Catalyst support materials for prominent mineralization of bisphenol A ...

Environ Sci Pollut Res (2016) 23:10223–10233 DOI 10.1007/s11356-016-6251-y

RESEARCH ARTICLE

Catalyst support materials for prominent mineralization of bisphenol A in catalytic ozonation process Magda Cotman 1 & Boštjan Erjavec 1 & Petar Djinović 1 & Albin Pintar 1

Received: 4 November 2015 / Accepted: 4 February 2016 / Published online: 16 February 2016 # Springer-Verlag Berlin Heidelberg 2016

Abstract Degradation of aqueous solution of bisphenol A (BPA) has been investigated through non-catalytic and catalytic ozonation treatments conducted in a semi-batch reactor. Noncatalytic ozonation resulted in complete degradation of aqueous BPA in less than 3 min but did not completely convert the reaction intermediates of BPA ozonation into CO2 and H2O. The main goal of this study was to find an effective heterogeneous catalyst to increase the extent of BPA mineralization at different pH conditions. In this way, the most promising catalyst carrier was γ-Al2O3; at pH = 8.0, 68 % of total organic carbon (TOC) was removed in the period of 75 min, out of which 42 % was attributed to mineralization. Finally, 3.0 wt.% Ru/γ-Al2O3 catalyst exhibited over 82 % of TOC removal after 240 min of ozonation at pH = 5.9, of which 56 % was mineralized. Keywords Catalytic ozonation . BPA removal . Mineralization . γ-Al2O3 . Ru/γ-Al2O3

Introduction Endocrine disrupting chemicals (EDCs) are an emerging class of persistent contaminants released from several industrial Responsible editor: Santiago V. Luis Electronic supplementary material The online version of this article (doi:10.1007/s11356-016-6251-y) contains supplementary material, which is available to authorized users. * Magda Cotman [email protected]

1

Laboratory for Environmental Sciences and Engineering, National Institute of Chemistry, Hajdrihova 19, SI-1001 Ljubljana, Slovenia

processes (Belgiorno et al. 2007; Esplugas et al. 2007). One such EDC of concern is bisphenol A (BPA), a well-known compound used in polycarbonate polymers and epoxy resins. BPA finds its way into the environment through different sources, such as discharge from wastewater treatment plants and effluents from BPA production units, BPA consuming facilities, and from landfill waste sites. For example, in waste landfill leachates in Japan, a maximum concentration of 17.2 mg/L was reported (Yamamoto et al. 2001). Due to relatively slow biodegradability of this compound, it is imperative to find an effective alternative to remove BPA from polluted water streams. Currently, the research activities are focused on advanced oxidation processes (AOPs) for the destruction of synthetic organic species resistant to conventional treatment methods. The removal techniques of EDCs from waters and wastewaters have been comprehensively reviewed (Belgiorno et al. 2007; Basile et al. 2011). For the elimination of aqueous BPA, several methods have been investigated, including biological treatment (Melcer and Klecka 2011; Li et al. 2012); adsorption (Pan et al. 2008); chemical oxidation (Lin et al. 2009); and different AOPs, such as Fenton reaction (Li et al. 2008), photocatalysis (Watanabe et al. 2003), or combinations of these techniques (Torres-Palma et al. 2010; Erjavec et al. 2013; Mezohegyi et al. 2013). Among AOPs, ozonation has proved to be of real interest as an efficient tool for degrading aqueous organic contaminants. Ozone is one of the most active oxidants available; therefore, ozonation processes have a wide range of applications in drinking water disinfection, bacterial sterilization, and oxidative removal of micropollutants in water (Xu et al. 2002). Mineralization of BPA refers to the conversion of BPA and its by-products in an aquatic environment into CO2 and H2O. The application of ozonation has been increasing in recent years, yet the main disadvantage of this type of treatment

10224

remains related to the excessive formation of by-products, which can exert toxic and carcinogenic properties. Despite these, the use of ozonation as a pre-treatment process to transform refractory organic compounds into substances which are more easily removed by conventional methods seems to be economically attractive (Garoma and Matsumoto 2009). The gas–liquid process itself is a heterogeneous system, since it relies on contacting of gaseous ozone and waterdissolved organic pollutants. The role of the solid catalyst in the ozonation process is to improve the oxidation efficiency, and a variety of solids have been shown to accelerate the degradation of organic compounds. There are many claims of catalytic activity observed at a laboratory scale; however, there are no reports of upscaling the processes, which might be because of a poor separation of catalyst slurry from treated water. Catalytic ozonation is often classified as an advanced oxidation process; i.e., in situ generation of hydroxyl radicals is expected. In heterogeneous catalytic ozonation process, a catalyst is in a solid form, while the reaction may proceed in bulk liquid phase or on the catalyst surface (Kasprzyk-Hordern et al. 2003; Nawrocki and Kasprzyk-Hordern 2010; Nawrocki 2013). We can talk about the catalytic ozonation when the efficiency of ozonation in the presence of a catalyst is higher than without it at the same pH value. However, when investigating catalytic activity of a solid catalyst, another condition must be also fulfilled; the total effect of catalytic ozonation must be higher than a combined effect of adsorption on the catalyst surface and ozonation. Monitoring of pH values in both ozonation and catalytic ozonation processes is a basic requirement, which allows discriminating between two effects: (i) decomposition of ozone and (ii) formation of OH• radicals. It should be noted that both of these processes are highly pH dependent. A variety of solid catalysts has been described in the literature as effective for ozonation of organic molecules in aqueous solutions. Various metal oxides are claimed to show catalytic activity in ozonation. The main representatives of metal oxides and transition metal oxides are all forms of Al2O3 and TiO2, Fe2O3, and MnO2, respectively. Alumina and alumina-based catalysts have been reported to be effective in enhancing the removal efficiency of different organic pollutants in the presence of ozone. Batch experiments of catalytic ozonation of refractory organic compounds (initial dissolved organic carbon content was 60 mg/L) were performed to draw distinction between adsorptive and reaction processes (Ernst et al. 2004). The results of using an alumina catalyst in concentration of 1 g/L showed the total organic carbon (TOC) removal up to 90 % (Keykavoos et al. 2013); a significant improvement was achieved by reducing the particle size from pellets to powder. Mineralization of benzotriazole (BTZ) by heterogeneous catalytic ozonation was studied in aqueous phase. The performances of three catalysts (Mn/Al2O3, Cu/Al2O3, and Mn-Cu/ Al2O3) were compared at various pH levels and operating

