1 Assessing the effect of inorganic anions on TiO2-photocatalysis and ozone oxidation treatment 1 efficiencies 2 3 Helen Barndõk, Daphne Hermosilla*, Luis Cortijo, Carlos Negro, and Ángeles Blanco 4 5 Department of Chemical Engineering, Complutense University of Madrid, Avda. Complutense s/n, 28040 6 Madrid (Spain). E-mail addresses (in order of appearance): hbarndok@estumail.ucm.es, 7 dhermosilla@quim.ucm.es, lcortijo@estumail.ucm.es, cnegro@quim.ucm.es, ablanco@quim.ucm.es 8 9 *Corresponding author; Telephone: +34 91 394 4245; Fax: +34 91 394 4243; E-mail address: 10 dhermosilla@quim.ucm.es 11 12 2 Abstract: Considering the application of AOPs might be limited for the treatment of industrial wastewater 1 with high inorganic load, and that partial results reported to date regarding this particular are 2 inconclusive, even opposite in some cases, the effect of inorganic anions on the oxidation efficiency of 3 photocatalysis and ozonation has been further assessed with statistical significance. While the presence of 4 sulphate and chloride did not appreciably affect the photocatalytic oxidation of phenol, nitrate 5 significantly enhanced the removal of COD (≈ 8-15%). The addition of carbonate simply increased the 6 pH, which strongly inhibited the photocatalytic process; whereas if pH=5 was kept constant, the reduction 7 of the COD was not affected by the presence of carbonate. On the other hand, sulphate, chloride and 8 nitrate did not significantly affect the degradation of phenol by ozonation; whereas the presence of 9 carbonate apparently enhanced the reduction of COD. It is actually proved that this improvement in the 10 efficiency of the treatment was produced by the pH buffering effect of these ions, rather than to its 11 presence itself, which actually significantly reduced the removal of COD (5-10%) by radical scavenging 12 action in comparison to when the treatment was performed in the absence of anions in the solution 13 adjusting the pH to similar basic values (≈9.5-13.5). When ozonation was performed at a pH close to 14 neutral (6.5 ± 0.2) or basic (12 ± 0.2), at which the indirect oxidation of hydroxyl radical is surely widely 15 active, the results were significantly enhanced in any case (COD removal ≈ 70-75%), whether in the 16 absence or the presence of these anions; despite the significant slight radical scavenging effect (COD 17 removal ≈ 65-70%) that was attributed to the addition of carbonate. 18 19 Key-words: photocatalysis; ozonation; radical scavengers; carbonate; sulphate; nitrate; chloride 20 21 3 Introduction 1 Advanced oxidation processes (AOPs) involving in situ generation of highly reactive transitory 2 species like H2O2, OH·, O3, O2 −· are applied when conventional wastewater treatment techniques become 3 insufficient to treat persistent contaminants (1-3). Particularly, ozonation and photocatalytic oxidation 4 with semiconductor catalysts, typically TiO2, have been widely assessed and applied at industrial scale for 5 this purpose (4-7). 6 Inorganic anions have been identified to reduce the efficiency of different AOPs (8-10), but actual 7 comparison of their effect at different ion concentration values has been poorly assessed. In general, the 8 produced loss of treatment efficiency attributed to the presence of inorganic anions has been explained by 9 the scavenging effect of reactive radical ionic species and the generation of other by-radicals with weaker 10 oxidation potentials (11-14). For example, chloride (Cl−) has been addressed to scavenge hydroxyl 11 radicals producing hydroxide and chlorine radical (15-17). Moreover, Cl− may generate additional “hole”-12 scavenging in photocatalytic processes. 13 Similar inhibition mechanisms have been also attributed to other anionic species with a 14 significant presence in wastewater, like sulphate (SO4 2−) and nitrate (NO3 −) (16-18); although other 15 authors have addressed an insignificant effect of Cl−, SO4 2−, and NO3 − on the efficiency of TiO2-16 photocatalysis (19, 20). In addition, several authors have reported an improved oxidation efficiency of 17 several AOPs caused by the strong oxidation potential of SO4·− radical itself (10, 21-24). 