2.1 Reagents

NaBrO₃ and NaBr from Acros Organics were of extra pure grade. Anion standards of Br^- and ${\text{BrO}}_3^{-}$, and other anions, for uses in ion chromatography, were purchased from AccuStandard Inc. All other chemicals were of analytical grade and used without further purification. N₂O, O₂ and N₂ gases were of high purity (99.99%). De-ionized water produced by Millipore Q system was used throughout the experiments.

2.2 Irradiation

Bromate solutions in 80-mL sealed Pyrex glass tubes were irradiated at ambient temperatures in a ⁶⁰Co γ-ray source at Shanghai Institute of Applied Physics, Chinese Academy of Sciences. Adsorbed doses were measured by the ceric sulfate dosimetry system. A series of NaBrO₃ aqueous solutions were prepared by dissolving certain amount of NaBrO₃ into de-ionized water to predetermined concentrations. Inorganic anions of nitrate (${\text{NO}}_3^{-}$), sulfate (${\text{SO}}_4^{2-}$), chloride (Cl^-) and carbonate (${\text{HCO}}_3^{-}/{\text{CO}}_3^{2-}$), were added to the solution if necessary in the form of their stock solutions of sodium salts. The pH value was adjusted by adding perchloric acid or sodium hydroxide, and the initial pH was 7.0 unless otherwise stated. The solutions were air-equilibrated or bubbled with high-purity N₂, O₂ and N₂O for 20 min, respectively, and were irradiated to 1 kGy–10 kGy.

2.3 Analysis methods

Concentrations of bromate and bromide were determined by a Dionex ICS-2000 reagent-free ion chromatograph with an IonPac AS19 analytical column (250×4 mm ID), using 20 mmol/L potassium hydroxide eluent at a flow rate of 1 mL/min. The detection limits of ${\text{BrO}}_3^{-}$ and Br^- were 1 and 10 μg/L, respectively. All the experiments were duplicated and the results were averaged.

3.1 Effect of dose and initial concentration

Air-equilibrated solutions with initial ${\text{BrO}}_3^{-}$ concentrations of 30 μg/L–210 μg/L and with dissolved oxygen of about 8 mg/L were irradiated to 1 kGy–10 kGy. The removal rates of ${\text{BrO}}_3^{-}$ are illustrated in Fig. 1(a) and percentage of bromine recovery (considering only ${\text{BrO}}_3^{-}$ and Br^- in the reaction systems) is shown in Fig. 1(b).

Fig. 1.Full-screen View

(Color online) Effect of dose and initial content on bromate removal (a) and bromine recovery (b), irradiated under air-saturated condition.

From Fig. 1, ${\text{BrO}}_3^{-}$ was degraded effectively by gamma-rays. The removal rate of ${\text{BrO}}_3^{-}$ increased with the dose and decreased with increasing initial ${\text{BrO}}_3^{-}$ concentrations. At 30.7 μg/L and 4.0 kGy, 73% ${\text{BrO}}_3^{-}$ was decomposed and the residual was 8.3 μg/L, which is below the maximum contaminant level of 10 μg/L. Figure 1 shows that at a given dose, the ${\text{BrO}}_3^{-}$ recovery increased with the initial concentration, but the removal rates did not change significantly. At higher ${\text{BrO}}_3^{-}$ levels, there were less recombination of reactive radicals (such as $\cdot {\text{e}}_{\text{aq}}^{-}+\cdot \text{OH}\to \cdot {\text{OH}}^{\text{-}}$, ·OH+·OH → H₂O₂, etc.) with more radicals reacting with ${\text{BrO}}_3^{-}$. However, the formation of intermediate compounds, and the oxidation of Br^- and the intermediates by ⋅OH radicals, made the reaction system complicated [17-19]. It was reported that bromide could be oxidized to ${\text{BrO}}_3^{-}$ when ⋅OH radicals acted as the only oxidant and that HOBr/OBr^- are the requisite intermediates [18]. ${\text{HBrO}}_{\text{2}}/{\text{BrO}}_2^{-}$ may exist in our reaction systems because of the oxidation of HOBr/OBr^- by ⋅OH radicals [3]. In Fig. 1(b), at 30.7 and 86.2 μg/L of ${\text{BrO}}_3^{-}$, the Br recovery decreased at first and then increased with the dose, indicating that the amount of bromine intermediates would agree with the opposite tendency. On the other hand, Zhou et al. [17] and LaVerne et al. [19] reported that Br^- could not be oxidized to ${\text{BrO}}_3^{-}$ by irradiating air-equilibrated NaBr solutions due to recycling of the oxidized species of bromine element by $\cdot {\text{e}}_{\text{aq}}^{-}$, ⋅H and/or ${\text{HO}}_{\text{2}}\cdot /{\text{O}}_2^{-}\cdot$, which implies that the reduction of ${\text{BrO}}_3^{-}$ and bromine intermediates is inevitable under similar operation condition. Therefore, we can infer that the amount of intermediates at 165 and 210 μg/L ${\text{BrO}}_3^{-}$ follows a similar tendency to that at 30.7 and 86.2 μg/L ${\text{BrO}}_3^{-}$, and that all the ${\text{BrO}}_3^{-}$ and the intermediates are reduced to Br^- at sufficiently higher absorbed doses.

