1 Introduction

Reduced graphene oxide (rGO), the product of oxidation and subsequent reduction of graphite [1,2,3], serves as a replacement to the pristine graphene in industrial applications owing to its large-scale production capabilities [4, 5]. Though it supports scalable production, it compromises in quality due to the presence of defects and oxygen functional groups (epoxy and hydroxyl groups on basal plane and carbonyl and carboxylic groups at the plane edge) [6] incepted during oxidation/reduction process. Thus, it is not widely commercialized in electronics and other energy-related applications even after these many years of discovery. However, not all the graphene-related applications require high-quality rGO and some application even desires rGO with specific defect types [7] and functional groups, e.g. Drug delivery [8], Water treatment [9, 10] and Microwave absorption [11].

In addition, industrial applications of rGO demand a simple, fast, cost-effective and eco-friendly synthesis method [12]. Conventionally, rGO is synthesized in 3 steps via a top-down approach. (1) Oxidation of graphite into graphite oxide, (2) exfoliation of graphite oxide into graphene oxide and (3) reduction of graphene oxide [2]. Quality of the rGO largely depends on the adopted reduction method. Most commonly used reduction methods are chemical, thermal and hydrothermal reduction. Chemical reduction is a largely practiced method due to its ability for mass production of desired quality rGO. However, it involves hazardous chemicals like hydrazine or sulfonate [13, 14], whose residues affect the structure and properties of the prepared rGO and makes it toxic and non-eco-friendly. Whereas, thermal reduction doesn’t use any toxic chemicals but it requires high thermal energy (temperature of 900–1000 °C) and also it introduces undesired defects [15].

In comparison, hydrothermal reduction offers a simple, fast, eco-friendly route as it utilizes only water and sometimes a mild reducing agent [16,17,18]. Furthermore, hydrothermal experimental set up contains a Teflon autoclave and a furnace that suits industrial large-scale production. However, like chemical and thermal methods, hydrothermal reduction itself is a separate step along with oxidation and exfoliation, which makes the process time-consuming. On the other hand, bottom-up hydrothermal synthesis is simple and can serve as a better alternative. In this method, a carbon source is subjected to carbonization via hydrothermal treatment to obtain graphene-related materials [19, 20].

Recently, glucose was proposed as a carbon source to produce graphene oxide [21]. Under high temperature and pressure, glucose gets converted into graphene oxide through successive chemical modification. Furthermore, glucose is also shown to reduce graphene oxide under hydrothermal conditions. Interestingly, glucose selectively reduces hydroxyl and epoxide groups of the graphene oxide whereas carbonyl functional groups remain unreduced [22, 23]. This suggests that glucose can transform into carbonyl-rich partially reduced graphene oxide if subjected to sufficiently long hydrothermal treatment. However, the reaction temperature should not exceed 160 °C as it results in carbon spheres formation [24]. This process is cost-effective and eco-friendly because it involves glucose, an abundant material and is carried out at a moderate temperature. Moreover, the carbonyl functional groups present in the sample offer a possibility for further functionalization and also provide good dispersibility in water which makes them useful in biomedical [25,26,27] and water treatment related applications [9].

In this context, the present study describes a simple, one-step, cost-effective, eco-friendly synthesis of carbonyl-rich partially reduced graphene oxide by hydrothermal method using glucose as a sole precursor as well as the reducing agent. The study discusses the detailed characteristics of the synthesized rGO evaluated using X-ray Diffraction (XRD), X-ray Photoelectron Spectroscopy (XPS), Fourier Transform Infrared Spectroscopy (FTIR), UV–Vis Spectroscopy, Raman Spectroscopy, Scanning Electron Microscopy (SEM), and Dynamic Light Scattering (DLS) characterization techniques. We have also proposed a plausible reaction mechanism for the formation of carbonyl-rich partial reduced graphene oxide to provide insight into the structural evolution.

2 Experimental section

2.1 Materials

β–d glucose was purchased from Sisco Research Laboratories Pvt. Ltd, India and used as received. Double distilled water was used throughout the process.

