Facile synthesis of reduced graphene oxide/trimethyl chlorosilane‐coated cellulose nanofibres aerogel for oil absorption

Abstract

A hydrophobic and oleophilic trimethyl chlorosilane/reduced graphene oxide‐coated cellulose nanofibres (TMCS/rGO/CNFs) aerogel with a three‐dimensional structure was fabricated through a facile dip‐coating process. The prepared aerogel exhibited several advantageous properties for absorption and expulsion of oils from water surfaces, such as a high specific surface area, low density (6.78 mg/cm³) and good porosity (99.12％). In addition, the TMCS/rGO/CNFs aerogel demonstrated good absorption capacities up to 39 times its own weight over a short time (1.5 min) for a broad range of oils. This research suggests that practical application of TMCS/rGO/CNFs aerogel in the cleanup of an oil spill is feasible.

1 Introduction

With the growth of oil production and transportation, significant attention has been focused on oil spills after a series of accidental releases of oil into the ecosystem [1, 2]. Such releases come from different sources, such as underwater extraction pipelines [3], tankers [4], oil rigs [5], refineries [6] and wells [7]. In 2010, the Deepwater Horizon incident released an estimated 4.1 million barrels of crude oil into the Gulf of Mexico and shocked the world [8]. Therefore, there is an urgent need to develop new materials for the collection and separation of large amounts of oil from water surfaces [9, 10]. Recently, a large number of absorbents including zeolites [1], active carbon [11], and natural fibres [12] have been used, but they suffer from low absorption and poor selectivity. Thus, many other materials have been developed to absorb oil from water, such as active carbon [13], carbon nanotube sponges [14, 15], carbon composites [16, 17], and micro‐porous polymers [18, 19]. However, the high production costs and complicated processing of these absorbents have limited their practical applications. Therefore, there is an urgent need for novel materials that have a high absorption capacity and selectivity while being lightweight, low cost, and environmentally friendly [20].

As a unique two‐dimensional (2D) carbon nanomaterial, graphene and its derivatives have been growing in popularity because of their intriguing physicochemical properties, such as their high surface area, exciting hydrophobic properties, excellent electronic conductivity, and good chemical, mechanical and environmental stability [21, 22]. Their outstanding physical and chemical properties offer great technological promise for a variety of sustainable applications, such as catalyst supports [23, 24], absorbents [25], gas sensors and supercapacitors [26]. Due to its extreme mechanical values (a 1 TPa Young's modulus and 130 GPa ultimate strength) and its special 2D structure, graphene is also an excellent reinforcing agent in polymers [27]. Incorporation of graphene nanoflakes into a cellulose matrix provides enhanced mechanical and functional properties [28]. However, little research has been reported on graphene‐based cellulose materials that have hydrophobic and oleophilic properties for separating and absorbing oils from water.

Cellulose is the most abundant natural polysaccharide in the world and is attracting increasing attention as an inexpensive, green and sustainable substrate material [29]. 1D cellulose nanofibres (CNFs) are mesoscopic material with micrometre‐long lengths and diameters of only a few nanometres, which makes it ideal candidates as nanobuilding blocks for constructing macroscopic structures [30]. Moreover, CNFs have abundant hydroxyl groups, which can facilitate the formation of aerogel by either physical or chemical crosslinking. Furthermore, abundant hydroxyl groups make the surface of CNFs hydrophilic [31, 32], enabling them to be easily combined with other nanoparticles to construct novel hybrid materials [33, 34].

Herein, we developed a facile and inexpensive dip‐coating method to fabricate a hydrophobic and oleophilic material based on the trimethyl chlorosilane/reduced graphene oxide (GO)‐coated composite aerogel. A 3D, porous CNFs aerogel with an absorption capacity for water, oils, and organic solvents was used as a frame for the rGO to coating. Recently, 3D graphene‐based macroporous architectures have been intensively explored as promising absorbents with remarkable absorption capabilities. This is owing to they have a lot of good performances, such as high porosity, available surface area, excellent chemical, mechanical stability, as well as the diverse chemical modifications of their carbon surfaces [35, 36]. Therefore, the addition of rGO would endow the CNFs aerogel improved thermal properties and absorption capabilities, whereas the CNFs aerogel would provide rGO a low cost template and scaffold for large‐scale applications. In previous studies, trimethyl chlorosilane (TMCS) was commonly chosen as the coating agent to fabricating hydrophobic and oleophilic materials. Because TMCS has the advantages of low cost, simple coating process and so on. Hydrophobic TMCS groups were controllably anchored onto the rGO/CNFs aerogel skeletons to regulate their hydrophilicity. Therefore, we have produced a porous TMCS/rGO/CNFs aerogel with sufficient porous structure and high specific surface area, which could achieve higher oil absorption capacities. As far as we know, this is the first report on the TMCS/rGO/CNFs composite aerogel.

