One-step preparation of CdS-modified mesoporous titanate nanobelts and their application as high-performance cationic dye adsorbents

Abstract

CdS-modified mesoporous titanate nanobelts (CTNS) were prepared successfully by one step, using TiO₂ nanobelts and CdS quantum dots, and applied as an adsorbent for the first time for the removal of typical cationic dyes. The exceptionally high adsorption performance for cationic dyes is exhibited by sample CTNS-3 with a typical mesoporous structure. Its adsorption capacity for methylene blue (MB) and neutral red (NR) could reach 2444.9 mg g⁻¹ and 3741.8 mg g⁻¹ respectively, which is extraordinary more than the other reported adsorbents, and the corresponding adsorption yields are also very high. The equilibrium absorption capacities of both MB and NR are very high no matter if the solution is acidic, neutral or alkaline, suggesting a great application prospect for CTNS-3 in removing cationic dyes. The adsorption kinetics and mechanism are also studied. The results indicate that the whole adsorption process of MB on CTNS-3 is attributed to the Freundlich adsorption model. The second-order kinetic model conforms to the actual adsorption process of MB. The adsorption process follows the intra-particle diffusion model along with the existence of other adsorption types, such as chemical adsorption.

---

Introduction

In the current era, the upcoming threats of pollution and environmental issues surrounding the presence of colors in effluent are enduring problems for dyestuff manufacturers, dyers, finishers and water companies. The effluents from dyestuff manufacturing and textile industries are generally highly colored with a large amount of suspended organic solids and hence are vital key sources of water pollution.¹ In addition, stringent color consent standards ask for the treatment of effluents containing organic dyestuff, which are harmful to fish and other aquatic organisms.²

Methylene blue (MB) and neutral red (NR) dyes are model compounds and also the typical cationic dyes for the researchers for removing organic dye contaminants from aqueous solutions, and are also the most frequently used dyestuff for dying cotton, wood and silk.^3–6 Moreover, dyes may also be responsible for permanent injury to the eyes of human and animals owing to eye burning aptitudes.⁴ Therefore, it is necessary to treat effluents containing organic dyes prior to their discharge out, preventing their damages on environment and aquatic lives. So far, the most commonly adopted methods for removing organic dyes include adsorption,⁷ biological treatment,⁸ chemical oxidation,⁹ cation exchange membranes¹⁰ and photocatalytic or electro-catalytic degradation^11,12 etc. Amongst the numerous methods of removing dyes, adsorption is as an efficient and economic choice, giving the best removal efficiency, and it can also be used to remove different types of coloring materials (e.g. dyes, pigments and other colorants) at the same time and control the bio-chemical oxygen demand.^13,14

Numerous studies have focused on the research for removing of cationic dyes by using different adsorbents. The usually used adsorbents include activated carbons,^3,15 agricultural by-products,^4,16 nanoparticles,¹⁷ mesoporous materials¹⁸ and so on. Foo et al.¹⁹ studied the MB adsorption by oil palm fiber activated carbon using microwave heating as an alternative activation technique. Results showed that the monolayer adsorption capacity of oil palm fiber activated carbon for MB is 312.5 mg g⁻¹, providing a strong evidence to support the potential use of microwave heating as an alternative activation technique. Mak et al.¹⁷ investigated the fast adsorption of MB using magnetic nanoparticles, and the results demonstrate that the maximum adsorption amount reaches 199 mg g⁻¹ and the adsorption isotherm data is fit to Langmuir isotherm. Mesoporous materials with high surface area usually own better adsorption performance for dyes.²⁰

Nanomaterials or micro-nano materials are widely used in various fields due to their excellent physical, chemical and photoelectric properties.²¹ TiO₂, CdS or their modification materials and their nanocomposites usually applied as photocatalytic materials.^22–25 Their adsorption capacity is usually lower as compared with the common adsorbents and primarily emphasize on the advantages of catalytic performances, not adsorption performance. However, some reports depict the adsorption on nano-TiO₂ with adsorbents usually composed of some inorganic ions, but their removal efficiencies are not very high.^26,27

Previous report²⁸ had found that the CdS-modified titanate nanocomposites can enhance visible light photocatalytic activity. Although some mesoporous materials have been applied into the removal of organic dyes, so far CdS-modified mesoporous titanate nanocomposites as adsorbents for removing organic dyes have not been reported. Here in, a series of mesoporous or micropore CdS-modified mesoporous titanate nanocomposites (CTNS) by changing the ratio of CdS quantum dots to TiO₂ nanobelts, and first applied CTNS into the removal of organic cationic dyes. MB and NR are chosen as the typical cationic dyes to study the adsorption performance, adsorption kinetics and adsorption mechanism of CTNS.

