A Novel Homogeneous Fenton-like System with Fe(III)-Phosphotungstate for Oxidation of Organic Compounds at Neutral pH Values

Abstract

Ferric ion (Fe[III]) catalyzes the decomposition of hydrogen peroxide (H₂O₂) into strong oxidants such as hydroxyl radical (^•OH) and ferryl ion (Fe[IV]) through the redox cycling of the iron couple (Fe[II]/Fe[III]). The use of these reactions for the catalytic oxidation of organic compounds is usually limited to the acidic pH region due to the low solubility of Fe(III) and the low efficiency of oxidant production at neutral pH values. The addition of phosphotungstate (PW₁₂O₄₀³⁻), a polyoxometalate extends the working pH range of the Fe(III)/H₂O₂ system up to pH 8.5. PW₁₂O₄₀³⁻ forms a soluble complex with iron that converts H₂O₂ into oxidants. The coordination of Fe(II) by PW₁₂O₄₀³⁻ also alters the mechanism of the reaction of Fe(II) with H₂O₂ at neutral pH, resulting in formation of an oxidant capable of oxidizing aromatic compounds. The base-catalyzed hydrolysis of PW₁₂O₄₀³⁻ gradually results in inactivation of the catalyst. In the absence of Fe(III), PW₁₂O₄₀³⁻ was completely hydrolyzed after 1 day at pH 7.5, whereas the Fe(III)-PW₁₂O₄₀³⁻ complex was active for at least four days under the same conditions.

1. Introduction

Over the last two decades, the advanced oxidation processes that employ hydrogen peroxide (H₂O₂) and iron have been extensively studied as methods for the oxidation of recalcitrant contaminants in drinking water, industrial wastewater, and soils [1]. Under acidic conditions, Fenton’s reagent (i.e., Fe(II)/H₂O₂) rapidly converts H₂O₂ into a stoichiometric amount of hydroxyl radical (^•OH) (reaction 1).

$$\mathrm{Fe}(\text{II})+{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{Fe}(\text{III})+{}^{•}\mathrm{O}\mathrm{H}+\mathrm{O}{\mathrm{H}}^{-}{k}_{\mathit{2}}=63{\mathrm{M}}^{-1}{\mathrm{s}}^{-1}[2]$$ (1)

While this process has been used for the treatment of water containing toxic organic compounds [3–5], the formation of ferric precipitates after the treatment produces large amounts of waste. To avoid this problem, a low concentration of Fe(III) relative to H₂O₂ can be used for the catalytic conversion of H₂O₂ into ^•OH (i.e., reaction 2 followed by reaction 1). The application of this treatment process has been limited because the rate of reaction 2 is slow.

$$\mathrm{Fe}(\text{III})+{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{Fe}(\text{II})+\mathrm{H}{{\mathrm{O}}_2}^{•}+{\mathrm{H}}^{+}{k}_{\mathit{1}}\approx {10}^{-3}{\mathrm{M}}^{-1}{\mathrm{s}}^{-1}\text{at pH}2.5[6]$$ (2)

Both the Fe(II)/H₂O₂ and the Fe(III)/H₂O₂ systems are typically used only under acidic pH conditions (< pH 3), and as a result the addition of strong acids and bases is required before and after the process. The low efficiency of these systems under neutral pH conditions is attributable to two factors. First, Fe(III) forms insoluble precipitates of amorphous Fe(III)-oxyhydroxides, which produce very low yields of oxidants from reactions with H₂O₂ [7, 8]. Second, at these pH values, the reaction of Fe(II) with H₂O₂ does not always produces ^•OH. The oxidant formed at neutral pH values, which appears to be ferryl ion (Fe[IV]), is capable of oxidizing short chain aliphatic alcohols and arsenite [9–11], but cannot oxidize aromatic organic compounds and other recalcitrant contaminants [11].