2 Experimental

2.1 Materials

Bamboo powder (purchased from Zhejiang Lishui, China) was used as a source of cellulose and sieved through a 60 mesh. Potassium hydroxide (KOH), acetic acid (CH₃ COOH), hydrochloric acid (HCl), hydrogen peroxide (H₂ O₂, 30%), concentrated sulphuric acid (H₂ SO₄, 98%), potassium permanganate (KMnO₄), sodium nitrate (NaNO₃), phosphorus pentoxide (P₂ O₅), potassium persulfate (K₂ S₂ O₈), sodium chlorite (NaClO₂), ammonium hydroxide (NH₃ ·H₂ O), and vitamin C (L‐ascorbic acid, Vc) were obtained from Nanjing Chemical Reagent Co., Ltd., and natural graphite powder (40 μm) was purchased from Qingdao Henglide Graphite Co., Ltd. They were all used without further purification.

2.2 Methods

2.2.1 Cellulose nanofibres

Chemical purification of the bamboo powder was performed with small modifications to a literature method [37]. First, the bamboo powder was dispersed in 30 ml deionised (DI) water by stirring the mixtures for 10 min. In order to remove pectin, hemicelluloses, and residual starch, the suspensions were treated with 2 wt% aqueous KOH at 90°C for 2 h. Second, the samples were treated with a solution of NaClO₂ and CH₃ COOH at 75°C for 1 h to remove the lignin. Then, the above two steps were repeated five times to get highly purified cellulose. Finally, the samples were treated with a 1 wt% HCl solution at 80°C for 2 h to remove excess metal ions. Thereafter, the cellulose was washed with DI water until the rinse water was neutral. Throughout the chemical treatment, the purified cellulose was grinded by a grinder (MKCA6–2, Masuko Sangyo Co., Ltd., Japan) at 1500 rpm to obtain a slurry of 1 wt% CNFs [38].

2.2.2 Preparation of GO

GO was synthesised from graphite powder using the modified Hummers method [39]. Graphite powder (3 g) was treated with a solution of concentrated H₂ SO₄ (25 mL), K₂ S₂ O₈ (4 g), and P₂ O₅ (4 g) at 80°C for 5 h. The resulting slurry was stirred at 0°C, and distilled water was added dropwise until no additional smoking was observed. The slurry was diluted with distilled water, filtered, and washed until the rinse water became neutral. The product was dried under vacuum at room temperature for one day and is referred to as preoxidised graphite. The preoxidised graphite was then oxidised by the Hummers method [40]. After that, concentrated H₂ SO₄ (100 mL), preoxidised graphite powder (3 g) and NaNO₃ (1 g) were added, followed by the gradual addition of KMnO₄ (10 g) at 0°C with stirring over 2 h. The mixture was then stirred at 35°C for 4 h. Then, distilled water (800 mL) was added to the mixture and allowed to stand for 30 min. The mixture was stirred at 90°C for 30 min and a 30% H₂ O₂ solution (20 mL) was added dropwise. The resulting bright yellow mixture was washed with concentrated HCl. The mixture was then subjected to centrifugation (8000 rpm, 70 min) and washed with distilled water. The centrifugation procedure was repeated until the rinse water became neutral. After the addition of water, the supernatant was collected, and the GO film formed after evaporation of the water at ambient temperature over two weeks.

2.2.3 Reduced GO

The GO solution (100 mL, 1.98 mg/mL) was dispersed in DI water using an ultrasonic wave cell pulverizer (XO‐1200, Nanjing Xianou Biological Technology Co., Ltd., China) for 2 h. Then, Vc (100 mg) was added and stirred for 30 min to get a homogeneous dispersion. Afterwards, a solution of NH₃ ·H₂ O was added dropwisely to the above dispersion (pH≈11) with stirring at 90°C for 2 h. Finally, rGO was washed with distilled water using vacuum filtration until the rinse water was neutral.

