- Email: [email protected]
Photocatalytic oxidation of dinitronaphthalenes: Theory and experiment
Chemosphere 75 (2009) 1008–1014
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Photocatalytic oxidation of dinitronaphthalenes: Theory and experiment Miray Bekbolet a, Zekiye Çınar b,*, Murat Kılıç b, Ceyda Senem Uyguner a, Claudio Minero c, Ezio Pelizzetti c a b c
Bogazici University, Institute of Environmental Sciences, 34342 Bebek, Istanbul, Turkey Yıldız Technical University, Department of Chemistry, 34220 Istanbul, Turkey Universita di Torino, Dipartimento di Chimica Analitica, Via P. Giuria 5, 10125 Torino, Italy
a r t i c l e
i n f o
Article history: Received 25 July 2008 Received in revised form 16 January 2009 Accepted 16 January 2009 Available online 20 February 2009 Keywords: Dinitronaphthalenes Photocatalytic oxidation DFT descriptors Reactivity indices Adsorption model
a b s t r a c t A combination of photocatalytic oxidation experiments and quantum mechanical calculations was used in order to describe the mechanism and the nature of the photocatalytic oxidation reactions of dinitronaphthalane isomers and interprete their reactivities within the framework of the Density Functional Theory (DFT). The photocatalytic oxidation reactions of three dinitronaphthalene isomers, 1,3-dinitronaphthalene, 1,5-dinitronaphthalene and 1,8-dinitronaphthalene in the presence of TiO2 Degussa P-25 grade were investigated experimentally. The reactions were carried out in a Solarbox photoreactor equipped with a Xenon lamp. The removal of the individual substrates was followed by means of a gas chromatographic method. Nonpurgable organic carbon contents of the samples were determined by means of the catalytic oxidation method using Total Organic Carbon analyzer. With the intention of determining the best reactivity descriptors to explain the differences in the photocatalytic oxidation rates in terms of the molecular properties, geometry optimizations of the compounds were performed with the Density Functional Theory DFT at B3LYP/6-31G* level. In order to take the effect of adsorption on the oxidation rate, a cluster Ti9O18 cut from the anatase bulk structure was modeled. The binding energies for the compounds were calculated by using the double-zeta basis set. Global hardness, softness, Fukui functions, local hardness–softness and local softness differences were calculated. The results show that the reactions investigated are orbital-controlled and electrophilic in nature. Local DFT descriptors reflect the reactivities of the dinitronaphthalene isomers better than the global ones, due to the differences in their adsorptive capacities. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction Policyclic aromatic hydrocarbons (PAHs) and their derivatives have become of enormous concern in the last few years (Gramatica et al., 2007; Hien et al., 2007; Hemelsoet et al., 2007; Gerasimov, 2007; Shemer and Linden, 2007a; Jonker, 2008). PAHs have been recognized as priority pollutants, due to their toxicity, carcinogenicity and mutagenicity. Although they are naturally originating in volcano eruptions, forest fires and fossilization processes, anthropogenic combustion processes are the main source of PAHs. Nitrated polyaromatic hydrocarbons (nitro-PAHs) are derivatives of PAHs that are ubiquitous environmental pollutants found in the exhaust fumes of gasoline and diesel combustion engines, in certain food products as a result of incomplete combustion and in combustion source emissions. They are also formed in the atmosphere through photochemical reactions of the parent PAHs with OH and NO3 radicals. These compounds exhibit higher mutagenicity (2 105 times) and carcinogenicity (10 times) than their parent PAHs (Hien et al., 2007). * Corresponding author. Tel.: +90 212 383 4179; fax: +90 212 383 4106. E-mail address: [email protected] (Z. Çınar). 0045-6535/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2009.01.051
The vast use of nitroderivatives of naphthalenes as explosives during the World Wars I and II received a recent interest on the environmental fate of these compounds (Bausinger et al., 2004). The knowledge of the environmental behavior and the dispersal of nitronaphthalenes at explosives-contaminated sites is very limited. PAHs are also formed in steam cracking units used in the petroleum industry for the production of light olefines such as ethylene and propylene. So far, most of the studies on PAHs and nitro-PAHs are on such growth processes of these chemicals, in order to eliminate some possible reaction paths (Hemelsoet et al., 2004, 2007). In all PAH growth processes various classes of elementary reactions such as hydrogen abstraction, addition, cyclization and dehydrogenation reactions can be distinguished. Among these reaction paths, hydrogen abstractions by methyl radicals and intra- or intermolecular additions to double bonds are the most widely and theoretically studied. Correlations between the mutagenicity of PAHs and their molecular structure have also been developed by the QSAR (quantitative structure–activity relation) regression (Gramatica et al., 2007; Ohura et al., 2008). The majority of the reports were focused on the air pollution perspectives rather than aquatic systems as well as soil media (Feilberg et al., 1999).