Environ Sci Pollut Res (2016) 23:10223–10233

conditions. Compared to the non-catalytic ozonation, a higher level of mineralization of BTZ was achieved over a wide range of pH values (Roshani et al. 2014). The use of nano-TiO2 prepared by the sol–gel method as a catalyst for the ozonation of nitrobenzene in water significantly improved the removal of this compound compared to ozonation alone. TiO2 in the crystallographic form of rutile showed higher catalytic activity compared to anatase (Yang et al. 2007). The removal of clofibric acid from aqueous solution has been investigated in semi-continuous ozonation runs with 1 g/L of a commercial TiO2 catalyst (Rosal et al. 2009). Ozonation using 10.0 wt.% TiO2 and 87.3 wt.% Al2O3 was developed as an effective means to eliminate dimethyl phthalate (DMP) from aqueous solutions. The TOC removal was the most efficient in the process employing the TiO2/Al2O3 catalyst, followed by the Al2O3 catalyst (Chen et al. 2011). Oxalic acid was efficiently removed from water in the presence of ozone and a 15 wt.% TiO2/Al2O3 catalyst at pH = 2.5; the process of oxalic acid ozonation likely occurs over both alumina and titania (Beltrán et al. 2004). In this study, the following materials were examined in comparative catalytic ozonation experiments: γ-Al2O3, θ/αAl2O3, boehmite (AlO(OH)), amorphous silica alumina (ASA), β-zeolite, and γ-Al2O3-supported TiO2 (3.25, 7.5, and 15 wt.% of TiO2). Based on the ratio between catalytic BPA degradation and accumulation of its reaction derivatives, the most promising support for catalytic ozonation was proposed. Finally, γ-Al2O3-supported RuO2 was evaluated in this process as well.

Materials and methods Catalyst efficiency was determined by withdrawing liquidphase samples at pre-selected time intervals and determining the remaining BPA and TOC content by using highperformance liquid chromatography (HPLC) and TOC analyses, respectively. Morphological, textural, and surface properties of tested materials were examined by means of various advanced characterization techniques. The carbon, hydrogen, nitrogen, and sulfur (CHNS) analysis of fresh and spent catalysts was performed in order to evaluate the amount of carbonaceous deposits on catalyst surface during the reaction, which enabled us to estimate the suitability of prepared catalysts for a long-term purification of BPA in the process of ozonation. Catalyst support selection The mixed phase theta/alpha alumina (θ/α-Al2O3) was prepared by calcination (5 h at 1100 °C, 2 °C/min) of commercially procured γ-Al2O3 (Nikki-Universal Co., Tokyo, Japan). ASA was synthesized from silica gel (SBET = 313 m2/g, pore diameter = 9.4 nm, pore volume = 0.74 cm 3 /g;


manufactured by Kemika Zagreb) using the deposition precipitation method. A nominal loading of 15 wt.% Al3+ from aluminium nitrate precursor was deposited at 90 °C under reflux for 22 h. Urea was used to induce slow precipitation followed by filtration, washing, drying overnight, and calcination for 12 h at 800 °C. More details about the synthesis of ASA can be found in the work of Hensen et al. (2010). A sample of boehmite was kindly provided by Conden Chemie. The protonated β-zeolite sample (SiO2/Al2O3 ratio between 23 and 25) designated as TZB-212 was provided by Tricat Zeolites. γ-Al2O3-supported TiO2 catalysts were synthesized with three different loadings; 3.25, 7.5, and 15 wt.% of TiO2 were deposited over commercially procured γ-Al2O3 (NikkiUniversal Co., Tokyo, Japan) by means of wet impregnation (WI) technique, using an aqueous solution of TiOSO4 · xH2SO4 · xH2O (Sigma-Aldrich). The suspensions were dried at 60 °C overnight and calcined at 600 °C, both at a heating ramp of 2 °C/min for 3 h from the start at room temperature (RT) until the final calcination temperature was achieved. Ruthenium containing catalyst was prepared by impregnating the γ-Al2O3 (purchased from Nikki-Universal Company) with an aqueous solution of ruthenium trichloride (SigmaAldrich). The solution was added dropwise to the γ-Al2O3 powder (SBET = 167 m2/g) while mixing and being heated inside a ceramic crucible, placed on an electric plate. Following the impregnation, the prepared catalyst was calcined for 3 h at 350 °C in still air. Nominal noble metal content in the prepared Ru/Al2O3 catalyst was equal to 3 wt.%. Low noble metal loading as selected in this study can be more efficiently dispersed over the high surface area support, enabling the formation of very fine metal particles, as reported by Djinović et al. (2011). Characterization of supports and Ru/γ-Al2O3 catalyst The obtained supports and Ru/γ-Al2O3 catalyst were characterized with emphasis on a detailed examination of textural, structural, and surface chemical properties using scanning electron microscopy (SEM), X-ray powder diffraction (XRD), CHNS elemental analysis, and N2 physisorption. The morphology of selected supports was analyzed by means of a scanning electron microscope (FE-SEM Zeiss SUPRA 35VP). SEM-EDX mapping was performed using energy-dispersive detector (Inca 400, Oxford Instruments). For morphology determination of γ-Al2O3-supported Ru catalyst, a scanning electron microscope (FE-SEM Zeiss Ultra Plus) equipped with energy-selective backscattered (ESB) detector for clear compositional contrast was used to visualize RuO2 particles deposited on γ-Al2O3. XRD patterns were acquired using PANalytical X’pert PRO MPD diffractometer with CuKα1 radiation (1.54056 Å) in reflection geometry. Data collection was conducted in the range between 10 and

10225

90° (steps of 0.034°). Specific surface area of materials, total pore volume, and average pore size were determined by measuring nitrogen adsorption and desorption isotherms at −196 °C (Micromeritics, model TriStar II 3020). Brunauer-Emmett-Teller (BET) theory was applied in order to calculate the specific surface area of materials, while pore size distribution was calculated from the desorption branch of the corresponding nitrogen isotherms, employing the BJH method. The amount of carbon accumulated on the catalyst surface during catalytic ozonation experiments was determined by means of CHNS elemental analysis (Perkin Elmer, model 2400 Series II). At the end of each ozonation experiment, the catalyst was recovered from the solution by filtering through a 0.45-μm filter followed by gentle rinsing with deionized water. The washed catalysts were dried in an oven at 105 °C for 60 min to remove molecular water present on the surface of the catalysts. CHNS elemental analysis of fresh and spent solids was performed. Considering the carbon-based elemental analysis of examined catalyst supports and measured corresponding TOC removals (TOCrem), one can estimate the amounts of TOC accumulated (TOCaccu) on the catalyst surface. If TOCaccu is then subtracted from TOCrem, a true TOC mineralization (TOCmin) can be determined. In this way, TOCaccu and TOCmin values were determined for all samples used in catalytic ozonation experiments. To measure the pH of the point of zero charge (pHPZC) above which the catalyst surface is negatively charged, carefully weighed amounts of examined solids were added sequentially to 50 mL of aqueous 0.005 M NaCl until the pH value of the solution did not change with further solid additions. The equilibrium pH value at the plateau of the curve corresponds to the pHPZC. The well-mixed cell in which the measurements were performed was thermostated at T = 25 °C and continuously purged with pure nitrogen. Catalytic ozonation experiments Non-catalytic and catalytic ozone degradation of BPA (c0 = 10.0 mg/L in ultrapure water with 18.2 MΩ cm resistance) was studied in a 1000-mL batch slurry reactor at atmospheric pressure. The reactor unit was filled with 500 mL of BPA solution for each ozonation experiment and thermostated at T = 15 °C (Julabo, model F25/ME), magnetically stirred (200 rpm), and continuously aerated with the 10 vol.% ozone-oxygen mixture. The flow rate of the ozone-oxygen mixture was 60 L/h. The catalyst powder was suspended in aqueous solution in concentration of 200 mg/L by means of ultrasonification. Before introducing an ozone-oxygen stream into the model solution, the solution was kept for 30 min under stirring at 15 °C to determine sorption equilibrium between the solid and dissolved BPA. Ozone was generated using the ozonizer A2ZS-4GLAB M (A2Z ozone), where air was used

