- Email: [email protected]
Photocatalytic ozonation of aniline with TiO2-carbon composite materials
Journal of Environmental Management xxx (2016) 1e8
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Research article
Photocatalytic ozonation of aniline with TiO2-carbon composite materials C.A. Orge*, J.L. Faria, M.F.R. Pereira lise e Materiais e Laborato rio de Cata rio Associado, LSREeLCM, Departamento de Engenharia Química, Faculdade de Engenharia, LCM e Laborato Universidade do Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 February 2016 Received in revised form 21 July 2016 Accepted 29 July 2016 Available online xxx
The photocatalytic ozonation of aniline (ANL) aqueous solutions was carried out in the presence of neat titanium dioxide (TiO2), multi-walled carbon nanotubes (MWCNT) and a composite of TiO2 and MWCNT. Independent tests for catalytic ozonation and photocatalysis were also carried out in order to explore the potential occurrence of a synergetic effect. Photocatalytic and catalytic ozonation carried out with an ozone dose of 50 g m3 converted ANL in 15 min. Photocatalysis using P25, commercial TiO2, and an 80:20 (w/w) composite of P25 and MWCNT also led to total ANL conversion, but at longer reaction times. Removal of TOC was higher than 70% for all photocatalytic ozonation systems at 1 h of reaction. With the exception of neat MWCNT, photocatalytic ozonation in the presence of the selected samples led to nearly complete mineralization after 3 h of reaction. Photocatalytic ozonation completely removed oxalic acid (OXA) formed during ANL degradation. The concentration of oxamic acid (OMA, other ANL degradation by-product more refractory than OXA) generally increased with time, and in the photocatalytic ozonation with P25 based materials its concentration decreased earlier. The presence of nitrates and ammonium was confirmed during ANL degradation by all tested treatments, with the exception of the cation in TiO2 catalysed reactions. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Aniline Photocatalytic ozonation Titanium dioxide Carbon nanotubes Composites
1. Introduction Aniline (ANL) and its derivatives are widely used in the chemical industry, predominantly in the production of rubber additives, polymers, pharmaceuticals, agricultural chemicals and pesticides. Furthermore, ANL is an important intermediate in the synthesis of dyes and pigments. Thus is not surprising that such compounds find their way to several types of industrial waste waters. ANL itself is refractory to biodegradation and it is one of the aromatic amines produced from the anaerobic biodegradation of some aromatic compounds, including azo dyes (Faria et al., 2009a, 2009b; Tan, 2001). It is toxic to humans and to the aquatic life, being easily adsorbed in sediments, which extends its persistence in the aquatic environment (Sarasa et al., 2002). Due to the negative environmental impact caused by the toxicity of ANL, waste waters containing such compound demand a thorough treatment before discharge in natural water courses. Due to its refractory character
* Corresponding author. E-mail addresses: [email protected] (C.A. Orge), [email protected] (J.L. Faria), [email protected] (M.F.R. Pereira).
towards oxidation, the treatment of ANL aqueous solutions, even at low concentrations, requires strong oxidising agents such as ozone (O3) (Faria et al., 2009a, 2009b, 2007; Orge et al., 2011). Concerning the ozonation of ANL, which is an aromatic compound with an unshared electron pair, the direct electrophilic attack by ozone is favoured. In this case, the by-products obtained are mainly aldehydes, ketones, acids and nitro compounds. ANL can also be attacked by hydroxyl radicals originated from O3 selfdecomposition, leading to the formation of azo-derivatives and hydroxylated compounds. Both mechanisms are favoured at basic pH where the electrophilic character of the molecule is enhanced (Oliviero et al., 2003). As a result of ANL oxidation, several classes of organic by-products have been identified. Some of them consist of nitrogenated compounds such as nitrosobenzene, nitrobenzene, aminophenol, etc. Another group is composed by condensation products such as azobenzene (Karunakaran et al., 2005) and azoxybenzene resulting from the reaction between ANL molecules and products from its partial oxidation, which are usually coloured. The rupture of the NeC bond leads to the formation of hydroquinone and benzoquinone (Oliviero et al., 2003), and the cleavage of the aromatic ring leads to the formation of significant amounts of
http://dx.doi.org/10.1016/j.jenvman.2016.07.091 0301-4797/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
2
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
saturated organic compounds such as aldehydes, ketones and mostly carboxylic acids. The latter are highly resistant to O3 attack and tend to accumulate in water solutions (Orge et al., 2011). According to the different oxidizing species reported, a scheme of the mechanism proposed for ANL degradation is presented in Fig. 1. The combination of advanced oxidation processes should give better results as compared to individual techniques, since some drawbacks of the individual techniques can be eliminated by some characteristics of other techniques (Gogate and Pandit, 2004). Therefore, the removal of such compounds can be achieved by O3 based Advanced Oxidation Processes (AOPs), which combine O3 with UV irradiation, H2O2 or a solid catalyst in all producing enhanced amounts of HO. These technologies are of great impact because in addition to O3 direct reactions, they involve the action of the generated HO (Rodríguez et al., 2013), allowing to explore different synergies. Among O3 based oxidation processes, photocatalytic ozonation has been proven to be an interesting technology due to its higher hydroxyl radical (HO) yield (Aguinaco et al., 2012). In addition to high production of HO due to the powerful oxidant character of O3 compared to oxygen, O3 captures electrons in the conduction band of the catalyst to yield the ozonide ion radical that eventually gives rise to HO (Aguinaco et al., 2012; García-Araya et al., 2010) This technology has the advantage of minimizing the undesirable recombination reaction of the conduction band electrons with valence band positive holes, which results in the inhibition of the process rate. The ANL degradation by UV-illuminated aqueous TiO2 suspensions strongly depends on pH. The best performance is verified either at very acidic or basic media, as well as at pH values near the nchez et al., 1997). When O3 is point of zero charge of TiO2 (Sa introduced, photocatalytic ozonation led to a TOC decrease of about 85% after 8 h at pH 3 and 6. In the same work it was observed that the ozonation pretreatment followed by photocatalysis increased the TOC removal in comparison to either ozonation or photocatalysis alone; however, the highest TOC removal was achieved by the combined ozonation and photocatalysis process (S anchez et al.,
Fig. 1. Mechanism proposed for ANL degradation. Adapted from (Karunakaran et al., nchez et al., 1997). 2005; Sa
1998). The formation of an ozonide anion radical previous to the generation of HO radicals was suggested to explain the synergic effect between O3 and TiO2 under illumination. ANL removal by photocatalytic reactions was studied in the past; however, the reports available in the literature have been focused only TiO2, namely the commercial sample P25, as active catalyst. In this study, the performance of carbon materials, neat or combined with TiO2, is considered. Therefore, the present work seeks to establish a better catalyst for the photocatalytic ozonation of ANL. In this sense, bare TiO2, multi-walled carbon nanotubes (MWCNT) and a composite of TiO2 and MWCNT were used as (photo)catalysts. In order to verify the presence of a synergetic effect, independently photocatalysis and catalytic ozonation experiments were also carried out. Composites of TiO2 and CNT have also attracted much attention in recent years for their synergetic effect enhancing the overall photocatalytic efficiency of TiO2 (Wang et al., 2012). Photocatalytic studies using CNTeTiO2 slurries have suggested that the role of the CNT phase in the composite catalyst can be ascribed to three distinct mechanisms: (i) CNT may act as a dispersing media for TiO2 nanoparticles, (ii) CNT can act as a co-adsorbent, or (iii) CNT can act as a photosensitizer (Di Paola et al., 2012; Gao et al., 2009; Sampaio et al., 2011; Silva and Faria, 2010; Tian et al., 2011; Wang et al., 2005; Yao et al., 2008). The first mechanism is more significant when TiO2 particles are generated simultaneously during the synthesis of the composite catalyst. In this case, chemical groups at the surface of the CNT may act as anchoring points to TiO2 nanoparticles. Due to their electronic properties, CNT can promote e diffusion, reduce e/hþ recombination and act as photosensitizers, being possible to take advantage of CNT properties to prepare somewhat more efficient photocatalysts (Marques et al., 2013). 2. Experimental 2.1. Reagents and materials ANL (C6H5NH2 99.5%), nitric acid (HNO3, 65%) oxalic acid (C2H2O4, 99%), oxamic acid (C2H3O3N, 98%) and titanium (IV) isopropoxide (Ti[OCH(CH3)2]4, 97%) were purchased from SigmaAldrich Química, S.L., Portugal. Ethanol (C2H5OH, 99.5%) was obrio, Lda. Sulfuric tained from Panreac PVL e Produtos para Laborato acid was supplied by Fisher Scientific, Portugal. Ultrapure water was produced on a Direct-Q milipore system. For this study five catalysts were selected. Two bare TiO2 samples, one commercially rica Sa, sample obtained and used as supplied by Evonik Degussa Ibe P25, and the other prepared by the sol-gel method, sample TiO2 (Silva and Faria, 2009). The preparation of TiO2 consisted of slowly adding Ti[OCH(CH3)2]4 to ethanol. After 30 min under continuous stirring, nitric acid was added. The solution was loosely covered and kept stirring until a homogenous gel was formed. After grinding the xerogel, a fine powder was formed and afterwards it was calcined at 400 C in a nitrogen flow for 2 h. Multi-walled carbon nanotubes (MWCNT) were purchased from Nanocyl 3100 and used as received from NANOCYL SA Belgium. Composite of P25 and MWCNT with 80% of P25 (optimal composition previous determined) was prepared by the hydration-dehydration technique, sample P25/ MWCNT (Sampaio et al., 2011). The preparation of the composite consisted in the dispersion of a determined amount of MWCNT in water under ultrasonication and 30 min later P25 was added to the suspension. The mixture was heated up to 80 C and magnetically stirred until the water was completely evaporated. The resulting composite was dried at 110 C overnight in order to eliminate the remaining humidity. Tested materials were previously characterized and the conditions of these analyses, as well as the results, are described in our recent work (Orge et al., 2014).
