Ethylenediamine-Functionalized Reduced Graphene Oxide for Enhanced Methylene Blue Removal

Abstract

Ethylenediamine-functionalized reduced graphene oxide (GO) was synthesized by reacting chlorine-functionalized reduced GO and ethylenediamine. Chlorine atoms in reduced GO are susceptible to substitution by ethylenediamine, thereby causing their elimination. The product exhibits improved efficiency in methylene blue (MB) removal. Adsorption kinetics of the reaction fit a pseudo-second order model. Adsorption isotherms of MB on the product could be well described by the Langmuir model; physisorption may dominate the adsorption process. Batch experiments show an adsorption capacity of 143.6 mg/g, which is superior to those of reduced GO (77.2 mg/g) and activated carbon (AC, 46.5 mg/g).

Introduction

Several efforts have been made toward the covalent and noncovalent functionalization of graphene oxide (GO) and reduced graphene oxide (RGO) (Guo et al., 2014; Criado et al., 2015). The covalent functionalized forms of GO and RGO are preferred because covalent functionalization adds a wide variety of functional groups on the graphene sheet and causes the most significant changes in its properties.

GO and RGO contain numerous functional groups that can react with various compounds. The epoxide groups and carboxyl groups of GO can react with amines, secondary amines, and other N-containing compounds (Ai et al., 2014; Yang et al., 2014; Xu, et al., 2016). During transformation of the hydroxyl and epoxide groups in GOs into other functional groups, these functional groups can also react with other compounds having C–Cl bond (Kumar et al., 2014; Li, et al., 2015; Mangadlao et al., 2015). The hydrogen atom of the hydroxyl group in GO can be substituted by other compounds (Wang et al., 2013, 2014a, 2014b). When growing a polymer with GO and RGO, the hydroxyl group is used as the anchor site (Zhao and Liu, 2014; Liu et al., 2015c; Qin et al., 2015; Rajender and Suresh, 2016). To form the derivatives of carboxyl and hydroxyl functional groups, isocyanates are applied to create the anchor site on GO (Gupta et al., 2006).

The C = C bond of GO is functionalized by cysteamine hydrochloride through a thiol–ene click reaction (Luong et al., 2015). Carbene can also react with the C = C of graphene (Sainsbury et al., 2016). Another approach for the functionalization of graphene is the addition of aryl radicals (Liu et al., 2015b; Fortgang et al., 2016; Voylov et al., 2016). GO and RGO can also be functionalized by cycloaddition (Ko et al., 2015). The siloxane coupling agent is also used to modify GO sheets (Madadrang et al., 2012; Liu et al., 2015a). The fluoro-functionalized graphene is a useful intermediate compound (Dubecký et al., 2015; Zhan et al., 2015; Xing et al., 2016). This compound can react with amine, alcohol, and sulfur nucleophiles (Whitener et al., 2015).

Likewise, chlorine-functionalized reduced graphene oxide (ClRGO) has been prepared (Wang et al., 2015). The preference for organic compounds, such as halogenoalkanes, of the C–Cl bond in ClRGO has yet to be determined. The carbon and chlorine bond in halogenoalkanes may react with nucleophiles, such as amines. To the best of our knowledge, such occurrences have not been reported. In this work, ethylenediamine-functionalized reduced graphene oxide (ERGO) was prepared.

Materials and Methods

Materials

Methylene blue (MB) was purchased from the Beijing Chemical Reagents Company (Beijing, China). Natural graphite, with a mean particle size of 44 μm, was purchased from the Qingdao Zhongtian Company (Qingdao, China). Other chemicals were supplied by the Guangfu Industry of Fine Chemicals Institute (Tianjin, China). All chemicals were used without further purification. GO used in this study was synthesized from natural graphite powder based on the Hummers method (William et al., 1958). ClRGO was synthesized according to literature (Wang et al., 2015).

Synthesis of ERGO and RGO

Ethylenediamine (1.0 mL, 0.015 mol) and Na₂CO₃ (1.59 g, 0.015 mol) were added to a mixture of ClRGO (0.5 g) and acetonitrile (30 mL). The mixture was stirred at room temperature for 1 h. The mixture was heated under reflux for 8 h and poured into a container with water (50 mL). The product was collected by filtration and washed with deionized water. The black product was dried at 50°C for 1 day to produce ERGO (0.37 g).

