Article pubs.acs.org/est
Rapid Photooxidation of As(III) through Surface Complexation with Nascent Colloidal Ferric Hydroxide Jing Xu, Jinjun Li,* Feng Wu,* and You Zhang Department of Environmental Science, Hubei Key Lab of Biomass Resource Chemistry and Environmental Biotechnology, School of Resources and Environmental Science, Wuhan University, Wuhan, 430079, P. R. China S Supporting Information *
ABSTRACT: Contamination of water and soils with arsenic, especially inorganic arsenic, has been one of the most important topics in the fields of environmental science and technology. The interactions between iron and arsenic play a very significant role in the environmental behavior and effect of arsenic species. However, the mechanism of As(III) oxidation in the presence of iron has remained unclear because of the complicated speciation of iron and arsenic. Photooxidation of As(III) on nascent colloidal ferric hydroxide (CFH) in aqueous solutions at pH 6 was studied to reveal the transformation mechanism of arsenic species. Experiments were done by irradiation using light-emitting diodes with a central wavelength of 394 nm. Results show that photooxidation of As(III) and photoreduction of Fe(III) occurred simultaneously under oxic or anoxic conditions. Photooxidation of As(III) in the presence of nascent CFH occurred through electron transfer from As(III) to Fe(III) induced by absorption of radiation into a ligand-to-metal charge-transfer (LMCT) band. The estimated quantum yield of photooxidation of As(III) at 394 nm was (1.023 ± 0.065) × 10−2. Sunlight-induced photooxidation of As(III) also occurred, implying that photolysis of the CFH−AsIII surface complex could be an important process in environments wherein nascent CFH exists.
■
for this process could be fitted with a pseudo-first-order rate equation. Bhandari et al.9 reported that during the oxidation of As(III) in the ferrihydrite system, generated As(V) strongly binds to the ferrihydrite surface and hinders As(III) oxidation. In their study on As(III) photooxidation with goethite,10 As(III) oxidation has been found to occur even under anoxic conditions; a further study also suggests that Fe(II) could be released to the solution during As(III) oxidation. However, ferric oxides obtained by various methods under various conditions show different features (size, shape, etc.),11 the ferric oxides used in their work were obtained from synthesis rather than as colloid generated by hydrolysis of dissolved iron ion. This difference might affect the crystallinity of ferric oxides and thus the efficiency or mechanism of As(III) oxidation or both. Even though complexation between arsenic and iron oxides has been widely investigated, the mechanism of redox reactions in the complexation has not been elucidated. Complexes between Fe(III) ions and ligands such as citrate,12 oxalate,13 and hydroxyl ions14 have been intensively studied during the past decades. Ultraviolet−visible (UV−vis) spectroscopy has been used to observe the formation of such complexes. Reports15−17 indicate that ligand-to-metal charge
INTRODUCTION The acute and chronic toxicity of arsenic is well-known. Therefore, widespread arsenic contamination calls for an effective method for controlling levels of arsenic in the environment. In aqueous environments, arsenic mainly exists in inorganic forms, namely, in the valence states As(III) and As(V).1 As(III) is much more toxic than As(V),2 and the latter is more easily removed through adsorption and precipitation methods.3 Thus, oxidation of As(III) to As(V) is one feasible measure in arsenic detoxification. In acidic and circumneutral aqueous media, iron occurs as dissolved ion and as colloid, respectively. Previous studies4−6 investigated the role of Fe(II) and Fe(III) ions in catalyzing As(III) oxidation induced by light. The mechanisms of As(III) photooxidation have been widely studied. Reactive oxygen species (ROS), mainly HO•, are considered crucial in the oxidation reaction. However, the mechanism of As(III) oxidation through interaction between As(III) and Fe(III) ions at circumneutral condition has not been clearly elucidated. In heterogeneous reactions, As(III) binds to ferric oxides by outer-sphere and inner-sphere complexation. Manning et al.7 found that the As(III) complexation on goethite (α-FeOOH) occurs through formation of a bidentate−binuclear bridge, resulting in an AsIII−Fe interatomic distance of 3.378 ± 0.014 Å. Liu et al.8 reported that, in the presence of iron and dissolved organic matter (DOM), As(III) could complex with DOMstabilized Fe colloids (As−Fe−DOM complexes). Kinetic data © 2013 American Chemical Society
