Removal of dyes by a novel fly ash–chitosan–graphene oxide composite adsorbent

Abstract

A novel graphene oxide-based adsorbent (FCGO) is synthesized from fly ash cross-linked with chitosan and graphene oxide and characterized through scanning electron microscopy, X-ray diffraction and Fourier transform infrared analyses. The adsorbent effectively removes anionic and cationic dyes, namely, acid red GR (ARG) and cationic red X-5GN (CRX), respectively. The removal ratio of ARG by FCGO decreases with increasing initial pH in acidic solutions but is not affected at pH higher than 6. By contrast, initial pH minimally influences the removal ratio of CRX. After the adsorption of ARG and CRX by FCGO, the final pH becomes close to neutral in acidic medium because of the protonation effect. The adsorption kinetics of ARG and CRX follow the pseudo-second-order kinetic model. Adsorption is mainly controlled by boundary diffusion and internal diffusion for ARG but only by boundary diffusion for CRX, and the process satisfactorily fits with the Redlich–Peterson model. The maximum adsorption capacities are 38.87 and 64.50 mg g⁻¹ for ARG and CRX, respectively. Negative ΔG⁰ values (ARG: −2.909 kJ mol⁻¹ to −4.592 kJ mol⁻¹, CRX: −2.526 kJ mol⁻¹ to −4.738 kJ mol⁻¹) and positive ΔH⁰ values (ARG: 22.299 kJ mol⁻¹ and CRX: 15.669 kJ mol⁻¹) indicate that adsorption is a spontaneous and endothermic process.

---

1. Introduction

Graphene oxide (GO), an important graphene derivative, is a highly-oxidized planar material with abundant oxygen-containing functional groups, such as hydroxyl, carboxyl, carbonyl, and epoxy,¹ and presents a high theoretical surface area of 2630 m² g⁻¹.² GO is widely studied and used as an adsorbent with a strong capacity to remove dyes, metal ions, and toxic refractory organic compounds.³ Sun et al.⁴ reported the removal of acridine orange by GO with an adsorption capacity of 1428 mg g⁻¹. However, GO is difficult to collect and separate from treated water because the hydrophilic oxygenous functional groups of the compound easily disperse in water, which can lead to serious secondary pollution. As such, GO is combined with organic materials, such as chitosan, and calcium alginate, as well as inorganic particulate matter, including magnetic Fe₃O₄, CoFe₂O₄, sand and ceramics.³

Fe₃O₄/GO composite is synthesized through self-assembly of aminated Fe₃O₄ nanoparticles on GO, and the fabricated material effectively removes methylene blue and neutral red and exhibits good magnetic separation property.⁵ Core–shell adsorbent GO/sand granules are synthesized through assembly of aqueous GO over sand particles and collected from the treated water through gravity sedimentation.⁶ The combination of GO with chitosan can be used to form spherical, flake, fibrous, or spongy-shape adsorbents.^7,8 GO foam is formed by direct freeze drying and demonstrates a high adsorption capacity for rhodamine B (72.5 mg g⁻¹) and methylene blue (184 mg g⁻¹).⁸ GO–chitosan is also synthesized and applied for the removal of reactive black 5,⁹ Au(III), and Pd(II).¹⁰ A novel magnetic GO–chitosan nanocomposite is also prepared and used to remove methylene blue,¹¹ lead ions,¹² chromium,¹³ hydroquinone,¹⁴ and other pollutants.

Positively charged GO can more easily absorb cationic dyes or ions compared with anionic dyes or ions. In this study, a new fly ash–chitosan–graphene oxide (FCGO) composite adsorbent is prepared by wrapping GO on fly ash with chitosan as combiner and glutaric aldehyde as cross-linking agent. The synthesized FCGO is then used to remove anionic and cationic dyes.

2. Materials and methods

2.1. Materials

Flake graphite recycled from desulfurization slag and fly ash was obtained from Maanshan Iron and Steel Co., Ltd, China. After passing through 0.08 mm sieve, the amounts of fly ash and flake graphite residue were 12.3% and 27.8%, respectively. Biological reagents, namely, chitosan (deacetylation degree of 80.0% to 95.0%) and glutaraldehyde (25% aqueous solution), were purchased from Sinopharm Chemical Reagent Co., Ltd. The other reagents used were of analytical grade. Cationic red X-5GN (CRX) and acid red GR (ARG) were purchased from Shanghai Chemical Industry of Jia Ying Co., Ltd, China and used directly without any purification.

