1 Introduction

Cellulose aerogels are a class of important cellulose products and have attracted much attention since the first report by Kistler [1] in 1931. Their porous three-dimensional (3D) network structure endows themselves numerous merits like large specific surface area, high porosity, and low density [24]. Compared with inorganic aerogels, cellulose aerogels are gaining more interest due to their environment friendliness, biodegradability, and biocompatibility. As a result, cellulose aerogels have many potential applications such as catalyst carriers, adsorbents, biomedicines, tissue engineering, fuel cells, and sound or heat insulation materials [58]. Cellulose aerogels are generally prepared by the cellulose dissolution, regeneration, and dry process. However, the poor dissolution of cellulose in common aqueous or organic solutions, which is attributed to the complexity of biopolymeric network, the partially crystalline structure, and the extended noncovalent interactions among molecules [9], hastens the development of cellulose solvents. Especially, among the various cellulose solvents, several green, nontoxic, and nonvolatile solvent systems are high profile such as ionic liquid, N-methylmorpholine-N-oxide (NMMO), and some alkali-based systems like NaOH/PEG solution [1013].

For their more widespread utilization, the critical weakness related to the inferior mechanical properties of cellulose aerogels, which is possibly derived from the defects such as dangling ends and loops in their structures [14], is the main challenge requiring urgent improvement. Besides, some authors also acclaim that the fragility is ascribed to the weak hydrogen bonding interaction and the quick rearrangement of cellulose chains [15]. Pre-gelation processing of cellulose solution is an interesting approach to enhancing the mechanical properties of cellulose aerogels, which is conducted by storing cellulose solution at pre-gelation temperatures leading to the formation of network units with stronger cross-linking structure [15, 16]. However, the pre-gelation process is usually extraordinarily time-consuming, and the improvement in mechanical properties is also rather limited. In contrast, incorporating with reinforcement additives is a relatively wider technique to improve the mechanical properties of cellulose aerogels. The reported fillers include cellulose whisker [17], clay [18], silica [19], organic polymers (e.g., polystyrene, poly(methyl methacrylate), and polypyrrole) [20, 21], metal nanoparticles or their related compounds (e.g., Ag and CoFe2O4) [22, 23]. However, the exploration on the appropriate reinforcement additives still needs to be continued, which is attributed to not only the demand for improving mechanical strength of cellulose aerogels, but also various fantastic performances of the composites inherited from both aerogels and fillers.

Graphene is composed of sp 2-bonded carbon atoms arranged in hexagonal pattern in a 2D plane, which has high aspect ratio, supernormal mechanical property, and intriguing transport phenomena. Therefore, graphene is wildly considered to be useful in different fields such as composites, optical devices, ultrasensitive sensors, and genie widget [24]. Moreover, graphene has been extensively reported as a kind of good filler for various polymer matrixes, and the resulting products exhibit many striking features in structure, strength, and macro–micro performances [25, 26]. Therefore, incorporation of graphene nanosheets into cellulose aerogels might effectively improve the inferior mechanical properties of cellulose aerogels and introduce new properties.

Herein, we show a facile, simple, and green preparation method of graphene/cellulose (GC) aerogels by incorporating graphene nanosheets into the cellulose aerogel matrix using NaOH/PEG aqueous solution as the processing solvent. The resulting GC aerogels were subsequently characterized by scanning electron microscope (SEM), energy-dispersive X-ray spectroscopy (EDX), N2 adsorption measurements, X-ray diffraction (XRD), Raman spectroscopy, and thermogravimetric analysis (TGA). The mechanical properties were investigated using a mechanical tester. As a kind of green porous materials, the application potential of GC aerogels was also studied as eco-friendly oil adsorbents. The oil adsorption efficiency of GC aerogels was measured and compared with that of the pure cellulose (PC) aerogels.

2 Materials and methods

2.1 Materials

Waste coconut shell was grinded and then screened through a 60-mesh sieve, and the resulting powder was subsequently collected and dried in a vacuum oven at 60 °C for 24 h before used. All chemical reagents with analytical grade were supplied by Shanghai Boyle Chemical Co., Ltd. (China) and used without further purification.

