Introduction

With the rapid development of industry, there is more and more concern about toxic dyes and heavy metal ions in untreated waste water from industrial production processes1. Most dyes and their intermediates have teratogenic, carcinogenic or mutagenic effects and high biological toxicity. Some dyes even become chemicals for carcinogenicity tests. Meanwhile, due to their wide application, the released dyes and dye intermediates have caused serious damages to the external environment, which are very difficult to control2. As one of the most common heavy metal ions, too much Cu2+ in the human body will cause gastrointestinal problems, liver and kidney damage, nausea, hair loss, severe headache and even death3. Therefore, how to remove organic dyes and heavy metal ions in wastewater has become a hot topic in environmental protection. There are many removal methods, such as adsorption method4, ion exchange method5 and chemical precipitation method6. Among these methods, the adsorption method is widely used because of its high adsorption efficiency, simple operation and recoverability7.

In view of the adsorption method, scholars have studied the adsorption performance of various adsorbent materials for contaminants. Tang et al.4 studied the simultaneous adsorption of atrazine and Cu2+ by magnetic carbon nanotubes. Asuha et al.7 investigated the adsorption performance of TiO2 for methyl orange and Cr(VI). Among these adsorbent materials, titanium dioxide is very promising for environment-purifying applications since ion doping and immobilization8,9,10,11. However, titanium dioxide is present in the form of powder and is difficult to be separated from aqueous solution for recovery and reuse. Due to its high mechanical strength, wide pore size distribution and high adsorption capacity, granular activated carbon can be effectively used as a carrier of TiO2. The combination of granular activated carbon and titanium dioxide can accelerate the settling rate and enhance the adsorption capacity, making up for the shortcomings of TiO2 and thereby allowing wide application in wastewater treatment12. Most of the previous literature focused on the removal of contaminants with this material in a single system12,13,14. However, the application of this material in more complex multivariate systems is rarely reported. In this paper, Cu2+, MO and methylene blue (MB) were selected to create a mixed system of heavy metals and dyes. The adsorption capacity of TiO2@GAC for Cu2+ and dyes was investigated. The adsorption performance of TiO2@GAC under the influence of pH, initial concentration of dyes/Cu2+ and time was studied in detail.

Experimental Sections

Materials and Instruments

Ethyl titanate, granular activated carbon, anhydrous ethanol, acetic acid, hydrochloric acid, methyl orange and metallic copper (Sinopharm Chemical Reagent Co., Ltd., analytical reagent); methylene blue (Guangdong Xilong Scientific Co., Ltd., analytical reagent). Secondary deionized water was used for all experiments.

Fourier transform infrared spectroscopy (Bruker, Germany); XRD-6000 diffractometer (Cu2+ Kα radiation, λ = 0.15406 nm, PANalytical, Holland); scanning electron microscope (SSX-550, Shimadzu Corporation); UV visible spectrophotometer (model 712, Shanghai Third Analytical Instrument Factory); atomic absorption spectrometer (TAS-990, Beijing Persee General Instrument Co., Ltd.).

Pretreatment of granular activated carbon

The granular activated carbon (with an average particle size of 3 mm) was first washed with deionized water until the washings were colorless, so as to remove the ash. Then it was soaked in nitric acid for 24 h to remove organic matter and other impurities. Finally, it was washed with deionized water until the pH was neutral, and dried in a vacuum oven at 80 °C.

Preparation of supported titanium dioxide

18 mL of titanium tetrabutyl titanate, 45 mL of anhydrous ethanol and 3 mL of acetic acid were mixed to prepare solution A; 45 mL of anhydrous ethanol and 8 mL of deionized water (adjusted to pH 2–3 with 0.1 mol nitric acid) was mixed to prepare solution B. 5 g of granular activated carbon was weighed and added to solution A. Solution B was slowly added to solution A with a separatory funnel under vigorous stirring. After the addition of solution B, the mixture was stirred to form a sol and was then allowed to stand for 2 days to form a jelly-like gel. The gel was dried in a vacuum oven at 90 °C, calcined at 250 °C for 1 h in an air atmosphere and then calcined at 600 °C for 2 h in a nitrogen atmosphere in a tube furnace to obtain TiO2@GAC.

