Worldwide attention has been attracted to 1,4-dioxane because of its probable human carcinogenicity and frequent occurrence in surface waters and wastewaters. Thus, many countries and organizations have set limits for the amount of this material in drinking water and wastewater effluent. However, the removal of 1,4-dioxane during traditional treatment processes, even ozonation (pH < 7), has been limited. Therefore, 1,4-dioxane removal during catalytic ozonation was investigated in this study, and activated carbon was selected as the ideal catalyst. The removal efficiency of 1,4-dioxane by ozonation was promoted significantly using activated carbon compared with that of ozonation only. Tert-butyl alcohol significantly reduced the removal efficiency of 1,4-dioxane during catalytic ozonation, which suggested that hydroxyl radicals (·OH) were formed during catalytic ozonation and played an important role in decomposing 1,4-dioxane. Additionally, results concerning the stability of activated carbon indicated that the catalytic activity of this catalyst remained steady during ozonation.
The non-volatile organic substance 1,4-dioxane (CAS No. 123-91-1) is widely used as a solvent, solvent stabilizer, emulgator and detergent in industries, particularly in pesticide, dye and medicine production. Because of its wide usage and high aqueous solubility (4.31 × 105 mg/L), 1,4-dioxane is frequently detected in wastewaters, surface waters and groundwaters. For example, 1,4-dioxane was detected in Bedford, Massachusetts, USA, at a maximum concentration of 2,100 μg/L in four municipal supply wells (Weimar 1980). At a 2,4,6-trichloroanisole-impacted site, 38 out of 57 wells contained 1,4-dioxane, and the concentrations of 1,4-dioxane in 21 wells were more than 43 ± 1 μg/L with a maximum concentration of 2,800 μg/L (Isaacson et al. 2006). Between 1995 and 1998, 83 out of 95 samples of river water, seawater and groundwater were found to contain 1,4-dioxane between 1.9 μg/L and 94.8 μg/L in the Kanagawa Prefecture of Japan (Abe 1999). Additionally, the concentrations of 1,4-dioxane in the Main, Rhine and Oder Rivers were 2.2, 0.86 and 0.86 μg/L, respectively (Stepien et al. 2014).
Notably, 1,4-dioxane is classified as a probable human carcinogen listed in group 2B by the International Agency for Research on Cancer and group B2 by the US Environment Protection Agency (EPA). This pollutant can lead to nasal tumours (Haseman & Hailey 1997), liver tumours and skin cancers in mice and rat experiments (King et al. 1973; Bull et al. 1986; Lundberg et al. 1987). Therefore, some countries and organizations have set limits for 1,4-dioxane. A provisional guideline value for 1,4-dioxane based on a specific cancer risk level of 10−5 was reported as 30 μg/L by Schriks et al. (2010). The US EPA has set 30 μg/L as a discharge limit for wastewater treatment plants (Stefan & Bolton 1998). The water quality standards of Japan for drinking water set a tolerable limit of 1,4-dioxane at 50 μg/L (Ministry of Health, Labour and Welfare, Japan). The German Federal Environmental Agency sets 0.1 μg/L as the precautionary guideline limit in drinking water (Stepien et al. 2014).
The removal efficiency of 1,4-dioxane in wastewater treatment plants is only 2.12%, according to the EPISuite™ website of the US EPA. Thus, conventional water treatment processes, such as physical pre-treatment, biological treatment and several chemical treatments, including chemical coagulation and ozonation, are ineffective at removing 1,4-dioxane.
The fate of 1,4-dioxane in domestic sewage and drinking water treatment processes has been investigated in previous studies. Concentrations of 1,4-dioxane in the influent and effluent of sewage water treatment plants ranged from 0.262 ± 0.032 μg/L to 0.834 ± 0.48 μg/L and 0.267 ± 0.035 μg/L to 62.26 ± 36 μg/L, respectively. In drinking water, 1,4-dioxane concentrations after bank filtration and drinking water facilities were 0.6 μg/L and 0.49 μg/L with a concentration range of 0.65 to 0.67 μg/L in raw water, respectively (Stepien et al. 2014). The results indicated that traditional sewage and drinking water treatment processes did not effectively remove 1,4-dioxane.
Research concerning the fate of 1,4-dioxane in nine on-site industrial wastewater treatment plants also found that the removal efficiencies of 1,4-dioxane in most treatment processes were low (activated sludge, 16.2 ± 15.2%; rotating biological contactor, 10%; chemical coagulation reactor, 7.7 ± 3.0% and sand filtration, 1.1 ± 1.1%) (Lee et al. 2011).