The addition of iron-chelating agents such as ethylenediaminetetraacetate (EDTA) [12], tetraamino macrocyclic ligands (TAML) [13], and porphyrins [14] can be used to extend the useful range of the Fe(III)/H₂O₂ system to neutral pH conditions by preventing iron precipitation and producing a more reactive oxidant [15, 16]. However, these organic ligands scavenge the oxidants produced by the system, lowering the efficiency of contaminant oxidation, and resulting in a need to continuously supply ligands. The ligands also pose secondary environmental risks because they can mobilize toxic metals [17].

The addition of phosphotungstate (PW₁₂O₄₀³⁻, a type of polyoxometalates) may provide an alternative approach for extending the useful range of the Fe(III)/H₂O₂ system without using organic ligands. PW₁₂O₄₀³⁻ forms soluble complexes with Fe(III) under neutral pH conditions [18], and is relatively nontoxic and resistant to oxidation [19, 20]. It also retains its catalytic activity when it is covalently bound to silica surfaces [21]. The objective of the present study was to evaluate the potential for PW₁₂O₄₀³⁻ to enhance oxidant production in the Fe(III)/H₂O₂ system under neutral pH conditions. For this purpose, the oxidation of compounds that are known to react with ^•OH or Fe(IV) (i.e., benzoate and methanol) by the Fe(III)/H₂O₂ system was studied under various conditions in the absence and presence of PW₁₂O₄₀³⁻.

2. Experimental

2.1. Reagents

All chemicals except for 2,4-dinitrophenyl hydrazine (DNPH) were of reagent grade and were used without further purification. DNPH was recrystallized three times from acetonitrile. All chemicals were obtained from Fisher Scientific Inc. except for DNPH, ferrous sulfate, and benzoic acid, which were obtained from Sigma-Aldrich Co.. All solutions were prepared using 18 MΩ. Milli-Q water from a Millipore system. The Fe(III) stock solution (50 mM) was prepared daily by dissolving ferric perchlorate (Fe(ClO₄)₃·9H₂O) in 0.1 N HClO₄. When necessary, 50 mM ferrous sulfate (in 0.1 mM HCl) and 100 mM H₂O₂ stock solutions were also prepared prior to experiments. Silica-immobilized PW₁₂O₄₀³⁻ (SiO₂-PW₁₂O₄₀³⁻) was prepared by a sol-gel hydrothermal method, and the catalyst prepared by this method typically has a PW₁₂O₄₀³⁻ content of 20 wt% and a BET surface area of 326 m²/g [18, 21].

2.2. Oxidation of probe compounds

Benzoate and methanol were used as probe compounds for detecting oxidants. Details regarding the reactivity of probe compounds with ^•OH and Fe(IV) are described elsewhere [11]. In most of the experiments, the disappearance of 0.5 mM benzoate was observed to quantify oxidant production. For some experiments, an excess of methanol (i.e., 200 mM) was employed to ensure that all of the oxidants reacted with the probe compound, and its major oxidation product (i.e., HCHO) was quantified.

All experiments on the oxidation of probe compounds were performed in 60-mL Pyrex vials at room temperature (20 ± 2°C) under air-saturation ([O₂]₀ = 0.25 mM). Borate buffer solutions (2 mM) were used for most of the experiments (pH 6–9). 1.5 mM Piperazine-N,N’-bis(ethanesulfonic acid) (PIPES; 1.5 mM) was used as a buffer for the experiments with 200 mM methanol at pH 7. The solution pH usually decreased during the reaction with borate buffer solutions (by 0.2 to 1 units). The pH was measured throughout each experiment, and the average pH value was calculated for each experiment.

The experiments were initiated by adding an aliquot of H₂O₂ stock solution to a pH-adjusted solution containing the probe compound, Fe(III), and PW₁₂O₄₀³⁻. Samples were withdrawn at pre-determined time intervals using a 5-mL glass syringe and immediately filtered through a 0.22-µm nylon syringe filter. For the experiments with benzoate, 25 µL of methanol was added to the 5 mL samples to quench the reaction. All of experiments were carried out in triplicate, and average values and standard deviations are presented.

