Surface modification of highly hydrophobic polyester fabric coated with octadecylamine-functionalized graphene nanosheets

This study focuses on the design of highly hydrophobic polyester fabrics (PET) coated with organophilic graphene nanosheets (G-ODA) through a simple, cost-effective and scalable coating method. The organophilic graphene oxide was successfully synthesized by covalently grafting a long chain fatty amine on its surface and was fully characterized by various physicochemical techniques. G-ODA was coated at different loadings onto the PET fabric ranging from 1 to 7 wt% to produce uniformly dispersed PET@G-ODA fabrics with multifunctional performances. FTIR has confirmed the formation of strong interfacial interaction between the PET and G-ODA functional groups. Moreover, the produced PET@G-ODA fabrics resulted in achieving enhanced thermal stability as well as excellent water repellency compared to the pristine PET. Water contact angle measurements showed a tremendous enhancement of surface hydrophobicity up to 148° with 7 wt% loading of G-ODA. Tensile strength tests revealed that our fabric exhibited excellent mechanical properties compared to neat PET. In addition, the designed PET@G-ODA fabrics demonstrated excellent oil/water separation efficiency for different oil/water mixtures. The obtained results are very promising in terms of designing and producing functional PET fabrics with improved thermal and surface proprieties.


Introduction
Textile materials have been extensively used thanks to their outstanding properties, such as low cost, good exibility, light weight, good feel, easy processing features and high porosity which is required in various applications. [1][2][3][4] However, their utilization for various industrial and domestic applications is restricted because of their low thermal stability and hydrophilic surface; they can be quickly wetted by various liquids including water and this subsequently limits their usage for several applications. Therefore, modication of the textile fabric is of interest to improve the quality of the textile material and expand its use for multiple applications. Recently a large number of researches have reported coated textile fabrics with different materials to introduce several signicant multifunctional properties such as: electrical conductivity, ultraviolet blocking, photo catalytic properties, photo-thermal conversion, 5 farinfrared emission 6-8 and hydrophobicity. [9][10][11][12][13] Hydrophobic textiles akin to lotus leaf and rose petal surfaces have recently attracted so much attention, [14][15][16][17] and offer a signicant potential for implication for various scientic and industrial applications. Several methods of coating have been developed to mime the high hydrophobicity on the textile fabrics, among them: solution-immersion coating, sol gel based coating, wet-chemical process, spray coating [18][19][20][21][22][23][24] etc. However, most of these methods have limitations in large scale applications due to their high cost, toilsome fabrication processes and could be employed effectively depending on a requirement for particular applications. In this context, the major challenge is to elaborate a simple and scale-up approach to produce durably high hydrophobic textile fabrics without damaging their native properties. Dip coating method [25][26][27][28] is recognized to be a promising alternative due to its facility, reproducibility and suitability for mass production in a large scale.
Recently, graphene and related materials have attracted enormous attention in the scientic community during the past decades in view of their unique, unprecedented and countless physicochemical properties that have permitted access to these materials in almost every domain, [29][30][31][32] including textile coating. 33,34 One of the most effective and widely used approaches to synthesis graphene could be chemical reduction of graphene oxide (GO). Furthermore, the presence of various kinds of oxygen-based functional groups on the GO basal planes and edges (hydroxyl, carboxyl, oxygen functional groups and epoxy) provide active sites for chemical functionalization to achieve new fascinating proprieties such as hydrophobicity, oleophobicity, superhydrophobicity and so on. For instance, Lin et al. synthesized functionalized graphene oxide by covalently graing octadecylamine on its surface. 35 They reported that the octadecylamine functionalized GO achieved high hydrophobicity with a water contact angle up to 132.4 . Moreover, Xu et al. reported the synthesis of hydrophobic modied graphene oxide with a water contact angle of 125.8 . 36 The overall approach is to introduce new chemical moieties to hydrophobize GO's surface which implies a decrease in the surface energy thus highly hydrophobic GO based materials can be obtained. 37 Yet, the critical challenge of GO functionalization is to carry the procedure so that it only modies its surface, while preserving its intrinsic proprieties i.e. structural and morphological characteristics. So far, few studies have been conducted on the functionalization of graphene oxide to enhance the hydrophobicity of polyester based textile materials. 33,38 In the other hand, the international maritime traffic of oils continues to increase in order to meet industrial and food demands. Moreover, the increase in the world's population, which consumes oil for its food and the diversication of industrial applications involving the processing of oils, especially in cosmetics and pharmaceuticals industry, leads to intense traffic inevitably oil spill accidents. 39,40 Despite the taken measures by the International Maritime Organization (IMO) to make oils transportation safer and protect the environment by regulating discharges. Accidents resulted in the dumping of oils such as rapeseed oil into the Port of Vancouver in 1989 and in 1998 in Hong Kong city. Wales in 1991 (Mudge, 1995), palm kernel oil in 1997 in the channel or palm oil in the Mississippi River (USA) in 1998. Oils, although inherently non-hazardous, have environmental and ecological effects on the marine fauna and ora, because of the importance of the spilled quantities. The spread-out layers of oil, can dri over several tens of nautical miles, emulsify to form a product that is difficult to remove. However, prevention of damage to the environment can be facilitated through a development of new energy-efficient solutions, which can enhance and control specic wettability and absorption. Hydrophobic surfaces have drawn a lot of interest lately as for the oil/water separation due to their capacity to repelling water and absorbed only oil.