10226

as a cooling medium. The ozone-oxygen mixture was then fed into the solution through a porous disk (sparger) located at the bottom of the reactor. The reactor was equipped with a monitoring system for instantaneous measurements of temperature and pH value. A pH range of 5.9–11.0 was used to study the reaction. The initial pH value was adjusted with 5.0 M NaOH or HCl solution using a pH meter (Metrohm, model 781) equipped with a combined pH electrode. The pH meter was calibrated using standard buffer solutions purchased from Merck before every measurement. The representative liquid-phase samples were withdrawn from the reactor suspension in pre-determined time intervals and filtered through a membrane filter (Sartorius, 0.45 μm) in order to remove catalyst particles prior to BPA content determination by means of HPLC analysis. The level of mineralization, i.e., the extent of complete removal of parent and intermediate organic substances in withdrawn aqueous-phase samples, was determined by measuring the TOC content. For determination of ozone concentration dissolved in the aqueous solution, indigo colorimetric method was applied. Chemical analysis of end-product solutions The effectiveness of catalytic ozonation was evaluated by determining temporal BPA conversions as a model pollutant, and concentrations of potential intermediates formed during the oxidative destruction using various analytical techniques (HPLC, TOC, and ion chromatography (IC)). Measurements performed with a HPLC apparatus (Thermo, model Spectra System™) were conducted in the isocratic analytical mode using a 100 × 4.6 mm BSD Hypersil C12 2.4-μm column, the flow rate of 0.5 mL/min and a methanol:ultrapure water ratio (M:W) of 70:30 (UV detection at λ = 210 nm) for measuring temporal concentration of BPA. The level of mineralization was determined by measuring the TOC content by means of an advanced TOC analyzer (Teledyne Tekmar, model Torch), which was equipped with a high-pressure NDIR detector and applying a hightemperature catalytic oxidation method carried out at 750 °C. In all analyses, three to four repeated measurements were taken for each liquid-phase sample, and the average value of TOC concentration was reported. The error of analysis was never greater than ±0.5 %. IC was employed to determine the composition of liquid-phase samples using Dionex ICS3000 apparatus and an IONPAC AG11-HC analytical column.

Results and discussion Characterization of supports and Ru/γ-Al2O3 catalyst SEM micrographs of examined supports (γ-Al2O3, θ/αAl2O3, AlO(OH), ASA, β-zeolite, and γ-Al2O3-supported


TiO2 (3.25, 7.5, and 15 wt.% of TiO2)) and Ru/γ-Al2O3 catalyst are presented in Fig. S1. γ-Al2O3 (Fig. S1a) constituents were elongated and flake-like particles 200 to 300 nm in length and mixed with substantially smaller rounded particles. The latter seems to be fully transformed in larger and high aspect ratio particles upon calcination at 1100 °C due to recrystallization and phase transformation from γ to mixed phase θ/α alumina (Fig. S1b). Boehmite exhibited fine and uniform, 10–20-nm-large particles, which agglomerated in several micron-sized rounded particles (Fig. S1c). On the contrary, ASA sample is evidently comprised of two phases. The small, few nanometer-sized particles belong to amorphous silica, while larger flakes formed during the precipitation reaction can be related to alumina (Fig. S1d). The SEM image of β-zeolite revealed the presence of almost uniformly sized particles (of about 15 nm) with similar morphology (Fig. S1e). Furthermore, these particles are congested in porous agglomerates with interparticle voids of about 100 nm. Figure S1f shows the 7.5 wt.% TiO2/γ-Al2O3 support. Lower and higher amounts of TiO2 (i.e., 3.25 and 15 wt.% of TiO2) resulted in no discernible difference in morphology compared to 7.5 wt.% of TiO2 and are not shown. However, SEM-EDX mapping of TiO2/γ-Al2O3 samples resolved the questionable existence of different amounts of TiO2 phase. Element distribution images of Ti showed even dispersions and its intensities in agreement with the loading (Fig. S2). Upon calcination of ruthenium precursor in air, Ru in the form of RuO2 particles is expected to prevail on the surface of γ-Al2O3. Unfortunately, RuO2 particles were not visible under normal electron scanning conditions (Fig. S1g), since images similar to pure γ-Al2O3 were obtained. However, the ESB detector enhanced the contrast and enabled visualization of RuO2 particles as brighter spots (Fig. S1h). Table 1 summarizes BET-specific surface area (SBET), total pore volume (Vpore), and average pore size (dpore) values of selected supports and Ru/γ-Al2O3 catalyst. The principal support, γ-Al2O3, exhibits the BET surface area of 166 m2/g, Table 1 Specific surface area (SBET), total pore volume (Vpore), and average pore width (dpore) of selected supports and 3Ru/γ-Al2O3 Sample

SBET (m2/g)

Vpore (cm3/g)

dpore (nm)

γ-Al2O3 θ/α-Al2O3 AlO(OH) ASA β-zeolite 3.25TiO2/γ-Al2O3 7.5TiO2/γ-Al2O3 15TiO2/γ-Al2O3 3Ru/γ-Al2O3

166 61 229 222 340 125 131 140 160

0.8 0.3 0.4 0.5 0.4 0.5 0.5 0.6 0.7

16.5 19.0 5.6 9.7 6.0 16.8 17.0 17.3 16.1

Environ Sci Pollut Res (2016) 23:10223–10233 12 γ-Al2O3

ASA AlO(OH)