2.2. Characterization

The surface morphology of the adsorbent was determined using a scanning electron microscope (SEM, Nano SEM430; FEI, America). X-ray diffraction (XRD) analysis was performed on a D8 ADVANCE diffractometer (Bruker, German) with Cu Kα radiation. Step scan was conducted over the 2θ range of 5–75°, with a stepping interval of 0.02° and a count time of 0.24 s (scanning rate of 5.00 deg min⁻¹, 40 kV, 40 mA). The diffraction peaks on the XRD spectra were detected using the software package MDI Jade 6.0 with automatic and manual peak-seeking. The functional groups on the adsorbent surface were assessed using Fourier transform infrared (FT-IR) spectrometer (TENSOR 27; Bruker, German). The adsorbent sample was ground with KBr to fine powder and then compressed into thin pellets for the FT-IR analysis. Scanning was performed within the range of 400 cm⁻¹ to 4000 cm⁻¹ at a resolution of 4 cm⁻¹ with 64 scans.

2.3. Preparation of GO

GO was synthesized according to Hummer's method.¹⁵ Briefly, 2.0 g of flake graphite was mixed with 1.0 g of NaNO₃ and 50 mL of H₂SO₄ in a beaker and stirred for 30 min in an ice bath. The mixture was slowly added with 10.0 g of KMnO₄ and then vigorously stirred. The reaction temperature was maintained below 10 °C. Subsequently, the mixture was stirred for 2 h at 35 ± 3 °C to oxidize graphite and added with 100 mL of deionized water to form a brownish paste. The temperature was increased to approximately 98 °C until effervescence was observed. The diluted mixture was stirred for another 15 min and added with 200 mL of deionized water and 50 mL of 30% H₂O₂. The color of the mixture turned yellow. The mixture was consequently washed with 150 mL of 3% HCl to remove MnO₂ and sulfate and then rinsed with ultrapure water until sulfate was completely removed. The mixture was finally separated using high-speed centrifuge, and the solid product was dried at 40 °C for 48 h under vacuum.

2.4. Preparation of FCGO

The preparation process of FCGO is shown in Scheme 1. Fly ash particles were wrapped by chitosan, whose –NH₂ groups form hydrogen bonds, electrostatic interactions, and/or nucleophilic substitutions with the oxygen-containing groups of GO.⁹ A chitosan solution (2.5%, w/v) was prepared by dissolving 2.5 g of chitosan powder into 100 mL of 2% (v/v) acetic acid solution under ultrasonic stirring for 1 h at room temperature. Briefly, 0.5 g of GO was dissolved into 100 mL of ultrapure water under ultrasound for 6 h to prepare 5 g L⁻¹ GO solution. Subsequently, 2.0 g of fly ash was added to 2 mL of 2.5% (w/v) chitosan solution, impregnated for 2 h, and then dried at 60 °C. Subsequently, 1.0 g of fly ash cross-linked with chitosan and 15 mL of GO solution were added into a three-necked flask. The mixture was stirred continuously for 3 h and stored overnight at room temperature. The mixture was then added with 10 mL of ultrapure water and stirred for another 15 min. Afterward, the mixture was added with 1 mL of 25% (v/v) glutaraldehyde. The flask was kept in a water bath for 3 h at 50 °C. The black product was sequentially washed with ethanol and distilled water until the pH became neutral. The product was dried at 60 °C for 3 h.

Scheme 1 Preparation process of FCGO adsorbent.

2.5. Adsorption experiment

Anionic (ARG) and cationic dyes (CRX) were used for batch adsorption by the FCGO adsorbent. The effect of the initial pH of the dye solution was investigated from pH from 3.0 to 10.0. The pH of the suspension was adjusted by adding HCl or NaOH solution (0.01 mol L⁻¹), and pH was monitored using a pH meter (μ-pH system 362, Systronics) calibrated with standard buffer solutions. Briefly, 25 mL of CRX and ARG solutions (initial concentration of 50.0 mg L⁻¹) and 0.08 g of the FCGO adsorbent were added in a 100 mL glass bottle. The bottle was placed in a thermostatic water bath at 25 °C and then agitated for 1 and 6 h for CRX and ARG solutions, respectively, at an agitation speed of 150 rpm. After the reaction, the suspension was centrifuged at 3500 rpm for 10 min. The centrifugal liquid was detected at a wavelength of 512 nm for CRX and 506 nm for ARG by using a spectrophotometer (722-type, Shanghai Analytical Instruments General Factory, China). The amounts of CRX and ARG adsorbed onto the FCGO were determined using the following equation:

