NRGO was synthesized through a simple two-step reaction, including the doping of nitrogen and the reduction of graphene oxide (GO). The morphologies of NRGO and GO were observed by Scanning Electron Microscopy (SEM). The results showed that the microstructure of NRGO (Figure 1B) exhibited a loose, multifold, three-dimensional stacked layer with wrinkles and pores while that of GO (Figure 1A) was a relatively flat layer with some wrinkles, which is consistent with the previous study (Gordon-Nuez et al., 2019). Meanwhile, the Energy Disperse X-ray spectroscopy (EDX) spectrum ((Figure 1C and elemental mapping images (Figures 1D–F) confirmed that nitrogen was clearly observed and uniformly distributed in the synthesized NRGO, distinguishing it from GO, which is composed solely of carbon and oxygen. This observation confirms that NRGO was successfully synthesized by through nitrogen doping.

Surface charge and Raman were used to further verify the reduction and functionalization of GO. As shown in Figure 2A, GO showed the negative potential (−38.25 mV), while NRGO showed the positive potential (25.07 mV), which possibly indicated graphene oxide was aminated. Meanwhile, the change of carbon frame structure was characterized by Raman spectrum. As shown in Figure 2B, the D and G band were clearly observed for GO and NRGO at 1,349 cm⁻¹, 1587 cm⁻¹, and 1349 cm⁻¹, 1601 cm⁻¹, respectively. The G band of NRGO was slightly split and had a slight peak shift, which indicated GO was reduced and functionalized (Song et al., 2012). Also, the I _D /I _G of NRGO was higher than that of GO, which indicated the amino groups were modified to the edge or surface of the graphene.

Moreover, X-ray powder diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR) were used to characterize the changes in structure and functional groups before and after the modification of GO. As shown in Figure 3A, the XRD patterns exhibited an intense characteristic diffraction peak at 2θ = 10.10° for GO and a weak characteristic peak at 2θ = 19.95° for NRGO, attributed to the reduction of oxygen-containing functional groups of GO, leading to a significant decrease in interlayer spacing (Zheng et al., 2013; He et al., 2013). Furthermore, FTIR was used to characterize the changes in functional groups of GO and NRGO. As shown in Figure 3B, the FTIR spectrum of GO displayed a broad and intense peak at 3,420 cm⁻¹, corresponding to -OH stretching vibration peak. Additionally, the other four adsorption peaks of GO are displayed at 1720 cm⁻¹, 1,630 cm⁻¹, 1,380 cm⁻¹ and1050 cm⁻¹, indicative of the stretching vibration absorption peak of C=O in aryl carboxylic acids, the stretching vibration absorption peak of the C=C form six-membered ring skeleton, the deformation vibration absorption peak of O-H in C-OH group, and the stretching vibration absorption peak of C-O, respectively (Liu et al., 2014; Yang et al., 2020).In contrast, the characteristic absorption peaks of oxygen-containing functional groups in NRGO at 3,420 cm⁻¹, 1720 cm⁻¹, 1,630 cm⁻¹, and 1,380 cm⁻¹ O were significantly weakened or absent as expected, confirming the reduction of GO. Instead, a new adsorption peak representing a carbon-nitrogen heterocyclic ring appears at 1,070 cm⁻¹ (Yuan et al., 2017), and 1,560 cm⁻¹ for bending vibration peak of N-H, indicating the successful introduction of N in the graphene lattice. These results suggested that the oxygen-containing groups in GO were decreased, and the N were doped during the synthesis process.

