Reduction of hexavalent chromium using epigallocatechin gallate in aqueous solutions: kinetics and mechanism

Abstract

The reduction of hexavalent chromium (Cr(VI)) to trivalent chromium (Cr(III)) by epigallocatechin gallate (EGCG) as a reductant in aqueous phase was studied in a batch experiment. Effects of EGCG concentration, reaction time, pH and temperature were examined. Acidic conditions and high temperature enhanced the reduction efficiency greatly. 96% of initial 100 μM Cr(VI) was reduced within 3 hours in the presence of equimolar EGCG at pH 3 and 25 °C. Cr(VI) reduction kinetics was evaluated under various initial Cr(VI) concentrations with excess EGCG over the pH range of 3–7.56, and results indicated that the reaction was first-order with respect to both the concentration of Cr(VI) and EGCG. Accordingly, two concurrent redox pathways involving zero and one proton may contribute to the observed overall rate dependence on pH. EGCG quinone, which has two hydrogen atoms removed from the trihydroxyphenyl B ring of the EGCG molecule, was identified as one of the oxidation products of EGCG. Our results highlight the potential of using EGCG for Cr(VI) reduction with both efficiency and safety aspects.

---

Introduction

Chromium is one of the most frequently used metal in various industrial processes such as electroplating, steel manufacturing, leather tanning, dyes, etc.^1,2 Chromium contamination mainly results from improper disposal and illegal discharge, and causes health problems and poses a severe threat to the environment in China.^3,4 The most common valence state of chromium existing in natural water and soils is trivalent (Cr(III)) and hexavalent (Cr(VI)). Generally, Cr(VI) is considered to be highly toxic to humans, animals and plants. In this oxidation state, the metal is water soluble in the full pH range and possesses a high mobility in the aqueous and soil phase.^4,5 In contrast, Cr(III) exhibits much less toxicity and is ready to form insoluble precipitates or bind to the soil surface under predominantly alkaline conditions. Hence, Cr(VI) reduction is of great significance in the remediation of environmental sites contaminated by hexavalent chromium.

Chemical methods are known as the most common strategy applied, converting Cr(VI) to Cr(III),^4,6 and numerous studies have been carried out on Cr(VI) reduction using reductants such as SO₂, H₂S,^7–9 divalent iron,^10,11 metallic iron,^12,13 zero-valent iron,^14,15 zero-valent magnesium,¹⁶ etc. These reducing agents, however, may exhibit toxicity themselves or produce solid wastes, which could introduce additional environmental problems without proper subsequent disposal.⁴ In comparison, Cr(VI) reduction using natural or naturally-derived reductants (ascorbic acid, flavonoids, leaves extract, etc.)^17–19 eliminate the worries of secondary pollution, providing a more environmental-friendly way for chromium contamination management.

Green tea is one of the most widely consumed beverages and famous for its abundant content of natural occurring polyphenols, especially the epigallocatechin gallate (EGCG) which consists of more than 50% mass of catechins and is the most potent antioxidant in green tea.²⁰ The reducing property of EGCG is considered of great value for its efficacies of anti-aging, cancer prevention, improving cardiovascular health.^20,21 In the field of medicine and food science, the antioxidant effects and oxidation mechanism of EGCG have been extensively studied in biological systems. Yet, to our knowledge, neither studies on application of EGCG for environmental purpose nor the interaction of hexavalent chromium with EGCG has been reported.

Also, numerous studies of Cr(VI) ‘adsorption-coupled reduction’ by green tea wastes^22–24 and other biomaterials have been published.²⁵ These studies clearly suggested that Cr(VI) was reduced to Cr(III) and phenolic or tannin groups may act as electron-donor in this process.^26,27 However, no works have been implemented to further evaluate the effects of these compounds on Cr(VI) reduction. Thus, our study may give more details of kinetics and mechanism of hexavalent chromium reduction by green tea polyphenols.

In this study, Cr(VI) reduction by EGCG in aqueous phase was investigated using potassium dichromate as the model pollutant. Variables including EGCG concentration, reaction time, pH and temperature were examined and reaction kinetics was measured in the pH range from 3 to 7.56. Meanwhile, an empirical kinetic equation and partial mechanism of Cr(VI) reduction by EGCG was also proposed.