Q_e = (C₀ − C_e) × V/m (1)

where Q_e (mg g⁻¹) is the adsorption capacity of the adsorbent; C₀ (mg L⁻¹) and C_e (mg L⁻¹) are the initial and final concentrations of CRX and ARG, respectively; V (L) is the solution volume; and m (g) is the adsorbent mass.

The adsorption kinetics of CRX and ARG onto FCGO was determined at pH 5 under an agitation rate of 150 rpm at 25 °C to 45 °C.

3. Results and discussion

3.1. Characterization of the absorbent

The SEM images of fly ash and FCGO are shown in Fig. 1. The fly ash particle is spherical and smooth (Fig. 1a), whereas the FCGO surface (Fig. 1b) is rough, uneven, and wrinkled. The composite surface is wrapped by a thin layer, namely, chitosan, and a sheet-like material, which is GO. Hence, GO is assembled on the surface of fly ash particles by chitosan.

Fig. 1 SEM photograph of (a) fly ash and (b) FCGO.

Fig. 2 shows that the main mineral components of fly ash and FCGO are quartz and mullite. No characteristic diffraction peaks of GO are detected in FCGO because of the low amounts of GO added on fly ash.

Fig. 2 XRD patterns of FCGO and fly ash.

The FT-IR spectrum of the FCGO adsorbent differs from that of fly ash (Fig. 3). A new band appears at 3367 cm⁻¹, which is due to the N–H and O–H groups from the adsorbed chitosan and GO. The bands at 1735 cm⁻¹ and 1628 cm⁻¹ are attributed to the stretching vibration of the C=O (carboxyl group) of GO and the C=C group stretching model of the sp² carbon skeletal network in GO, respectively.^4,9 The appearance of new bands confirms that chitosan and GO are successfully fixed on the surface of fly ash particles.

Fig. 3 FT-IR spectra of (a) FCGO, (b) fly ash, (c) chitosan, and (d) GO.

3.2. Effect of pH

Fig. 4 shows that the removal ratio of ARG decreases with increasing initial solution pH. The adsorption capacity minimally changes and is maintained at approximately 10 mg g⁻¹ at pH higher than 6. The adsorption capacity of CRX (approximately 11 mg g⁻¹) is not affected by initial pH, ranging from 3 to 9. After adsorption, the final pH values of ARG and CRX solutions are almost neutral.

Fig. 4 Effects of initial pH on the adsorption of ARG and CRX by FCGO (pH = 3–9; 50 mg L⁻¹ dye concentration; 4 g L⁻¹ adsorbent addition; T = 25 °C; 150 rpm; 1 h adsorption).

The structural formulas of ARG and CRX are abbreviated as ARG-SO₃Na and CRX=NCl, respectively. The hydrolysis processes of the two ionic dyes are as follows:^11,16

ARG-SO₃Na ↔ ARG-SO₃⁻ + Na⁺ (2)

CRX=NCl ↔ CRX=N⁺ + Cl⁻ (3)

In acidic solutions, the amino, hydroxyl, and amide groups of FCGO are readily protonated and positively charged, as described in eqn (4)–(6).^9,11

R_FCGO–NH₂ + H⁺ ↔ R_FCGO–NH₃⁺ (4)

R_FCGO–OH + H⁺ ↔ R_FCGO–OH₂⁺ (5)

R_FCGO–CONH₂ + H⁺ ↔ R_FCGO–CONH₃⁺ (6)

where R_FCGO is the alkyl group originated from FCGO. The protonation of these groups consume H⁺ in the solution, thereby increasing the pH.

In basic solutions, the protonation of amino, hydroxyl, and amide groups reduces. Coexisting ions in the solution also change with increasing amounts of OH⁻. Subsequently, H⁺ and NH₃⁺, which are ionized from the FCGO composite, combine with OH⁻, thereby decreasing the pH (eqn (7)–(9)).