In this respect, the fabrication of highly hydrophobic polyester fabrics was achieved using the dip coating approach in an organophilic graphene oxide solution followed by an in situ chemical reduction. The organophilic graphene oxide (GO-ODA) was synthesized via chemical graing of octadecylamine and was fully characterized by various physicochemical techniques. Optical, structural, thermal, morphological, and mechanical characterizations of the produced PET@G-ODA fabrics were performed by means of FTIR, TGA, SEM and tensile strength. Furthermore, water contact angle measurement was also performed to characterize the water-repelling property of treated fabrics. Since the produced PET@G-ODA fabrics have the potential to be used in numerous domains such oil/water separation, their separation performances for different oil/water and organic solvents mixtures was also investigated and discussed. The obtained ndings showed the effectiveness of GO-ODA coating achieving enhanced proprieties compared to the pristine PET while maintaining the intrinsic characteristics of the fabric such as lightweight and good feel.

Preparation of graphene oxide nanosheets
In the present work, the graphene oxide was synthesized from natural graphite following modied Hummer's method. 41 The obtained aqueous graphene oxide dispersion was used for the preparation of ogranophilic graphene oxide nanosheets (GO-ODA). The preparation process of GO-ODA is illustrated schematically in Fig. 1. In short, 200 mg of graphene oxide nanosheets (GO) was dispersed in 20 mL of distilled water. In the mean time 200 mg of octadecylamine (ODA:CH 3 (CH 3 ) 17 -NH 2 ) was dissolved in 20 mL of ethanol. Then, ODA suspension was added to the GO solution and the mixture was reuxed for 24 h, during which GO nanosheets react with ODA molecules to form GO-ODA. Upon reaction completion, the solid product was repeatedly washed with ethanol to remove unreacted ODA moieties then isolated by centrifugation and nally dried in a vacuum oven at 70 C. The PET@G-ODA samples have been prepared using an easy "dip-coating" approach followed by in situ chemical reduction. The process occurs in three consecutive steps as depicted in Fig. 2. Before coating, knit polyester with dimensions of 10 cm in length and 4 cm in width was pretreated with an aqueous solution containing 3 g L À1 of ammonia and 2 g L À1 of nonionic detergent for 1 h at 100 C. Then, the PET fabrics were washed with distilled water and dried in an oven at 80 C for 4 h. Aer that, the pretreated fabrics were dipped in an aqueous solution of exfoliated GO-ODA in tetrahydrofurane (THF) (1 mg mL À1 ) and then dried in an oven at 90 C for 2 h. This process of dip coating and drying was performed several times in order to increase the amount of GO-ODA loaded on the samples. In this step GO-ODA sheets were tightly deposited on the surface of the PET fabrics. An obvious color change from white to black was visualized as the number of coating cycle's increases. Subsequently, the as-obtained PET@GO-ODA samples were chemically reduced by simple immersion in 100 mL aqueous solution of Na 2 S 2 O 4 (50 Â 10 À3 M) while stirring for 1 h at 90 C to reduce GO-ODA into G-ODA.