2.3. Analytical methods

Benzoate (BA) was analyzed using high-performance liquid chromatography (HPLC) with UV absorbance detection at 270 nm. The concentration of HCHO was determined by HPLC with UV absorbance detection at 350 nm after DNPH derivatization [22]. Separation was performed on a Waters Symmetry C18 column (150 mm × 4.6 mm, 5 µm), using water with 10 mM nitric acid and acetonitrile as the eluent, at a flow rate of 1.0 mL min⁻¹. UV/visible absorbance of PW₁₂O₄₀³⁻ and iron solutions were analyzed using a Perkin Elmer Lambda 14 spectrophotometer with a quartz cell with a 1-cm optical path length. The concentration of H₂O₂ was quantified spectrophotometrically by either the titanium sulfate method [23] or the N,N-diethyl-p-phenylenediamine (DPD) method (for measuring H₂O₂ concentration less than 1 mM [24]).

3. Results

3.1. Enhanced oxidation of benzoate in the Fe(III)/H₂O₂ system in the presence of PW₁₂O₄₀³⁻

The oxidative degradation of benzoate (BA) and the decomposition of H₂O₂ in the Fe(III)/H₂O₂ system were investigated in the absence and presence of PW₁₂O₄₀³⁻. In the absence of PW₁₂O₄₀³⁻, the concentration of BA decreased by only 14% over 4 hr with 1 mM Fe(III) and 50 mM H₂O₂ (Figure 1a), while 62% of H₂O₂ was decomposed (Figure 1b). Increasing the concentration of Fe(III) to 4 mM resulted in the loss of approximately 33% of BA as the rate of H₂O₂ decomposition increased (complete decomposition after 2 hr). Control experiments conducted in the absence of H₂O₂ indicate that adsorption of BA on Fe(III)-oxyhydroxides accounted for approximately 33% of the observed loss of BA (4 mM Fe(III) only in Figure 1a).

FIGURE 1.

BA loss and H₂O₂ decomposition in the Fe(III)/H₂O₂ and the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ systems ([H₂O₂]₀ = 50 mM; [PW₁₂O₄₀³⁻]₀ = 1 mM; [BA]₀ = 0.5 mM; pH_i = 7.5, pH_ave = 7.3 for the Fe(III)/H₂O₂ system; pH_i = 8.0, pH_ave = 7.0 for the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system)

The addition of PW₁₂O₄₀³⁻ (1 mM) enhanced the rate of BA loss (Figure 1c), and decreased the rate of H₂O₂ decomposition (Figure 1d). With 1 mM Fe(III) in the presence of PW₁₂O₄₀³⁻, 43% of BA was transformed in 4 hr, whereas the H₂O₂ concentration decreased by only 8%. Increasing the concentration of Fe(III) enhanced both the rate of BA loss and the rate of H₂O₂ decomposition. No significant removal of BA was observed in the absence of H₂O₂ (Figure 1c).

3.2. pH Dependence

The Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system exhibited a higher degree of BA oxidation and less H₂O₂ decomposition than the Fe(III)/H₂O₂ system after 3 hours over pH values ranging pH 6.3–8.8. In both the Fe(III)/H₂O₂ and Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ systems, BA removal decreased, while the loss of H₂O₂ increased with increasing pH (Figure 2a & 2b). These trends were most clearly evident in the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system above pH 7.5. The calculated H₂O₂ utilization efficiency (i.e., Δ[BA]/Δ[H₂O₂]×100) in the Fe(III)/PW₁₂O₄₀³⁻/ H₂O₂ system was approximately 3.5% at pH values ranging from 6.3 to 7.7, and decreased to approximately 0.5% at pH 8.4 after that region (Figure 2c). The H₂O₂ utilization efficiency in the Fe(III)/H₂O₂ system was less than 0.5% over the entire pH range.

FIGURE 2.