Materials and methods

Materials and reagents

Epigallocatechin gallate was obtained from Jiamu biotechnology Co., Ltd (Hunan, China). Analytical-grade HCl, NaOH, H₂SO₄, H₃PO₄, K₂HPO₄, KH₂PO₄ and K₂Cr₂O₇ were purchased from Tianjin Kermel Chemical Reagent Co., Ltd., China. Formic acid and acetonitrile (both HPLC-grade) were from Sigma-Aldrich. All the chemicals were used without further purification. Milli-Q water (Millipore Corp., 18.2 MΩ cm) was used for all experiments. EGCG stock solution was prepared freshly prior to each experiment.

Experimental procedure

Batch experiments were performed in 100 mL polyethylene bottles, agitated on a rotary shaker at 180 rpm. Reaction mixtures were obtained by diluting potassium dichromate stock solution to an appropriate concentration, adjusting pH (by HCl or NaOH), and then adding different amount of EGCG, finally bring volume to 50 mL. After designed contact time, residual Cr(VI) concentrations were measured by a modified diphenylcarbazide colorimetric method at 540 nm. All testes were conducted in triplicate with a control experiment that no EGCG was added to the chromium solution.

Cr(VI) reduction kinetics study performed as variables initial Cr(VI) concentration, EGCG concentration and pH. During the examination of a specific variable, all other parameters were held constant. These parameters and their variability are presented in Table 1. Solution pHs were controlled by phosphate buffers: PBS1 (K₂HPO₄ and KH₂PO₄) for pH 4.7–7.5 and PBS2 (H₃PO₄ and K₂HPO₄) for pH 4–4.6. Solutions of pH under 4 were unbuffered and the initial pHs were adjusted by HCl. The concentration of each buffer is 0.025 M and the pH changes in all tests are less than 0.05 units, suggesting a sufficient buffer capacity. All other procedures were performed in accordance with the steps mentioned above. In addition, ionic strength was not examined as a variable in this paper since our preliminary experiments suggested that the effect of ionic strength is insignificant when the value is below 0.1 M. In this study, the ionic strength was always less than 0.03 M.

Table 1 Range of parameters in kinetics study

Parameter	Range studied
[Cr(VI)] (μM)	20–40
[EGCG] (μM)	200–800
pH	3.07–7.56
Temperature (°C)	25
Rotating rate (n/min)	180

---

Hexavalent chromium analytical method

Aqueous Cr(VI) concentration was analyzed by a modified diphenylcarbazide colorimetric method.²⁸ Matrix contains EGCG may interfere with the diphenylcarbazide test in two ways: (a) EGCG may presents absorbance at test wavelength, and (b) parallel Cr(VI) reduction may occur at the low pH of the test, preventing the color development of the diphenylcarbazide reagent. Blank experiment shows that no detectable absorbance was observed in EGCG solution at 540 nm, suggesting that EGCG did not disturb the measuring procedure. Whereas it was found that problem (b) did occur due to the acidification process before the addition of diphenylcarbazide (especially in the kinetics study where the amount of EGCG was far more than Cr(VI)). To solve this problem, a slightly modified colorimetric method was designed in our study. For influencing factor tests, an appropriate amount of sample was added directly into a 50 mL colorimetric tube containing 2 mL of 0.5% solution of diphenylcarbazide in 50% acetone and 0.5 mL (1 + 1) H₂SO₄. After 1 min development of the red-violet complex, its absorbance was analyzed photomatrically at 540 nm within 2 min. Kinetics study followed the same procedure described above except 5 mL of 1% solution of diphenylcarbazide in acetone were used for complexing. Owing to the large concentration of excess diphenylcarbazide used, the amount of Cr(VI) reduced by EGCG in the sample preparation process was minimized to a negligible level.

