Introduction

The incessant progress of nanotechnology in material science has opened new pathways for developing new functional materials. Among them, reduced graphene oxide (RGO) (a one-atom-thick and two-dimension planar monolayer of sp2-hybridized carbon into a two-dimensional (2D) honeycomb lattice) applications in has great promise for potential many fields (Novoselov et al. 2004; Geim and Novoselov 2007; Erickson et al. 2010) due to its unique chemical, physical, electrical, and mechanical properties (Wang et al. 2009; Xiao et al. 2014). A number of synthetic methods have so far been developed for preparing high-quality graphene (Mao et al. 2011; Kwon et al. 2011). Among them, the chemical reduction of graphene oxide (GO) is considered one of the most promising methods for the large-scale synthesis (Chen et al. 2010; Park and Ruoff 2009). The method includes the steps of oxidation of graphite, dispersion/suspension of oxidized graphite sheets, and reduction of the graphite oxide to graphene. The oxidation introduces the hydroxyl and epoxide groups on the basal planes, as well as the carboxyl and carbonyl groups mostly at the sheet edges of graphene, which increases the hydrophilicity of graphene and the layer distance between GO sheets. This in turn facilitates the exfoliation and dispersion of GO sheets. The reduction of the dispersed GO to remove the oxygen-containing groups that has been previously introduced in the oxidation step and in order to return the C=C bonds to the molecule, various types of reductants such as hydrazine monohydrate (Tung et al. 2009; Some et al. 2013), NaBH4 (Shin et al.2009), and hydroquinone (Wang et al. 2008) have been used.

Nowadays, research on graphene is a hot topic because of its novel and unique properties such as extremely high carrier mobility, high optical transparency, and superior mechanical flexibility (Novoselov et al. 2012). Graphene-based materials hold strong potential for practical applications in nanoelectronics (Soldano et al. 2013; Brownson et al. 2011), energy-storage (Pumera 2011; Wu et al. 2012), polymer composites (Hwang et al. 2013), liquid crystal devices (Behabtu et al. 2010), and biosensors (Zhu et al. 2012). Moreover, graphene has many biological and medical applications (Cho et al. 2013), which mainly require aqueous stability and biocompatibility. The combination of carbohydrate and GO leads to the formation of hybrid nanostructures with high dispersibility in aqueous media mainly due to the high hydrophilicity of carbohydrates.

Among carbohydrates, cellulose (Siqueira et al. 2010; Namazi and Jafarirad 2008; Namazi et al. 2014) is one of the most widely used biopolymers that exhibits excellent physical properties such as good flexibility, excellent biocompatibility, and biodegradability, and it is also used as insulating component (Moon et al. 2011; Yadollahi and Namazi 2013; Namazi et al. 2012). Nanocrystalline cellulose (NCC) is an organic nanomaterial that has recently gained interest as a renewable, environmentally friendly and low cost reinforcing agent for composite materials (Lalia et al. 2013; Abdul Khalil et al. 2012). In addition, it has some appealing intrinsic properties such as high surface area, unique morphology, low density, and high mechanical strength (Habibi et al. 2010). Unlike many nanomaterials, NCC is not synthesized from molecular or atomic components; rather it is obtained by controlled acid hydrolysis of natural cellulose (Bai et al. 2009; Corrêa et al. 2010). The size of nanocrystals depends on the biological origin of cellulose. On the other hand, the use of NCC with exceptionally high mechanical properties instead of microcrystalline cellulose (MCC) can improve the mechanical properties of compounds for the high crystallinity of NCC (Moon et al. 2011; Lavoine et al. 2012; Corrêa et al. 2010).

More recently, some monosaccharides including glucose, fructose, and sucrose were proved to be good stabilizers of graphene in aqueous suspension (Zhu et al. 2010). Furthermore, some polysaccharide molecules were used as both stabilizer and physical cross-linking agents for the preparation of graphene-based multifunctional hydrogels (Shan et al. 2009; Rodríguez-González et al. 2012). Sodium carboxymethyl cellulose/graphene oxide nanocomposite films were fabricated recently (Yadav et al. 2013). The covalent grafting of chitosan (Pham et al. 2010; Bustos-Ramírez et al. 2013) and hydroxypropyl cellulose (Yang et al. 2011) onto the surface of GO has been done by esterification and click reactions (Ryu et al. 2013). The covalent functionalization of RGO with NCC employing click chemistry or other methods has not been reported yet.