2.2.4 Fabrication of TMCS/rGO/CNFs aerogel

Our facile dip‐coating process is shown in Fig. 1. First, a 1 wt% slurry of CNFs (6 g) was dispersed in DI water using an ultrasonic wave cell pulveriser for 30 min. DI water was added until the sample had a volume of 10 mL. The sample was then frozen in a refrigerator at −18°C for >24 h. After that, freeze drying was carried out for two days using a freeze drier (XIANOU‐10, Nanjing Xianou Biological Technology Co., Ltd., China) to obtain the desired CNFs aerogel (3.5 × 1 cm). Then, the CNFs aerogel was dipped into the rGO solution (1.33 mg/mL) for 10 min and dried at 80°C for 2 h to produce the rGO/CNFs aerogel (65 mg). Finally, the rGO/CNFs aerogel was dipped in the TMCS solution and heated in an oven at 70°C for 2 h so that silanation produced a TMCS/rGO/CNFs aerogel that had good hydrophobic and oleophilic properties. In addition, the density of the TMCS/rGO/CNFs aerogel was ∼6.78 mg/cm³, and the whole sample was able to be placed on the top of a flower without deforming it.

Fig. 1.

Schematic showing the synthetic steps for preparing TMCS/rGO/CNFs aerogel

2.3 Characterisation

Surface morphologies of the developed aerogels were investigated using a field emission scanning electron microscopy (FE‐SEM, S‐4800, HITACHI, Japan). The X‐ray diffraction (XRD) profiles were measured using an Ultima IV multipurpose XRD system and Cu‐Ka irradiation (40 kV and 30 mA). Raman spectra results were confirmed by a DXR Raman spectrophotometer (Thermo Scientific, USA) with a 532‐nm laser source. The Fourier transform infrared spectroscopy (FTIR) profiles were measured using an FTIR spectrometer (Nicolet iS10, Thermo Electron Corp., USA) with an attenuated total reflectance device. The water contact angles (CAs) of the aerogels were measured using an optical CA meter (OCA, Data Physics) with 4 µL water droplets. The absorption capacities of the aerogels for oils were measured by immersing the modified aerogels into various oils for 30 min until they were saturated. The absorption capacity, C, was determined by the equation C  = (W ₂ –W ₁)/W ₁, where W ₁ and W ₂ are the weights of the modified aerogels before and after absorption, respectively.

3 Results and discussion

3.1 Microstructures

Fig. 2 a shows the FE‐SEM images of the developed CNFs aerogel. The porous network structure had continuous nanofibres with diameters ranging from 10 to 50 nm. As shown in Fig. 2 b, rGO was similar to transparent gauze. Figs. 2 c and d show the morphologies of the TMCS/rGO/CNFs aerogel. In contrast to the mesopores of the aerogel formed from CNFs alone, macropores are clearly observed in the SEM images of the highly porous structures of the TMCS/rGO/CNFs aerogel (Fig. 2 c). The molecules or atoms on the surface of TMCS/rGO/CNFs aerogel have residual surface energy due to force imbalance. When the oil is gathered on the aerogel, the aerogel surface energy decreases and the oil is stuck to the aerogel by the attraction of unbalanced forces. TMCS/rGO/CNFs aerogel have the advantages of highly porous structures, which makes it easy to achieve oil absorption.

Fig. 2.

