A. Materials

NaBr of extra pure grade was purchased from Acros Organics. Anion Standards of Br^- and BrO₃^- and other anions were purchased from AccuStandard Inc. N₂O, O₂ and N₂ gases were of high purity (99.99%). All other chemicals were of analytical grade and obtained through J&K Chemical Ltd and used as received. De-ionized water by Millipore Q system was used throughout the experiments.

B. Sample irradiation

The irradiation experiments were performed in a ⁶⁰Co gamma source at Shanghai Institute of Applied Physics, Chinese Academy of Sciences. Adsorbed doses were measured by a ceric sulfate dosimetry system. Prepared NaBr aqueous solutions under air-equilibration or saturated with N₂, O₂ or N₂O by bubbling for 20 min of high-purity N₂, O₂ or N₂O gases in 80 mL Pyrex glass tubes, were irradiated to 0.5∼10 kGy. Inorganic anions of nitrate, sulfate, chloride, and (bi)carbonate, were added in the form of their stock solutions of sodium salts. The solution pH was adjusted by adding perchloric acid or sodium hydroxide, and was adjusted to 7.0 unless otherwise stated. All experiments were carried out at ambient temperature.

C. Analysis methods

The concentrations of bromate and bromide and other anions were determined by a Dionex ICS-2000 reagent-free ion chromatograph, with an IonPac AS19 analytical colum (250 mm×4 mm ID), using 20 mmol l^-1 potassium hydroxide eluent at a flow rate of 1 ml min^-1. Sample volume loaded for all analysis was 200 μl each. The detection limits of bromate and bromide were below 1 μgl^-1 and 10 μgl^-1, respectively. The pH values were measured by a PHSJ-4A pH meter.

A. Effect of atmosphere and initial concentration

The results from samples saturated with air, N₂, or O₂ and without other additives, are illustrated in Fig. 1. For both initial Br^- concentration of approximately 80 μgl^-1 and 1650 μgl^-1, the concentration of Br^- almost remained unchanged for all samples, and the BrO₃^- concentrations were all below the detection limit. It could be conclude from Fig. 1 that Br^- could not be oxidized into BrO₃^- under such conditions. In Ar saturated solutions with initial Br^- concentrations from 800 μgl^-1 to 80 μgl^-1. LaVerne et al. [7] found that the Br^- concentrations kept constant up to 100 kGy.

Fig. 1.Full-screen View

(Color online) Br^- concentration as a function of absorbed dose and atmosphere. The datum points in square (□) and circle (∘) are of initial Br^- concentration of approximately 80 μgl^-1 and 1650 μgl^-1, respectively.

It is well known that ·OH radicals could oxidize Br^- to BrOH^-., which then transforms to Br., Br₂^-. and other bromine species [8, 9, 10]. From Fig. 1, the oxidized bromine species could be reduced effectively by $e_{aq}^{-}$, H· and/or O2^-·/HO₂· andBr^- reformed. Possible reactions are listed below [7, 9]:

The numbers in Eq. (1) are called G-values, defined as the number of formed or decomposed molecules per 100 eV absorbed energy. The reactions in Eqs. (8) – (10) take place in aqueous solutions saturated with air or O₂.

On the other hand, as shown in Fig. 2, the majority of bromide was oxidized to bromate in N₂O saturated solutions even at 0.5 kGy. The formation of bromate increased with increasing doses and initial bromide concentrations. At 2.0 kGy, for initial bromide concentrations of 180.2 μgl^-1, 416.9 μgl^-1, 663.1 μgl^-1 and 823.9 μgl^-1 approximately 79.5%, 84.0%, 87.3% and 88.3% of bromide ions were transformed to bromate, respectively. And the analysis of bromide indicated that over 95% of the oxidized bromide was transformed to bromate for all the samples in Fig. 2. This means that the concentrations of intermediates were low.

Fig. 2.Full-screen View

(Color online) Effect of initial Br^- concentration and absorbed dose on bromate formation in N₂O saturated bromide solutions.

In the irradiating N₂O-saturated solutions, the primary reactive radicals were ·OH radicals due to reactions (11) and (12).