10

β−zeolite θ/α-Al2O3

3.0 wt.% Ru/γ-Al2O3 7.5 wt.% TiO2/γ-Al2O3

8 pH (/)

which is sufficient to support and achieve high dispersion of active components. In accordance to SEM-ESB image (Fig. S1h), RuO2 particles with the average size of 15.5 nm were evenly distributed over the surface of γ-Al2O3. The BET surface area of this supported catalyst was slightly lower compared to the bare alumina carrier, due to the heat treatment at 350 °C during the synthesis. Furthermore, the TiO2/γ-Al2O3 supports were annealed at even higher temperature (i.e., 600 °C), which caused further decrease of BET-specific surface area. However, it is obvious that the higher the amount of TiO2 was, the higher BET surface area was retained. Surface coating can act as a physical inhibitor of phase transition and recrystallization of solid particles; therefore, the BET surface area of nanoparticles can be efficiently preserved (Jamnik et al. 2009). Calcination of γ-Al2O3 at 1100 °C resulted in phase transformation toward more thermodynamically stable phases (θ/α), and accompanied particle growth led to notable loss of BET-specific surface area (SBET = 61 m2/g). ASA sample is based on silica gel, which possesses BET-specific surface area higher than 300 m2/g. However, after the deposition of alumina, the BET-specific surface area of the product was determined to be 222 m2/g. Comparable value of BET-specific surface area was found with AlO(OH), while β-zeolite exhibited the highest BET-specific surface area (i.e., 340 m2/g) among tested supports. Structure identification of supports was carried out by means of X-ray powder diffraction (Fig. S3). The diffractograms of TiO2/γ-Al2O3 samples resembled the pattern of bare γ-Al2O3, and, contrary to expectations, no distinctive TiO2 (e.g., in the form of anatase) XRD peaks could be observed. At 2θ = 25.3°, where the most intensive anatase Bragg reflection (101) is positioned, only minor indication of the peak was perceivable, which indicates that TiO2 was deposited on γ-Al2O3 as amorphous phase. On the contrary, RuO2 crystallized into large enough crystals, which gave distinct XRD peaks denoted with asterisk in Fig. S3. XRD pattern of ASA indicates the precipitation of γ-Al2O3 in the matrix of amorphous silica, which is confirmed with Bragg reflections at 2θ = 37.2, 46.0, and 66.9° in the pattern of otherwise amorphous solid. The heat treatment of γ-Al2O3 at 1100 °C resulted in formation of biphasic material with the predominant θ-Al2O3 and α-Al2O3 (denoted with asterisk in Fig. S4a) in minority. Calcination at even higher temperatures (e.g., 1200 °C) would result in production of pure α-Al2O3 with considerably reduced BET-specific surface area. The XRD patterns of AlO(OH) and β-zeolite are presented in Fig. S4b, S4c. The experimental results of the pHPZC determination are illustrated in Fig. 1. These equilibrium pH values might have a crucial role in sorption processes of water-dissolved BPA and its reaction intermediates on the surface of employed solids. It is immediately apparent from the illustrated data that β-zeolite is the most acidic solid with the equilibrium pH of

10227

6 4 2 0 0.0

0.2

0.4

0.6

0.8

1.0

Mass (g)

Fig. 1 Mass titration curves (measured in 0.005 M NaCl) for different supports (γ-Al2O3, θ/α-Al2O3, AlO(OH), ASA, and β-zeolite) and γAl2O3-supported catalysts containing 7.5 wt.% of TiO2 and 3.0 wt.% of RuO2

about 3.0. On the contrary, γ-Al2O3 and AlO(OH) exhibited more neutral character with pHPZC values of 7.4, and the pH PZC value of θ/α-Al 2 O 3 sample was equal to 6.7. Apparently, both Ru and TiO2 addition lowered the pHPZC to acidic region (values of 4.7 and 4.0 were measured, respectively). Finally, the pHPZC for ASA was found to be 4.8. Determination of operating parameters in ozonation experiments In the preliminary experiments, optimal operating parameters (such as initial BPA concentration, catalyst concentration, ozone concentration, and temperature) were determined (Cotman et al. 2013). The initial BPA concentration was 10.0 mg/L, which was in accordance with relevant environmental determinations found in waste landfill leachates (Yamamoto et al. 2001). In all non-catalytic and catalytic ozonation runs performed in the present study, BPA was completely degraded in 3 min, based on the concentration of model pollutant in withdrawn liquid-phase samples being under the limit of detection of employed HPLC method, which was 0.01 mg/L. The results of BPA mineralization using ozone in the absence of a catalyst under pre-determined reaction conditions are shown in Fig. 2 (blank experiment) as a TOC removal vs. reaction time dependency. It can be seen that the maximum TOC removal of 12 % was achieved after 75 min of reaction. The reaction proceeded very fast in the first 10 min upon introduction of BPA into the reactor leading to fast consumption of aqueous ozone and also fast removal of TOC, which is in accordance with very high reaction rate constant for reaction between ozone and BPA; the latter is reported to be in the range of 1.3 × 104 and 2.7 × 106 M−1 s−1 (Esplugas et al. 2007). Fast reaction between ozone and some other aromatic

10228

Environ Sci Pollut Res (2016) 23:10223–10233 Ozonation

1.0

TOC/TOC0 (/)

0.8

0.6

0.4

7.5 wt.% TiO2/γ-Al2O3 ASA θ/α-Al2O3 AlO(OH)

0.2

γ-Al2O3 β -zeolite

Blank

0.0

-20

0

20

40

60

80

Time (min)

Fig. 2 TOC degradation profiles obtained in catalytic ozonation of BPA using different catalytic supports: γ-Al2O3 (pHfinal = 4.7), θ/α-Al2O3 (pHfinal = 4.3), AlO(OH) (pHfinal = 4.2), ASA (pHfinal = 4.9), β-zeolite (pHfinal = 3.9), and 7.5 wt.% TiO2/γ-Al2O3 (pHfinal = 4.0). The pHinitial was equal to 5.9 in all experiments

compounds has also been reported (Deborde et al. 2008; Pocostales et al. 2011). It is evident from Fig. 2 that despite very fast degradation of BPA, a considerable amount of unreacted TOC (88 %) remained in the solution at the end of reaction course. This confirms the ability of ozone to react fast with organic pollutants including double bonds, aromatic groups, and amino groups (such as BPA) but inability to react with linear compounds (intermediates; Skoumal et al. 2006). During the reaction between ozone and BPA, pH value of the solution dropped from 5.9 to 4.7 after 75 min of ozonation. This can be attributed to the formation of resistant acidic intermediates. Deborde et al. (2008) reported that muconic acid derivatives, benzoquinone, 2-(4-hydroxyphenyl)propan-2-ol, orthoquinone, catechol, and acids or aldehydes are possible by-products of BPA ozonation. The levelling off of the TOC vs. time curve following long reaction times indicates that non-catalytic ozonation only leads to incomplete mineralization of BPA, producing linear acidic intermediates which cannot be removed by ozonation.