SEM images of the

(a) CNFs, (b) rGO and, (c) TMCS/rGO/CNFs aerogel

3.2 Chemical properties

Fig. 3 shows the XRD patterns of the (a) CNFs, (b) GO, (c) rGO, (d) rGO/CNFs, and (e) TMCS/rGO/CNFs aerogel. As shown in Fig. 3 a, the CNFs exhibit two peaks at 2θ  = 15.7° and 22.5°, corresponding to the (110) and (200) crystal planes, which were attributed to a typical cellulose I structure [41]. GO exhibited one peak at 2θ  = 11.3° (Fig. 3 b), corresponding to the (002) crystal plane. The interlayer spacing of GO was 0.781 nm, which was attributed to completely peeled GO [42]. When the GO was reduced to rGO, the diffraction peak of rGO widened, which meant that the crystal planes improved [43]. As shown in Fig. 3 c, rGO exhibited one peak at 2θ  = 25.4°, corresponding to the (002) crystal plane, and an interlayer spacing of 0.351 nm. As the oxygen‐containing functional groups of GO were eliminated, the strong van der Waals force between rGO sheets reunited them and led to decreases in the interlayer spacing [44]. The CNFs and rGO peaks appeared in the rGO/CNFs aerogel (Fig. 3 d), but the rGO peak was extremely weak, which may be due to less rGO being in the sample. The surface treatment of the rGO/CNFs aerogel with TMCS did not change the diffraction peaks of the aerogel (Fig. 3 e) because TMCS only forms a hydrophobic layer on the surface of the aerogel and does not change its chemical structure.

Fig. 3.

XRD patterns of the

(a) CNFs, (b) GO, (c) rGO, (d) rGO/CNFs and, (e) TMCS/rGO/CNFs aerogel

In Fig. 4 a, we observed a typical FTIR spectrum of CNFs: C–O–C pyranose ring skeletal vibrations (1026 cm⁻¹), O–H bending (1368 cm⁻¹), –CH₂ bending (1429 cm⁻¹), H–O–H bending of the absorbed water (1639 cm⁻¹), C–H stretching (2897 cm⁻¹), and hydroxyl groups (3327 cm⁻¹) [37]. In Fig. 4 b, the characteristic peaks of various carbon–oxygen functional groups are observed in the FTIR spectrum of GO: C–O vibration (1034 cm⁻¹), C–OH (1588 cm⁻¹), C  = O in carboxylic acid, and carbonyl moieties (1714 cm⁻¹). In the rGO spectrum (Fig. 4 c), most signals corresponding to oxidative species, especially the C  = O peak at 1714 cm⁻¹, disappeared indicating a complete reduction of GO. Fig. 4 d shows the FTIR spectrum of the rGO/CNFs aerogel. All of the diffraction peak intensities decreased substantially, which may manifest a strong interaction between the two components. After the surface treatment of the rGO/CNFs aerogel with TMCS, no characteristic peaks of the aerogel changed (Fig. 4 e), which is consistent with the XRD data.

Fig. 4.

FTIR spectra of the

(a) CNFs, (b) GO, (c) rGO, (d) rGO/CNFs and, (e) TMCS/rGO/CNFs aerogel

Fig. 5 a shows the Raman spectrum of the CNFs aerogel. The band at 883, 1092, 1366, 2897 cm^−1, are due to the in‐plane symmetric stretching of C–O–C, due to C–O–C glycosidic link asymmetric stretching modes, CH₂ deformation vibrations, the symmetric and asymmetric stretching vibrations of CH₂, respectively. A corresponding Raman spectrum of cellulose has been noted in the document [45, 46]. The Raman spectrum of GO and rGO is shown in Figs. 5 b and c, respectively. The disorder‐induced D band is at 1340 cm⁻¹, and the tangential G band is at 1574 cm⁻¹ [47]. In addition, the relative intensity of the D band and G band (I _D /I _G) of GO and rGO is 1.06 and 1.34, respectively. After GO was reduced to rGO, a large number of sp³ hybridised carbon atoms deoxidise to form new regions with sp² hybridisation. However, the total sp² area of rGO was less than that of GO, and the sp² regions in rGO were smaller and more numerous, as reflected in the Raman spectra and the increased I _D /I _G ratio; the same results have been reported in the literature [48, 49]. For the rGO/CNFs aerogel (Fig. 5 d), the peaks of CNFs became very weak or even disappeared completely as they were covered by the rGO peaks. After the surface treatment of the rGO/CNFs aerogel with TMCS, none of its characteristic peaks changed (Fig. 5 e), which is in agreement with the SEM, XRD, and FTIR data.

Fig. 5.