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
PAHs and nitro-PAHs are removed from the environment by photo- and chemical oxidation and biodegradation. Chemical oxidation methods are expensive and nitro-PAHs are known to withstand biodegradation. Photolytic oxidation processes, such as UV/ ozone, UV/H2O2 or UV/perchlorate have been proven to be successful in destroying many PAHs in wastewater. However, these processes result in the formation of products of higher polarity, and solubility in water which are more toxic than the parent compound (Shemer and Linden, 2007a,b). Therefore, in recent years attention has been directed towards heterogeneous photocatalysis, an advanced oxidation process (AOP) in order to avoid these highly toxic pollutants. Heterogeneous photocatalysis is based on the combined use of low energy UV-A light and semiconductor photocatalysts. The anatase form of TiO2 is the current preferred semiconductor, since it is inexpensive, non-toxic, chemically and biologically inert and commercially available. The most important feature of this advanced oxidation process is the generation of the .OH radicals upon irradiation. Band-gap excitation of TiO2 (3.2 eV) generates electron– hole pairs that can initiate redox reactions on the surface of TiO2 particles. The oxidation reactions of the adsorbed OH ions or H2O molecules with the photogenerated holes yield .OH radicals which can degrade a great variety of organic compounds (Bahnemann et al., 1994; Mills and Hunte, 1997; Pichat, 1997). The photocatalytic degradation of nitrogenous organic compounds has been investigated extensively (Maurino et al., 1997; Piccinini et al., 1997a,b; Waki et al., 2000). In light of the relevant studies on the concurrent oxidative and reductive processes occurring during photocatalysis, an evaluation of the system has been illustrated by Pelizzetti and Minero, (1999). In order to use heterogeneous photocatalysis as an effective water treatment technique, and to eliminate certain reaction paths in the degradation process yielding more hazardous compounds than the original pollutant, the mechanism and the nature of the photocatalytic degradation reaction of the pollutant under consideration should be known. As for the mixtures of different pollutants, the reactivities of the individual molecules are also needed. A detailed understanding of the reactivities of the pollutant molecules and the mechanism of the photocatalytic degradation reactions can only be achieved through the use of experimental techniques together with theoretical calculations. The aim of this study was to investigate the photocatalytic oxidation mechanism of three selective dinitronaphthalene isomers namely 1,3-dinitronaphthalene (1,3-diNN), 1,5-dinitronaphthalene (1,5-diNN) and 1,8-dinitronaphthalene (1,8-diNN). This study was conducted to investigate the photocatalytic oxidation of diNNs in combination with the quantum mechanical calculations in order to describe the mechanism and the nature of the photocatalytic oxidation reactions of diNNs and interpret their reactivities within the framework of the Density Functional Theory (DFT). With this aim, the photocatalytic oxidation reactions of the three selected dinitronaphthalene isomers as 1,3-diNN, 1,5-diNN, 1,8diNN were modeled and DFT calculations were performed in order to obtain their electronic properties. Then by the application of the Conceptual DFT (Geerlings et al., 2003), reactivity descriptors providing information on the mechanism of the reactions and the reactivities of the diNN molecules were calculated. In order to compare with the calculated results, photocatalytic oxidation experiments were carried out and the apparent rate constants were determined by following the removal of the individual substrates as well as the corresponding organic carbon contents. And finally, the DFT-based reactivity descriptors were assessed whether they are capable of providing reliable information about the experimentally obtained reactivity sequence of the three diNN molecules.