HPLC-MS/MS condition

HPLC-MS/MS analyses were performed on a 1290 infinity HPLC instrument coupled with 6400 series triple quadrupole mass spectrum (both from Agilent, USA). HPLC separation was carried out at 38 °C with isocratic elution using acetonitrile/water (14.5 : 85.5, v/v) with 0.1% formic acid. The mobile phase flow rate was 0.2 mL min⁻¹. Electrospray ionization (ESI) source was operated in positive ion detection mode with a desolvation temperature of 350 °C, a nebulizer pressure of 15 psi and a capillary voltage of 4000 V. Single ion monitoring (SIM) was performed by monitoring particular precursor ion from full-scan spectra with dwell time of 150 ms.

UV-VIS analysis

The UV/visible spectra were measured with a Hitachi Model 3900 spectrophotometer. 1.5 mL of Cr(VI) solution (200 μM) was mixed with equivalent volume of EGCG solution (200 μM) in a 1 cm quartz cuvette. pH of each solution was adjusted by HCl to 3 before mixing and spectra were recorded repetitively.

Results and discussion

Influencing factor tests

Effect of EGCG dosage on Cr(VI) reduction. Fig. 1a displays the relationship between initial EGCG concentration and residual concentration of Cr(VI) in solution. With the increase of EGCG dosage, Cr(VI) reduction efficiency raised dramatically from 5.5% to 91% as the EGCG concentration ranged from 10 μM to 100 μM. Further increase in EGCG dosage did not result in notable reduction enhancement since the original 100 μM Cr(VI) was almost completely removed within 2 hours as the EGCG concentration exceeded 100 μM. Therefore, an equimolar of initial Cr(VI) and EGCG (100 μM) was used in the following tests.

Fig. 1 Effect of (a) EGCG dosage, (b) reaction time, (c) pH, and (d) temperature on Cr(VI) reduction in aqueous solution. Experiments performed at Cr[(VI)] = 100 μM, [EGCG] = 100 μM, ([EGCG] = 20 μM for inset of (b)), pH = 3 (for a, b, and d), temperature = 25 °C (for a, b, and c), reaction time = 2 h (for a), and rotating rate = 180n/min. Error bars represent the standard deviation of triplicate runs.

Effect of reaction time on Cr(VI) reduction. The variations of Cr(VI) concentration as a function of time are illustrated in Fig. 1b. The reduction rate of Cr(VI) was fast at the starting 20 min, reducing more than 50% of the initial Cr(VI), and then slowed down gradually due to the depletion of Cr(VI). In general, a contact time of 3 hours was sufficient to achieve equilibrium and the reduction efficiency did not change significantly with further increase in reaction time. In fact, in a prolonged reaction time, a mole EGCG was capable of reducing at least more than 3 moles Cr(VI). As the inset of Fig. 1b shows, in the case of 20 μM EGCG added to solution containing excess Cr(VI) (100 μM), the reduction efficiency achieved 33% within 4 hours and increased gradually to 65% after 5 days. This result indicates that several steps may involved in this redox interaction: (a) one or several reactions with fast reaction rates finished quickly within the initial 4 hours and (b) successive reactions with slow rates last for days.

Effect of solution pH on Cr(VI) reduction. Solution pH is of vital importance affecting the Cr(VI) reduction speed since dichromate exhibits higher redox potential in acidic condition while in basic solutions are less oxidizing. As a whole, lower pH value resulted more Cr(VI) removed in solution (Fig. 1c). Within 2 hours of reaction, a high reduction efficiency was obtained and more than 90% Cr(VI) were reduced at pH 3 and lower. When pH increased to 4, the reduction efficiency declined sharply from 91% to 26%, and few Cr(VI) were reduced at neutral and moderately alkaline conditions. It is worth to mention that EGCG still possessed a certain reducibility to Cr(VI) in the weakly alkaline solutions after a contact time of 24 hours, proving the feasibility of chromium contaminated underground water and soil remediation using EGCG in a wide range of pH. At pH > 11, insignificant Cr(VI) reduction occurred, even after 96 hours. This is probably due to the autoxidation of EGCG in strong alkaline environment.²⁹