2.2 Synthesis of carbonyl-rich partially reduced graphene oxide

Carbonyl-rich partially reduced graphene oxide was prepared by the hydrothermal method with modified reaction parameters [19,20,21]. Briefly, 50 ml of 0.5 M glucose solution was stirred for 30 min. Later, the solution was transferred to 100 ml Teflon autoclave and heated in a hot air oven at 160 °C for 12 h. Then, the sample was allowed to cool to room temperature. After that, the sample was washed 3 times with double distilled water and dried in a hot air oven at 60 °C for 48 h. Finally, the as-prepared sample was stored in a desiccator.

2.3 Characterization

The X-ray Diffraction (XRD) patterns were recorded by Rigaku Miniflex600 X-Ray diffractometer with Cu Kα radiation (λ = 1.54 A°) at a scan rate of 2°/min in scan range 5°–45°. The X-ray Photoelectron Spectroscopy (XPS) analysis was performed on KRATOS Axis Ultra DLD system with a monochromatic Al Kα X-ray source (1486.6 eV). Fourier Transform Infrared (FTIR) spectra were collected using SHIMADZU IRPrestige-21 FTIR spectrometer in wavenumber range 4000–500 cm−1. UV–Vis absorption spectra were recorded using SHIMADZU 1800 UV–Vis spectrometer in the wavelength range of 190–800 nm. Raman spectroscopy measurements were taken using HORIBA LabRAM HR Raman microscope with a 532 nm laser. Scanning Electron Microscope (SEM) images were taken by Carl Zeiss Sigma field-emission scanning electron microscope (FE-SEM). Particle size and Zeta potential were measured by BROOKHAVEN Zeta PALS Zeta potential analyzer utilizing phase analysis of Dynamic Light Scattering (DLS).

3 Results and discussion

The crystalline structure of the as-prepared sample was studied by XRD (Fig. 1a) [28]. In Fig. 1a, broad diffraction peak at 2θ = 22.5° corresponds to the (002) reflection plane. The flake size D and interlayer spacing d002 of the sample calculated using Scherrer Formula and Bragg’s Law are 5.91 A° and 3.95 A°, respectively. The average number of layers per single rGO flake thus calculated using d002 and D is 2.5. The broad diffraction peak at 22.5° confirms the formation of reduced graphene oxide [20]. The broadening of the peak is due to the shorter stack height of the produced sample.

Fig. 1
figure 1

a XRD pattern and XPS, b survey spectrum, deconvoluted, c C1s and d O1s core level spectrum of carbonyl-rich partially reduced graphene oxide

Full size image

The elemental composition and chemical environment of the elements in the sample were determined by XPS [29]. Figure 1b confirms the presence of carbon and oxygen elements in the sample through the characteristic binding energy peaks at ~ 285 eV (C1s) and ~ 531 eV (O1s), respectively. In order to study the oxygen functionalization in the sample, the high-resolution XPS core-level spectra of C1s and O1s were deconvoluted and corresponding oxygen functional groups were assigned based on theoretical prediction of core-level shift [30].

Figure 1c shows the C1s core-level XPS spectra of the sample. A well-defined intense peak at binding energy 284 eV corresponds to the unoxidized sp2 carbon with C–C/C=C in the aromatic rings of rGO. The binding energies of deconvoluted peaks at 285.2 eV, 286.9 eV, and 287.8 eV are assigned to sp3 carbon of C–O–C/C–OH (epoxy/hydroxy), C=O (carbonyl) and HO–C=O (carboxylic) [16] oxygen functional groups, respectively. The chemical shifts of the C1s to higher binding energies are due to the attachment of more electronegative oxygen functional groups. The 291 eV component stems from the shake-up satellite peak (π–π*) of sp2 carbon bonds [31]. Figure 1d shows the O1s XPS spectra of rGO. The presence of C–O, C=O and O–C=O functional groups were again confirmed by the deconvoluted peaks centered at 533.5 eV, 531.2 eV and 530.2 eV [16], respectively.