Analysis of TiO2 on TiO2@GAC

The method for analyzing TiO2 on GAC was described in detail by El-Sheikh et al.15. 0.1 g dried TiO2@GAC sample were weighed (±0.1 mg) in a Teflon tube, and 3.0 ml 18.0 M H2SO4, 0.04 g CuSO4 and 0.35 g K2SO4 were added to the tube. The Teflon tube with sample was digested in a microwave oven for 5 min. Then the tube was added in 7 ml water and centrifuged at 3000 rpm to remove residual carbon. The supernatant was mixed with 1.00 ml 30% H2O2, and diluted with water to 10.00 ml. Absorbance of the solution was detected at 410 nm.

Adsorption experiment

10 mL of a single or binary solution with a certain concentration was added to a centrifuge tube. After the addition of 10 mg of TiO2@GAC particles, the tube was centrifuged. The supernatant was then removed and the concentration was measured. In the experiment, the absorbance of MO and MB was measured with a UV-Vis spectrophotometer (the maximum absorption wavelength of MO was 464 nm and the MB was 664 nm). The concentration of Cu2+ was measured with an atomic absorption spectrophotometer. The adsorption rate was calculated with the following formula:

$$\eta =\frac{{C}_{0}-{C}_{t}}{{C}_{0}}\times 100 \% $$
(1)

where C0 represents the concentration before adsorption and Ct represents the concentration after adsorption.

Results and Discussion

Figure 1 shows the infrared absorption spectra (FT-IR) of TiO2@GAC and GAC. It can be observed from Fig. (1b) that GAC has four main absorption bands in the wavelength range of 4000–400 cm−1. The absorption peaks at 3400 cm−1 and 1600 cm−1are due to the O-H stretching vibrations. The absorption peak at 1726 cm−1 is due to the C=O stretching vibration, while the absorption peak at 1060 cm−1 is due to skeletal stretching vibrations16. In Fig. (1a), the absorption bands of TiO2@GAC in the range of 400–800 cm−1 were different from those of GAC. This is caused by the Ti-O stretching vibrations. The absorption band at 1060 cm−1 disappears because GAC was covered by TiO2. The FT-IR analysis gives preliminary evidence that titanium dioxide has been loaded on granular activated carbon.

Figure 1
figure 1

FT-IR spectra of TiO2@GAC (a) and GAC (b).

Figure 2 shows the SEM and EDS images of the prepared supported titanium dioxide. It can be seen from Fig. 2(a) that the surface of GAC features a mesoporous structure and has been loaded with TiO2. It can be known from Fig. 2(b) that the EDS spectrum only contains the element C. Figure 2(c) suggests that three elements (Ti, C and O) are present in the EDS spectrum. Therefore, it can be concluded that TiO2 has been successfully loaded on GAC. The TiO2 contents on synthetic TiO2@GAC were in the range from 43.4 mg/g to 45.1 mg/g TiO2@GAC.

Figure 2
figure 2

SEM (a) and EDS (b and c) images of TiO2@GAC.

Figure 3 shows the XRD analysis of TiO2@GAC. As shown in Fig. 3, in the 2θ range of 10° to 80°, there are six characteristic peaks of TiO2, which are 25.2°, 37.6°, 47.8°, 53.8°, 54.9° and 62.7°, respectively. According to JCDPS Card #16-629, they are the characteristic diffraction peaks of (101), (004), (200) (105), (211) and (204) planes of anatase TiO2, respectively17. There is a significant peak at 43.5°, which is a characteristic peak of activated carbon. This is probably because part of activated carbon has not been fully loaded. However, the above analysis already shows that TiO2 has been loaded onto GAC.

Figure 3
figure 3

XRD patterns of TiO2@GAC.