Raman spectra of the

(a) CNFs, (b) GO, (c) rGO, (d) rGO/CNFs, and, (e) TMCS/rGO/CNFs aerogel

In Figs. 6 a and b, the water CA of the untreated CNFs and the rGO/CNFs are close to 0°, indicating that the hydrophobicity did not change after treatment with rGO. However, in Fig. 6 c, after the surface treatment with TMCS, the hydrophobicity greatly increased and the water CA increased to 117°. As shown in Fig. 7 a, the untreated CNFs aerogel sank below the water interface, but the TMCS/rGO/CNFs aerogel floated on the water's surface. Thus, TMCS/rGO/CNFs aerogel have good hydrophobicity. In Fig. 7 b, the TMCS/rGO/CNFs aerogel sank below the oil–water interface. Thus, it was shown that the TMCS/rGO/CNFs aerogel had better hydrophobicity and oleophilicity. In addition, the TMCS/rGO/CNFs aerogel remained at the oil–water interface and did not sink below the water, even after two weeks, illustrating its structural stability.

Fig. 6.

CA of the

(a) CNFs, (b) rGO/CNFs, and, (c) TMCS/rGO/CNFs aerogel

Fig. 7.

Photograph of the TMCS/rGO/CNFs aerogel in

(a) Water and, (b) Mixture of water–soybean oil (dyed with Sudan red)

As expected, the TMCS/rGO/CNFs aerogel absorbed oil easily. By dipping the aerogel into soybean oil (dyed with Sudan red), oil was quickly absorbed by the aerogel within 1.5 min (as shown in Fig. 8). The absorption capacity of each oil on the TMCS/rGO/CNFs aerogels is plotted as a function of absorption time in Fig. 9. The absorption rate is faster at the first 10 s and gradually slow down as time goes on. The initial rapid adsorption is attributed to the availability of the active adsorption sites suitably exposed on surface as well as interior of TMCS/rGO/CNFs aerogels which becomes saturated with duration of time [50]. Then, the absorption reaches the equilibrium state at 30 s for both of the three oils, which is attributed to swelling equilibrium (the amount of absorbed oil and spilled oil is equal) of aerogels [51].

Fig. 8.

Optical images for the removal of soybean oil (dyed with Sudan red)

Fig. 9.

Oil absorption capacities of the CNFs, rGO/CNFs, and TMCS/rGO/CNFs aerogel

To further demonstrate the oil‐absorption ability of the TMCS/rGO/CNFs aerogel, its absorption capacities for different oils were investigated. The sorption efficiency was determined by weighing the sample before and after oil absorption. The absorption capacities of the CNFs, rGO/CNFs, and TMCS/rGO/CNFs aerogel for different oils were measured and are reported in Fig. 10. Three common oils (soybean oil, corn germ oil, and vacuum pump oil) were used for testing the oil absorption capacities of those aerogels. The absorption capacities range from 14 to 17 times the weight of the untreated CNFs for a variety of oils. This can likely be attributed to the fact that the CNFs aerogel had a higher porosity and a porous structure. However, the absorption capacities range from 29 to 38 times the weight of the rGO‐treated CNFs for a variety of oils, which indicates that the addition of rGO can significantly improve the oil absorption capacities of aerogel. After the surface treatment of the rGO/CNFs aerogel with TMCS, the absorption capacities range from 33 to 39 times the weight of the untreated aerogel for a variety of oils. However, the absorption capacity of TMCS/rGO/CNFs composite aerogel (33–39 times) is higher than studied aerogel [52] (18–20 times). There have several factors may be contributed to the high absorption capacities observed with TMCS/rGO/CNFs aerogel including extremely high porosities, interconnected pores, and highly hydrophobic aerogel surfaces. Oils were mainly stored in the macropores of the aerogel, so the differences of absorption capacities between kinds of oils were related to the densities of oils.

Fig. 10.

Oil absorption capacities of the CNFs, rGO/CNFs, and TMCS/rGO/CNFs aerogel

4 Conclusions

We demonstrated a facile approach to fabricate hydrophobic TMCS/rGO/CNFs aerogel through a dip‐coating process. TMCS was used to form a hydrophobic coating on the surface of the CNFs. The as‐prepared aerogel had a high specific surface area and good porosity, and it was able to remove oil spills from water surfaces with a high oil‐absorption capacity. In addition, this oil‐absorption material is green and reproducible and will not cause secondary pollution to the environment. Thus, TMCS/rGO/CNFs have a strong potential to become versatile, efficient, and safe absorbers for treating oil spills.