Received: Revised: Accepted: Published: 272
August 17, 2013 November 26, 2013 December 6, 2013 December 6, 2013 dx.doi.org/10.1021/es403667b | Environ. Sci. Technol. 2014, 48, 272−278
Environmental Science & Technology
Article
total concentration of arsenic was determined by using 5% HCl−2% KBH4 with thiourea and ascorbic acid. During the oxic photoinduced reaction, 10 mL of solution was taken and added to colorimetric tubes for Fe(II) analysis. The detection method was based on procedures according to Zuo et al.20 and previously used by our group.19 HCl [2 mL, 1:1 (v/v)] was used to adjust the acidity and to dissolve the iron colloid, and NH4F was used to complex Fe(III). Subsequently, 2 mL of 2 g L −1 phenanthroline and 5 mL of NH4(CH3CO2)−H(CH3CO2) buffer solution were successively added to the colorimetric tubes. After dilution to 25 mL and 15 min of color development, the absorbance of the samples at 510 nm was analyzed on a spectrophotometer (Shimadzu UV-1601) at room temperature using a 5 cm quartz cuvette. Experiments were conducted to determine various periods for the anoxic reaction. After each period, the lamps were switched off and a specific amount of concentrated sulfuric acid was immediately added to acidify each solution. Afterward, 10 mL of solution was taken for Fe(II) determination. The procedures were the same with those in the oxic experiments. Spectra of the Fe(III), As(III), and Fe(III)−As(III) solutions were recorded on a Shimadzu UV-1601 spectrophotometer. Because of the low concentration and weak absorption of Fe(III), the determination was carried out in a 10 cm quartz cuvette at room temperature to amplify the absorption. A specific amount of H2SO4 was added to the ferric stock solution to stabilize the ferric species; the reaction solution was therefore very acidic before pH adjustment and iron was present as dissolved ions. After adjusting the pH (within 2 min) to 6.0 and storing in the dark for 1 h, 25 mL of the solution was taken for UV−vis spectrum analysis.
transition (LMCT), the electron transfer (ET) from the ligand to the central metal ion, could occur in such complexes and result in the photoreduction of Fe(III) ions to Fe(II) ions. The goal of this study was to investigate the mechanism of As(III) photooxidation in the colloidal ferric hydroxide (CFH) system. Complexation and photochemical reaction between Fe(III) and As(III) ions through LMCT were carried out to reveal their interactions. The complexation between As(III) and CFH was also investigated to determine their ability to form a complex. The present study on the mechanism of As(III) photooxidation could help in understanding the geochemical cycle and fate of arsenic in the environment.
■
MATERIALS AND METHODS Chemicals. NaAsO2 (99.5%) was obtained from Gracia Chemical Technology Co. Ltd. (Chengdu, China) and then used after 24 h of drying in desiccators. Fe2(SO4)3 (analytical reagent (AR) grade) was purchased from Guangdong Taishan Chemical Co. Ltd. and used as the iron source. Nascent CFH was produced by adjusting the pH of Fe2(SO4)3 solutions to a circumneutral value. NaOH (AR), H2SO4 (AR), HCl (AR), KBH4 (95%), KOH (AR), KH2PO4 (AR), FeSO4·7H2O (AR), NH4F (AR), CH3COOH (AR), and phenanthroline (AR) were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). tert-Butyl alcohol (TBA, AR) and ethanol (EtOH, AR) were purchased from Shanghai Shiyi Chemicals Reagent Co. Ltd. and Tianjin Kermel Chemical Reagent Co. Ltd., respectively. Ultrapure water obtained through a water purification system (Liyuan Electric Instrument Co., Beijing, China) was used in all experiments. All prepared solutions were stored at 4 °C and protected from light. Photooxidation Reaction System. A photograph of the reactor is shown in Figure S1 in the Supporting Information (SI). Experiments were conducted in a 250 mL beaker. During each experiment, 200 mL of solution was placed in the beaker. Four 15 cm ×15 cm light-emitting diode (LED) lamps used to irradiate in the near-UV (NUV) region were placed around the beaker. The emission spectrum for the LED lamp (λmax = 394 nm) is shown in Figure S2 (SI). Since the lamps were not calorigenic, the reactions, which were conducted at room temperature, did not require the use of cooling water. The average intensity of emitted light was about 0.13 mW cm−2. The reaction solutions were