The RGO was prepared according to the literature (Wang et al., 2014b).

Characterization of ERGO

Scanning electron microscopy (SEM) and energy-dispersive X-ray (EDX) spectroscopy were performed with a FEI-Quanta 200 scanning electron microscope. High-resolution scanning electron microscopy (HRSEM) was performed on a JEOL JEM-2011 electron microscope operating at 200 kV with a Gatan 794 camera. Fourier-transform infrared spectrometry (FTIR) were obtained with an FTS-40 (Bio-Rad, CA). Raman spectra were recorded on a Renishaw InVia multi-channel confocal microspectrometer with a 532 nm excitation laser. X-ray diffraction (XRD) measurements were obtained on an X'pert PRO diffractometer with Co Kα radiation. X-ray photoelectron spectroscopy (XPS) with monochromatized Al Kα X-ray (hν = 1486.6 eV) radiation (ESCALAB 250; ThermoFisher Scientific Co.) was used to investigate the surface properties of ERGO. The shift in binding energies was corrected, with the C1s signal at 284.6 eV as internal standard.

MB adsorption

Batch adsorption experiments were performed in 250 mL glass bottles with a mixture of the adsorbent (∼2 mg) and MB solution (200 mL, 4 mg/L to 12 mg/L). The glass bottles were sealed with Teflon. The experiments were performed at 180 rpm for 12, 24, and 48 h, and other time periods. The temperature of the solution was set to the desired value, with a variation of ±0.02°C, by adjusting the flow rate of the thermostatically controlled water running through an external glass-cooling spiral. The mixture was sampled in a series of intervals, and the concentration of the remaining MB in the mixture was measured after centrifugation. The absorbance was obtained with a UV/vis spectrophotometer. The concentration was calculated by the standard spectrophotometric method at a maximum absorbance of MB. All the samples were centrifuged at 10,000 rpm for 10 min.

Results and Discussion

Structural features of ERGO

Morphologies and microstructures of ERGO were characterized by SEM and HRTEM, as shown in Fig. 1. The graphene sheets are wrinkled and twisted, which might be a characteristic feature of the few-layered sheets in Fig. 1a–c. The corresponding EDX spectrum revealed the presence of C, O, N, and Cl, thereby implying that some chlorine atoms in ClRGO were possibly substituted by ethylenediamine.

FIG. 1.

Images and EDX spectrum of ERGO. (a, b) Scanning electron microscopy images and EDX spectrum of ERGO; (c) HRSEM image of ERGO. EDX, energy-dispersive X-ray; HRSEM, high-resolution scanning electron microscopy.

The FTIR spectra of RGO, ClRGO, and ERGO are shown in Fig. 2a. The characteristic peaks of RGO are present at 3,410, 1,560, and 1,040 cm⁻¹. The peaks of ClRGO can be observed at 1,713, 1,580, 1,092, and 653 cm⁻¹. ERGO displays an evident peak at 3,390 cm⁻¹, which might have originated from the stretching vibrations of −NH₂,–NH–, and the residual −OH groups. The skeletal vibration of unoxidized graphite is represented by the peak at 1,581 cm⁻¹. The peak of C–H is at 1,452 cm⁻¹; this peak was caused by the–CH₂–group of ethylenediamine. The peak of C–N is located at 1,221 cm⁻¹, which also implies the substitution of C–Cl and ethylenediamine.

FIG. 2.

FTIR spectra (a) and Raman spectra (b) of RGO, ClRGO, and ERGO. ClRGO, chlorine-functionalized reduced graphene oxide; ERGO, ethylenediamine-functionalized reduced graphene oxide; FTIR, Fourier-transform infrared spectrometry; RGO, reduced graphene oxide.