Characterization techniques
The chemical structure of the GO-ODA and the produced PET@G-ODA fabrics was studied by infrared spectroscopy using an Affinity-1S SHIMADZU spectrometer equipped with a golden gate single reection attenuated total reectance (ATR) accessory. FTIR spectra were recorded in the range of 500-4000 cm À1 at a resolution of 16 cm À1 . X-ray diffraction (XRD) measurements were performed on a Bruker D8 Discover diffractometer tted with a Cu-Ka irradiation source (l ¼ 1.5438Å) operating at 40 kV with a scan speed of 2 min À1 . The color changes of the produced fabrics were measured by color coordinates (L*, a*, b*) and reectance spectra using color spectrophotometer BYK in the spectral range 400-700 nm with 20 nm resolution. The surface morphology of PET@G-ODA was observed by Scanning Electron Microscopy (SEM), using an FEI Quanta 200-ESEM operating at an accelerating voltage of 20 kV. The thermal stability of the uncoated and coated polyester fabric was investigated by thermo-gravimetric analysis (TGA) using a TGA-Q500 (TA instruments) at the heating rate of 10 C min À1 in the air atmosphere from 25 to 700 C. The water contact angles measurements of the polyester fabrics were determined using an automatic video contact-angle testing apparatus (OCA 40, DATAPHYSICS) by the sessile drop method. A drop of 3 mL of distilled water was applied on the surface of the PET@G-ODA and the contact angle was determined from the video camera images of the drop. The tensile properties of the PET@G-ODA fabrics were determined on a Shimadzu EZ-SX apparatus, at a tensile speed of 10 mm min À1 . Measurements were performed upon rectangular specimens of 80 mm length and 40 mm width and 0.2 mm thickness.

Characterization of GO-ODA
The crystal structure of all the samples was determined using Xray diffraction (XRD). Fig. 4 shows the XRD patterns of natural graphite, graphite oxide (GO) and organophilic graphene oxide (GO-ODA). XRD diffractogram of graphite powder exhibits a strong and sharp diffraction signal localized at 2q ¼ 26.5 assigned to the (002) reection of an arranged graphite structure (JCPDS no. 75-1621), 43 and corresponding to an interlayer distance (d 002 ) of 0.33 nm calculated by Braggs law. Aer oxidation and ODA functionalization, the diffraction peak observed at 2q ¼ 26.5 has shied to 2q ¼ 21.05 , indicating an increase in the interlayer d-spacing (0.42 nm) compared to natural graphite (0.33 nm), this increase in the interlayer spacing is attributed to the covalently attachment of long chain fatty amine ODA on the GO surface moieties 44 through an amidation reaction. The main reactive sites for the chemical graing of ODA molecule onto GO are the epoxy groups and the carboxyl groups located at the basal and the edges of the graphitic structure. 45 For G-ODA the d-spacing shis to 0.44 nm with 2q ¼ 20.16 , suggesting that during the chemical reduction, ODA molecules were bonded to the exogenous groups of reduced graphene oxide and prevented the nanosheets from agglomeration.   This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 24941-24950 | 24943 FTIR analysis was performed to dene the chemical changes that can possibly happen during the chemical reaction between GO and ODA. Fig. 5 showed FTIR spectra of GO, ODA and GO-ODA. As expected, the GO spectrum shows the specic absorption bands of GO that appear respectively at 3411, 1738, 1629, and 1073 cm À1 , corresponding to the hydroxyl groups, C]O in carboxyl group, C]C in aromatic ring, and C-O-C in epoxide. 46 The existence of these functional groups endows GO with possibilities for functionalization. Aer the chemical reaction with ODA, the successful graing of ODA molecule on GO is clearly observed via the emerging bands at 2917 cm À1 and 2846 cm À1 (-CH 2 stretching in the octadecyl chain). 47 The new bands at 1587 and 1469 cm À1 arising, respectively, from N-H bending and C-N stretching of amide in GO-ODA gave further evidence to the successful amidation reaction between carboxyl groups of GO and the amine groups (NH 2 ) of ODA molecules. Moreover, the decrease in the relative intensity of the peaks observed at 1073 and 1738 cm À1 (C-O-C/-COOH) suggests that these groups are involved in a chemical reaction with the amine group of ODA molecule through an amidation reaction. We believe that the reaction between GO and ODA occur favorably between the COOH groups of GO and NH 2 groups of ODA, but it could also occur via nucleophilic substitution between epoxy group of GO (C-O-C) and the amine group of ODA (NH 2 ). In this case the NH 2 group of ODA molecule serves as a nucleophile and attacks the carbon atom in epoxy group leading to oxirane ring opening reaction, thus graing the ODA hydrocarbon chains in GO surface. These results conrmed the chemical graing of ODA into GO's surface, which is similar to what it was reported in previous works. These results conrmed the chemical graing of ODA into GO's surface, which is similar to what it was reported in previous works. 