18 Besides radical scavenging, some anions (e.g. NO3 −, SO4 2−) may raise the turbidity of the 19 solution, which causes the screening of UV radiation when applying photocatalytic treatments (10, 22); 20 and the competitive adsorption of inorganic anions has been also proposed as an additional potential 21 mechanism of inhibition, “stealing” surface active sites from organic molecules (10, 16, 25). 22 Nevertheless, Wang et al. (26) observed no significant relationship between adsorption inhibition and 23 photodegradation rate, attributing all the observed oxidation loss to the radical scavenging action exerted 24 by NO3 −, Cl− and SO4 2−. In addition, the competitive adsorption of anions is unlikely to occur under basic 25 conditions because the amphoteric nature of TiO2 will lead to the repulsion of negative electrostatic forces 26 (22, 27, 28). Particularly, Epling and Lin (29) observed that photo-bleaching of anionic dyes was partially 27 inhibited in the presence of HNO3 and NaHCO3, while it was accelerated for cationic ones. 28 Finally, most research assessing the effect of the presence of inorganic anions on the performance 29 of ozone oxidation has been devoted to bicarbonate (HCO3 −) and carbonate (CO3 3−) action scavenging 30 OH· radical (30-33): 31 CO3 2− + OH·  CO3 3− + OH− (1) 32 HCO3 − + OH·  HCO3· + OH− (2) 33 HCO3 − + OH·  CO3·− + H2O (3) 34 In fact, Chiang et al. (33) used HCO3 − and CO3 2− to scavenge OH· radical in order to study direct 35 oxidation by molecular ozone; and Song et al. (35) justified the reduced mineralization of CI Reactive 36 Yellow 145 by ozone oxidation at pH>11.0 by the radical scavenging effect of HCO3 −/CO3 2− ions. On the 37 other hand, Lair et al. (36) found that the presence of HCO3 −/CO3 2− inhibited the degradation of 38 naphthalene by photocatalysis; but HCO3 − has been reported to produce a negligible effect on the 39 photocatalytic oxidation of Acid Orange 7 (26) and TNT (37). 40 In short, several authors have previously assessed in part the effect of inorganic anions on the 41 degradation efficiency of several AOPs, although the reported results are inconsistent and have not been 42 analyzed in terms of statistical significance. Therefore, the main objective of this essay aims to further 43 assess the significance of the effect the most common anions present in industrial wastewater (SO4 2−, Cl−, 44 NO3 − and CO3 2−) have on the treatment efficiency of phenol by ozonation and TiO2-photocatalysis. These 45 treatments were selected because they are AOPs widely applied with success in several industrial 46 applications (4-7); while phenol was chosen because it is a commonly used model compound for the 47 assessment of AOPs, and its nature and behaviour have been widely described when it is treated by 48 ozonation or TiO2-photocatalysis (1, 22, 38-43). 49 50 4 Experimental and methods 1 Materials 2 All used chemicals were analytical grade, provided by PANREAC S.A. (Barcelona, Spain). 3 Phenol was available in its purest form (99.99%) and diluted with ultra-pure deionized water to the 4 concentration of 200 mg·L−1 prior to experiment performance (initial COD=480 mg·L−1). One of the 5 inorganic anions selected to perform this essay (SO4 2−, Cl−, NO3 − or CO3 2−) was then added to the solution 6 as the corresponding sodium salt at one of the following concentrations: 250, 500, 1000, and 2000 7 mg·L−1. Every experiment was repeated 3-6 times to minimize the standard deviation of the results. 8 9 UV/TiO2 oxidation 10 Aeroxide® TiO2 P25 (Evonik, Essen, Germany) photocatalyst, with a specific surface area of 50 11 ± 15 m2·g−1 and an average primary particle size of 21 nm, was first added to the solution at the 12 concentration of 5 g·L−1. This optimum TiO2 dosage resulted from preliminary trials performed using 13 catalyst concentration values ranging from 1 g·L-1 to 10 g·L-1; and has been previously reported to 14 produce optimal results in the treatment of industrial wastewater (44-46). 15 The oxidation treatment was carried out in a magnetically-stirred and specially shaped glass 16 reactor (3 L) provided with the necessary ports in the upper part to insert pH and redox potential probes, 17 which furthermore made possible the withdrawal of 5-ml aliquots from the reacting suspension for 18 monitoring the evolution of the COD. The source of UV irradiation was a vertically located medium-19 pressure mercury vapour lamp (450 W; model 7825-34, ACE Glass Inc., Vineland, NJ, USA), inserted in 20 a quartz cooling jacket, which radiates a total light power of 175.8 W covering from the infrared to the far 21 ultraviolet wavelength regions. The whole assembly was placed in a photochemical safety cabin assisted 22 with a cooling system. The treated solution was 2 L, and every experimental run was conducted for 2 23 hours. The photocatalyst was immediately separated from every collected sample by means of a 0.45 μm-24 pore filter at 15, 30, 45, 60, 90, and 120 min from the beginning of the treatment. 