Synthesis of citric acid functionalized magnetic graphene oxide coated corn straw for methylene blue adsorption

Accepted Manuscript Case Study Synthesis of citric acid functionalized magnetic graphene oxide coated corn straw for methylene blue adsorption Heyi Ge, Cuicui Wang, Shanshan Liu, Zhen Huang PII: DOI: Reference:

S0960-8524(16)31317-7 http://dx.doi.org/10.1016/j.biortech.2016.09.060 BITE 17082

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

29 July 2016 6 September 2016 12 September 2016

Please cite this article as: Ge, H., Wang, C., Liu, S., Huang, Z., Synthesis of citric acid functionalized magnetic graphene oxide coated corn straw for methylene blue adsorption, Bioresource Technology (2016), doi: http:// dx.doi.org/10.1016/j.biortech.2016.09.060

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis of citric acid functionalized magnetic graphene oxide coated corn straw for methylene blue adsorption Heyi Ge 1,2,*, Cuicui Wang 1,2, Shanshan Liu 1,2, Zhen Huang1,2 1

Shandong Provincial Key Laboratory of Preparation and Measurement of Building

Materials, University of Jinan, Jinan 250022, P.R. China 2

School of Material Science and Engineering, University of Jinan, Jinan 250022, P. R.

China * Corresponding author: Heyi Ge; Tel: 86-531- 82769579; email: [email protected] Abstract: The citric acid functionalized magnetic graphene oxide coated corn straw (CA-mGOCS) as a new adsorbent was synthesized in this work for the elimination of methylene blue (MB) from waste water. The as-prepared CA-mGOCS was tested by SEM, FTIR, XRD, Roman spectrum, TGA, particle size analyzer, BET and magnetic properties analyzer. Some factors affecting adsorption removal efficiency were explored. As a result, the addition of 5 g CS (CA-mGO5CS) had the better adsorption performance than other adsorbents. The pseudo-second-order model and the Freundlich described the adsorption behavior well. The equilibrium adsorption capacity was 315.5 mg g-1 for MB at pH=12 and 298 k. The electrostatic incorporation as well as hydrophobic interactions between CA-mGO5CS and MB determined the favourable adsorption property. Besides, the thermodynamic studies results △G<0, △H<0, △S<0 suggested that the adsorption was a spontaneous, exothermic and randomness decrease process. Finally, reusability studies imply that CA-mGO5CS has an excellent reproducibility. Keywords: corn straw; magnetic graphene oxide; citric acid; adsorption kinetics;

1

adsorption isotherm 1. Introduction The synthetic dyes have been widely used in plastic, leather, communication, food, paper printing and textile industries. Dyes wastewater is affecting the ecological environment and human health seriously because of its complex composition, depth color, high content organic pollutants, poor biodegradability and most of them as well as their intermediate holding mutagenicity, carcinogenicity and other toxicity (Hoa et al., 2016; González et al., 2015; Gan et al., 2015). Therefore, the treatment of dyes effluents has increasingly become the focus attention. Methylene blue (MB) is a representative compound of water-soluble azo dyes. MB forms a monovalent organic cationic quaternary ammonium ionic group, which has serious pollution for environment (Pirbazari et al., 2014; Feng et al., 2012; Li et al., 2016). At present, various ways have been commonly explored for the treatment of dyes effluents, including of biological treatment, chemical oxidation, flocculation, precipitation, adsorption and photocatalytic degradation methods. Among them, adsorption method with the low cost, strong operability, easy design and environmental protection has been recognized as the most suitable approach to separate various dyes from wastewater (Çelekli et al., 2013; Qei et al., 2009; Baldikova et al., 2014). Therefore, an adsorbent owning various excellent characteristics is urgently expected. The straw is a rich source of agricultural by-product. Its main component is hemicellulose and lignin (Pirbazari et al., 2014; El-Bindary et al., 2014). So far, the

2

most common way to deal with about 300 million tons of straw in China every year is incineration or deserting, which destroys the soil and atmosphere as well as is the waste of resources (Pirbazari et al., 2014; Zhang et al., 2012; Zhang et al., 2012). The straw holds the porous structure and rich available reactive groups, which can promote MB solution to penetrate easily into its interior. Therefore, preparation of straw-based adsorbents is one of the most promising ways to make full use of this abundant bioresource (Zhao et al., 2014; Xu et al., 2010; Zhang et al., 2011; El-Bindary et al., 2015). In recent years, in order to easily separate and collect adsorbent from water after adsorption, magnetic adsorbent shows a predominance future due to accomplishment the separation rapidly under an portable magnet, omitting the complex filtration or centrifugation (Tian et al., 2011; Yao et al., 2015). There are many reports about magnetic straw as adsorbents, such as magnetic rye straw for acridine orange and methyl green (Baldikova et al., 2015), magnetic wheat straw for MB (Pirbazari et al., 2014), magnetic wheat straw for arsenic (Tian et al., 2011), magnetic rice straw for heavy metals (Khandanlou et al., 2016), magnetic wheat straw biochar for MB (Li et al., 2016), and so on. However, the adsorption capacity of raw straw is inefficiency. Normally, chemical treatment particularly acid modification is one of the most universal methods to prepare straw owning the typical functional groups with the high adsorption capacity (Feng et al., 2012). So far, citric acid (CA) is frequently utilized to deal with various pollutions in waste water due to environment friendly, efficient and low-cost compound (Gong et al., 2008; Franklin et al., 2015). Sun et al. reported that the

3

maximum adsorption capacity of CA coated biochar derived from eucalyptus saw dust for MB was 178.57 mg g-1 (Sun et al., 2015). Leyva-Ramos et al. found that the biochar originated in corncob treatment with CA was more efficient than that of treatment with nitric acid, 55.2 and 19.3 mg g-1 adsorption capacities for Cd(II), respectively (Leyva-Ramos et al., 2005). Graphene oxide (GO) and graphene are considered good candidates for adsorbent due to unique two-dimension structure (Xing et al., 2015). GO is an oxidation of graphene. GO has a high surface area and abundant functional groups on surface introduced by the oxidation process of graphite. The groups are the key chemical skeletons which can be used as anchoring sites for dyes connection, making it become a potential material as a super and ideal adsorbent (Wang et al., 2015; Zhao et al., 2015). Our aim is to find a novel, cost-effective and high-efficiency CA modified magnetic GO coated corn straw (CA-mGOCS) adsorbent for removal of MB. There are many reports on absorption performance of mGO (Zhao et al., 2015; Cui et al., 2015; Ouyang et al., 2015; Ali et al., 2015; Hu et al., 2015), magnetic straw (Pirbazari et al., 2014; Li et al., 2016; Baldikova et al., 2015), CA coated straw (Gong et al., 2008). However, to the best of our knowledge, CA-mGOCS even GO coated CS as adsorbents has not been studied. In this work, mGOCS was synthesized using a simple one-step solvothermal method and then CA was decorated on mGOCS by adjusting the reaction pH, temperature and reaction time. The relevant factors that influenced the adsorption performance were discussed. The dynamics and thermodynamics were utilized to