3.2 Effect of saturated atmosphere

In order to explore the atmosphere effect, aqueous solutions with high level ${\text{BrO}}_3^{-}$ and saturated with air, N₂, O₂ or N₂O were irradiated to 1 kGy–10 kGy. As shown in Fig. 2, for the solution degassed by N₂, the removal rate of ${\text{BrO}}_3^{-}$ was near 98% and its concentration reduced remarkably to 7.6 μg/L at 4.0 kGy. At a given irradiation dose, the removal efficiency of ${\text{BrO}}_3^{-}$ decreased in the order of N₂, air, O₂, and N₂O.

Fig. 2.Full-screen View

(Color online) Effect of atmosphere on ${\text{BrO}}_3^{-}$ removal by γ-rays.

The general reaction of radiolysis of N₂-saturated water can be written as Eq. (1) [20, 21], where the numbers are G-values, which are defined as the number of formed or decomposed molecules per 100 eV absorbed energy.

As a powerful reducing agent with a standard reduction potential of -2.9 V, hydrated electron ($\cdot {\text{e}}_{\text{aq}}^{-}$) react quickly with ${\text{BrO}}_3^{-}$, as shown in Eq. (2), while ⋅H reacts in a much slow rate with ${\text{BrO}}_3^{-}$, as shown in Eq. (3) [21].

The reactions of ⋅H and $\cdot {\text{e}}_{\text{aq}}^{-}$ with O₂ produce HO₂. and ${\text{O}}_2^{-}$. as shown in Eqs. (4) and (5), respectively [20], which also reduce ${\text{BrO}}_3^{-}$ possibly via Eqs. (6) and (7). However, according to Fig. 2, it is reasonable to infer that the reduction of ${\text{BrO}}_3^{-}$ by ${\text{HO}}_{\text{2}}\cdot /{\text{O}}_2^{-}\cdot$ is slower than that by $\cdot {\text{e}}_{\text{aq}}^{-}$.

On the other hand, the ${\text{BrO}}_3^{-}$ concentration decreased just slightly and no Br^- ion was detected in the N₂O-saturated solutions, when $\cdot {\text{e}}_{\text{aq}}^{-}$ and ⋅H generated through Eq. (1) were effectively converted into ⋅OH radicals by N₂O, as shown in Eqs. (8) and (9) [21]. It was reported that ${\text{BrO}}_3^{-}$ could react with ⋅OH radical and form ${\text{BrO}}_3^{-}$ radical, which was further dissociated to BrO. radical, as shown in Eqs. (10) and (11) [22]. However, BrO^- radicals could be disproportionate to ${\text{HBrO}}_{\text{2}}/{\text{BrO}}_2^{-}$ and HOBr/OBr^- and then reform ${\text{BrO}}_3^{-}$ through oxidation by ⋅OH radicals. As a result, the variation of the ${\text{BrO}}_3^{-}$ concentration was small in the N₂O-saturated solutions (Fig. 2).

3.3 Effect of pH

To further elucidate efficiency of different reactive radicals in reducing ${\text{BrO}}_3^{-}$, air- and N₂-saturated solutions of different pH values were irradiated. Figure 3 shows the results. The pH increase had a positive impact on ${\text{BrO}}_3^{-}$ removal in the N₂-saturated conditions, with the ${\text{BrO}}_3^{-}$ removal rate being 88% at pH 11 and only 28% at pH 3.3. By radiolysis of N₂-saturated water, the concentrations of ⋅H and $\cdot {\text{e}}_{\text{aq}}^{-}$ radicals generated varied with the pH value, with more $\cdot {\text{e}}_{\text{aq}}^{-}$ but less ⋅H formed under high pH conditions [20]. The amount of $\cdot {\text{e}}_{\text{aq}}^{-}$ was negligible at pH 3.3 and ⋅H almost disappeared at pH 11, while the generation of ⋅OH radicals remains nearly constant in the pH range studied [20]. Therefore, from Fig. 3, $\cdot {\text{e}}_{\text{aq}}^{-}$ is much more effective than ⋅H in reducing ${\text{BrO}}_3^{-}$, and $\cdot {\text{e}}_{\text{aq}}^{-}$ is the main reactive radical account for the removal of ${\text{BrO}}_3^{-}$.

Fig. 3.Full-screen View

(Color online) Removal rate of ${\text{BrO}}_3^{-}$ at various pH values.

