A facile strategy to increase the selectivity of heterogeneous Fenton oxidation is investigated. The increase was reached by increasing selective adsorption of heterogeneous Fenton catalyst to a target pollutant. The heterogeneous Fenton catalyst was prepared by a two-step process. First, zeolite particles were imprinted by the target pollutant, methylene blue (MB), in their aggregations, and second, iron ions were loaded on the zeolite aggregations to form the molecule imprinted Fe-zeolites (MI-FZ) Fenton catalyst. Its adsorption amount for MB reached as high as 44.6 mg g−1 while the adsorption amount of un-imprinted Fe-zeolites (FZ) is only 15.6 mg g−1. Fenton removal efficiency of MI-FZ for MB was 87.7%, being 33.9% higher than that of FZ. The selective Fenton oxidation of MI-FZ for MB was further confirmed by its removal performance for the mixed MB and bisphenol A (BPA) in solution. The removal efficiency of MB was 44.7% while that of BPA was only 14.9%. This fact shows that molecular imprinting is suitable to prepare the Fe-zeolites (FZ)-based Fenton catalyst with high selectivity for removal of target pollutants, at least MB.
Composites of wastewater have become more complex in industrial development, leading to more difficulty in wastewater treatment (Chen & Zhang 2013; Sharma et al. 2013; Zhang et al. 2013). However, it is noticed that only parts of these composites are recalcitrant and more toxic, while others are low in toxicity. On the other hand, for example, the color of wastewater containing residual dye is very high, and it is difficult to reach discharge standards. Many chemical synthetic dyes are difficult to biodegrade, thus, it is a favorable potential strategy to treat these wastewaters by selective oxidation for special pollutants followed by a biological process (Inumaru et al. 2005; Shen et al. 2008). Therefore, to develop an effective selective oxidation technology is of significance.
In chemical oxidation, heterogeneous Fenton process was successfully used in past decades as an effective treatment to degrade toxic organic pollutants and to enhance the biodegradability of the wastewater (Zhang et al. 2010, 2011). As Fenton process yields highly reactive and non-selective hydroxyl radical (•OH) by iron interaction with hydrogen peroxide, it is still difficult to realize selective removal of harmful materials from complicated wastewaters in the presence of other pollutants (Grebel et al. 2010; Maddila et al. 2014). Considering pollutants were adsorbed on the surface of the catalyst first and then oxidized in the heterogeneous Fenton process, it is expected that the specific selectivity of its catalytic oxidation can be increased by increasing selectivity of the pollutant adsorption on the heterogeneous catalysis.
Molecularly imprinted (MIP) materials have been extensively studied due to their molecular recognition ability (Fan et al. 2015; Tang et al. 2015), specific adsorption (Zhao et al. 2014; Wei et al. 2015), and wide applications in separation (Serrano et al. 2015; Wu et al. 2015) and sensors (Chen et al. 2012; Gupta et al. 2015; Kamra et al. 2015). Generally, MIP materials are organic polymers whose stability is limited by chemical oxidation. Therefore, many researchers recently have focused on MIP inorganic catalysts. For example, Shen prepared an inorganic MIP polymer-coated photocatalyst for photodegradation of diethyl phthalate (DEP) by coating a layer of MIP silica/alumina on the surface of TiO2 nanoparticles with DEP as the template (Shen et al. 2009). In the MIP photocatalysis, Al3+ ions as a Lewis acid on the particles interacted with the template DEP as a Lewis base to form surface coordination compounds. Brage et al. (2014) prepared a bi-functionalized MIP silica for solid phase extraction of quercetin. To ensure the specificity of molecular recognition, aluminum ions were inserted in the silica matrix to form Lewis acid sites and allow the interaction with quercetin. All the above inspired us to study the MIP modified heterogeneous Fenton catalysts for selective removal of toxic organic pollutants.
Considering that zeolite contains a large number of Lewis acid centers generated by Na+ or non-skeletal Al (Yu et al. 2011), the Lewis acid sites are expected to combine with the target toxic organic pollutant which is also a Lewis base. The MIP modification was obtained on zeolite through Lewis acid–base action. After being imprinted, the template molecule was washed out and the ‘molecular footprint’ cavities on the zeolite remained. On the other hand, zeolite is a crystalline aluminosilicate that has uniform pore size distribution fixed by the atomic arrangement of the unit cells. The existence of unvarying pore diameter endows zeolite with extraordinary molecular discrimination ability (Yonli et al. 2012; Gonzalez-Olmos et al. 2013; Ikhlaq et al. 2014). The ion exchange of zeolite was carried out in the framework of the Si2AlO6-, and the negative charge of Si2AlO6- could not only be combined with the sodium ion, but also other cations. Fe ions could enter the large crystal hole of the sodium ion, and replace the sodium ions in the zeolite. Therefore, loading of iron ions were obtained on MIP-modified zeolite as the Fe-immobilized zeolite was reported to be a typical Fenton-like catalyst (Doocey et al. 2004; Chen et al. 2010; Rache et al. 2014). Then, the molecule imprinted Fe-zeolites (MI-FZ) Fenton catalyst was prepared.