4

discuss the interactions between CA-mGOCS and MB. The recyclability and reproducibility of CA-mGOCS were measured to explore its potential environment and economic benefits. 2. Experimental section 2.1. Materials Graphite powder (8000 meshes, 99.95%) and all chemical agents with analytical reagent grade were supplied by Aladdin Industrial Corporation (Shanghai, China). CS was gained from countryside of Jinan. The 100 mesh CS was obtained according to the method of (Chen et al., 2015) mentioned. 2.2. Synthesis of CA-mGOCS The GO aqueous suspension was prepared using the modified Hummers method (Wang et al., 2015, a; Wang et al., 2015, b; Chen et al., 2014; Chen et al., 2015). The mGO was synthesized using a solvothermal system (Cui et al., 2015). Firstly, 16.2 g ferric trichloride hexahydrate (FeCl3.6H2O) was added into ethylene glycol under magnetic stirring for dissolution completely. Then, 81.1 g GO aqueous suspension (10 mg/g) was dispersed into the above solution. With continuous magnetic stirring, 15 ml ammonium hydroxide was added to the solution by continuous dipping method. After forming the uniform mixture solution, 1 g polyvinylpyrrolidone was added slowly. Next, the above mixture was pour into six teflon reactors for 20 h at 180

. After that, the

reactors were cooled naturally and then the prepared mGO was obtained using magnetic separation and washing by ultrapure water and ethanol repeatedly. Finally, after the

5

oven dry at 50 ℃ and grind process, the mGO power was collected for the following experiments. mGO5CS and mGO10CS were prepared by the same way with mGO except added 5 g and 10 g CS fiber during added GO. The CA-mGO and CA-mGOCS materials were synthesized according to the method of (Lin et al., 2014) mentioned. 2.3. Characterizations The morphology properties of GO, CA-mGO and CA-mGOCA were characterized by SEM (FEI, QUANTA FEG 250, USA) system. FTIR measurement was carried out by a FTIR spectrometer (Nicolet 380, Thermo electron corporation, USA). The structure of the adsorbent was performed by XRD (D8 ADVANCE, Bruker, Germany). Raman spectra were measured using an inVia Reflex confocal Laser MicroRaman spectrometer (Renishaw, Britain). The TGA (TGA/DSC1/1600HTA) produced by Mettler-Toledo was utilized for recording thermal property of adsorbent with a heating rate of 10 ℃/min under argon atmosphere. Particle size and distribution were recorded by dynamic laser particle analyzer (LS-13320, Beckman Coulter, United States). The pore size distribution and specific surface area were characterized by the Micromeritics ASAP 2020 and porosity analyzer (Quantachrome, United States). Before measurement, the sample was dried under vacuum at 200 ℃ for 12 h. The magnetic performances were detected by a vibrating sample magnetometer (VSM, Lakeshore, Model 7400 series). The UV-vis spectra of MB solution were operated using a Shimadzu UV2550 spectrometer. The zeta potentials of the adsorbents with the different pH values were

6

tested by Zeta PALS (Phase Analysis Light Scattering) made by Brookhaven Instruments Corporation (USA). 2.4. MB adsorption The concentration of the MB stock solution for adsorption experiment was 2000 mg L-1. The magnetic adsorbent was added to 75 mL MB solution (100-1000 mg L-1) at pH=8, 298 k and oscillated for 3 h. The adsorbent was removed from the solution using a portable magnet. The supernatant MB amount was determined at 664 nm by UV-vis spectrum. The equilibrium adsorption amount qe (mg g-1) of MB and the removal efficiency onto adsorbent were calculated by the following equations (Li et al., 2016; Cui et al., 2015): qe =

(C0 − Ce )V m

Removal efficiency (%) =

(1)

C0 − Ce ×100% C0

(2)

Where V (L) represents the volume of the solution. C0 and Ce (mg L-1) represent the initial and equilibrium concentrations of solution, respectively. m (g) is the adsorbent dose (Li et al., 2016). 2.5. Desorption experiments The MB-loaded adsorbents were washed by the 25 mL HCl (0.1 mol L-1) solution firstly shaking for 90 min and then washed with distilled water three times. Next, the adsorbent was collected via magnetic separation then reused in the followed experiments. The adsorption-desorption cycles were carried out for five times. The removal efficiency of MB was measured by the UV-vis spectrum as description of

7

section 2.4. All experiments were performed at 298 k. 3. Results and discussion 3.1. Characterization of the adsorbent From Supplementary Fig. S2a, the SEM image exhibit the GO sheets characteristic of flat yet crumpled, which is in good agreement with our previous report (Chen et al., 2014). In addition, in our previous work, atomic force microscope results had demonstrated that the complete exfoliation GO sheet about 1 nm had been prepared successfully (Wang et al., 2015, a; Wang et al., 2015, b; Chen et al., 2014; Chen et al., 2015). The SEM image of CA-mGO composite (Fig. S2b) shows that the regular sphere is immobilized uniformly onto GO substrate. Fig. S2c shows that the fiber surface of CS treated by NaOH is unconfined and clean, which is attributed to the remove of the redundant paste by NaOH (Chen et al., 2015). Fig. S2d-g show the surface morphologies of CA-mGO5CS and CA-mGO10CS. Fig. S2e and Fig. S2g are the larger magnification of Fig. S2d and Fig. S2f, respectively. The images present the uniform coverage of the CA-mGO in the CS surface. In addition, after hybrid treatment, CS surface morphology does not change. Fig. S2h and Fig. S2i exhibit energy dispersive X-ray spectroscopy (EDXRF) for CA-mGO5CS and CA-mGO10CS, respectively. The characteristic peaks at 0.68, 6.20 and 7.30 keV are derived from Fe elements of Fe3O4 (Khandanlou et al., 2016). Therefore, from the above analysis, Fe3O4 existence on the CS has been confirmed. The chemical structures of CS treated by NaOH, mGO, mGO5CS, mGO10CS and