Effect of pH on BA degradation (a), H₂O₂ decomposition (b), and the H₂O₂ utilization efficiency (c) in the Fe(III)/H₂O₂ and the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ systems ([Fe(III)]₀ = 2 mM; [H₂O₂]₀ = 50 mM; [PW₁₂O₄₀³⁻]₀ = 1 mM; [BA]₀ = 0.5 mM; reaction time = 4 hr)

3.3. Effect of aging

PW₁₂O₄₀^3– hydrolyzes into PO₄^3– and WO₄²⁻ under neutral pH conditions [25, 26]. To investigate the stability of the Fe(III) complexes with PW₁₂O₄₀^3–, a stock solution containing 2 mM Fe(III) and 1 mM PW₁₂O₄₀³⁻ was prepared, and aged in borated-buffered solution at 7.5. An aliquot of the stock solution was withdrawn on a daily basis, and the transformation of BA and the decomposition of H₂O₂ were examined after adding 0.5 mM BA and 50 mM H₂O₂.

The catalytic activity of the Fe(III) complexes with PW₁₂O₄₀³⁻ gradually decreased over 7 days as PW₁₂O₄₀³⁻ underwent hydrolysis. The rate of the BA transformation decreased (Figure 3a), and the rate of H₂O₂ decomposition increased (Figure 3b) as Fe(III)-PW₁₂O₄₀³⁻ aged. The solution of Fe(III)-PW₁₂O₄₀³⁻ turned turbid as the solution aged possibly due to the formation of iron precipitates. PW₁₂O₄₀³⁻ aged for 1 day in the absence of Fe(III) lost all of its catalytic activity with respect to BA oxidation when Fe(III) and H₂O₂ were added to the aged solution (no. 12 in Table 1), indicating that PW₁₂O₄₀³⁻ is stabilized by Fe(III) complexation.

FIGURE 3.

Effect of aging (at pH_ave 7.5) on BA degradation (a) and H₂O₂ decomposition (b) in the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system ([Fe(III)]₀ = 2 mM; [H₂O₂]₀ = 50 mM; [PW₁₂O₄₀³⁻]₀ = 1 mM; [BA]₀ = 0.5 mM; pH_i = 8.0, pH_ave = 7.0)

TABLE 1.

Effects of reagent concentrations, a ^•OH scavenger (tert-butanol), and SiO₂-PW₁₂O₄₀³⁻ on the BA degradation, the H₂O₂ decomposition and the H₂O₂ utilization efficiency.

				Conditions			BA removal (%)	H2O2 decomposition (%)	Efficiency (%) (Δ[BA]/Δ[H2O2] × 100)
No.	[Fe(III)]0 (mM)	[H2O2]0 (mM)	[PW12O403−]0 (mM)	[BA]0 (mM)	pH (pHi, pHf, pHave)	reaction time, additives, etc.
1	2	50		0.5	7.5, 7.2, 7.3	4 hr	14±4.6	70±5.8	0.20±0.064
2	2	50	1	0.5	8.0, 6.6, 7.0	4 hr	87±7.4	26±3.2	3.4±0.64
3	2	50	1	2	8.0, 6.3, 6.4	4 hr	46±2.8	23±1.6	8.2±0.39
4	2	50	1	5	8.0, 6.1, 6.2	4 hr	27±1.3	20±1.1	14±0.50
5	2	25	1	0.5	8.0, 6.6, 6.8	4 hr	74±9.0	25±0.38	5.9±0.36
6	2	5	1	0.5	8.0, 7.0, 7.1	4 hr	56±5.1	35±2.5	16±0.94
7	2	50	2	0.5	8.0, 6.2, 6.5	4 hr	87±4.4	14±2.5	6.2±0.58
8	2	50	0.5	0.5	8.0, 6.5, 6.8	4 hr	89±8.0	29±3.8	3.1±0.30
9	2	50	0.25	0.5	8.0, 6.8, 7.2	4 hr	67±4.3	31±3.0	2.1±0.053
10	2	50		0.5	7.5, 7.2, 7.3	4 hr,200 mM tert-butanol	13±4.3	85±0.18	0.15±0.050
11	2	50	1	0.5	8.0, 6.8, 7.1	4 hr, 200 mM tert-butanol	16±0.21	20±2.1	0.81±0.075
12	2	50	1a	0.5	7.0, 8.1, 8.5	4 hr	11±0.34	46±9.1	0.23±0.023
13	2	50		0.5	7.0, 8.3, 8.4	4 hr, 10 mM WO42−	17±3.0	69±2.5	0.24±0.022
14	2	50	1b	0.5	8.0, 7.2, 7.4	4 hr	31±2.5	54±2.9	0.58±0.067
15	2	50	5b	0.5	8.0, 6.8, 7.0	4 hr	50±4.6	33±3.8	1.5±0.020

