To remove As(III) in water, the composite material of TiO2@ZIF-8 was prepared by a sol-gel method with zeolitic imidazolate framework-8 (ZIF-8) as the matrix. The structure of TiO2@ZIF-8 was characterized with scanning electron microscopy (SEM), X-ray powder diffraction (XRD), and Fourier transform infrared spectroscopy (FT-IR). The results indicated that the best loading efficiency of TiO2 on ZIF-8 occurred when it was calcined at 300 °C for 3 h. This material was used to remove As(III) from aqueous solution, and the effect of the initial concentration of As(III), pH, and the illumination condition on the removal of As(III) was investigated. The results showed that the removal rate of As(III) was as high as 100% under a pH of 4–7, an initial As(III) concentration of less than 2 mg/L, and UV-light irradiation for 2 h. The repeated experiments were also performed for the investigation of the stability of TiO2@ZIF-8.
Due to the development of mining industries and the exclusive use of arsenical pesticides, the area of arsenic contamination is rising in developing countries (Mohan & Pittman 2007). So for people living in this area, long-term exposure to arsenic, even ingestion of contaminated water, will lead to diseases like skin lesions and cancers of the brain, liver, kidney, and stomach. Therefore, the maximum contaminant level of arsenic in drinking water was recommended by the World Health Organization (WHO) to be 10 μg/L in 1993. Arsenic contamination in water is mainly caused by nonionic trivalent (As(III)) and ionic pentavalent (As(V)) arsenic, and the different proportions of As(III) and As(V) depend on the environmental conditions of the aquifer, such as redox conditions, pH, biological activity, and adsorption reactions. The toxicity of As(III) is over 60 times greater than that of As(V), and, in addition, As(III) is more difficult to remove from water by most techniques (Zhang et al. 2017). At present, the mainstream technology of removal of As(III) from water is peroxidation to As(V) by an oxidant and adsorption of the As(V). Among different oxidation treatments (such as ozone, hydrogen peroxide, potassium ferrate, and photocatalytic oxidation), photocatalytic oxidation has been proven to be a useful approach in oxidizing As(III) to As(V) (Guan et al. 2012). Among all photocatalysts, titanium dioxide (TiO2), the most common semiconductor photocatalyst, is a promising material for removing arsenic, especially As(III) (Bissen et al. 2001; Pirila et al. 2011).
Photocatalysis with TiO2 offers a relatively inexpensive, environmentally benign way to achieve As(III) oxidation. Moreover, TiO2 can also work as an adsorbent to remove arsenic from water. The bi-functional property of TiO2 gives it an additional advantage, and many studies have been carried out to investigate arsenic removal by TiO2 (Bang et al. 2005). However, the low adsorption capacity and the problem of separating TiO2 powder from an aqueous solution usually limit its application in arsenic removal (Qu 2008). Furthermore, the direct application of suspended TiO2 powders in drinking water treatment may be problematic due to the difficulty of separation and recovery of the tiny particles. In order to overcome this drawback and combine the advantages of TiO2, the supported TiO2 is becoming a better choice for the field application of the photocatalyst. As is usual, the support material has the properties of high surface area, absorption capacity, and chemical stability. Compared with conventional adsorbents (such as activated carbon, activated alumina, and zeolite), zeolitic imidazolate framework-8 (ZIF-8) has a higher surface area and adsorption activity (Liu et al. 2015).
ZIF-8, formed by imidazole ligands and metal ions, is a kind of metal-organic framework, and has potential selectivity in adsorption, gas separation, and capturing target materials, etc. ZIF-8 has perfect adsorption ability for As(V), but has a very low adsorption affinity for As(III) as it is a neutral molecular species in a large pH range. Jian et al. (2015) used ZIF-8 to adsorb As(V) up to 99.94%, in contrast the adsorption rate of As(III) was only 27% under ZIF-8 dosage of 0.2 g/L for 100 μg/L As(III) or As(V). Considering the adsorption ability of ZIF-8 for As(V) and the oxidation capacity of TiO2 for As(III), we have a proposition to synthesize the composite material of TiO2@ZIF-8 for removal of As(III). In recent years, TiO2 and adsorbent heterostructures have been reported for removal of As(III) in water, such as TiO2-activated carbon (Liu et al. 2014), magnetic Fe3O4@TiO2 (Lan 2015), and TiO2-chitin (Miller & Zimmerman 2010). Moreover, photocatalysts loaded on the surface of ZIF-8 were also reported for degradation of organic dye, such as ZnO@ZIF-8 and TiO2@ZIF-8 (Chandra et al. 2015; Yu et al. 2015).