Raman spectra of RGO, ClRGO, and ERGO are presented in Fig. 2b. The G band of RGO at ∼1,565 cm⁻¹ is associated with the vibration of sp² carbon atoms. The G band of ClRGO is at 1,587 cm⁻¹. ERGO shows a G band at 1,598 cm⁻¹, which was caused by the formation of a conjugated system. This result implies that more graphene domains were formed in ERGO. Some chlorine atoms were possibly eliminated in the process, that is, ClRGO was continuously reduced by ethylenediamine. The D band of RGO at ∼1,348 cm⁻¹ is related to the vibration of defective and disordered sp³ carbon atoms. The D band of ClRGO is present at 1,346 cm⁻¹. The intensity of the D band at 1,354 cm⁻¹ of ERGO substantially increased, thereby indicating the increased size of sp³ domains, which was possibly caused by extensive substitution. The chlorine atoms were probably substituted by ethylenediamine, and the intensity of the D band might have been further increased. Ethylenediamine may have caused defects and disorderliness in ERGO. More sp³ carbon atoms can be observed in ERGO than in RGO because of the substitution reaction.

Figure 3a shows the XRD patterns of the RGO, ClRGO, and ERGO. The peaks of RGO and ClRGO are at 27.6° and 24.8°, and their average interlayer spacings are 0.36 and 0.41 nm, respectively. ERGO has a strong peak at 26.9° (0.38 nm) and a weak peak at 12.6° (0.72 nm). The average interlayer spacing of ERGO is slightly wider than those of graphite (0.334 nm) and RGO (0.36 nm). The N-containing groups, along with some residual chlorine and oxygenated functional groups, are attached on both sides of the ERGO sheets. These groups acted as spacers and/or as pillaring agents; the sheet might be thicker than graphene. The broad peak implies poor ordering of the sheets along the stacking direction, thereby indicating that ERGO might be composed of a few layered sheets.

FIG. 3.

(a) X-ray diffraction patterns of RGO, ClRGO, and ERGO. (b) XPS survey spectra of ERGO. (c, d) C1s (c) and N1s (d) XPS spectra of ERGO. XPS, X-ray photoelectron spectroscopy.

Composites of ERGO were also analyzed by XPS. The XPS survey spectra of ERGO show the presence of C, O, N, and Cl elements, as shown in Fig. 3b and Table 1. The carbon and oxygen content of RGO, ClRGO, and EGRO are similar. ERGO has a higher nitrogen content than ClRGO, which implies the substitution of C–Cl and ethylenediamine. Theoretically, the nitrogen content should be twice as much as the chlorine content of ClRGO. As observed, the nitrogen content of ERGO is lower than the chlorine content of ClRGO. A similar result was reported in literature (Whitener et al., 2015). Some chlorine atoms in ClRGO have a structure similar to that of primary halogenoalkanes; these atoms may be substituted by ethylenediamine. The epoxide group in RGO can also react with ethylenediamine. Other chlorine atoms in ClRGO have structures similar to those of tertiary halogenoalkanes; most of the atoms may be eliminated, and some parts of the ClRGO sheet may be reduced, thereby producing conjugated system form. Meanwhile, a third set of chlorine atoms in ClRGO have structures similar to those of secondary halogenoalkanes; a fraction of these chlorine atoms may be substituted by ethylenediamine, whereas other atoms could be eliminated (Whitener et al., 2015).

Table 1.

Atomic Percentage of Elements of Reduced Graphene Oxide, Chlorine-Functionalized Reduced Graphene Oxide, and Ethylenediamine-Functionalized Reduced Graphene Oxide

Element (atomic %)	C1s	O1s	N1s	Cl 2p
RGO	87.53	12.47	—	—
ClRGO	80.67	13.34	—	5.22
ERGO	84.79	10.72	4.24	0.25

ClRGO, chlorine-functionalized reduced graphene oxide; ERGO, ethylenediamine-functionalized reduced graphene oxide; RGO, reduced graphene oxide.

The core-level XPS signals of C1s are shown in Fig. 3c. The peak observed at 282.9 eV is assigned to the–CH₂–group of ethylenediamine. The peak at 284.6 eV was caused by graphitic sp² carbon atoms. The peak at 287.6 eV originates from C–N and C–O. The peak at 294.3 eV is associated with C = O in carboxylic acids. The peak at 399.6 is assigned to N1s in Fig. 3d, which implies that N-containing groups were attached to the graphene sheet.