^aAged for 1 day at pH_ave 7.5 in the absence of Fe(III),

^bSiO₂-PW₁₂O₄₀³⁻ (1 mM as PW₁₂O₄₀³⁻ ≈ 2.9 g/L SiO₂-PW₁₂O₄₀³⁻)

3.4. Effects of reagent concentrations, a ^•OH scavenger, SiO₂-PW₁₂O₄₀³⁻, and methanol as a probe compound

The efficiency of the BA degradation in the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system was investigated varying the initial concentrations of H₂O₂, BA, and PW₁₂O₄₀³⁻ (no. 1–9 in Table 1). Increasing the initial concentration of BA from 0.5 to 5 mM increased the H₂O₂ utilization efficiency from 3.4 to 14% mainly through increased loss of BA (no. 2–4 in Table 1). Decreasing the concentration of H₂O₂ from 50 to 5 mM resulted in a similar degree of enhancement (no. 2, 5–6 in Table 1) because H₂O₂ competes with BA for the oxidant. When the concentration of PW₁₂O₄₀³⁻ was reduced from 2 to 0.25 mM, the H₂O₂ utilization efficiency decreased from 6.2 to 2.1% (no. 2, 7–9 in Table 1).

In the presence of PW₁₂O₄₀³⁻, the addition of 200 mM tert-butanol, which is an efficient ^•OH scavenger, lowered the extent of BA transformation, and reduced the H₂O₂ utilization efficiency (no. 11 in Table 1). In the absence of PW₁₂O₄₀³⁻, there was little change in BA transformation or H₂O₂ decomposition upon addition of tert-butanol (no. 1 & 10 in Table 1).

Relative to homogeneous PW₁₂O₄₀³⁻, SiO₂-PW₁₂O₄₀³⁻ exhibited a much lower H₂O₂ utilization efficiency (no. 14 & 15 in Table 1)

To measure maximum yields of oxidant, an excess of methanol (i.e., 200 mM), which is known to react with both ^•OH and Fe(IV) [11], was used as a probe compound, and its oxidation product, formaldehyde (HCHO) was quantified. In the Fe(III)/H₂O₂ system, the yields of HCHO were approximately 6% at neutral pH values (no. 1 & 2 in Table 2), whereas the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system with 1–2 mM PW₁₂O₄₀³⁻ exhibited HCHO yields ranging from 42–64% (no. 3 & 4 in Table 2).

TABLE 2.

Methanol oxidation into HCHO, the H₂O₂ decomposition, and the H₂O₂ utilization efficiency

				Conditions			HCHO formed (mM)	H2O2 decomposed (mM)	Efficiency (%) (Δ[HCHO]/ Δ[H2O2]×100)
No.	[Fe(III)]0 (mM)	[H2O2]0 (mM)	[PW12O403−]0 (mM)	[methanol]0 (mM)	pH (pHi, pHf, pHave)	reaction time
1	2	1		200	7.0, 6.5, 6.7	4 hr	(5.7±0.38)×10−2	1	5.7±0.38
2	2	1		200	8.0, 7.8, 7.9	4 hr	(5.5±0.26)×10−2	1	5.5±0.26
3	2	1	1	200	8.0, 6.8, 7.0	10 hr	0.42±0.029	1	42±2.9
4	2	1	2	200	8.0, 6.1, 6.4	10 hr	0.64±0.017	1	64±1.7
5	2	1		200	2.5, 2.6, 2.6	10 hr	0.69±0.037	1	69±3.7

