The pollution of water resources has become a serious problem and is causing public concern. Chlorophenols (CPs) are one of various organic contaminants in the environment. These toxic compounds are widespread and suspected carcinogens [1]. CPs have been widely used as raw materials and synthetic intermediates in many industrial processes. Because of their broad-spectrum antimicrobial properties, CPs have also been widely used as preservatives and disinfectants. CPs are generated as side products in the disinfection of drinking water and bleaching of pulp with chlorine as well. CPs are resistant to biodegradation; therefore, discharge of large amounts of these compounds into the environment is a threat to human health and ecosystems [2].
The Fenton process is an effective method for degrading a wide range of organic pollutants in water. The Fenton reaction is based on the catalytic decomposition of H2O2 by Fe2+ ions in aqueous solution to produce HO•, which can non-selectively oxidize most organic compounds. Catalysis by Fe2+ ions involves the Fe2+ ⇆ Fe3+ redox cycle shown in Eqs. (1) and (2) [3]:
However, because the reduction of Fe3+ to Fe2+ via Eq. (2) is much slower than the oxidation of Fe2+ to Fe3+ via Eq. (1), the traditional homogeneous Fenton process needs a high concentration of ferrous salt (50-80 mg/L). The subsequent treatment of iron-containing water and ferric hydroxide sludge therefore consumes large amounts of chemicals. Additionally, the homogeneous Fenton process typically works under acidic conditions (pH = 2.0-3.0), which is unfavorable in practice because of the costs of acidification during processing and neutralization after treatment [4, 5, 6]. To overcome these disadvantages of the homogeneous Fenton process, and to achieve catalyst recycling, much effort has been made to develop heterogeneous Fenton-like catalysts [7, 8, 9, 10, 11, 12]. However, many of the heterogeneous Fenton-like catalysts reported in the literature still require acidic pH conditions, and some of them show significant deactivation in consecutive reaction cycles [13, 14, 15, 16, 17]. The development of reusable and efficient heterogeneous Fenton-like catalysts with wide working pH ranges is therefore highly desirable.
Polyoxometalates (POMs), which are multi-metal oxygen clusters, have been intensively studied as catalysts because of their ability to undergo multi-electron transfer reversibly. Irradiation with light of energy higher than or equal to that of the highest occupied molecular orbital-lowest unoccupied molecular orbital (HOMO-LUMO) gap of a POM results in charge transfer from O2− to M6+ (M = W, Mo) at M-O-M bonds, leading to the formation of an excited POM with strong oxidizing power [18, 19, 20, 21, 22]. Recently, the use of POMs in Fenton or related processes has been attracting increasing attention. Choi’s group [23] reported that SiW12O404− or PW12O403− in the Fe0/O2 system can mediate electron transfer from Fe0 to O2 to produce Fe2+ and H2O2. Sedlak’s group [24] showed that addition of PW12O403− to the Fe3+/H2O2 homogeneous system extends the working pH range up to pH = 8.5. Taghdiri et al. [25] reported that silicotungstic acid catalyzes the oxidation of hexamethylenetetramine in the presence of H2O2 or H2O2/Fe2+. These studies suggest potential synergies between iron and POM for catalytic oxidation of pollutants by H2O2. It is therefore reasonable to infer that water-insoluble complexes containing both iron and heteropolyanions would be efficient heterogeneous Fenton-like catalysts. However, binary complexes of iron ions (Fe2+ or Fe3+) with many heteropolyanions (e.g., SiW12O404− and PW12O403−) are soluble in water [24].
Recently, we found that some amino acids can link Fe3+ ions with various heteropolyanions in aqueous solution to form water-insoluble ternary complexes; these complexes show Fenton-like catalytic activity [26, 27]. Here, we report a new FeIII-glutamate-silicotungstate ternary complex (FeIIIGluSiW), which acts as a highly active heterogeneous Fenton-like catalyst in the degradation of 4-chlorophenol (4-CP) at neutral pH value. The catalytic mechanism of FeIIIGluSiW was also investigated.
Ferric chloride hexahydrate, glutamic acid, silicotungstic acid, sodium silicate, sodium tungstate, and H2O2 (30%) were provided by the Sinopharm Chemical Reagent Beijing Co., Ltd. 4-CP was purchased from the Tianjin Guangfu Fine Chemical Industry Research Institute. All reagents were analytical grade and used without further purification.