Ozonation performed over different catalyst supports In order to determine sorption equilibrium between dissolved BPA and dispersed solids, the solution was kept for 30 min under stirring at 15 °C prior to introducing the ozone-oxygen gaseous stream into the suspension. Figure 2 compares TOC conversions as a function of time obtained in the presence of six different supports (γ-Al2O3, θ/α-Al2O3, AlO(OH), ASA, β-zeolite, and 7.5TiO2/γ-Al2O3) to non-catalytic ozonation of BPA (blank sample). The obtained results demonstrated that boehmite was the most efficient among selected solids in

catalytic ozonation experiments (51 % TOC removal). Less prominent and somehow similar removal efficiencies were observed for pure γ-Al2O3- and γ-Al2O3-supported TiO2 (7.5 wt.% of TiO2) with 40 and 42 % removal of TOC, respectively. The rest of supports, ASA, θ/α-Al2O3, and β-zeolite, were considerably less effective in terms of TOC removal, giving rise to 29, 14, and 18 % conversion, respectively. However, the major removal of BPA reaction derivatives occurred within the first 15 min, which is characteristic to all examined solids. Presumably, this indicates that the organic matter was merely accumulated on the surface of solids and negligible transformation to CO2 and H2O occurred. Indeed, CHNS elemental analysis of fresh and spent solids confirmed this assumption (see Table 2). Considering the carbon-based elemental analysis of examined catalyst supports and measured corresponding TOC removals (TOCrem), one can estimate the amounts of TOC accumulated (TOCaccu) on the catalyst surface. If TOCaccu is then subtracted from TOCrem, a true TOC mineralization (TOCmin) can be determined. In this way, TOCaccu and TOCmin values were determined for six different support samples subjected to catalytic ozonation experiments (Fig. 3). The decomposition of aqueous ozone in the presence of three aluminum oxides-based solids (AlO(OH), γ-Al2O3, and α-Al2O3) was reported by Qi et al. (2008). All three materials enhanced the rate of ozone decomposition. The greatest effect on catalyzed ozone decomposition was observed when the solution pH value was close to the pH PZC . The results of catalytic ozonation of 2,4,6trichloroanisole in the presence of aluminum oxides (AlO(OH), γ-Al2O3, and α-Al2O3) were reported by Qi et al. (2009). It was demonstrated that the efficiency of catalytic ozonation in the presence of aluminum oxides can Table 2 Carbon content (measured by means of CHNS elemental analysis) on the surface of fresh and spent catalyst samples used in the catalytic ozonation process of BPA and TOC mineralization and TOC accumulation percentages Sample

TCfresh (mg/g)

TCspent (mg/g)

TOCaccu (%)

TOCmin (%)

pHfinal (/)

γ-Al2O3 θ/α-Al2O3 AlO(OH)

0.2 0.2 0.3

1.4 0.5 1.6

30.5 6.9 34.2

12.3 6.1 17.0

4.7 4.3 4.2

ASA β-zeolite 3.25TiO2/γ-Al2O3 7.5TiO2/γ-Al2O3 15TiO2/γ-Al2O3 γ-Al2O3a 3Ru/γ-Al2O3a

0.2 0.3 0.2 0.2 0.2 0.2 0.1

0.5 1.0 0.8 0.8 0.8 1.4 1.2

8.40 17.9 14.8 14.5 14.8 30.5 26.4

20.5 0.12 16.3 27.5 27.7 12.3 56.0

4.9 3.9 3.9 4.0 4.1 4.2 5.1

Operating conditions tdark = 30 min, treac = 75 min, ccat = 200 mg/L, c(BPA)0 = 10.0 mg/L, T = 15 °C a

treac = 240 min


10229

100

Ozonation

TOCaccu TOCmin

TOC/TOC0 (/)

β -zeolite

γ -Al2O3

θ/α -Al2O3

ASA

20

0.8

AlO(OH)

40

7.5wt. TiO2/γ-Al2O3

TOC removal (%)

80

60

1.0

substantially be enhanced compared with the sole ozonation. The catalytic activity of the aluminum oxides was related to highly hydroxylated surface. The surface hydroxyl groups on the aluminum oxides were active sites in catalytic ozonation reaction. The most promising result regarding the TOC mineralization was achieved using the 7.5 wt.% TiO2 supported on γalumina. The addition of titania caused radical turn in TOCaccu to TOCmin ratio (14.5/27.5) compared to the pure γ-Al2O3 sample, which was more dominant in accumulation (30.5/ 12.3). Besides γ-Al2O3-supported TiO2, suppressed TOCaccu and pronounced TOCmin (8.4/20.5) were also observed in the presence of ASA sample. The TOCaccu to TOCmin ratio of θ/ α-Al2O3 material approached to unity, showing no tendency to either side. On the contrary, similar behavior to pure γAl 2O 3 (TOC min is overwhelmed by TOC accu) was also witnessed with AlO(OH) and β-zeolite samples with their TOC accu to TOC min ratios of 34.2/17.0 and 17.9/0.12, respectively.

0.4 3.25 wt.% TiO2/γ-Al2O3 7.5 wt.% TiO2/γ-Al2O3 15 wt.% TiO2/γ-Al2O3

0.2

γ-Al2O3

Blank

0

Fig. 3 TOCmin and TOCaccu values for particular catalyst supports (γAl2O3 (pHfinal = 4.7), θ/α-Al2O3 (pHfinal = 4.3), AlO(OH) (pHfinal = 4.2), ASA (pHfinal = 4.9), β-zeolite (pHfinal = 3.9), and 7.5 wt.% TiO2/γ-Al2O3 (pHfinal = 4.0)) determined after 75 min of catalytic ozonation of aqueous BPA solution. The pHinitial was equal to 5.9 in all runs

0.6

0.0

-20

0

20 40 Time (min)

comparing adsorbed and mineralized TOC fractions (Fig. 5). It can be seen that the accumulation of BPA reaction intermediates is already hindered with the lowest amount of TiO2 (i.e., 3.25 wt.%). Increasing the amount of TiO2 exhibited no influence on further minimization of carbonaceous deposits formed but provoked the transformation of organic matter into CO2 and H2O. The highest loading of TiO2 (i.e., 15 wt.%) resulted in comparable TOCmin (about 27 %) to the amount obtained with 7.5 wt.% of TiO2. Accordingly, the medium amount of TiO2 represented the optimum loading of TiO2 on γ-Al2O3 for catalyst support in catalytic ozonation process. Results of removal of oxalic acid from water in the presence of ozone and a TiO2/Al2O3 catalyst at pH = 2.5 were reported by Beltrán et al. (2004). The reaction led to total mineralization of oxalic acid according to the TOC analysis. The process 100 TOCaccu TOCmin

TOC removal (%)

15 wt.% TiO2/γ-Al2O3

20

7.5 wt.% TiO2/γ-Al2O3

40

3.25 wt.% TiO2/γ-Al2O3

60 γ-Al2O3

Contrary to pure γ-Al2O3, TiO2/γ-Al2O3 catalyst samples resulted in more progressive removal of TOC. Above all, this can be attributed to catalytic ozonation reaction, which dominated over adsorption of dissolved organic compounds on the catalyst surface. Various amounts of TiO2 (3.25, 7.5, and 15 wt.% of TiO2) on the surface of γ-Al2O3 were examined in catalytic ozonation process. Figure 4 demonstrates that in the presence of 7.5 and 15 wt.% of TiO2 deposited on γAl2O3, TOC can be removed to the same extent as achieved with pure γ-Al2O3 (i.e., around 42 %). However, the benefit of using combined TiO2/γ-Al2O3 support is revealed when

80

Fig. 4 TOC degradation profiles obtained in catalytic ozonation of BPA using γ-Al2O3 and γ-Al2O3-supported TiO2 solids with different titania loadings: 3.25 (pHfinal = 3.9), 7.5 (pHfinal = 4.0), and 15 (pHfinal = 4.1) wt.% of TiO2. The pHinitial was equal to 5.9 in all runs

