	–
GO–F	68.09	10.21	4.05	17.65	6.67	2.3

Sample	XPS peak (eV)	Assignment
CNT-OH	284.6	C–C
285.8	C–OH
287.6	O=C–O
289.5	COOH
CNT–F	284.6	C–C
286	C–O–C
288.5	O=C–O
290.2	–CF₂
292	–CF₃
GO	284.7	C–C
285.6	C–OH
286.7	Epoxy
287.8	O=C–O
288.9	COOH
GO–F	284.6	C–C
285.8	C–O–C
287.3	C–N
288.9	C=O
291.3	–CF₂
293	–CF₃

Footnotes

Acknowledgements

The authors acknowledge Yarong Wu, College of Chemistry, Chemical Engineering and Biotechnology, Donghua University, for her assistance with the routine laboratory work. Suxia Wang is also acknowledged for the GO preparation during her master’s degree thesis work.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Huang

Yin

et al.

Graphene-based materials: synthesis, characterization, properties, and applications. Small 2011; 7: 1876–1902.

Kaner

. Materials science: graphene-based materials. Science 2008; 320: 1170–1171.

Geim

. Graphene: status and prospects. Science 2009; 324: 1530–1534.

Mkhabela

Mishra

Mbianda

. Thermal and mechanical properties of phosphorylated multiwalled carbon nanotube/polyvinyl chloride composites. Carbon 2011; 49: 610–617.

Jiang

Wang

et al.

Preparation of superhydrophobic and superoleophilic polypropylene fibers with application in oil/water separation. J Text Inst 2013; 104: 790–797.

Liu

Wang

et al.

Functionalization of cotton with carbon nanotubes. J Mater Chem 2008; 18: 3454–3460.

Wang

et al.

Self-Assembly of octadecyltrichlorosilane on graphene oxide and the tribological performances of the resultant film. J Phys Chem C 2011; 115: 10080–10086.

Wan

Zhang

et al.

The tunable wettability in multistimuli-responsive smart graphene surfaces. Appl Phys Lett 2013; 102: 011603:1–5.

Liu

Tang

Xin

. Fabrication of nanowires with polymer shells using treated carbon nanotube bundles as macro-initiators. Chem Commun (Camb) 2004; (24): 2828–2829.

10.

Liu

Tang

Wang

et al.

Artificial lotus leaf structures from assembling carbon nanotubes and their applications in hydrophobic textiles. J Mater Chem 2007; 17: 1071–1078.

11.

Zhang

Zha

et al.

Formation of superhydrophobic microspheres of poly(vinylidene fluoride- hexafluoropropylene)/graphene composite via gelation. Langmuir 2011; 27: 8943–8949.

12.

Lin

Ehlert

Bukowsky

et al.

Superhydrophobic functionalized graphene aerogels. ACS Appl Mater Interfaces 2011; 3: 2200–2203.

13.

Georoakilas

Bourlinos

Zboril

et al.

Synthesis, characterization and aspects of superhydrophobic functionalized carbon nanotubes. Chem Mater 2008; 20: 2884–2886.

14.

Deng

B-J

et al.

Preparation and characterization of long-chain alkyl silane-functionalized graphene film. J Mater Sci 2012; 48: 156–161.

15.

Singh

Chen

Houshmand

et al.

Superhydrophobic graphene foams. Small 2013; 9: 75–80.

16.

Shateri-Khalilabad

Yazdanshenas

. Preparation of superhydrophobic electroconductive graphene-coated cotton cellulose. Cellulose 2013; 20: 963–972.

17.

Liu

Wang

Gao

et al.

Synthesis and characterization of photoreactive silica nanoparticles for super-hydrophobic cotton fabrics application. Text Res J 2015; 85: 795–803.

18.

Kramar

Obradovic

Vesel

et al.

Preparation of hydrophobic viscose fabric using nitrogen DBD and copper ions sorption. Plasma Proc Polym 2015; 12: 1095–1103.