To the best of our knowledge, no one has ever reported the removal of As(III) by TiO2@ZIF-8, and the concentration of arsenic in drinking water is generally less than 500 μg/L, even in the two worst affected areas, Bangladesh and West Bengal, India (Mazumder et al. 1998). Furthermore, the pH of drinking water is close to neutral, so we chose TiO2@ZIF-8 to deal with the low concentration of As(III) in water at neutral conditions. In this paper, the composite material of TiO2@ZIF-8 was synthesised, and its properties were investigated. Then the adsorption kinetic and isotherm of TiO2@ZIF-8 for As(III) were conducted under protection from light. Finally, we carried out an experiment to remove As(III) from aqueous solution, and discussed the effect on As(III) removal of the different factors.
MATERIALS AND METHODS
Chemicals and materials
Analytical grade chemicals were used without any further purification. Tetrabutyltitanate Ti(OBu)4 was purchased from Shanghai Zhanyun Chemical Co., Ltd (Shanghai, China); 2-methylimidazole was obtained from Shanghai Baiao Biological Technology Co., Ltd (Shanghai, China); and Zn(NO3)2·6H2O was purchased from Xilong Chemical Co., Ltd (Guangdong, China). A standard solution of 1,000 mg/L As (III) was purchased from O2si smart solutions (Charleston, SC, USA), and this was diluted just before use. All volumetric flasks and vessels were soaked in 10% HNO3 for at least 24 h, and then rinsed for several times with ultra-pure water from a Sartorius water purification system (Mini plus, Sartorius, Germany).
Preparation of ZIF-8 nanoparticles and TiO2@ZIF-8 materials
ZIF-8 nanoparticles were synthesized by a liquid phase diffusion method. In a typical synthesis, 1.5 g Zn(NO3)2·6H2O and 3.3 g 2-methylimidazole were dissolved in 70 mL anhydrous methanol. Then the solution of 2-methylimidazole was slowly added to another solution under stirring at room temperature (25 ± 2 °C). After stirring for 24 h, the product was collected by centrifugation at 8,000 rpm for 10 min, and then washed with anhydrous methanol three times. The obtained ZIF-8 nanoparticles were dried in a vacuum drying oven at 110 °C for 24 h to remove residual methanol, and kept in a desiccator before use.
TiO2@ZIF-8 was prepared by a sol-gel method at room temperature using Ti(OBu)4 as the precursor. First, 0.6 mL acetic acid was added to 10 mL absolute alcohol as a stabilizer of the solution, and then 4 mL Ti(OBu)4 and 10 g ZIF-8 were added to the solution and mixed well by a magnetic stirrer for 2 h. Next, 0.6 mL acetic acid and 1.6 mL ultra-pure water were added to 5 mL absolute alcohol, and the pH was adjusted by concentrated nitric acid to 3 under vigorous stirring. Then the second solution was slowly added to the first solution at room temperature followed by stirring for 4 h. The titanium sol⁄ZIF-8 turned into titanium gel⁄ZIF-8 after aging for 24 h at 25 °C, and it was then dried at 60 °C. The powder was calcined at different temperatures ranging from 100 °C to 400 °C for 3 h. The obtained composite materials were kept in a desiccator before use.
The morphologies of TiO2@ZIF-8 and ZIF-8 were examined by a scanning electron microscope (SEM, Ultra-55, Carl Zeiss, Germany). X-ray powder diffraction (XRD, D/max-r B, Rigaku Corporation, Japan) analysis was carried out on a diffractometer with Cu-Kα target (40 kV, 100 mA, λ = 1.5418 Å) from 5° to 80°. The functional groups on the TiO2@ZIF-8 were identified at room temperature by Fourier transform infrared spectroscopy (FT-IR, Nicolet 6700, Thermo Fisher, USA). All samples were dried for 24 h at 60 °C prior to the KBr preparation in order to avoid water related bands interference.
