Improving Fenton-like System with Catechin, an Environmental-friendly Polyphenol: Effects and Mechanism

Fenton-like system had long been modified to promote its adaptability and efficiency by adding extra components like metal chelators or reducers. Nonetheless, chelator such as ethylene diamine tetraacetic acid (EDTA) was harmful in high dosage. While strong reducers could efficiently promote the generation speed of hydroxyl radicals (HO • ), they could also get oxidized easily during storage time, for which further application was costly. In this study, (+)-catechin (CAT), an environmentalfriendly component, was proved to be an efficient promoter of Fenton-like system on Bisphenol A (BPA) degradation. The modified system achieved 92% of BPA removal in one hour. The mechanism was studied, and the reaction procedure between CAT and Fe(III) together with the degradation pathways of CAT were identified. Fe(III) got chelated first by CAT and quickly reduced to Fe(II) via a single inner-compound electron transfer. The intermediates of CAT could further reduce one and a half more potion of Fe(III) on average. Consequently, massive amount of Fe(II) turned Fenton-like procedure to Fenton reaction and thus promoted the generation of HO • .  Corresponding author. Tel.: +86 21-31242030. E-mail addresses: wuyanlin@fudan.edu.cn (Y. Wu), wbdong@fudan.edu.cn (W. Dong)


Introduction
Since the Fenton reagent was firstly introduced by H.J.H Fenton in 1894 [1], it had been developed to several branches, such as Fenton-like process, photo-Fenton process, and ligand-modified Fenton process to extend its applicability and improve efficiency. Those processes had shown high efficiency of degrading some persistent organic pollutants (POPs) or endocrine disruptor chemicals (EDCs) by producing large amount of HO • based on the reaction illustrated as Eqs (1)-(4) (Fenton-like process: 1-4; Fenton process: 4). In latest cases, the Fenton and Fenton-like processes had already been applied in industrial wastewater treatment in real situation. Bautista et al. [2] utilized traditional Fenton system to treat wastewater of low BOD 5 /COD rate from cosmetic industry. Changotra et al. [3] treated real pharmaceutical wastewater with three different Fenton processes combined with aerobic biological treatment.
Although those applications were proved to be successful to some extent, there were still some defects yet to improve. The major limitation of Fenton processes was pH value. To overcome the limitation on pH, chelators such as EDTA, oxalate, citrate and [S,S]-Ethylene diamine-N,N′-disuccinic acid (EDDS) were added to promote the system. Hu et al. [4] achieved 92.7% malachite green degradation in neutral pH with EDTA-modified Fenton-like system. Manenti et al. [5] tested oxalate, citrate and EDDS as organic ligands and accomplished 69.1% mineralization with modified photo-Fenton system. However, EDTA had low biodegradability [6][7][8]. Hence, it was crucial to find an efficient chemical compound which i) improved the Fenton process efficiency and ii) was easy to degrade.
Plant polyphenols were a group of organic compounds with multiple phenol structures generated in plant metabolism procedure. They were often detected in a vast amount in roots, leaves, fruits, mass-produced food and beverage made from plants. According to previous researches, fruits such as grapes and berries might contain 0.3‰ of polyphenol in fresh weight, especially in tea leaves [9]. The concentration of polyphenols could reach up to 30% in dry weight of green tea [10].
The naturally high productivity showed the potential for exploitation and manufacture of certain polyphenol. Polyphenols were also believed to bring health benefits on cardiovascular protection and anti-oxidation, and they were becoming popular as the additive or health care products [11][12][13]. The associated benefits of these compounds were often based on the reducibility of polyphenols [14]. From the perspective of chemical structure, the sensitivities toward oxidant of both o-phenolic hydroxyl group and p-phenolic hydroxyl group were accounted for the reducibility of polyphenol which raised the possibility to modify traditional Fenton-like system as a reducer.
Catechin was polyphenol classified under flavonoid category. They were typical flavan-3-ols, characterized by a resorcinol group on its A ring, and a catechol group on its B ring. Due to the two chiral carbons, there were four isomers, including (+)-catechin, (-)-catechin, (-)-epicatechin and (+)-epicatechin. However, (+)-catechin (Scheme 1) was the major catechins present in green tea [15,16] and was chosen as the research target in this study. As stated in previous studies [17], the catechol group on its B ring provides possible chemical reactivity both in chelation and reduction, whereas the process was not clear. Wang et al. [18] studied the photodegradation of inderal in a UV/Fe(III)/CAT system at near neutral pH. The high degradation efficiencies were obtained and they were strongly dependent on pH, following the order pH 3.0 < 5.0 < 7.0 < 6.0. The reaction between Fe(III) and CAT was considered as chelation. Wu et al. [19] reported the degradation of atenolol in Fe(III)/CAT /S 2 O 8