48,49 3.2. Characterization of PET@G-ODA Fig. 3 showed the digital images of PET, PET@G-ODA-1%, PET@G-ODA-3% PET@G-ODA-5% and PET@G-ODA-7%. It is clear from the photographic images that the color of PET fabric was signicantly affected aer coating with G-ODA. The blank sample, PET turned from white to dark grey aer dipping in G-ODA solution, and changed to black with increased dipping cycles to reach 7% of G-ODA loaded. These observations    indicate that the coating of PET fabrics with G-ODA was successfully occurred. The color changes of the PET fabrics were measured by color coordinates (L*, a*, b*). Table 1 illustrates color coordinates (data COLOR) of coated and uncoated polyester samples. It was found that color coordinates values changed aer G-ODA coating, indicating the color differences between the pristine PET samples and the coated ones. The pristine PET shows the highest lightness (L* ¼ 88.74) with values of a* (rednessgreenness) and b* (yellowness-blueness) equal to 0.91 and 2.41, respectively, describing its white color. Aer coating PET samples with G-ODA, an obvious decrease was observed in L* values (31.56, 29.33, 19.02 and 18.90) respectively, for the PET@G-ODA-1%, PET@G-ODA-3%, PET@G-ODA-5% and PET@G-ODA-7%, and for a* values (from 0.91 for the pristine PET to 0.23 for the PET@G-ODA-7%). Moreover, by comparing the L*a*b* values for each sample, it can be noticed that the samples don't match in color and that the greater the amount of G-ODA is, the darker is the PET@G-ODA fabric, conrming the successful coating of the PET fabric.
The Scanning Electron Microscope (SEM) was employed in order to investigate the bers and the surface morphology of treated and untreated fabrics. Fig. 6 shows the SEM images of PET, PET@G-ODA-1% (b), PET@G-ODA-3%, PET@G-ODA-5% (d) and PET@G-ODA-7%. From this gure, it can be clearly seen that the surface morphology of the uncoated polyester fabric (PET) (Fig. 6a) presents a relatively smooth and sleek bers texture. Aer coating, G-ODA sheets were clearly deposited onto the surface of PET fabrics and also inserted into the pores between ber-ber of PET fabrics, as shown in Fig. 6b-f. The surface of the fabric's bers became completely covered with the rough layer of G-ODA on each ber (Fig. 6b-f). These results conrm the successful attachment of G-ODA on the surface of PET. In addition, SEM images of PET@G-ODA fabrics revealed that the PET bers were almost intact aer coating with G-ODA, which is indicating that the fabrics did not undergo structural changes during the coating. From all of this results, coating PET with G-ODA can generate a roughness on the surface of PET bers that resembles the surface texture of a lotus leaf or a rose petal, and thus it provides the responsible texture leading to the formation of high hydrophobic surfaces, based on recent investigations. 50,51 The FTIR spectrum of the uncoated knit polyester fabric (Fig. 7), shows characteristic bands at 969 and 1014 cm À1 corresponding to the C-O stretching of glycol and benzene in-plane vibrations, respectively. The bands observed at 1091 and 1247 cm À1 represent the ester C]O stretching vibration and the band centered at 1714 cm À1 is assigned to C]O group of aromatic ester. 52 The absorption bands at about 1460 cm À1 is attributed to C]C stretching vibration. 53 The FT-IR spectra of PET@G-ODA samples revealed some noticeable changes including notable reduction in the intensity of carbonyl groups stretching vibration (1714 cm À1 ) suggesting that G-ODA strongly interacted with the PET fabric, occurring probably between the ester group of PET and NH, C-N group of the amide, or through hydrophobic interaction between G-ODA molecules and C]C and benzene rings of PET that interact through p-p stacking, 54 leading to the formation of a uniform G-ODA coating on PET surface. Furthermore, the apparition of a new absorption bands at 2847 and 2916 cm À1 assigned to the symmetric and asymmetric stretching vibration modes of CH 2 groups in octadecylamine graed onto GO's surface gave further proof to the establishment of strong interactions between the PET and G-ODA.