When tert-butanol (acted as an ·OH scavenger) was added to the solution of 0.1–25 mM concentration, no bromate was detected after 4.0 kGy irradiation for N₂O saturated solutions with 823.9 μgl^-1 Br^-. From the results, one knows that ·OH radicals can oxidize bromide to bromate. Von Gunten and Oliveras reported that bromate formed with ·OH radicals being the only oxidants and HOBr/OBr^- are requisite intermediates [11].

B. Effect of pH

Figure 3 shows the effect of pH value on the formation of bromate in N₂O saturated bromide-containing waters. The formation of bromate was nearly the same at pH 3.5–7.0, and decreased by about 5% at pH 2.5. At the pH 10.8, however, the bromate concentration decreased to 46.0 μgl^-1 from about 550 μgl^-1 in neutral conditions.

Fig. 3.Full-screen View

Effect of pH on bromate formation for N₂O saturated bromide-containing waters.

This shows the pH value dependence of G-values of the species formed in water radiolysis [12]. While the G-values of ·OH, $e_{aq}^{-}$ and ·H are almost constant in near neutral conditions, the H⁺ in higher concentration can react with e_aq^-(H⁺ + e_aq^- →·H), hence the decreased concentration of ·OH radicals according to Eq. (11) [12], which resulted in a small decrease of bromate concentration compared to neutral conditions. In alkaline conditions, the G-value of ·OH radicals decreases as they are transformed to less reactive ·O^- species, and more reductive species, like e_aq^- and less reactive HO₂^-·, were formed [12]. The oxidized bromide ions may be reduced by these reductive species. Therefore, the bromate formed at pH 10.8 was much less than that at lower pH values.

C. Effect of common anions

Natural waters are complex matrixes, containing anions that may interfere with the oxidation of bromide by competing with the free radical species. To investigate the influence of common anions in drinking waters, 0.1–25 mM of Cl^-, CO₃^2-/HCO₃^-, NO₃^- or SO₄^2- were added to bromide-containing solutions. The results from solutions saturated with air, N₂ or O₂ showed that no bromate was found at 0.5–6.0 kGy for solutions in initial bromide concentrations of 81.7 μgl^-1 and 1636.6 μgl^-1.

Figure 4 depicts the influence of the selected common anions on the formation of bromate in N₂O saturated waters. The amounts of bromate ion formed were unaffected by SO₄^2- ion, and increased slightly with the Cl^- ions. However, in the presence of CO₃^2-/HCO₃^- (predominantly HCO₃^- when the solution was neutral), the bromate concentration decreased notably to 44.2 μgl^-1 from 554.0 μgl^-1 added with 1 mM CO₃₂^-/HCO₃^- after 4.0 kGy irradiation. The bromate concentration decreased significantly only at high NO₃^- concentrations, as depicted in Fig. 4. This shows that the bromate concentration was not substantially affected by 1 mM NO₃^- addition, whereas added with 10 mM NO₃^-, it decreased from 554.0 μgl^-1 to 190.1 μgl^-1.

Fig. 4.Full-screen View

(Color online) Effect of common anions on bromate formation for N₂O saturated bromide-containing waters.

Cl^- ions could react with ·OH radical quickly to form ClOH.(k= 4.3×10⁹ mol^-1 s^-1), but ClOH^-· may reform ·OH radical by the fast reverse reaction(k = 6.1×10⁹ s^-1) in neutral solutions [13]. Then, the slightly increased bromate formation in the presence of Cl^- ion might be due to less radical recombination of ·OH radicals (such as ·OH+·H → H₂O, ·OH+·OH → H₂O₂, etc.)

Both CO₃^2- and HCO₃^- react with ·OH radicals and form ·CO₃^- radicals [13]. The results in Fig. 4 indicated that the ·CO₃^- radical only cannot oxidize Br^- and other bromine species to bromate. In the ozonation process, however, ·CO₃^- radicals, which are also produced from ⋅ OH, can oxidize BrO^- to BrO. (k = 4.3×10⁷ mol L^-1 s^-1) and potentially lead to an increase in bromate due to the presence of O₃ [14].

NO₃^- ion is not as efficient an ·OH scavenger as CO₃^2-/HCO₃^-, as revealed in Fig. 4, but for higher NO₃^- ion concentrations, NO₃^- ion can compete with N₂O for ·e_aq^- (NO₃^- + ·e_aq^- → NO₃^2-, k= 9.7×10⁹ mol L^-1 s^-1), resulting in the notably decrease of ·OH concentration and then the decrease of bromate formed.