The XPS result (Fig. 1c, d) clearly shows that the present sample contains high amount of carbonyl functional groups compared to hydroxyl, epoxy and carboxylic groups. The C/O ratio of the sample is 2.1, which suggests the formation of graphene oxide [21] (for rGO C/O > 7 [30]). Whereas, XRD result suggested the formation of rGO and we also didn’t observe any diffraction peak around 10°, which is generally observed in the case of oxygen-rich graphene derivatives [32]. The contrast in XPS and XRD results is because the majority of the oxygen in the sample is of carbonyl functional groups that exist at the plane edge. Glucose and the hydrothermal method itself selectively reduce hydroxyl and epoxy groups on the basal plane of intermediate graphene oxide leaving carbonyl groups unreduced [22, 23, 33].

The FTIR characterization also confirmed the presence of oxygen functional groups along with H2O and CO2 released during reduction. In the FTIR spectrum (Fig. 2a), strong characteristic peak at 1534 cm−1 due to C=C stretching skeletal vibrations of aromatic domains [32] and peak at 1680 cm−1 corresponding to C=O stretching vibrations in carbonyl groups [10] confirms the formation of carbonyl-rich partially reduced graphene oxide. Peaks at 1305 cm−1, 1207 cm−1, and 1028 cm−1 are due to C–O bending vibrations, epoxy C–O–C stretching vibrations and alkoxy C–O stretching vibrations [20], respectively. The weak broad peak from 3491 to 2920 cm−1 is due to O–H stretching vibrations in hydroxyl groups present on the basal plane and of intercalated free H2O molecules [20]. Residual CO2 adsorbed on to the sample results in a small peak at 2333 cm−1 [34]. 1403 cm−1 peak is attributed to O–H bending vibrations incurred by the angular deformation of water adsorbed on the sample [32]. FTIR result (Fig. 2a) also clearly shows the presence of a high amount of C=O functional groups through corresponding intense peak at 1680 cm−1 which is consistent with XPS.

Fig. 2
figure 2

a FTIR, b UV–Vis, c Raman spectrum and d SEM image of carbonyl-rich partially reduced graphene oxide

Full size image

The extent of formation and reduction of graphene oxide is evaluated using UV–Vis spectroscopy. Figure 2b shows the UV–Visible absorption spectrum of the sample. In the spectrum, absorbance peaks at 225 nm and 285 nm corresponds to π–π* transition of sp3 and sp2 hybridized carbon bonds, respectively [35]. sp3 carbon atoms confirm the presence of oxygen functional groups. This result is in agreement with XPS and FTIR results.

Raman spectroscopy characterizes the structure and disorder/defect in graphene-related materials. In a typical Raman spectrum of rGO, the position and relative intensity (ID/IG) of D and G band determine the quality of the sample as well as the efficiency of the reduction method [36]. D band (~ 1350 cm−1) measures the degree of disorder/defect in the sample and it arises from the breathing mode of K-point phonons of A1g symmetry in sp3 carbon atoms associated with oxygen functional groups. G (~ 1575 cm−1) is the characteristic band of graphene materials and it arises from the first-order scattering of the E2g phonons of sp2 carbon atoms of C–C/C=C bonds [37]. In Fig. 2c, the blueshift of D band (1374 cm−1) and G band (1593 cm−1) can be attributed to the induced structural defects by oxygen functional groups. The ID/IG of the sample is 1.39. The relatively high ID/IG ratio is due to the smaller average size of the sample resulted from high temperature and pressure during hydrothermal synthesis.

The morphology of the sample was investigated using FE-SEM. SEM image (Fig. 2d) shows a thick flake-like structure formation which may have resulted from the agglomeration of much thinner flakes. The agglomeration of the rGO might be due to edge to edge interactions between oxygen functional groups via hydrogen bonding [17]. We can also observe a small number of carbon spheres that contain hydrophobic core and hydrophilic shell [38]. The average flake diameter of the sample observed is ~ 200 nm.