R_FCGO–COOH ↔ R_FCGO–COO⁻ + H⁺ (7)

R_FCGO–OH ↔ R_FCGO–O⁻ + H⁺ (8)

R_FCGO–CONH₂ ↔ R_FCGO–COO⁻ + NH₃⁺ (9)

As a result, the final pH is close to neutral because H⁺ and OH⁻ are consumed in acidic and alkaline solutions, respectively.

In acidic solutions, protonated functional groups, such as R_FCGO–NH₃⁺ and R_FCGO–CONH₃⁺ (eqn (4)–(6)), easily combine with ARG-SO₃⁻ (eqn (2)) because of electrostatic attraction force. This combination benefits the adsorption of ARG. In neutral and alkaline solutions, the electrostatic force becomes weak. The π–π interaction existing between the dye aromatics and the GO benzene ring on the FCGO surface is the main force for ARG adsorption. However, the π–π interaction force is not affected by pH.⁹ As a result, ARG adsorption is almost unchanged at initial pH values higher than 6.0.

The adsorption capacity of CRX does not change with increasing initial pH of the solution because the π–π interaction is the main force between FCGO and CRX.⁹

However, in neutral and alkaline solutions, the adsorption of CRX is larger than ARG, because the weak electrostatic attraction between FCGO and CRX is promoting π–π interaction and the weak electrostatic repulsion between FCGO and ARG is inhibiting π–π interaction.¹⁷

3.3. Adsorption kinetics

As shown in Fig. 5, the adsorption capacity of ARG increases rapidly within 60 min and then increases slowly, reached the adsorption equilibrium after 4 h. CRX adsorption rate increases rapidly within 30 min and then slowly increases. The adsorption rate of CRX is faster for the 25 mg L⁻¹ CRX solution and then reaches the adsorption equilibrium after 60 min. However, longer time is needed to achieve the adsorption equilibrium for CRX with high initial concentrations. During FCGO preparation, the fly ash particle is wrapped by chitosan before loading GO onto chitosan. Therefore, the functional groups of GO are found on the FCGO surface but the amino groups of chitosan are possibly overlaid by GO. The negative charge of the functional groups of GO accelerates the adsorption of the cationic CRX dye because of electrostatic attraction and inhibits the adsorption of the anionic ARG dye because of electrostatic repulsion.^3,11 As a result, the adsorption rate of CRX is faster than that of ARG. The anionic dye remains adsorbed by the amino functional group of chitosan through internal diffusion and therefore requires long adsorption duration.

Fig. 5 Kinetic fitting for the adsorption of (a) ARG and (b) CRX (pH: no change; 4 g L⁻¹ adsorbent addition; T = 25 °C; 150 rpm).

Three kinetic models, namely, pseudo-first-order,¹⁸ pseudo-second-order,¹⁹ and Elovich,²⁰ are selected to fit the experimental data by using the non-linear fitting method with the software Origin 8.5. The fitting results are shown in Table 1. Based on the value of the statistical error chi-square (χ²) and the linear regression coefficient (R²), the adsorption of ARG and CRX better fits the pseudo-second-order equation than the pseudo-first-order and Elovich equations. The calculated equilibrium adsorption capacities (Q_{e, cal}) of ARG and CRX based on the pseudo-second-order model are in good agreement with the experimental equilibrium adsorption capacities (Q_e). Table 1 show that the reaction rate constant (k₂) of CRX is higher than that of ARG, which confirms the faster adsorption rate of CRX.

Table 1 Kinetic constants for the adsorption of ARG and CRX by FCGO^a

Kinetic model	Parameter	Acid red GR			Cationic red X-5GN
		25 mg L^−1	50 mg L^−1	100 mg L^−1	25 mg L^−1	50 mg L^−1	100 mg L^−1
	Qe	6.118	10.470	15.433	5.808	10.659	19.661
a Note: Qe (mg g^−1), the experimental saturation adsorption capacity; Qe, cal (mg g^−1), the theoretical equilibrium adsorption capacity; k1 and k2 are the rate constants for the pseudo-first order model and pseudo-second order model, respectively; α, β are constants of Elovich equation, reflecting the initial adsorption rate (g mg^−1 min^−1) and the desorption constant (g mg^−1).
Pseudo-first order model Qt = Qe(1 − e^−k1t)	k1	5.210	3.879	3.500	30.677	25.999	20.353
	Qe, cal	5.958	9.966	14.267	5.659	10.362	18.773
	R^2	0.769	0.814	0.751	0.553	0.683	0.638
Pseudo-second order model	k2	1.673	0.618	0.357	18.496	7.128	2.284
	Qe, cal	6.235	10.612	15.348	5.794	10.680	19.672
	R^2	0.970	0.979	0.945	0.953	0.974	0.966
Elovich equation	α	29427.184	2035.794	1164.848	4.10 × 10^13	3.23 × 10^10	8.11 × 10^6
	β	2.013	0.855	0.524	5.763	2.379	0.817
	R^2	0.897	0.957	0.989	0.908	0.862	0.933