Experimental

Materials synthesis

The synthesis protocol for CdS-modified mesoporous titanate nanocomposites was followed as. Typically, 5 mg CdS quantum dots and 500 mg TiO₂ nanobelts were dispersed in the mixed solution containing 40 mL NaOH (1 mol L⁻¹) and 40 mL ethanol. After ultrasonication for 15 min, the suspension was transferred into 100 mL of Teflon-lined stainless autoclave and maintained at 180 °C for 2 h. The obtained yellow precipitates were washed by deionized water and ethanol alternately until pH = 7. A series of CTNS materials including CTNS-1, 2, 3, 4, 5, 6 and 7 were prepared by varying the ratio of CdS to TiO₂ nanobelts. The detailed reaction conditions are listed in Table S1.† For the detail synthesis process of TiO₂ nanobelts and CdS quantum dots, please see the ESI (ESI-1 Experimental section†).

Material characterization

Transmission electron microscope (TEM) images of the CTNS were obtained with Tecnai G2 F30 (FEI, Holland) transmission electron microscopy. X-Ray Diffraction (XRD) patterns were measured on an X-ray powder diffractometer (PANalyticalX'pert PRO-DY2198, Holland) operating at 40 kV and 40 mA using Cu Kα radiation (λ = 0.15406 nm). X-ray photoelectron spectroscopy (XPS) data were obtained using an AXIS-ULTRA DLD-600W instrument (Shimadzu, Japan). Zeta potentials of CTNS were measured using a Zetasizer Nano (Malvern Instruments). And zeta potentials can indicate the charge on the CTNS surface.

Adsorption studies

The adsorption performances of CTNS were investigated in distilled water with MB and NR as targets. The initial parameters including dye concentration, experimental temperature and pH of MB solution were 100 mg L⁻¹, 25 °C and 6.47 respectively. For the experiments of adsorption studies, 50 mg CTNS sample was added into 100 mL of dyes solution at desired temperature and pH. The preliminary experiment revealed that about 30 min was required for the adsorption process to reach the equilibrium concentration. A magnetic stirrer consistently agitated the mixture at 25 °C for 80 min. A water bath was used to keep the temperature constant. The pH of the solution was adjusted with NaOH or HCl solution by using a PHS-3C pH-meter (Leici, China) with a combined pH electrode. The amount of adsorbed dye on CTNS versus time was determined from absorbency on a Specord 50 UV-vis spectrophotometer (Analytic Jena) at a 665 nm wavelength, where the maximum absorbency occurred.

All the experiments repeated for two times, and each experimental point was an average of two independent adsorption tests.²⁹

Results and discussion

TEM analysis

Fig. 1 and 2 depict the morphology and structure of CdS quantum dots, TiO₂ nanobelts and typical CTNS sample. Fig. 1a and b, illustrate the size of CdS quantum dots is about 5 nm, and its uniform dispersity without obvious aggregation. The HRTEM image shows that CdS quantum dots exhibit integrity crystal structure long with the lattice spacing parallel to the top faces of CdS is 0.24 nm, corresponding to the (102) planes of anatase CdS. Fig. 1c and d show the TEM images of TiO₂ nanobelts. The integrity crystal structures also can be witnessed clearly with lattice spacing 0.35 nm, parallel to the top faces of TiO₂, which corresponding to (110) plane of anatase TiO₂.

Fig. 1 TEM images of CdS (a and b) and TiO₂ (c and d).

Fig. 2 TEM images of CTNS-1 (a and b) and CTNS-3 (c and d).