On the other hand, the removal rate of ${\text{BrO}}_3^{-}$ in air-saturated solutions was lower and it increased a bit with the pH value. According to Eqs. (1), (4) and (5), the main reactive species formed under these conditions were ${\text{O}}_2^{-}\cdot /{\text{HO}}_{\text{2}}\cdot$ and ⋅OH, with more ${\text{O}}_2^{-}\cdot$ radicals, and less ${\text{HO}}_{\text{2}}\cdot$ radicals formed at high pH values, but amount of ⋅OH radicals was nearly constant. Therefore, it can be sure that ${\text{O}}_2^{-}\cdot$ radicals are less powerful than HO₂. radicals in reducing ${\text{BrO}}_3^{-}$. Based on the results of Figs. 2 and 3, the efficiency of decomposing ${\text{BrO}}_3^{-}$ by reactive species was in the order of $\cdot {\text{e}}_{\text{aq}}^{-}>\cdot {\text{H>HO}}_{\text{2}}\cdot >{\text{O}}_2^{-}\cdot$.

3.4 Effect of coexisting anions

The presence of coexisting anions in water is known to affect the radiolytic degradation of pollutants [21]. The effects of 1 mmol/L Cl^-, ${\text{NO}}_3^{-}$, ${\text{SO}}_4^{2-}$ or ${\text{HCO}}_3^{-}/{\text{CO}}_3^{2-}$ anions on the radiolytic degradation of ${\text{BrO}}_3^{-}$ solutions are shown in Fig. 4(a). For N₂O-saturated solutions, they had little effect on degradation of ${\text{BrO}}_3^{-}$, as they could neither compete with N₂O for reacting with $\cdot {\text{e}}_{\text{aq}}^{-}$ or ⋅H, nor act as effective scavengers of the main existing radical (namely, ⋅OH). For similar reason, the addition of Cl^- or ${\text{SO}}_4^{2-}$ anions showed little influence, too, on ${\text{BrO}}_3^{-}$ removal under ${\text{N}}_2^{-}$ or air-saturated conditions.

Fig. 4.Full-screen View

(Color online) Effect of anions (a) and ${\text{NO}}_3^{-}$ concentration (b) on ${\text{BrO}}_3^{-}$ removal.

The effects of ${\text{CO}}_3^{2-}/{\text{HCO}}_3^{-}$ anions on ${\text{BrO}}_3^{-}$ removal are more complex. On one hand, ${\text{CO}}_3^{2-}/{\text{HCO}}_3^{-}$ anions react with ⋅OH and play a role as ⋅OH scavenger, as shown in Eqs. (12) and (13) [21], hence the indirect promotion of ${\text{BrO}}_3^{-}$ removal [17, 19] . On the other hand, the formed active carbonate radicals (${\text{CO}}_3^{-}\cdot$) through Eqs. (12) and (13) can oxidize the intermediates of bromine and form ${\text{BrO}}_3^{-}$ precursors again, as shown in Eqs. (14) and (15) [23] . In Fig. 4(a), adding ${\text{CO}}_3^{2-}/{\text{HCO}}_3^{-}$ slightly increased the removal of ${\text{BrO}}_3^{-}$.

However, it was observed that the presence of ${\text{NO}}_3^{-}$ anion suppressed significantly the ${\text{BrO}}_3^{-}$ reduction in both N₂^-- and air-saturated solutions. According to Fig. 4(a), the removal rate of ${\text{BrO}}_3^{-}$ decreased from 74.7% to 10.9% by adding 1 mmol/L ${\text{NO}}_3^{-}$ in the N₂-saturated atmosphere. By adding 1 to 10 mg/L nitrate (measured on basis of nitrogen content), the results of ${\text{BrO}}_3^{-}$ removal under N₂-, O₂- or air-saturated conditions, are shown in Fig. 4(b). In all the three atmospheres, there existed an inhibition effect, which was enhanced with increasing concentrations of ${\text{NO}}_3^{-}$. For example, in air-saturated solutions, the removal rate decreased more than a half when 10 mg/L ${\text{NO}}_3^{-}$ was added (the drinking water standard of nitrate in China is 10 mg/L). The reason of inhibition can be attributed to the scavenging of $\cdot {\text{e}}_{\text{aq}}^{-}$ by ${\text{NO}}_3^{-}$ ions, as shown in Eq. (16),

and the reaction rate is very fast according to Ref. [21]. From Fig. 4, we know that the product of $\cdot {\text{e}}_{\text{aq}}^{-}$ and ${\text{NO}}_3^{-}$, namely $\cdot {\text{NO}}_3^{2-}$, and its further decay products, are incapable or ineffective in decomposing ${\text{BrO}}_3^{-}$. As a result, the efficiency of ${\text{BrO}}_3^{-}$ removal is reduced, so the operation cost shall be carefully considered in irradiation removal of ${\text{BrO}}_3^{-}$ from water of high nitrate levels.