At the start of the investigation, the present work is aimed at selective characteristic research of the MIP Fenton catalyst MI-FZ. The specific adsorption and oxidation of the target pollutant within or without the interference of the nontarget pollutants were investigated. Methylene blue (MB) was chosen as the target pollutant as it is one of the most commonly used substances for dyeing and is potentially harmful to the eco-environment (Bai et al. 2015; Dutta et al. 2015; Soniya & Muthuraman 2015). Bisphenol A (BPA) was chosen as the interference matter because it is also an aromatic compound similar to MB. The Fe-immobilized zeolite catalyst without MIP-FZ was also prepared to emphasize the selectivity of MI-FZ by comparison.
MATERIALS AND METHODS
Chemicals and materials
Hydrogen peroxide solution (6%, v/v), Fe(NO3)3·9H2O, MB, and BPA were supplied by Guangzhou Chemical Reagent Company, China. NaOH, HNO3, and HCl were provided by Tianjin Chemical Reagent Co., Ltd. The 4A zeolite was supplied by Tianjin Kermel Chemical Reagent Co. Ltd, China. All the chemicals were of analytical reagent grade and used as received without further purification.
Preparation of MI-FZ catalyst
MI-FZ catalysts were prepared via a routine given in a graphical abstract. First, the molecular imprinting was carried out by adding 10 g zeolite powder into 200 mL saturated solution of MB through Lewis acid–base action. The mixed liquid was kept in a water bath at 80 °C for one week after adjusting the pH to 4. To remove the template (MB), a Soxhlet extraction system was used (Jafari et al. 2012). The extraction solvent (500 mL) was a mixture of methanol and acetic acid (9:1) and the extraction was continued for 24 h. Then, the obtained precipitate was washed by ethanol, dilute hydrochloric acid, and water in turn. The complete removal of the template was verified by using spectrophotometric determination of MB in the final filtrate. Absorbance less than 0.1 was taken as the complete removal of the template. Then, the MIP Fenton-like particles based on modified zeolite were prepared according to our previous research (Shang et al. 2016). Twenty grams of imprinted 4A zeolite was dispersed into 200 mL FeSO4 solution (0.36 M) to exchange ion at 30 °C. After continuously stirring for 24 h, the precipitate was thoroughly washed with distilled water. MI-FZ catalysts were then obtained after drying at 105 °C and calcined at 350 °C for 2 h. The amount of iron introduced into molecular imprinted zeolite was determined by energy dispersive X-ray spectrometer (EDS) to be 17.13 wt.%. Fe-zeolite without imprinting (FZ) was prepared by ion exchange of zeolite and FeSO4 solution with the same conditions as MI-FZ to make a comparison (Chen & Zhang 2013).
Characterization of MI-FZ Fenton catalysts
The Fe content in the catalyst was determined by an EDS (JEOL, Japan scanning electron microscope). Fourier transform-infrared spectroscopy (FT-IR) and X-ray diffraction (XRD) spectra were recorded on a Bruker VERTEX 70 spectrophotometer and a Shimadzu UV-2550 spectrophotometer, respectively. The Fe concentration in the effluent was determined by inductively coupled plasma optical emission spectrometry (ICP-OES, Optima 5300DV, Perkin-Elmer). Specific Brunauer–Emmet–Teller (BET) surface area, Barrett–Joyner–Halenda (BJH) and Horváth–Kawazoe (HK) analysis for pore size distribution were determined with a Micromeritics ASAP 2010 apparatus by nitrogen adsorption at 77 K. The particle size distribution was measured by laser particle size analyzer (MasterSizer 3000). FT-IR spectra were recorded on a spectrophotometer (Bruker VERTEX 70).
Adsorption and Fenton-like experiments were carried out in a cylindrical Pyrex vessel. In a typical run, the reaction suspension was prepared by adding a given amount of MI-FZ and FZ powder into 100 mL MB or BPA solution which has been adjusted to the desired concentration and pH value. The initial molar concentration of MB and BPA was 0.027 Mm and 0.439 mM, respectively. The molar concentration of BPA was fixed about 16 times higher than MB to confirm the selectivity. Prior to Fenton-like reaction, the suspension was magnetically stirred in the dark for 3 h to establish the adsorption/desorption equilibrium. A Fenton-like reaction was initiated by adding H2O2 (6%) to the solution with constant stirring. Samplings were obtained at given time intervals during the reaction. Then, samples were analyzed immediately after filtration through 0.45 μm Millipore membrane filters to remove suspended particles.
The concentration of MB was measured by UV-vis spectrophotometer (UV-3150, Shimadzu) with a maximum absorbance at 664 nm. BPA was measured by high-performance liquid chromatography on a PU-2089 HPLC (JASCO) equipped with a C18ODS column and an ultraviolet detector.
