Environ. Sci. Technol. 2004, 38, 5246-5252
Degradation of tert-Butyl Alcohol in Dilute Aqueous Solution by an O3/UV Process TEMESGEN GAROMA* AND MIRAT D. GUROL Department of Civil and Environmental Engineering, San Diego State University, San Diego, California 92182
This paper describes the degradation of tert-butyl alcohol (TBA) in dilute aqueous solution by an O3/UV process. The degradation process was investigated experimentally in a semi-batch reactor under various operational conditions, i.e., ozone gas (O3) dosage, UV light intensity, and water quality in terms of varying bicarbonate concentration. TBA was oxidized rapidly in the O3/UV system, and acetone, hydroxy-iso-butyraldehyde, and formaldehyde were identified as primary intermediates, whereas pyruvaldehyde and acetic, formic, pyruvic, and oxalic acids were generated as a result of further oxidation process. A good organic carbon balance was obtained, indicating that most reaction intermediates have been identified and quantified.
Introduction tert-Butyl alcohol (TBA) is a fuel oxygenate and a common co-contaminant in methyl tert-butyl ether (MTBE) contaminated groundwater. TBA can be found as coproduct during the manufacturing of MTBE, and more importantly it is produced in groundwater as a degradation byproduct of MTBE (1-3). TBA is completely soluble in water and therefore very mobile in groundwater systems. Currently, TBA is not an EPA priority groundwater pollutant, however, the state of California and New Jersey have proposed action levels of 12 and 100 µg/L, respectively, for TBA in drinking water (4, 5). The detection of TBA in groundwater has created great concern in the scientific community because TBA is a known toxin and suspected possible human carcinogen (6). Because of TBA’s special physical properties, conventional treatment methods are not quiet effective for its removal from contaminated waters. Some of the chemical and physical properties of TBA are summarized in Table 1 along with those of MTBE. Its complete solubility in water and low log Koc value limit the extent of its adsorption on activated carbon surfaces; as a result the carbon must be replaced or regenerated very often to reach desired treatment efficiencies. Adsorption experiments conducted in our laboratory (unpublished data) showed that TBA has at least four times less affinity for carbon surfaces than MTBE at equal concentrations. Due to its low Henry’s constant, air stripping is not easily applicable for removal of TBA, i.e., the air-to-water ratio must be very high to reach desired treatment efficiencies, which also increases the cost of treatment. Similarly, a number of studies indicated that under reducing states, conditions under which most contaminated groundwater exists, TBA biodegrades very slowly (1, 7-9). On the other * Corresponding author e-mail:
[email protected]. Currently affiliated with MWH Soft, Pasadena, CA. 5246
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 19, 2004
TABLE 1. Chemical and Physical Properties of TBA and MTBE property
TBA
MTBE
ref
molecular structure molecular weight (g/mol) density (20 °C) (g/mL) water solubility (mg/L) log Kow (25 °C) [-] log Koc (25 °C) [-] Henry’s law constant (25 °C) [-]
(CH3)3COH 74.12 0.79 miscible 0.35 1.57 0.00059
(CH3)3COCH3 88.15 0.74 48000 1.20 1.04 0.045
10 10 10 10 11 11 11