Removal of As(III)
The experiments on As(III) removal were carried out using a xenon lamp (300 W) as the irradiation source at room temperature. An initial As(III) concentration of 2 mg/L and a pH of 7.0 ± 0.1 were used in most experiments, while the amount of TiO2@ZIF-8 powder was 0.04 g/L. The suspension of As(III)-TiO2@ZIF-8 was in the dark for 1.5 h to allow adsorption equilibrium before it was illuminated by UV-light (λ < 400 nm) or visible light (λ > 400 nm). Samples of the suspension were withdrawn after 30 min intervals and filtered through a 0.22 μm membrane. The filtrates were analyzed for the residual concentration of the different species of arsenic, especially As(III) and As(V). To compare the removal efficiency of As(III) by TiO2@ZIF-8, the pure TiO2 and ZIF-8 were tested. Additionally, the effect of the media pH, the initial concentration of As(III), the interaction time, and the illumination condition were also determined. Adsorption isotherm experiments of As(III) on TiO2@ZIF-8 were conducted in dark conditions, and with the initial concentration of As(III) ranging from 0.2 mg/L to 10 mg/L, pH 7.0 ± 0.1, solid⁄solution ratio of 0.04 g/L, and stirring for 5 h at 25 ± 1 °C. Furthermore, the adsorption kinetics of As(III) by TiO2, ZIF-8, and TiO2@ZIF-8 were also tested in dark conditions.
To detect the different species of arsenic in samples, an atomic fluorescence spectrometer coupled with liquid chromatography (LC-AFS, LC-AFS 9600, Beijing Haiguang instrument, China) was used, which was equipped with a Hamilton PRP-X100 anion-exchange column (250 mm × 4.1 mm). The mobile phase was a pH 5.6 phosphate buffer prepared with 2.26 g Na2HPO4·12H2O and 3.03 g KH2PO4 with ultra-pure water to 500 mL. The reducing solution was a mixture of 5 g/L NaOH and 10 g/L KBH4, with 5% (v/v) hydrochloric acid as the carrier solution.
The recycling method of TiO2@ZIF-8 was as follows: the TiO2@ZIF-8 that had been used for the removal of As(III) was filtered, desorbed using 0.1 mol/L NaOH under stirring for 24 h, and then washed with ultra-pure water three times, and finally dried in a vacuum drying oven at 110 °C for 24 h.
RESULTS AND DISCUSSION
Characterization of TiO2@ZIF-8
Calcination is a common treatment used to improve the crystallinity of TiO2 powders, and the XRD patterns of TiO2@ZIF-8 calcined at the different temperatures are shown in Figure 1. We found that the strength of the characteristic peaks of TiO2 increased with the rise in calcination temperature when the calcination temperature ranged from 100 °C to 300 °C. However, when the calcination temperature was up to 400 °C, the TiO2@ZIF-8 was amorphous, because the thermo-stability of ZIF-8 is about 350 °C, which was in agreement with previous reports (Tran et al. 2011). The XRD patterns of TiO2@ZIF-8 calcined at 300 °C and ZIF-8 are shown in Figure 2, and the TiO2@ZIF-8 was composed of a TiO2/ZIF-8 hybrid structure: the peaks at 2θ values of 25.3°, 37.9°, 48.1°, 54.16°, 55.32°, and 62.7° could be indexed to the (101), (004), (200), (105), (211), and (204) crystal planes of TiO2, respectively (Wei et al. 2016). The peaks at 7.26°, 10.32°, 12.64°, 17.96°, and 26.62° corresponded to the (011), (002), (112), (222), and (134) of the structure of ZIF-8, respectively (Zhu et al. 2011). Both the characteristic peaks of TiO2 and ZIF-8 were found in TiO2@ZIF-8, which indicated the combination of ZIF-8 and TiO2. The peak strength of TiO2 was weaker than that of ZIF-8, which may be caused by the lower content of TiO2 than ZIF-8 in the composite material, because the molar ratio of Zn to Ti was about 16.8:1. On the other hand, the peak strength of ZIF-8 in the composite material was weaker than the ZIF-8 nanoparticle, which may be attributed to the effect of calcination.