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
In short, prepared TiO2 has a higher BET specific surface area (SBET) than commercial P25, respectively SBET ¼ 80 m2 g1 and SBET ¼ 50 m2 g1. As expected, the BET surface area of the composite, SBET ¼ 100 m2 g1, is higher than the neat P25, due to the contribution of carbon phase. No significant differences were observed between the expected and the measured masses of TiO2 (as determined by TG), indicating that the composite was successfully synthetized (Orge et al., 2014).
3
7-250/4.0 column (250 40 mm) with a mobile phase of 0.117 g of 2,6-pyridine dicarboxylic acid and 0.11 mL of HNO3 for 1 dm3 of solution was used for nitrate and nitrite. These analyses were carried out at 45 C using a Suppressed CD (J002) suppressor. Ammonium was analysed in a Metrosep C4-250/4.0 column working at 25 C with a solution of Na2CO3 3.6 mM as the mobile phase. In Table 1 the retention time (tr) and detection limit (DL) for each compound are presented.
2.2. Kinetic experiments Experiments of photocatalytic ozonation (PCO), catalytic ozonation (CO) and photocatalysis (PC) were evaluated following the degradation of aniline (ANL). The experiments were carried out in a glass immersion photochemical reactor under the optimal conditions previously determined. The reactor with 60 mm of diameter and 250 mm of height was loaded with 250 mL of suspension. The initial concentration of ANL was 1 mM (pH z 5.6). The solution was prepared from ultrapure water obtained in a Milli-Q Millipore system. The experiments were carried out in a glass immersion photochemical reactor typically loaded with 250 mL of suspension. The initial concentration of ANL was 1 mM. The irradiation source consisted in a Heraeus TQ 150 medium-pressure mercury vapour lamp (lexc ¼ 254, 313, 366, 436 and 546 nm). A DURAN® glass cooling jacket was used as a filter in order to work with irradiation in the near-UV to visible light range (lmax ¼ 366, 436 and 546 nm). Light intensity and photon flux were determined with an UVeVis spectroradiometer (USB2000þ, OceanOptics, USA) and the obtained values were 43 mW cm2 and 9.7 1016 photons cm2 s1, respectively. The agitation was maintained constant at 400 rpm in order to keep the reactor content perfectly mixed. The gas was introduced in the reactor by a diffuser with 1 cm of diameter. The experiments were performed at constant gas flow rate (150 cm3 min1) and constant inlet O3 concentration (50 g m3). In O3 containing experiments, the gas was produced from pure oxygen in a BMT 802X ozone generator. The concentration of O3 in the gas phase was monitored with a BMT 964 ozone analyser. O3 in the gas phase leaving the reactor was removed in a series of gas washing bottles filled with potassium iodide solution. In catalytic reactions, 125 mg of catalyst were introduced in the reactor. The gas was introduced in the reactor at the same time that light was turned on. Samples for analysis were withdrawn regularly from the reactor and centrifuged immediately for separation of the suspended solids with a VWR MicroStar 12 centrifuge. In direct photolysis and photocatalytic reactions, the O3 containing stream was replaced by an oxygen stream. The absence of internal and external diffusional limitations was previously confirmed. 2.3. Analytical methods For ANL a Lichrocart Purospher Star column (250 mm 4.6 mm, 5 mm) working at room temperature under isocratic elution with 40% of water and 60% of methanol with a flow rate of 1 mL min1, an injection volume of 40 mL was used. The detector's wavelength was 200 nm. For oxalic acid (OXA) and oxamic acid (OMA) an Altech AO-100 column (300 mm 6.5 m) working at room temperature under isocratic elution with H2SO4 5 mM with a flow rate of 0.5 mL min1, an injection volume of 15 mL was used. The detector's wavelength was 200 nm. Nitrate, nitrite and ammonium were analysed by ion chromatography in a MetrOHM 881 Compact IC Pro with 863 Compact Autosampler using an injection volume of 20 mL. A Metrosep A Supp
3. Results 3.1. Aniline degradation The conversion of ANL by PCO in the presence of the selected catalysts followed a first-order kinetic decay. The photocatalytic degradation of ANL can be described by a Langmuir-Hinshelwood kinetic model (Ollis et al., 2015) and assuming that due to combined effects of light and O3 the concentration of HO with respect to ANL is quasi constant. According to this model, the evolution of ANL concentration during non-catalytic run is described by the following equation:
dC ¼ khom C dt
(1)
where khom (min1) represents the first-order apparent rate constant and C (mM) is the concentration of ANL in each instant. Integration of Eq. (1), considering C ¼ C0, when t ¼ 0, leads to:
ln
C0 ¼ khom t C
(2)
In the presence of prepared materials, both homogeneous and heterogeneous degradation occur. Therefore, ANL removal rate is the sum of the two contributions:
dC ¼ ðkhom þ khet ÞC dt
(3)
where khet (min1) represents the first-order apparent rate constant for the heterogeneous degradation. Integration of Eq. (3), considering kapp ¼ khom þ khet and C ¼ C0, when t ¼ 0, leads to:
ln
C0 ¼ kapp t C
(4)
The corresponding rate constants are given in Table 2. CO and PC were also considered, as well as the non-catalytic methods. The so obtained apparent first-order rate constant values are included in Table 2. For the fitting procedure, the value of khom previously calculated was fixed. The evolution of ANL degradation during the tested processes in the presence of P25/MWCNT composite is depicted in Fig. 2. The degradation by non-catalytic methods is also included. The combination of any tested material with O3 and light Table 1 Retention time (tr) and detection limit (DL) of analyzed compounds. Compound
tr (min)
DL (mM)
ANL OXA OMA NO 3 NO 2 NHþ 4
4.5 5.8 9.3 8.1 12.6 20.0
0.05 0.05 0.03 0.005 0.008 0.005
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
4
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
Table 2 First-order rate constants and half-life (t1/2) for degradation of ANL in homogeneous (k ¼ kapp) or heterogeneous processes. Balance of nitrogen-containing species resultant from the oxidation of ANL (C0 ¼ 1 mM) at 3 h of reaction.
System
PCO
Catalyst
t1/2 k (min-1) (min) ±22% ±22%
NO3(mM) ±6.7%
NH4+ (mM) ±0.41%
OMA (mM) ±1.5%
ANL (mM) ±7.2%
NT (mM)
MWCNT
0.29
3.01
0.11
0.54
0.43
-
1.1
TiO2
0.34
2.78
0.040
-
0.43
-
0.47
P25
0.60
1.61
0.078
0.64
0.20
-
0.91
P25/MWCNT
0.79
1.38
0.15
0.28
0.62
-
1.1
O3+Light
-
0.19
3.92
0.064
0.33
0.42
-
0.81
O3
-
0.14
4.60
0.024
0.30
0.35
-
0.0016
-
0.012
0.11
0.01
0.54
0.66
MWCNT
0.72
1.41
0.080
0.30
0.27
-
0.65
Light
CO
PC
0.68
TiO2
0.50
1.46
0.039
-
0.26
-
0.30
P25
0.43
1.98
0.085
0.45
0.27
-
0.81
P25/MWCNT
0.91
1.35
0.086
0.37
0.25
-
0.71
MWCNT
0.0016
-
0.048
0.085
0.007
0.69
0.83
TiO2
0.0020
-
0.040
-
0.26
0.65
0.95
P25
0.023
31.9
0.061
0.58
0.27
-
0.91
P25/MWCNT 0.010
33.9
0.054
0.28
0.12
0.027
0.47
accelerates the degradation of ANL, especially with P25 containing catalysts. Both PCO and CO led to complete removal of ANL from the aqueous solutions. The P25/MWCNT and MWCNT catalysts were the most efficient with complete conversion after 5 min of reaction; however, the other catalysts also led to total ANL removal after 10 min. On the other hand, PC with TiO2 and MWCNT had a performance similar to photolysis, removing approximately 30% of ANL after 3 h of reaction. Only PC in the presence of P25 and P25/MWCNT led to total conversion of ANL after 2 and 3 h, respectively. Compared to PCO and CO, ANL degradation by PC is slower with no complete degradation for all the tested catalysts. The efficiency of the composite catalyst is governed by the success in promoting efficient separation of the charge carriers on the resulting semiconductor material. So, an optimal ratio between
the amounts of TiO2 or P25 and MWCNT must be found for each particular situation. Experiments for different amounts were previously reported (Sampaio et al., 2011, 2013) and the ratio of 80:20 between the metal oxide and carbon phase proved to be the most effective to promote enhanced electronic interaction of the materials. In addition, at this composition the metal oxide dispersion is also maximized. O3 containing reactions were proven to be effective for the degradation of ANL. O3 is an extremely powerful oxidant capable of reacting with a vast range of compounds, and it attacks selectively aromatic moieties and unsaturated bonds. Aromatic compounds possessing electron donor groups, such as amino group (-NH2),
Fig. 2. Evolution of dimensionless ANL concentration during non-catalytic and catalytic reactions with P25/MWCNT composite.