80

Ozonation over TiO2/γ-Al2O3 catalysts

60

0

Fig. 5 TOCmin and TOCaccu values obtained in the presence of γ-Al2O3supported TiO2 solids with different titania loadings (3.25 (pHfinal = 3.9), 7.5 (pHfinal = 4.0), and 15 (pHfinal = 4.1) wt.% of TiO2) in comparison with pure γ-Al2O3. The pHinitial was equal to 5.9 in all runs

10230


Influence of pH value and different supports on catalytic ozonation It is widely recognized that pH value of aqueous solution affects the efficiency of catalytic ozonation process to a great extent. Namely, pH value influences not only ozone decomposition reaction in aqueous solution but also the surface properties of catalysts. At basic pH (pH > pHPZC), the surface of alumina is negatively charged; at acidic pH (pH < pHPZC), its surface is positively charged, while at pH = pHPZC, the surface of alumina is neutral (surface hydroxyl groups with no charge; Ikhalaq et al. 2012). Figure 6 compares temporal TOC degradation profiles in catalytic ozonation experiments obtained over pure γ-Al2O3 at different pH values (5.9, 8.0, and 11.0). pH = 5.9 was established at the beginning of ozonation process over γ-Al2O3; however, this value was gradually lowered to pH = 4.7 at the end of reaction run, which is due to the formation of aliphatic acids (formic, acetic, oxalic, and malonic acids). pHPZC of γ-Al2O3 was 7.4, which resulted in negatively charged surface at initial pH values equal to 8 and 11, respectively. In accordance, the sorption process of dissolved organic compounds was affected, leading to lower TOC accu values (Fig. 7). The experiment conducted at pH = 8.0 enabled 68 % removal of TOC, out of which 42 % could be attributed to TOC min . Comparing to the run Ozonation 1.0

TOC/TOC0 (/)

0.8

0.6

0.4 pH=11.0 pH=8.0 pH=5.9 Blank

0.2

0.0

-20

0

20

40

60

80

Time (min)

Fig. 6 Influence of pH value on TOC degradation profiles in catalytic ozonation of BPA using γ-Al2O3. The pHinitial = 11.0 changed to pH final = 10.1. The pH initial = 8.0 changed to pH final = 5.9. The pHinitial = 5.9 changed to pHfinal = 4.7

100 TOCaccu TOCmin

80 TOC removal (%)

likely involves adsorption of oxalic acid on two different active sites, alumina active sites and titania active sites. Final step of the mechanism is the surface reaction between oxalic acid adsorbed on titania active sites and ozone remaining nonadsorbed in solution.

60

40

20

0

pH=11.0

pH=8.0

pH=5.9

Fig. 7 TOCmin and TOCaccu values obtained in the presence of γ-Al2O3 in catalytic ozonation experiments performed at different pH values. The pHinitial = 11.0 changed to pHfinal = 10.1. The pHinitial = 8.0 changed to pHfinal = 5.9. The pHinitial = 5.9 changed to pHfinal = 4.7

performed at uncontrolled initial pH value, 42 % of TOC was removed in both TOC accu and TOC min processes. Contrary to these were results of the run conducted at pH = 11.0. At these strongly basic conditions in the model aqueous solution, only 18 % of TOC was removed, which was comparable to values obtained under non-catalytic ozonation conditions. At pH = 10 and above, the concentration of molecular ozone is low, due to increased decomposition rate. Hence, catalysts are less effective compared to other pH values (Roshani et al. 2014). Observations of Ernst et al. (2004), who conducted batch and semicontinuous experiments, confirm that γ-Al2O3 can be an efficient heterogeneous catalyst in ozonation of succinic acid. The effectiveness and the quantity of direct carbon removal depend not only on the properties of the applied alumina but also on chemical properties and structure of the dissolved organic molecules as well as on the matrix of the solution. For succinic acid, the catalytic effect was superior in a batch test, in which 80 % of TOC was removed. The enhancement of catalytic activity of γ-Al2O3 at pH = 8.0 was also verified with ASA, which showed high potential for mineralization of BPA. The activity of both supports (ASA and γ-Al2O3) was compared at pH values equal to 5.9 and 8.0 (Fig. 8). Interestingly, the TOC degradation profiles of ASA at these pH conditions were very similar, indicating that more basic conditions were not beneficial in the presence of this sample. Accordingly, the sums of TOCaccu and TOCmin at both pH conditions were alike as well (28.9 and 28.4 % at pH values of 5.9 and 8.0, respectively; see Fig. 9). However, the process of TOC accumulation was more pronounced at higher pH values, which was right the opposite to the sorption behavior observed in the presence of γ-Al2O3 sample. The removal and mineralization of TOC using γAl2O3 were more successful at basic pH conditions.


10231 Ozonation

1.0

1.0

0.8

0.8

TOC/TOC0 (/)

TOC/TOC0 (/)

Ozonation

0.6

0.4 ASA, pH=5.9 ASA, pH=8.0 γ-Al2O3, pH=5.9

0.2

0.6

0.4 3.0 wt.% Ru/γ-Al2O3, pH=8.0

0.2

3.0 wt.% Ru/γ-Al2O3, pH=5.9 γ-Al2O3, pH=5.9

γ-Al2O3, pH=8.0

0.0

γ-Al2O3, pH=8.0

Blank

-20

0

20

40

60

0.0 -30

80

Blank

0

30

60

90

Time (min)

Fig. 8 TOC degradation profiles obtained in the presence of γ-Al2O3 and ASA solids at pHinitial = 5.9 (changed to pHfinal = 4.7 for γ-Al2O3 and pHfinal = 4.9 for ASA) and pHinitial = 8.0 (changed to pHfinal = 5.9 for γAl2O3 and pHfinal = 6.2 for ASA) during catalytic ozonation of BPA

Ozonation experiments over Ru/γ-Al2O3 catalyst Ruthenium was selected as the active metal since it is known to be strongly resistant to corrosion from both acidic and basic solutions and, generally, exhibits high activities in various AOPs. The catalyst was examined in prolonged 240-min ozonation runs conducted at both pH = 5.9 and pH = 8.0 (Fig. 10). At pH = 5.9, the Ru/γAl2O3 solid showed meaningful increase in TOC degradation compared to the pure γ-Al2O3 sample. The Ru/γ-Al2O3 catalyst provoked progressive removal of TOC, which reached steady state after 3 h of operation. On the contrary to γAl2O3, the increase in pH value had no perceivable influence on TOC degradation. The most prominent difference in using Ru/γ-Al2O3 catalyst is its mineralization potency. In the

240

TOC removal (%)

3.0 wt.% Ru/γ-Al2O3, pH=8.0

40

γ -Al2O3, pH=8.0

60

TOCmin

γ -Al2O3, pH=5.9

ASA, pH=8.0

ASA, pH=5.9

γ -Al2O3, pH=8.0

0

Fig. 9 TOCmin and TOCaccu values obtained in the presence of ASA and γ-Al2O3 solids in catalytic ozonation experiments performed at two different pH values

80

3.0 wt.% Ru/γ-Al2O3, pH=5.9

TOCaccu

TOCmin

γ -Al2O3, pH=5.9

TOC removal (%)

210

100

80

20

180

presence of this catalyst, it was possible to remove over 82 % of TOC and 56 % can be attributed to TOC min (Fig. 11). Slightly lower TOC removal (79 %) was achieved at pH = 8.0, mineralizing 51 % of TOC. Nevertheless, the TOCaccu value remained almost unchanged and comparable to the values determined in the presence of γ-Al2O3, which clearly indicated that this contribution to TOC removal is strongly related to the surface properties of the support. These experiments demonstrated that the activity of 3 wt.% Ru/γ-Al2O3 catalyst is stable at different pH conditions, and high degree of organic matter transformation to ultimate oxidation products can be efficiently achieved in catalytic ozonation process.