2.2 Preparation of graphene

Typically, graphene was synthesized from natural graphite powder by a modified Hummers method [27, 28]. Briefly, graphite powder was first oxidized by concentrated nitric acid and sulfuric acid (1:2, v/v). The reaction was conducted in an ice bath, and potassium chlorate was added slowly. Then the mixture was heated to 65 °C and kept at this temperature for 120 h to complete the oxidation reaction. After being thoroughly washed and filtered by distilled water (until pH 7), the graphite oxide was obtained. Thereafter, a mild 2-h ultrasonication treatment was carried out to exfoliate the graphite oxide in a mixture of ethanol and water. The resultant was subsequently reduced to graphene by hydrazine hydrate at 100 °C for 24 h. After being centrifuged, washed, and vacuum-dried, the graphene powder was obtained.

2.3 Preparation of GC aerogel

The cellulose was isolated from the coconut shell powder by a facile chemical treatment method, and the purified cellulose was then dissolved in the NaOH/PEG solution to form a 2 wt% homogeneous cellulose aqueous solution. The detailed process could be referred to our previous reports [2931]. The graphene aqueous dispersion (0.2 mg mL−1) obtained by 4-h ultrasonication (500 W) was added in the aforementioned homogeneous cellulose solution at a 99.5:0.5 mass ratio of cellulose to graphene. Then, the mixture was magnetically stirred for 30 min and immediately frozen at −15 °C for 3 h. After that, the frozen cake was subjected to the regeneration treatment by immersing it in a 1 % hydrochloric acid for 6 h at room temperature, repeating this process until the formation of a cylindrical GC hydrogel. Finally, GC hydrogel was successively rinsed with distilled water and tert-butyl alcohol several times to remove residual ions and water, and then underwent a freeze drying treatment at −35 °C for 48 h in an approximate vacuum (25 Pa), and the following GC aerogel was obtained. Moreover, PC aerogel was prepared following the aforementioned process without adding the graphene.

2.4 Characterizations

SEM images were obtained by a FEI Quanta 200 SEM coupled with EDX. N2 adsorption measurements were carried out on a Micromeritics Tristar II 3020 surface area analyzer. The specific surface area and pore size distribution were calculated by the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods. XRD technique was performed on an XRD instrument (D/max 2200, Rigaku) using Ni-filtered Cu Kα radiation (λ = 1.5406 Å) at 40 kV and 30 mA. Scattered radiation was detected in the range of 2θ = 5–70° at a scan rate of 4° min−1. Raman spectra were recorded using a Raman spectrometer (Renishaw inVia, Germany) with a helium/neon laser (785 nm) as the excitation source. TGA was implemented on a TG analyzer (TA, Q600) under N2 atmosphere at a heating rate of 10 °C min−1. The compression tests were performed on a mechanical tester (CMT6503, MTS) with a compressing velocity of 2 mm min−1. The Young’s modulus was calculated as the slope of the linear region of the stress–strain curve. The energy absorption was taken as the area below the stress–strain curve between 0 and 70 % strain.

3 Results and discussion

Figure 1 shows the SEM images of PC aerogels and GC aerogels, respectively. It is observed in Fig. 1a that the nanoporous PC aerogels have dense and interconnected 3D network structure, which is formed by spontaneous physical cross-linking of cellulose chains due to many surface hydroxyl groups that can easily form hydrogen bonding linked networks [32]. Meanwhile, the developed 3D network also demonstrates the good dissolution capacity of NaOH/PEG solvent. Whereas for GC aerogels, it is obvious that the 3D architecture is almost completely disappeared and covered tightly with the dense sheet structure (Fig. 1b). Moreover, from the EDX spectra (the inset in Fig. 1), the C, O, and Au elements are detected. The Au element originates from the coating layer used for electric conduction during the SEM observation. Besides, the mass ratio of C/O of GC aerogels is approximately 2.77, much higher than that of PC aerogels (ca. 1.44). The remarkable improvement in carbon proportion reveals the potential presence of graphene.