---

The intra-particle diffusion model is used for fitting the experimental data to further investigate the limiting factor in the adsorption of the ionic dye by FCGO.²¹ The adsorption of solid particles is generally controlled by three steps: boundary diffusion, internal diffusion, and surface adsorption.

The intra-particle diffusion model is shown as follows:

Q_t = k_pt^0.5 + C (10)

where k_p (mg g⁻¹ min^−0.5) and C are the intra-particle diffusion rate constant and the boundary layer thickness, respectively. The kinetic data of ARG fitted by the intra-particle diffusion model is divided into three linear parts (Fig. 6). This finding indicates that the adsorption process is complex and internal diffusion is not the only rate-limiting step. In the first stage, the adsorption capacity of ARG adsorbed by FCGO increases with increasing t^0.5, and this phenomenon corresponds to boundary diffusion.²² The adsorption capacity continues to increase with increasing t^0.5 in the second stage, but the adsorption rate is lower than that of the first stage because adsorption is limited by internal diffusion.²³ Q_t does not increase in the third stage, which indicates that adsorption reaches the equilibrium stage. The adsorption of CRX is divided into two phases: boundary diffusion and surface adsorption, which are rapidly completed in the first stage and the adsorption process gradually reaches the equilibrium in the second stage.²⁴ These findings show that the adsorption rate of ARG adsorbed by FCGO is first controlled by boundary diffusion, followed by internal diffusion, whereas the adsorption of CRX is mainly limited by boundary diffusion. The adsorption of ARG at the internal diffusion stage is mainly hindered by GO.

Fig. 6 Fitting of the intra-particle diffusion model (pH: no change; 4 g L⁻¹ adsorbent addition; T = 25 °C; 150 rpm).

3.4. Adsorption isotherm

Fig. 7 shows the nonlinear fitting of the experimental data with Langmuir, Freundlich, and Redlich–Peterson (R–P) models by using the software Origin 8.5.^25–27 Table 2 presents the resulting maximum adsorption capacity (Q_{e, cal}) and isothermal parameters. According to the regression coefficient (R²) and the statistical error chi-square (χ²) values,^13,28 the adsorption process best fits the R–P isotherm model (ARG: 0.963 ≤ R² ≤ 0.993 and 1.025 ≤ χ² ≤ 12.510; CRX: 0.989 ≤ R² ≤ 0.995 and 2.266 ≤ χ² ≤ 3.139). The adsorption process also fits with the Freundlich model (ARG: 0.937 ≤ R² ≤ 0.971 and 3.362 ≤ χ² ≤ 15.828; CRX: 0.973 ≤ R² ≤ 0.996 and 1.889 ≤ χ² ≤ 7.297) within the range of 25 °C to 60 °C. The Langmuir model cannot fit with the experimental data at low temperatures (ARG: 25 °C, 35 °C; CRX: 25 °C, 45 °C) but fits well at high temperatures (ARG: 45 °C; CRX: 60 °C).

Fig. 7 Fitting of the adsorption isotherm model for the adsorption of (a) ARG and (b) CRX (pH: no change; 4 g L⁻¹ adsorbent addition; T = 25 °C; 150 rpm).