As an unstable and strong oxidant (E = +2.07 V), ozone has a high potential for discoloration, disinfection and organic matter removal; however, ozone cannot decompose several micro-pollutants such as 1,4-dioxane (pH < 7). The reaction kinetic constant of 1,4-dioxane (kO3) is as low as 0.32 M−1 s−1 (Hoigné & Bader 1983). Advanced oxidation processes (AOPs) that attempt to produce more powerful oxidants including hydroxyl radicals (·OH) are better options. TiO2 photocatalysis, O3/H2O2 and O3/electrolysis (Suh & Mohseni 2004; Kishimoto et al. 2008; Vescovi et al. 2010) have been proven to be effective in decomposing 1,4-dioxane and enhancing its biodegradability; however, these AOPs have several disadvantages, including high energy consumptions.
Catalytic ozonation is an effective AOP with a high potential for decomposing persistent organic pollutants that can improve ozone efficiency (Kasprzyk-Hordern et al. 2003). Nano-TiO2 powders, particularly rutile TiO2, have significant catalytic activity in removing nitrobenzene during catalytic ozonation (Yang et al. 2007). Activated carbon has been widely used in treating drinking water because of its high adsorption capacity and abundant support for microorganism growth. Several studies concerning hazardous pollutant removal report that activated carbon is an effective catalyst in removing oxamic acid, oxalic acid and 1,3,6-naphthalenetrisulfonic acid by ozonation (Sánchez-Polo & Rivera-Utrilla 2003; Faria et al. 2008). However, research concerning 1,4-dioxane removal by catalytic ozonation with activated carbon has not been reported.
Therefore, this study attempted to evaluate 1,4-dioxane removal characteristics during activated carbon catalytic ozonation. The effect of tert-butyl alcohol (TBA) during activated carbon catalytic ozonation was also investigated. The stability of activated carbon during ozonation has also been evaluated.
MATERIALS AND METHODS
Materials and chemicals
Certified 1,4-dioxane (purity >99.5%) was purchased from Dr. Ehrenstorfer GmbH of Germany. Briquetted carbon supplied by the ShanXi XinHua Activated Carbon Co., Ltd was made of anthracite. Multiwall carbon nanotubes (MWCNTs) were purchased from Chengdu Organic Chemistry Co., Ltd, China. Other activated carbons including coconut carbon and charcoal (activated carbon) were purchased from ANPEL (Shang Hai, China). Methylene chloride was purchased from JT Baker with a purity for high-performance liquid chromatography. Other chemicals used in this study had analytical reagent purity.
Gaseous ozone provided by an O3 generator (Ozocenter, China) was added into the glass batch reactor (a cylinder that measured Φ = 0.09 m with a volume of 3 L, 0.50 m in height) that contained the 1,4-dioxane solution (C0 = 50 mg/L) and activated carbon at about 25 °C. The concentration of the inlet gaseous ozone was detected using an O3 detector (IDEAL-2000, China) and maintained at 27 ± 2 mg/L. The pH value of the water solution was adjusted and maintained at 7.0 ± 0.1 using a phosphate buffer. Water samples of approximately 50 mL were taken during ozonation, and the residual ozone in the water samples was quenched using sodium thiosulfate for 1,4-dioxane and dissolved organic carbon (DOC) analysis. The DOC value was determined using a DOC analyzer (TOC-VCPH, Shimadzu).
Experiments for evaluating the contribution of hydroxyl radicals
TBA, an effective hydroxyl radical scavenger at a concentration of 1 g/L was added into the solution that contained activated carbon and 1,4-dioxane to determine whether the hydroxyl radical was an oxidant in this catalytic ozonation system. Residual 1,4-dioxane concentrations in the water samples after ozonation were determined to compare the reaction rates of 1,4-dioxane with and without TBA addition.
Stability evaluation of the selected activated carbon
Activated carbon (3 g/L) was added into a water solution that contained 1,4-dioxane with an initial concentration of 50 mg/L. Approximately 50 mL of the solution was sampled for 1,4-dioxane analysis after ozonation for 0.5 h. The residual solution in the batch reactor after sedimentation for 10 min was discarded. Fresh 1,4-dioxane solution (50 mg/L) was added into the reactor that contained residual activated carbon, and then gaseous ozone was added into the mixture of activated carbon and 1,4-dioxane to repeat the reaction. The experiments were repeated six times with the same activated carbon and fresh 1,4-dioxane solution.
Given that activated carbon could adsorb pollutants, the removal efficiency of 1,4-dioxane during catalytic ozonation could possibly be influenced by adsorption. Original activated carbon and adsorbed activated carbon, which was used to adsorb 1,4-dioxane for 2 h, were collected and added into the 1,4-dioxane solution in the presence and absence of ozone, respectively. The concentrations of residual 1,4-dioxane in solutions were measured.
The water samples after ozonation or adsorption were filtered by 0.22 μm filters to eliminate residual activated carbon. Thereafter, the liquid–liquid extraction method was used to transfer 1,4-dioxane from the water solution to methylene chloride, according to a previous study (Park et al. 2005).