FeШGluSiW was synthesized by precipitation in aqueous solution at ambient temperature. Ferric chloride hexahydrate (0.27 g) and glutamic acid (0.15 g) were dissolved in water (25 mL) to form a red soluble complex of Fe3+ ions and glutamate. A solution of silicotungstic acid (25 mL, 0.02 mol/L) was added dropwise, precipitating FeШGluSiW. The precipitate was aged for 24 h at room temperature, and red-brown FeШGluSiW powder was obtained by washing the precipitate with water and drying at 50 °C under vacuum.
The prepared FeIIIGluSiW was characterized using inductively coupled plasma atomic emission spectroscopy (ICP-AES; Shimadzu ICPS-7510), thermogravimetry (TG; Seiko TG/DTA6300), Fourier-transform infrared (FT-IR) spectroscopy (Shimadzu IRAffinity-1), ultraviolet-visible diffuse reflectance spectroscopy (UV-Vis DRS; Shimadzu UV2550), X-ray diffraction (XRD; XD-2, Cu Kα radiation), and field-emission scanning electron microscopy (FE-SEM; Hitachi S-4800).
Degradation of 4-CP was performed at 25 ± 2 °C in an FeIIIGluSiW/H2O2 system under both dark and irradiated conditions. All experiments were performed in a cylindrical quartz tube under magnetic stirring. A 400 W high-pressure mercury lamp was used as the irradiation source; it was surrounded by a quartz jacket to enable water cooling. A cutoff filter was placed around the lamp to remove radiation below 350 nm. In a typical procedure, FeIIIGluSiW powder was added to 4-CP solution (100 mg/L) with stirring to maintain a uniform suspension. The suspension pH was adjusted using dilute H2SO4 or NaOH solution, and then H2O2 was added. No pH control was applied during the process. In the photo-assisted Fenton-like process, H2O2 addition and irradiation of the reaction system were performed simultaneously. Specimens were withdrawn at regular time intervals and analyzed immediately after filtration through a 0.22 μm filter to remove the FeIIIGluSiW powder. Experiments were performed in triplicate, and all the results were expressed as mean values.
The 4-CP concentration was determined using a high- performance liquid chromatography (HPLC) system (Shimadzu LC-20AD) equipped with C18 reverse-phase column and UV detector. The mobile phase was a mixture of methanol (65%) and 1% acetic acid aqueous solution (35%) at a flow rate of 0.5 mL/min. The total organic carbon (TOC) content was determined using a TOC analyzer (Shimadzu TOC-V CPH).
The ICP-AES, TG, and FT-IR results show that the chemical formula of FeШGluSiW is [Fe(C5H8NO4)(H2O)]2SiW12O40∙13H2O. The ICP-AES data for the Fe, Si, and W contents in FeШGluSiW are close to the values calculated based on the suggested chemical formula. Calcd for FeIIIGluSiW (%): Fe 3.15, Si 0.79, W 62.2; found (%): Fe 3.62, Si 0.75, W 61.1.
The TG curve of FeIIIGluSiW in Fig. 1 shows a multi-step mass loss in the range 30-425 °C. The 6.4% mass loss in the first step, from 30 to 174 °C, arises from loss of the 13 crystallization water molecules (anal., calcd: 6.6%). The 9.5% mass loss in the subsequent steps, from 174 to 425 °C, is caused by loss of the coordinated water molecule and stepwise destruction of the glutamate moiety (anal., calcd: 9.3%). The TG data are in agreement with the proposed chemical formula.
The FT-IR spectrum of a POM reflects the structure of the heteropolyanion, i.e., the primary structure. The FT-IR spectrum of FeШGluSiW is shown in Fig. 2(1) and the assignments of the main absorption peaks, based on literature values [28, 29], are listed in Table 1. The FT-IR spectrum of FeШGluSiW has the characteristic peaks of Keggin-structured SiW12O404− at 986, 920, 881, and 793 cm-1; these are respectively attributed to the stretching vibrations of terminal W=O bonds, tetrahedral Si-O bonds, and two types of bridging W-O-W bonds in SiW12O404−. Structural information on the glutamate moiety can also be obtained from the FT-IR spectrum. The peaks at 1630 and 1617 cm−1 correspond to the asymmetric vibrations of two COO− groups of glutamate. The absorption at 1507 cm−1 is assigned to the bending vibration of NH3+. The CH2 and CH bending modes are observed at 1459, 1437, and 1349 cm−1. The FT-IR data verify the presence of Keggin-structured SiW12O404− and glutamate moiety in FeШGluSiW.