8

CA-mGO10CS are confirmed by FT-IR analysis (Fig. S3a). The peaks at 1060 cm-1 and 2907 cm-1 are attributed to CS epoxy group C-O and C-H stretching vibration, which is enhanced with the addition of CS weight. For mGO, the characteristic peaks at 574 cm-1, 1380 cm-1 and 1633 cm-1 are attributed to Fe-O stretching, C-O-H bending and aromatic ring C=C stretching derived from GO. As for CA-mGO10CS, the obvious peak at 1756 cm-1 belongs to C=O of CA, which indicates that CA was successfully grafted into the surface (Li et al., 2013; Kingsley et al., 2015). The XRD patterns of synthesized products are shown in Fig. S3b. It is clear that the prominent diffraction peaks at 2θ values of about 30.30°, 35.55°, 43.25°, 55.55°, 57.21° and 62.80° of the Fe3O4 nanoparticles correspond to the (220), (311), (400), (422), (511) and (440) planes, respectively [(JPPDS No. 19-0629)] (Hoa et al., 2016; Pirbazari et al., 2014; Chen et al., 2015). As for mGO and CA-mGO, two new characteristic peaks at 2θ values of about 33.15° and 48.48° of the Fe2O3 nanoparticles correspond to the (104) and (024) crystallographic planes growth, respectively (Sultan et al., 2016). It is worth noting that with the introduction of CS, the above new characteristic peaks disappear and at 2θ values of about 16.1° and 22.5° of CS appear (Pirbazari et al., 2014; Tian et al., 2011). Fig. S3c shows the Raman spectrum of four magnetic samples. The intensity ratio of ID/IG reveals the cluster size of sp 2 (Ouyang et al., 2015; He et al., 2013). The intensity

ratios

of

ID/IG

of

CA-mGO,

CA-mGO5CS,

CA-mGO10CS

and

CA-mGO5CS-MB are 1.08, 1.05, 0.97 and 0.92, respectively, which is due to the

9

introduction of CS and MB increasing the order of carbonaceous material structure and sp2 cluster size. The TGA curves of Fe3O4 and its derivative are presented in Fig. 1a. As shown in Fig. 1a, from the temperature of 30

to 900

, Fe3O4 only has about 2.1% mass loss.

The decrease of mass can be attributed to the evaporation of surface water of magnetic particles. As for the other materials, the mass loss process can be divided into three stages. The small proportions of mass loss below 200

of mGO and CA-mGO are

attributed to the evaporation of absorbed solvent. Then, a slow mass loss appearing from 400 to 450

is assigned to the removal of the residual oxygen functionality of

GO sheets. With the increasing temperature, a rapid mass loss appears at around 700 which is attributed to pyrolysis of carbon skeleton (Shen et al., 2010). As for CA-mGO5CS and CA-mGO10CS, a fast mass loss appearing from 300 to 400

is due

to CS pyrolysis of thermally functional groups. The percentages of residues are 66.7%, 60.1%, 41.6% and 24.6% corresponding to the mGO, CA-mGO, CA-mGO5CS and CA-mGO10CS, respectively. The obvious weightlessness peak exceeds 300

, implies

that CA-mGOCS owns a well thermal stability and can be applied in room temperature normally. Particle size distributions of different adsorbents are shown in Fig. 1b, which reveals that the different kinds of magnetic materials have the various diameter sizes. The mGO mean diameter is 0.756 µm, while CA-mGO, mGO5CS, CA-mGO5CS, mGO10CS and CA-mGO10CS mean diameters are 0.361 µm, 36.77 µm, 34.42 µm,

10

127.8 µm and 56.62 µm respectively, which demonstrates that the CS increased the diameter size, and CA decreased the diameter size by modified the dispersion of Fe3O4 nanoparticles (Lin et al., 2014). Normally, the specific surface area, surface functional group and pore size distribution are the factors to affect the adsorption property. BET analyses were executed to test the porosity of CA-mGO, CA-mGO5CS and CA-mGO10CS. Fig. S3e, Fig. 1c and Fig. S3f present the N2 adsorption-desorption curves and pore size distribution of CA-mGO, CA-mGO5CS and CA-mGO10CS. It can be noticed that the three adsorbents have the porosity structure and the narrow pore size distribution of 2-30 nm, 2-18 nm and 2-12 nm corresponding to CA-mGO, CA-mGO5CS and CA-mGO10CS, respectively. The isotherms inset of Fig. S3e, Fig. 1c and Fig. S3f also exhibit the three samples confirmed to the type IV hysteresis loop. The specific surface area of CA-mGO, CA-mGO5CS and CA-mGO10CS are 49.40 m2/g, 50.74 m²/g and 32.01 m²/g, respectively. Among the three adsorbents, CA-mGO5CS has the highest specific surface area. The perfect pore size distribution and specific surface area might promote the CA-mGO5C material possessing outstanding adsorption capacity (Cui et al., 2015). Fig. 1d shows the magnetization spectrum of magnetic materials to investigate its specific saturation magnetization (Ms), remnant magnetization (Mr) and coercivity (Hc). The hysteresis loops of all samples are S type showing the typical ferromagnetic of materials (Ouyang et al., 2015; Lin et al., 2014). The values of Ms, Mr and Hc of

11

materials are given in Table 1. We can observe in Table 1 that the Fe3O4 has the maximum of Ms about 81.57 emu g-1. However, CA-mGO, CA-mGO5CS and CA-mGO10CS Ms decrease to 37.15 emu g-1, 31.21 emu g-1 and 16.35 emu g-1, respectively. Moreover, the Mr and Hc also decline, which means that CS decreased magnetic performance. In addition, Fig. S3d shows a simple experiment to evaluate magnetic property of CA-mGO5CS by a portable magnet. The first flask is a uniform MB solution. The second flask is introduction of CA-mGO5CS and separation by external magnet. 3.2. Factors affecting adsorption The adsorbent dosage is the important factor to evaluate the adsorption properties. Therefore, different masses of magnetic adsorbents were added to 75 mL MB solution (100 mg L-1) at pH=8, 298 k and oscillated for 3 h. The removal efficiency is shown in Fig. 2a. The CA-mGO5CS has the fastest speed in the adsorption process and the equilibrium adsorbent dosage is 0.051 g. However, the equilibrium adsorbent dosages of CA-mGO and CA-mGO10CS are 0.065 g and 0.057 g, respectively. CA-mGO has the worst adsorption performance than that of CA-mGO5CS and CA-mGO10CS due to lack of CS. CS holds the porous structure and abundant available reactive groups, which can promote MB solution to penetrate easily into its interior. Besides, CS is rich in hydroxyl groups providing the sufficient hydrogen bond with MB. However, CA-mGO5CS adsorption capacity is more than CA-mGO10CS because of the lesser mGO. Fig. 2b shows the three different adsorbents removal efficiency and zeta potential