RESULTS AND DISCUSSION
Characterization of MI-FZ catalyst
Figure 1(b) shows the FT-IR spectrum of MI-FZ and FZ. The peak values of the two lines were not significantly changed, which indicated that the modified loading and the imprinting process had no effect on the structure of zeolite. The peaks at 3,450 cm−1 and 1,640 cm−1 correspond to the stretching bands and bending vibration of Si–OH–Al groups (Montanari & Busca 2008), respectively. The peak at 1,002 cm−1 is assigned to the asymmetric stretching of Si–O–Al (Zhong et al. 2006). Compared with the FZ, due to the imprinted hole of the template, MI-FZ spectrum shows new weak peaks of 1,600 cm−1 and 1,330 cm−1 which was assigned to the stable polyaromatic C = C (Zheng et al. 2016) and bending band of adsorbed water, respectively. The Al3+ ions in MI-FZ as Lewis acid would act as a functional group for the inorganic polymer to combine with the template.
Selective adsorption of MI-FZ catalyst
To further confirm the molecular recognition of MI-FZ, the sorption kinetics of BPA as the interference matter on MI-FZ and FZ are presented in Figure 4(b). The time for both MI-FZ and FZ to achieve the adsorption/desorption equilibrium was about 1 h. Compared with the target molecule MB, BPA was suggested only to be adsorbed on the surface of the catalysts. While the cavities were especially in MB, the effective adsorption sites of MI-FZ for BPA were less than FZ; the adsorption capacity of MI-FZ for BPA was about 223.1 mg g−1 after sorption equilibrium, much lower than that of FZ of 311.5 mg g−1. The above results also indicated that MI-FZ has a special adsorption ability regarding the target molecule, but other interference matters. Moreover, according to one study, the molecular diameter of MB is 0.8 nm (Hameed & Ahmad 2009). However, from the literature, we cannot find the exact molecular size of BPA. A single BPA molecule was assumed to be a regular small ball, and molecules were assumed to be compact with no space. According to the Avogadro constant (NA = 6.02*1023/mol) and the molar volume of BPA (V = 199.5cm3/mol), the molecular diameter of BPA was roughly calculated to be about 0.86 nm by the formula: d = (6*V/NA*π)−3. Therefore, MB and BPA were similar in size, and the special adsorption may be caused by different spatial structures.
|Binary system||Single MB||Single BPA|
|Catalyst||Equilibrium time (h)||Percent %||Equilibrium time (h)||Percent %||Equilibrium time (h)||Percent %|
|Binary system||Single MB||Single BPA|
|Catalyst||Equilibrium time (h)||Percent %||Equilibrium time (h)||Percent %||Equilibrium time (h)||Percent %|
Figure 6(b) shows that the adsorption of MB by MI-FZ could also be described by the Dubinin–Radushkevich isothermal model (R2 > 0.99). The calculated E of 12.91 kJ/mol indicated that the adsorption was chemical adsorption mainly due to the interaction between Lewis acid Al3+ and Lewis alkali MB.
Selective Fenton-like activity of catalysts in single and binary systems
Effect of pH on the adsorption and Fenton-like removal of MB by MI-FZ
Considering the toxicity of by-products, the effluent was tested using liquid chromatography–mass spectrometry (LC-MS); however, no by product was found as the apparatus could only detect materials with a molecular weight greater than 50. The reason may be due to the fact that N-CH3 in MB molecules, which connects externally to the benzene ring (7C and 12C), was easily interrupted by bombardment of the active particle, and then the generated -CH3 is oxidized to small molecular organic compounds such as HCHO and HCOOH. Hence, it may be supposed that there are no toxic by products in the effluent.
Stability of MI-FZ
A new molecular imprinted Fenton-like catalyst MI-FZ with high selectivity was successfully synthesized based on FZ by using target pollutant MB as the template. Several characterizations were carried out for the catalyst such as XRD, FT-IR, particle size distribution, BET surface area, BJH and HK pore size analysis. The results proved the generation of new holes produced by the imprinting and which was beneficial for selective catalytic oxidation of the target pollutant. The calculated energy of Dubinin–Radushkevich isothermal model indicated that the adsorption of target pollutant by MI-FZ was chemical adsorption, mainly due to the interaction between Lewis acid of Al3+ and Lewis alkali of MB. Both the adsorption and Fenton-like removal efficiencies of MB over MI-FZ were higher than FZ, whether in the single system or in the binary system coexisting with the interference matter BPA. The best pH value for Fenton-like reaction was 4, with low ion dissolution and, hence, MI-FZ was confirmed to have high stability.
This research was supported by Nature Science Foundations of China (51668005, 21267002, 21367003, 41473118, and 41273139), the China Postdoctoral Science Foundation (No. 2016M592607, 2016M590846), the National Key Basic Research Program of China (No. 2013CB956102), Nature Foundations of Guangxi Province (2015GXNSFBB139009, 2014GXNSFAA118296 2014GXNSFBA118217, 2013GXNSFEA053001) and the BaGui Scholars Program Foundation (2014).