Environ. Sci. Technol. 2009, 43, 2054–2059
Bromate Ion Removal by Electrochemical Reduction Using an Activated Carbon Felt Electrode NAOYUKI KISHIMOTO* AND NOBUAKI MATSUDA Faculty of Science and Technology, Ryukoku University, Otsu 520-2194, Japan
Received November 6, 2008. Revised manuscript received January 13, 2009. Accepted January 15, 2009.
The electrochemical removal of bromate ion (BrO3-) was investigated using a two-compartment electrolytic flow cell with activated carbon felt electrodes. Bromate ion removal and corresponding Br- increase was observed during electrochemical treatment, whereas the activated carbon felt used possessed no catalytic effect on BrO3- reduction. The BrO3reduction rate was accelerated at lower pH, which also improved current efficiency. Transition of chemical equilibrium of the BrO3- reductive reaction was theorized as the reason for pH dependencyoftheBrO3- reduction.Theelectrochemicaltreatment of BrO3--contaminated tap water resulted in a rapid decrease in BrO3- concentration from 100 to 48 µg/L with a contact time of 9.2 s. Thus, electrochemical treatment allowed the rapid removal of BrO3-. However, competitive hydrogen evolution at the cathodes reduced current efficiency of BrO3- reduction. Standard potentials of corresponding anodic and cathodic reactions suggested that electrolysis at a terminal voltage less than 1.229 V would promote BrO3- reduction without hydrogen evolution. However, the activated carbon felt electrode did not function well at a terminal voltage of 1.0 V. Accordingly, the development of an electrode material with high catalytic activity will be required to improve current efficiency.
Introduction Bromate ion (BrO3-) is a disinfection byproduct in water supplies. The toxicity of BrO3- was summarized by Butler et al. (1). Although evidence of carcinogenicity in humans is inadequate, the International Agency for Research on Cancer (IARC) has classified it as a Group 2B substance (possibly carcinogenic to humans). The World Health Organization (WHO) recommends a provisional guideline value of 0.01 mg/L for drinking water (2). Production of BrO3- occurs through ozonation of water containing bromide ion (Br-), in which ozone and/or hydroxyl radical oxidize Br- into hypobromous acid (HOBr) and/or hypobromite radical (BrO•). Then, HOBr and BrO• are further oxidized into BrO3- by ozone and/or hydroxyl radical (3-6). In addition, Weinberg et al. (7) reported that commercially produced solutions of sodium hypochlorite (NaOCl) used for disinfection of drinking water contain BrO3- as a contaminant (median value: 254.8 µg of BrO3-/g of free chlorine). They reported that the increase in BrO3- concentration in water plants without ozonation was due mainly to BrO3- contamination in feedstock NaOCl solutions. Thus * Corresponding author phone: +81 77 544 7107; fax: +81 77 544 7130; e-mail:
[email protected]. 2054
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 6, 2009
both BrO3- formation during ozonation and BrO3- contamination through feedstock NaOCl solutions play important roles in the increase in BrO3- concentration in water plants. There are two approaches to control BrO3- pollution in water supply, bromate minimization before formation and bromate removal after formation (8). The most popular technology to minimize BrO3- contamination before formation is optimization of ozonation operating conditions. For example, pH depression during ozonation reduces the generation of hydroxyl radicals and limits BrO3- formation through the hydroxyl radical pathway (9), and Kim et al. (10) succeeded in reducing BrO3- formation during ozonation of secondary effluents by maintaining the ozone dose at less than 0.1 mg/L. However, fewer hydroxyl radicals and low ozone dose results in the slower degradation rate of micropollutants. Moreover, pH depression leads to the formation of HOBr and brominated organic compounds (8). Ammonia addition is an alternative technology to reduce BrO3- formation during ozonation (8, 11). Ammonia does not alter ozone stability and reacts with HOBr, which is an intermediate of BrO3- formation pathways from Br- (11). This is an economical countermeasure for minimizing BrO3-. However, the ammonia residue increases chlorine demand. Therefore, post-treatment