1009
2. Theoretical background In any chemical reaction, the reactivity of a molecule depends upon its response to the perturbations caused by the attacking chemical species, in our case OH radicals. The typical perturbations for a chemical reaction are changes in external potential v(r) and the number of electrons N. The Conceptual DFT provides a framework to discuss reactions in terms of these changing properties. This approach leads to a series of DFT-based reactivity descriptors, such as; the electronic chemical potential, hardness, softness and Fukui function. These descriptors then may be connected to the different reactivity principals to be used in the determination of the reactivities of the compounds under investigation (Torrent-Sucarrat et al., 2005). The calculations based on DFT-based reactivity descriptors are computationally less intensive because all information is obtained through study of the reactants only. DFT-based reactivity descriptors are defined as derivatives of the electronic energy with N and v(r). Using the finite-difference approach together with Koopman’s theorem, global hardness g and global softness S can be written in terms of the first ionization potential I and the electron affinity A of the molecule (Chermette, 1999; Vos et al., 2002; Torrent-Sucarrat et al., 2005). Whereas, in the frozen-core approximation g equals the gap between the frontier orbitals HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital);
g¼
I A ELUMO EHOMO 1 ¼ ¼ 2 2S 2
ð1Þ
Global hardness is a measure of the stability of the molecule. It is also a measure of the resistance of a chemical species to change its electronic configuration. According to the ‘‘Maximum Hardness Principle (MHP)” molecules arrange themselves so as to be as hard as possible. Therefore, stable molecules are likely to be harder than less stable molecules and thus they have low reactivities. On the other hand; global softness is related with the polarizability of the molecule. Soft molecules have a high polarizability, which can allow a large deformation of the electron cloud. Thus, these two DFT descriptors can be used to discuss the reactivities of the molecules by means of ‘‘Hard and Soft Acid Base Principle (HSAB)” which states that hard acids prefer to react with hard bases and soft acids prefer to react with soft bases (Chermette, 1999). Generally, local properties are used in the determination of the reactivities of different sites of a molecule. Fukui function f ðrÞ is the most important local DFT descriptor. It is based on the frontier molecular orbital (FMO) Theory and defined as the mixed second derivative of the energy of the molecule with respect to N and v(r). The Fukui function reflects the reactivity of a site of the molecule and it is the change in the electron density driven by a change in the number of electrons. The Fukui function is the reactivity index for orbital-controlled reactions, the larger the value of the Fukui function, the higher the reactivity. Fukui functions per atom i in a molecule are defined as (Vos et al., 2002);
fi ¼ ½qi ðNÞ qi ðN 1Þ
ð2Þ
fio
ð3Þ
¼
½fiþ
þ
fi =2
where qi is the electron population of atom i in the molecule. f ðrÞ is used when the system undergoes an electrophilic attack, whereas f o ðrÞ is valid when the system undergoes a radical attack. The Fukui function contains relative information about different sites of a molecule, but the local softness sðrÞ is more important when comparing different sites in different molecules. It is related to the Fukui function through sðrÞ ¼ S f ðrÞ. This equation indicates that f ðrÞ redistributes the global softness among different parts of the molecule.
1010
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
3. Experimental and computational details
used for the preparation of the potassium hydrogen phthalate standard solutions for the respective calibration curves.