25 In the absence of inorganic anions, and when SO4 2−, Cl− and NO3 − ions were added to the 26 solution, the initial pH after the addition of the catalyst was approximately 5.0, and decreased thereafter to 27 approximately 4.0 along the initial 30 min of reaction; keeping this value constant until the end of the 28 experiment. The addition of Na2CO3 caused the initial pH to rise to about 10.0 to 12.0, depending on the 29 added salt concentration, and was kept buffered by the presence of HCO3 −/CO3 2− ions. Therefore, 30 additional experiments were carried out for testing the effect of pH alone: a) continuously adjusting pH to 31 5.0 (± 0.2) and 4.0 (± 0.2) after the addition of Na2CO3; and b) keeping pH=12.0 (± 0.2) when treating 32 phenol in the absence of anions. pH was adjusted to acid values adding 1M H2SO4, as the effect of SO4 2- 33 was first checked to be non significative on the performance of the treatment; whereas 1N NaOH was 34 added to perform the experiments keeping basic pH values. 35 36 Ozonation 37 Ozonation experiments were conducted in a glass jacketed cylindrical bubble reactor (height=1 38 m, diameter=5 cm) with a continuous feed of ozone gas (4.0 L·min−1) produced from ordinary grade air 39 passed through polycarbonate filters, and subsequently enriched with oxygen. The system consisted of an 40 ozone generator (Model 6020, Rilize, Gijón, Spain), a flow controller Bronkhorst® (Model F-201AV, 41 Ruurlo, The Netherlands), and an ozone on-line analyzer (Model 964C, BMT Messtechnik GMBH, 42 Berlin, Germany). Ozone consumption was pH dependent, but did not resulted significantly different 43 among the tested ions presence, resulting 0.39 ± 0.05 g/L when the initial pH value of the solution was 44 not adjusted (pH=5.7); 0.48 ± 0.06 g/L when the pH was maintained at 6.5; and 0.67 ± 0.07 g/L for a 45 constant pH = 12.0. Unconsumed ozone was sent to a catalytic ozone destructor. 46 A peristaltic pump (Masterflex® Console Drive, Cole-Parmer Instrument Company, Illinois, 47 USA) was used to recirculate the solution under treatment (1 L) through the reactor, and probes for pH 48 and redox potential and dissolved oxygen (ProODO, YSI Inc., Ohio, USA) measurement. Temperature 49 was kept at 25ºC using a thermostatic bath (Model FL300, JULABO Labortechnik GmbH, Seelbach, 50 Germany), which was aided by the reactor glass jacket itself. 51 5 Every experiment was performed for 30 min. The initial pH value of the prepared phenol solution 1 (200 mg·L−1) averaged 5.7 (±0.2), and was not modified by the addition of SO4 2−, Cl− or NO3 −. During the 2 first 5 min of ozone treatment, pH decreased to approximately 3.2, and was kept more or less constant 3 thereafter. On the other hand, the addition of Na2CO3 caused pH to rise to about 12 to 13 (depending on 4 the added concentration), and was kept buffered by the presence of HCO3 −/CO3 2− ions, as it was 5 previously described when performing photocatalytic trials. When the initial pH was adjusted to 5.7 after 6 the addition of Na2CO3, it resulted to rise to a 6.5 constant value after 5 min. In order to set proper 7 comparisons, additional experiments were therefore carried out for every considered ion keeping the pH 8 value constant at 6.5 and 12.0. 1M H2SO4 was added to acidify the solution in the experiments performed 9 in the presence of HCO3 −/CO3 2−; whereas 1N NaOH was used for increasing and keeping constant the pH 10 in the experiments developed at pH=6.5 and 12 when no ions or SO4 2−, Cl− and NO3 − were added. 11 12 Analytical methods 13 On-line pH, redox potential, and dissolved oxygen measurements were carried out every minute. 14 The degradation of phenol was assessed as the achieved reduction in chemical oxygen demand (COD), 15 which was measured according to the Standard Methods for the Examination of Water and Wastewater 16 (APHA, 2005) by the colorimetric method at 600 nm using an Aquamate-spectrophotometer (Thermos 17 Scientific AQA 091801, Waltham, USA). 