The UV-Vis DR spectra of silicotungstic acid and FeШGluSiW are shown in Fig. 3. The spectrum of silicotungstic acid has a wide absorption band in the region 200-450 nm, with peaks at 210 and 265 nm; these are attributed to charge transfer from O2− to W6+ in the Keggin unit for W=O and W-O-W bonds, respectively [30]. It is worth noting that the spectrum of FeШGluSiW shows a clear red shift of the absorption onset compared with that of silicotungstic acid. This may be favorable for photocatalysis because of the relatively narrow HOMO- LOMO gap of FeШGluSiW.
The XRD patterns of H4SiW12O40·xH2O and FeШGluSiW are shown in Fig. 4. The XRD pattern of a POM reflects its secondary structure, i.e., the three-dimensional arrangements of the heteropolyanions, cations, and other components. Usually, the primary structure of POM is stable (Fig. 2); however, the secondary structure is variable. As shown in Fig. 4, H4SiW12O40·xH2O is highly crystalline, whereas FeШGluSiW is less crystalline and possibly contains an amorphous phase.
The SEM image of FeШGluSiW in Fig. 5 shows irregular particles of size 100-200 nm.
Zero-valent iron (ZVI) has been used as a heterogeneous Fenton-like catalyst for the degradation of 4-CP at ambient temperature [13]. For comparison, the catalytic activity of FeШGluSiW was evaluated under similar conditions: 4-CP 100 mg/L, H2O2 20 mmol/L, FeШGluSiW 1.0 g/L, and temperature 25 ± 2 °C. The natural starting pH was 6.5, and no pH control was applied throughout the process. All experiments were conducted under dark and irradiated conditions separately.
Control experiments indicated that no significant 4-CP degradation occurred in 40 min when FeШGluSiW or H2O2 was used alone. However, as shown in Fig. 6, rapid degradation was observed when both FeШGluSiW and H2O2 were present, and the reaction was significantly accelerated by irradiation. The time required to degrade 4-CP completely was ca. 40 min in the dark and ca. 15 min under irradiation. Similar to most heterogeneous Fenton-like systems, the reaction rate is also influenced by experimental parameters such as catalyst dosage, H2O2 concentration, and initial pH value; these issues will be discussed in Section 3.4.
The catalytic performance of FeШGluSiW at initial pH = 6.5 was also evaluated based on TOC removal. The results (Fig. 7) show that TOC removal after 2 h was 27% in the dark and 72% under irradiation. The FeШGluSiW ternary complex therefore showed high Fenton-like catalytic activity in the degradation of 4-CP, without any pH adjustment (at natural pH = 6.5), and the reaction rate clearly increased under irradiation. In contrast, it has been reported that ZVI has no catalytic activity in 4-CP degradation at pH > 4.0 [13].
The stability of a solid heterogeneous catalyst during reactions is crucial to its reusability. In solution, SiW12O404− can be hydrolyzed to SiO32− and WO42− at neutral or alkaline pH values [18]. In reaction systems of initial pH = 6.5, FeШGluSiW is therefore expected to be unstable and prone to decomposition through hydrolysis of SiW12O404− in FeШGluSiW. Furthermore, the glutamate moiety in FeШGluSiW is possibly oxidized during this procedure, which also results in the decomposition of FeШGluSiW. If FeШGluSiW decomposes during 4-CP degradation, Fe, Si, and W at appropriate concentrations should be detected in the bulk solution by ICP-AES. The stability of FeШGluSiW during 4-CP degradation at pH = 6.5 was investigated by separating the FeШGluSiW powder from the solution after the reaction (recovery yield ca. 80%-90%). The concentrations of Fe, Si, and W in the solution were ca. 0.018, 0.0086, and 0.096 mmol/L, respectively, for the dark system, and ca. 0.030, 0.012, and 0.14 mmol/L, respectively, for the irradiated system. This shows that in addition to the unavoidable loss of catalyst during separation, ca. 5% of FeШGluSiW decomposed during the reaction. The separated FeШGluSiW was characterized using FT-IR; the results are shown in Fig. 2 and Table 1. There are no obvious differences between the FT-IR spectra of the freshly prepared FeШGluSiW and used FeШGluSiW samples. Experiments confirmed that there were no significant differences in the FeШGluSiW activity after three cycles for both the dark and irradiated systems. The FeШGluSiW catalyst is therefore stable during 4-CP degradation.
SiW12O404− immobilized in water-insoluble FeШGluSiW is more resistant to hydrolysis than SiW12O404− dissolved in water. This is similar to the reported observation that PW12O403− immobilized in a SiO2 matrix is stable in a neutral pH medium [31]. Also, in contrast to the fast oxidation of 4-CP, oxidation of the glutamate moiety in the FeШGluSiW catalyst is retarded. The exact reasons for the stabilities of the SiW12O404− and glutamate moiety in FeШGluSiW remain unknown, and further investigation is needed.