MB adsorption and its adsorption mechanism

Effect of pH

The pH of a mixture is an important parameter that controls the ion adsorption process. To investigate the effect of pH, a series of experiments were performed at 293 K for 24 h using 200 mL of MB, with an initial concentration of 8 mg/L and 2 mg of adsorbent. The protonated form MBH²⁺ dominates when pH = 1. The monomer MB⁺ is dominant in the pH range between 2 and 10. At high pH conditions, the dimer (MB⁺)₂ and trimer (MB⁺)₃ are dominant (Saad et al., 2015). The pH range between 2 and 10 was used in this study, and the results are shown in Fig. 4a. The values of q_e increased as the increasing initial pH increased from 2 to 6. The optimum MB adsorption was observed at pH 7–10. The adsorption capacity increased with the increasing pH, which can be attributed to the electrostatic attraction of positively charged MB with the negatively charged surface of ERGO. With the decreasing pH, the concentration of H⁺ increased, whereas the ionization of –COOH and −OH also decreased, that is, the number of negatively charged functional groups decreased. A coordinate bond can also form between H⁺ and the nitrogen atom of –NH–and–NH₂. MB competed with the hydrogen ions among the adsorption sites. Therefore, the adsorption capacity decreased. Furthermore, the pH of the ERGO solution without adjustment by HCl or NaOH was approximately eight in this study. Therefore, pH = 8 was selected in the succeeding experiments.

FIG. 4.

Effect of pH (a) and ionic strength (b) on the adsorption of MB by ERGO. MB, methylene blue.

Effect of ionic strength

Dyehouse waste commonly includes dyes and electrolytes; thus, the effect of electrolytes should be tested to estimate the actual adsorption capacity of ERGO in the removal of MB. To investigate the salt effect on the adsorption of MB, NaCl was used to adjust the solution salinity (Jia et al., 2015). A series of experiments were performed at 293 K for 24 h using 200 mL of MB with an initial concentration of 12 mg/L and 2 mg of adsorbent. The salt can greatly improve the ionic strength, which might have influenced the adsorption of MB by the adsorbent. As shown in Fig. 4b, the ability to absorb MB slightly increased as the NaCl concentration increased from 0 to 0.1 mol/L. This result indicates that MB is relatively more competitive than metal ions and can be adsorbed in the presence of high salt concentrations.

Comparison of absorption capacity

Adsorption capacities of ERGO, RGO, and commercial AC were compared by MB adsorption. Figure 5a presents the adsorption curves of MB by ERGO, RGO, and AC. The adsorption capacity of ERGO for MB is 143.6 mg/g, which is much larger than that of RGO (77.2 mg/g) and approximately thrice as much as that of AC (46.5 mg/g). ERGO contains several functional groups, such as −OH, −COOH, −NH, and −NH₂. The hydroxyl group of phenols and–COOH can form negatively charged groups; these groups may absorb MB. A coordinate bond could also form between cationic dyes and nitrogen atom of −NH and −NH₂.

FIG. 5.

(a) Adsorption of MB by ERGO, RGO, and AC. Conditions: 8 mg/L concentration and room temperature. (b) Adsorption kinetics of MB by ERGO for pseudo-second order reaction. Conditions: 8 mg/L concentration. (c) Langmuir plot (C_e/q_e vs. C_e) for MB adsorption by ERGO. (d) Dubinin–Radushkevich plot (ln(q_e) vs. Epsilon²) for MB adsorption by ERGO. (Epsilon = ɛ). AC, activated carbon.

Adsorption kinetics

Adsorption kinetics to describe MB uptake rate is one of the important characteristics that control the adsorbate uptake rate at the solid–liquid interface. To clarify the adsorption behavior of ERGO, the kinetics of MB removal was observed. The available kinetic models include pseudo-first order models, pseudo-second order models, and intraparticle diffusion models (Liu et al., 2012; Chao et al., 2013; Zhai et al., 2013).

The correlation coefficient (R², which is almost 1) was used to evaluate the suitability of the different models. A more applicable model for the kinetics of MB adsorption is the use of high-R² indicators. The related parameters of the different models are shown in Table 2. The pseudo-first order models have high correlation coefficients (0.949, 0.983, 0.974, 0.928, and 0.991). However, significant differences between the calculated q_cal and experimental q_exp values were observed. For MB adsorption with high R² values (0.999, 0.997, 0.997, 0.999, and 0.997, as shown in Fig. 5b), the pseudo-second order model proved to have good correlation for ERGO; the differences between the calculated q_cal and experimental q_exp values were small. The correlations for the intraparticle diffusion models are low (0.845, 0.929, 0.919, 0.854, and 0.972), thereby implying that MB adsorption on ERGO does not fit the model.