transferred into the reactor and then the lamps were switched on. Samples were taken for analysis at selected time intervals. Experiments were conducted on solutions that were exposed to the atmosphere. Anoxic experiments were conducted in a 1000 mL hermetic cylindrical reactor. Solutions were purged with N2 (99.9%) for 45 min before and during photoreaction to ensure that oxygen was excluded. A dissolved oxygen (DO) meter (8403, AZ Instrument Co. Ltd.) was used to determine the DO of each solution. During anoxic reaction, the DO was 0.01 mg L −1. In sunlight-induced photooxidation experiments, a solar power meter (SM206, Shenzhen Sanpometer Ltd., Shenzhen, China) was used to determine the irradiation flux of solar power. Analysis. The determination of As(III) concentration was based on our previous work.18,19 Analysis was done by hydride generation−atomic fluorescence spectrometry (Titan Instrument Co. Ltd., Beijing, China) using 1.5% HCl−1.5% KBH4 solution. Argon (99.99%) was used as carrier gas and shielding gas during the determination. In the mass balance experiment,
■
RESULTS AND DISCUSSION Effect of pH on As(III) Oxidation. Iron exists as different species at different pH values. Thus, it is expected that the efficiency and mechanism of As(III) photooxidation could be significantly affected by the pH of the solution. Experiments were therefore carried out to study the effect of pH on As(III) photooxidation at pH 1.0, 3.0, 4.0, 6.0, and 8.0. Figure 1 shows the oxidation results at different pH values. No obvious As(III) oxidation was observed at pH 6.0 when the reaction was conducted in a dark environment within 30 min, whereas oxidation within the same period was significant under NUV
Figure 1. Change of As(III) concentrations during photooxidation at various initial pH values and in the dark at pH 6. [Fe(III)] = 0.1 mM, [As(III)] = 6.67 μM. 273
dx.doi.org/10.1021/es403667b | Environ. Sci. Technol. 2014, 48, 272−278
Environmental Science & Technology
Article
irradiation. The mass balance results at pH 6.0 are shown in Figure S3 (SI), which confirms that the change of As(III) concentration was not due to precipitation. The oxidation efficiency was enhanced from 54.2% to 63.6% when the pH value was increased from 6.0 to 8.0. Under acidic conditions, Fe(III) also rapidly oxidized As(III); 57.7% As(III) was oxidized to As(V) after 30 min of irradiation at pH 3.0, whereas only 34.7% As(III) was oxidized at pH 4.0 within the same time interval. The distribution of ferric species is shown in Figure S4 (SI). As the solubility product constant (pKsp) of Fe(OH)3 is 38.55,21 ferric ion was mainly in the form of FeOH2+ at pH 3.0 when the total iron concentration was 0.1 mM. FeOH2+ has the highest photoactivity among FeIII− hydroxyl complexes (Fe(OH)2+, Fe2(OH)24+, etc.)22 and generates HO• according to eq 1.23 Upon an increase in the pH to 4.0, the dissolved iron ion started to convert to CFH, and at pH 6.0, ferric ion was mainly in the form of CFH. Results of determination of dissolved ferric iron, as well as images obtained by transmission electron microscopy (TEM) and light scattering, are shown in Table S1 and Figures S5 and S6 (SI), respectively. These results prove the formation of CFH particles, which is in accordance with the species distribution in Figure S4 (SI). As(III) oxidation occurred in the presence of both Fe(OH)2+ (at pH 3 or 4) and CFH (at pH 6 or 8) under irradiation; therefore, the mechanism of As(III) oxidation might be pH-dependent. The lower efficiency of oxidation at pH 4.0 than at pH 3.0 and 6.0 could be explained by the lower concentration of both Fe(OH)2+ and CFH in solution. Fe(OH)2 + + hν → Fe2 + + HO•
Figure 2. Quenching effect of TBA or EtOH on the photooxidation of As(III) at pH 6 and 3 and the competing reaction between As(III) and PO43‑. [Fe(III)] = 0.1 mM, [As(III)] = 6.67 μM, [TBA] = [EtOH] = 10 mM, [PO43−] = 66.7 μM.
reaction with HO• has a rate constant of (3.8−7.6) × 108 M−1 s−1.28 The reported rate constants of the reactions between As(III) and HO• are presented in eqs 2−4.29−31 HO• + AsO2− + 2H 2O → As IV (OH)4 + OH− k 2 = 9.0 × 109 M−1 s−1
(2)
HO• + HAsO2 + H 2O → As IV (OH)4 k 3 = (1−1.8) × 109 M−1 s−1
(1)
Nevertheless, no As(III) oxidation was observed within 30 min at pH 1.0. Under such highly acidic conditions, the amount of Fe(OH)2+ is very low and Fe(III) mainly exists as Fe3+; thus, little HO• is generated for oxidation. Therefore, the reaction was terminated by adding samples to acid liquor; 1.5% HCl (pH