TOCaccu

40

150

Fig. 10 TOC degradation profiles obtained in the presence of γ-Al2O3 and γ-Al2O3-supported 3 wt.% Ru catalyst at pHinitial = 5.9 (changed to pHfinal = 4.2 for γ-Al2O3 and pHfinal = 5.1 for γ-Al2O3-supported 3 wt.% Ru catalyst) and pHinitial = 8.0 (changed to pHfinal = 5.3 for γ-Al2O3 and pHfinal = 5.8 for γ-Al2O3-supported 3 wt.% Ru catalyst) during catalytic ozonation of BPA

100

60

120

Time (min)

20

0

Fig. 11 TOCmin and TOCaccu values obtained in the presence of γAl 2 O 3 - and γ-Al 2 O 3 -supported 3 wt.% Ru catalysts in catalytic ozonation runs carried out at different pH values

10232

Catalyst stability and reusability The essential factors that have to be taken into account before a full-scale implementation of heterogeneous catalytic processes in an industrial application are stability and reusability of the catalyst. Weak interactions between the catalyst support and metal clusters could result in leaching of active metal ingredient into the reaction solution. Loss of active sites on the catalyst surface will in turn reduce the removal efficiency of organic pollutants. In addition to the catalyst deactivation due to metal leaching into the liquid phase, the presence of metals in water can become a source of secondary pollution and health risk. In order to quantify the extent of metal leaching from the catalyst into the reaction suspension, metal concentration in water was determined by means of ICP-MS analysis after 240 min of ozonation performed with 200 mg/L of examined 3 wt.% Ru/γ-Al2O3 catalyst at pHinitial = 5.9. The amount of leached Ru was equal to 55 μg/L, which corresponded to 0.21 % of noble metal content deposited on the alumina support. An ideal catalyst should be easily recoverable from the reaction mixture without any metal leaching occurrence. Further, it should be reusable without a significant loss of catalytic activity. In order to study the performance of the Ru-based catalyst, the reusability tests were carried out by using the same catalyst batch in three consecutive ozonation runs. As illustrated in Fig. 12, the reusability of 3 wt.% Ru/γ-Al2O3 catalyst sample indicates that there is no significant decrease of catalyst activity after three reaction runs. However, in the last ozonation run, a slight decrease in TOC removal was observed. This may be either due to the blockage of catalyst active sites by products of BPA oxidation or catalyst leaching effect. Since the abovementioned extent of Ru leaching into the aqueous solution was very low, the blockage of active sites seems to be more probable. However, further investigation is required to understand the role of adsorption, surface properties of the catalysts used, and the effect of pH value of liquid phase on the efficiency of catalytic ozonation.

Ozonation 1.0

0.8 TOC/TOC0 (/)

Accordingly to Nawrocki (2013), the mechanism of catalytic ozonation carried out in the presence of a noble metal catalyst is based on the formation of very reactive complex between a metal and ozone, followed by the reaction of the complex with organic molecule in the solution. Ru/Al2O3 as catalyst and DMP as the model pollutant in the catalytic ozonation process were systematically investigated by Yunrui et al. (2007). Although the degradation rates of DMP were almost the same in the direct ozonation, the TOC removal efficiency of catalytic ozonation was much higher than that of ozonation. The catalyst increased the TOC removal to 72 % after 120 min, compared to only 15 % for direct ozonation.


Run #1 Run #2 Run #3 Blank

0.6

0.4

0.2

0.0 -30

0

30

60

90

120

150

180

210

240

Time (min)

Fig. 12 TOC degradation profiles obtained in catalytic ozonation of BPA (pHinitial = 5.9) in the presence of γ-Al2O3-supported 3 wt.% Ru catalyst used in three consecutive runs

Conclusions To sum up, supports (γ-Al 2 O 3 , θ/α-Al 2 O 3 , AlO(OH), ASA, and β-zeolite) and γ-Al2O3-supported TiO2 (3.25, 7.5, and 15 wt.% of TiO2) solids with different physicochemical properties were used in catalytic ozonation experiments to mineralize BPA dissolved in water. The most promising catalyst carrier and reaction conditions were determined in order to obtain pronounced mineralization of BPA and reaction intermediates. Fresh and spent solids were thoroughly examined in terms of TOC accumulation, since it was revealed that substantial amounts of carbonaceous deposits could be formed and deposited on the catalyst surface during the reaction course. Non-catalytic ozonation resulted in complete degradation of aqueous BPA in less than 3 min; however, very low extent of TOC removal (12 %) was observed. The extent of BPA mineralization in the presence of γ-Al2O3 is strongly pH dependent. The most pronounced TOC mineralization was achieved in the presence of γ-Al2O3 at pH = 8.0; in this manner, 68 % of TOC was removed, out of which 42 % could be attributed to TOCmin. Other alumina polymorphs and silica-based supports (ASA and β-zeolite) exhibited lower BPA mineralization activity. Furthermore, a benefit of using a combined 7.5 % TiO2/γ-Al2O3 support was revealed in terms of more progressive mineralization of parent organic matter, compared to pure γ-Al2O3. The use of 3.0 wt.% Ru/γ-Al2O3 as catalyst at different pH values showed promising results regarding both the kinetics of TOC removal as well as extent of mineralization of BPA and reaction intermediates. In the presence of this catalyst tested at pH = 5.9, it was possible to remove over 82 % of TOC, of which 56 % was mineralized to H2O and CO2.

Environ Sci Pollut Res (2016) 23:10223–10233 Acknowledgments The authors gratefully acknowledge the financial support of the Ministry of Education, Science, and Sport of the Republic of Slovenia through Research Program No. P2-0150.