Effect of pH

Figure 4 shows the effect of pH on the adsorption performance of TiO2@GAC and GAC. It can be seen from Fig. 4 that, with constant changes in pH, the adsorption rate of TiO2@GAC for dyes and Cu2+ is higher than that of GAC. It can be concluded that the adsorption performance of TiO2@GAC for dyes and Cu2+ is better than that of GAC. It can be seen from Fig. 4(a) that, with the decrease of pH, the adsorption of MO on TiO2@GAC is not conducive to the adsorption of MB. With the increase of pH (1–10), the adsorption rate of MO decreases from 95.55% to 48.13%, while the adsorption rate of MB increases from 42.50% to 90.54%. These results can be explained by the theory of isoelectric point (pHpzc). According to literature, the pHpzc of TiO2@GAC is about 6.018.

Figure 4
figure 4

Effect of pH on the adsorption of MO, MB (a) and Cu2+ on TiO2@GAC and GAC (b).

When the pH value of the solution is less than the pHpzc, the surface of TiO2@GAC is positively charged (TiOH2+). On the contrary, when the pH value of the solution is greater than the pHpzc, the surface of TiO2@GAC is negatively charged (TiO). Under acidic conditions, the MO molecule features a quinone structure with its sulfonate terminal negatively charged, facilitating its adsorption on the negatively charged surface of TiO2@GAC; while under alkaline conditions, the MO is negatively charged, resulting in an electrostatic repulsion toward the negatively charged TiO2@GAC, which hinders the adsorption of MO. In the case of MB, its molecular structure is positively charged. Under alkaline conditions, the negatively charged TiO2@GAC strongly adsorbed the positively charged MB, accelerating the removal of MB; while under acidic conditions, there is an electrostatic repulsion between the positive charges on the surface of TiO2@GAC and the positive charges on the MB molecule, which becomes one of the causes to the decreased removal efficiency. In addition, the decrease in the removal efficiency of MB under acidic conditions may also be due to the competition between H+ and MB on TiO2@GAC19.

It can be seen from Fig. 4 (b) that the adsorption rate of TiO2@GAC for Cu2+ increases with the increase of pH. The adsorption of Cu2+ can also be explained by pHpzc. As the pH increases, the Zeta potential of TiO2@GAC decreases. Because of the electrostatic attraction, the negatively charged TiO2@GAC (TiO) (pH > 6) is conducive to the adsorption of Cu2+. Similarly, the positively charged TiO2@GAC (TiOH2+) (pH < 6) is not conducive to the adsorption of Cu2+. Furthermore, coprecipitation of Cu2+ occurs when the pH is higher than 64. Therefore, when the pH value is in the range from 6 to 10, both adsorption and coprecipitation contribute to the significant increase of the removal efficiency of Cu2+, where coprecipitation plays a leading role. For this reason, a pH of 6 should be used as the best experimental condition in future studies. Figure 4 also shows that adsorption capacities of TiO2@GAC were higher than those of AC for dyes and Cu2+. Previous work showed that mesoporous TiO2 was an excellent adsorbent for dyes and heavy metal, with higher adsorption capacities for dyes and heavy metal7 than those of GAC or TiO2@GAC. Nevertheless, mesoporous TiO2 is difficult to be separated from aqueous solution for recovery and reuse. The combination of GAC and TiO2 can make up for the shortcomings of TiO2 and thereby allowing wide application in wastewater treatment12.

Effect of time on adsorption

Figure 5 depicts the effect of time on adsorption of MO (MB) and Cu2+ on TiO2@GAC in single systems. The time for MO and MB to reach the adsorption equilibrium is 4 h, and the time for Cu2+ to reach the adsorption equilibrium is 5 h. At the initial stage, the adsorption rates of all the three substances increase rapidly, which may be due to the fact that there are abundant adsorption sites on TiO2@GAC for the adsorption of dyes and Cu2+. With the lapse of time, more dye molecules and Cu2+ are adsorbed on the surface of TiO2@GAC, resulting in less available sites. Meanwhile, the concentration of dyes and Cu2+ in the solution also decreases. Therefore, the adsorption effect is reduced.

Figure 5
figure 5

Effect of time on adsorption of MO, MB and Cu2+ on TiO2@GAC (a) and Fit of kinetic data to pseudo-second-order model in single component systems (b).