Fig. 3. Percentage of TOC removal during ANL degradation at 60 min and 180 min of reaction.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
have a high electronic density in ortho- and para-positions and react actively with O3 by electrophilic attack. In fact, the eNH2 present in the molecule is an electron donating group, activating the aromatic ring by increasing its electronic density (Orge et al., 2012). Furthermore, aromatic compounds, such as ANL, have a higher delocalization of electrons and then exhibit high reactivity towards O3. Under these experimental conditions, the prepared composite is the most effective catalyst for ANL degradation by PCO (kapp ¼ 0.79 min1), presenting a removal rate 4.2 times higher than the verified with the non-catalytic combined method. Neat P25 also presented high degradation rate, 0.60 min1. MWCNT presented the lowest catalytic activity during PCO, but still showing a degradation rate 1.5 fold higher than photo-ozonation. 3.2. TOC removal Although ANL was easily removed either by PCO or CO, its degradation results in the formation of several intermediates that remain in solution, such as small chain carboxylic acids. Therefore, the evaluation of the mineralization degree by TOC measurements was considered. The TOC removal achieved at 60 and 180 min of reaction by all studied systems is depicted in Fig. 3. The integrated treatment in the presence of any catalyst allowed better TOC elimination than the individual processes. The combined processes removed 90% of TOC after 180 min of reaction, with the exception of MWCNT, which achieved a TOC removal slightly lower (84%). The advantage of using a catalyst is clear when observing the TOC removal at 60 min of reaction. While photoozonation allowed a TOC removal of around 30% at 60 min of reaction, all PCO runs led to higher mineralization degrees, which means that in addition to ANL, some oxidation by-products were also degraded. Analyzing the experimental data for the individual treatments, CO allowed a better TOC removal than PC. This result is in accordance with what was expected, since ANL was slowly degraded by PC and in some cases a considerable amount of ANL is still in solution after 180 min of reaction. For the P25 sample no significant differences were observed in the TOC removal after 180 min between CO and PCO. On the other hand, for the remaining samples the introduction of irradiation increases the mineralization degree. The high performance verified at 60 min of reaction when these two oxidation processes, ozonation and photocatalysis, were carried out together may be explained by the high production of HO resulted from the combination of several reactions (Mehrjouei et al., 2015). Predominantly, photocatalytic reactions begin by photoexciting the surface of photocatalyst with UVeVis radiation, which can provide the appropriate band gap energy to generate photoactivated electronehole pairs. In parallel, O3 molecules can adsorb on the surface of the photocatalyst through three different interactions: physical adsorption, formation of weak hydrogen bonds with surface hydroxyl groups, and molecular or dissociative adsorption into Lewis acid sites (Bulanin et al., 1995), each interaction resulting in the production of active oxygen radicals (O). Huang and Li showed these active oxygen radicals react with water molecules to produce HO (Huang and Li, 2011). The photogenerated electrons on the photocatalyst surface react with adsorbed O2 and O3 molecules as electron acceptors, and these reactions are important to decrease the recombination rate of electronehole (e/hþ) pairs (Augugliaro et al., 2006; Rivas et al., 2006). The recombination of e/hþ negatively affects reduction and oxidation reactions on the photocatalyst surface, quantitatively decreasing the effective interactions and consequently reducing the performance of this technology. The electron affinity of O3 (z2.1 eV) is higher than that of O2 (z0.44 eV), which can promote
5
photocatalytic reactions in the presence of O3 (PCO) more efficiently than in the presence of O2 (PC) (Pichat et al., 2000). The H2O2 molecules that are generated can also react with photoexcited electrons on the photocatalyst surface, forming hydroxyl radicals (Mena et al., 2012). Heterogeneous photocatalytic reactions begin by photoexciting the surface of the photocatalyst with UVeVis radiation, which provides the appropriate band gap energy to generate photoactivated electronehole pairs. In parallel, O3 molecules can adsorb on the surface of the photocatalyst ([O3]A) through three different interactions: physical adsorption, formation of weak hydrogen bonds with surface hydroxyl groups, and molecular or dissociative adsorption into Lewis acid sites (Bulanin et al., 1995), each interaction resulting in the production of active oxygen radicals ([O]A), Eq. (5). Huang and Li showed these active oxygen radicals react in the bulk, or with neighborhood water molecules to produce HO (Huang and Li, 2011).
½O3 A /½O$A þ ½O2 A
(5)
The photogenerated electrons on the photocatalyst surface react with adsorbed O2 and O3 molecules as electron acceptors, and these reactions are important to decrease the recombination rate of electronehole (e/hþ) pairs, as described in Eqs. (6) and (7) (Augugliaro et al., 2006; Rivas et al., 2006).
½O2 A þ e /O$ 2
(6)
½O3 A þ e /O$ 3
(7)
The recombination of e/hþ pairs negatively affects reduction and oxidation reactions on the photocatalyst surface, quantitatively decreasing the effective interactions and consequently reducing the performance of this technology. The electron affinity of O3 (z2.1 eV) is higher than that of O2 (z0.44 eV), thus promoting photocatalytic reactions in the presence of O3 (PCO) more efficiently than in the presence of O2 (PC) (Pichat et al., 2000). The H2O2 molecules that are generated can also react with photoexcited electrons on the photocatalyst surface, forming hydroxyl radicals (Eq. (9)) (Mena et al., 2012).
2HO$2 /H2 O2 þ O2
(8)
½H2 O2 A þ e /HO$ þ OH
(9)
The generated ozodine radical (O3-) rapidly reacts with Hþ in the solution to give HO3 radical, which then decomposes into O2 and HO as shown in the following equations: þ $ O$ 3 þ H /HO3
(10)
HO$3 /O2 þ HO$
(11)
In contrast with HO3, O2- is not likely to give HO radicals in a single step since requires the presence of H2O2 from Eq. (8): H2 O2 þ O$ 2 /HO$ þ HO þ O2
(12)
However, this mechanism needs a total of three electrons for the generation of a single HO species, which is a less favored situation if compared with the one electron needed through the reaction nchez et al., 1998). pathway (Agustina et al., 2005; Sa Additionally, the adsorbed oxygen in the form of superoxide ion (O 2 ) promotes the decomposition of ozone into free radicals in water (Eq. (14)e(16)):
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
6
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
Fig. 4. Evolution of OXA (a) and OMA (b) concentrations during ANL degradation by photocatalytic ozonation.
þ hþ þ ½O2 A þ H2 O/O 2 þ 2H
(13)
O3 þ O 2 /O3 þ O2
(14)
þ O 3 þ H /HO$ þ O2
(15)
The photogenerated holes are also assumed to directly attack the pollutant molecules adsorbed on the photocatalyst surface (Eq. (16)) or to react with water molecules (at acid conditions) or more efficiently with hydroxide anions, þ
½ANLA þ h
/½ANLþ A /…/COx ;
NOx ; SOx ; etc:
(16)
In summary, the high performance observed during ANL
degradation by PCO in the presence of prepared materials, in contrast to the least significant catalytic activity verified in the individual methods, is in accordance with the results reported in the literature. The increase of the HO radicals production, in addition to the decrease of e/hþ pairs recombination verified when O3, radiation and a suitable catalyst are used together may justify the improvement observed during PCO.