3.1. Materials 3.4. Computational set-up Dinitronaphthalene isomers as 1,3-dinitronaphthalene (1,3diNN, purity 97%, kmax = 340 nm), 1,5-dinitronaphthalene (1,5diNN purity 95.4%, kmax = 331 nm) and 1,8-dinitronaphthalene (1,8-diNN purity 99.5%, k max = 328 nm) were used as substrates. TiO2 Degussa P-25 was used as a photocatalyst and in order to avoid interferences from adsorbed organic and inorganic species, the powder was suspended in water (about 10 g L1) preliminarily irradiated for 24 h using the below described light source, washed several times with bidistilled water and dried at 80 °C for 1 h. Predeposition method was employed using 10 mg of each of the individual substrates and 40 mg of TiO2 in dichloromethane (DCM) followed by evaporation to dryness and stored under dark conditions in a glass bottle. Working aqueous solutions were prepared by dispersing 12.5 mg of the TiO2 powder loaded with the substrate in 250 mL of bidistilled water, mixing was provided by sonication followed by a continuous magnetic stirring. The sample volume taken for photocatalytic experiments was 5 mL. Under all conditions the slurries were kept in dark by covering with Al-foil. 3.2. Photoreactor Solarbox, CO.FO.Megra, (Milan, Italy) equipped with a 1500 W Xenon Lamp was used as the photocatalytic reactor. 340 nm cutoff filter was employed to avoid the possibility of direct photolytic action on the substrates (kmax 6 340 nm). Cylindrical Pyrex cells with flat parallel optical windows (u: 40 mm and h: 30 mm) and a lateral aperture (i.d.:10 mm) near the upper window which was closed with a cap were used (Pelizzetti et al., 1989). Total photonic fluence was determined by potassium ferrioxalate actinometer as 5.8 105 Einstein min1 cm2.
Geometry optimizations of the dinitronaphthalane isomers under investigation were performed with the Density Functional Theory DFT method within the GAUSSIAN 03 package (Frisch et al., 2003) due to the fact that it takes electron correlation into account. The DFT calculations were performed by the hybrid B3LYP functional which combines Hartree–Fock HF and Becke exchange terms with the Lee–Yang–Parr correlation functional, in combination with the 6-31G* basis set (Lee et al., 1988; Becke, 1993). Vibrational frequencies were calculated for the determination of the diNN structures as stationary points and true minima on the potential energy surfaces. All the possible stationary geometries located as minima were generated by free rotation around single bonds (Stewart, 1989). The geometric parameters, the frontier orbital energies EHOMO and ELUMO were calculated. With the intent to determine the role of adsorption at TiO2 particles on the photocatalytic reactivities of the dinitronaphthalenes, an adsorption model for each of the compounds was developed. For this purpose, a cluster Ti9O18, cut from the anatase bulk structure was modeled by quantum mechanical techniques. In the developed cluster model, all the distances are fixed at the bulk values. The total energies of the bare and diNNs adsorbed cluster models were calculated by means of the DFT/B3LYP method with the double-zeta LanL2DZ effective core potential, in order to take the relativistic effects into account. Throughout the calculations, the cluster geometry was kept frozen and the adsorption processes of the diNN molecules on TiO2 surface were examined. The orientations and the positions of the molecules on the cluster surface were determined by the molecular mechanics MMFF (molecular mechanics force field) method (Foresman and Frisch, 1996), followed by the geometry optimizations of the adsorbed molecules through the utilization of the DFT/B3LYP method.
3.3. Analytical methods 4. Results and discussion Nine samples for each irradiation period were collected (total 45 mL) and subjected to extraction with DCM using a continuous liquid extractor for 40 min. Following the liquid–liquid extraction step, samples were evaporated using a vacuum rotary evaporator and stored refrigerated in closed cap vials. The same procedure was also employed for the determination of the initial concentrations of the substrates. Duplicate experiments were performed unless otherwise stated. The disappearance of the substrates; i.e. dinitronaphthalenes isomers were followed by means of a gas chromatographic method using Gas Chromotograph Varian 3400CX equipped with a FID detector. A methylphenyl silicone 5–10%ph DB5 column was used. Appropriate amounts of the parent compounds were dissolved in dichloromethane in order to prepare the calibration curves; 9-nitroantracene was used as the internal standard. The following conditions were employed during analysis: detector temperature: 350 °C, injection temperature: 300 °C, oven temperature, temperature programme: 50 °C 3 min., 20 °C min1, 300 °C, injection mode: splitless, carrier gas: nitrogen, flow rate of 20 mL min1. The retention times (min) for the substrates were 14.3, 14.2 and 15.5 for 1,3diNN, 1,5-diNN and 1,8-diNN, respectively. Nonpurgable organic carbon (NPOC) contents of the samples were determined by means of the catalytic oxidation method (Pt/ 680 °C) using Total Organic Carbon analyzer, Shimadzu Model TOC-5000. After appropriate irradiation periods, the suspensions were passed through 0.45 lm cellulose acetate filters (Millipore HA). Duplicate analyses were performed. Deionized water was
4.1. Disappearance of 1,3-, 1,5-, and 1,8-dinitronaphthalenes The parent substrate removal experiments were carried out for the assessment of the reaction kinetics. Although the reactivities of the isomers in the complex redox medium could be visualized as specific to the substitution positions of the nitro groups on the aromatic moieties, dinitronaphthalene isomers followed a rather basic trend of pseudo first order removal kinetics as depicted in Fig. 1. Thus the disappearance of the parent substrate diNN isomer could be described as:
Fig. 1. Removal of dinitronaphthalene isomers via photocatalysis during the course of irradiation (() 1,3-diNN, (j) 1,5-diNN, (N) 1,8-diNN).