Additionally, X-ray photoelectron spectroscopy (XPS) was utilized to verify the elemental composition, and the valence bond form of NRGO. As shown in Figure 4A, the N element was clearly observed after N-doping on GO, whereas GO only contained O and C, indicating successfully doped of N element. Additionally, the C:O ratio in NRGO increased significantly compared to GO, which indicated GO was also reduced in obtaining N-doped reduced graphene oxide. Also, C1s (Figure 4B), N1s (Figure 4C), and O1s (Figure 4D) spectra were studied to analyze the possible valence bond. As shown in C 1s spectrum (Figure 4B), the four characteristic peaks were observed, including C-C (284.6 eV), C-N/C-O (285.7 eV), C=O (286.8 eV), and π-π^* (288.9 eV). The higher peak area of C-C indicated the reduction of GO. The appearance of C-N indicates that the amino group in oleoamine may be linked to GO. In Figure 4C, the split peaks of N 1s were observed as C-N (401.3 eV) and N-H (399.5 eV), which indicated the GO was aminated. These results indicated GO was reduced and functionalized by oleylamine through the formation of C-N and N-H, which was also verified by the FTIR results.

The possible adsorption mechanism of NRGO for MB can be roughly delineated into three points: 1) The microstructure of NRGO. SEM image in Figure 1B showed that NRGO had more folds and pores, and the three-dimensional lamellar structure of NRGO was conducive to improving the adsorption efficiency of MB. 2) The π-π interaction between NRGO and MB. According to the previous reports, a large number of benzene rings on NRGO have been proved to be effective adsorption of phenols through π-π bonds (Zhao et al., 2020), potentially facilitating interactions with the aromatic groups of MB molecules. 3) The formation of new chemical bonds. The kinetic study results indicated that the adsorption of MB by NRGO was more in accordance with pseudo-second-order kinetics, suggesting that the chemisorption predominantly governs the adsorption rate. Therefore, some necessary characterizations were conducted to further elucidate the adsorption mechanism.

The intuitive observation of microscopic morphologies could directly reflect the adsorption of MB. Thus, SEM was employed to observe the morphologies change of before and after MB adsorption. As shown in Figure 8A, MB presented a stacked particles form. After the adsorption of MB (Figure 8B), NRGO-MB still kept a stacked lamellar structure, similar to that of NRGO. However, MB was not observed clearly on the surface of NRGO-MB because of the encapsulation of MB by the lamellar structure of NRGO. To further confirm the adsorption of MB by NRGO, EDX analysis was conducted. As shown in Figure 8D, the result showed the presence of carbon, oxygen, nitrogen, chlorine, and sulfur in NRGO-MB, which was consistent with the elements present in MB (Figure 8C). Additionally, the mapping images showed uniformly dispersion of chlorine and sulfur elements on the surface of NRGO-MB (Figures 8E, F). Furthermore, as depicted in the inset image of (Figure 8D), the blue coloration associated with MB appeared lighter or even disappeared after the adsorption by NRGO, indicating successfully adsorption of MB by NRGO.

In addition, XPS was utilized to further analyze the elements before and after MB adsorption. As shown in Figure 9A, the NRGO surface was predominantly composed of carbon, oxygen and nitrogen. Apart from carbon, oxygen and nitrogen, the MB surface also contained chlorine and sulfur. Compared with NRGO, sulfur was observed on the surface of NRGO-MB, which aligned with the elemental analysis results of MB, suggesting the successful adsorption of MB onto NRGO. The XRD patterns, as shown in Figure 9B, revealed a shift in the characteristic peak from 2θ = 19.98° for NRGO to 2θ = 19.87° for NRGO-MB, indicating an increased interlayer spacing due to the adsorption of MB on NRGO surface. Moreover, three other characteristic peaks at 2θ = 7.45°, 2θ = 12.33° and 2θ = 29.11° were clearly observed in NRGO-MB, which closely matched with XRD pattern of MB, which indicated that MB was successfully loaded onto NRGO to form NRGO-MB. Furthermore, FTIR was used to characterize the functional groups before and after MB adsorption. As shown in Figure 9C, no significant changes or new peaks were observed in the infrared spectra of NRGO after adsorbing MB, which indicated that the adsorption process was attributed to physisorption through π-π interactions between NRGO and MB.