3.3. Product analysis The oxidation of ANL leads to the formation of saturated aliphatic compounds such as aldehydes and carboxylic acids. Among the large number of intermediates formed during the oxidation of the selected compound, we focused our analysis in
Fig. 5. Evolution of OXA (a) and (b) and OMA (c) and (d) concentrations during ANL degradation by CO and PC, respectively.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
following those that persist for longer reaction periods, namely OXA and OMA, which evolution during ANL degradation by PCO is presented in Fig. 4. The OXA concentration profile is similar, independently of the process. The concentration of OXA rises to reach a maximum at early reaction times and, in some cases it remains there for a few minutes, before starting to disappear from the solution. As expected, the less active systems need more time to remove OXA; however, it is important to take into account that the concentration depends on the amount of OXA formed, in addition to its removal rate, which is different in each process. In the case of OMA, with MWCNT its concentration increases to a maximum at 90 min, decreasing therafter. For P25 the turning point is less pronounced and occurs earlier, at 60 min. The evolution of OXA and OMA during ANL degradation by CO and PC was also monitored and the concentration profiles are presented in Fig. 5. During CO, OXA concentration increased during the first 30e120 min, depending on the system, decreasing thereafter, but at variance with PCO it was not totally removed. On the other hand, OMA concentration always increased with time. In PC, the concentration of OXA is low and almost no changes were verified along the time, with the exception of the reaction with P25/MWCNT where OXA concentration increased during the 3 h of reaction. The production of OMA was only observed for the systems with P25 containing catalysts. This result was expected, because these two samples degraded all ANL presented in solution and, consequently, several intermediates have been formed and further transformed into saturated compounds, such as OXA and OMA. The mineralization of ANL resulted in the conversion of its initial nchez et al., 1997). nitrogen moieties into ions containing N (Sa þ Therefore, NO , NO and NH concentrations were also monitored 3 2 4 during the reaction. In order to verify the presence of other nitrogen-containing intermediates, balance of N-total (NT) at 3 h of reaction was carried and the results are presented in Table 2. NO 2 was not detected during ANL degradation by all tested systems. As previously reported, only photocatalysis with TiO2 and MWCNT still had considerable amount of ANL presented in solution after 3 h of reaction. The concentration of NHþ 4 is higher than NO3 , independently of the catalytic reactions tested, with exception of reactions carried out with TiO2 where no NHþ 4 was detected. This result suggests that this sample had an oxidation mechanism different of the other samples, independently of the type of reaction, once TiO2 with O3 and/or light did not produce NHþ 4 . Photocatalytic ozonation in the presence of MWCNT and P25/MWCNT did not release other Ncontaining materials, since the balance is closed. On the other hand, the presence of unidentified nitrogen-containing species must be taken into account for the other catalytic systems. 4. Conclusions Titanium dioxide, carbon nanotubes and a composite of both have significant catalytic activity for the degradation of ANL in aqueous solution under irradiation in the near-UV to visible spectral range and in the presence of O3. ANL was 100% converted in 5e10 min by PCO. Complete ANL conversion was also easily attained in the absence of irradiation. However, the presence of light was crucial to achieve high mineralization degrees during the first hour of reaction. Although no significant differences were observed between TiO2 based heterogeneous photocatalysts, the combination of these catalysts with ozonation and near UV/Vis light allowed a TOC removal of
7
approximately 90% after 3 h of reaction. It is important to emphasize that the presence of MWCNT, even pure, during photocatalytic ozonation promoted a significant mineralization of the ANL solution, emerging therefore as an alternative to neat TiO2. The absence of NHþ 4 during reactions with TiO2 suggested a different oxidation mechanism of ANL in the presence of this catalyst. PCO presented higher mineralization degrees than the individual methods, especially than photocatalytic treatment, under equivalent operating conditions. Acknowledgements This work was financially supported by: Project POCI-01-0145FEDER-006984 e Associated Laboratory LSRE-LCM funded by FEDER funds through COMPETE2020 e Programa Operacional Competitividade e Internacionalizaç~ ao (POCI) e and by national ~o para a Cie ^ncia e a Tecnologia. C. A. funds through FCT e Fundaça Orge acknowledges the research fellowship BPD/90309/2012 by FCT. References n, F.J., García-Araya, J.F., Oropesa, A., 2012. Photocatalytic Aguinaco, A., Beltra ozonation to remove the pharmaceutical diclofenac from water: influence of variables. Chem. Eng. J. 189e190, 275e282. Agustina, T.E., Ang, H.M., Vareek, V.K., 2005. A review of synergistic effect of photocatalysis and ozonation on wastewater treatment. J. Photochem. Photobiol. C Photochem. Rev. 6, 264e273. Augugliaro, V., Litter, M., Palmisano, L., Soria, J., 2006. The combination of heterogeneous photocatalysis with chemical and physical operations: a tool for improving the photoprocess performance. J. Photochem. Photobiolog. C 7, 127e144. Bulanin, K.M., Lavalley, J.C., Tsyganenko, A.A., 1995. IR spectra of adsorbed ozone. Colloid. Surf. A 101, 153e158. pez, E., Marcì, G., Palmisano, L., 2012. A survey of photoDi Paola, A., García-Lo catalytic materials for environmental remediation. J. Hazard. Mater. 211e212, 3e29. a ~o, J., Pereira, M., 2009a. Mineralization of substituted aromatic comFaria, P., Orf pounds by ozonation catalyzed by cerium oxide and a cerium oxide-activated carbon composite. Catal. Lett. 127, 195e203. a ~o, J.J.M., Pereira, M.F.R., 2009b. Cerium, mangaFaria, P.C.C., Monteiro, D.C.M., Orf nese and cobalt oxides as catalysts for the ozonation of selected organic compounds. Chemosphere 74, 818e824. a ~o, J.J.M., Pereira, M.F.R., 2007. Ozonation of aniline promoted by Faria, P.C.C., Orf activated carbon. Chemosphere 67, 809e815. Gao, B., Chen, G.Z., Li Puma, G., 2009. Carbon nanotubes/titanium dioxide (CNTs/ TiO2) nanocomposites prepared by conventional and novel surfactant wrapping solegel methods exhibiting enhanced photocatalytic activity. Appl. Catal. B Environ. 89, 503e509. n, F.J., Aguinaco, A., 2010. Diclofenac removal from water by García-Araya, J.F., Beltra ozone and photolytic TiO2 catalysed processes. J. Chem. Technol. Biotechnol. 85, 798e804. Gogate, R. Parag, Pandit, Aniruddha B., 2004. A review of imperative technologies for wastewater treatment II: hybrid methods. Adv. Environ. Res. 8, 553e597. Huang, H., Li, W., 2011. Destruction of toluene by ozone-enhanced photocatalysis: performance and mechanism. Appl. Catal. B Environ. 102, 449e453. Karunakaran, C., Senthilvelan, S., Karuthapandian, S., 2005. TiO2dphotocatalyzed oxidation of aniline. J. Photochem. Photobiol. A Chem. 172, 207e213. Marques, R.R.N., Sampaio, M.J., Carrapiço, P.M., Silva, C.G., Morales-Torres, S., Dra zi c, G., Faria, J.L., Silva, A.M.T., 2013. Photocatalytic degradation of caffeine: developing solutions for emerging pollutants. Catal. Today 209, 108e115. € ller, D., 2015. A review on photocatalytic ozonation Mehrjouei, M., Müller, S., Mo used for the treatment of water and wastewater. Chem. Eng. J. 263, 209e219. Mena, E., Rey, A., Acedo, B., Beltr an, F.J., Malato, S., 2012. On ozone-photocatalysis synergism in black-light induced reactions: oxidizing species production in photocatalytic ozonation versus heterogeneous photocatalysis. Chem. Eng. J. 204e206, 131e140. Oliviero, L., Barbier Jr., J., Duprez, D., 2003. Wet Air Oxidation of nitrogen-containing organic compounds and ammonia in aqueous media. AppAppl. Catal. B Environ. 40, 163e184. Ollis, D., Silva, C.G., Faria, J., 2015. Simultaneous photochemical and photocatalyzed liquid phase reactions: dye decolorization kinetics. Catal. Today 240 (Part A), 80e85. Orge, C.A., Faria, J.L., Pereira, M.F.R., 2014. Removal of oxalic acid, oxamic acid and aniline by a combined photolysis and ozonation process. Environ. Technol. 36,
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
8
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
1075e1083. ~ Orge, C.A., Orf ao, J.J.M., Pereira, M.F.R., Duarte de Farias, A.M., Fraga, M.A., 2012. Ceria and cerium-based mixed oxides as ozonation catalysts. Chem. Eng. J. 200e202, 499e505. a ~o, J.J.M., Pereira, M.F.R., Duarte de Farias, A.M., Neto, R.C.R., Orge, C.A., Orf Fraga, M.A., 2011. Ozonation of model organic compounds catalysed by nanostructured cerium oxides. Appl. Catal. B Environ. 103, 190e199. Pichat, P., Cermenati, L., Albini, A., Mas, D., Delprat, H., Guillard, C., 2000. Degradation processes of organic compounds over uv-irradiated TiO2. effect of ozone. Res. Chem. Intermed. 26, 161e170. n, F.J., Gimeno, O., Carbajo, M., 2006. Fluorene oxidation by coupling Rivas, F.J., Beltra of ozone, radiation, and semiconductors: a mathematical approach to the kinetics. Ind. Eng. Chem. Res. 45, 166e174. rquez, Gracia, Leo n, Elena A., Alvarez, Rodríguez, Eva M., Ma Pedro M., Amat, Ana M., n, Fernando J., 2013. Mechanism considerations for photocatalytic Beltra oxidation, ozonation and photocatalytic ozonation of some pharmaceutical compounds in water. J. Environ. Manage. 127, 114e124. Sampaio, M.J., Marques, R.R.N., Tavares, P.B., Faria, J.L., Silva, A.M.T., Silva, C.G., 2013. Tailoring the properties of immobilized titanium dioxide/carbon nanotube composites for photocatalytic water treatment. J. Environ. Chem. Eng. 1, 945e953. Sampaio, M.J., Silva, C.G., Marques, R.R.N., Silva, A.M.T., Faria, J.L., 2011. Carbon nanotubeeTiO2 thin films for photocatalytic applications. Catal. Today 161, 91e96. nech, X., 1997. Photocatalyzed destruction of aniline in S anchez, L., Peral, J., Dome UV-illuminated aqueous TiO2 suspensions. Electrochim. Acta 42, 1877e1882.