The widely accepted catalytic mechanism in heterogeneous Fenton-like systems using iron-based solid catalysts is similar to that in homogeneous Fe2+/H2O2 or Fe3+/H2O2 systems; these involve a redox cycle between Fe3+ and Fe2+ ions on the catalyst surface and production of HO• from H2O2 decomposition. However, previous studies have indicated the formation of ferryl (FeIV) species under some conditions; this is a weaker oxidant than HO• and cannot oxidize aromatic compounds [32]. It has been suggested that in POM photocatalysis, HO• can be formed through H2O activation by an excited POM [33].
The oxidative intermediate species formed in the FeIIIGluSiW/H2O2 system were detected by performing control experiments involving addition of 0.3 mol/L n-butanol (an HO• scavenger) to the reaction system at pH = 6.5. The results show that 4-CP degradation was significantly retarded in both the dark and irradiated systems, indicating that HO• is the dominant oxidative intermediate species in the FeIIIGluSiW/H2O2 system.
Control experiments were performed in two homogeneous mixed solutions, each containing 100 mg/L of 4-CP, to investigate possible degradation of 4-CP in the bulk solution. One solution contained 0.028 mmol/L Fe3+ (ferric chloride), 0.014 mmol/L SiO32− (sodium silicate), and 0.168 mmol/L WO42− (sodium tungstate), and the other contained 0.028 mmol/L Fe3+ and 0.014 mmol/L SiW12O404− (silicotungstic acid). The two aqueous solutions are equivalent to the bulk solution phase of the FeШGluSiW/H2O2 system, on the basis that 5% of the catalyst decomposes during the reaction, as mentioned in Section 3.3. The former solution corresponds to the case in which FeШGluSiW decomposes through hydrolysis of SiW12O404−, and the latter represents decomposition of FeШGluSiW through oxidation of glutamate. Both solutions were adjusted to pH = 6.5, followed by addition of H2O2 to a final concentration of 20 mmol/L. The changes in 4-CP concentration were monitored. The results show that 4-CP degradation was negligible under both dark and irradiated conditions. The formation of HO• and degradation of 4-CP in the heterogeneous FeШGluSiW/H2O2 system therefore probably occur on or near the FeШGluSiW surface.
For a heterogeneous Fenton-like process, the adsorption of H2O2 on the catalyst surface is crucial to the formation of HO•. Tang’s group [34] reported that ethylenediaminetetraacetic acid (EDTA) can significantly enhance the catalytic activity of nano-BiFeO3 for oxidative degradation of bisphenol A. The carboxyl and amino groups in EDTA can concentrate H2O2 within local areas on the catalyst surface by hydrogen bonding with H2O2; this facilitates HO• formation. In the case of FeIIIGluSiW, because of the presence of glutamate moieties and SiW12O404− heteropolyanions on the catalyst surface, H2O2 is also prone to adsorption on the surface via hydrogen bonding. This may be an important reason for the high catalytic activity of FeIIIGluSiW.
We propose a possible mechanism for FeШGluSiW catalysis, based on our experimental results and those reported in the literature. When the reaction is performed in the dark, the catalytic mechanism for FeШGluSiW is similar to those for iron-based materials, in which H2O2 reacts with iron ions on the catalyst surface (denoted by ≡FeIII and ≡FeII), resulting in the formation of HO• and oxidation of 4-CP (Eqs. (3)-(5)) [13, 14, 15, 16, 17]:
The hydrogen-bonded H2O2sub> on the FeШGluSiW surface is highly reactive with FeIII and FeII, resulting in fast production of HO•, and therefore rapid degradation of 4-CP.