Table 2 Equilibrium parameters for the adsorption of ARG and CRX by FCGO^a

Isotherm equation	Parameter	Acid red GR			Cationic red X-5GN
		25 °C	35 °C	45 °C	25 °C	45 °C	60 °C
	Qe	26.880	33.191	38.866	45.768	47.355	64.504
a Note: Qe (mg g^−1), the experimental saturation adsorption capacity; Qe, cal (mg g^−1), the theoretical equilibrium adsorption capacity, KL (L mg^−1) is the Langmuir adsorption equilibrium constant, Ce (mg L^−1) is the equilibrium concentration in the solution; KF (mg^1−1/n L^1/n g^−1) is the Freundlich constant, n (dimensionless) is the constant; A and B are Redlich–Peterson isotherm constants, g is a empirical constant between 0 and 1.
Langmuir equation	Qe, cal	28.721	29.947	35.024	54.326	56.473	68.063
	KL	0.079	0.621	0.912	0.032	0.030	0.087
	R^2	0.979	0.924	0.894	0.989	0.964	0.930
	χ^2	2.509	13.581	26.804	2.908	10.227	37.059
Freundlich equation Qe = KFCe^n−1	KF	5.909	10.590	14.186	6.429	6.816	15.261
	n	3.126	4.028	4.246	2.522	2.559	3.128
	R^2	0.971	0.964	0.937	0.973	0.989	0.996
	χ^2	3.362	6.515	15.828	7.297	3.068	1.889
Redlich–Peterson (R–P) equation	A	5.537	47.265	91.501	2.155	352.066	5449.765
	B	0.493	3.167	4.663	0.069	51.160	355.987
	g	0.811	0.831	0.850	0.897	0.611	0.681
	R^2	0.993	0.988	0.963	0.990	0.989	0.995
	χ^2	1.025	2.929	12.510	3.139	3.680	2.266

---

The g value of the R–P model is not close to 1.0 (0.811–0.850 for ARG and 0.611–0.897 for CRX), which indicates that the data do not fit with the Langmuir model. This finding is confirmed by the calculated regression coefficient. Thus, the adsorption sites on the FCGO surface are homogeneous; meanwhile, ARG and CRX are not removed through mono-layer adsorption.¹³ The A and B values of the R–P model for ARG and CRX increase with increasing reaction temperature; as such, high temperature is favourable for adsorption. The A and B values of CRX are higher than those of ARG, which indicates that FCGO adsorbs CRX through electrostatic attraction force.²⁹ The adsorption capacity of CRX and ARG also increase with increasing adsorption temperature, thereby confirming the endothermic nature of the adsorption process.

The comparison of the adsorption capacity between FCGO and other reported works are listed in Table 3. The adsorption capacity of only GO for CRX was much higher than the FCGO, because the loading of GO on FCGO was lower than 7.5% (FCGO prepared by 1 g fly ash with 0.075 g GO). As a word, FCGO composite prepared easily by fly ash with little GO can remove anionic and cationic dyes with a considerable adsorption capacity.

Table 3 The maximum adsorption capacities of ARG and CRX by different adsorbent

Acid red GR			Cationic red X-5GN
Adsorbent	Qe (mg L^−1)	Reference	Adsorbent	Qe (mg L^−1)	Reference
Chitosan	21.26	This study	Chitosan	2.73	This study
GO	40.69	This study	GO	694.40	This study
FCGO	38.87	This study	FCGO	64.50	This study
Organo-bentonite	91.48	33	Activated carbon	263.16	37
TA-TiO2	32.99	34	Cation-exchange membrane	93.46	38
Sludge-based activated carbon	14.95	35	Micro–mesoporous molecular sieve	19.56	39
Activated fly ash	9.60	36	Zeolite	5.44	40

---

3.5. Adsorption thermodynamics

Thermodynamic parameters, namely, Gibbs free energy (ΔG⁰), enthalpy (ΔH⁰), and entropy (ΔS⁰), are used to evaluate the feasibility and nature of adsorption, which are determined by the following equations.³⁰

ΔG⁰ = ΔH⁰ − TΔS⁰ (11)

where R (8.314 J (mol K)⁻¹) is the ideal gas constant, T (K) is the absolute temperature, and K_c is the thermodynamic equilibrium constant and determined using the method of Khan and Singh by plotting ln(Q_e/C_e) versus Q_e.³¹ The values of ΔH⁰ and ΔS⁰ are calculated from the slope and intercept of the Van't Hoff linear plots of ln K_c versus T⁻¹ by using eqn (12). The parameters obtained are shown in Table 4.