18 19 Statistical analysis 20 One-way ANOVA was run (Statplus, 2009) to determine the significance of the observed 21 differences among experiments. Post hoc all pairwise comparisons were performed using Tukey's test (P 22 < 0.05). Linear regression was used to explain the strong relationship between some treatment variables. 23 24 Results and discussion 25 UV/TiO2 oxidation 26 The UV/TiO2 oxidation treatment of phenol achieved a 23% reduction of the COD in the absence 27 of inorganic anions, and the process was not significantly affected by the presence of Cl−, SO4 2− or 28 HCO3 −/CO3 2− (Figure 1). In absolute terms, while the removal of COD decreased ≤ 3% when Cl− or 29 HCO3 −/CO3 2− were added at any of the tested concentrations, the presence of SO4 2− increased the 30 efficiency of the treatment an additional 2-3%. These non-significant effects were shown along the whole 31 performance of the treatment (e.g. Cl−, Figure 2A), indicating that SO4 2−, Cl−, and HCO3 −/CO3 2− were 32 non-effective radical scavengers, nor competitors for active sites on TiO2, in the performed photocatalytic 33 degradation of phenol; whereas the oxidative contribution of SO4·− radical was confirmed (10, 21-24). 34 On the other hand, a significant increase in the reduction of the COD was observed when NO3 − 35 was added to the solution (Figure 1). While the addition of 250 mg·L−1 of NO3 − enhanced the removal of 36 COD up to a 31%, the presence of the highest tested dosage of 2000 mg·L−1 achieved an almost 37%. 37 Differences among results produced at different NO3 − concentration were not statistically significant; so 38 the presence of 250 mg·L−1 of NO3 − was enough to almost saturate the potential effect. In addition, 39 almost all the improvement of the oxidation efficiency was produced during the first 15-30 minutes of 40 treatment, as denoted by the observed changes in the slopes of the curves illustrating the evolution of the 41 removal of COD along reaction time (Figure 2B). All the curves, whether in the presence or the absence 42 of NO3 −, keep more or less parallel thereafter; thus showing no further treatment enhancement. 43 A brown colour that progressively darkened the solution was observed along photocatalytic 44 treatment in the presence of NO3 −, suggesting the formation of some chromatic phenol derivatives (e.g. 45 benzoquinone (39, 47, 48)); whereas no remarkable colour change was noticed in the presence of other 46 anions. The presence of NO3 − (or NO3· radical) may therefore be able to drift phenol’s route of 47 degradation towards the production of certain intermediates that accumulate near TiO2 particles and 48 accelerate further intermediate coupling reactions on its surface (18, 49). In addition, some authors have 49 proved the ability of NO3 − to absorb UV-light yielding OH·, which may represent a strong homogenous 50 “accelerating” effect for the ongoing photo-degradative process (25, 50, 51). 51 6 When the solution was not acidified after the addition of Na2CO3 , the produced buffered alkaline 1 pH values (10-12) strongly reduced the removal of COD in comparison to the experiments performed at 2 an initial pH=5.0 (COD removal ≈ 22.5%, Figure 3A), which results were identical to those achieved 3 adjusting pH=4.0. As a higher concentration of CO3 2− was added, a higher pH value was reached, and a 4 lower removal of COD was achieved. In the presence of 250 mg·L−1 of CO3 2− (pH≈10.5), the reduction of 5 the COD dropped to ≈12%; whereas at the highest concentration of 2000 mg·L−1 (pH≈11.5), a poor 5% 6 removal was just achieved after a 2-hours oxidation treatment. 7 This effect of the presence of CO3 2− has been previously attributed to the production of CO3 −· 8 radical, which scavenges OH·, and therefore causes poorer COD removal efficiencies (15, 26, 35-37). In 9 order to check this out, further trials were performed in the absence of anions adjusting pH to 12.0 (± 0.2). 