In Figure 3, the SEM micrographs of TiO2@ZIF-8 calcined at the different temperatures and the original ZIF-8 are shown. From Figure 3 and combined with the analytic results of XRD (Figure 1), the larger particles of the precursor of TiO2 but not TiO2 were found to densely adhere to the surface of ZIF-8 when the calcination temperature was 100 °C. When the calcination temperature was elevated to 300 °C, the precursor of TiO2 could hardly be seen, and TiO2 clearly adhered to the surface of ZIF-8. However, when the calcination temperature was up to 400 °C, the pore structure of the ZIF-8 had been damaged (Tran et al. 2011), and the large amount of TiO2 that adhered to the surface of ZIF-8 had been dropped. From the above observations, the crystallinity of TiO2 increased while the crystallinity of ZIF-8 decreased as the calcination temperature increased in the range from 100 °C to 300 °C, so we may come to the conclusion that the best calcination temperature is 300 °C, and TiO2@ZIF-8 calcined at 300 °C was used in the following experiments.
The FT-IR spectra of TiO2@ZIF-8 calcined at 300 °C and ZIF-8 nanoparticles in the range of 2,000–400 cm−1 are plotted in Figure 4. The FT-IR analysis of TiO2@ZIF-8 revealed the presence of an adsorption band at 420 cm−1 (Zn-N stretch), another adsorption band at 1,584 cm−1 (C = N stretch), and also two bands at 996 cm−1 and 1,425 cm−1 (C-N stretch), which were typically characteristic for ZIF-8 (Hu et al. 2011). The absorption peak around 1,337 cm−1 was the typical vibration of the Ti-OH bond (Tan et al. 2011). No new bands, such as Ti-N stretch at 510 cm−1 (Zeng et al. 2016), could be observed on the composite material spectrum, which indicated that no new chemical groups were formed. From the XRD pattern and FT-IR spectra of TiO2@ZIF-8 calcined at 300 °C, we affirmed that TiO2 had adhered to the surface of ZIF-8.
Adsorption kinetics and isotherms of As(III)
To analyse the adsorption kinetics of As(III) by TiO2, ZIF-8, and TiO2@ZIF-8 in dark conditions, pseudo-first-order and pseudo-second-order models were used to describe the sorption kinetic data.
As shown in Figure 5, the adsorption of As(III) on TiO2@ZIF-8 and ZIF-8 was fast in the beginning, and then quickly levelled off, and the time required to reach the adsorption equilibrium was only 60 min. The adsorption capacity of ZIF-8 was higher than TiO2@ZIF-8, which is because the porosity and specific surface of ZIF-8 was higher than that of TiO2. The adsorption rate constants k1 (see Table 1) of ZIF-8 and TiO2@ZIF-8 were similar, and the rate of TiO2@ZIF-8 was about 3.5 times faster than that of TiO2.
|qeq (mmol/g) .||k1 (h−1) .||R2 .||qeq (mmol/g) .||k2 (g/mmol·h) .||R2 .|
|qeq (mmol/g) .||k1 (h−1) .||R2 .||qeq (mmol/g) .||k2 (g/mmol·h) .||R2 .|
To measure the adsorption capacities and behaviours of As(III) on TiO2@ZIF-8 in dark conditions, the Langmuir and Freundlich isotherm models were applied to explain the adsorption equilibriums.
The adsorption isotherms are shown in Figure 6, and the parameters are summarized in Table 2. The adsorption equilibrium data were better fitted by the Freundlich model than the Langmuir model, which means a heterogeneous sorption (Jian et al. 2015). The maximum adsorption capacity for As(III) from the Langmuir model was determined at 0.51 mmol/g.
|Material .||Langmuir model|
|Q (mmol/g) .||KL (L/mg) .||R2 .||n .||KF (mmol/g) .||R2 .|
|Material .||Langmuir model|
|Q (mmol/g) .||KL (L/mg) .||R2 .||n .||KF (mmol/g) .||R2 .|
Influential factors for the removal of As(III)
The effect of illumination condition was analysed in the results of UV-light (λ < 400 nm) or visible light (λ > 400 nm) irradiation (see Figure 7). The results indicated that the removal rate increased rapidly under UV-light irradiation, and other species of arsenic were not found except As(III). By comparison of the amounts of As(III) removed with and without UV-light, it can be affirmed that the disappearance of As(III) is mainly caused by photocatalytic oxidation instead of only adsorption. The mechanism of the oxidation reaction catalyzed by TiO2@ZIF-8 was proposed as the following four steps (Yao et al. 2010):
The adsorption of As(III) on the surface of TiO2;
The oxidation of As(III) to As(V);
The desorption of As(V) from the surface of TiO2;
The adsorption of As(V) on the surface of ZIF-8 or TiO2.