The thermal degradation behavior of PET and PET@G-ODA fabrics was investigated by means of TGA and DTG with  10 C min À1 heating rate under air atmosphere. TGA curves of all the samples being presented in Fig. 8, while the different parameters obtained from the thermograms are listed in Table  2. It is clearly observed that all the samples, regardless the G-ODA content, undergo mainly two distinct decomposition stages and were identied as follows, (i) a primary degradation step was occurred in the range of 300-470 C corresponding to the ester linkage decomposition in the PET fabric which took place via the depolymerization of polyester through chain scission producing carboxy-and vinyl-terminated chain fragments 55 simultaneously occurring with the decomposition of both physically adsorbed ODA molecules remaining aer washing by ethanol and chemically graed ODA chains, (ii) the second degradation step undergoes up to 600 C corresponding to C-C bond cleavage leading to the formation of volatile compounds such as carbon mono-and dioxide, methane, ethylene, benzene, benzaldehyde, formaldehyde and acetaldehyde, as well as the pyrolysis of the residual covalently graed ODA chains. Compared to pristine PET, the PET@G-ODA fabrics displayed enhanced thermal stabilities, which can be ascribed to the strong interactions established between the G-ODA coating and the PET bers. Overall, the thermal decomposition of PET@G-ODA fabrics took place at higher temperature than that of uncoated PET. The signicant increase in T 10% (358.6 C for PET vs. 386.7 C for PET@G-ODA 7%) and T 80% (448.9 C for PET vs. 463.2 C for PET@G-ODA 7%) gave further supports that G-ODA coating has successfully improved the overall thermal stability of the PET fabric. Such enhancement can be attributed to the presence of G-ODA as coating, improving the overall thermal resistance through the establishment of strong interactions with the PET bers. Furthermore, this improvement may be due to the contribution of graphene's intrinsic properties that improves the thermal properties of textile fabrics. 56 As reported previously, the surface roughness plays an important role in determination of the material's wetting properties. According to Wenzel model, 57 the surface roughness of a material will usually increase its apparent hydrophobicity. The inuence of surface roughness on the high hydrophobicity and wetting property of PET@G-ODA fabrics is affirmed by the contact angle measurements (Fig. 9). As shown in Fig. 9a, the native PET textile was rapidly wetted by water droplets with apparent contact angle of 0 , this "hydrophilicity" is due to the fact that water droplet tend to diffuse through the woven PET fabric via the wicking effect 58 and capillary force governed by the porous structure of PET fabric. Aer coating with G-ODA (Fig. 9b), all the PET samples turned to be highly hydrophobic which can be explained by the deposition of graphene moieties bearing long hydrophobic hydrocarbon chains on the PET surface. 37,59 Water droplets with spherical shape appeared to remain durable on the treated textile's surface, this property was achieved for all loaded samples and the contact angle value increases from 128.7 to 148.7 by increasing G-ODA contents from 1 to 7% (Fig. 10). Moreover, a drop on the surface of treated PET doesn't roll off even on ipping it upside down (Fig. 9c). As results, PET@G-ODA exhibits a "petal effect" which is similar to a rose petal i.e. hydrophobic character along with high adhesion forces (Fig. 9c).
The water contact angle measurement, as a function of time was monitored in order to investigate the change in the   hydrophobicity of the different prepared PET@G-ODA fabrics (Fig. 11). As clearly observed in the gure, the treated PET maintained its high hydrophobicity and only a slight decrease in the contact angle for all samples was detected, suggesting the excellent durability of the PET@G-ODA textiles as highly hydrophobic surfaces.
The prepared fabrics in our study exhibit high hydrophobic properties, which make them suitable for different applications. Due to their excellent uid absorption capacity, textile materials have been extensively used for cleaning and mopping. Therefore, to demonstrate absorption capacity of the elaborated PET@G-ODA; an oil/water separation experiment was performed. Fig. 13a shows the selective absorption of the oil from an oil/water mixture by the treated PET@G-ODA-7% sample.
The baker contains frying oil (blue) and an aqueous solution of distilled water. According to this gure, it can be clearly observed that the highly hydrophobic PET@G-ODA coated samples, when immersed in the oil/water mixture, absorbed the oil selectively and completely, within a few seconds, without the requirement of extra force, leaving behind a clear aqueous solution. Aer the procedure, the water volume remained the same with no change, leading to a selective oil/water separation. The coated samples were found to be efficient as an absorbent and absorbed oil almost 4 times of their initial weight. The oil absorption capacity (C oil ) by different samples was calculated and quantied using following formula: 60,61 C oil (g/g) ¼ (W f À W i )/W i where W i : weight of the sample and W f : weight of the sample with absorbed oil.