10 Very poor results (COD removal < 5%), equal to those produced in the presence of a high amount of 11 CO3 2−, were achieved. A radical scavenging effect may therefore be neglected; whereas a strong effect of 12 pH itself was supported (Figure 3B). As TiO2 surface is negatively charged at pH>6.8, and the presence 13 of phenol as negatively-charged phenolate species is significant at higher pH values, its adsorption on the 14 surface of the catalyst is therefore hindered by the action of repulsive electrostatic forces (22, 27, 28). In 15 addition, a lower degradation rate at higher pH values has been attributed to the fact that a higher 16 concentration of OH− prevents UV-light from penetrating the solution to effectively reach the surface of 17 the catalyst (52). 18 19 Ozonation 20 In the absence of inorganic anions, the ozone oxidation of phenol produced a 55% removal of 21 COD after a 30-minute-treatment without adjusting the pH (Figure 4A); whereas the addition of SO4 2−, 22 Cl−, or NO3 − resulted in a non-significant detrimental effect on the treatment efficiency (COD removal 23 ≥50% in any case). A similar pH evolution along the treatment was observed in the absence of anions and 24 when SO4 2−, Cl−, or NO3 − were added to the solution; that is, initial pH averaged 5.7 ± 0.2, and then 25 decreased to 3.2 ± 0.2 along the first 5 min of treatment, keeping more or less constant thereafter (the 26 formation of carboxylic acids in the degradation process of phenol may acidify the solution, e.g. (53-56)). 27 In addition, whether in the absence of anions or in the presence of SO4 2−, Cl−, or NO3 − in the solution, no 28 significant differences were shown in the reduction of COD along the whole treatment time, as shown for 29 all tested SO4 2− concentration values in Figure 4B. SO4 2−, Cl−, and NO3 − were therefore ineffective radical 30 scavengers when phenol was treated by ozone oxidation under acid conditions. 31 Provided there is a pH threshold below which the leading degradation mechanism of the 32 ozonation treatment of every organic substance in the solution is direct oxidation by molecular ozone, and 33 above which indirect oxidation after O3 decomposition to OH· is predominant (11), and considering the 34 above reported results under pretty strong acid conditions, it is reasonable to suppose that direct ozonation 35 is widely dominating the oxidation process; thus potential radical scavengers could not produce much 36 effect on the results. 37 On the other hand, the addition of CO3 2− to the solution increased pH to basic values and the 38 removal of COD was improved in comparison to when no anions were added and pH was not adjusted 39 (Figure 5A). While the addition of 250 mg·L−1 of CO3 2− (pH≈12.5-9.5) increased the efficiency of the 30-40 minute ozonation treatment close to a 60% COD removal, the addition of 500 mg·L−1 (pH≈13.0-10.5) 41 significantly increased the efficiency of the treatment to about the 65%. No significant greater 42 enhancement was achieved adding up to 2000 mg·L−1 of CO3 2−. 43 When ozonation treatment was carried out in the absence of CO3 2−, but pH was adjusted to similar 44 basic values (pH≈12-11), the reduction of the COD resulted even higher (≈70%; Figure 5A); which 45 supports the hypothesis that this apparent removal enhancement attributed to the presence of CO3 2− was 46 really caused by the increase of the pH value itself, which promotes the production of OH· radical when 47 OH− anion concentration is higher (33, 34). In fact, the ozonation process seems actually partially 48 hampered after the addition of CO3 2− when the results achieved at similar pH values are compared, which 49 has been previously explained by an OH· scavenging effect (32, 34, 35). Finally, the effect of CO3 2− on 50 the ozonation efficiency of phenol was mainly produced during the first 10 minutes of treatment; after 20 51 7 minutes, no further effect was accounted for, as denoted by the changes in the slopes of the curves 1 representing the evolution of COD removal along the time this treatment was performed (Figure 5B). 2 In order to compare further results under acid conditions, pH adjustment to 5.7 was set after the 3 addition of Na2CO3. After several attempts, it was checked out that the actual pH value was buffered to 4 6.5 (±0.2) by the air present in the bubble column and the action of HCO3 −/CO3 2− equilibrium. The 5 addition of 500 mg·L−1 of CO3 2− under these conditions enhanced significantly the degradation of phenol 6 by ozone, producing up to a 68% COD removal; whereas in the absence of ions, or when the addition of 7 an equivalent concentration of the other tested anions, it resulted in about a 55% reduction of the COD if 8 pH was not adjusted (pH=5.7 ± 0.2; Figure 6). On the other hand, when all the experiments were repeated 9 adjusting the pH to 6.5 or 12, it resulted that about the 75% of the initial COD was removed, whether no 10 inorganic anion was added to the solution, whether SO4 2−, Cl−, or NO3 − were added. 