Taking into account the adsorption kinetic, isotherms, and aforementioned characterizations, the adsorption mechanism was attributed to physisorption attracted by the π-π bonds between NRGO and MB.

Graphene oxide (GO) was prepared in our previous study. The chemicals used in the experiments, including N, N-Dimethylformamide (DMF), 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC), cis-9-Octadecenylamine, trichloromethane, petroleum, ethanol, sodium hydroxide (NaOH), hydrochloric acid (HCl) and methylene blue (MB), were all purchased from Chengdu Kelong Chemical Reagent Company (China), and used without further purification. Deionized water (18.25 MΩ cm) was obtained using ULUPURE Water Purification System (Chengdu, China) and was used to prepare all aqueous solutions.

NRGO was prepared according to the previous reports. In brief, 0.30 g of GO was weighed and dispersed in 50.0 mL DMF under sonication for 30 min and stirring for 1 h. Then, 0.46 g of EDC was dissolved in 20.0 mL DMF by sonication. The EDC solution was added to GO dispersion, stirred for 2 h and sonicated for 30 min. Subsequently, 0.27 g of NHS was dissolved to 10.0 mL DMF under sonication and then was added to the above reaction system, stirred for 2 h and sonicated for 30 min to ensure complete reaction. Next, 1.28 g of olamine was weighed and added to 10.0 mL of chloroform. After the olamine completely dissolved, the solution was added to the above reaction system. Then the mixture was treated by sonicated at 40°C for 30 min and stirred overnight. The obtained material was filtered through organic membrane (0.22 μm), followed by washing with petroleum ether and ethanol. Finally, the modified graphene oxide was obtained by freeze-drying and marked as N doped reduced graphene oxide (NRGO).

Reduced graphene oxide was prepared as follows:the 0.5 mg/mL GO solution was transferred to polytetrafluoroethylene autoclave at 120°C for 8 h, and the reduced material was obtained as RGO.

The surface morphologies of GO, NRGO, MB, and the composite after MB adsorption were examined by environmental scanning electron microscope (ESEM, JSM-IT500, JEOL Company, Japan). Materials compositions were analyzed by energy disperse X-ray spectroscopy (EDX, JEOL Company, Japan) and X-ray photoelectron spectroscopy (XPS, Kratos, Britain). Changes in the crystal structure of the materials were measured by X-ray powder diffraction (XRD, X’Pert Pro, Philips, Netherlands). The alterations of functional groups of the material were determined using a Fourier transform infrared spectrometer (FTIR, Nicolet IS10, Thermo, America) with KBr pellets in a scanning range of 500–4000 cm⁻¹. The change of frame structure was measured by Raman spectrometer (Jobin Yvon S.A.S., HORIBA). The surface charge was detected by zeta potential analyzer (Zetasizer-ZS series, England).

Batch adsorption tests were performed to assess the impact of various factors on the adsorption property of MB onto NRGO. The prepared NRGO was grinded into powder and the solid MB powder was dissolved in deionized water to prepare a methylene blue solution for subsequent experiments. For the adsorption experiments, different doses of NRGO with 300 µL of 0.3 mg/mL of MB solution were firstly optimized to ensure the adsorption efficiency as high as possible. Subsequently, a certain amount of NRGO was introduced into MB solution of varying initial concentrations (to study the effect of concentration) at various pH values (for pH experiment) for different contact time (forvadsorption kinetic experiment). After adsorption at room temperature (25°C), the absorbance of their supernatant was measured by U-2910 UV–Vis spectrophotometer. The final concentration of MB after adsorption was calculated to determine the adsorption rate (E,%) and adsorption capacity (q, mg/g) according to experimental Equations 5, 6 as follows: E = C 0 − C / C 0 × 100 % (5) q = C 0 − C × V / M (6) Where E is the adsorption efficiency of MB (%), q is the adsorption capacity of MB (mg/g), C ₀ and C are the initial and final concentration of MB solution (µg/mL), respectively, V is the volume of MB solution (L), M is the mass of the adsorbent (g).