Effect of temperature on Cr(VI) reduction. The effect of temperature on Cr(VI) reduction efficiency was studied in the range of 5–45 °C. As Fig. 1d shows, high temperatures accelerated the reaction rate and higher reduction efficiencies were achieved, owing to more successful collisions between reaction particles. Generally, compared with ambient temperature, only a half time was needed to achieve the same reduction efficiency at 45 °C. Poonkuzhali et al. applied aqueous extract of Aerva lanata L for Cr(VI) reduction and found reduction efficiency dropped dramatically when reaction temperature exceeds 40 °C at pH 3. Continuous increase of Cr(VI) reduction rate rather than decline observed in our study would attribute to the thermal stability of EGCG at pH below 4.³⁰

Reaction kinetics

The results of influencing factor test suggests that the kinetics of Cr(VI) reduction by EGCG depends on reductants concentration and solution pH. To further explore the specific relationship between these variables and the Cr(VI) reduction rate, a certain assumption was made that the reaction kinetics adhere to the following expression:

−d[Cr(VI)]/dt = k_H[Cr(VI)]^x[EGCG]^y (1)

where [Cr(VI)] and [EGCG] are the concentrations of Cr(VI) and EGCG, x and y is the reaction order and k_H is the reaction rate (M⁻¹ s⁻¹) at given pH. When EGCG concentration is much higher than Cr(VI) concentration at fixed pH (controlled by phosphate buffer), the equation can be simplified as

−d[Cr(VI)]/dt = k_obs[Cr(VI)]^x (2)

where

k_obs = k_H[EGCG]^y (3)

Various initial concentrations (20–40 μM) of Cr(VI) were reduced by EGCG (800 μM) in phosphate-buffered solution at pH 6.86. Since the concentration of EGCG used was far more than Cr(VI) concentrations, it is reasonable to consider EGCG concentrations as constant during the entire procedure. As depicted in Fig. 2, the plots of [Cr(VI)] vs. time were linear (0.997 < R² < 0.998) for more than 4 half-lives. Also, the slope of the fitting equation presented a similar value (0.0153 < k_obs < 0.0165) under various initial Cr(VI) concentrations, indicating a first-order reaction with respect to Cr(VI). Similar reaction orders were found in earlier studies using scrap iron,¹³ H₂S,^9,31 and various organic reductants.³²

Fig. 2 Values of ln[Cr(VI)] versus t plots under various initial Cr(VI) concentrations with EGCG (800 μM) at pH 6.86 (phosphate buffered). The reaction is pseudo-first-order with respect to Cr(VI).

Effect of different initial EGCG concentrations (200–800 μM) on Cr(VI) reduction was tested in two groups buffered by phosphate at pH 6.86 and 4.01 respectively. As shown in Fig. 3a, the rate constant k_obs (the slope of the fitting curve) increased as the initial EGCG concentrations raised at pH 6.86. The inset shows a plot of log k_obs as a function of log[EGCG] and the data can be fitted by a straight line with a slope of 0.886 (R² = 0.999). At pH 4.01, likewise, a slope value of 0.946 (R² = 0.994) was obtained (Fig. 3b). These results suggest that the true value of y should be 1.0 and the reaction also adhere to first-order kinetics with respect to EGCG concentration.

Fig. 3 Values of ln[Cr(VI)_t/Cr(VI)₀] versus t plots, [Cr(VI)] = 20 μM with various initial EGCG concentrations at (a) pH 6.86 and (b) pH 4.01 (phosphate buffer). The results indicate that the reaction is pseudo-first-order with respect to EGCG.

The effect of pH on the k_obs value was examined over the pH range 3.07–7.56 with [Cr(VI)] = 20 μM and [EGCG] = 400 μM (Table 2). As Fig. 4 shows, the plots of log k_obs were not linear and displayed a complex order with respect to [H]⁺. In the range pH 3.07–4.29, the plot had a high correlation of linearity with a slope of 1.039 (R² = 0.996), whilst the value of log k_obs appeared to level off as the pH rose from 4.86 to 7.05, with an approaching value of 0.051 (R² = 0.902). Then, k_obs as a function of pH can be empirically expressed by the linear equation: k_obs = a[H]⁺ + b, namely, the overall rate equation can be described as follows:

−d[Cr(VI)]/dt = {k₁[H⁺] + k₀}[Cr(VI)][EGCG] (4)

and

k_H = k₁[H⁺] + k₀ (5)

where k₁, k₀ represent the first-order and zero-order rate constant.