Table 4 Thermodynamic parameters for the adsorption of ARG and CRX by FCGO

Dye	T (K)	ΔG^0 (kJ mol^−1)	ΔH^0 (kJ mol^−1)	ΔS^0 (J mol^−1 K^−1)
ARG	298	−2.909	22.299	85.019
	308	−4.160
	318	−4.592
CRX	298	−2.526	15.669	60.605
	318	−3.242
	333	−4.738

---

The ΔG⁰ values of CRX and ARG are negative and decrease with increasing adsorption temperature, which indicates the feasibility and spontaneity of the adsorption of dye on FCGO. The positive values of ΔH⁰ indicate the endothermic nature of the adsorption, which is in agreement with the conclusions presented in Section 3.4. These positive values are caused by desorption of water molecules adsorbed onto the dye molecule and the adsorption of dye molecules onto the adsorbent surface.⁹ The positive value of ΔS⁰ reflects the increase in the number of species and randomness at the solid–liquid interface. This behavior could be due to the displacement of the coordinated water molecules by dye molecules, resulting in higher translational entropy gained than that lost by dye molecules (this process occurs during adsorption).³² The ΔS⁰ value of ARG is higher than that of CRX, which implies that the adsorption of ARG by FCGO is more complicated because it involves boundary diffusion and internal diffusion (explained in Section 3.3).

4. Conclusions

(1) The removal ratio of ARG by FCGO decreases with increasing initial pH in acidic solutions but is not affected in basic solutions. Initial solution pH also exerts minimal effect on the removal ratio of CRX. After ARG and CRX are adsorbed by FCGO, the final pH becomes close to neutral in acidic medium because of the protonation effect.

(2) The adsorption kinetics of ARG and CRX follow the pseudo-second-order kinetic model. Adsorption is mainly controlled by boundary diffusion and internal diffusion for ARG and by boundary diffusion only for CRX.

(3) The removal efficiencies of ARG and CRX are temperature dependent and thus increase with increasing temperature. The adsorption capacities (Q_e) of ARG is 38.87 mg g⁻¹ at 45 °C and the adsorption capacities of CRX is 64.50 mg g⁻¹ at 60 °C. The results of equilibrium isotherm investigation indicate that the adsorption of ARG and CRX is in agreement with the Redlich–Peterson and Freundlich models.

(4) The thermodynamic parameters of ARG and CRX are obtained. The value of ΔG⁰, ΔH⁰, and ΔS⁰ for ARG is −2.909 kJ mol⁻¹ to −4.592 kJ mol⁻¹, 22.299 kJ mol⁻¹, and 85.019 J mol K⁻¹, respectively. The value of ΔG⁰, ΔH⁰, and ΔS⁰ for CRX is −2.526 kJ mol⁻¹ to −4.738 kJ mol⁻¹, 15.669 kJ mol⁻¹, and 60.605 J mol K⁻¹, respectively. These findings indicate that the adsorption process is spontaneous and endothermic.

Acknowledgements

The authors are grateful for financial support from the Research Fund of Shaanxi Key Laboratory of Comprehensive Utilization of Tailings Resources (2014SKY-WK007) and the graduate innovation fund of Anhui University of Technology (2014099).