2-
system at acid condition (pH=3.3). The mechanism was concluded as the reduction of Fe(III) by CAT to produce Fe(II) and PS was activated by Fe(II) to produce large amount SO 4 •− . In general, there were three major explanations of the reaction between Fe(III) and CAT respectively, regarding CAT as a ligand [18], a reducer [19,20] or both a ligand and reducer at the same time [17], yet none of these theories had been fully confirmed. Based on the chemical reactivity and its environmental-friendly feature as a plant polyphenol, CAT was potentially suitable to modify the conventional Fenton system.

Scheme 1.
Structure of (+)-catechin BPA was a kind of EDCs which had estrogenic activity by interacting with estrogen receptors in human body. As reported in previous study, the maximum concentration in wastewater could reach 0.15 mg/L in some Canadian and US municipal wastewater treatment plants [21], and because it was hard to biodegrade [22][23], leftover BPA in effluent could still be detected. In this study, BPA was used as a model pollutant. The effect of CAT on the modified Fenton-like system on BPA degradation was investigated. Mechanism of the reaction between CAT and Fe(III), radical identification, effect of pH value and key influencing factors were carefully studied. The degradation pathway of BPA was also reported. added at the concentration level of this study [24]. Fe(III) were the last one to be mixed into the solution of BPA, CAT and H 2 O 2 , working as a trigger of the degradation procedure and making this step the time zero point. Immediately after each sampling, methanol was added to the samples in order to scavenge radicals and stop the degradation. To prevent photolysis, the experiments were conducted in dark brown glass vials and HPLC vials were sheltered from light using an auto-sampler with a hood. All degradation experiments were conducted at least twice.

Analysis method
The concentration of BPA was determined by high-performance liquid chromatography (U3000, Dionex, USA) with Diode Array Detector (HPLC-DAD).
The detecting wavelengths were set at 226 nm and 280 nm to preclude the interference from the signal of intermediates. Samples of 10 μL each, pretreated with methanol were injected into a C-18 reversed phase column (150 mm × 4.6 mm × 5 μm, Kromasil, Sweden). The temperature of the column oven was set at 40°C. The mobile phase containing 70% of methanol (phase A) and 30% of water (phase B) was pumped into the system at the flow rate of 1 ml min -1 [25]. Fe(II) concentration were spectrophotometrically measured based on the reaction between Fe(II) and ferrozine which formed a complex that was detected by UV-vis spectrophotometer at 562 nm [26].

Stopped-flow Method for Intermediates Analysis.
Intermediates produced by the reaction between CAT and Fe(III) over the time span of several seconds were identified by UV-vis spectrum equipped with Stopped-flow mixing system (Applied Photophysics, SX20). The system was set to take 1000 scans in each time gap from 350 nm to 750 nm.

Computational method.
The main computational methods were density functional theory (DFT) and time-dependent density functional theory (TDDFT). Gaussian09 program (Frisch et al., 2009) was used to perform both DFT and TDDFT calculations, and the three-parameters hybrid density-functional (B3LYP) was used in all DFT calculations setting to 6-31 + g (d,p) basis. All excited-state geometry optimizations as open shell species were described with an unrestricted approach, while ground-state geometry optimizations as closed-shell species were explained with a restricted method.

Effect of Fenton-like system modified by CAT on BPA degradation
The effect of CAT-modified Fenton-like system on degradation of BPA was compared with that of conventional Fenton-like system and photo

Role of ROS in the degradation process
Reactive oxygen species (ROS) were often considered as the core of traditional mT in magnetic field with four peaks in 1:2:2:1 ratio, which were the characteristic pattern for HO • [28]. Ethanol was used as solvent in another test to quench HO • , so that patterns from other radicals were not affected by signal overlapping. As illustrated in Fig. 2b, the dominant signal was in 1:1:1:1 ratio, accompanied with two minor signals after the second and third peak which was often interpreted as the existence of superoxide anion radicals (O 2 •-) [29,30]. It was concluded from the ESR analysis that both HO • and O 2 •were produced.