1011
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
rate ¼ ðd½diNN=dtÞ ¼ kapp ½diNN
ð4Þ
and, the normalized concentration (C) term as [diNN]t/[diNN]0 with respect to irradiation time t and 0 would yield the apparent rate constant upon linear regression analysis. The first order removal rate constants kapp (min1) were 5.70 102, 3.51 102 and 3.92 102 for 1,3-dinitronaphthalene, 1,5-dinitronaphthalene and 1,8-dinitronaphthalene respectively (R2 > 0.92). The corresponding half-life values were 12 min, 20 min, and 18 min. Relatively higher removal efficiency was achieved for 1,3-dinitronaphthalene isomer followed by considerably similar rates for 1,5-and 1,8-isomers. 4.2. Removal of Total Organic Carbon In the course of photocatalysis, degradation of the parent substrates would not necessarily lead to the removal of the total organic compounds that are formed via radical attacks. Therefore, removals of NPOC were monitored throughout the irradiation periods even for longer periods than 1 half-life (Fig. 2). During the early stages of the photocatalytic oxidation of the dinitronaphthalene isomers, the removal of organic carbon from the reaction medium could be assessed as a rather fast process. On the other hand, due to the predeposition of the substrates on the TiO2 surface followed by the desorption and readsorption of the substrates as well as the products, the measured NPOC contents in solutions could reveal a questionable removal efficiency of the photocatalytic oxidation. The release of organic carbon from the surface in addition to the presence of the formed organic products revealed a considerable amount of the organic carbon in solution even up to the irradiation periods of 60 min. Considering the initial loading amount of the dinitronaphthalene isomers, approximately 45%, 60% and 40% removals were obtained for 1,3-dinitronaphthalene, 1,5-dinitronaphthalene and 1,8dinitronaphthalene respectively under irradiation periods of up to 3 h. No further irradiation periods were performed since substrate removal was followed by GC measurements. 4.3. Reaction model Fig. 3 shows the optimized structures of the three diNN molecules under investigation. Due to its symmetric structure, 1,5-diNN molecule is planar. In 1,3-diNN, the naphthalene ring and the nitrogen atoms of the nitro groups lie on the same plane. But, the oxygen atoms deviate around 10o from the ring plane. 1,8-diNN molecule is characterized by a nonplanar geometry, due to steric interactions between the two nitro groups. The angle between the two benzene rings was calculated to be 11°. The two nitrogen atoms deviate from the ring planes by around 10°, each one facing
Fig. 3. Optimized structures of the dinitronaphthalene isomers.