Figure 6 depicts the effect of time on adsorption of MO (MB) and Cu2+ on TiO2@GAC in binary systems. In the binary system with MO and Cu2+, the adsorption equilibrium times of MO and Cu2+ are 3 h and 5 h, respectively. Both substances show a higher adsorption rate compared with single systems. Furthermore, the increase of the adsorption rates during the initial stage is also significantly faster compared with single systems. This may be due to the synergistic effect between the positively charged Cu2+ and the negative charged MO. In the binary system with MB and Cu2+, the adsorption equilibrium times of MB and Cu2+ are both 5 h. Both substances show a lower adsorption rate compared with single systems. This may be due to the competition between the positively charged Cu2+ and the positively charged MB.

Figure 6
figure 6

Effect of time on the adsorption of MO, MB and Cu2+ on TiO2@GAC (a) and Fit of kinetic data to pseudo-second-order model in binary component systems (b).

The pseudo-first-order model and pseudo-second-order model are built to describe the adsorption kinetics. The equations of the two models are as follows20,

Pseudo-first-order model,

$$\mathrm{ln}({{\rm{q}}}_{{\rm{e}}}-{{\rm{q}}}_{{\rm{t}}})=\,\mathrm{ln}\,{{\rm{q}}}_{{\rm{e}}}-{{\rm{K}}}_{1}{\rm{t}}$$
(2)

Pseudo-second-order model,

$$\frac{{\rm{t}}}{{{\rm{q}}}_{{\rm{t}}}}=\frac{1}{{{\rm{K}}}_{2}{{\rm{q}}}_{{\rm{m}}}^{2}}+\frac{{\rm{t}}}{{{\rm{q}}}_{{\rm{m}}}}$$
(3)

where qm is the adsorption capacity (mg/g) at the equilibrium, qt is the adsorption capacity (mg/g) at time t, K1 is the adsorption equilibrium rate constant (h−1) of the pseudo-first-order model, and K2 is the adsorption equilibrium rate constant (g/(mg·h)) of the pseudo-second-order model.

Table 1 lists the kinetic model parameters for the adsorption of MO, MB and Cu2+ on TiO2@GAC in single systems and binary systems. According to the correlation (R2), MO, MB and Cu2+ in single systems and binary systems all comply with the pseudo-second-order kinetic model (Figs 5(b) and 6(b)). This indicates that the adsorption of MO, MB and Cu2+ on TiO2@GAC is a chemical adsorption process21.

Table 1 Pseudo-first-order, pseudo-second-order kinetics constants for MO, MB and Cu2+ adsorption in single and binary systems.

Effect of concentration on adsorption in single systems

Figure 7 depicts the effect of concentration on adsorption of dyes and Cu2+ on TiO2@GAC in single systems. It can be seen from Fig. 7 that, the adsorption rates of both dyes and Cu2+ on TiO2@GAC decrease with the increase of the initial concentration. When the adsorption time and the concentration of adsorbent are constant, the adsorption sites on the adsorbent surface decrease with the increase of the concentration of MO (MB) and Cu2+, thus reducing the adsorption rates.

Figure 7
figure 7

Effect of concentration on the adsorption of MO, MB (a) and Cu2+ (b) on TiO2@GAC in single component systems.

The Langmuir and Freundlich isotherm adsorption equations are used to process the experimental data. The linear equations of Langmuir22 (Eq. 4) and Freundlich23 (Eq. 5) isothermal models are as follows,

$$\frac{{{\rm{C}}}_{{\rm{e}}}}{{{\rm{q}}}_{{\rm{e}}}}=\frac{{{\rm{C}}}_{{\rm{e}}}}{{{\rm{q}}}_{{\rm{m}}}}+\frac{1}{{{\rm{q}}}_{{\rm{m}}}{{\rm{K}}}_{{\rm{L}}}}$$
(4)
$$\mathrm{ln}\,{{\rm{q}}}_{{\rm{e}}}=\,\mathrm{ln}\,{{\rm{K}}}_{{\rm{F}}}+\frac{1}{{\rm{n}}}\,\mathrm{ln}\,{{\rm{C}}}_{{\rm{e}}}$$
(5)

where qe is the equilibrium adsorption capacity per unit mass of TiO2@GAC for dyes and Cu2+, Ce is the equilibrium concentration, KL is the Langmuir equilibrium adsorption constant, qm is the maximum adsorption capacity per unit mass of TiO2@GAC, KF is the capacity coefficient and n is the intensity factor.