may be required for removal of residual ammonia. Although pH depression and ammonia addition are thought to be applicable in drinking water treatment (12), these BrO3- minimization technologies are not effective in controlling BrO3- contamination through feedstock NaOCl solutions. Bromate removal after formation is effective to decrease BrO3- loaded to a water stream by ozonation and prechlorination. Various technologies for BrO3- removal have been discussed, such as filtration, UV irradiation, photocatalytic decomposition, arc discharge, coagulation, reducing agents such as ferrous iron, activated carbon techniques, and biological remediation (1). Butler et al. (1) pointed out the most developed technologies were chemical removal by ferrous iron and biological activated carbon (BAC) treatment. Ferrous iron reduces BrO3- to Br- and is oxidized into ferric iron as follows: BrO3- + 6Fe2+ + 6H+ T Br- + 6Fe3+ + 3H2O
(1)
However, the issue of cost-effectively removing residual iron from the water stream must be resolved prior to development of a viable full-scale technique (1), because the WHO guidelines for drinking-water quality recommends iron concentrations less than 300 µg/L (2). Two processes are involved in the BAC treatment: catalytic reduction of BrO3- by activated carbon and biological remediation. The catalytic BrO3- removal by granulated activated carbon (GAC) was well discussed by many researchers and virgin GAC was effective for reducing BrO3to Br- (13-17). However, Asami et al. (14) and Huang and Chen (17) reported the transition of GAC to BAC apparently decreased BrO3- removal rate. Some research work demonstrated the successful bioremediation of BrO3-, but the bioremediation required addition of electron donors such as ethanol and glucose and/or removal of dissolved oxygen before biological treatment (18-20). Accordingly, Hijnen et al. (18) concluded that the bioremediation of BrO3- would not be feasible in water treatment because of the longer reaction time needed and the requirement for a posttreatment process to remove residual biomass and dissolved organic matter. Xie and Shang (8) also summarized BrO3- removal technologies and pointed out that the chemical reduction 10.1021/es803144w CCC: $40.75
2009 American Chemical Society
Published on Web 02/18/2009
by zerovalent iron might be an efficient, cost-effective method for BrO3- control. However, coexisting anions such as phosphate, nitrite, chlorate, etc., and dissolved oxygen inhibited the BrO3- reduction by zerovalent iron (21, 22), and the repeated use of iron resulted in the decrease in its reactivity (22). Accordingly, extensive pilot studies are recommended to realize its best application in drinking water production (8). Consequently, an all-round prevention technology of BrO3- contamination has not been established yet. Therefore, it is meaningful to provide a new option to remove BrO3from the water stream. Mussini and Longhi (23) summarized the standard potentials of BrO3- and HOBr as follows:
TABLE 1. Water Quality of Original Tap Water item cations Na+ K+ Ca2+ Mg2+ NH4+ anions FClBrNO3SO42PO4-P BrO3ClO3pH electric conductivity free chlorine combined chlorine
BrO3- + 5H+ + 4e- T HOBr + 2H2O 1.447 V vs SHE (standard hydrogen electrode) (2) HOBr + H+ + 2e- T Br- + H2O
1.341 V vs SHE (3)
These potentials clearly show BrO3- and HOBr act as oxidants and electrochemical reduction of BrO3- may proceed at the cathode in an electrochemical reactor. Several electrode materials catalyzing BrO3- reduction were reported such as molybdenum oxide (24), tungsten oxide (25), and polyoxometalate (26). These metallic oxides were successfully applied to amperometric detection of BrO3-. However, application of the metallic oxides for water treatment is thought to be difficult because of their unstability as a cathode. For example, catalytic reduction of BrO3- by tungsten oxide (WO3) was observed from 0.05 to -0.25 V vs saturated calomel electrode (SCE) (25). However, WO3 itself is electrochemically reduced at -0.22 V vs SCE as follows (25): WO3 + 2yH+ + 2ye- T WO3-y + yH2O
a
value 10.0 mg/L 1.5 mg/L 11.6 mg/L 2.4 mg/L N.D.a 0.11 mg/L 15.2 mg/L 0.3 mg/L 1.3 mg/L 12.7 mg/L 0.01 mg/L N.D. 0.08 mg/L 7.0 155 µS/cm 0.15 mg/L 0.10 mg/L
N.D.: not detected.