12

at the varied of pH values from pH=1 to pH=12. It is obvious that the removal efficiency of the adsorbents and zeta potential absolute value increase with pH growth and finally reach the maximum at pH=12. The main reason is that the -COOH groups became the -COO- on the CA and GO surface at the high pH and easy connected with MB+ cationic dye (Yu et al., 2013). In addition to electrostatic interactions, hydrophobic interactions (π-π stacking) and hydrogen bonding interactions were also existence occurring between adsorbent and the aromatic rings of the MB or the -COOH on the adsorbent surface and MB (Zhang et al., 2012; Zhang et al., 2011; Cui et al., 2015; Yu et al., 2013). The relevant mechanism is illustrated in Fig. 3. Fig. 4a, Fig. S4a and Fig. S5a present the influence of adsorption time and initial MB concentration on the adsorption efficiency. The three magnetic adsorbents with equilibrium adsorbent dosage were added into 75 mL MB solution with the different initial concentrations (100 mg L-1, 200 mg L-1, 300 mg L-1, 400 mg L-1, 500 mg L-1, 750 mg L-1 and 1000 mg L-1) at pH=12, 298 k. It can be seen that the adsorbing capacity rapid grows firstly and then reaches the equilibrium value at different initial MB concentration. The CA-mGO, CA-mGO5CS and CA-mGO10CS adsorption equilibrium time are 40 min, 20 min and 35 min, respectively. The maximum adsorption capacities in initial MB concentration of 1000 mg L-1 are 276.5 mg g-1, 300.3 mg g-1 and 315.5 mg g-1 according to CA-mGO, CA-mGO10CS and CA-mGO5CS, which indicated that CA-mGO5CS has the best adsorption performance than others. 3.3. Adsorption kinetics

13

The adsorption kinetics process of three adsorbents was investigated. The five models including of pseudo-first order kinetics model, pseudo-second order kinetics model, Elovich kinetics model, intra-particle diffusion kinetics model and Bangham kinetics model fitting results are shown in Fig. 4b-f, Fig. S4b-f and Fig. S5b-f. The relevant parameters are listed in the Table 2, Table 3 and Table 4. The pseudo-first-order assumes the adsorption rate linear declines with the increase of the removal efficiency. The pseudo-second-order holds the opinion that the rate controlling step of adsorption is the interaction between the adsorbent and adsorbate, such as ion sharing and transferring. The Elovich model describes ion exchange in liquid phase. The intra-particle diffusion and Bangham model assumes the internal diffusion and channel diffusion are the determining step to control the adsorption rate, respectively (Feng et al., 2012; Cui et al., 2015). From Fig. 4, Fig. S4 and Fig. S5, it can be found that the three magnetic adsorbents all fit the pseudo-second-order kinetics model well, and the determination coefficients (R2) are all greater than 0.99, which indicates that chemical reaction including of ion sharing and transferring mainly controlled the adsorption process (Cui et al., 2015). The relevant reaction mechanism is shown in Fig. 3. 3.4. Adsorption isotherms For study of the adsorption performance deeply, isothermal adsorption experiments were carried out at different temperatures. The isothermal adsorption equilibrium is the basis for describing the interaction behavior between the solute and the adsorbent

14

(Pirbazari et al., 2014; Li et al., 2016; Cui et al., 2015). The fitting isothermal adsorption of the CA grafted adsorbent with MB at different temperatures is shown in Fig. 5. From the Henry model (Fig. 5a), the removal efficiency of MB linear increases with the enhancement of the initial concentration of MB solution in the low concentration range. However, the growth rate becomes slow down when the concentration is about 400 mg L-1. It is indicated that adsorption sites were sufficient at low initial concentration but adsorption sites gradually saturated at high initial concentration and then reached the adsorption balanced gradually. The Langmuir model is a description of the uniform adsorption sites, constant adsorption energy and single molecule layer adsorption. The Freundlich is used to describe the uneven surface non ideal multi-layer adsorption. Theoretically, the adsorption capacity can reach infinity. The Temkin model is an assumption of adsorption heat decreasing linearly with adsorption capacity and used to explain chemical adsorption of the heterogeneous surface (Pirbazari et al., 2014; Xu et al., 2010; Tian et al., 2011). The fitting results of the four isothermal models are presented in Table 5. It can be found that the Freundlich R2 being much closer 1 compared to that of the other models demonstrated that multimolecular layer chemical and physical adsorption with ion sharing and transferring controlled the adsorption process (Cui et al., 2015). Through experimental data and the analyses of the pseudo-second-order and Freundlich models, the surface functional group was concluded as the most important factor to influence the adsorption ability in this study.

15

3.5. Thermodynamic parameters The Gibbs free energy change formula is
G = − RT ln K . Where K is adsorption equilibrium constant and calculated by the intercept of line of ln(qe/Ce) vs. q e (Fig. S6a).

qe (mg g-1) and Ce (mg L-1) are equilibrium adsorption capacity and concentrations of adsorbate. R is the gas constant (8.3145 J mol-1 K-1). T (K) is Kelvin temperature. Meanwhile, from the formula: ln K = −


G
H
S =− + , △H and △S were calculated. RT RT R

Where △S (J mol-1 K-1) is the entropy change, △H (kJ mol-1) is the enthalpy change (Xu et al., 2010; Cui et al., 2015). △S and △H were obtained by the intercept and the slope of line about lnK vs. 1/T (Fig. S6b). Table 5 lists the values of △S, △H and △G. The △G values are -9.25, -7.75 and -4.74 kJ mol-1 according to 298 k, 318 k and 338 k, respectively, which indicates that the removal MB procedure was a spontaneous reaction. In addition, the △G absolute values decrease with the increase of temperature, implied a higher temperature impeded the adsorption process. The △H and △S values are -42.66 kJ mol-1 and -111.37 J mol-1 K-1, respectively. The results indicate that the adsorption was an exothermic and randomness decrease process. 3.6. Desorption and reusability To explore the desorption and reusability of MB loaded on CA-mGO5CS, CA-mGO10CS and CA-mGO, 0.1 mol L-1 HCl was used as a eluants to carry out the recycle experiment (Fig. 6). The adsorbents surface would protonate and the connection between MB and adsorbents would be slow down and even forbidden. These phenomena were mentioned by the section of the pH influence on the removal