The absorption capacities of the PET@G-ODA for other types of oil compounds and organic solvents are illustrated in Fig. 13b. The PET coated samples were found to be highly efficient as absorbents and absorbed oil more than 4 times their weight.
Washing test was also preformed in order to investigate the PET@G-ODA coated samples stability aer washing cycles. As known, the ultrasound treatment would rapidly deteriorate the coating structure and speed up the coating removal from the substrate. The samples were washed for 40 min each, under ultrasound treatment. The PET@G-ODA-7% was immersed in a 100 mL solution of (i) hot water, (ii) detergent and (iii) organic solvent, for 5 cycles. Aer each washing cycle, the sample was washed with deionized water and dried. Contact angle measurements were carried out aer every washing cycle and the results are summarized in Fig. 12. PET@G-ODA-7% (C. A ¼  with hot water, THF and detergent. These results conrm the durability of G-ODA coating as well as the strong interaction between the coating and the bers, as it was mentioned previously. It is well reported that the density and viscosity of the absorbed oil usually affect the absorption capacity of absorbents. 62,63 They exhibit lower absorption capacity for organic compounds with high viscosity, this last resist the diffusion of oil towards the absorbent, which leads to slow absorption for oils. In the present study, the absorption capacity of the PET@G-ODA-7% coated sample was found to be higher for THF due to its low viscosity (h (THF) z 0.46 Â 10 À3 Pa s) compared to the other used organic compounds (h (DMF) z 0.80 Â 10 À3 Pa s and h (toluene) z 0.59 Â 10 À3 Pa s). In the case of oils, the absorbed volumes are almost similar; this is probably due to the approaching density values of rapeseed oil and frying oil, which are 0.916 g cm À3 and 0.910-0.920 g cm À3 , respectively.
The absorbed oil can be taken out of the PET@G-ODA textile easily by simple squeezing, and washing with acetone. The oilsoaked textile can be reused in order to absorb oil once more.
As mentioned previously, the dip-coating approach guaranties high homogeneity and adhesion of G-ODA to the PET samples. The aim of PET coating is adding new hydrophobic properties via graing organophilic graphene functionalities into the fabric's surface, while maintaining the high exibility, lightweight and without damaging the structure of textile bers.
Tensile test was performed to investigate the mechanical behavior of the PET coated samples. The tensile properties of pristine and coated PET samples were obtained from their stress-strain curves. The average tensile strength (MPa), elongation at break (%), and Young's modulus (MPa), are calculated and illustrated in Fig. 14, and the corresponding results are represented in Table 3. According to the gures, pristine PET showed an inferior tensile strength (Fig. 14a) and elongation at break (Fig. 14b)    In addition, the coated PET samples displayed higher elongation at break values (Fig. 14b), that increase from 9.16 AE 0.9% for pristine PET, to 35.90 AE 1.6% for PET@G-ODA-7%, which is nearly similar to PET@G-ODA-1% (37.08 AE 2.01%), showing that even a small amount of 1 wt% can interestingly improves the mechanical properties of PET fabrics. It was previously reported that coated GO fabrics show a decrease in Young's modulus compared to uncoated PET fabrics. 27 The Young's modulus of PET, PET@G-ODA-1%, PET@G-ODA-3%, PET@G-ODA-5% and PET@G-ODA-7%, was calculated from the elastic domain 64 in each of the stress-strain curves. Young's modulus for the pristine PET is 208.3 AE 0.40 MPa, while it decreases proportionally with increasing the amount of G-ODA loading to 75.2 AE 0.20 MPa for PET@G-ODA-7% (Fig. 14c). This decrease is probably due to the formation of aggregates that leads to discontinuous nanostructure in the PET surface as it was previously seen by SEM.

Conclusions
The elaboration of PET to PET@G-ODA coated textiles was carried out by dip-coating approach in an organophilic graphene oxide solution. The ODA-functionalized graphene was successfully synthesized by covalently graing a long chain fatty amine to graphene's surface. The prepared textiles exhibit an enhanced thermal stability and a proportional increase in the contact angle with the amount increasing of G-ODA from 1 to 7%. Compared to the pristine PET, the G-ODA coated PET was found to be highly efficient for oil/water separation. Tensile properties including tensile strength and elongation at break reected an improvement with G-ODA coating. The PET@G-ODA coated samples show promising results, which make them suitable to be used in different scientic and industrial applications.

Conflicts of interest
The authors declare that they have no competing interests.