4. Discussion

4.1. Catalytic decomposition of H₂O₂ by Fe(III)

The formation of ^•OH in the Fe(III)/H₂O₂ system under acidic pH conditions usually has been explained by a free radical mechanism (reactions 1–4).

$$\mathrm{Fe}(\text{III})+\mathrm{H}{{\mathrm{O}}_2}^{•}\to \mathrm{Fe}(\text{II})+{\mathrm{O}}_2+{\mathrm{H}}^{+}{k}_{\mathit{3}}=2.5\times {10}^5{\mathrm{M}}^{-1}{\mathrm{s}}^{-1}\text{at pH}2.5[27,28]$$ (3)

$$\mathrm{Fe}(\text{II})+\mathrm{H}{{\mathrm{O}}_2}^{•}\underrightarrow{H^{+}}\mathrm{Fe}(\text{III})+{\mathrm{H}}_2{\mathrm{O}}_2{k}_{\mathit{4}}=1.2\times {10}^6{\mathrm{M}}^{-1}{\mathrm{s}}^{-1}[29]$$ (4)

In this mechanism, the rate-determining step is the slow reduction of Fe(III) by H₂O₂ (reaction 2), which is followed by the Fenton reaction to produce ^•OH (reaction 1). The HO₂^• produced from reaction 2 either reduces Fe(III) or oxidizes Fe(II) (reactions 3 & 4). In the presence of high concentration of H₂O₂, Fe(III) is the predominant species of iron ([Fe(II)]/[Fe(III)] < 0.005 [6]), and the rate of reaction 4 is slow compared to reaction 3. According to reactions 1–3, the theoretical yield of ^•OH from H₂O₂ (Δ[^•OH]/Δ[H₂O₂]×100) is 67% because three moles of H₂O₂ are consumed for every two moles of ^•OH produced (i.e., reaction 1 + 2 × reaction 2 + reaction 3). The highest observed yield of HCHO from the oxidation of methanol in the Fe(III)/H₂O₂ system at pH 2.6 (69%; no. 5 in Table 2) is consistent with the theoretical value.

Under neutral pH conditions, Fe(III) forms the amorphous precipitates of Fe(III)-oxyhydroxides consisting of iron centers linked together by oxy and hydroxyl bridges. These heterogeneous Fe(III) species are known to generate very low yields of oxidants from the catalytic decomposition of H₂O₂ [7, 8], which is consistent with our observations for the oxidation of BA and methanol (Figure 1a & 1b; no. 1 & 2 in Table 2). The low yield of oxidants from the heterogeneous iron catalysts is believed to be due to reactions occurring on or near the surface that directly decompose H₂O₂ into water and oxygen [8].

In addition, the reaction of Fe(II) with H₂O₂ produces oxidants other than ^•OH under neutral pH conditions. In the absence of Fe(II)-complexing ligands, Fe(IV) appears to be the predominant oxidant produced at pH values above 5 (reaction 5 [9, 10]). Fe(IV) rapidly oxidizes methanol to HCHO, but does not react with BA or other aromatic compounds [11].

$$\mathrm{Fe}(\text{II})+{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{Fe}(\text{IV})(\mathrm{Fe}{\mathrm{O}}^{2+})+{\mathrm{H}}_2\mathrm{O}$$ (5)