NCC is a nontoxic, sustainable, biodegradable, recyclable material, which is produced from abundant renewable resources. The mentioned characteristics for NCC together with its unique mechanical and optical properties have generated a great interest in manufacturing NCC-based products in industrial scale. Therefore, we decided to use NCC for functionalization of RGO. On the other hand, it has been found that the completely protonated acid-form of NCC (H-NCC) is not well dispersible in water once it has been fully dried, even under fairly gentle conditions, while the NCC functionalized with GO has enhanced its dispersibility.

In this study, a stable aqueous suspension of graphene nanosheets fabricated with covalent grafting of NCC on graphene through click reaction. In brief, PG-NCC and GO-N3 were synthesized. Subsequently, the novel GO-click-NCC was created via the one-step alkyne-azide click reaction in the presence of Cu (I) (Scheme 1). At the end, the dispersed GO-NCC was chemically reduced with hydrazine to obtain a stable aqueous suspension of RGO-NCC.

Scheme 1
scheme 1

Synthesis of click-coupled RGO-NCC: i 2-chloroethyl isocyanate, DMF, RT, 24; ii NaN3, DMSO, 50 °C, 48 h; iii NaH, LiCl, DMF, 40 °C, 96 h; iv CuSO4·5H2O, sodium ascorbate, DMSO, RT, 52 h; and v hydrazine monohydrate, 70 °C, 36 h

Full size image

Materials

Medium molecular weight microcrystalline cellulose (MCC), 2-chloroethyl isocyanate, LiCl, NaNO3, KMnO4, propargyl chloride (PGCl), H2O2, H2SO4, HCl, methylene chloride, sodium diethyldithiocarbamate trihydrate, and other chemical reagents were obtained from Merck. Sodium ascorbate, copper sulfate pentahydrate, sodium hydride, and sodium azide were purchased from Sigma-Aldrich. Graphite (average particle size 30 µm) is commercially available, which was used without further purification. Dimethyl sulphoxide (DMSO), N,N-Dimethylformamide (DMF), and N, N-dimethylacetamide (DMA) were dried with molecular sieve before use.

Preparation

Preparation of NCC

20 grams of MCC were mixed with the aqueous solution of 64 wt % sulfuric acid (350 ml), and the mixture was stirred vigorously at 45 °C for 3 h. Fivefold dilution was then applied to the mixture to stop the hydrolysis reaction. The suspension was centrifuged to separate the crystals existing in the suspension. The crystals were then washed with distilled water, the mixture was centrifuged, and the crystals were separated again. The process was repeated ten times for each sample. The precipitate was finally placed in regenerated cellulose dialysis tubes (Fisher Scientific, Pittsburgh, PA, USA) having a molecular weight cut-off of 12,000–14,000, and dialyzed against distilled water for several days until the water pH reached a value of 7.0 (Liu et al. 2010).

Preparation PG-NCC

The propargyl cellulose was synthesized by Pai Peng et al.’s method (Peng et al. 2012) with some modification. Briefly, 0.1 g of NCC was stirred in 15 mL DMA at 120 °C for 2 h to allow it swell sufficiently; then 15 % LiCl (w/v) was added. Stirring was continued for 6 h to undertake uniform solubilization. Afterward, the mixture was cooled down to room temperature, followed by the addition of required quantities of NaH (the molar ratio of NaH to anhydrous glucose unit (AGU) was 4:1). After 30 min, the required quantity of propargyl chloride (the molar ratio of propargyl chloride (PG-Cl) to AGU was 10:1) was added slowly under ice cooling. The reaction temperature was kept at 40 °C, and the reaction was run for 96 h. After the required time, the product was precipitated with four volumes of ethanol. The precipitate (color: off-white) was filtered off, and washed with ethanol and distilled water until the pH of the filtrate was neutral. Finally, it was freeze dried.

Preparation of GO

GO was synthesized from the natural graphite powders using a modified Hummers’ and Offenman’s method (Hummers and Offeman 1958). In brief, 0.5 g of graphite, 0.5 g of NaNO3, and 23 mL of concentrated H2SO4 were mixed together and stirred in an ice bath for around 1 h. Then 1.5 g of KMnO4 was added slowly into the solution followed by stirring at 35 °C for 1 day to form thickened paste. Afterward, 23 mL of de-ionized water was added slowly into the solution to avoid the reaction temperature rising to a limit of 98 °C. This solution was then kept stirring for 30 min. Finally, 70 mL of de-ionized water and 5 mL of H2O2 (30 %) were poured into the mixture in sequence. This solution was then filtered by gravity filtration, and the filter cake was washed with de-ionized water and 3 % HCl solution alternatively. This filtered cake was then dispersed and washed with de-ionized water for several times by repeated centrifugation. In order to avoid the aggregation of GO in the drying process, the sample was obtained by freeze-drying. Normal drying process produces aggregated GO layers.