Fig. 2a and b show that CTNS-1 retains nanobelt-like morphology with the width of 25 nm, which is smaller comparing with its precursor TiO₂ nanobelts, indicating the formation of new nanobelts during hydrothermal treatment, namely titanate nanobelts. Fig. 2c and d exhibit the TEM image of CTNS-3 that owns the highest removal efficiency for MB and NR dyes. The CTNS-3 also possesses nanobelts, but the length and width of nanobelts turned more smaller. The HRTEM image shows abundant mesopores in CTNS-3. CdS quantum dots could not be found, and the reasons will be discussed later.

XRD analysis

To understand the structure of the CdS, TiO₂ and CTNS, XRD is conducted and results are shown in Fig. 3. The CdS diffraction peaks at 25.2°, 26.7°, 28.5°, 36.2°, 44.09° and 48.1° are consistent with the (100), (002), (101), (102), (110), (103) and (112) planes (JCPDS no. 00-001-0783). The diffraction peaks at 24.9°, 28.79° and 48.53° corresponding to (110), (002) and (020) planes (JCPDS no. 00-046-1237), are attributed to TiO₂ nanobelts, and the diffraction peak at 12.1° is ascribed to Na_0.57Ti₂O₄. The presence of Na_0.57Ti₂O₄ can be due to unremoved residues of Na element during the washing process. The blue line and magenta line show the XRD pattern of typical samples of CTNS-1 and CTNS-3. The main crystal phase is consistent with Na₂Ti₉O₁₉, and the diffraction peaks at 25.0°, 29.4°, 34.4°, 48.3°, 49.4°, 52.5° and 60.8° are corresponding to the (110), (−401), (−404), (008), (512), (−223) and (−803) planes (JCPDS no. 01-073-2256). For CTNS-1, the diffraction peak at 38.4° belongs to the peaks of anatase TiO₂ (JCPDS no. 01-083-2243) which can be produced by the incomplete reaction of TiO₂ and NaOH. For CTNS-3, the weak diffraction peaks credited to CdTiO₃ at 34.1° and 60.5° are identified, confirming the formation of insoluble CdTiO₃.

Fig. 3 XRD patterns of CdS (a), TiO₂ (b), CTNS-1 (c) and CTNS-3 (d).

XPS analysis

To identify the oxidation states and chemical composition, CTNS-3 was chosen for analysis and further evidences were integrated by XPS analysis. Fig. 4 illustrates the high-resolution XPS spectra of Ti 2p, O 1s, Cd 3d, and S 2p in CTNS-3, respectively. The Ti 2p spectrum can be curve-fitted with four peak components at binding energies of 458.1, 458.8, 461.4 and 464.5 eV. Furthermore, the peak signals at 458.8 and 464.5 eV correspond to TiOS and TiO₂ respectively, and the peaks centered at 458.1 and 461.4 eV are ascribed to the Ti–O bond in the Na₂Ti₉O₁₉. The O 1s spectrum can be curve-fitted with three peak components at bonding energies of 530.02, 531.49 and 535.04 eV. The peaks at 530.02 and 531.49 eV can be attributed to the Ti–O bond in Na₂Ti₉O₁₉ and the Cd–O bond in CdTiO₃, consistent with the XRD analysis of CTNS-3 (Fig. 3). The peak at 535.04 eV can be attributed to the O–H bond in H₂O on the sample surface. The Cd 3d spectrum splits into three peak components signals at bonding energies of 404.6, 405.5 and 408.0 eV, exposing Cd²⁺ signal in CdS centered at 405.5 eV. And the peaks at 404.6 and 408.0 eV can be attributed to the Cd–O bond in the CTNS-3, corresponding to the analysis results of O 1s spectra. The S 2p spectrum can be curve-fitted with four components signals at bonding energies of 161.29, 164.39 and 165.58 eV, analogous to CdS, Ti–O–S and S–O bond in the CTNS-3 respectively. The peak intensities of different elements in XPS spectra authenticate the higher amounts of titanate precursor but a little CdS. In addition, XPS analysis results indicating the atomic ratio of Cd and S 0.72 : 0.42, further endorse the reaction of CdS and titanate. From the XPS analysis, it can be concluded that the S and Cd in the CdS have inserted into the titanate and only amount of CdS as particles are remaining. These analysis results also correlate to the XRD analysis.