4.2. Production of oxidants by iron-PW₁₂O₄₀³⁻ complexes

PW₁₂O₄₀³⁻ forms soluble complexes with Fe(III) and Fe(II) [18]. The stoichiometry between iron and PW₁₂O₄₀³⁻ in these complexes is not precisely known, but it appears that a single molecule of PW₁₂O₄₀³⁻ is capable of coordinating several iron atoms because the rate of BA transformation continues to increase as the ratio of Fe(III) to PW₁₂O₄₀³⁻ increases from 1:1 to 4:1 (Figure 1c). The Fe(III)/Fe(II)-PW₁₂O₄₀³⁻ complexes are believed to catalyze the conversion of H₂O₂ into oxidants through a mechanism analogous to reactions 1–4. An equimolar mixture of Fe(III) and PW₁₂O₄₀³⁻ under neutral pH conditions produced a 64% yield of HCHO (no. 4 in Table 2), which is approximately equal to the predicted maximum, whereas the yields of HCHO were around 5.6% under similar conditions in the absence of PW₁₂O₄₀³⁻ (no. 1 & 2 in Table 2).

The identity of the oxidant produced in the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system is still unclear. The significant degradation of BA observed at neutral pH values and the inhibition of the BA degradation by the addition of tert-butanol (no. 11 in Table 1) suggest that the oxidant produced in the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system is much more reactive than the Fe(IV) species formed by the Fenton reaction under neutral pH conditions (reaction 5). Nonetheless, solution-phase ^•OH does not appear to be the only oxidant formed in this system because the BA degradation did not behave as expected when the concentrations of BA, H₂O₂, and ^•OH scavengers were varied: the fraction of ^•OH that react with BA (f_•OH,BA) can be expressed by equation 6, and the H₂O₂ utilization efficiency for BA degradation (Δ[BA]/Δ[H₂O₂]) should be approximately proportional to the f_•OH,BA value.

$$f_{•OH,BA}=k_{•OH,BA}[\mathrm{B}\mathrm{A}]/(k_{•OH,BA}[\mathrm{B}\mathrm{A}]+k_{•OH,H2O2}[{\mathrm{H}}_2{\mathrm{O}}_2]+\mathrm{\Sigma }{k}_{•OH,S}[\mathrm{S}])$$ (6)

Where, k_{•OH, H2O2} is 2.7 × 10⁷ M⁻¹ s⁻¹, k_{•OH, BA} is 5.9 × 10⁹ M⁻¹ s⁻¹ [30], and Σk_{•OH, S}[S] represents the rate constants and concentrations of other ^•OH scavengers (e.g., tert-butanol).

In the presence of 0.5 mM BA and 50 mM H₂O₂, the f_•OH,BA value is calculated to be 0.68. When the concentration of BA increases by an order of magnitude (from 0.5 to 5 mM) or the concentration of H₂O₂ decreases by an order of magnitude (from 50 to 5 mM), the calculated f_•OH,BA value increases from 0.68 to 0.96. However, the observed H₂O₂ utilization efficiency increased by approximately a factor of five under these conditions (from 3.4 to 14–16%; no. 2, 4, & 6 in Table 1), which indicates that the oxidant formed in this system is more reactive with H₂O₂ than BA. In addition, the inhibitory effect of tert-butanol on the BA degradation is not as high as predicted by its ^•OH rate constant (k_{•OH, tert-butanol} = 6.6 × 10⁸ M⁻¹ s⁻¹ [30]): addition of 200 mM tert-butanol should have decreased the H₂O₂ utilization efficiency of BA by almost two orders of magnitude in the presence of 0.5 mM BA and 50 mM H₂O₂, compared to the factor of four decrease observed in Table 1 (no. 11). The inconsistency between experimental observations and predictions based on ^•OH reaction rates suggests that other oxidants such as caged ^•OH [31] or a more reactive Fe(IV) species coordinated by PW₁₂O₄₀³⁻ are produced in the presence of PW₁₂O₄₀³⁻.