When the reaction is conducted under irradiation, the reduction of ≡FeIII to ≡FeII via Eq. (3) is accelerated, resulting in enhanced production of HO• via Eq. (4) [35]. More importantly, 4-CP is also oxidized through a photocatalytic route mediated by SiW12O404− in FeШGluSiW. For POM photocatalysis, the oxidation of organic compounds is suggested to occur via two parallel reaction routes, i.e., oxidation by an excited POM (POM) and by HO• produced via activation of H2O by an excited POM [33]. The reduced POM (POM−) is oxidized by O2 to close the catalytic cycle. Photocatalysis by FeШGluSiW of the degradation of 4-CP is expressed by Eqs. (6)-(8):
We noted that the FeШGluSiW powder changed to dark brown from red-brown during the irradiated reaction. This confirms reduction of SiW12O404- during the reaction. It is worth noting that most POMs by themselves are not good photocatalysts because of slow oxidation of POM− by O2 (Eq. (9)) [18]. This may be why 4-CP degradation under irradiation is slow when FeШGluSiW is used alone without addition of H2O2. However, in the presence of both FeШGluSiW and H2O2, H2O2 can oxidize SiW12O405− to SiW12O404−, producing additional HO• species (Eq. (10)) [25]. This proposal is supported by the fact that the redox potential of H2O2 to HO• and H2O (+0.71 V versus the normal hydrogen electrode [36]) is more positive than that of SiW12O404- to SiW12O405- (+0.057 V versus the normal hydrogen electrode [37]). Additionally, the hydrogen bonding of H2O2 on the FeШGluSiW surface may promote oxidation of SiW12O405− by H2O2. The photocatalytic activity of FeШGluSiW is therefore enhanced by the presence of H2O2. In summary, the increased rate of 4-CP degradation under irradiation is ascribed to simultaneous oxidations of 4-CP through Fenton-like and photocatalytic routes mediated by ferric iron and SiW12O404− heteropolyanions, respectively, in FeШGluSiW.
The proposed catalytic mechanism explains the effects of catalyst dosage, H2O2 concentration, and initial pH value. As shown in Fig. 6(a) and (d), 4-CP degradation was hindered by decreasing the catalyst dosage from 1.0 to 0.2 g/L, and did not change significantly with increasing dosage from 1.0 to 2.0 g/L. The formation of HO• and oxidation of 4-CP mainly occur on the catalyst surface. A lower catalyst dosage (e.g., 0.2 g/L) provides fewer active sites; therefore, smaller amounts of H2O2 and 4-CP are attached to the catalyst surface, resulting in a lower reaction rate. However, addition of an excess of catalyst to the reaction system (e.g., 2.0 g/L) leads to considerable agglomeration of catalyst powder, reducing the increase in the number of active sites caused by increasing the catalyst dosage. When the reaction is carried out under irradiation, the presence of excess catalyst reduces light penetration through water, and this retards the photocatalytic degradation of 4-CP.
Fig. 6(b) and (e) shows that decreasing the H2O2 concentration from 20 to 10 mmol/L inhibited the reaction. A lower H2O2 concentration results in decreased adsorption of H2O2 on the catalyst surface, leading to reduced production of HO• via Eqs. (3) and (4), and slower oxidation of SiW12O405− by H2O2 via Eq. (10). The oxidation of 4-CP through both the Fenton-like and photocatalytic routes is therefore hindered. However, the reaction rate did not increase correspondingly when the concentration of H2O2 was increased from 20 to 30 mmol/L because of scavenging of HO• by excess H2O2 [7, 13]. Furthermore, coverage of too many active sites on the catalyst by H2O2 molecules, i.e., a too high H2O2 concentration, restrains the adsorption of 4-CP, which is unfavorable for 4-CP oxidation.
The degradation of 4-CP was enhanced by lowering the initial pH from 6.5 to 3.0, and was significantly retarded by increasing the initial pH to 8.0 (Fig. 6(c) and (f)). This is because of the higher redox potential of HO• and less decomposition of H2O2 to O2 and H2O at lower pH [14, 38, 39]. Although a low pH (e.g., 3.0) favors 4-CP degradation in the FeIIIGluSiW/H2O2 system, the reaction rate at an initial pH = 6.5, with no pH adjustment, is still considerable. FeIIIGluSiW is therefore a promising heterogeneous Fenton-like catalyst for the treatment of CPs.
An iron-glutamate-silicotungstate ternary complex, FeIIIGluSiW, was synthesized from ferric chloride, glutamic acid, and silicotungstic acid. The chemical formula of FeIIIGluSiW is [Fe(C5H8NO4)(H2O)]2SiW12O40∙13H2O, with glutamate moiety and Keggin-structured SiW12O404− heteropolyanion. The prepared FeШGluSiW showed high catalytic activity in 4-CP oxidation by H2O2 at neutral pH under both dark and irradiated conditions. The high catalytic activity of FeШGluSiW is related to the formation of hydrogen-bonded H2O2 on the catalyst surface. The enhanced degradation of 4-CP under irradiation is ascribed to the synergetic effect of Fenton-like and photocatalytic oxidations catalyzed by ferric iron and SiW12O404− heteropolyanions, respectively, in FeШGluSiW.