In order to evaluate the effect of the removal of As(III) by TiO2@ZIF-8, the comparative experiments with TiO2 and ZIF-8 were carried out, and the results are shown in Figure 8. Under the UV-light irradiation, the concentrations of As(III) rapidly decreased for the pure TiO2 and TiO2@ZIF-8, and the concentration decreased slightly for ZIF-8, which was due to the ozone that was produced by the xenon lamp oxidizing As(III) to As(V), which was then adsorbed by ZIF-8. The more significant finding was that TiO2@ZIF-8 was more effective for the removal of As(III) in comparison to TiO2. The reason may be that the band gap of TiO2 decreased, which enhanced the effect of photocatalytic oxidation (Chandra et al. 2015). Compared to TiO2 and ZIF-8, almost 100% arsenic removal was achieved by TiO2@ZIF-8, indicating a great treatment efficiency to satisfy the WHO's drinking water standard for arsenic.
The removal of As(III) by TiO2@ZIF-8 under UV-light irradiation was also studied at the various initial As(III) concentrations ranging from 0.5 mg/L to 10 mg/L with the solid/solution ratio of 0.04 g/L (see Figure 9). The findings indicated that the removal amount (mmol/g) of As(III) increased with the increase of the initial concentration, from 0.16 mmol/g (0.5 mg/L) to 2.7 mmol/g (10 mg/L). However, the removal rates of As(III) gradually decreased with the concentration increase. The removal rate of As(III) was 100% when the initial As(III) concentration was below 2 mg/L, and 72% when the initial concentration was 10 mg/L. The decrease in the percentage of removal rate can be explained by the fact that the active sites of TiO2@ZIF-8 reached saturation with the increase in initial concentration. In general, TiO2@ZIF-8 has the advantage of producing less polluted water with a low concentration of As(III), especially less than 2 mg/L.
The solution pH influences both the binding sites on the surface of the adsorbent and the distribution of arsenic species, thus it may have an impact on the removal of As(III). The effect of media pH on the removal behaviour was analysed in the pH range 4–10. (see Figure 10). The results show that the adsorption rate was similar at different pH values in dark conditions, and the removal rate was not significantly different at a pH of 4–7 under UV-light irradiation. The removal rate was 100% and 89.34%, corresponding to pH 4 and 10, respectively. So TiO2@ZIF-8 adapts to a comparatively broad range of pH, and similar As(III) removal behavior by nanocrystalline TiO2 was observed (Pena et al. 2006).
In order to assess the feasibility of reutilization of TiO2@ZIF-8 calcined at 300 °C, it was subjected to a series of adsorption/desorption cycles (Figure 11). The results show that the removal rate of As(III) remained above 90% after the TiO2@ZIF-8 had been used for five times. Thus, it is suggested that the deposited TiO2 particles had firmly attached to the surface of ZIF-8, and cannot be easily exfoliated from the ZIF-8 with mechanically stirred solutions for a long period.
The optimum synthesis condition of TiO2@ZIF-8 was calcination at 300 °C for 3 h by the sol-gel method. In contrast with the single ZIF-8 or TiO2, TiO2@ZIF-8 had a better removal ability to deal with a low concentration of As(III), especially less than 2 mg/L, under UV-light irradiation. This material could adapt to a wide pH range, and the removal rate was up to 100% in the pH range from 4 to 7. In addition, the composite material can be used repeatedly and the high removal properties were maintained with a slight decline. Through the above analysis and discussion, we propose that TiO2@ZIF-8 is a good prospect for the removal of As(III) from water containing low initial As(III) concentrations.