19.

Aminayi

Abidi

. Ultra-oleophobic cotton fabric prepared using molecular and nanoparticle vapor deposition methods. Surf Coating Technol 2015; 276: 636–644.

20.

Tissera

Wijesena

Perera

et al.

Hydrophobic cotton textile surfaces using an amphiphilic graphene oxide (GO) coating. Appl Surf Sci 2015; 324: 455–463.

21.

Joung

Buie

. Antiwetting fabric produced by a combination of layer-by-layer assembly and electrophoretic deposition of hydrophobic nanoparticles. ACS Appl Mater Interface 2015; 7: 20100–20110.

22.

Chen

Zhang

et al.

Preparation and characterization of novel hydrophobic cellulose fabrics with polyvinylsilsesquioxane functional coatings. Cellulose 2016; 23: 941–953.

23.

Olsen

Burris

Mandel

et al.

Serum perfluorooctane sulfonate and hepatic and lipid clinical chemistry tests in fluorochemical production employees. J Occup Environ Med 1999; 41: 799–806.

24.

Lau

Butenhoff

Rogers

. The developmental toxicity of perfluoroalkyl acids and their derivatives. Toxicol Appl Pharmacol 2004; 198: 231–241.

25.

Zhang

Niu

et al.

Evaluating the sub-lethal toxicity of PFOS and PFOA using rotifer Brachionus calyciflorus. Environ Pollut 2013; 180: 34–40.

26.

Liu

et al.

Surface covalent encapsulation of multiwalled carbon nanotubes with poly(acryloyl chloride) grafted poly(ethylene glycol). J Polym Sci A Polym Chem 2006; 44: 6880–6887.

27.

Zhou

Lin

. Preparation and characterization of functionalized graphene with ethanolamine. Acta Chim Sinica 2011; 69: 1463–1468.

28.

Conrad

Burkholder

WRDJ

Medebielle

. The syntheses of nonnucleoside, HIV-1 reverse transcriptase inhibitors containing a CF₂ group. The S_RN1 reactions of 2-(bromodifluoromethyl)benzoxazole with the anions derived from heterocyclic thiols and phenolic compounds. J Fluorine Chem 1999; 102: 369–376.

29.

Conrad

Burkholder

WRDJ

Medebielle

. Synthesis and reactivity of halogeno-difluoromethy aromatics and heterocycles application to the synthesis of gem-difluorinated bioactive compounds. J Fluorine Chem 2001; 109: 39–48.

30.

Tuan Anh Pham

NAK

Jeong

. Covalent functionalization of graphene oxide with polyglycerol and their use as templates for anchoring magnetic nanoparticles. Synth Met 2010; 160: 2028–2036.

31.

Gao

Huang

Feng

et al.

Novel superhydrophobic and highly oleophobic PFPE-modified silica nanocomposite. J Mater Sci 2009; 45: 460–466.

32.

Nakamura

Ishihara

Ohana

et al.

Sidewall modification of single-walled carbon nanotubes using photolysis of perfluoroazooctane. Chem Commun (Camb) 2004; (11): 1336–1337.

33.

Ago

Kugler

Cacialli

et al.

Work functions and surface functional groups of multiwall carbon nanotubes. J Phys Chem B 1999; 103: 8116–8121.

34.

Baker

Cai

Lasseter

et al. Covalently bonded adducts of deoxyribonucleic acid (DNA) . oligonucleotides with single-wall carbon nanotubes: Synthesis and hybridization. Nano Lett 2002; 2: 1413–1417.

35.

Bahners

Textor

Opwis

et al.

Recent approaches to highly hydrophobic textile surfaces. J Adhes Sci Technol 2008; 22: 285–309.

36.

Pasta

La Mantia

et al.

Stretchable, porous, and conductive energy textiles. Nano Lett 2010; 10: 708–714.