nchez, L., Peral, J., Dome nech, X., 1998. Aniline degradation by combined phoSa tocatalysis and ozonation. Appl. Catal. B Environ. 19, 59e65. s, S., Ormad, P., Gracia, R., Ovelleiro, J.L., 2002. Study of the aromatic Sarasa, J., Corte by-products formed from ozonation of anilines in aqueous solution. Water Res. 36, 3035e3044. Silva, C.G., Faria, J.L., 2009. Anatase vs. rutile efficiency on the photocatalytic degradation of clofibric acid under near UV to visible irradiation. Photochem. Photobiol. Sci. 8, 705e711. Silva, C.G., Faria, J.L., 2010. Photocatalytic oxidation of phenolic compounds by using a carbon nanotube-titanium dioxide composite catalyst. ChemSusChem 3, 609e618. Tan, N.C.G., 2001. Integrated and Sequential Anaerobic/aerobic Biodegradationof Azo Dyes. Wageningen University, Wageningen, The Netherlands. Tian, L., Ye, L., Deng, K., Zan, L., 2011. TiO2/carbon nanotube hybrid nanostructures: solvothermal synthesis and their visible light photocatalytic activity. J. Solid State Chem. 184, 1465e1471. Wang, H., Dong, S., Chang, Y., Faria, J.L., 2012. Enhancing the photocatalytic properties of TiO2 by coupling with carbon nanotubes and supporting gold. J. Hazard. Mat. 235e236, 230e236. Wang, W., Serp, P., Kalck, P., Faria, J.L., 2005. Visible light photodegradation of phenol on MWNT-TiO2 composite catalysts prepared by a modified solegel method. J. Mol. Catal. A Chem. 235, 194e199. Yao, Y., Li, G., Ciston, S., Lueptow, R.M., Gray, K.A., 2008. Photoreactive TiO2/carbon nanotube composites: synthesis and reactivity. Environ. Sci. Technol. 42, 4952e4957.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Research article
Photocatalytic ozonation of aniline with TiO2-carbon composite materials C.A. Orge*, J.L. Faria, M.F.R. Pereira lise e Materiais e Laborato rio de Cata rio Associado, LSREeLCM, Departamento de Engenharia Química, Faculdade de Engenharia, LCM e Laborato Universidade do Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 February 2016 Received in revised form 21 July 2016 Accepted 29 July 2016 Available online xxx
The photocatalytic ozonation of aniline (ANL) aqueous solutions was carried out in the presence of neat titanium dioxide (TiO2), multi-walled carbon nanotubes (MWCNT) and a composite of TiO2 and MWCNT. Independent tests for catalytic ozonation and photocatalysis were also carried out in order to explore the potential occurrence of a synergetic effect. Photocatalytic and catalytic ozonation carried out with an ozone dose of 50 g m3 converted ANL in 15 min. Photocatalysis using P25, commercial TiO2, and an 80:20 (w/w) composite of P25 and MWCNT also led to total ANL conversion, but at longer reaction times. Removal of TOC was higher than 70% for all photocatalytic ozonation systems at 1 h of reaction. With the exception of neat MWCNT, photocatalytic ozonation in the presence of the selected samples led to nearly complete mineralization after 3 h of reaction. Photocatalytic ozonation completely removed oxalic acid (OXA) formed during ANL degradation. The concentration of oxamic acid (OMA, other ANL degradation by-product more refractory than OXA) generally increased with time, and in the photocatalytic ozonation with P25 based materials its concentration decreased earlier. The presence of nitrates and ammonium was confirmed during ANL degradation by all tested treatments, with the exception of the cation in TiO2 catalysed reactions. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Aniline Photocatalytic ozonation Titanium dioxide Carbon nanotubes Composites
1. Introduction Aniline (ANL) and its derivatives are widely used in the chemical industry, predominantly in the production of rubber additives, polymers, pharmaceuticals, agricultural chemicals and pesticides. Furthermore, ANL is an important intermediate in the synthesis of dyes and pigments. Thus is not surprising that such compounds find their way to several types of industrial waste waters. ANL itself is refractory to biodegradation and it is one of the aromatic amines produced from the anaerobic biodegradation of some aromatic compounds, including azo dyes (Faria et al., 2009a, 2009b; Tan, 2001). It is toxic to humans and to the aquatic life, being easily adsorbed in sediments, which extends its persistence in the aquatic environment (Sarasa et al., 2002). Due to the negative environmental impact caused by the toxicity of ANL, waste waters containing such compound demand a thorough treatment before discharge in natural water courses. Due to its refractory character
* Corresponding author. E-mail addresses: [email protected] (C.A. Orge), [email protected] (J.L. Faria), [email protected] (M.F.R. Pereira).
towards oxidation, the treatment of ANL aqueous solutions, even at low concentrations, requires strong oxidising agents such as ozone (O3) (Faria et al., 2009a, 2009b, 2007; Orge et al., 2011). Concerning the ozonation of ANL, which is an aromatic compound with an unshared electron pair, the direct electrophilic attack by ozone is favoured. In this case, the by-products obtained are mainly aldehydes, ketones, acids and nitro compounds. ANL can also be attacked by hydroxyl radicals originated from O3 selfdecomposition, leading to the formation of azo-derivatives and hydroxylated compounds. Both mechanisms are favoured at basic pH where the electrophilic character of the molecule is enhanced (Oliviero et al., 2003). As a result of ANL oxidation, several classes of organic by-products have been identified. Some of them consist of nitrogenated compounds such as nitrosobenzene, nitrobenzene, aminophenol, etc. Another group is composed by condensation products such as azobenzene (Karunakaran et al., 2005) and azoxybenzene resulting from the reaction between ANL molecules and products from its partial oxidation, which are usually coloured. The rupture of the NeC bond leads to the formation of hydroquinone and benzoquinone (Oliviero et al., 2003), and the cleavage of the aromatic ring leads to the formation of significant amounts of
http://dx.doi.org/10.1016/j.jenvman.2016.07.091 0301-4797/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
2
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
saturated organic compounds such as aldehydes, ketones and mostly carboxylic acids. The latter are highly resistant to O3 attack and tend to accumulate in water solutions (Orge et al., 2011). According to the different oxidizing species reported, a scheme of the mechanism proposed for ANL degradation is presented in Fig. 1. The combination of advanced oxidation processes should give better results as compared to individual techniques, since some drawbacks of the individual techniques can be eliminated by some characteristics of other techniques (Gogate and Pandit, 2004). Therefore, the removal of such compounds can be achieved by O3 based Advanced Oxidation Processes (AOPs), which combine O3 with UV irradiation, H2O2 or a solid catalyst in all producing enhanced amounts of HO. These technologies are of great impact because in addition to O3 direct reactions, they involve the action of the generated HO (Rodríguez et al., 2013), allowing to explore different synergies. Among O3 based oxidation processes, photocatalytic ozonation has been proven to be an interesting technology due to its higher hydroxyl radical (HO) yield (Aguinaco et al., 2012). In addition to high production of HO due to the powerful oxidant character of O3 compared to oxygen, O3 captures electrons in the conduction band of the catalyst to yield the ozonide ion radical that eventually gives rise to HO (Aguinaco et al., 2012; García-Araya et al., 2010) This technology has the advantage of minimizing the undesirable recombination reaction of the conduction band electrons with valence band positive holes, which results in the inhibition of the process rate. The ANL degradation by UV-illuminated aqueous TiO2 suspensions strongly depends on pH. The best performance is verified either at very acidic or basic media, as well as at pH values near the nchez et al., 1997). When O3 is point of zero charge of TiO2 (Sa introduced, photocatalytic ozonation led to a TOC decrease of about 85% after 8 h at pH 3 and 6. In the same work it was observed that the ozonation pretreatment followed by photocatalysis increased the TOC removal in comparison to either ozonation or photocatalysis alone; however, the highest TOC removal was achieved by the combined ozonation and photocatalysis process (S anchez et al.,
Fig. 1. Mechanism proposed for ANL degradation. Adapted from (Karunakaran et al., nchez et al., 1997). 2005; Sa
1998). The formation of an ozonide anion radical previous to the generation of HO radicals was suggested to explain the synergic effect between O3 and TiO2 under illumination. ANL removal by photocatalytic reactions was studied in the past; however, the reports available in the literature have been focused only TiO2, namely the commercial sample P25, as active catalyst. In this study, the performance of carbon materials, neat or combined with TiO2, is considered. Therefore, the present work seeks to establish a better catalyst for the photocatalytic ozonation of ANL. In this sense, bare TiO2, multi-walled carbon nanotubes (MWCNT) and a composite of TiO2 and MWCNT were used as (photo)catalysts. In order to verify the presence of a synergetic effect, independently photocatalysis and catalytic ozonation experiments were also carried out. Composites of TiO2 and CNT have also attracted much attention in recent years for their synergetic effect enhancing the overall photocatalytic efficiency of TiO2 (Wang et al., 2012). Photocatalytic studies using CNTeTiO2 slurries have suggested that the role of the CNT phase in the composite catalyst can be ascribed to three distinct mechanisms: (i) CNT may act as a dispersing media for TiO2 nanoparticles, (ii) CNT can act as a co-adsorbent, or (iii) CNT can act as a photosensitizer (Di Paola et al., 2012; Gao et al., 2009; Sampaio et al., 2011; Silva and Faria, 2010; Tian et al., 2011; Wang et al., 2005; Yao et al., 2008). The first mechanism is more significant when TiO2 particles are generated simultaneously during the synthesis of the composite catalyst. In this case, chemical groups at the surface of the CNT may act as anchoring points to TiO2 nanoparticles. Due to their electronic properties, CNT can promote e diffusion, reduce e/hþ recombination and act as photosensitizers, being possible to take advantage of CNT properties to prepare somewhat more efficient photocatalysts (Marques et al., 2013). 2. Experimental 2.1. Reagents and materials ANL (C6H5NH2 99.5%), nitric acid (HNO3, 65%) oxalic acid (C2H2O4, 99%), oxamic acid (C2H3O3N, 98%) and titanium (IV) isopropoxide (Ti[OCH(CH3)2]4, 97%) were purchased from SigmaAldrich Química, S.L., Portugal. Ethanol (C2H5OH, 99.5%) was obrio, Lda. Sulfuric tained from Panreac PVL e Produtos para Laborato acid was supplied by Fisher Scientific, Portugal. Ultrapure water was produced on a Direct-Q milipore system. For this study five catalysts were selected. Two bare TiO2 samples, one commercially rica Sa, sample obtained and used as supplied by Evonik Degussa Ibe P25, and the other prepared by the sol-gel method, sample TiO2 (Silva and Faria, 2009). The preparation of TiO2 consisted of slowly adding Ti[OCH(CH3)2]4 to ethanol. After 30 min under continuous stirring, nitric acid was added. The solution was loosely covered and kept stirring until a homogenous gel was formed. After grinding the xerogel, a fine powder was formed and afterwards it was calcined at 400 C in a nitrogen flow for 2 h. Multi-walled carbon nanotubes (MWCNT) were purchased from Nanocyl 3100 and used as received from NANOCYL SA Belgium. Composite of P25 and MWCNT with 80% of P25 (optimal composition previous determined) was prepared by the hydration-dehydration technique, sample P25/ MWCNT (Sampaio et al., 2011). The preparation of the composite consisted in the dispersion of a determined amount of MWCNT in water under ultrasonication and 30 min later P25 was added to the suspension. The mixture was heated up to 80 C and magnetically stirred until the water was completely evaporated. The resulting composite was dried at 110 C overnight in order to eliminate the remaining humidity. Tested materials were previously characterized and the conditions of these analyses, as well as the results, are described in our recent work (Orge et al., 2014).