Table 2 Values of k_obs for the reduction of Cr(VI) (20 μM) with EGCG (400 μM) at various pH

[EGCG] = 400 μM
pH	kobs (s^−1)	pH	kobs (s^−1)
7.56	5.3 × 10^−5	5.32	1.7 × 10^−4
7.31	6.2 × 10^−5	4.86	1.8 × 10^−4
7.05	1.4 × 10^−4	4.59	2.1 × 10^−4
6.86	1.5 × 10^−4	4.25	3.2 × 10^−4
6.29	1.6 × 10^−4	4.01	5.0 × 10^−4
6.05	1.5 × 10^−4	3.49	2.0 × 10^−3
5.80	1.6 × 10^−4	3.31	2.8 × 10^−3
5.53	1.7 × 10^−4	3.07	5.0 × 10^−3

---

Fig. 4 Values of log k_obs as a function of solution pH. The dots are the experimental results, and the solid line is the empirical fit to k_obs = a[H⁺] + b, corresponding to k_H = k₁[H⁺] + k₂.

The value of k₁ and k₀ and the indicated errors (standard deviations) were determined from a least squares nonlinear curve fit with Matlab (Table 3). Also, the nonlinear fitting curve was depicted in Fig. 4 in red solid line, and the statistical goodness of fit (R² = 0.984) verified the suitability of the empirical rate expression. Michael S. Elovitz et al.^33,34 investigated the redox interaction of Cr(VI) with substituted phenol and found a similar result of log k_obs as a function of pH. They pointed out that the presence of parallel reaction pathways involving different stoichiometries of proton probably responsible for the different pH dependence of the reaction order. This theory provided a reasonable explanation of the experimental phenomenon observed in our study.

Table 3 The value of k₁ and k₀ determined by least squares nonlinear curve fit of eqn (5)

k1 (M^−2 s^−1)	k0 (M^−1 s^−1)	R^2
1.2 ± 0.10 × 10^4	0.35 ± 0.02	0.984

---

As the pH value exceeded 7.05, the further decline of log k_obs may attribute to the enhanced autoxidation of EGCG in alkalescent environment^30,35 and the variation of Cr(VI) species (HCrO₄⁻ = H⁺ + CrO₄²⁻, pK_a = 6.52). But it did not necessarily suggests that HCrO₄²⁻ is the sole reactive species since Cr(VI) reduction still occurs at alkaline condition as we mentioned above. In terms of conditions at pH < 3, the reaction kinetics were not tested due to the reaction was extremely fast and accomplished in dozens seconds.

Reaction mechanism

Oxidation products analysis. LC-MS/MS analyses were carried out to obtain more information about the oxidation products of EGCG by Cr(VI) at pH = 3. Although catechin and its related compounds achieve better sensitivity and selectivity in negative ESI mode at alkaline conditions,^36,37 the process of alkalization would induce further oxidation of analytes and is detrimental to the sample preservation. Hence, positive ion detection mode was chosen in MS analysis. The total ion chromatography (TIC) showed four peaks over the mass range from m/z 50 to 500 (Fig. 5). EGCG corresponded to the fourth peak, yielding a protonated ion [M + H]⁺ at m/z 459 (where M represents the EGCG molecule). The first peak consisted of several ion fragments probably came from impurities since the same peak was also found in the standard sample of EGCG. The second peak produced a maximum response at m/z 145 and, regrettably, we failed to propose a plausible source of this fragment. In terms of the third peak, it contains several ions corresponded to [M-2 + H]⁺ at m/z 457, [M-2 + 18 + H]⁺ at m/z 475 and [M-2 + 36 + H]⁺ at m/z 493 respectively. Actually, the molecule ions at m/z 457, 475, and 493 yielded almost the same mass spectra in MS/MS analysis (data not show), it can be inferred that these ions were derived from the same quasi-molecule ion with a successive obtain of H₂O molecule. Namely, the molecular weight of this compound was 456, probably formed by loss of two hydrogen molecules from the EGCG molecule.