Fig. 1
figure 1

SEM images of a PC aerogels and b GC aerogels, respectively. The insets show the corresponding EDX spectra

Figure 2a presents the N2 adsorption–desorption isotherms of PC aerogels and GC aerogels, respectively. Apparently, both aerogel samples exhibit similar change tendency belonging to type-IV adsorption isotherms according to the IUPAC classification [33], which indicates the presence of mesopores (2–50 nm). Moreover, the curves rise slowly at low relative pressure, which reveals the presence of little or no micropore (<2 nm) [34]. At higher relative pressure (P/P 0 at 0.6–0.9), the obvious hysteresis loops could be found, which is caused by capillary condensation, suggesting the characteristic of mesoporous structure [35]. Furthermore, the large amount of adsorption (without apparent limitation) occurring at relative pressures above 0.9, demonstrates the presence of macropores (>50 nm) [36]. Moreover, it could be seen in Fig. 2b that the pore diameters are within the scope of 3–110 nm with several peaks concentrated at 3–20 nm, which further demonstrates that both PC aerogels and GC aerogels are primarily composed of mesopores and macropores. The BET calculation gives the BET surface area for PC aerogels equal to 156 m2 g−1; and the BJH method gives the pore volume equal to 0.95 cm3 g−1, respectively. For GC aerogels, after being embedded with graphene nanosheets with high specific surface area, GC aerogels show larger BET surface area (205 m2 g−1) and total pore volume (1.11 cm3 g−1). The results suggest that the incorporation of graphene makes contribution to the significant enhancement in the pore characteristic parameters.

Fig. 2
figure 2

a N2 adsorption–desorption isotherms and b pore diameter distributions of PC aerogels and GC aerogels, respectively

The XRD patterns of as-prepared graphene, PC aerogels, and GC aerogels are shown in Fig. 3a, respectively. The graphene presents a broad diffraction peak at around 25.0° with an obvious disappearance of the characteristic peaks (generally at 2θ = 12° and 2θ = 26.5°). It can be attributed to the exfoliation of graphene into a monolayer or few layers and long-range disorder in graphene [3739]. Moreover, PC aerogels and GC aerogels exhibit similar XRD patterns, in which the all diffraction peaks at around 11.7°, 20.1°, and 21.9° could be indexed to typical cellulose II crystalline structure corresponding to (\(\bar{1}10\)), (110), and (200) planes [40]. In addition, the XRD pattern of GC aerogels does not show any characteristic peaks derived from graphene, possibly attributed to the disorder, low content, and good dispersion of graphene [41, 42].

Fig. 3
figure 3

a XRD patterns of as-prepared graphene, PC aerogels and GC aerogels, respectively. b Raman spectra of PC aerogels and GC aerogels, respectively

Figure 3b presents the Raman spectra of PC aerogels and GC aerogels, and the fast Fourier transform (FFT) filtering algorithm was used to create smooth spectra. As shown, the peaks at 897, 1103, 1377, and 1459 cm−1 in the Raman spectrum of PC aerogels are originated from HCC and HCO bending at C6, CC and CO stretching, HCC, HCO and HOC bending, and HCH and HOC bending of cellulose structure [43], respectively. For GC aerogels, apart from the characteristic peaks from cellulose (899 and 1095 cm−1), two strong peaks at around 1340 and 1600 cm−1 are attributed to the D-band (K-point phonons of A 1g symmetry) and G-band (E 2g phonons of Csp 2 atoms) of graphene [44], which demonstrates that the graphene nanosheets have been successfully embedded into the cellulose aerogels. Furthermore, the broadening of D and G bands with a strong D line reveals disordered graphitic crystal stacking of the graphene nanosheets [45], in accordance with the aforementioned XRD results.