Hydrodynamic mean-size and size distributions of sample dispersion were studied by Dynamic Light Scattering (DLS) technique (Fig. 3). The mean hydrodynamic thickness of the sample is 654.2 nm with the polydispersity index of 0.005. Further, the stability of the sample in a colloidal solution was determined by Zeta potential analyzer. A moderate Zeta potential of − 27.40 mV was observed for the sample, which may have resulted from the static repulsive force developed due to the presence of negatively charged oxygen functional groups on the surface and along the edges [17]. This suggests that the present sample can be used in biomedical and water treatment related applications.

Fig. 3
figure 3

Hydrodynamic size distributions and Zeta potential of carbonyl-rich partially reduced graphene oxide

Full size image

4 Reaction mechanism

Glucose, when hydrothermally treated transforms into carbonyl-rich graphene oxide through carbonization [39] and subsequently gets reduced into carbonyl-rich partially reduced graphene oxide (Fig. 4).

Fig. 4
figure 4

Mechanism of formation of carbonyl-rich partially reduced graphene oxide from glucose

Full size image

4.1 Carbonyl-rich graphene oxide formation

Carbonization takes place mainly via three steps (1) Dehydration and Fragmentation, (2) Polymerization/Condensation and (3) Aromatization.

4.1.1 Dehydration and fragmentation of glucose into soluble products

Under the hydrothermal condition, glucose converts into 5-Hydroxymethylfurfural (5-HMF) through dehydration reaction [40]. 5-HMF further converts to 1,2,4 benzenetriol [41], erythrose [42], 2,5-dioxo-6-hydroxy hexanol [43], furfural [44], and glyceraldehyde [42]. Then, these products partially decompose into acids, aldehydes, and phenols via. ring opening and C–C bond breaking [45].

4.1.2 Polymerization/condensation of the soluble products

5-HMF and its decomposition products undergo polymerization via intermolecular dehydration or aldol addition/condensation [24, 43, 46]. The acids produced in the previous step accelerates the polymerization process by making the reaction medium slightly acidic.

4.1.3 Aromatization of the polymers, nucleation, and growth

The polymers undergo aromatization via ring substitution and ring fusion reactions [44]. When the concentration of aromatic clusters in the aqueous solution reaches the critical supersaturation point, burst nucleation takes place [47]. The nuclei so formed grow outwards by diffusion and linkage [48] to form graphene oxide.

As observed from the FTIR and XPS spectra, the synthesized rGO contains a high amount of C=O functional groups along with an aromatic C=C carbon network. C=O functional group forms due to dehydration of equatorial hydroxyl groups. Whereas, C=C network results from keto-enol tautomerism of the dehydrated species or via intramolecular dehydration [49].

4.2 Reduction of carbonyl-rich graphene oxide

Unutilized glucose in the solution acts as reducing agent during graphene oxide reduction. Glucose reacts with hydroxyl and epoxy groups on the basal plane of graphene oxide and gets oxidized to gluconic acid [22, 23]. Moreover, hydrothermal reduction itself preferentially reduce carbon single-bonded oxygen functional groups i.e. O–H and C–O–C [33]. Therefore, the carbonyl functional groups (C=O) remain unreduced during reduction and hence results in production of carbonyl-rich partially reduced graphene oxide.

5 Conclusion

In the present study, we synthesized carbonyl-rich partially reduced graphene oxide using glucose by hydrothermal method. The as-prepared sample was characterized by XRD, XPS, FTIR, UV–Vis, Raman, SEM, and DLS techniques. The sample contains 2–3-layers and has an interlayer spacing of 3.95 A° and it contains a high amount of carbonyl functional groups at the plane edge. C/O and ID/IG ratio of the sample is 2.1 and 1.39, respectively. It has flake-like structure and hydrodynamic size 654.2 nm with a moderate zeta potential of − 27.40 mV. We also proposed a plausible reaction mechanism for the formation of carbonyl-rich partial reduced graphene oxide to provide insight into the structural evolution. The present work offers a simple, one-step, cost-effective and eco-friendly method for the mass production of carbonyl-rich partially reduced graphene oxide. The presence of peripheral carbonyl groups supports further functionalization and also improves the water dispersibility of the rGO samples which makes them useful candidates for biomedical and water treatment related applications.