Preparation GO-Cl

5 mL of GO in anhydrous DMF (10 mg mL−1) was loaded into a 10-mL round-bottom flask equipped with a magnetic stirring bar. Then organic 2-chloroethyl isocyanate (2 mmol) was added, and the mixture was stirred under nitrogen. After 24 h, the slurry reaction mixture was poured into methylene chloride (50 mL) to coagulate the product. The brown product was then filtered, washed with additional methylene chloride (50 mL), and dried under vacuum (Wang et al. 2012).

Preparation of GO-N3

To prepare azide-functionalized GO, 6 mmol sodium azide powder was added to 50 mg of GO-Cl dissolved in 10 ml of post-ultrasonication DMSO (kept in an ice bath to prevent heating) for 10 min. Then the mixture was stirred and refluxed for 48 h at 50 °C in a constant temperature oil bath, thus the azide group was introduced on the GO-Cl sheets via the nucleophile substitution reaction of alkyl halide. Finally, ethyl acetate was used to eliminate any residual DMSO, and the black product was filtered and dried under vacuum.

Preparation of GO-NCC-GO click reaction

The general procedure for azide-alkyne [3 + 2] dipolar cycloaddition to prepare the NCC-click-GO polymer is as follows: after 0.1 g of PG-NNC was dispersed in 10 mL of DMSO, CuSO4-5H2O (5 mg in 0.5 mL water), sodium ascorbate (10 mg in 0.5 mL water), and GO-N3 (0.1 g) were added. The mixture was stirred at room temperature for 48 h. 15 mg of sodium diethyldithiocarbamate trihydrate was then added to the mixture to remove the copper catalyst. Isolation took place by precipitation in 400 mL of acetone followed by centrifugation. Then the precipitate was washed with water. The gray product was finally obtained by freeze-drying.

Reduction of GO-NCC to RGO-NCC

RGO-NCC was fabricated by reducing the GO-NCC prepared above. 20 mg of GO-NCC was dispersed in 10 mL of deionized water under sonication for 3 h. Then the resulting yellow exfoliated GO-NCC suspension was reduced with 20 µl hydrazine. The reduction was carried out on a shaker at 400 rpm and 70 °C for 36 h. At the end of the reduction, the RGO-NCC was filtered out from the corresponding suspension and washed twice with water. The dark gray product was finally obtained by freeze-drying.

Characterization and methods

Infrared spectra were obtained on a FTIR spectrometer (Bruker Instruments, model Aquinox 55, Germany) in the 4,000–400 cm−1 range at a resolution of 0.5 cm−1 as KBr pellets. The carbon nuclear magnetic resonance (13C NMR) and DEPT135 NMR spectra were recorded in DMSO-d6 in LiCl in 5-mm NMR tubes with a Bruker 100 MHz spectrometer. Raman spectroscopy (Bruker Instruments, model SENTERRA (2009), Germany, Laser wavenumber: 785 nm) was used to confirm the functionalization of GO. The XRD patterns of the samples were obtained by Siemens diffractometer with Cu-ka radiation at 35 kV in the scan range of 2 h from 2 to 70∘ and scan rate of 1∘/min. All of the analyzed samples were in powdery form. Thermo gravimetric analysis (TGA) was performed with a TGA-PL thermal analyzer under air atmosphere from room temperature up to 700 °C at a heating rate of 10 °C/min. Scanning electron micrographs (SEM) were obtained with a LEO 1430VP scanning electron microscope operating at 15 kV. UV–Vis spectroscopy was carried out on a Perkin-Elmer Lambda 35 UV–Vis absorption spectrometer at room temperature. The particle sizes of samples were determined by dynamic light scattering (DLS) using a 90 Plus particle size analyzer equipped with diode laser operating at 658.0 nm. The samples were diluted with distilled water to adjust their solid content to 0.05 wt %, and directly placed in the cell. All measurements were carried out at 25 °C.