In view of the importance of thermal stabilities of nanocomposites in practical applications, the thermal stabilities of PC aerogels and GC aerogels were investigated by TGA and derivative thermogravimetry (DTG). As shown in Fig. 4, for the both samples, the small weight loss below 150 °C is attributed to the evaporation of adsorbed water; besides, only one exothermic peak is centered at 335–370 °C, which is related to cellulose pyrolysis [46]. Compared with PC aerogels, GC aerogels apparently show more superior thermal stability. In detail, PC aerogels start to decompose at 216.8 °C, whereas GC aerogels begin at 253.7 °C; at maximum degradation rate, the decomposition temperature occurs at 337.9 °C for PC aerogels and 366.7 °C for GC aerogels. This rising trend of decomposition temperature implies that the incorporation of graphene with high heat resistance in the aerogel is helpful to significantly increase the heat resistance.

Fig. 4
figure 4

a TG and b DTG curves of PC aerogels and GC aerogels, respectively

The stress–strain curves of PC aerogels and GC aerogels are presented in Fig. 5, respectively. The reinforcing effect of graphene is rather significant. Compared with PC aerogels, the Young’s modulus and energy absorption of GC aerogels increase by 168 % from 508 to 1361 kPa and 77 % from 104 to 184 kJ m−3, respectively. The significant improvement further confirms that the graphene is a good reinforcing additive for the cellulose aerogels. In addition, for the curves, the compression stress–strain behavior could roughly be divided into four regions [47]. First, at low strains (<6 %), the stress–strain behavior is linear elastic in nature primarily due to elastic cell wall bending. Second, a stark transformation from linear behavior to nonlinear behavior could be found. Third, in higher strain, a horizontal plateau region appears after reaching yield stress, and the collapse behavior would happen due to the plastic hinges resulted from plastic yielding of the cell wall [48]. Final, the bent cell walls start to touch, and the materials show considerable stiffening due to densification of the loose porous 3D network structure. Actually, the two-dimensional (2D) nanopaper-like structure of graphene has stronger load-bearing capacity than the 3D skeleton of cellulose aerogels [14]. Besides, the high aspect ratio of graphene also plays an important role in the enhancement in the resistance to pressure.

Fig. 5
figure 5

Stress–strain curves of PC aerogels and GC aerogels, respectively

Frequently occurred oil spills have brought great hazard to natural environment and human health. As a class of green biomaterials with large specific surface area and pore volume, cellulose aerogels are frequently considered as promising adsorbents for the treatment of oil-containing wastewater or oil spills. Therefore, it has some significance to research the oil adsorption property of PC aerogels and GC aerogels. An easily operated oil adsorption test was designed, i.e., immersing the samples (ca. 30 mm in diameter and 15 mm in height) into 100-mL beakers with 80 mL waste machine oil for 6 h. The weights of the samples were measured before and after the tests. The oil adsorption property is estimated by the oil adsorption ratio calculated by dividing the weight after the test by the original dry weight. After the 6-h adsorption, as shown in Fig. 6, both the aerogels naturally sank below the liquid surface indicating that a large number of oil molecules had entered into the pores of the aerogels. The absorption efficiency of GC aerogels is estimated to be 20.4 g g−1, which is 33.3 % higher than that of PC aerogels (15.3 g g−1). Besides, the aerogels maintain the original sizes and shape before and after the tests, and no obvious cracks and collapse can be seen, which suggests the favorable shape stability of the aerogels in the oil. In addition, apart from the superior oil adsorption ability, the stronger mechanical strength of GC aerogels is also considered as a more attractive alternative for the fabrication of green oil adsorbent, which is beneficial to resist various potential damages such as impact and compression. Therefore, GC aerogels draw more advantages than PC aerogels to develop green high-performance adsorbents.

Fig. 6
figure 6

PC aerogels and GC aerogels at the beginning a and ending b of the oil adsorption tests

4 Conclusions

A mild, easy and environmentally friendly method is proposed to fabricate GC aerogels nanocomposites using the green NaOH/PEG aqueous solution as the processing solution. The stacked self-assembly sheet structure is tightly covered on the 3D network skeleton of cellulose aerogels, leading to the significant improvement in the specific surface area, pore volume, thermal stability, mechanical properties, and oil adsorption efficiency of GC aerogels. Considering the stronger mechanical strength and thermal property as well as oil-absorbing ability, GC aerogels are more suitable for the treatment of oil spills under the harsh environment comparing to PC aerogels.