16

efficiency. After five recycles, CA-mGO5CS, CA-mGO10CS and CA-mGO still holds good reusable adsorption property (84.02%, 82.41% and 77.01% removal efficiency, respectively). The good regeneration indicated that the environmental, cost-effective CA-mGO5CS adsorbent for dealing with effluent has been synthesized. Finally, Table 6 lists the maximum capacity of comparative tests between CA-mGO5CS and the reported sorbents. The qm of CA-mGO5CS for MB are clearly more than that of other reported adsorbents. The CA-mGO5CS with the excellent adsorption capacity as well as the feasible synthesis method and simple magnetic separate after adsorption process will lead it to a broad application prospects. 4. Conclusion

The excellent adsorbent CA-mGO5CS with the favourable adsorption performance for MB was synthesized in this work. The adsorbent reached the adsorption equilibrium at adsorbent dose 0.051 g for 20 min and had the best adsorption performance at pH=12 and temperature 298 k. The pseudo-second-order described adsorption kinetic behavior well. The adsorption isotherm behavior of between CA-mGO5CS and MB was fitted to the Freundlich model. Besides, the thermodynamic studies results △G<0, △H<0, △S<0 suggested the adsorption was a spontaneous, exothermic and randomness

decrease process. Finally, desorption and reusability studies imply that CA-mGO5CS has an excellent reproducibility in water treatment.

Acknowledgments

17

The authors sincerely acknowledge the financial support from the major special project of science and technology in Shandong province (Grant NO. 2015JMRH0110), and the major special project for independent innovation and achievements transformation of Shandong province, China (Grant NO. 2014ZZCX05302). References

Hoa, N.V., Khong, T.T., Quyen, T.T.H., Trung, T.S., 2016. One-step facile synthesis of mesoporous graphene/Fe3O4/chitosan nanocomposite and its adsorption capacity for a textile dye. J. Water Proc. Eng. 9, 170-178. González, J.A., Villanueva, M.E., Piehl, L.L., Copello, G.J., 2015. Development of a chitin/graphene oxide hybrid composite for the removal of pollutant dyes: Adsorption and desorption study. Chem. Eng. J. 280, 41-48. Gan, L., Shang, S.M., Hu, E.L., Yuen, C.W.M., Jiang, S.X., 2015. Konjac glucomannan/graphene oxide hydrogel with enhanced dyes adsorption capability for methyl blue and methyl orange. Appl. Surf. Sci. 357, 866-872. Pirbazari, A.E. Saberikhah, E., Kozani, S.S.H., 2014. Fe3O4-wheat straw: preparation, characterization and its application for methylene blue adsorption. Water Resour. Ind. 7-8, 23-37. Feng, Y.F., Zhou, H., Liu, G.H., Qiao, J., Wang, J.H., Lu, H.Y., Yang, L.Z., Wu, Y.H., 2012. Methylene blue adsorption onto swede rape straw (Brassica napus L.) modified by tartaric acid: Equilibrium, kinetic and adsorption mechanisms. Bioresour. Technol. 125, 138-144.

18

Li, G.T., Zhu, W.Y., Zhang, C.Y., Zhang, S., Liu, L.L., Zhu, L.F., Zhao, W.G., 2016. Effect of a magnetic field on the adsorptive removal of methylene blue onto wheat straw biochar. Bioresour. Technol. 206, 16-22. Çelekli, A., Bozkurt, H., Geyik, F., 2013. Use of artificial neural networks and genetic algorithms for prediction of sorption of an azo-metal complex dye onto lentil straw. Bioresour. Technol. 129, 396-401. Oei, B.C., Ibrahim S., Wang, S.B., Ang, H.M., 2009. Surfactant modified barley straw for removal of acid and reactive dyes from aqueous solution. Bioresour. Technol. 100, 4292-4295. Baldikova, E., Safarikova, M., Safarik, I., 2015. Organic dyes removal using magnetically modified rye straw. J. Magn. Magn. Mater. 380, 181-185. El-Bindary, A.A., El-Sonbati, A.Z., Al-Sarawy, A.A., Mohamed, K.S., Farid, M.A., 2014. Adsorption and thermodynamic studies of hazardous azocoumarin dye from an aqueous solution onto low cost rice straw based carbons. J. Mol. Liq. 199, 71-78. Zhang, W.X., Li, H.L., Kan, X.W., Dong, L., Yan, H., Jiang, Z.W., Yang, H., Li, A.M., Cheng, R.S., 2012. Adsorption of anionic dyes from aqueous solutions using chemically modified straw. Bioresour. Technol. 117, 40-47. Zhang, W.X., Yang, H., Dong, L., Yan, H., Li, H.J., Jiang, Z.W., Kan, X.W., Li, A.M., Cheng, R.S., 2012. Efficient removal of both cationic and anionic dyes from aqueous solutions using a novel amphoteric straw-based adsorbent. Carbohyd.

19

Polym. 90, 887-893. Zhao, B.L., Shang, Y., Xiao, W., Dou, C.C., Han, R.P., 2014. Adsorption of Congo red from solution using cationic surfactant modified wheat straw in column model. J. Environ. Chem. Eng. 2, 40-45. Xu, X., Gao, B.Y., Yue, Q.Y., Zhong, Q.Q., 2010. Preparation and utilization of wheat straw bearing amine groups for the sorption of acid and reactive dyes from aqueous solutions. J. Hazard. Mater.182, 1-9. Zhang, WX., Yan, H., Li, H.J., Jiang, Z.W., Dong, L., Kan, X.W., Yang, H., Li, A.M., Cheng, R.S., 2011. Removal of dyes from aqueous solutions by straw based adsorbents: Batch and column studies. Chem. Eng. J.168, 1120-1127. El-Bindary, A.A., El-Sonbati, A.Z., Al-Sarawy, A.A., Mohamed, K.S., Farid, M.A., 2015. Removal of hazardous azopyrazole dye from an aqueous solution using rice straw as a waste adsorbent: Kinetic, equilibrium and thermodynamic studies. Spectrochim. Acta A 136, 1842-1849. Tian, Y., Wu, M., Lin, X.B., Huang, P., Huang, Y., 2011. Synthesis of magnetic wheat straw for arsenic adsorption. J. Hazard. Mater. 193, 10-16. Yao, T., Guo, S., Zeng, C.F., Wang, C.Q., Zhang, L.X., 2015. Investigation on efficient adsorption of cationic dyes on porous magnetic polyacrylamide microspheres. J. Hazard. Mater. 292, 90-97. Khandanlou, R., Masoumi, H.R.F., Ahmad, M.B., Shameli, K., Basri, M., Kalantari, K., 2016. Enhancement of heavy metals via nanocomposites of rice straw and Fe3O4