to the opposite sides of the naphthalene ring. The oxygen atoms deviate more ca. 35° than the nitrogen atoms of the two nitro groups. During photocatalysis, the governing role of active species leading to the initial photoreaction process is still a matter of active controversy. The interfacial transfer of conduction band electrons to dissolved dioxygen present in aqueous reaction matrix acting as primary electron acceptor, would probably be accepted as the rate determining step of the whole photocatalytic reaction. Hydroxyl radicals either adsorbed or present in the bulk solution are generally considered to be the principle reactive species responsible for the photocatalytic reaction. Therefore, the reaction model used in the theoretical part of this study is based on the primary attack of the photogenerated OH radicals (Turchi and Ollis, 1990). The most plausible reaction pathway of hydroxyl radical expressing a strong electrophilic character would be a direct attack on one of the carbon atoms of the naphthalene ring, generally the one with the highest electron density, forming a C–O bond while a p-bond of the aromatic ring is broken. In order to elucidate qualitative data on the nature of the photocatalytic oxidation reaction, the energies of the frontier
Table 1 DFT-based reactivity descriptors, frontier orbital and binding energies for dinitronaphthalene isomers. 1,3-diNN Frontier orbital energies (eV) EHOMO ELUMO
0.2608 0.1151
1,5-diNN 0.2610 0.1140
1,8-diNN 0.2582 0.1007
DFT-based reactivity descriptors Global descriptors
g S Local descriptors f fo s so Ds Fig. 2. NPOC removal via photocatalysis during the course of irradiation. (() 1,3diNN, (j) 1,5-diNN, (N) 1,8-diNN).
Binding energy (eV)
0.07286 6.862
0.07349 6.804
0.07878 6.347
0.187 0.231 1.283 1.585 0.723
0.167 0.201 1.136 1.368 0.870
0.182 0.199 1.155 1.263 0.851
4092.03
4091.49
4097.44
1012
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
molecular orbitals (FMO) of the reactants were calculated and the overlapping orbitals were determined. Due to its highly electrophilic character, OH radical has a very low-lying SOMO (singly occupied molecular orbital), the energy of which was calculated to be 0.324 eV. The calculated EHOMO and ELUMO energies for the diNN molecules investigated are presented in Table 1. As can be seen, EHOMO values are in the range (0.258 eV)–(0.261 eV), whereas ELUMO values are much higher in the range (0.115 eV)– (0.101 eV). Therefore, in the hydroxylation of the diNNs, the SOMO of the OH radical interacts with the HOMOs of the molecules, as displayed in Fig. 4. By the application of the FMO Theory (Eberhardt and Yoshida, 1973), it may be concluded that the reactions under consideration are orbital-controlled reactions. 4.4. Global DFT-based reactivity descriptors The values for the global hardness and global softness values of the three diNN isomers calculated by using Eq. (1) are presented in Table 1. The experimentally determined photocatalytic oxidation rates of the three diNN isomers are in the order 1,3-diNN > 1,8diNN > 1,5-diNN. As seen in Table 1, the highest S value belongs
to 1,3-diNN which is consistent with its highest oxidation rate constant. Soft molecules are more reactive, thus their oxidation rates are faster than the hard molecules. As for the remaining two compounds, softness values exhibit a different order than the experimental one. 1,8-diNN undergoes oxidation faster than 1,5-diNN, however it has a lower S value. The reason may be attributed to the different adsorptive capacities of the diNNs under investigation which will be explained in the following section. 4.5. Adsorptive capacities The efficiency of the photocatalytic oxidation processes of organic compounds in the presence of TiO2 depends upon their adsorption behavior. The TiO2 surface consists of titanium atoms whose coordination is incomplete. These atoms are fourfold coordinated to oxygen and have two unfilled orbitals. Therefore, they can accept two lone electron pairs from electron donors to complete the octahedral coordination (Tunesi and Anderson, 1991). As the active species, OH radicals are formed at the surface of TiO2 particles; adsorption process enhances the photocatalytic oxidation rate of the organic contaminant. In the case of diNNs, the oxygen atoms of the nitro groups may form chelates on TiO2 by forming bonds to titanium cations. The adsorption model for the diNN molecules developed in this study is displayed in Fig. 5. Binding energies Ebind defined as the difference in total energies ET at the optimized geometries, were obtained using the following equation:
Ebind ¼ ET ðTi9 O18 Þ þ ET ðdiNNÞ ET ðTi9 O18 diNNÞ
Fig. 4. Frontier molecular orbitals of the OH radical and 1,3-dinitronaphthalene.