References

1. S. Virendra, J. Daeha, L. Zhai, D. Soumen, K. Saifull and S. Sudipta, Prog. Mater. Sci., 2011, 56, 1178–1271.
2. M. D. Stoller, S. Park, Y. Zhu, J. An and R. S. Ruoff, Nano Lett., 2008, 8, 3498–3502.
3. S. Chowdhury and R. Balasubramanian, Adv. Colloid Interface Sci., 2014, 204, 35–56.
4. L. Sun, H. Yu and B. Fugetsu, J. Hazard. Mater., 2012, 203, 101–110.
5. G. Q. Xie, P. X. Xi, H. Y. Liu, F. J. Chen, L. Huang, Y. J. Shi, F. P. Hou, Z. Z. Zeng, C. W. Shao and J. Wang, J. Mater. Chem., 2012, 22, 1033–1039.
6. W. Gao, M. Majumder, L. B. Alemany, T. N. Narayanan, M. A. Ibarra, B. K. Pradhan and P. M. Ajayan, ACS Appl. Mater. Interfaces, 2011, 3, 1821–1826.
7. V. Chabot, D. Higgins, A. Yu, X. C. Xiao, Z. W. Chen and J. J. Zhang, Energy Environ. Sci., 2014, 7, 1564–1596.
8. Y. He, J. Hazard. Mater., 2013, 260, 796–805.
9. N. A. Travlou, G. Z. Kyzas, N. K. Lazardis and E. A. Deliyanni, Chem. Eng. J., 2013, 217, 256–265.
10. L. Liu, C. Li, C. L. Bao, Q. Jia, P. F. Xiao, X. T. Liu and Q. P. Zhang, Talanta, 2012, 93, 350–357.
11. L. L. Fan, C. N. Luo, X. J. Li, F. G. Lu, H. M. Qiu and M. Sun, J. Hazard. Mater., 2012, 215, 272–279.
12. L. L. Fan, C. N. Luo, M. Sun, X. J. Li and H. M. Qiu, Colloids Surf., B, 2013, 103, 523–529.
13. S. Debnath, A. Maity and K. Pillay, J. Environ. Chem. Eng., 2014, 2, 963–973.
14. L. L. Li, L. L. Fan, M. Sun, H. M. Qiu, X. J. Li, H. M. Duan and C. N. Luo, Int. J. Biol. Macromol., 2013, 58, 169–175.
15. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80(6), 1339.
16. L. L. Fan, C. N. Luo, M. Sun, X. J. Li, F. G. Lu and H. M. Qiu, Bioresour. Technol., 2012, 114, 703–706.
17. Y. Z. Wang and A. X. Wu, Chin. J. Org. Chem., 2008, 28, 997–1011.
18. L. S. Zur, K. Sven. Vetenskapsakad. Handl., 1989, 24, 1–39.
19. W. J. Weber Jr and J. C. Morris, J. Sanit. Eng. Div., Am. Soc. Civ. Eng., 1963, 89, 31–59.
20. S. H. Chien and W. R. Clayton, Soil Sci. Soc. Am. J., 1980, 44, 256–268.
21. Y. S. Ho and G. McKay, Chem. Eng. J., 1998, 70, 115–124.
22. Y. Ma, S. G. Wang, M. Fan, W. S. Gong and B. Y. Gao, J. Hazard. Mater., 2009, 168, 1140–1146.
23. S. J. Allen, G. McKay and K. Y. H. Khader, Environ. Pollut., 1989, 56, 39–50.
24. Q. Chang, G. D. Jiang, M. X. Hu, J. Huang and H. Q. Tang, Environ. Sci. Technol., 2014, 5, 1804–1809.
25. I. Langmuir, J. Am. Chem. Soc., 1916, 38, 2221–2295.
26. H. M. F. Freundlich, J. Phys. Chem., 1906, 57, 385–470.
27. O. Redlich and D. L. Peterson, J. Phys. Chem., 1959, 63, 1024.
28. D. Fafchamps, G. Rodriguez and P. Gallion, Photonics Switching, 2008, 4–7.
29. K. V. Kumar, M. M. Castro, M. Martinez-Escandell, J. Silvwstre-Albero and F. Rodriguez-Reinoso, Chem. Phys. Lett., 2010, 492, 187–192.
30. X. C. Fu, W. X. Shen and T. Y. Yao, Physical Chemistry, Higher Education Press, Beijing, 5th edn, 2005.
31. A. A. Khan and R. P. Singh, Colloids Surf., 1987, 24, 33–42.
32. Y. Li, J. Sun, Q. Du, L. Zhang, X. Yang, S. Wu, Y. Xia, Z. Wang, L. Xia and A. Cao, Carbohydr. Polym., 2014, 102, 755–761.
33. D. Shen, J. Fan, W. Zhou, B. Gao, Q. Yue and Q. Kang, J. Hazard. Mater., 2009, 172, 99–107.
34. J. T. Feng, J. W. Zhu, W. Liu, J. J. Li and W. Yan, Chem. Eng. J., 2015, 269, 316–322.
35. X. Wang, N. Zhu and B. Yin, J. Hazard. Mater., 2008, 153, 22–27.
36. W. J. Yao, Z. Liu and J. Liang, Environ. Sci. Technol., 2011, 34, 173–176.
37. Y. H. Li, Q. J. Du, T. H. Liu, Y. Qi, P. Zhang, Z. H. Wang and Y. Z. Xia, Appl. Surf. Sci., 2011, 257, 10621–10627.
38. B. Xu, K. Hu, X. W. Chen, Y. P. Xu and S. Xu, Modern Silk Science & Technology, 2014, 29, 9–12.
39. X. H. Li, W. J. Fan and Y. Cheng, J. Chin. Ceram. Soc., 2013, 32, 2238–2243.
40. X. Y. Zhang, Metal Mine, 2009, 11, 154–156.

---