Fig. 5 HPLC total ion chromatography of reaction products derived from the oxidation of EGCG by Cr(VI).

Fig. 6 enumerates the MS2 spectra of the precursor ions at m/z 459 and m/z 475 (m/z 475 was selected owing to its highest abundance). The product ions of EGCG were assigned as follows: the fragments m/z 307 and 289 corresponded to [M + H-galloy + H]⁺ and [M + H-galloy + H − H₂O]⁺ respectively,³⁸ the m/z 139 ion can derived from the phenolic ring A or B,³⁹ and m/z 153 ion originated from the cleavage of ester bond between EGC and gallic acid.³⁷ Interestingly, the MS/MS spectrum of m/z 475 ([M-2 + H₂O + H]⁺) presented three product ions at m/z 457, 305 and 287, 2 Da less than the EGCG fragments at m/z 459, 307 and 289 respectively, and the galloy moiety at m/z 153 was also found in the spectrum. Moreover, the dramatic decrease of intensity at m/z 139 directly implied the structure change at the phenolic B-ring. These results suggest that the oxidation of EGCG occurs at the EGC part rather than the galloy moiety. Also, Susanne et al.⁴⁰ reported that the trihydroxyphenyl B ring is the principle site of antioxidation reaction on the EGCG molecule. Midori et al. indicated that catechol moiety show more antioxidative activity than resorcinol.⁴¹ Considering the oxidation features of phenol moiety and 2 Da mass unit lose in the EGCG phenolic B ring, it is inferred that EGCG quinone (Fig. 6b) is the preliminary oxidation product of EGCG. Besides, the peak at m/z 261 (26 Da less than 287) in MS/MS of m/z 457 was probably attribute to the loss of CO fragment at benzoquinone moiety in the collision cell (Fig. 6b inset), which was not found in the spectrum of EGCG and further convinced the production of EGCG quinone.

Fig. 6 MS/MS spectrum of: (a) EGCG and (b) EGCG quinone in positive ion detection mode (collision energy 10).

EGCG quinone is an important intermediate mentioned prevailingly in EGCG autoxidation process and considered of great significance to antioxidant effects of EGCG in biological system.³⁵ Thus, this finding further ensured the safety of using EGCG as reductant for chromium(VI) reduction.

Proposed reaction pathways. UV-VIS spectra of Cr(VI) and EGCG mixture exhibits absorption band in wavelength range from 250 to 400 nm (Fig. 7) where EGCG responsible for the absorbance peak at 272 nm and Cr(VI) corresponded to the absorbance around 350 nm. Repetitive scanning of the reaction shows the peak at 272 nm decreased gradually, indicating the depletion of EGCG. On the contrast, the absorption band from 300 nm to 370 nm increased continuously rather than declined as Cr(VI) reduced, where result from the absorption of adjacent quinones group derived from EGCG oxidation.⁴² In other words, the absorption of oxidation products of EGCG overlapped the spectrum of Cr(VI) coincidently and raised the overall absorbance. The inset of Fig. 7 shows the spectrum of 100 μM partly autoxidized EGCG under acidic condition, the absorption band range was quite in accordance with the spectra of reaction mixture.

Fig. 7 UV-visible spectra changes for the reaction of EGCG with Cr(VI) at [EGCG] = 100 μM, Cr(VI) = 100 μM, pH = 3, temperature = 25 °C.

Numerous studies have been reported on Cr(VI) reduction by organic compounds which extensively elucidated the mechanistic details of Cr(VI) redox chemistry. Formation of chromate-ROH ester has been well-established in the Cr(VI) oxidation of alcohol,⁴³ ascorbic acid⁴⁴ and substituted phenols.³⁴ It is reasonable to assume that an analogous intermediate formation may also apply to polyphenol. Unfortunately, as described above, we did not find any new peak that may come from intermediate formation of chromate-ROH ester in UV spectra. Perhaps the reason is that the trihydroxyphenyl group of EGCG, as electron-donating substituents, greatly increases π-electron density and leads to the electron transfer processes much faster than ester formation. Thus once the ester intermediate formed, it decomposed so quickly that a detectable concentration does not develop. Nevertheless, we still proposed a simple two electrons transfer mechanism of the redox interaction between Cr(VI) and EGCG based on stoichiometry (Fig. 8). In acidic conditions, a one-proton pathway may get involved in the reaction (eqn (6)), corresponding to the first-order reaction with respect to [H⁺] as mentioned in kinetics study section. Meanwhile, a concurrent redox pathway independent of proton dominated at weakly acidic and neutral conditions (eqn (7)), in accordance with the [H⁺] independent reaction rate. Afterwards, Cr(IV) disproportionated and/or reacted with Cr(VI) to produce Cr(V) (eqn (8) and (9)), then Cr(V) through a similar two electrons transfer mechanism to finished the chromium reduction to Cr(III) (eqn (10)).