11 These results show that, regardless whether direct or indirect oxidation is the leading mechanism 12 in the ozonation treatment of phenol, the presence of SO4 2−, Cl−, and NO3 − is not significantly scavenging 13 radicals and does not significantly affect the degradation process. It is additionally shown that ozonation 14 treatment is significantly enhanced at a pH value close to neutral (6.5), or under basic conditions (12.0), 15 which was surely caused by the promotion of OH· indirect oxidation (11). On the other hand, it has been 16 also addressed that the ozonation rate of phenolic compounds increases at higher basic pH values, as it 17 also does the degree of deprotonation and dissociation into phenolate species (57). The oxidation 18 treatment efficiency is therefore also improved under these conditions because the ozone-phenolate 19 reaction is faster than the ozone-phenol one (40). Finally, the significant lower COD removal rate 20 addressed to the addition of CO3 2− supports a radical scavenging effect that has been previously attributed 21 to the presence of HCO3 −/CO3 2− (31, 32). 22 23 Conclusions 24 Although other authors have reported inconclusive to even contradictory conclusions regarding 25 the effect of inorganic anions on the performance of several AOPs, it has been really demonstrated with 26 statistical significance that the addition of SO4 2− and Cl− did not reduce the efficiency of the 27 photocatalytic treatment of phenol when the pH of the solution was not adjusted (pH≤5); and an identical 28 negligible effect was found when CO3 2− was added and pH was adjusted to similar acid values. 29 On the other hand, the presence of NO3 − significantly enhanced the photocatalytic reduction of 30 the COD (≈15%). This effect may be explained by the ability of NO3 − to absorb UV light, leading to an 31 additional production of OH· radicals; the enhancement of the adsorption of phenol on the surface of 32 TiO2; and the acceleration of intermediate reactions on the catalyst surface. 33 The addition of CO3 2− without pH adjustment resulted in an apparent loss of efficiency of the 34 photocatalytic treatment of phenol, which was really caused by the generated higher pH value itself 35 (pH≈10.0-12.0), as phenol adsorption on the catalyst surface was hampered under alkaline conditions due 36 to the repulsive electrostatic forces that are manifested (both TiO2 surface and phenolate species are 37 negatively charged at such alkaline pH values). 38 Likewise, the presence of SO4 2−, Cl−, or NO3 − did not reduce the efficiency of the treatment to a 39 significant extent when phenol was oxidized by ozone; whether results without adjusting the initial pH 40 value of the solution (pH=5.7) were always 15-20% worse in terms of the achieved reduction of the COD 41 than those performed under adjusted close to neutral (pH=6.5) or basic (pH=12.0) conditions. 42 Finally, the addition of CO3 2− negatively affected the efficiency of ozonating phenol by a 5-10% 43 COD removal due to the manifested radical scavenging effect under both close to neutral (pH=6.5) and 44 basic (pH=9.5-13.5) conditions; whereas there was no such an effect when the pH was adjusted to keep its 45 initial acid value (pH=5.7). 46 47 Acknowledgements 48 This research was developed in the framework of the following projects: “PROLIPAPEL” (S-49 0505/AMB-0100), funded by the Regional Government of Madrid (Comunidad Autónoma de Madrid), 50 Spain; “AGUA Y ENERGÍA” (CTM2008-06886-C02-01), funded by the Ministry of Science and 51 8 Innovation of Spain (Ministerio de Ciencia e Innovación); and “AQUAFIT4USE” (211534), funded by 1 the European Comission. Archimedes Foundation (Estonia) sponsors H. Barndõk’s PhD at Complutense 2 University of Madrid. 3 4 References 5 (1) Esplugas, S.; Giménez, J.; Contreras, S.; Pascual, E.; Rodríguez, M. Water Res. 2002, 36, 1034-1042. 6 (2) Comninellis, C.; Kapalka, A.; Malato, S.; Parsons, S. A.; Poulios, I.; Mantzavinos, D. J. Chem. 7 Technol. Biotechnol. 2008, 83, 769-776. 8 (3) Hermosilla, D.; Cortijo, M.; Huang, C. P. Sci. Total Environ. 