According to R2 in Table 2, the isothermal adsorption models of MO, MB and Cu2+ in single systems all comply with the Langmuir model, which indicates that the adsorption process of MO, MB and Cu2+ on TiO2@GAC is monomolecular adsorption24. The maximum adsorption capacities of MO, MB and Cu2+ are 32.36 mg/g, 25.32 mg/g and 23.42 mg.

Table 2 Adsorption isotherm constants for MO, MB and Cu2+ adsorption in single component systems.

Effect of concentration on adsorption in binary systems

It can be seen from Fig. 8(a) that, in the binary system with a constant concentration of MO, the adsorption rate of MO first increased and then decreased with concentration of Cu2+. In addition to the adsorption of TiO2@GAC itself, the increase in the adsorption rate of MO may also be due to the synergistic effect between positively charged Cu2+ and negatively charged MO. However, as the concentration of Cu2+ increased, the adsorbed MO may be replaced by Cu2+, causing decrease in removal rate of MO. Although the removal rate of MO decreased, the adsorption rate was still above 85%. However, in the presence of MO, with a constant concentration of Cu2+, the adsorption rate of Cu2+ TiO2@GAC remained almost unchanged with concentration of MO, as shown in Fig. 8(b). This was probably because Cu2+ entered the pores in activated carbon and got adsorbed earlier than organic substance MO with a higher molecular weight. Therefore, the effect of MO on the adsorption of Cu2+ was relatively small and almost negligible.

Figure 8
figure 8

Effect of initial concentration Cu2+ on adsorption MO (a) and effect of initial concentration of MO on adsorption Cu2+ (b).

It can be seen from Fig. 9 that, in the binary system consisting of MB and Cu2+, in the presence of Cu2+, with a constant concentration of MB, the adsorption rate of MB decreases with the increase of the concentration of Cu2+. This is because there is a competitive relationship between positively charged Cu2+ and positively charged MB. As the concentration of Cu2+ increases, the removal rate of MB decreases. In the presence of MB, the adsorption rate of Cu2+ also remains almost unchanged as the concentration increases. This is also probably because Cu2+ get adsorbed earlier than the organic matter MB with a higher molecular weight.

Figure 9
figure 9

Effect of initial concentration Cu2+ on adsorption MB (a) and effect of initial concentration of MB on adsorption Cu2+ (b).

Conclusions

Granular activated carbon-supported titanium dioxide particles were prepared with the sol-gel method and their adsorption performance for dyes and Cu2+ were studied. In both single systems and binary systems, the adsorption behaviors of MO, MB and Cu2+ by TiO2@GAC all complies with the pseudo-second-order kinetic model. In single systems, the adsorption isotherms of both dyes and Cu2+ on TiO2@GAC comply with the Langmuir model, which indicates that the adsorption process of TiO2@GAC is monomolecular chemisorption. The pH has significant effect on the adsorption of dyes and Cu2+. For Cu2+, coprecipitation also contributes and plays a leading role. In the binary system consisting of MO and Cu2+, the adsorption rate of MO increases first and then decrease while the adsorption rate of Cu2+ remains almost unaffected.

The increase in the adsorption rate of MO is due to the adsorption and the synergistic effect between Cu2+ and MO, while the unaffected adsorption rate of Cu2+ and the subsequent decrease in the adsorption rate of MO may be due to the preferential adsorption of Cu2+ on TiO2@GAC. Similarly, in the binary system consisting of MB and Cu2+, the adsorption rate of MB increases first and then decrease while the adsorption rate of Cu2+ remains almost unaffected. The increase in the adsorption rate of MB is due to the adsorption of TiO2@GAC, while the subsequent decrease in the adsorption rate of MB is due to the competition between Cu2+ and MB. The above experimental results provide a certain theoretical basis for the removal of dyes and heavy metal ions with TiO2@GAC in practical applications.