Fig. 8 Proposed reaction pathways for Cr(VI) reduction by EGCG. Note that the formation of the quinonic structure has been indicated at the 3′,4′ phenol position in the gallate, it may also occur in the 4′,5′ phenol position.

Conclusions

This work presents the first effort to use epigallocatechin gallate as reductant for hexavalent chromium reduction in aqueous. The reduction of Cr(VI) by EGCG can occur not only in acidic conditions, but also in weakly alkaline conditions (pH 7–9), and more rapid reduction rate was obtained as the solution acidity increased. Meanwhile, the variation of pH dependence probably ascribed to the existence of parallel reactions with different reaction orders with respect to [H⁺]. High temperature enhanced the Cr(VI) reduction efficiency. Kinetics studies indicated that the reaction kinetics are first-order in both Cr(VI) and EGCG concentrations. The trihydroxyphenyl B ring of EGCG may be the most reactive moiety to Cr(VI) and EGCG quinone was identified as a preliminary oxidation product of EGCG. The identification of EGCG quinone ensured the safety of Cr(VI) reduction using EGCG initially. However, the redox interaction of Cr(VI) with EGCG is complicated and the successive oxidation details of EGCG by Cr(VI) are still ambiguous, further works are needed to establish a comprehensive mechanism and explore the potentials of practical application.

Acknowledgements

This work was financially supported by the National “Twelfth Five-Year” plan for Science & Technology support (Grand No: 2012BAJ24B03). The authors are thankful to the fellows of College of Chemistry and Chemical Engineering, Hunan University for supporting the HPLC-MS/MS analysis.