Table 2.

Parameters Obtained from Different Kinetic Models of Methylene Blue

		Parameters
Models and equations	Temperature (K)	K₁ ( × 10⁻² min⁻¹)	q_cal (mg/g)	q_exp (mg/g)	R²
pseudo-first order ln (q_e – q_t) = ln (q_e) – k₁t	293	4.62	39.4	81.3	0.949
	303	4.49	46.1	89.2	0.983
	313	4.40	57.4	102.6	0.974
	323	3.74	82.0	143.1	0.928
	333	3.23	131.3	189.1	0.991

		K₂ ( × 10⁻⁴ g/[mg·min])	q_cal (mg/g)	q_exp (mg/g)
pseudo-second order t/q_t = 1/k₂q_e² + t/q_e	293	4.11	80.0	81.3	0.999
	303	4.04	88.5	89.2	0.997
	313	3.18	102.0	102.6	0.997
	323	2.80	142.9	143.1	0.999
	333	2.71	188.7	189.1	0.997

		k_dif (mg/[g·min^1/2])	C (mg/g)
Intraparticle diffusion q_t = k_dif t^1/2 + C	293	6.19	33.0	0.845
	303	6.62	36.6	0.929
	313	9.34	34.8	0.919
	323	12.6	43.1	0.854
	333	17.2	41.0	0.972

Concentration = 8 mg/L.

Adsorption isotherms

Freundlich, Langmuir, and Dubinin–Radushkevich (D–R) isotherm models are widely used to simulate and understand the adsorption mechanism (Liu et al., 2012; Chao et al., 2013; Zhai et al., 2013). These models were also adopted in the present study.

Constants (b, q_m) for the Langmuir isotherm are obtained from the slope and intercept of the plot between C_e/q_e and C_e (Fig. 5c). The Langmuir model parameters and the statistical fits of the adsorption data to this equation are compiled in Table 3. The regression coefficient is high; the adsorption isotherms of MB on the product are well described by the Langmuir model. The adsorption process occurs in the monolayer at the ERGO surface. The parameters (R_L) are within the favorable limit (between 0 and 1), thereby implying a favorable process. The q_m and Langmuir constant (b) increased with temperature. Thus, these data clarify the endothermic nature of the existing process.

Table 3.

Constants, Parameters, and Correlation Coefficients Calculated for Different Adsorption Isotherm Models at Different Temperatures for Methylene Blue Adsorption

Models and equations	Constants		293K	303K	313K	323K	333K
Langmuir	q_m (mg/g)		434.8	454.5	476.2	500.1	526.3
	b (L/mg)		0.767	0.957	1.31	2.00	3.17
C_e/q_e = 1/q_mb + C_e/q_mR_L = 1/(1 + bC_o)	R_L	C₀ = 4 mg/L	0.246	0.207	0.160	0.111	0.0731
		C₀ = 6 mg/L	0.179	0.148	0.113	0.0769	0.0499
		C₀ = 8 mg/L	0.140	0.116	0.0871	0.0588	0.0379
		C₀ = 10 mg/L	0.115	0.0947	0.0709	0.0476	0.0306
		C₀ = 12 mg/L	0.0980	0.0801	0.0598	0.0400	0.0256
	R ²		0.997	0.995	0.996	0.997	0.997
Freundlich	K_F (mg/g [L/mg]^1/n)		188.4	219.7	262.6	312.3	391.0
ln (q_e) = ln (K_F) + (1/n)ln (C_e)	n ⁻¹		0.428	0.416	0.395	0.364	0.324
	R ²		0.957	0.933	0.923	0.906	0.854
Dubinin-Radushkevich	q_s ( × 10⁻³ mol/g)		0.914	1.01	1.07	1.16	1.22
ln (q_e) = ln (q_s) – Bɛ²ɛ = RTln (1 + 1/C_e)E = (2B)^−1/2	B ( × 10⁻² mol²/KJ²)		0.198	0.155	0.104	0.0694	0.0416
	E (kJ/mol)		1.59	1.80	2.20	2.68	3.47
	R ²		0.984	0.995	0.995	0.996	0.998

The Freundlich model is an empirical model that allows multilayer adsorption with nonuniform distribution of the heat of sorption on the surface. The values of K_F and 1/n were calculated from the intercept and slope of the plot of ln q_e versus ln C_e; these values are given in Table 3, with the regression coefficients at the studied temperatures. The parameters of K_F and n are related to the adsorption capacity and adsorption intensity, respectively. The value of 1/n is lower than 1, thereby implying a normal Langmuir isotherm.