Fig. 4 XPS spectrum of Ti 2p (a), O 1s (b), Cd 3d (c), and S 2p (d) peaks of CTNS-3.

So, CdS nano-particles could not be found in the Fig. 3, which are mainly due to two reasons. One is that the size of CdS was very small around 5 nm, and easily covered by titanate nanobelts; and the other is that the inserted amount of CdS was very small and most CdS nano-particles turned into CdTiO₃ or the other compounds which has been verified by XRD and XPS, so that the amounts of CdS quantum dots in the composites was further decreased.

BET analysis

Fig. 5 shows nitrogen adsorption/desorption isotherms of TiO₂, CTNS-3 and CTNS-4. The isotherms exhibit type IV behavior, indicating mesoporous morphology. CTNS-3 and CTNS-4 possess much larger surface area of 56.8 m² g⁻¹ and 34.4 m² g⁻¹ respectively than pure TiO₂ nanobelts with 26.9 m² g⁻¹, which is attributed to the addition of trace amount of CdS quantum dots into composites. The insertion of small CdS quantum dots improves the mesoporosity, enhancing the surface area (Fig. 2 and 5). However, the further increment in the ratio of CdS to TiO₂ corresponding to CTNS-4, the surface area decreases compared with CTNS-3. The main contributor to the surface area of composites is titanate nanobelts. And the proportion of CdS quantum dots is very high, corresponding to less titanate nanobelts content of composites, resulting in the decrease of surface area of composites (CTNS-4). So there must be a right ratio of CdS to TiO₂ to improve both the mesoporosity and surface area of composites. The improved surface area of CTNS can contribute to enhance the adsorption capacity. The pore-size distributions calculated by the Barret–Joyner–Halender (BJH) method from the adsorption branches of the nitrogen isotherms display deferent character for the samples. It can be seen that the pore size of CTNS-3 and CTNS-4 are larger than pure TiO₂ nanobelts.

Fig. 5 Nitrogen adsorption/desorption isotherm and corresponding pore size distribution of mesoporous TiO₂ nanobelts (a), CTNS-3 (b) and CTNS-4 (c).

Zeta potential analysis

Zeta potential measurements are done to verify the charge of CTNS surface, and the corresponding results are shown in the Fig. 6. It can be seen that CTNS-3 surface possess the highest negative charge, while the charge of CTNS-1 surface is positive. The charge of material surface closely related to the adsorption process of dyes. In the paper, the more negatively charged surface of CTNS, the more dyes can be adsorbed due to the strong electrostatic attraction between the cationic dyes (MB and NR) and the negatively charged surface of CTNS.

Fig. 6 Zeta potentials of CTNS with deionized water as the dispersant.

Adsorption studies

Effect of initial dye concentration, temperature and pH on adsorption process. Fig. 7a exhibits adsorption capacity of CTNS at the initial MB concentration of 100 mg L⁻¹, 25 °C and pH = 6.47. It can be seen that the CTNS-3 exposes the maximum adsorption capacity of 1644 mg g⁻¹ for MB. The adsorption capacity of MB is very low, about 426 mg g⁻¹ and 340 mg g⁻¹ for CTNS-1 (the pure TiO₂ nanobelts) and CTNS-2 respectively. CTNS-3 possesses highest adsorption effect for MB. BET results have verified that CTNS-3 (CdS : TiO₂ = 1 : 100) owns an admirable mesoporous structure with maximal surface area of 56.79 m² g⁻¹. The improved surface area and the mesoporous structures are two key factors contributing towards MB adsorption. In addition, the more negatively charged surface of CTNS-3 (Fig. 6) also can be a factor that enhancing the improvement of dyes adsorption capacity of CTNS-3. Therefore, CTNS-3 has been chosen as a typical sample for the following adsorption studies.

Fig. 7 The effect of CTNS (a), the initial MB concentration (b), temperature (c) and pH (d) on the adsorption capacity of MB.