20

nanoparticles using artificial neural network (ANN). Ecol. Eng. 91, 249-256. Gong, R.M., Zhu, S.X., Zhang, D.M., Chen, J., Ni, S.J., Guan, R., 2008. Adsorption behavior of cationic dyes on citric acid esterifying wheat straw: kinetic and thermodynamic profile. Desalination 230, 220-228. Franklin, D.S., Guhanathan, S., 2015. Investigation of citric acid-glycerol based pH-sensitive biopolymeric hydrogels for dye removal applications: A green approach. Ecotox, Environ, Safe, 121, 80-86. Sun, L., Chen, D.M., Wan, S.G., Yu, Z.B., 2015. Performance, kinetics, and equilibrium of methylene blue adsorption on biochar derived from eucalyptus saw dust modified with citric, tartaric, and acetic acids. Bioresour. Technol. 198, 300-308. Leyva-Ramos, R., Bernal-Jacome, L.A., Acosta-Rodriguez, I., 2005. Adsorption of cadmium(II) from aqueous solution on natural and oxidized corncob. Sep. Purif. Technol. 45(1), 41-49. Xing, H.T., Chen, J.H., Sun, X., Huang, Y.H., Su, Z.B., Hu, S.R., Weng, W., Li, S.X., Guo, H.X., Wu, W.B., 2015. NH2-rich polymer/graphene oxide use as a novel adsorbent for removal of Cu (II) from aqueous solution. Chem. Eng. J. 263, 280-289. Wang, W., Cheng, Y.L., Kong, T., Cheng, G.S., 2015. Iron nanoparticles decoration onto three-dimensional graphene for rapid and efficient degradation of azo dye. J. Hazard. Mater. 299, 50-58. Zhao, C., Guo, J.H., Yang, Q., Tong, L., Zhang, J.W., Zhang, J.W., Gong, C.H., Zhou,

21

J.F., Zhang, Z.J., 2015. Preparation of magnetic Ni@graphene nanocomposites and efficient removal organic dye under assistance of ultrasound. Appl. Surf. Sci. 357, 22-30. Cui, L.M., Wang, Y.G., Gao, L., Hu, L.H., Yan, L.G., Wei, Q., Du, B., 2015. EDTA functionalized magnetic graphene oxide for removal of Pb(II), Hg(II) and Cu(II) in water treatment: Adsorption mechanism and separation property. Chem. Eng. J. 281, 1-10. Ouyang, K., Zhu, C.H., Zhao, Y., Wang, L.C., Xie, S., Wang, Q., 2015. Adsorption mechanism of magnetically separable Fe3O4/graphene oxide hybrids. Appl. Surf. Sci. 355, 562-569. Ali, M., Nader, K., Ali, J., 2015. Fabrication of graphene/Fe3O4@polythiophene nanocomposite and its application in the magnetic solid-phase extraction of polycyclic aromatic hydrocarbons from environmental water samples. Anal. Chim. Acta 868, 1-9. Hu, X.J., Liu, Y.G., Wang, H., Zeng, G.M., Hu, X., Guo, Y.M., Li, T.T., Chen, A.W., Jiang, L.H., Guo, F.Y., 2015. Adsorption of copper by magnetic graphene oxide-supported β-cyclodextrin: Effects of pH, ionic strength, background electrolytes, and citric acid. Chem. Eng. Res. Des. 93, 675-683. Chen, J., Teng, Z.Y., Wu, J.J., 2015. Recycling of Waste FRP and Corn Straw in Wood Plastic Composite. Polym. Compos. DOI 10.1002/pc.23789. Wang, C.C., Ge, H.Y., Liu, H.S., Liang, J.J., 2015. Properties of carbon fiber and

22

composites modified with different-sized graphene oxide sheets. Polym. Compos. DOI: 10.1002/pc.23466. Wang, C.C., Ge, H.Y., Liu, H.S., Guo, S., 2015. Microstructure and properties of carbon fiber sized with pickering emulsion based on graphene oxide sheets and its composite with epoxy resin. J. Appl. Polym. Sci. 132, 42285-42292. Chen, J., Zhao, D., Jin, X., Wang, C.C., Wang, D.Z., Ge, H.Y., 2014. Modifying glass fibers with graphene oxide: Towards high-performance polymer composites. Compos. Sci. Technol. 97, 41-45. Chen, J., Zhao, D., Ge, H.Y., Wang, J., 2015. Graphene oxide-deposited carbon fiber/cement composites for electromagnetic interference shielding application. Constr. Build. Mater. 84, 66-72. Lin, J.F., Tsai, C.C., Lee. M.Z., 2014. Linear birefringence and dichroism in citric acid coated Fe3O4 magnetic nanoparticles. J. Magn. Magn. Mater. 372, 147-158. Li, L., Mak, K.Y., Leung, C.W., Chan, K.Y., Chan, W.K., Zhong, W., Pong, P.W.T., 2013. Effect of synthesis conditions on the properties of citric-acid coated iron oxide nanoparticles. Microelectron. Eng. 110, 329-334. Kingsley, M.P., Desai, P.B., Srivastava, A.K., 2015. Simultaneous electro-catalytic oxidative determination of ascorbic acid and folic acid using Fe3O4 nanoparticles modified carbon paste electrode. J. Electroanal. Chem. 741, 71-79. Sultan, S., Kareem, K., He, L., 2016. Synthesis, characterization and resistant performance of α-Fe2O3@SiO2 composite as pigment protective coatings. Surf.

23

Coat. Tech. 300, 42-49. He, F., Lam, K., Ma, D., Fan, J.T., Chan, L.H., Zhang, L.M., 2013. Fabrication of graphene nanosheet (GNS)-Fe3O4 hybrids and GNS–Fe3O4/syndiotactic polystyrene composites with high dielectric permittivity. Carbon 58, 175-184. Shen, J., Shi, M., Li, N., Yan, B., Ma, H., Hu, Y., Ye, M., 2010. Facile synthesis and application of Ag-chemically converted graphene nanocomposite. Nano Res. 3, 339-349. Yu, Y., Murthy, B.N., Shapter, J.G., Constantopoulos, K.T., Voelcker, N.H., Ellis, A.V., 2013. Benzene carboxylic acid derivatized graphene oxide nanosheets on natural zeolites as effective adsorbents for cationic dye removal. J. Hazard. Mater. 260, 330-338.