To distinguish between the physical and chemical adsorption of MB, the D–R model was also applied (Vasudevan and Lakshmi, 2012). The values of the constants (q_s, B) calculated from the intercept and slope of the plot of ln q_e versus ɛ², with the regression coefficients, are shown in Table 3 and Fig. 5d. The constant B indicates the mean sorption energy (E). The value of E is lower than 8.0 kJ/mol for all the tested temperatures, thereby indicating that physisorption probably dominated the adsorption process (Luo et al., 2010, 2016; Vasudevan and Lakshmi, 2012; He et al., 2016). At the initial adsorption stage, partial chemical adsorption may also occur because of the strong chemical interactions between the amine and negatively charged groups of ERGO and MB. By increasing the MB loadings, the number of active adsorption sites greatly decreased. The van der Waals and hydrophobic interaction might be important factors that affect adsorption. At this stage, ERGO tends to physically absorb MB.

Conclusions

Some chlorine atoms in RGO can be substituted by ethylenediamine, whereas other atoms may be eliminated. ERGO could be an important intermediate for graphene-based materials and other compounds. The product exhibits improved efficiency in MB removal. Moreover, the adsorption process fits the pseudo-second order rate model. Adsorption isotherms of MB on the product are well described by the Langmuir model, which implies the occurrence of monolayer adsorption. The physisorption process possibly dominated the adsorption of MB.

Footnotes

Acknowledgments

This work was supported by the Science and Technology Department of Henan Province (No. 162300410015 and 162102110028), the Education Department of Henan Province (No. 12B210022), the Science and Technology Bureau of Xinxiang (Nos. 13SF39 and ZG15022), and the Xinxiang University Fund (No. 15ZP05).

Author Disclosure Statement

No competing financial interests exist.

References

, Cao

X.H.

, Sun

Z.P.

, Jiang

, Du

Z.Z.

, Xie

L.H.

, Wang

Y.L.

, Wang

X.J.

, Zhang

, Huang

, and Yu

(2014). Redox-crosslinked graphene networks with enhanced electrochemical capacitance. J. Mater. Chem. A, 2, 12924.

Chao

, Liu

J.D.

, Wang

J.T.

, Zhang

Y.W.

, Zhang

Y.T.

, Xiang

, and Chen

R.F.

(2013). Surface modification of halloysite nanotubes with dopamine for enzyme immobilization. ACS Appl. Mater. Interfaces, 5, 10559.

Criado

, Melchionna

, Marchesan

, and Prato

(2015). The covalent functionalization of graphene on substrates. Angew. Chem. Int. Ed., 54, 10734.

Dubecký

, Otyepková

, Lazar

, Karlický

, Petr

, Čépe

, Banáš

, Zbořil

, and Otyepka

(2015). Reactivity of fluorographene: A facile way toward graphene derivatives. J. Phys. Chem. Lett., 6, 1430.

Fortgang

, Tite

, Barnier

, Zehani

, Maddi

, Lagarde

, Loir

, Jaffrezic-Renault

, Donnet

, Garrelie

, and Chaix

(2016). Robust electrografting on self-organized 3D graphene electrodes. ACS Appl. Mater. Interfaces, 8, 1424.

Guo

, Creighton

, Chen

Y.T.

, Hurt

, and Külaots

(2014). Porous structures in stacked, crumpled and pillared graphene-based 3D materials. Carbon, 66, 476.

Gupta

, Chen

, Joshi

, Tadigadapa

and Eklund

P.C.

(2006). Raman scattering from high-frequency phonons in supported n-graphene layer films. Nano Lett. 6, 2667.

J.S.

, Dai

J.D.

, Zhang

, Sun

, Xie

A.T.

, Tian

S.J.

, Yan

Y.S.

, and Huo

P.W.

(2016). Preparation of highly porous carbon from sustainable alpha-cellulose for superior removal performance of tetracycline and sulfamethazine from water. RSC Adv. 6, 28023.