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
In short, prepared TiO2 has a higher BET specific surface area (SBET) than commercial P25, respectively SBET ¼ 80 m2 g1 and SBET ¼ 50 m2 g1. As expected, the BET surface area of the composite, SBET ¼ 100 m2 g1, is higher than the neat P25, due to the contribution of carbon phase. No significant differences were observed between the expected and the measured masses of TiO2 (as determined by TG), indicating that the composite was successfully synthetized (Orge et al., 2014).
3
7-250/4.0 column (250 40 mm) with a mobile phase of 0.117 g of 2,6-pyridine dicarboxylic acid and 0.11 mL of HNO3 for 1 dm3 of solution was used for nitrate and nitrite. These analyses were carried out at 45 C using a Suppressed CD (J002) suppressor. Ammonium was analysed in a Metrosep C4-250/4.0 column working at 25 C with a solution of Na2CO3 3.6 mM as the mobile phase. In Table 1 the retention time (tr) and detection limit (DL) for each compound are presented.
2.2. Kinetic experiments Experiments of photocatalytic ozonation (PCO), catalytic ozonation (CO) and photocatalysis (PC) were evaluated following the degradation of aniline (ANL). The experiments were carried out in a glass immersion photochemical reactor under the optimal conditions previously determined. The reactor with 60 mm of diameter and 250 mm of height was loaded with 250 mL of suspension. The initial concentration of ANL was 1 mM (pH z 5.6). The solution was prepared from ultrapure water obtained in a Milli-Q Millipore system. The experiments were carried out in a glass immersion photochemical reactor typically loaded with 250 mL of suspension. The initial concentration of ANL was 1 mM. The irradiation source consisted in a Heraeus TQ 150 medium-pressure mercury vapour lamp (lexc ¼ 254, 313, 366, 436 and 546 nm). A DURAN® glass cooling jacket was used as a filter in order to work with irradiation in the near-UV to visible light range (lmax ¼ 366, 436 and 546 nm). Light intensity and photon flux were determined with an UVeVis spectroradiometer (USB2000þ, OceanOptics, USA) and the obtained values were 43 mW cm2 and 9.7 1016 photons cm2 s1, respectively. The agitation was maintained constant at 400 rpm in order to keep the reactor content perfectly mixed. The gas was introduced in the reactor by a diffuser with 1 cm of diameter. The experiments were performed at constant gas flow rate (150 cm3 min1) and constant inlet O3 concentration (50 g m3). In O3 containing experiments, the gas was produced from pure oxygen in a BMT 802X ozone generator. The concentration of O3 in the gas phase was monitored with a BMT 964 ozone analyser. O3 in the gas phase leaving the reactor was removed in a series of gas washing bottles filled with potassium iodide solution. In catalytic reactions, 125 mg of catalyst were introduced in the reactor. The gas was introduced in the reactor at the same time that light was turned on. Samples for analysis were withdrawn regularly from the reactor and centrifuged immediately for separation of the suspended solids with a VWR MicroStar 12 centrifuge. In direct photolysis and photocatalytic reactions, the O3 containing stream was replaced by an oxygen stream. The absence of internal and external diffusional limitations was previously confirmed. 2.3. Analytical methods For ANL a Lichrocart Purospher Star column (250 mm 4.6 mm, 5 mm) working at room temperature under isocratic elution with 40% of water and 60% of methanol with a flow rate of 1 mL min1, an injection volume of 40 mL was used. The detector's wavelength was 200 nm. For oxalic acid (OXA) and oxamic acid (OMA) an Altech AO-100 column (300 mm 6.5 m) working at room temperature under isocratic elution with H2SO4 5 mM with a flow rate of 0.5 mL min1, an injection volume of 15 mL was used. The detector's wavelength was 200 nm. Nitrate, nitrite and ammonium were analysed by ion chromatography in a MetrOHM 881 Compact IC Pro with 863 Compact Autosampler using an injection volume of 20 mL. A Metrosep A Supp
3. Results 3.1. Aniline degradation The conversion of ANL by PCO in the presence of the selected catalysts followed a first-order kinetic decay. The photocatalytic degradation of ANL can be described by a Langmuir-Hinshelwood kinetic model (Ollis et al., 2015) and assuming that due to combined effects of light and O3 the concentration of HO with respect to ANL is quasi constant. According to this model, the evolution of ANL concentration during non-catalytic run is described by the following equation:
dC ¼ khom C dt
(1)
where khom (min1) represents the first-order apparent rate constant and C (mM) is the concentration of ANL in each instant. Integration of Eq. (1), considering C ¼ C0, when t ¼ 0, leads to:
ln
C0 ¼ khom t C
(2)
In the presence of prepared materials, both homogeneous and heterogeneous degradation occur. Therefore, ANL removal rate is the sum of the two contributions:
dC ¼ ðkhom þ khet ÞC dt
(3)
where khet (min1) represents the first-order apparent rate constant for the heterogeneous degradation. Integration of Eq. (3), considering kapp ¼ khom þ khet and C ¼ C0, when t ¼ 0, leads to:
ln
C0 ¼ kapp t C
(4)
The corresponding rate constants are given in Table 2. CO and PC were also considered, as well as the non-catalytic methods. The so obtained apparent first-order rate constant values are included in Table 2. For the fitting procedure, the value of khom previously calculated was fixed. The evolution of ANL degradation during the tested processes in the presence of P25/MWCNT composite is depicted in Fig. 2. The degradation by non-catalytic methods is also included. The combination of any tested material with O3 and light Table 1 Retention time (tr) and detection limit (DL) of analyzed compounds. Compound
tr (min)
DL (mM)
ANL OXA OMA NO 3 NO 2 NHþ 4
4.5 5.8 9.3 8.1 12.6 20.0
0.05 0.05 0.03 0.005 0.008 0.005
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
4
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
Table 2 First-order rate constants and half-life (t1/2) for degradation of ANL in homogeneous (k ¼ kapp) or heterogeneous processes. Balance of nitrogen-containing species resultant from the oxidation of ANL (C0 ¼ 1 mM) at 3 h of reaction.
System
PCO
Catalyst
t1/2 k (min-1) (min) ±22% ±22%
NO3(mM) ±6.7%
NH4+ (mM) ±0.41%
OMA (mM) ±1.5%
ANL (mM) ±7.2%
NT (mM)
MWCNT
0.29
3.01
0.11
0.54
0.43
-
1.1
TiO2
0.34
2.78
0.040
-
0.43
-
0.47
P25
0.60
1.61
0.078
0.64
0.20
-
0.91
P25/MWCNT
0.79
1.38
0.15
0.28
0.62
-
1.1
O3+Light
-
0.19
3.92
0.064
0.33
0.42
-
0.81
O3
-
0.14
4.60
0.024
0.30
0.35
-
0.0016
-
0.012
0.11
0.01
0.54
0.66
MWCNT
0.72
1.41
0.080
0.30
0.27
-
0.65
Light
CO
PC
0.68
TiO2
0.50
1.46
0.039
-
0.26
-
0.30
P25
0.43
1.98
0.085
0.45
0.27
-
0.81
P25/MWCNT
0.91
1.35
0.086
0.37
0.25
-
0.71
MWCNT
0.0016
-
0.048
0.085
0.007
0.69
0.83
TiO2
0.0020
-
0.040
-
0.26
0.65
0.95
P25
0.023
31.9
0.061
0.58
0.27
-
0.91
P25/MWCNT 0.010
33.9
0.054
0.28
0.12
0.027
0.47
accelerates the degradation of ANL, especially with P25 containing catalysts. Both PCO and CO led to complete removal of ANL from the aqueous solutions. The P25/MWCNT and MWCNT catalysts were the most efficient with complete conversion after 5 min of reaction; however, the other catalysts also led to total ANL removal after 10 min. On the other hand, PC with TiO2 and MWCNT had a performance similar to photolysis, removing approximately 30% of ANL after 3 h of reaction. Only PC in the presence of P25 and P25/MWCNT led to total conversion of ANL after 2 and 3 h, respectively. Compared to PCO and CO, ANL degradation by PC is slower with no complete degradation for all the tested catalysts. The efficiency of the composite catalyst is governed by the success in promoting efficient separation of the charge carriers on the resulting semiconductor material. So, an optimal ratio between
the amounts of TiO2 or P25 and MWCNT must be found for each particular situation. Experiments for different amounts were previously reported (Sampaio et al., 2011, 2013) and the ratio of 80:20 between the metal oxide and carbon phase proved to be the most effective to promote enhanced electronic interaction of the materials. In addition, at this composition the metal oxide dispersion is also maximized. O3 containing reactions were proven to be effective for the degradation of ANL. O3 is an extremely powerful oxidant capable of reacting with a vast range of compounds, and it attacks selectively aromatic moieties and unsaturated bonds. Aromatic compounds possessing electron donor groups, such as amino group (-NH2),
Fig. 2. Evolution of dimensionless ANL concentration during non-catalytic and catalytic reactions with P25/MWCNT composite.