The effects of initial concentrations of MB (50 mg L⁻¹, 75 mg L⁻¹, and 100 mg L⁻¹) on the adsorption capacity of CTNS-3 at pH = 6.47 and 25 °C have been studied and shown in Fig. 7b. It can be noted that the adsorption capacity for MB increases with augmenting MB concentration, as the higher initial MB concentration provides higher driving force to overcome the mass transfer resistance.² Fig. 7c shows the effect of temperature on the MB adsorption process at the initial MB concentration of 100 mg L⁻¹ and pH = 6.47. Results indicate that higher temperature improves the adsorption rates and the equilibrium absorption capacity. Adsorption capacity of MB adsorbed increases from 1632.2 to 1850.9 mg g⁻¹ (Table 1) as the temperature increases from 25 °C to 65 °C, indicating an endothermic process. The higher temperature promotes the higher adsorption for MB due to two reasons. First, higher temperature accelerates the rate of diffusion of MB from the external boundary layer to the internal pores of adsorbent particles. Second, the increase of temperature promotes the increase of equilibrium capacity of the adsorbent for MB.³⁰ Fig. S1a† shows the color change of MB solution upon different initial temperatures at the initial MB concentration of 100 mg L⁻¹. It can be seen that superb adsorption performances and removal yields are acquired by the color change of MB solution at the higher initial temperature. Almost whole MB can be adsorbed by CTNS-3 at 65 °C with the initial MB concentration of 100 mg L⁻¹. Fig. 7d shows the effect of initial pH on the MB adsorption at the initial MB concentration of 100 mg L⁻¹ and 25 °C. The results show that adsorption capacity of MB are 1625 mg g⁻¹, 1632 mg g⁻¹ and 1480 mg g⁻¹ (Table 1) at pH = 3.00, 6.47 and 11.00 respectively. Generally, pH is one of the most significant factors controlling the adsorption of MB onto adsorbent, and with the increase of pH, the adsorption capacity usually is expected to increase.^3,31 In our studies, the adsorption capacities at different pH are contrary, as the equilibrium absorption capacity is almost same. As the equilibrium absorption capacity are very high (≥1480.00 mg g⁻¹) no matter the solution is acid, neutral or alkaline, the CTNS-3 can be applied for MB removal from the effluents. It can also be found a rapid uptake of dyes in the Fig. 7, which can be due to the chemical adsorption.³²

Table 1 The equilibrium absorption capacities and adsorption yields for CTNS-3 obtained at different temperature

Dyes	T (°C)	C0 (mg L^−1)	pH	qe (mg L^−1)	Adsorption yields (%)
MB	25	50	6.47	978	97.77
	25	75	6.47	1354	90.23
	25	100	6.47	1632	81.61
	25	150	6.47	1962	67.77
	25	200	6.47	2064	52.60
	45	100	6.47	1783	89.14
	65	100	6.47	1851	92.54
	25	100	3.00	1625	81.25
	25	100	11.00	1480	74.00
NR	25	100	4.49	1777	88.86
	25	150	4.49	2755	91.83
	25	200	4.49	3742	93.54
	45	100	4.49	1747	87.36
	65	100	4.49	1598	79.89
	25	100	7.00	1879	93.96
	25	100	11.00	1523	76.15

---

Fig. S2a† shows the effect of initial concentration of NR (50 mg L⁻¹, 75 mg L⁻¹, and 100 mg L⁻¹) on the adsorption capacity for CTNS-3 at pH = 4.49 and 25 °C. It is shown by the results that the adsorption capacity for NR also rises with the increase of the concentration of NR solution from 100 to 200 mg L⁻¹, and the maximum adsorption capacity and removal yields can reach 3741.8 mg g⁻¹ and 93.54% (Table 1). Fig. S2b† shows the effect of temperature on the NR adsorption process at the initial NR concentration of 100 mg L⁻¹ and pH = 4.49. Contrary to the adsorption process of MB, the absorption capacity of NR decreases with the increase of temperature as the higher temperature results in desorption of NR from the CTNS-3. Fig. S3† expresses the color change of NR solution with different initial temperature at the initial NR concentration of 100 mg L⁻¹. In the absence of CTNS-3, the color of solution is very dark. After adding CTNS-3, the NR can be removed efficiently at the lower temperature (25 °C). Fig. S2† shows the effect of pH on the NR adsorption process at the initial NR concentration of 100 mg L⁻¹ and 25 °C. Results indicate that the maximum absorption capacity is obtained at pH = 7.10, and all the equilibrium absorption capacity are also very high (≥1523.04 mg g⁻¹) no matter the solution was acid, neutral or alkaline, which is similar to the adsorption process of MB. Table 1 summaries the equilibrium absorption capacities and adsorption yields obtained at different initial concentration of MB and NR, temperature and pH. Results show that CNTS-3 has higher adsorption capacity and removal yields for both MB and NR. Hence, CTNS-3 will has great application prospect in removing the cationic dyes.