Figure and Table captions:

24

Fig. 1 Characterization of as-prepared adsorbents: (a) TGA analysis of Fe3O4, mGO, CA-mGO, CA-mGO5CS and CA-mGO10CS; (b) Particle size distribution of mGO, CA-mGO, mGO5CS, CA-mGO5CS, mGO10CS and CA-mGO10CS; (c) Pore size distribution of CA-mGO5CS, inset is BET analysis; (d) Magnetization curve of Fe3O4, CA-mGO, CA-mGO5CS and CA-mGO10Cs, inset is the expanded low field hysteresis curves. Fig. 2 Effect of adsorbent dosage and pH on the removal efficiency: (a) Effect of adsorbent dosage on the removal efficiency; (b) Effect of pH on the removal efficiency and zeta potential Fig. 3 The possible interactions between CA-mGO5CS and MB: (a) Electrostatic interactions, (b) Hydrophobic interactions Fig. 4 The fitting of adsorption kinetic models: (a) The three-dimensional graph for MB adsorption amount by CA-mGO5CS at different initial MB concentration and adsorption time; (b) Titting of pseudo-first order; (c) Pseudo-second order; (d) Elovich; (e) Intra-particle diffusion; (f) Bangham kinetic models for MB adsorption on CA-mGO5CS under different initial concentrations (100 mg L-1, 200 mg L-1, 300 mg L-1, 400 mg L-1, 500 mg L-1, 750 mg L-1 and 1000 mg L-1) Fig. 5 The adsorption isotherm models: (a) Henry; (b)Langmuir; (c) Freundlich; (d) Temkin adsorption isotherm fit of MB (mCA-mGO5CS=0.051 g, V=75 mL, pH at 12, temperature at 298 K-338 K) Fig. 6 Adsorption-desorption recycles (m=0.05 g, V=75 mL, pH=12, 298 K)

25

Table 1 Magnetic properties of magnetic materials Sample Ms (emu/g) Hc (Oe) Fe3O4 81.57 104.55 CA-mGO 37.15 68.12 CA-mGO5CS 31.21 23.52 CA-mGO10CS 16.35 0.24

Mr (emu/g) 9.28 2.80 1.50 0.06

Table 2 Kinetic parameters of various models fitted to experimental data of CA-mGO MB c0 Kinetic model Parameter (mg L-1) 100 200 300 400 500 750 1000 k1 (mg min 0.06623 0.06979 0.07265 0.07394 0.0721 0.08771 0.09644 -1 g ) Pseudo-first-order q e (mg g-1) 128.59 165.44 178.78 194.846 226.01 267.83 279.51 2 R 0.98806 0.98487 0.97645 0.97861 0.99199 0.98113 0.95107 k2 (mg min 0.00039 0.00047 0.00038 0.00035 0.00033 0.00036 0.00037 g-1) Pseudo-second-order q e (mg g-1) 150.38 177.30 211.86 240.96 271.74 295.86 315.46 2 R 0.9983 0.99611 0.99411 0.99322 0.99536 0.99564 0.99442 α (mmol g-1 23.17 29.18 39.65 52.37 61.84 76.69 86.90 -1 min ) Elovich β (g 0.03187 0.02607 0.022596 0.020426 0.018126 0.016685 0.015547 mmol-1) R2 0.96388 0.94262 0.95439 0.96398 0.96711 0.96766 0.95744 -1 kdif (mg g 15.89 19.65 22.71 25.33 28.62 31.24 33.59 Intra-particle min-1/2) diffusion C 1.79 2.19 9.07 17.15 21.41 29.91 35.1 2 R 0.92817 0.93015 0.91452 0.90587 0.90427 0.89048 0.87734 kb (mg g-1) 7.94 12.93 18.85 25.85 31.08 38.83 45.12 Bangham m 1.41 1.59 1.73 1.87 1.92 2.03 2.11 2 R 0.92164 0.95225 0.94322 0.93192 0.93348 0.92516 0.92985

26

Table 3 Kinetic parameters of various models fitted CA-mGO5CS MB c0 Kinetic model Parameter (mg L-1) 100 200 300 k1 (mg min 0.12697 0.14719 0.15363 g-1) Pseudo-first-order qe (mg g-1) 166.34 215.02 239.23 R2 0.9866 0.97826 0.96633 k2 (mg min 0.00056 0.00048 0.00046 -1 g ) Pseudo-second-order qe (mg g-1) 197.24 235.85 271.00 2 R 0.99107 0.98346 0.98489 -1 α (mmol g 49.66 61.61 76.12 min-1) Elovich β (g mmol-1) 0.02333 0.01966 0.01690 2 R 0.9695 0.94717 0.94964 kdif (mg g-1 27.59 33.13 38.56 -1/2 Intra-particle min ) diffusion C 6.52 8.44 13.10 R2 0.92323 0.92206 0.91001 kb (mg g-1) 21.65 30.66 36.93 Bangham m 1.64 1.71 1.82 R2 0.91845 0.94684 0.93512

to experimental data of

400 0.15054

500 0.1358

750 0.15606

1000 0.17356

261.96 276.93 318.35 0.98377 0.99052 0.98621 0.00048 0.00043 0.00044

310.23 0.94286 0.00047

298.51 333.33 359.71 0.99049 0.99121 0.99184 95.22 106.45 125.40

374.53 0.98802 140.73

0.01536 0.01379 0.01277 0.95939 0.95849 0.95824 42.79 47.70 51.84

0.01217 0.9388 54.59

20.90 23.48 30.44 0.903 0.90197 0.89479 45.75 51.40 60.57 1.91 1.92 1.99 0.92462 0.92615 0.92228