hand, MTBE biodegrades at a much higher rate than TBA under both reducing and oxidizing conditions (2, 3), and since MTBE biodegrades to TBA in groundwater, it is likely that TBA will accumulate in groundwater. Therefore, there is a need for an alternative treatment process for the removal of TBA from contaminated waters. Advanced oxidation processes (AOPs), which involve the application of various combinations of ozone (O3), hydrogen peroxide (H2O2), ultraviolet light (UV), and semiconductors (e.g., TiO2), provide promising treatment options. The major advantage of AOPs is attributed to the hydroxyl radical (OH•) which attacks indiscriminately almost any organic compound with a very high reaction rate constant. In addition, the OH• oxidizes a wide variety of organic pollutants into innocuous intermediates and end products, as opposed to conventional treatment methods, such as activated carbon and air stripping, in which pollutants are transferred from water to another medium that may need further treatment before final disposal. To date, no detailed study primarily focused on the removal of TBA from contaminated waters using any advanced oxidation processes (AOPs) has been reported in the literature. While investigating degradation pathways of MTBE by a UV/H2O2 process, Stefan et al. (12) oxidized TBA in a UV/H2O2 system and reported the formation of acetone, formaldehyde, and hydroxy-iso-butyraldehyde (HiBA) as primary reaction intermediates. The sonolysis of TBA in aqueous solution has also been reported in the literature (13). In the sonolysis study, TBA was employed to determine the extent of hydroxyl radical (OH•) scavenged in the gas phase within a collapsing cavitational bubble; and acetone, formaldehyde, and a multitude of other products have been reported as reaction intermediates. Considering the environmental concern posed by TBA and the limited data available in the literature on the treatment of TBA contaminated waters, the main objective of this study is to investigate application of an O3/UV process for the removal of TBA in dilute aqueous solution under different operational conditions, and at the same time identify the degradation products formed. This is an original research, and the authors are not aware of any other publications or unpublished research on the treatment of TBA by an O3/UV process. The decomposition of aqueous ozone in pure waters is initiated by its reaction with hydroxide ion (OH-), and this reaction leads to the production of radicals that propagate the decomposition process by chains of radical reactions and produce hydroxyl radical (OH•) (14-16). However, this process is known to occur very slowly (14) and the decomposition process can be further accelerated with ultraviolet radiation. In aqueous phase, the photolysis of ozone produces hydrogen peroxide (H2O2) as primary product (17), which then can photolyze into hydroxyl radical (OH•) directly and can react with aqueous ozone to produce hydroxyl radical 10.1021/es0353210 CCC: $27.50
2004 American Chemical Society Published on Web 09/03/2004
TABLE 2. Principal Reactions in O3/UV System in Natural Waters
reaction
O3 +
OH -
constant
-
f HO2 + O2
rate constants (M-1s-1)/ equil. constants (pKa)/ primary quantum yields
k1 ΦP
70 0.48
H2O2 98 2OH•
ΦP′
0.50
H2O2 T HO2- + H+ O3 + H2O2 f HO2-• + OH• + O2 O3 + HO2- f O2-• + OH• + O2 O3 + O2-• f O3-• + O2 O3-• + H+ f OH• + O2 OH• + O3 f HO2• + O2 HO2• T O2-• + H+ OH• + H2O2 f HO2• + H2O OH• + HO2- f HO2• + OHOH• + HCO3- f CO3-• + H2O OH• + CO32- f CO3-• + OHH2CO3 T HCO3- + H+ HCO3- T CO32- + H+ CO3-• + H2O2 f HO2• + HCO3CO3-• + HO2- f O2-• + HCO32OH• f H2O2 OH• + O2-• f O2 + OHOH• + HO2• f H2O + O2 OH• + O3-• f HO2• + O2-• O2-• + HO2• + H2O f H2O2 + O2 + OH2HO2• f H2O2 + O2 O2-• + CO3-• f O2 + CO32O3-• + CO3-• f O3 + CO32OH• + Mi f productsa OH• + Sj f productsa O3 + Mi f productsa
pKa3 k2 k3 k4 k5 k6 pKa4 k7 k8 k9 k10 pKa1 pKa2 k11 k12 k13 k14 k15 k16 k17 k18 k19 k20 kbi kj kdi
hv
O3 + H2O 98 H2O2 + O2 hv
11.8 1.0 × 10-2 5.5 × 106 1.6 × 109 9.0 × 1010 1.1 × 108 4.8 2.7 × 107 7.5 × 109 8.5 × 106 3.9 × 108 6.3 10.3 4.3 × 105 1.0 × 107 5.5 × 109 9.4 × 109 6.6 × 109 8.5 × 109 9.7 × 107 8.3 × 105 6.5 × 108 6.0 × 107 -
ref 14 18 19 20 14 14 15 21 22 23 24 24 25 25 26 26 27 27 25 28 28 21 23 23 29 30 -
a Where M represents the target organic compounds and S represents the scavengers of OH• other than the target organic compounds and i j carbonates.