2009, 407, 3473–3481. 9 (4) Diebold, U. Surf. Sci. Rep. 2003, 48, 53-229. 10 (5) Chong, M. N.; Jin, B.; Chow, C. W. K.; Saint, C. Water Res. 2010, 44, 2997-3027. 11 (6) Lucas, M. S.; Peres, J. A.; Lan, B. Y.; Puma, G. L. Water Res. 2009, 43, 1523-1532. 12 (7) Lovato, M. E.; Martín, C. A.; Cassano, A. E. Chem. Eng. J. 2009, 146, 486-497. 13 (8) Pignatello, J. J. Environ. Sci. Technol. 1992, 26, 944-951. 14 (9) Lu, M. C. Chemosphere 1997, 35/10, 2285-2293. 15 (10) Burns, R. A.; Crittenden, J. C.; Hand, D. W.; Selzer, V. H.; Sutter, L. L.; Salman, S. R. J. Environ. 16 Eng. 1999, 125/1, 77-85. 17 (11) Hoigné, J.; Bader, H. Water Res. 1976, 10, 377-386. 18 (12) Matthews, R. W. J. Chem. Soc., Faraday Trans. 1 1984, 80/2, 457-71. 19 (13) Bahnemann, D.; Cunningham, J.; Fox, M. A.; Pelizzetti, E.; Serpone, N.; Pichat, P. In Aquatic and 20 Surface Photochemistry; Helz, G. R.; Zepp, R. G.; Crosby D. G., Eds.; Lewis: Boca Raton, 1994; pp. 21 261–316. 22 (14) Guillard, C.; Lachheb, H.; Houas, A.; Ksibi, M.; Elaloui, E.; Herrmann, J. M. J. Photochem. 23 Photobiol., A 2003, 158, 27-36. 24 (15) Abdullah, M.; Low, G. K. C.; Matthews, R. W. J. Phys. Chem. 1990, 94/17, 6820-6825. 25 (16) Yalap, K. S.; Balcioglu, I. A. J. Adv. Oxid. Technol. 2009, 12/1, 134-143. 26 (17) Papadam, T.; Xekoukoulotakis, N. P.; Poulios, I.; Mantzavinos, D. J. Photochem. Photobiol., A 27 2007, 186, 308-315. 28 (18) Wong, C. C.; Chu, W. Chemosphere 2003, 50, 981-987. 29 (19) Wenhua, L.; Hong, L.; Sao’an, C.; Jianquing, Z.; Chunan, C. J. Photochem. Photobiol., A 2000, 131, 30 125-132. 31 (20) Rincón, A. G.; Pulgarin, C. Appl. Catal., B 2004, 51, 283-302. 32 (21) Serrano, K.; Michaud, P. A.; Comninellis, C.; Savall, A. Electrochim. Acta 2002, 48, 431-436. 33 (22) Kashif, N.; Ouyang, F. J. Environ. Sci. 2009, 21, 527-533. 34 (23) Méndez-Díaz, J.; Sánches-Polo, M. ; Rivera-Utrilla, J.; Canonica, S.; von Gunten, U. Chem. Eng. J. 35 2010, 165, 300-306. 36 (24) Rastogi, A.; Al-Abed, S. R.; Dionysiou, D. D. Appl. Catal., B 2009, 85, 171-179. 37 (25) Chen, H. Y.; Zahraa, O.; Bouchy, M. J. Photochem. Photobiol., A 1997, 108, 37-44. 38 (26) Wang, K.; Zhang, J.; Lou, L.; Yang, S.; Chen, Y. J. Photochem. Photobiol., A, 2004, 165, 201-207. 39 (27) Hu, C.; Yu, J. C.; Hao, Z.; Wong, P. K. Appl. Catal., B, 2003, 46, 35-47. 40 (28) Habibi, M. H.; Hassanzadeh, A.; Mahdavi, S. J. Photochem. Photobiol., A 2005, 172, 89-96. 41 (29) Epling, G. A.; Lin, Chemosphere 2002, 46, 937-944. 42 (30) Hoigné, J.; Bader, H. Water Res. 1985, 19 (8), 993-1004. 43 (31) Glaze, W. H.; Kang, J. W. Ind. Eng. Chem. Res. 1989, 28 (11), 1573-1580. 44 (32) Ma, J.; Graham, N. J. D. Water Res. 2000. 34 (15), 3822-3828. 45 (33) Alaton, I. A.; Kornmüller, A.; Jekel, M. R. J. Environ. Eng. 2002, 128 (8), 689-696. 46 (34) Chiang, Y. P.; Liang, Y. Y.; Chang, C. N.; Chao, A. C. Chemosphere 2006, 65, 2395-2400. 47 (35) Song, S.; Xu, X.; Xu, L.; He, Z.; Ying, H.; Chen, J. Ind. Eng. Chem. Res. 2008, 47, 1386-1391. 48 (36) Lair, A.; Ferronato, C.; Chovelon, J. M.; Herrmann, J. M. J. Photochem. Photobiol., A 2008, 193, 49 193-203. 50 (37) Schmelling, D. C.; Gray, K. A.; Kamat, P. V. Water Res. 1997, 31 (6), 1439-1447. 51 9 (38) Ahmed, S.; Rasul, M. G.; Martens, W. N.; Brown, R.; Hashib, M. A. Desalination 2010, 261, 3-18. 1 (39) Li, K. Y.; Kuo, C. H.; Weeks, J. L. AIChE J. 1979, 25, 583–591. 2 (40) Matheswaran, M.; Moon. I. S. J. Ind. Eng. Chem. 2009, 15, 287-292. 3 (41) Chiou, C. H.; Wu, C. Y.; Juang, R. S. Chem. Eng. J. 2008, 139, 322-329. 4 (42) Siedlecka, E. M.; Więckowska, A.; Stepnowski, P. J. Hazard. Mater. 2007, 147, 497-502. 5 (43) Kusic, H.; Koprivanac, N.; Bozic, A. L. Chem. Eng. J. 2006, 123, 127-137. 6 (44) Chang, C. N.; Ma, Y. S.; Fang, G. C.; Chao, A. C.; Tsai, M. C.; Sung, H. F. Chemosphere 2004, 56, 7 1011-1017. 8 (45) Muneer, M.; Qamar, M.; Saquib, M.; Bahnemann, D. W. Chemosphere 2005, 61, 457–468. 9 (46) Shifu, C.; Yunzhang, L. Chemosphere 2007, 67, 1010–1017. 10 (47) Abbas, O.; Rebufa, C.; Dupuy, N.; Kister, J. Talanta 77 (2008) 200–209. 11 (48) Sobczynski, A.; Duczmal, L.; Zmudzinski, W. J. Mol. Catal. 2004, 213, 225–230. 12 (49) Makarova, O. V.; Rajh, T.; Thurnauer, M. C. Sci. Technol. 2000, 34, 4797-4803. 13 (50) Fox, M. In Concepts of Inorganic Photochemistry; Adamson, A. W.; Fleishauer, P. D., Eds.; Wiley-14 Interscience: New York, 1975; pp. 333-380. 