4.3. Stability of Fe(III)-PW₁₂O₄₀³⁻ complexes

The production of oxidants during the catalytic decomposition of H₂O₂ does not alter the structure or reactivity of PW₁₂O₄₀³⁻ as verified by the observation that the UV absorption band of Fe(III)-PW₁₂O₄₀³⁻ (ε_255nm ≈ 35000 M⁻¹ cm⁻¹) was not significantly changed after the reaction (under the conditions of no. 2 in Table 1; data not shown). However, PW₁₂O₄₀³⁻ undergoes stepwise hydrolysis, to ultimately produce PO₄^3– and WO₄²⁻, under neutral and alkaline pH conditions [25, 26]. The hydrolysis rate increases with pH, with half lives of PW₁₂O₄₀³⁻ at pH 7 and 8 estimated to be 75 and 19 min, respectively, at 20°C [25]. This process explains the reduction in H₂O₂ utilization efficiency observed when a PW₁₂O₄₀³⁻ solution was stirred for 1 day at pH 7.5 before adding Fe(III) (0.58%; no. 12 in Table 1). A similar H₂O₂ utilization efficiency was observed when 10 mM WO₄²⁻ (i.e., the major hydrolyzed product of PW₁₂O₄₀³⁻) was added to the Fe(III)/H₂O₂ system under identical conditions (no. 13 in Table 1).

The Fe(III)-PW₁₂O₄₀³⁻ complexes appear to be more resistant to hydrolysis than uncomplexed PW₁₂O₄₀³⁻. The catalytic activity of Fe(III)-PW₁₂O₄₀³⁻ gradually decreased when the catalyst was aged for 7 days at pH 7 (Figure 3). The Fe(III)-PW₁₂O₄₀³⁻ complexes likely consist of a sandwich-type structure with two Keggin fragments linked by metal atoms. Such structures are known to be more stable under neutral pH conditions than the typical Keggin-type polyoxometalates, XW₁₂O₄₀ⁿ⁻ [32].

Despite the enhanced stability of the catalyst, the gradual loss of the Fe(III)-PW₁₂O₄₀³⁻ complexes by hydrolysis may still restrict the use of this system. PW₁₂O₄₀³⁻ is not used commonly in water treatment or manufacturing applications and it is therefore difficult to predict the cost of a full-scale treatment system. Further research may be needed to develop more stable forms of iron-polyoxometalate catalysts or less expensive ways of producing POM.

4.4. Reuse of PW₁₂O₄₀³⁻

The retention and subsequent reuse of PW₁₂O₄₀³⁻ is also necessary for the cost-effective application of the Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system. It may be possible to use nanofiltration to retain the Fe(III)-PW₁₂O₄₀³⁻ complexes while allowing water, salts, and dissolved solutes to pass through a treatment system. For example, Kim et al. (2008) have recently demonstrated the hybrid system of the homogeneous Fenton-like oxidation using Fe(III)-tetrasulfophthalocyanine (Fe(III)-TsPc) with nanofiltration under acidic pH conditions [33]. Use of a supported catalyst such as SiO₂-PW₁₂O₄₀³⁻ also could be used to retain PW₁₂O₄₀³⁻. However, SiO₂-PW₁₂O₄₀³⁻ exhibited a much lower H₂O₂ utilization efficiency than PW₁₂O₄₀³⁻ (i.e., 3.4% for PW₁₂O₄₀³⁻ vs. 0.58% for SiO₂-PW₁₂O₄₀³⁻; no. 2 & 14 in Table 1), probably due to the limited sites for the Fe(III) complexation on the catalyst surface. As a result, a supported catalyst may require a larger amount of PW₁₂O₄₀³⁻ than the homogeneous system.

5. Conclusions

The Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system is a new homogeneous Fenton-like system that is capable of oxidizing recalcitrant organic compounds under neutral pH conditions. Catalytic reactions of iron-PW₁₂O₄₀³⁻ complexes convert H₂O₂ into strong oxidants with a relatively higher efficiency. Due to the use of PW₁₂O₄₀³⁻, which is a relatively nontoxic and resistant to oxidation, this system offers many advantages over other Fenton-like systems using organic ligands in terms of the system efficiency and environmental safety. The Fe(III)/PW₁₂O₄₀³⁻/H₂O₂ system can be useful for treatment of water contaminated with toxic organic compounds, but issues associated with the eventual loss of activity and approaches for retaining and reusing the catalyst need to be further investigated for its effective application.