(4)
A stability of an electrode is an important factor in water treatment. Accordingly, BrO3- reduction by an electrochemically stable electrode should be discussed for water treatment. One of the purposes of this study was to investigate effects of pH and terminal potential on electrochemical reduction of BrO3- using activated carbon felt electrodes, which are electrochemically stable, have large surface area, and may have a catalytic capability of reducing BrO3- like that of GAC. The other purpose is to demonstrate an electrolytic flow cell applicable to water with low electric conductivity like drinking water.
Experimental Section
Materials. A 0.100 mmol/L BrO3- solution was prepared by dilution of potassium bromate solution (1/60 mol/L KBrO3, analytical grade) with distilled and ion-exchanged water. The solution pH was adjusted by addition of sulfuric acid (H2SO4, guaranteed grade) and potassium hydroxide (KOH, guaranteed grade). All chemicals were purchased from Nacalai Tesque, Kyoto, Japan, and used without further purification. Tap water contaminated with BrO3- was prepared by addition of the KBrO3 solution to a final BrO3- concentration of 100 µg/L. The pH was adjusted to 2.0 by addition of H2SO4. Water quality of the original tap water is summarized in Table 1. Apparatus. Figure 1 shows a schematic view of the experimental setup and the structure of the two-compartment electrolytic flow cell divided by a cation exchange membrane (Nafion N-117, DuPont, Wilmington, DE). Table 2 summarizes the experimental conditions at all runs. At runs A and C through N the both anodic and cathodic compartments were filled with an electrochemically conductive activated carbon felt (XF308, Toyobo, Osaka, Japan), which contacted the cation exchange membrane and a metallic electrode, a platinum-coated titanium anode, and a stainless steel (ANSI304) cathode, with 50 cm2. The area,
FIGURE 1. A schematic view of experimental setup using activated carbon felt electrodes for batch experiments and the structure of the electrolytic flow cell. thickness, weight, and specific surface of the felt were 50 cm2, 5 mm, 1.98 g, and 50 m2/g, respectively. Empty compartment volume and actual compartment volume of each compartment were 25 and 8.2 cm3, respectively. The BrO3- solution was fed into the cathodic compartment, and distilled and deionized water was fed into the anodic compartment. The flow rate was 298 mL/min for batch VOL. 43, NO. 6, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2055
TABLE 2. Experimental Conditionsa run mode A B
batch
b
batch
C
batch
D
batch
E
batch
F
batch
G batch H
batch
I
batch
J
batch
K
flow
L
flow
M flow N flow
sample 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq 0.100 mmol/L KBrO3aq tap water with 100 µg/L BrO3tap water with 100 µg/L BrO3tap water with 100 µg/L BrO3tap water with 100 µg/L BrO3-
flow rate (mL/min)
pH
terminal current voltage (V) (A)
298
4.2 ( 0.3
298
2.3 ( 0.1
(5.9 ( 0.5)
0.20
298
2.2 ( 0.1
(7.1 ( 0.3)
0.20
298
3.4 ( 0.1
(7.6 ( 0.4)
0.20
298
4.9 ( 0.3
(7.7 ( 0.6)
0.20
298
9.1 ( 0.1
(7.8 ( 0.1)
0.20
298
9.9 ( 0.1 (10.0 ( 0.4)
0.20
298
2.3 ( 0.1
1.0
(