Jia

Y.Y.

, Jin

, Li

, Sun

Y.X.

, Huo

J.Z.

, and Zhao

X.J.

(2015). Investigation of the adsorption behaviour of different types of dyes on MIL-100(Fe) and their removal from natural water. Anal. Methods-UK, 7, 1463.

10.

, Kim

, Lim

M.Y.

, Nam

S.Y.

, Kim

T.H.

, Kim

S.K.

, and Lee

J.C.

(2015). Sulfonated poly (arylene ether sulfone) composite membranes having poly(2,5-benzimidazole)-grafted graphene oxide for fuel cell applications. J. Mater. Chem. A, 3, 20595.

11.

Kumar

, Sain

, and Jain

S.L.

(2014). Photocatalytic reduction of carbon dioxide to methanol using a ruthenium trinuclear polyazine complex immobilized on graphene oxide under visible light irradiation. J. Mater. Chem. A, 2, 11246.

12.

, Alain-Rizzo

, Galmiche

, Audebert

, Miomandre

, Louarn

, Bozlar

, Pope

M.A.

, Dabbs

D.M.

, and Aksa

I.A.

(2015). Functionalization of graphene oxide by tetrazine derivatives: A versatile approach toward covalent bridges between graphene sheets. Chem. Mater., 27, 4298.

13.

Liu

H.M.

, Guo

, Wang

X.S.

, Li

Y.K.

, Liang

X.J.

, and Liu

(2015a). Double carboxyl silicane modified graphene oxide coated silica composite as sorbent for solid-phase extraction of quarternary alkaloids. Anal. Methods, 7, 135.

14.

Liu

, Wan

Y.Z.

, Xie

Y.D.

, Zhai

, Zhang

, and Liu

J.D.

(2012). The removal of dye from aqueous solution using alginate-halloysite nanotube beads. Chem. Eng. J., 187, 210.

15.

Liu

X.B.

, Zheng

Y.Y.

, and Wang

X.L.

(2015b). Controllable preparation of polyaniline–graphene nanocomposites using functionalized graphene for supercapacitor electrodes. Chem. Eur. J., 21, 10408.

16.

Liu

Z.Z.

, Zhu

S.J.

, Li

Y.J.

, Li

Y.S.

, Shi

, Huang

, and Huang

X.Y.

(2015c). Preparation of graphene/poly (2-hydroxyethyl acrylate) nanohybrid materials via an ambient temperature ‘grafting-from’ strategy. Poly. Chem., 6, 311.

17.

Luo

, Zhao

Y.F.

, Zhang

, Liu

J.D.

, Yang

, and Liu

J.F.

(2010). Study on the adsorption of neutral red from aqueous solution onto halloysite nanotubes. Water Res. 44, 1489.

18.

Luo

S.H.

, Chen

S.Y.

, Chen

S.X.

, Ma

N.F.

, and Wu

Q.H.

(2016). Sisal fiber-based solid amine adsorbent and its kinetic adsorption behaviors for CO₂. RSC Adv. 6, 72022.

19.

Luong

N.D.

, Sinh

L.H.

, Johansson

, Campell

, and Seppälä

(2015). Functional graphene by thiol-ene click chemistry. Chem. Eur. J., 21, 3183.

20.

Madadrang

C.J.

, Kim

H.Y.

, Gao

G.H.

, Wang

, Zhu

, Feng

, Gorring

, Kasner

M.L.

, and Hou

S.F.

(2012). Adsorption behavior of EDTA-graphene oxide for Pb (II) removal. ACS Appl. Mater. Interfaces, 4, 1186.

21.

Mangadlao

J.D.

, Leon

A.C.C.D.

, Felipe

M.J.L.

, Cao

P.F.

, Advincula

P.A.

, and Advincula

R.C.

(2015). Grafted carbazole-assisted electrodeposition of graphene oxide. ACS Appl. Mater. Interfaces, 7, 10266.

22.

Qin

, Yuan

, Li

, Chen

D.C.

, Kong

, Chu

F.Q.

, Tao

Y.G.

, and Liu

M.L.

(2015). Crosslinking graphene oxide into robust 3D porous N-doped graphene. Adv. Mater., 27, 5171.

23.