Results and discussion

Covalent functionalization of GO with NCC

In a common preparation, GO was synthesized from graphite powder by the Hummers’ method. The functional groups containing abundant oxygen make GO sheets vigorously hydrophilic, which improves their solubility in water and provides possibility for more functionalization. As shown in Fig. 6, DLS analysis demonstrated that the average particle size of GO was less than 200 nm. To make use of the Cu-catalyzed Huisgen cycloaddition click reaction, either graphene should contain azide moieties or the second reagent should contain alkynyl moieties (Kolb et al. 2001). In the present study, azide and alkynyl moieties were attached to GO and NCC, respectively (Scheme 1). The azide groups on GO were generated in two steps. Firstly, GO-Cl was prepared by the reaction of GO with 2-chloroethyl isocyanate. The chemical treatment with 2-chloroethyl isocyanate converted hydrophilic functional groups of GO into hydrophobic functional groups, which enabled GO-Cl to be well dispersed in DMSO, DMF, and NMP. In the second step, GO-N3 was prepared by reacting of GO-Cl with NaN3. The presence of an absorption azide band in the FT-IR spectra confirmed the presence of an azide moiety in GO-N3. 13C NMR analysis was used in order to get more detailed information of the grafted azide on functionalized graphene. Then NCC was prepared through acid hydrolysis of MCC. DLS analysis demonstrated that the average particle size of NCC was in the range of 120–150 nm (Fig. 6). To covalently attach propargyl moieties to the surface of the nanocrystals, we followed the reaction pathway illustrated in Scheme 1. The alkynyl groups on NCC were generated by reaction of NCC and propargyl chloride using NaH in DMA/LiCl. The FT-IR spectrum of PG-NCC indicated the successful incorporation of terminal alkyne groups onto the cellulose chains (two absorption bands at 3,286 and 2,190 cm−1, corresponding to ≡C–H stretching and C≡C stretching, respectively). Finally, the click reaction was conducted at room temperature with CuSO4·5H2O and sodium ascorbate as the catalyst between alkynyl-functionalized NCC and azidated GO (Scheme 1). Ascorbic acid has been reported as an effective reducing agent for GO that can compete with hydrazine (Fernandez-Merino et al. 2010). Hence, it has an added advantage for click-coupling reactions. FT-IR, 13CNMR, and Raman spectra confirmed a successful coupling reaction as depicted by the five-membered triazole ring with loss of the azide and alkynyl peaks. For the complete reduction of GO in GO-NCC to RGO-NCC, the dispersed GO-NCC was chemically reduced with hydrazine to obtain a stable aqueous suspension of RGO-NCC. This was confirmed by the elimination of GO peaks in the FT-IR spectrum and the red-shift of absorbance peak from 230 to 272 nm in UV–Vis spectroscopy. DLS analysis showed that the average particle size of RGO-NCC was in the range of 70–90 nm (Fig. 6).

FTIR analysis

FT-IR spectra in the 4,000–400 cm−1 wave number range in the equal amounts of GO, GO-N3, NCC, PG-NCC, and GO-NCC are shown in Fig. 1. In the spectrum of GO, the absorbance bands at 1,061, 1,380, 1,620, 1,716, and 3,375 cm−1 can be attributed to C–O epoxy, C–O carboxylic acid, C=C aromatic, C=O carboxylic acid, and O–H (broad-coupling hydroxyl group), respectively (Chen et al. 2010). The spectrum of GO-N3 shows a very broad and intense peak of the O–H group’s stretch at about 3,400 cm−1, which is also present in the starting material, and a peak at 2,115 cm−1 that is consistent with the asymmetric stretching of the azide group. The absorbance bands at 3,410, 2,900, 1,370, 1,313, 1,264, 1,201, 1,154, 1,030, and 898 cm−1 are associated with NCC (Khan et al. 2012). In the spectrum of PG-NCC, two new absorption bands at 3,286 and 2,190 cm−1 (corresponding to≡C–H stretching and C≡C stretching, respectively) indicate the successful inclusion of terminal alkyne groups onto the cellulose chains. In the spectrum of RGO-NCC, the absorbance band at 1,655 cm−1 corresponds to aromatic ring (five-member triazole ring). The elimination of azide and alkynyl peaks shows that the reaction between alkyne groups on the NCC and azide groups on the GO has taken place via click reaction. Furthermore, the intensity of OH group’s strength peaks in the RGO-NCC spectrum was decreased (not eliminated because the OH strength of cellulose is located in this area), and CO peaks were eliminated. Which showed that GO has been reduced to RGO during the reaction.