References Basile T, Petrella A, Petrella M, Boghetich G, Petruzzelli V, Colasuonno S, Petruzzelli D (2011) Review of endocrine-disrupting-compounds removal technologies in water and wastewater treatment plants: an EU perspective. Ind Eng Chem Res 50:8389–8401 Belgiorno V, Rizzo L, Fatta D, Dellarocca C, Lofrano G, Nikolaou A, Naddeo V, Meric S (2007) Review on endocrine disruptingemerging compounds in urban wastewater: occurrence and removal by photocatalysis and ultrasonic irradiation for waste water reuse. Desal 215:166–176 Beltrán FJ, Rivas FJ, Mo-de-Espinosa RA (2004) TiO2/Al2O3 catalyst to improve the ozonation of oxalic acid in water. Appl Catal B Env 47: 101–109 Chen YH, Hsieh DC, Shang NC (2011) Efficient mineralization of dimethyl phthalate by catalytic ozonation using catalyst TiO2/Al2O3. J Haz Mat 192:1071–1025 Cotman M, Erjavec B, Tišler T, Pintar A (2013) Use of non-catalytic and catalytic ozonation for mineralization and toxicity removal of bisphenol A. In: EAAOP-3, Almería, 1–3. Deborde M, Rabouan S, Mazellier P, Duguet JP, Legube B (2008) Oxidation of bisphenol A by ozone in aqueous solution review paper. Water Res 42:4299–4308 Djinović P, Osojnik Črnivec IG, Batista J, Levec J, Pintar A (2011) Catalytic syngas production from greenhouse gasses: performance comparison of Ru-Al2O3 and Rh-CeO2 catalysts. Chem Eng Proc 50:1054–1062 Erjavec B, Tišler T, Kaplan R, Pintar A (2013) Titanate nanotubes as a novel catalyst for removal of toxicity and estrogenicity of bisphenol A in the CWAO process. Ind Eng Chem Res 52:12559–12566 Ernst M, Lurot F, Schrotter JC (2004) Catalytic ozonation of refractory organic model compounds in aqueous solution by aluminum oxide. Appl Catal B Env 47:15–25 Esplugas S, Bila DM, Gustavo L, Krause T, Dezotti M (2007) Ozonation and advanced oxidation technologies to remove endocrine disrupting chemicals (EDCs) and pharmaceuticals and personal care products (PPCPs) in water effluents. J Haz Mat 149:631–642 Garoma T, Matsumoto S (2009) Ozonation of aqueous solution containing bisphenol A: effect of operational parameters. J Haz Mat 167: 1185–1191 Hensen EJM, Poduval DG, Magusin PC, Coumans AE, van Veen JAR (2010) Formation of acid sites in amorphous silica-alumina. J Catal 269:201–218 Ikhlaq A, Brown DR, Kasprzyk-Hordern B (2012) Mechanisms of catalytic ozonation on alumina and zeolites in water: formation of hydroxyl radicals. Appl Catal B Env 123–124:94–106 Jamnik J, Dominko R, Erjavec B, Remškar M, Pintar A, Gaberšček M (2009) Stabilizers of particle size and morphology: a road towards high-rate performance insertion materials. Adv Mat 13:2715–2719 Kasprzyk-Hordern B, Ziółek M, Nawrocki J (2003) Catalytic ozonation and methods of enhancing molecular ozone reactions in water treatment. Appl Cat B Env 46:639–669

10233 Keykavoos R, Mankidy H, Mab P, Jones B, Soltan J (2013) Mineralization of bisphenol A by catalytic ozonation over alumina. Sep Purif Tech 107:310–317 Li C, Li XZ, Graham N, Gao NY (2008) The aqueous degradation of bisphenol A and steroid estrogens by ferrate. Water Res 42:109–120 Li G, Zu L, Wong PK, Hui X, Lu Y, Xiong J, An T (2012) Biodegradation and detoxification of bisphenol A with one newly isolated strain Bacillus sp. GZB: kinetics, mechanism and estrogenic transition. Bior Tech 114:224–230 Lin K, Liu W, Gan J (2009) Oxidative removal of bisphenol A by manganese dioxide: efficacy, products, and pathways. Env Sci Tech 43: 3860–3864 Melcer H, Klecka G (2011) Treatment of wastewaters containing bisphenol A: state of the science review. Water Env Res 83:650–666 Mezohegyi G, Erjavec B, Kaplan R, Pintar A (2013) Removal of bisphenol A and its oxidation products from aqueous solutions by sequential catalytic wet air oxidation and biodegradation. Ind Eng Chem Res 52:9301–9307 Nawrocki J (2013) Catalytic ozonation in water: controversies and questions. Appl Catal B Env 142–143:465–471 Nawrocki J, Kasprzyk-Hordern B (2010) The efficiency and mechanisms of catalytic ozonation. Appl Catal B Env 99:27–42 Pan B, Lin D, Mashayekhi H, Xing B (2008) Adsorption and hysteresis of bisphenol A and 17α-ethinyl estradiol on carbon nanomaterials. Env Sci Tech 42:5480–5485 Pocostales P, Álvarez P, Beltrán FJ (2011) Catalytic ozonation promoted by alumina-based catalysts for the removal of some pharmaceutical compounds from water. Chem Eng J 168:1289–1295 Qi F, Chen Z, Xu B, Shen J, Maa J, Joll C, Heitz A (2008) Influence of surface texture and acid–base properties on ozone decomposition catalyzed by aluminum (hydroxyl) oxides. Appl Catal B Env 84: 684–690 Qi F, Xu B, Chen Z, Ma J, Suna D, Zhanga L (2009) Influence of aluminum oxides surface properties on catalyzed ozonation of 2,4,6trichloroanisole. Sep Purif Tech 66:405–410 Rosal R, Mía S, Gonzalo AR, García-Calvo E (2009) Ozonation of clofibric acid catalyzed by titanium dioxide. J Haz Mat 169:411–418 Roshani B, McMaster I, Rezaei E, Soltan J (2014) Catalytic ozonation of benzotriazole over alumina supported transition metal oxide catalysts in water. Sep Purif Tech 135:158–164 Skoumal M, Cabot PL, Centellas F, Arias C, Rodriguez RM, Garrido JA, Brillas E (2006) Mineralization of paracetamol by ozonation catalyzed with Fe2+, Cu2+ and UVA light. Appl Catal B Env 66:228–240 Torres-Palma RA, Nieto JI, Combet E, Pétrier C, Pulgarin C (2010) An innovative ultrasound, Fe2+ and TiO2 photo assisted process for bisphenol A mineralization. Water Res 44:2245–2252 Watanabe N, Horikoshi S, Kawabe H, Sugie Y, Zhao J, Hidaka H (2003) Photodegradation mechanism for bisphenol A at the TiO2/H2O interfaces. Chemosphere 52:851–859 Xu P, Janex ML, Savoye P, Cockx A, Lazarova V (2002) Wastewater disinfection by ozone: main parameters for process design. Water Res 36:1043–1055 Yamamoto T, Yasuhara A, Shiraishi H, Nakasugi O (2001) Bisphenol A in hazardous waste landfill leachates. Chemosphere 42:415–421 Yang Y, Ma J, Qin Q, Zhai X (2007) Degradation of nitrobenzene by nano-TiO2 catalyzed ozonation. J Mol Cat A Chem 267:41–48 Yunrui Z, Wanpeng Z, Fudong L, Jianbing W, Shaoxia Y (2007) Catalytic activity of Ru/Al2O3 for ozonation of dimethyl phthalate in aqueous solution. Chemosphere 66:145–150