Mechanism of adsorption

For detailed information of MB adsorption isotherms and kinetics on CTNS-3, please see the ESI (ESI-3 Adsorption isotherms and ESI-4 Adsorption kinetics†). The analysis results adsorption isotherms and kinetics indicate that the whole adsorption process of MB for CTNS-3 is attributed to the Freundlich adsorption model (Fig. S4†) and the second-order kinetic model belong to the actual adsorption process of MB on CTNS-3 (see Fig. S5, S6 and Table S2†).

MB is also as model cationic dye for the study of adsorption mechanism. Previous reports have indicated that the mechanism of MB adsorption onto mesoporous adsorbents followed five steps:^33,34 first, the MB migrates from the bulk of the solution to the adsorbent surface. Second, MB diffuses on the adsorbent surface through the boundary layer, i.e. external diffusion. Third, the active sites of the adsorbent adsorb the MB. Fourth, MB diffuses into the interior pores of the adsorbent, i.e. intra-particle diffusion. Last step is the adsorption of adsorbent for MB. In all the five steps, the intraparticle diffusion generally is the rate controlling step. The intraparticle diffusion model is as below:

q_t = k_dift_1/2 + C (1)

where C is constant, k_dif is the intraparticle diffusion rate constant (mg min^−1/2 g⁻¹). Fig. S7† shows the intraparticle diffusion plots that q_t versus t^1/2. The relevant parameters of intraparticle diffusion model are calculated and listed in Table S3.† Results show that there is certain linear relationship between the values of q_t and t^1/2. But the linear relationships in Fig. S7† are not very good due to the lower values of R² which is from 0.8100 to 0.9796. In addition, all the linear plots did not pass through the origin, which indicated that the intraparticle diffusion is not the only rate controlling step. All the results indicated that the adsorption process followed the intraparticle diffusion model, but also existing other adsorption process, such as chemical adsorption.

Table 2 shows a comparison of the maximum adsorption capacity of MB and NR on different materials. Compared with the other materials, such as activated carbon, graphene oxide, nano-materials (including mesoporous materials) or their ramifications and composites, the adsorption performance of our materials (CTNS-3) is the best, and the maximum adsorption capacity and adsorption rate in the present work are far more than that of other materials. In addition, it is known that TiO₂ is an environmentally friendly nano-materials.³⁵ For CTNS-3, only containing little amounts of CdS, and most of CdS have bonded with titanate and can not be released in the solution. So, CTNS-3 is one of adsorbents which is safe and environment friendly. However, for the other CTNS, their adsorption capacity are low; on the other hand, the increase of ratio of CdS to TiO₂ can results in threatening of environment due to the toxicity of cadmium.