(OH•). The principal reactions involved in an O3/UV system in natural waters are summarized in Table 2, adapted from Gurol and Akata (18), who successfully modeled the O3/UV system from kinetic and mechanistic points of view. The list in the table includes reactions that initiate, propagate, and terminate the decomposition process. The rate constants for the reactions involved are also included in the table.
Experimental Approach The laboratory studies were conducted by using a quartz reactor placed inside a concentric photochemical chamber that contained 16 removable low-pressure mercury lamps. The reaction vessel has an internal diameter of 13 cm, and was filled with 3 L of solution during experiments. The lamps emit UV light primary at 254 nm at a power rated as 2.2 W per lamp. The schematic diagram of the experimental setup used in this study is shown in Figure 1. A glass diffuser was used to sparge ozone gas into the solution. Influent ozone gas concentration in the range of 25-50 mg/L, at constant flow rate of 1.5 L/min, was used to investigate the effect of ozone dosage on the removal rate of TBA in the O3/UV system. The alkalinity of the aqueous solution was adjusted by sodium bicarbonate (NaHCO3), and the effect of bicarbonate on removal efficiency of the TBA was investigated in the range of 2-8 mM. Dilute solutions of sulfuric acid (H2SO4) and sodium hydroxide (NaOH) were used to adjust the pH of the aqueous solution to a value of 7.0 at the start of each experiment. It was observed that pH increased in all experiments, recorded as high as 8.5 at the end of some experiments. Since CO2 is produced in the oxidation process, pH is expected to decrease. The increase in pH is most likely due to the scavenging of CO2 by OH•. Gurol and Akata (18) reported that the rate of ozone photolysis was not significantly
affected by varying pH in the range of 6-9. Since ozone photolysis is the major initiator of ozone decomposition in the O3/UV system and TBA is nonionizable, variations in pH are not expected to affect the rate of TBA removal in the O3/UV system. Therefore, no attempt was made in this study to investigate the effect of pH on the rate of TBA removal. The reactor was operated in the semi-batch mode and the contents were stirred continuously using a magnetic stirrer. For each experiment, the appropriate numbers of UV lamps were inserted into the photochemical chamber. The TBA solution prepared at desired concentration was placed inside the reactor. At time zero, ozone gas was introduced into the reactor at the desired flow rate and the UV lamps were switched on. The experiments were conducted at room temperature, and the increase in temperature due to UV light exposure was less than 2 °C. Samples were withdrawn periodically and analyzed for residual concentration of TBA, reaction intermediates formed, and total organic carbon (TOC), and these values were used to check the material balance for organic carbon. The degradation of TBA in O3/UV system was investigated under various ozone gas dosages, UV light intensities, and water quality in terms of bicarbonate at constant gas flow rate of 1.5 L/min and pH of 7.0. Other experimental conditions used in this study are summarized in Table 3. The initial concentration of TBA was prepared at 1.1 mM for all the experiments, and the slight variations in the range of 1.001.20 is due to errors involved in solution preparation. Experimental runs 1 and 2 were controls with no UV or no ozone, respectively. In the ozonation process, ozone gas was continuously bubbled into the solution at constant flow rate of 1.5 L/min, whereas for O3/UV process, ozone gas was continuously bubbled into the solution and at same time the VOL. 38, NO. 19, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5247
FIGURE 1. Experimental setup.
TABLE 3. Experimental Conditions exp. [TBA]o Io run (mM) (W/L) 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1.20 1.20 1.06 1.14 1.16 1.00 1.11 1.20 1.09 1.10 1.20 1.11 1.15 1.10
0.00 2.73 1.35 1.35 1.35 2.73 2.73 2.73 4.28 4.28 4.28 2.73 2.73
avg. (O3)inf (mg/L)
avg. (O3)eff [O3]water bicarbonate (mg/L) (µg/L) (mM)
39 0 (O3)inf* ) 27 39 49 27 (O3)inf* ) 39 49 27 39 (O3)inf* ) 49 39 39
>5 0 5 >5