Fig. 3. Percentage of TOC removal during ANL degradation at 60 min and 180 min of reaction.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
have a high electronic density in ortho- and para-positions and react actively with O3 by electrophilic attack. In fact, the eNH2 present in the molecule is an electron donating group, activating the aromatic ring by increasing its electronic density (Orge et al., 2012). Furthermore, aromatic compounds, such as ANL, have a higher delocalization of electrons and then exhibit high reactivity towards O3. Under these experimental conditions, the prepared composite is the most effective catalyst for ANL degradation by PCO (kapp ¼ 0.79 min1), presenting a removal rate 4.2 times higher than the verified with the non-catalytic combined method. Neat P25 also presented high degradation rate, 0.60 min1. MWCNT presented the lowest catalytic activity during PCO, but still showing a degradation rate 1.5 fold higher than photo-ozonation. 3.2. TOC removal Although ANL was easily removed either by PCO or CO, its degradation results in the formation of several intermediates that remain in solution, such as small chain carboxylic acids. Therefore, the evaluation of the mineralization degree by TOC measurements was considered. The TOC removal achieved at 60 and 180 min of reaction by all studied systems is depicted in Fig. 3. The integrated treatment in the presence of any catalyst allowed better TOC elimination than the individual processes. The combined processes removed 90% of TOC after 180 min of reaction, with the exception of MWCNT, which achieved a TOC removal slightly lower (84%). The advantage of using a catalyst is clear when observing the TOC removal at 60 min of reaction. While photoozonation allowed a TOC removal of around 30% at 60 min of reaction, all PCO runs led to higher mineralization degrees, which means that in addition to ANL, some oxidation by-products were also degraded. Analyzing the experimental data for the individual treatments, CO allowed a better TOC removal than PC. This result is in accordance with what was expected, since ANL was slowly degraded by PC and in some cases a considerable amount of ANL is still in solution after 180 min of reaction. For the P25 sample no significant differences were observed in the TOC removal after 180 min between CO and PCO. On the other hand, for the remaining samples the introduction of irradiation increases the mineralization degree. The high performance verified at 60 min of reaction when these two oxidation processes, ozonation and photocatalysis, were carried out together may be explained by the high production of HO resulted from the combination of several reactions (Mehrjouei et al., 2015). Predominantly, photocatalytic reactions begin by photoexciting the surface of photocatalyst with UVeVis radiation, which can provide the appropriate band gap energy to generate photoactivated electronehole pairs. In parallel, O3 molecules can adsorb on the surface of the photocatalyst through three different interactions: physical adsorption, formation of weak hydrogen bonds with surface hydroxyl groups, and molecular or dissociative adsorption into Lewis acid sites (Bulanin et al., 1995), each interaction resulting in the production of active oxygen radicals (O). Huang and Li showed these active oxygen radicals react with water molecules to produce HO (Huang and Li, 2011). The photogenerated electrons on the photocatalyst surface react with adsorbed O2 and O3 molecules as electron acceptors, and these reactions are important to decrease the recombination rate of electronehole (e/hþ) pairs (Augugliaro et al., 2006; Rivas et al., 2006). The recombination of e/hþ negatively affects reduction and oxidation reactions on the photocatalyst surface, quantitatively decreasing the effective interactions and consequently reducing the performance of this technology. The electron affinity of O3 (z2.1 eV) is higher than that of O2 (z0.44 eV), which can promote
5
photocatalytic reactions in the presence of O3 (PCO) more efficiently than in the presence of O2 (PC) (Pichat et al., 2000). The H2O2 molecules that are generated can also react with photoexcited electrons on the photocatalyst surface, forming hydroxyl radicals (Mena et al., 2012). Heterogeneous photocatalytic reactions begin by photoexciting the surface of the photocatalyst with UVeVis radiation, which provides the appropriate band gap energy to generate photoactivated electronehole pairs. In parallel, O3 molecules can adsorb on the surface of the photocatalyst ([O3]A) through three different interactions: physical adsorption, formation of weak hydrogen bonds with surface hydroxyl groups, and molecular or dissociative adsorption into Lewis acid sites (Bulanin et al., 1995), each interaction resulting in the production of active oxygen radicals ([O]A), Eq. (5). Huang and Li showed these active oxygen radicals react in the bulk, or with neighborhood water molecules to produce HO (Huang and Li, 2011).
½O3 A /½O$A þ ½O2 A
(5)
The photogenerated electrons on the photocatalyst surface react with adsorbed O2 and O3 molecules as electron acceptors, and these reactions are important to decrease the recombination rate of electronehole (e/hþ) pairs, as described in Eqs. (6) and (7) (Augugliaro et al., 2006; Rivas et al., 2006).
½O2 A þ e /O$ 2
(6)
½O3 A þ e /O$ 3
(7)
The recombination of e/hþ pairs negatively affects reduction and oxidation reactions on the photocatalyst surface, quantitatively decreasing the effective interactions and consequently reducing the performance of this technology. The electron affinity of O3 (z2.1 eV) is higher than that of O2 (z0.44 eV), thus promoting photocatalytic reactions in the presence of O3 (PCO) more efficiently than in the presence of O2 (PC) (Pichat et al., 2000). The H2O2 molecules that are generated can also react with photoexcited electrons on the photocatalyst surface, forming hydroxyl radicals (Eq. (9)) (Mena et al., 2012).
2HO$2 /H2 O2 þ O2
(8)
½H2 O2 A þ e /HO$ þ OH
(9)
The generated ozodine radical (O3-) rapidly reacts with Hþ in the solution to give HO3 radical, which then decomposes into O2 and HO as shown in the following equations: þ $ O$ 3 þ H /HO3
(10)
HO$3 /O2 þ HO$
(11)
In contrast with HO3, O2- is not likely to give HO radicals in a single step since requires the presence of H2O2 from Eq. (8): H2 O2 þ O$ 2 /HO$ þ HO þ O2
(12)
However, this mechanism needs a total of three electrons for the generation of a single HO species, which is a less favored situation if compared with the one electron needed through the reaction nchez et al., 1998). pathway (Agustina et al., 2005; Sa Additionally, the adsorbed oxygen in the form of superoxide ion (O 2 ) promotes the decomposition of ozone into free radicals in water (Eq. (14)e(16)):
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
6
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
Fig. 4. Evolution of OXA (a) and OMA (b) concentrations during ANL degradation by photocatalytic ozonation.
þ hþ þ ½O2 A þ H2 O/O 2 þ 2H
(13)
O3 þ O 2 /O3 þ O2
(14)
þ O 3 þ H /HO$ þ O2
(15)
The photogenerated holes are also assumed to directly attack the pollutant molecules adsorbed on the photocatalyst surface (Eq. (16)) or to react with water molecules (at acid conditions) or more efficiently with hydroxide anions, þ
½ANLA þ h
/½ANLþ A /…/COx ;
NOx ; SOx ; etc:
(16)
In summary, the high performance observed during ANL
degradation by PCO in the presence of prepared materials, in contrast to the least significant catalytic activity verified in the individual methods, is in accordance with the results reported in the literature. The increase of the HO radicals production, in addition to the decrease of e/hþ pairs recombination verified when O3, radiation and a suitable catalyst are used together may justify the improvement observed during PCO.
3.3. Product analysis The oxidation of ANL leads to the formation of saturated aliphatic compounds such as aldehydes and carboxylic acids. Among the large number of intermediates formed during the oxidation of the selected compound, we focused our analysis in
Fig. 5. Evolution of OXA (a) and (b) and OMA (c) and (d) concentrations during ANL degradation by CO and PC, respectively.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
following those that persist for longer reaction periods, namely OXA and OMA, which evolution during ANL degradation by PCO is presented in Fig. 4. The OXA concentration profile is similar, independently of the process. The concentration of OXA rises to reach a maximum at early reaction times and, in some cases it remains there for a few minutes, before starting to disappear from the solution. As expected, the less active systems need more time to remove OXA; however, it is important to take into account that the concentration depends on the amount of OXA formed, in addition to its removal rate, which is different in each process. In the case of OMA, with MWCNT its concentration increases to a maximum at 90 min, decreasing therafter. For P25 the turning point is less pronounced and occurs earlier, at 60 min. The evolution of OXA and OMA during ANL degradation by CO and PC was also monitored and the concentration profiles are presented in Fig. 5. During CO, OXA concentration increased during the first 30e120 min, depending on the system, decreasing thereafter, but at variance with PCO it was not totally removed. On the other hand, OMA concentration always increased with time. In PC, the concentration of OXA is low and almost no changes were verified along the time, with the exception of the reaction with P25/MWCNT where OXA concentration increased during the 3 h of reaction. The production of OMA was only observed for the systems with P25 containing catalysts. This result was expected, because these two samples degraded all ANL presented in solution and, consequently, several intermediates have been formed and further transformed into saturated compounds, such as OXA and OMA. The mineralization of ANL resulted in the conversion of its initial nchez et al., 1997). nitrogen moieties into ions containing N (Sa þ Therefore, NO , NO and NH concentrations were also monitored 3 2 4 during the reaction. In order to verify the presence of other nitrogen-containing intermediates, balance of N-total (NT) at 3 h of reaction was carried and the results are presented in Table 2. NO 2 was not detected during ANL degradation by all tested systems. As previously reported, only photocatalysis with TiO2 and MWCNT still had considerable amount of ANL presented in solution after 3 h of reaction. The concentration of NHþ 4 is higher than NO3 , independently of the catalytic reactions tested, with exception of reactions carried out with TiO2 where no NHþ 4 was detected. This result suggests that this sample had an oxidation mechanism different of the other samples, independently of the type of reaction, once TiO2 with O3 and/or light did not produce NHþ 4 . Photocatalytic ozonation in the presence of MWCNT and P25/MWCNT did not release other Ncontaining materials, since the balance is closed. On the other hand, the presence of unidentified nitrogen-containing species must be taken into account for the other catalytic systems. 