The concentration of 1,4-dioxane after extraction was determined by gas chromatography–mass spectrometry (GC system 7890A, Mass system 5975C inert XL MSD, Agilent Technologies, USA). The GC was equipped with a column of DB-5MS (30 m × 0.25 mm × 0.25 μm, Agilent Technologies, USA). Pulsed splitless injection of a 1 μL sample was applied at an injector temperature of 473 K. The initial oven temperature was 303 K and was held for 1 min. The oven was then run with a 280 K/min ramp to 363 K followed by a 303 K/min ramp to 573 K, which was held for 1 min. The retention time of 1,4-dioxane was approximately 4.0 min. The quantitation and confirmation ions were m/z 88 and m/z 58, respectively.
RESULTS AND DISCUSSION
Catalyst selection for 1,4-dioxane removal
The DOC releases from briquetted carbon and coconut charcoal during ozonation were negligible. Considering the price, briquetted carbon (0.33–0.66 US$/kg) made from anthracite was cheaper than coconut charcoal (1–2 US$/kg). Therefore, briquetted carbon was a better option as an ozonation catalyst than coconut charcoal. The removal characteristics of 1,4-dioxane during catalytic ozonation by activated carbon (activated carbon was used instead of briquetted carbon in following paragraphs) were further investigated in this study.
Degradation of 1,4-dioxane during catalytic ozonation
By fitting the data of ozone with activated carbon in Figure 3, the results showed that k′ was 2.97 × 10−4 s−1 (R2 = 0.997) and 9.13 × 10−4 s−1 (R2 = 0.995) for 1 g/L and 3 g/L of activated carbon, respectively. Beckett & Hua (2003) used ultrasound to decompose 1,4-dioxane by adding Fe(II) and found that the highest decomposition was obtained at 358 kHz; the first-order rate constants were 4.2 × 10−4, 8.2 × 10−4 and 11.5 × 10−4 s−1 with 0, 1 and 10 mM Fe(II), respectively. These constants were similar to the pseudo-first-order reaction kinetics results of this study. Approximately 3.4 mmol/L 1,4-dioxane reacted with 17 mmol/L H2O2 and UV illumination, and its pseudo-first-order kinetics constant was 5 × 10−4 s−1 (Kim et al. 2006), which was of the same order of magnitude as the results of this study.
Notably, the residual DOC after catalytic ozonation for 30 min was significantly lower than that after adsorption alone. However, the removal rate of DOC was significantly much lower than that of 1,4-dioxane, which indicated that catalyst ozonation could decompose 1,4-dioxane into several intermediate organic products. Research concerning whether the intermediate organic products are more easily adsorbed or could be further converted into CO2 with abundant O3 is still required.
Several studies concerning other AOPs found that DOC was difficult to remove completely (Nishijima & Okada 2003; Georgieva et al. 2011; Tzikalos et al. 2013) because some small-molecule acids, aldehydes and ketones were produced. Tzikalos et al. (2013) used UV-A irradiation (1.18 × 10−4 Einstein/min) and anatase/brookite TiO2 (200 mg/L) to decompose Reactive Red 195 (50 mg/L) from aqueous samples, whereas only approximately 55% of DOC could be removed after 420 min. Therefore, biological treatments after AOPs were required to decompose the residual organic matter.
Effects of radical scavengers
Other reaction mechanisms, including surface reactions, have been proposed for catalytic ozonation. Faria et al. (2008, 2009) stated that the oxidation of organic pollutants could possibly occur on the surface of activated carbon between the adsorbed reagent, surface (or aqueous phase) radical species and dissolved ozone.
The 1,4-dioxane degradation mechanism by catalytic ozonation was found to be very complex. Catalytic ozonation in this study was a three-phase reaction process that included adsorption, desorption, ozonation and free radical reactions. The possible pathways for activated carbon catalytic ozonation to decompose 1,4-dioxane are shown as following. Firstly, O3 generated ·OH in the presence of activated carbon, and ·OH converted 1,4-dioxane to other organic products and even CO2. Secondly, 1,4-dioxane was adsorbed on the surface of the catalyst, and the adsorbed 1,4-dioxane was then decomposed by oxidants (e.g., O3 or surface oxygenated radicals). Whether the adsorbed 1,4-dioxane on carbon could be removed by ozone still required further study.
Degradation stability of catalytic ozonation
Activated carbon, which is a cheap and stable catalyst, exhibited positive effects in removing 1,4-dioxane by ozonation. First, the removal rate of 1,4-dioxane was significantly increased by the combination of activated carbon and O3, unlike that of activated carbon or O3 alone. Catalytic ozonation with a high dose of O3 could remove the DOC in 1,4-dioxane solution. Second, the reaction kinetic constant was below 36% of its initial value in the presence of 1 g/L TBA, which certified that ·OH played an important role in catalytic ozonation in this study.
This study is supported by the National High-tech R&D Program of China (863 Program No. SS2013AA065205), the National Science Fund of China (No. 51138006) and the Collaborative Innovation Center for Regional Environmental Quality.