15 (51) Stumm, W.; Morgan, J. J. Aquatic Chemistry; Wiley-Interscience: New York, 1996, pp. 1040. 16 (52) Qamar, M.; Muneer, M.; Bahnemann, D. J. Environ. Manage. 2006, 80 (2), 99-106. 17 (53) Huang, C. R.; Shu, H. Y. J. Hazard. Mater.1995, 41, 47-64. 18 (54) Lan, B. Y.; Nigmatullin, R.; Puma, G. L. Water Res. 2008, 42, 2473-2482. 19 (55) Zhang, F. F.; Yediler, A.; Liang, X. M.; Kettrup, A. J. Environ. Sci. Health, Part A: Toxic/Hazard. 20 Subst. Environ. Eng. 2002, 37, 707-713. 21 (56) Hermosilla, D.; Cortijo, M.; Huang, C. P. Chem. Eng. J. 2009, 155, 637-646. 22 (57) Hoigné, J.; Bader, H. Water Res. 1983, 17 (2), 185-194. 23 24 10 Figure captions 1 2 Figure 1. Reduction of the COD after a 2h photocatalytic oxidation treatment of phenol (200 mg·L−1) in 3 the absence and after the addition of chloride, sulphate, or nitrate without pH adjustment (initial 4 pH=5.0±0.2); or in the presence of carbonate at pH=5.0 (Mean ± standard deviation, n=3-6. Letters label 5 homogeneous groups of values). 6 7 Figure 2. COD removal evolution along the UV/TiO2 oxidation treatment of phenol (200 mg·L−1) 8 without pH adjustment (initial pH=5.0±0.2) in the presence of (A) chloride, or (B) nitrate (Mean ± 9 standard deviation, n=3-6). 10 11 Figure 3. (A) Reduction of the COD after a 2h photocatalytic oxidation treatment of phenol (200 mg·L−1) 12 in the presence of carbonate without adjusting the pH (Mean ± standard deviation, n=3-6. Letters label 13 homogeneous groups of values). (B) A strong correlation was found between pH and COD removal when 14 UV/TiO2 treatment was performed under alkaline conditions in the presence or the absence of carbonate. 15 16 Figure 4. (A) Removal of COD after a 30-min ozone oxidation treatment of phenol (200 mg·L−1) in the 17 absence of inorganic anions and in the presence of chloride, sulphate and nitrate without adjusting the pH 18 (initial pH=5.7±0.2). (Mean ± standard deviation, n=3-6. n.s.=non-significant differences were found). 19 (B) The evolution in time of the reduction of the COD in the presence of sulphate is shown as an 20 example. 21 22 Figure 5. (A) Reduction of the COD after a 30-min ozone oxidation treatment of phenol (200 mg·L−1) in 23 the absence of anions at pH=5.7 or 12; and after the addition of carbonate without adjusting the pH (Mean 24 ± standard deviation, n=3-6. Letters label homogeneous groups of values). (B) Evolution in time of the 25 removal of COD in the absence of anions at pH=12, and in the presence of carbonate without pH 26 adjustment. 27 28 Figure 6. COD removal after a 30-min ozonation treatment of phenol (200 mg·L−1) in the absence or the 29 presence (2000 mg·L−1) of sulphate, chloride, nitrate, or carbonate with or without adjusting the pH. 30 11 FIGURE 1 inorganic anions, mg·L-1 0 250 500 1000 2000 C O D r em o va l, % 0 10 20 30 40 no anions Cl- CO3 2- SO4 2- NO3 -a a a a a b b b b Tukey's test P<0.05 12 FIGURE 2 time, min 0 20 40 60 80 100 120 C O D r e m o va l, % 0 10 20 30 40 no anions 250 mg·L-1 NO3 - 500 mg·L-1 NO3 - 1000 mg·L-1 NO3 - 2000 mg·L-1 NO3 - C O D r em o va l, % 0 10 20 30 40 no anions 250 mg·L-1 Cl- 500 mg·L-1 Cl - 1000 mg·L-1 Cl- 2000 mg·L-1 Cl- A B 13 FIGURE 3 pH 10.0 10.5 11.0 11.5 12.0 12.5 C O D r em o va l, % 0 3 6 9 12 15 COD removal (%) = 75.00 - 5.97·pH R2 = 97.65 % carbonate, mg·L-1 0 250 500 1000 2000 C O D r em o va l, % 0 10 20 30 40 p H 9 10 11 12 13 d cd cd bc b a Tukey's test P<0.05 no anions, pH= 5 no anions, pH=12 CO3 2- pH= A B 14 FIGURE 4 anions, mg·L-1 0 250 500 1000 2000 C O D r em o va l, % 0 20 40 60 80 100 no anions SO4 2- Cl- NO3 - Tukey's test P<0.05 n.s. time, min 0 5 10 15 20 25 30 C O D r e m o v a l, % 0 20 40 60 80 100 no anions 250 mg·L-1 SO4 2- 500 mg·L-1 SO4 2- 1000 mg·L-1 SO4 2- 2000 mg·L-1 SO4 2- A B 15   time, min 0 5 10 15 20 25 30 C O D r em o va l, % 0 20 40 60 80 100 no anions (pH adjusted =12) 250 mg·L-1 CO3 2- 500 mg·L-1 CO3 2- 1000 mg·L-1 CO3 2- 2000 mg·L-1 CO3 2- carbonate, mg·L-1 0 250 500 1000 2000 C O D r em o va l, % 20 40 60 80 100 b ab ab ababa Tukey's test P<0.05 no anions, pH= 5.7 no anions, pH=12 CO3 2- (13.5 < pH > 9.5) A B 13.5 < pH > 9.5 (13.5 > pH > 9,5) 13.5 > pH > 9,5 FIGURE 5 16 FIGURE 6 anions, 2000 mg·L-1 no anions sulphate chloride nitrate carbonate C O D r em o va l, % 20 40 60 80 100 Initial pH=5.7 Initial pH=5.7 (buffered) pH=6.5 (adjusted) pH=12 (adjusted) a a ab ab b Tukey's test P<0.05 c c c c ab