36.44 0.87387 68.05 2.07 0.91619

27

Table 4 Kinetic parameters of various models fitted CA-mGO10CS MB c0 Kinetic model Parameter (mg L-1) 100 200 300 k1 (mg min 0.09218 0.09713 0.10116 g-1) Pseudo-first-order qe (mg g-1) 124.90 160.07 173.67 2 R 0.97696 0.97592 0.96691 k2 (mg min 0.00072 0.00053 0.00049 -1 g ) Pseudo-second-order qe (mg g-1) 170.94 208.77 239.81 R2 0.99514 0.99272 0.98934 α (mmol g-1 43.68 53.24 61.46 -1 min ) Elovich β (g mmol-1) 0.02643 0.02183 0.01894 2 R 0.97809 0.97361 0.96704 -1 kdif (mg g 20.87 25.49 29.10 Intra-particle min-1/2) diffusion C 15.11 14.72 21.60 2 R 0.8782 0.89094 0.86689 kb (mg g-1) 18.98 23.11 29.03 Bangham m 1.70 1.728 1.789 2 R 0.85151 0.89005 0.8775

to experimental data of

400 500 750 0.10271 0.09997 0.11041

1000 0.11736

187.72 216.18 221.13 0.96667 0.97920 0.94009 0.00053 0.00049 0.00049

215.18 0.89245 0.00047

263.85 294.99 321.54 0.99366 0.99462 0.99327 74.19 87.02 99.08

343.64 0.99149 107.62

0.01699 0.01537 0.01398 0.96869 0.96515 0.95702 32.23 35.68 39.06

0.01297 0.9482 41.92

29.64 36.37 43.19 0.85562 0.8558 0.84025 35.13 45.95 53.30 1.84 2.02 2.08 0.85407 0.87924 0.87722

47.82 0.82535 57.33 2.07 0.87055

28

Table 5 Isotherm parameters for adsorption of MB onto CA-mGO5CS at 298 K, 318 K, 338 K △S (J Temperature Adsorption △G (kJ △H (kJ 2 Parameter Value R mol-1 (k) isotherm mol-1) mol-1) K-1) Henry Langmuir 298

Freundlich Temkin

Henry Langmuir 318

Freundlich Temkin

Henry Langmuir 338

Freundlich Temkin

KH b (L mg-1) qm(mg g-1) KF 1/n bT AT

0.1942 0.05640 267.38 75.65 0.21039 52.57 0.77188

0.86146 0.68874

KH b (L mg-1) qm(mg g-1) KF 1/n bT AT

0.184 0.02893 261.10 52.66 0.2571 50.89 0.30222

0.82256 0.81767

KH b (L mg-1) qm(mg g-1) KF 1/n bT AT

0.18167 0.01134 265.25 25.93 0.3527 46.87 0.09653

0.77296 0.9537

-9.25

0.9445

-

-

-42.66

-111.37

-

-

0.90365

-7.75

0.97226 0.95588

-4.74

0.96425 0.97740

29

Table 6 Comparison of adsorption capacities with various adsorbents for dye Capacit y qm(mg g-1)

Kinetic

Isotherm

Ref.

92.3, 51.6

Pseudo-second-order

Freundlich

Gan etal. (2015)

208.3, 384.6

-

Langmuir

Baldikova et al. (2015)

156.34

Pseudo-second-order

Freundlich

El-Bindary et al. (2014)

MB

249.25

Pseudo-second-order

Langmuir

Hoa et al. (2016)

MB

246.4

Pseudo-second-order

Acidic functional wheat straw biochar

MB

62.5

Amphoteric wheat straw

MB

138.0-1 51.7

Carboxymethylation modified wheat straw

MB

274.7

Citric acid esterifying wheat straw

MB

312.50

Eucalyptus saw dust modified with citric, tartaric and acetic acids Citric acid functionalized magnetic graphene oxide coated corn straw

Adsorbent

Konjac glucomannan/graphene oxide hydrogel Magnetically modified rye straw

Rice straw based carbons

Mesoporous graphene@Fe3O4@chitosan hybrids Swede rape straw modified by tartaric acid

Target dye

methyl blue, methyl orange acridine orange, methyl green A new hazardous azocoumarin dye

The Elovich and double-constant models Pseudo-second-order and Elovich models Pseudo-second order equation and Elovich models

Langmuir

Feng et al. (2012)

Freundlich

Li et at. (2016)

Langmuir

Zhang et al. (2012)

Langmuir

Zhang et al. (2011)

Pseudo-second-order

Langmuir

Gong et al. (2008)

MB

178.57, 99.01 and 29.94

Pseudo-second-order

Langmuir

Sun et al. (2015)

MB

315.5

Pseudo-second-order

Freundlich

This work

30

Figure

Fig. 1 Characterization of as-prepared adsorbents: (a) TGA analysis of Fe3O4, mGO, CA-mGO, CA-mGO5CS and CA-mGO10CS; (b) Particle size distribution of mGO, CA-mGO, mGO5CS, CA-mGO5CS, mGO10CS and CA-mGO10CS; (c) Pore size distribution of CA-mGO5CS, inset is BET analysis; (d) Magnetization curve of Fe3O4, CA-mGO, CA-mGO5CS and CA-mGO10Cs, inset is the expanded low field hysteresis curves.

Figure

Fig. 2 Effect of adsorbent dosage and pH on the removal efficiency: (a) Effect of adsorbent dosage on the removal efficiency; (b) Effect of pH on the removal efficiency

Figure

Fig. 3 The possible interactions between CA-mGO5CS and MB: (a) Electrostatic interactions, (b) Hydrophobic interactions

Figure

Fig. 4 The fitting of adsorption kinetic models: (a) The three-dimensional graph for MB adsorption amount by CA-mGO5CS at different initial MB concentration and adsorption time; (b) Titting of pseudo-first order; (c) Pseudo-second order; (d) Elovich; (e) Intra-particle diffusion; (f) Bangham kinetic models for MB adsorption on CA-mGO5CS under different initial concentrations (100 mg L-1, 200 mg L-1, 300 mg L-1, 400 mg L-1, 500 mg L-1, 750 mg L-1 and 1000 mg L-1)

Figure

Fig. 5 The adsorption isotherm models: (a) Henry; (b)Langmuir; (c) Freundlich; (d) Temkin adsorption isotherm fit of MB (mCA-mGO5CS=0.051 g, V=75 mL, pH at 12, temperature at 298 K-338 K)

Figure

Fig. 6 Adsorption-desorption recycles (m=0.05 g, V=75 mL, pH=12, temperature at 298 K)

Graphical Abstract (for review)

Graphical abstract: The schematic illustration of preparing CA-mGOCS and adsorption MB process

1. The CA-mGOCS as a novel magnetic adsorbent was synthesized for the removal of MB from aqueous solution. 2. The electrostatic incorporation as well as hydrophobic interactions between CA-mGO5CS and MB determined the favourable adsorption property. 3. The adsorption property obeyed the pseudo-second order kinetic model and the Freundlich isotherm well. 4. The thermodynamic studies results suggested that the adsorption mechanism was a spontaneous, exothermic and randomness decrease process essentially. 5. Desorption and reusability studies implied CA-mGO5CS has an excellent reproducibility.

31