Table 2 Comparison of the maximum adsorption capacity of MB and NR onto different materials

Dyes	Materials	Maximum adsorption capacity (mg g^−1)	Equilibrium adsorption time (min)	References
MB	CTNS-3	2444.9	30	Present work
	Commercial activated carbon	298.4	120	1
	Carbon nanotubes	64.7	60	36
	Lotus leaf	221.7	50	37
	Graphene oxide	714.0	150	38
	Fe3O4@C	44.4	150	39
	Polydopamine microspheres	90.7	100	40
	Mesoporous silica	462.0	200	32
NR	CTNS-3	3741.8	30	Present work
	Peanut husk	37.5	90	41
	Cottonseed hull	176.9	240	42
	Modified hectorite	393.7	100	43
	Halloysite nanotubes	65.5	30	44
	Rice husk	32.4	>300	45
	Zn3[Co(CN)6]2·nH2O nanospheres	285.7	60	46

---

Conclusions

A series of mesoporous CTNS nanobelts were prepared successfully by inserting CdS into TiO₂ nanobelts. The sample of CTNS-3 owned the exceptionally higher adsorption performance for the typical cationic dyes (MB and NR). The adsorption capacity of CTNS-3 for MB and NR could reach 1850.9 mg g⁻¹ and 3741.8 mg g⁻¹ respectively, which was remarkably higher than the other adsorbents, and their corresponding adsorption yields also could reach 92.54% and 93.54%. Hence, CTNS-3 will have great application prospect for removing the organic dyes in the effluent treatment. The mechanisms of adsorption for CTNS-3 could include physical adsorption and chemical adsorption.

Acknowledgements

This research was financially supported by Shenzhen strategic emerging industry development special fund project (Project No. JCYJ20130401144744190), the innovation foundation of Huazhong University of Science and Technology Innovation Institute (No. 2015TS150, 2015ZZGH010). We acknowledge the support of the Analytical and Testing Center of the Huazhong University of Science and Technology for TEM, XPS and XRD measurements.