Notes and references

1. L. Fishbein, Environ. Health Perspect., 1981, 40, 43–64.
2. Y. Gao and J. Xia, Environ. Sci. Technol., 2011, 45, 8605–8606.
3. H. Cheng, T. Zhou, Q. Li, L. Lu and C. Lin, PLoS One, 2014, 9, e87753.
4. C. E. Barrera-Diaz, V. Lugo-Lugo and B. Bilyeu, J. Hazard. Mater., 2012, 223–224, 1–12.
5. B. Dhal, H. N. Thatoi, N. N. Das and B. D. Pandey, J. Hazard. Mater., 2013, 250–251, 272–291.
6. J. G. J. A. J. C. P. Avakian, Chromium(VI) Handbook, CRC press, 2004.
7. M. Pettine, I. Barra, L. Campanella and F. J. Millero, Water Res., 1998, 32, 2807–2813.
8. B. Hua and B. Deng, Environ. Sci. Technol., 2003, 37, 4771–4777.
9. C. Kim, Q. Zhou, B. Deng, E. C. Thornton and H. Xu, Environ. Sci. Technol., 2001, 35, 2219–2225.
10. R. R. Patterson, S. Fendorf and M. Fendorf, Environ. Sci. Technol., 1997, 31, 2039–2044.
11. I. J. Buerge and S. J. Hug, Environ. Sci. Technol., 1997, 31, 1426–1432.
12. J. P. Gould, Water Res., 1982, 16, 871–877.
13. M. Gheju and A. Iovi, J. Hazard. Mater., 2006, 135, 66–73.
14. L.-n. Shi, X. Zhang and Z.-l. Chen, Water Res., 2011, 45, 886–892.
15. V. Madhavi, T. N. V. K. V. Prasad, A. V. B. Reddy, B. R. Reddy and G. Madhavi, Spectrochim. Acta, Part A, 2013, 116, 17–25.
16. G. Lee, J. Park and O. R. Harvey, Water Res., 2013, 47, 1136–1146.
17. X. R. Xu, H. B. Li, X. Y. Li and J. D. Gu, Chemosphere, 2004, 57, 609–613.
18. V. A. Okello, S. Mwilu, N. Noah, A. Zhou, J. Chong, M. T. Knipfing, D. Doetschman and O. A. Sadik, Environ. Sci. Technol., 2012, 46, 10743–10751.
19. K. Poonkuzhali, V. Rajeswari, T. Saravanakumar, P. Viswanathamurthi, S.-M. Park, M. Govarthanan, P. Sathishkumar and T. Palvannan, J. Hazard. Mater., 2014, 272, 89–95.
20. D. G. Nagle, D. Ferreira and Y.-D. Zhou, Phytochemistry, 2006, 67, 1849–1855.
21. Q. Meng, C. N. Velalar and R. Ruan, Free Radicals Biol. Med., 2008, 45, 1738.
22. A. B. Albadarin, C. Mangwandi, G. M. Walker, S. J. Allen, M. N. M. Ahmad and M. Khraisheh, J. Environ. Manage., 2013, 114, 190–201.
23. E. Malkoc and Y. Nuhoglu, Sep. Purif. Technol., 2007, 54, 291–298.
24. E. Malkoc and Y. Nuhoglu, Chem. Eng. Sci., 2006, 61, 4363–4372.
25. D. Park, S.-R. Lim, Y.-S. Yun and J. M. Park, Chemosphere, 2007, 70, 298–305.
26. S. K. Prabhakaran, K. Vijayaraghavan and R. Balasubramanian, Ind. Eng. Chem. Res., 2009, 48, 2113–2117.
27. R. Elangovan, L. Philip and K. Chandraraj, Chem. Eng. J., 2008, 141, 99–111.
28. A. W. W. A. American Public Health Association, Journal, 1989.
29. M. Mochizuki, S. Yamazaki, K. Kano and T. Ikeda, Biochim. Biophys. Acta, Gen. Subj., 2002, 1569, 35–44.
30. V. K. Ananingsih, A. Sharma and W. Zhou, Food Res. Int., 2013, 50, 469–479.
31. M. Pettine, F. J. Millero and R. Passino, Mar. Chem., 1994, 46, 335–344.
32. B. Deng and A. T. Stone, Environ. Sci. Technol., 1996, 30, 2484–2494.
33. M. S. Elovitz and W. Fish, Environ. Sci. Technol., 1994, 28, 2161–2169.
34. M. S. Elovitz and W. Fish, Environ. Sci. Technol., 1995, 29, 1933–1943.
35. S. Sang, I. Yang, B. Buckley, C.-T. Ho and C. S. Yang, Free Radicals Biol. Med., 2007, 43, 362–371.
36. G. K. Poon, J. Chromatogr. A, 1998, 794, 63–74.
37. Y. Mizooku, M. Yoshikawa, T. Tsuneyoshi and R. Arakawa, Rapid Commun. Mass Spectrom., 2003, 17, 1915–1918.
38. A. Kiehne and U. H. Engelhardt, Zeitschrift fur Lebensmittel-Untersuchung und -Forschung, 1996, 202, 48–54.
39. Z. Spacil, L. Novakova and P. Solich, Food Chem., 2010, 123, 535–541.
40. S. Valcic, A. Muders, N. E. Jacobsen, D. C. Liebler and B. N. Timmermann, Chem. Res. Toxicol., 1999, 12, 382–386.
41. M. Kumamoto, T. Sonda, K. Nagayama and M. Tabata, Biosci., Biotechnol., Biochem., 2001, 65, 126–132.
42. S. Baugh and T. Hnat, US Pat., Application Pat., 12/181, 285, 2008.
43. K. B. Wiberg and H. Schaefer, J. Am. Chem. Soc., 1969, 91, 927–932.
44. D. A. Dixon, N. P. Sadler and T. P. Dasgupta, J. Chem. Soc., Dalton Trans., 1993, 3489–3495,  10.1039/dt9930003489.

---