4. Conclusions Titanium dioxide, carbon nanotubes and a composite of both have significant catalytic activity for the degradation of ANL in aqueous solution under irradiation in the near-UV to visible spectral range and in the presence of O3. ANL was 100% converted in 5e10 min by PCO. Complete ANL conversion was also easily attained in the absence of irradiation. However, the presence of light was crucial to achieve high mineralization degrees during the first hour of reaction. Although no significant differences were observed between TiO2 based heterogeneous photocatalysts, the combination of these catalysts with ozonation and near UV/Vis light allowed a TOC removal of
7
approximately 90% after 3 h of reaction. It is important to emphasize that the presence of MWCNT, even pure, during photocatalytic ozonation promoted a significant mineralization of the ANL solution, emerging therefore as an alternative to neat TiO2. The absence of NHþ 4 during reactions with TiO2 suggested a different oxidation mechanism of ANL in the presence of this catalyst. PCO presented higher mineralization degrees than the individual methods, especially than photocatalytic treatment, under equivalent operating conditions. Acknowledgements This work was financially supported by: Project POCI-01-0145FEDER-006984 e Associated Laboratory LSRE-LCM funded by FEDER funds through COMPETE2020 e Programa Operacional Competitividade e Internacionalizaç~ ao (POCI) e and by national ~o para a Cie ^ncia e a Tecnologia. C. A. funds through FCT e Fundaça Orge acknowledges the research fellowship BPD/90309/2012 by FCT. References n, F.J., García-Araya, J.F., Oropesa, A., 2012. Photocatalytic Aguinaco, A., Beltra ozonation to remove the pharmaceutical diclofenac from water: influence of variables. Chem. Eng. J. 189e190, 275e282. Agustina, T.E., Ang, H.M., Vareek, V.K., 2005. A review of synergistic effect of photocatalysis and ozonation on wastewater treatment. J. Photochem. Photobiol. C Photochem. Rev. 6, 264e273. Augugliaro, V., Litter, M., Palmisano, L., Soria, J., 2006. The combination of heterogeneous photocatalysis with chemical and physical operations: a tool for improving the photoprocess performance. J. Photochem. Photobiolog. C 7, 127e144. Bulanin, K.M., Lavalley, J.C., Tsyganenko, A.A., 1995. IR spectra of adsorbed ozone. Colloid. Surf. A 101, 153e158. pez, E., Marcì, G., Palmisano, L., 2012. A survey of photoDi Paola, A., García-Lo catalytic materials for environmental remediation. J. Hazard. Mater. 211e212, 3e29. a ~o, J., Pereira, M., 2009a. Mineralization of substituted aromatic comFaria, P., Orf pounds by ozonation catalyzed by cerium oxide and a cerium oxide-activated carbon composite. Catal. Lett. 127, 195e203. a ~o, J.J.M., Pereira, M.F.R., 2009b. Cerium, mangaFaria, P.C.C., Monteiro, D.C.M., Orf nese and cobalt oxides as catalysts for the ozonation of selected organic compounds. Chemosphere 74, 818e824. a ~o, J.J.M., Pereira, M.F.R., 2007. Ozonation of aniline promoted by Faria, P.C.C., Orf activated carbon. Chemosphere 67, 809e815. Gao, B., Chen, G.Z., Li Puma, G., 2009. Carbon nanotubes/titanium dioxide (CNTs/ TiO2) nanocomposites prepared by conventional and novel surfactant wrapping solegel methods exhibiting enhanced photocatalytic activity. Appl. Catal. B Environ. 89, 503e509. n, F.J., Aguinaco, A., 2010. Diclofenac removal from water by García-Araya, J.F., Beltra ozone and photolytic TiO2 catalysed processes. J. Chem. Technol. Biotechnol. 85, 798e804. Gogate, R. Parag, Pandit, Aniruddha B., 2004. A review of imperative technologies for wastewater treatment II: hybrid methods. Adv. Environ. Res. 8, 553e597. Huang, H., Li, W., 2011. Destruction of toluene by ozone-enhanced photocatalysis: performance and mechanism. Appl. Catal. B Environ. 102, 449e453. Karunakaran, C., Senthilvelan, S., Karuthapandian, S., 2005. TiO2dphotocatalyzed oxidation of aniline. J. Photochem. Photobiol. A Chem. 172, 207e213. Marques, R.R.N., Sampaio, M.J., Carrapiço, P.M., Silva, C.G., Morales-Torres, S., Dra zi c, G., Faria, J.L., Silva, A.M.T., 2013. Photocatalytic degradation of caffeine: developing solutions for emerging pollutants. Catal. Today 209, 108e115. € ller, D., 2015. A review on photocatalytic ozonation Mehrjouei, M., Müller, S., Mo used for the treatment of water and wastewater. Chem. Eng. J. 263, 209e219. Mena, E., Rey, A., Acedo, B., Beltr an, F.J., Malato, S., 2012. On ozone-photocatalysis synergism in black-light induced reactions: oxidizing species production in photocatalytic ozonation versus heterogeneous photocatalysis. Chem. Eng. J. 204e206, 131e140. Oliviero, L., Barbier Jr., J., Duprez, D., 2003. Wet Air Oxidation of nitrogen-containing organic compounds and ammonia in aqueous media. AppAppl. Catal. B Environ. 40, 163e184. Ollis, D., Silva, C.G., Faria, J., 2015. Simultaneous photochemical and photocatalyzed liquid phase reactions: dye decolorization kinetics. Catal. Today 240 (Part A), 80e85. Orge, C.A., Faria, J.L., Pereira, M.F.R., 2014. Removal of oxalic acid, oxamic acid and aniline by a combined photolysis and ozonation process. Environ. Technol. 36,
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091
8
C.A. Orge et al. / Journal of Environmental Management xxx (2016) 1e8
1075e1083. ~ Orge, C.A., Orf ao, J.J.M., Pereira, M.F.R., Duarte de Farias, A.M., Fraga, M.A., 2012. Ceria and cerium-based mixed oxides as ozonation catalysts. Chem. Eng. J. 200e202, 499e505. a ~o, J.J.M., Pereira, M.F.R., Duarte de Farias, A.M., Neto, R.C.R., Orge, C.A., Orf Fraga, M.A., 2011. Ozonation of model organic compounds catalysed by nanostructured cerium oxides. Appl. Catal. B Environ. 103, 190e199. Pichat, P., Cermenati, L., Albini, A., Mas, D., Delprat, H., Guillard, C., 2000. Degradation processes of organic compounds over uv-irradiated TiO2. effect of ozone. Res. Chem. Intermed. 26, 161e170. n, F.J., Gimeno, O., Carbajo, M., 2006. Fluorene oxidation by coupling Rivas, F.J., Beltra of ozone, radiation, and semiconductors: a mathematical approach to the kinetics. Ind. Eng. Chem. Res. 45, 166e174. rquez, Gracia, Leo n, Elena A., Alvarez, Rodríguez, Eva M., Ma Pedro M., Amat, Ana M., n, Fernando J., 2013. Mechanism considerations for photocatalytic Beltra oxidation, ozonation and photocatalytic ozonation of some pharmaceutical compounds in water. J. Environ. Manage. 127, 114e124. Sampaio, M.J., Marques, R.R.N., Tavares, P.B., Faria, J.L., Silva, A.M.T., Silva, C.G., 2013. Tailoring the properties of immobilized titanium dioxide/carbon nanotube composites for photocatalytic water treatment. J. Environ. Chem. Eng. 1, 945e953. Sampaio, M.J., Silva, C.G., Marques, R.R.N., Silva, A.M.T., Faria, J.L., 2011. Carbon nanotubeeTiO2 thin films for photocatalytic applications. Catal. Today 161, 91e96. nech, X., 1997. Photocatalyzed destruction of aniline in S anchez, L., Peral, J., Dome UV-illuminated aqueous TiO2 suspensions. Electrochim. Acta 42, 1877e1882.
nchez, L., Peral, J., Dome nech, X., 1998. Aniline degradation by combined phoSa tocatalysis and ozonation. Appl. Catal. B Environ. 19, 59e65. s, S., Ormad, P., Gracia, R., Ovelleiro, J.L., 2002. Study of the aromatic Sarasa, J., Corte by-products formed from ozonation of anilines in aqueous solution. Water Res. 36, 3035e3044. Silva, C.G., Faria, J.L., 2009. Anatase vs. rutile efficiency on the photocatalytic degradation of clofibric acid under near UV to visible irradiation. Photochem. Photobiol. Sci. 8, 705e711. Silva, C.G., Faria, J.L., 2010. Photocatalytic oxidation of phenolic compounds by using a carbon nanotube-titanium dioxide composite catalyst. ChemSusChem 3, 609e618. Tan, N.C.G., 2001. Integrated and Sequential Anaerobic/aerobic Biodegradationof Azo Dyes. Wageningen University, Wageningen, The Netherlands. Tian, L., Ye, L., Deng, K., Zan, L., 2011. TiO2/carbon nanotube hybrid nanostructures: solvothermal synthesis and their visible light photocatalytic activity. J. Solid State Chem. 184, 1465e1471. Wang, H., Dong, S., Chang, Y., Faria, J.L., 2012. Enhancing the photocatalytic properties of TiO2 by coupling with carbon nanotubes and supporting gold. J. Hazard. Mat. 235e236, 230e236. Wang, W., Serp, P., Kalck, P., Faria, J.L., 2005. Visible light photodegradation of phenol on MWNT-TiO2 composite catalysts prepared by a modified solegel method. J. Mol. Catal. A Chem. 235, 194e199. Yao, Y., Li, G., Ciston, S., Lueptow, R.M., Gray, K.A., 2008. Photoreactive TiO2/carbon nanotube composites: synthesis and reactivity. Environ. Sci. Technol. 42, 4952e4957.
Please cite this article in press as: Orge, C.A., et al., Photocatalytic ozonation of aniline with TiO2-carbon composite materials, Journal of Environmental Management (2016), http://dx.doi.org/10.1016/j.jenvman.2016.07.091