Notes and references

1. N. Kannan and M. M. Sundaram, Dyes Pigm., 2001, 51, 25–40.
2. D. Özer, G. Dursun and A. Özer, J. Hazard. Mater., 2007, 144, 171–179.
3. S. Karagöz, T. Tay, S. Ucar and M. Erdem, Bioresour. Technol., 2008, 99, 6214–6222.
4. B. Hameed, A. M. Din and A. Ahmad, J. Hazard. Mater., 2007, 141, 819–825.
5. B. Hameed, A. Ahmad and K. Latiff, Dyes Pigm., 2007, 75, 143–149.
6. P. Luo, Y. Zhao, B. Zhang, J. Liu, Y. Yang and J. Liu, Water Res., 2010, 44, 1489–1497.
7. M. Rafatullah, O. Sulaiman, R. Hashim and A. Ahmad, J. Hazard. Mater., 2010, 177, 70–80.
8. S.-A. Ong, E. Toorisaka, M. Hirata and T. Hano, Environ. Chem. Lett., 2007, 5, 95–99.
9. K. Dutta, S. Mukhopadhyay, S. Bhattacharjee and B. Chaudhuri, J. Hazard. Mater., 2001, 84, 57–71.
10. J.-S. Wu, C.-H. Liu, K. H. Chu and S.-Y. Suen, J. Membr. Sci., 2008, 309, 239–245.
11. G.-S. Li, D.-Q. Zhang and J. C. Yu, Environ. Sci. Technol., 2009, 43, 7079–7085.
12. H. Ma, Q. Zhuo and B. Wang, Chem. Eng. J., 2009, 155, 248–253.
13. A. Jain, V. Gupta and A. Bhatnagar, J. Hazard. Mater., 2003, 101, 31–42.
14. Y.-S. Ho and G. McKay, Process Biochem., 2003, 38, 1047–1061.
15. A. L. Cazetta, A. M. Vargas, E. M. Nogami, M. H. Kunita, M. R. Guilherme, A. C. Martins, T. L. Silva, J. C. Moraes and V. C. Almeida, Chem. Eng. J., 2011, 174, 117–125.
16. D. Kavitha and C. Namasivayam, Bioresour. Technol., 2007, 98, 14–21.
17. S.-Y. Mak and D.-H. Chen, Dyes Pigm., 2004, 61, 93–98.
18. H. L. Parker, RSC Adv., 2012, 2, 8992–8997.
19. K. Foo and B. Hameed, Chem. Eng. J., 2011, 166, 792–795.
20. J. M. Hong, B. Lin, J. S. Jiang, B. Y. Chen and C. T. Chang, J. Ind. Eng. Chem., 2014, 20, 3667–3671.
21. M. Asif, A. Aziz, A. Q. Dao, A. Hakeem, H. Wang, S. Dong, G. Zhang, F. Xiao and H. Liu, Anal. Chim. Acta, 2015, 898, 34–41.
22. A. Franco, M. Neves, M. R. Carrott, M. Mendonça, M. Pereira and O. Monteiro, J. Hazard. Mater., 2009, 161, 545–550.
23. C. Burda, Y. Lou, X. Chen, A. C. Samia, J. Stout and J. L. Gole, Nano Lett., 2003, 3, 1049–1051.
24. Q. Xiang, J. Yu and M. Jaroniec, J. Am. Chem. Soc., 2012, 134, 6575–6578.
25. J. Zhu, D. Yang, J. Geng, D. Chen and Z. Jiang, J. Nanopart. Res., 2008, 10, 729–736.
26. P. Connor and A. J. McQuillan, Langmuir, 1999, 15, 2916–2921.
27. M. Rasmussen, L. Molina and B. Hammer, J. Chem. Phys., 2004, 120, 988–997.
28. Z.-R. Tang, X. Yin, Y. Zhang and Y.-J. Xu, Inorg. Chem., 2013, 52, 11758–11766.
29. X. Y. Yang and B. Al-Duri, Chem. Eng. J., 2001, 83, 15–23.
30. Z. Al-Qodah, Water Res., 2000, 34, 4295–4303.
31. M. Doğan, M. Alkan, A. Türkyilmaz and Y. Özdemir, J. Hazard. Mater., 2004, 109, 141–148.
32. J. R. Deka, Y. H. Lin and H. M. Kao, RSC Adv., 2014, 4, 49061–49069.
33. S. Kaur, S. Rani, R. Mahajan, M. Asif and V. K. Gupta, J. Ind. Eng. Chem., 2015, 22, 19–27.
34. Z. Chen, J. Zhang, J. Fu, M. Wang, X. Wang, R. Han and Q. Xu, J. Hazard. Mater., 2014, 273, 263–271.
35. W. Ming-Jie, Z. Luzheng, L. Linghong, Z. Yudan, K. E. Gubbins and L. Xiaohua, Phys. Chem. Chem. Phys., 2012, 14, 16536–16543.
36. Y. Yao, F. Xu, M. Chen, Z. Xu and Z. Zhu, Bioresour. Technol., 2010, 101, 3040–3046.
37. X. Han, W. Wang and X. Ma, Chem. Eng. J., 2011, 171, 1–8.
38. S.-T. Yang, S. Chen, Y. Chang, A. Cao, Y. Liu and H. Wang, J. Colloid Interface Sci., 2011, 359, 24–29.
39. Z. Zhang and J. Kong, J. Hazard. Mater., 2011, 193, 325–329.
40. J. Fu, Z. Chen, M. Wang, S. Liu, J. Zhang, J. Zhang, R. Han and Q. Xu, Chem. Eng. J., 2015, 259, 53–61.
41. R. P. Han, P. Han, Z. H. Cai, Z. H. Zhao and M. S. Tang, J. Environ. Sci., 2008, 20, 1035–1041.
42. Q. Zhou, W. Q. Gong, C. X. Xie, D. J. Yang, X. Q. Ling, X. A. Yuan, S. H. Chen and X. F. Liu, J. Hazard. Mater., 2011, 185, 502–506.
43. D. Y. Yue, Y. Jing, J. Ma, C. L. Xia, X. J. Yin and Y. Z. Jia, Desalination, 2011, 267, 9–15.
44. P. Luo, Y. F. Zhao, B. Zhang, J. D. Liu, Y. Yang and J. F. Liu, Water Res., 2010, 44, 1489–1497.
45. W. H. Zou, P. Han, Y. L. Lia, X. Liu, X. T. He and R. P. Han, Desalin. Water Treat., 2009, 12, 210–218.
46. W. Q. Wang, C. C. Li, J. L. Yao, B. Zhang, Y. T. Zhang and J. D. Liu, J. Mol. Liq., 2013, 184, 10–16.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra04653c

‡ Hongwei Liu and Lu Zhang contributed equally to this work.

---