Rajender

, and Suresh

K.I.

(2016). Surface-initiated atom transfer radical polymerization (SI-ATRP) from graphene oxide: Effect of functionalized graphene sheet (FGS) on the synthesis and material properties of PMMA nanocomposites. Macromol. Mater. Eng., 301, 81.

24.

Saad

, Al-Mawla

, Moubarak

, Al-Ghoul

, and El-Rassy

(2015). Surface-functionalized silica aerogels and alcogels for methylene blue adsorption. RSC Adv. 5, 6111.

25.

Sainsbury

, Passarelli

, Naftaly

, Gnaniah

, Spencer

S.J.

, and Pollard

A.J.

(2016). Covalent carbene functionalization of graphene: Toward chemical band-gap manipulation. ACS Appl. Mater. Interfaces, 8, 4870.

26.

Vasudevan

, and Lakshmi

(2012). The adsorption of phosphate by graphene from aqueous solution. RSC Adv. 2, 5234.

27.

Voylov

, Saito

, Lokitz

, Uhrig

, Wang

Y.Y.

, Agapov

, Holt

, Bocharova

, Kisliuk

, and Sokolov

A.P.

(2016). Graphene oxide as a radical initiator: Free radical and controlled radical polymerization of sodium 4-vinylbenzenesulfonate with graphene oxide. ACS Macro. Lett., 5, 199.

28.

Wang

C.B.

, Ni

, and Zhou

J.W.

(2014a). Enhanced pore size of graphene by modification for water purification. Mater. Technol., 29, 252.

29.

Wang

C.B.

, Ni

, Zhou

J.W.

, Wen

J.L.

, and Lü

X.B.

(2013). Strategically designed porous polysilicate acid/graphene composites with wide pore size for methylene blue removal. RSC Adv. 3, 23139.

30.

Wang

C.B.

, Zhou

J.W.

, and Chu

L.L.

(2015). Chlorine-functionalized reduced graphene oxide for methylene blue removal. RSC Adv. 5, 52466.

31.

Wang

C.B.

, Zhou

J.W.

, Ni

, Cheng

Y.L.

, and Li

(2014b). Design and synthesis of pyrophosphate acid/graphene composites with wide stacked pores for methylene blue removal. Chem. Eng. J., 253, 130.

32.

Whitener

K.E.

Jr. , Stine

, Robinson

J.T.

, and Sheehan

P.E.

(2015). Graphene as electrophile: Reactions of graphene fluoride. J. Phys. Chem. C, 119, 10507.

33.

William

, Hummers

J.R.

, and Richard

E.O.

(1958). Preparation of graphitic oxide. J. Am. Chem. Soc., 80, 1339.

34.

Xing

R.G.

, Li

Y.N.

, and Yu

H.T.

(2016). Preparation of fluoro-functionalized graphene oxide via the Hunsdiecker reaction. Chem. Comm., 52, 390.

35.

L.G.

, Xiang

, Liu

, Xu

, Luo

Y.C.

, Feng

L.Z.

, Liu

, and Peng

(2016). Functionalized graphene oxide serves as a novel vaccine nano-adjuvant for robust stimulation of cellular immunity. Nanoscale, 8, 3785.

36.

Yang

, Wu

J.X.

, Lü

Q.F.

, and Lin

T.T.

(2014). Facile preparation of lignosulfonate–graphene oxide–polyaniline ternary nanocomposite as an effective adsorbent for Pb (II) ions. ACS Sustain. Chem. Eng., 2, 1203.

37.

Zhai

, Zhang

, Wan

Y.Z.

, Li

C.C.

, Wang

J.T.

, and Liu

J.D.

(2013). Chitosan-halloysite hybrid-nanotubes: Horseradish peroxidase immobilization and applications in phenol removal. Chem. Eng. J., 214, 304.

38.

Zhan

H.F.

, Zhang

G.Y.

, Bell

J.M.

, and Gu

Y.T.

(2015). Graphene with patterned fluorination: Morphology modulation and implications. J. Phys. Chem. C, 119, 27562.

39.

Zhao

X.B.

, and Liu

(2014). Hydrophobic-polymer-grafted graphene oxide nanosheets as an easily separable adsorbent for the removal of tetrabromobisphenol A. Langmuir, 30, 13699.