Fig. 1
figure 1

FTIR spectra of GO, GO-N3, NCC, PG-NCC, and RGO-NCC

Full size image

13C NMR analysis

To further verify the occurrence of click reaction, PG-NCC, GO-N3, and RGO-NCC were investigated by DEPT135 and 13C NMR spectroscopies in DMSO/LiCl by the number scan of 10 k and d1 = 20 s (Fig. 2). In the 13C NMR spectrum of PG-NCC, the signals at 104.5, 76.2–74.3, 65.2, and 59.7 ppm are attributed to the C-1, C3-C-5, C-6, and C-2 of the functionalized glucose units resulting from cellulose, respectively. The signals at 104.8 and 66.7 ppm are attributed to the C-1′ and C-2′ of the unfunctionalized glucose units resulting from cellulose, respectively (Kono et al. 2002; Qi et al. 2009). Other signals of the unfunctionalized glucose units overlap with the functionalized glucose units’ peaks. The presence of signals at 82.0, 74.5, and 31.7 ppm is due to the carbons of propargyl groups corresponding to the C-8, C-9, and C-7 of PG-NCC, respectively. In the DEPT135 spectrum of PG-NCC, the resonances of methylene groups (C-7 and C-6) are seen in the reverse phase. The presence of these signals indicates the successful substitution of the propargyl groups on the cellulose chains. In the spectrum of GO-N3, the peaks at 59, 66, 73, and 78 ppm are attributed to the epoxide, hydroxyl, and two methylene groups, respectively. The resonance at 130 ppm belongs to the un-oxidized sp2 carbons of the graphene network, and at 155 ppm, it arises from the urethane groups. In the DEPT135 spectrum of GO-N3, the resonance of methylene groups is seen in the reverse phase. In the spectrum of RGO-NCC, the presence of signals at 127.4 and 147.1 ppm is attributed to the triazole ring carbons, which supports the successful click reaction between PG-NCC and GO-N3. Moreover, the signals at 38, 61, 77, and 81 ppm are attributed to the methylene groups on RGO-NCC. It was confirmed by DEPT135 analysis results.

Fig. 2
figure 2

13C NMR and DEPT135 NMR of PG-NCC, GO-N3, and GO-NCC

Full size image

Raman analysis

Raman spectroscopy, which is sensitive to the electronic structure, is a valuable tool to study the ordered/disordered crystal structures of carbon-based materials. The known characteristics of Raman spectra for carbon materials are D-band(C–C) and G-band(C=C) that are typically located at about 1,350 and 1,580 cm−1 (Stankovich et al. 2007). Figure 3 shows the Raman spectra of GO, GO-NCC, and RGO-NCC. The Raman spectrum of GO, as expected, displays a G peak at 1,585 cm−1 corresponding to the first-order scattering of the E2 g mode, and the D band at 1,363 cm−1 assigned to the defects introduced upon the oxidation process. The GO-NCC intensity of D-band is increased, supporting the statement that covalent bonding occurs between NCC and GO; this is expected as functionalization results in the increase of sp3-hybridized sidewall carbons. The Raman spectrum of RGO-NCC also contains both G and D bands; however, there is an increase in the D/G intensity ratio compared to that of GO-NCC. This change suggests a decrease in the average size of the sp2 domains upon the reduction of the exfoliated GO-NCC, and can be explained that the created RGO-NCC is smaller in size but is higher in number in comparison to GO-NCC.

Fig. 3
figure 3

Raman spectra of GO, GO-NCC and RGO-NCC

Full size image

UV–Vis analysis

GO, GO-NCC, and RGO-NCC were determined using UV–Vis spectroscopy in aqueous suspension. Figure 4 shows that GO (A) exhibits strong bands centered at 230 and 301 nm, corresponding to the π→π* transitions of the aromatic C=C band and the n→π* transitions of the-COOH band in GO, respectively (Shin et al. 2009). The absorption peak of GO-NCC (B) was red-shifted from 230 to 262 nm. The red-shifted value is smaller than that of fully reduced graphene sheets, which can reach 272 nm, suggesting that GO-NCC was only partially reduced, and its electronic conjugation level is still lower than that of perfect graphene sheets. This partial reduction of GO is probably due to the existence of sodium ascorbate (the salt of ascorbic acid, known as Vitamin C) in the click reaction, which is added originally in order to avoid the oxidization of Cu (I). Vitamin C, as a natural antioxidant, possesses a similar efficiency to hydrazine in the reduction of GO (Fernandez-Merino et al. 2010). The absorption peak of RGO-NCC (C) was red-shifted from 230 to 272 nm; this confirms the reduction of GO-NCC to perfect RGO-NCC.