Role of CAT
As the result above showed, CAT barely changed the way how traditional Fenton-like system worked which still relied on the oxidation by HO • , but it improved the efficiency of BPA removal. Although the mechanism was unclear, it was inferred that Fe(III) might react with CAT. To test the assumption, Fe(II) was determined in the mixture solution of Fe(III) (50 μM) and CAT (6.25-100 μM) (Fig. 4). As the result, more than 80% of Fe(III) were reduced to Fe(II) in an hour, when CAT : Fe(III) was more than 1:2. While the concentration of CAT was less than a quarter of Fe(III), the formed Fe(II) concentrations were varied depending on initial concentration of CAT, but with a constant proportional relationship of roughly 2.5 times of CAT. The degradations of CAT were also followed and the results were shown in Fig. SM2.
When the CAT concentrations were lower than 25 μM, they were quickly 100% consumed at the beginning 3 min of the reaction. Data of the experiment revealed a reduction process between CAT and Fe(III), which was in accordance to some studies on catechol and other polyphenol substances [34]. Nonetheless, previous studies also hinted that catechol or compounds with similar functional group tended to react as a complex agent when Fe(III) was involved [35,36]. The reduction procedure was, therefore, deduced to be a set of complicated reaction instead of a single elementary reaction. To gain a deeper insight into the reduction process, Stopped-flow UV-vis spectrum was obtained (as shown in Fig. 5). In the first second, absorbance rose steeply at 390 nm, 450 nm and 685 nm, implying formation of multiple compounds.
Four seconds after mixing, absorbance at two wavelengths (450 nm and 685 nm) peaked and diminished gently to a low level while the other one (390 nm) still kept on growing. Since the peaked times at those wavelength were different from each other, it could be confirmed that one terminal product and two intermediates were produced during the reaction. The structures were also double checked via computational method. To further confirm the structure of three products, the UV-vis spectrum of them were compared with the computational result of DFT model. The DFT model simulated the possible spectrum based on the input structure, and the results were in accordance with the spectrum obtained by the Stopped-follow UV-vis spectrum instrument (Table 1)

Effect of CAT/Fe(III) ratio, pH and common anions
To seek the best condition of the modified system on BPA degradation, several influencing factors were tested or introduced to the system. Ratio impact of CAT and Fe(III) was tested from 1/8:1 to 2:1. Based on the data, two different trends which were divided by inflection point between 1/2:1 and 1/4:1 were noticed (Fig. 6a).
When ratio went down from 2:1 to 1/2:1, BPA got removed much faster and more thoroughly over the experiment. As the ratio went deeper down to 1/4:1 or even lower, BPA removal gradually slowed down, but the similar removal effect was achieved.
The outcome was partially in conformity with the pattern of Fe(II) generation: as the CAT outnumbers Fe (III) in 1/3 or more, some CAT was spared and might compete on HO • that hindered BPA removal [37]. As for the similarity in removal effect in low CAT/Fe(III) ratio, it showed the efficiency of CAT modification was quite high.
Nonetheless, numerous products were created during the degradation process, which blocked further degradation. In order to get the best degradation result, following experiments applied the ratio of CAT and Fe(III) as 1/6:1. Concerning that Fe(III) formed different hydroxide species under different pH, the initial pH value was tested as another influencing factor. The data illustrated in Fig.   6b showed that the system worked well in both acidified and neutral environment.
There were more than 70% of BPA removal but the degradation of BPA was completely stopped if the pH value was above 8.0. It was known that Fe(III) and Fe(II) would completely precipitate in pH 8.0 or above, and as a result, the system stopped yielding HO • . On the other hand, less BPA was degraded as the pH went down in high acidity solution (pH=3). This phenomenon happened since H + competed with Fe(III) on CAT complexation and changed the distribution of Fe(III) species. Either of the factors might have negative impacts on BPA degradation.
In this study, all chemicals were solved in deionized water or HPLC class organic solvent, while in real water treatment, inlet water possibly contained varieties of anions that may bring unknown influence. To ensure the modified system was also suitable for water treatment accompanied with anions, several common anions were selected to add in the reaction solution (Fig. 6c)

Conclusion
The addition of CAT promoted the reduction reaction from Fe(III) to Fe(II) through a five-step process: complexation, reduction, release, second reduction and final reduction, leading to a higher productivity of HO • in the Fenton-like system.
Thus, CAT was proved to be an efficient agent to modify Fenton-like system in neutral pH on BPA degradation and very possibly on removal of many other compounds which was also sensitive towards HO • . Moreover, the result demonstrated that the modified system kept its efficiency not only in neutral pH in deionized water, but also in acidic environment or with common anions. The degradation pathway of BPA had also been studied (shown in Fig. SM3). Considering the high potential for massive production of CAT and other plant polyphenols, together with their negligible risk of ecosystem and human health, results of this study present a promising future of applying plant polyphenols in Fenton-like system to promote its efficiency in real water treatment.