Fig. 4
figure 4

UV-Vis spectra of a GO, b GO-NCC, and c RGO-NCC

Full size image

XRD analysis

NCC, GO, and RGO-NCC were determined using the XRD patterns. The diffractogram of NCC (A) shows the characteristic peaks at 2θ = 14.8∘ and 22.7∘ (Shang et al. 2013). The characteristic 2θ peak of GO (B) appearing at 12.1∘ corresponds to the 001 inter planar spacing of 0.73 nm caused by the oxygen-rich groups on both sides of the sheets and the water molecules trapped between the sheets (Ryu et al. 2013). For the GO-NCC (C), the peak at 2θ = 12.1∘ disappeared, which clearly demonstrates the formation of a fully exfoliated structure of GO sheets in the polymer matrix and the disappearance of the regular and periodic structure of graphene, whereas the peak at 2θ = 22.7∘ decreased drastically, which can be attributed to the decrease in crystallinity. This suggests that the hydrogen bonding ability of NCC was reduced after the grafting of NCC onto the RGO backbonen (Fig. 5).

Fig. 5
figure 5

XRD patterns of a NCC, b GO and c GO-NCC

Full size image

DLS analysis

In order to determine the size of NCC, GO, and RGO-NCC, dynamic light scattering (DLS) method was applied in water. The size of NCC and GO was found to be in nanometric range, which was decreased after click reaction. Figure 6 demonstrates that the average particle size of NCC is in the range of 120–150 nm (Lin et al. 2011). After preparation of GO by oxidation of graphite (average particle size 30 µm) and then using sonication, the average hydrodynamic diameters of the graphene oxide sheets were decreased to below 200 nm. The size of RGO-NCC was in the range of 80–90 nm. Reduction in the size of RGO-NCC compared to initial compounds is due to the formation of covalent bonding between RGO and NCC that causes the exfoliation of RGO sheets (Ryu et al. 2013).

Fig. 6
figure 6

DLS diagrams of NCC, GO and RGO-NCC

Full size image

Thermogravimetric analysis

The TGA curves of GO (A), NCC (B), and RGO-NCC (C) are shown in Fig. 7. It is apparent that RGO-NCC has different decomposition patterns, compared to pure NCC and GO. GO shows two-step thermal degradation process. The first step can be attributed to the loss of H2O around 100 °C, and the second step appeared around 225 °C for the loss of acidic functional groups and residues with total remaining of about 65 %. Above this temperature, GO is almost linearly stable (Moon et al. 2010). The NCC degradation started at slightly lower temperatures and occurred over a wider temperature range of 250–420 °C with the total remaining of about 22 %. It also has only one pyrolysis process (Li et al. 2012). Furthermore, RGO-NCC exhibits no weight loss at 225 °C, suggesting that the oxygen functional groups of RGO are reduced by functionalization. This disappearance of the oxygen functional groups can probably be attributed to their elimination in the functionalization process, as well as to their conversion into more stable triazole rings. The thermal stability of RGO-NCC is higher than that of pure NCC, which can be due to the reduction of GO to RGO. Furthermore, the degree of functionalization from the gradual mass loss of the RGO-NCC suggests that around 23 mass % has been functionalized covalently. It was calculated by the mass loss between 28 and 450 °C over the whole temperature range (Yang et al. 2011; Shen et al. 2010).

Fig. 7
figure 7

TGA curves obtained for a GO, b NCC and c RGO-NCC

Full size image

SEM analysis

The SEM images of NCC, RGO, and RGO-NCC are shown in Fig. 8 (A-C). The SEM image of NCC films (A) shows a layered pattern of NCC crystallites oriented in a similar direction. The SEM image of RGO (B) consists of randomly aggregated, thin, crumpled sheets closely connected with each other and forming a disordered solid. The SEM image of NCC-RGO (C) shows the successful grafting of RGO on NCC.

Fig. 8
figure 8

SEM images of a NCC, b RGO, and c RGO-NCC

Full size image

Conclusions

As showed, a simple method was developed, involving a three-step reaction pathway, to covalently attachment of NCC molecules onto the surface of GO. Structural details were provided with FTIR spectroscopy, Raman, DEPT135 and 13C NMR spectroscopy. The complete exfoliation of graphene in the NCC matrix was examined using the XRD measurements. The analyses results confirmed a successful coupling reaction as depicted by the five-membered triazole ring. The degree of functionalization from the gradual mass loss of RGO-NCC suggests that around 23 wt % it has been covalently functionalized.