ð5Þ
The calculated binding energies for the three diNN isomers, presented in Table 1, are in the order 1,8-diNN > 1,3-diNN > 1,5-diNN which is different than that of the experimentally determined rate constants. 1,8-diNN is the most strongly bound molecule on to the cluster surface among the three molecules investigated in this study. But, due to its small global softness value, it undergoes oxidation more slowly than 1,3-diNN which is the most reactive molecule with its highest S value. On the other hand, 1,5-diNN undergoes oxidation with the slowest rate in contrast to its high global softness, however it is weakly bound to the TiO2 surface. Consequently, the fact that the global DFT descriptors do not describe the photocatalytic reactivities of the diNN isomers is mainly
Fig. 5. Adsorption model for 1,3-dinitronaphthalene.
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
related with their different adsorptive capacities, which can be explained by the binding energies. 4.6. Local DFT-based reactivity descriptors Since the carbon atom with the highest fi is the site where the electrophilic attack will occur, the calculated fi values are used in the determination of the point of attack of the OH radical for each of the diNN molecules studied. Then, the highest fi values were used to calculate the local softnesses. The calculated local softness and Fukui functions are presented in Table 1. The local softness s and the Fukui function f governing the electrophilic attack are in the same order as the experimental one 1,3-diNN > 1,8-diNN > 1,5diNN, whereas f o and so values show a different trend. This finding indicates that the reactions of the diNN isomers with the photogenerated OH radicals are orbital-controlled reactions rather than radical attack, which is in agreement with our former prediction explained in Section 4.3. In order to obtain the best reactivity descriptor and the nature of the photocatalytic oxidation reactions of the diNN isomers, ‘‘softness-matching principle” (Vos et al., 2002) was used. This principle is based on the HSAB principle. According to the local HSAB principle, soft atoms react preferentially with other soft atoms and hard atoms with other hard atoms. The reactions investigated in this study are orbital-controlled reactions in which softsoft interactions dominate. For these reactions, Ds between the reacting atoms must be as small as possible. Therefore, Ds ¼ sþ ðOÞ s ðCÞ, the difference between the oxygen atom of the attacking OH radical and the carbon atom of the naphthalene ring that undergoes the electrophilic attack was calculated for each of the diNN molecules. The calculated Ds values agree well with the experimental oxidation rates. Therefore, it may be concluded that local DFT descriptors describe the reactivities of the diNN molecules in their photocatalytic oxidation reactions better than the global ones. 5. Conclusions The principal conclusions of the present study can be summarized as follows: 1. The photocatalytic oxidation of dinitronaphthalenes can be described by the simple first order kinetic model. The experimentally determined photocatalytic oxidation rates of the three diNN isomers are in the order 1,3-diNN > 1,8-diNN > 1,5-diNN. 2. The release of organic carbon from the surface in addition to the presence of the formed organic products revealed a considerable amount of the organic carbon in solution. 3. Electronic properties of the dinitronaphthalenes affect their photocatalytic reactivities in aqueous TiO2 suspensions. 4. The photocatalytic oxidation reactions of dinitronaphthalenes are orbital-controlled and electrophilic in nature. 5. The DFT-based reactivity descriptors can well describe the mechanisms of the reactions and the reactivities of the dinitronaphthalene isomers investigated. 6. Local DFT descriptors reflect the reactivities of the dinitronaphthalene isomers better than the global ones, due to the differences in their adsorptive capacities. The local softness values showing the same order as the reaction rates are the best reactivity descriptors to be used for these reactions. 7. The softness difference between the oxygen atom of the attacking OH radical and the carbon atom of the naphthalene ring that undergoes the electrophilic attack indicates that soft–soft interactions are more dominant than radical attack for the photocatalytic oxidation reactions of dinitronaphthalenes.
1013
References Bahnemann, D., Cunningham, J., Fox, M.A., Pelizzetti, E., Pichat, P., Serpone, N., 1994. In: Helz, G.R., Zepp, R.G., Crosby, D.G. (Eds.), Aquatic and Surface Photochemistry. Lewis, Baco Raton, FL, p. 261. Bausinger, T., Dehner, U., Preuß, J., 2004. Determination of mono-, di- and trinitronaphthalanes in soil samples contaminated by explosives. Chemosphere 57, 821–829. Becke, A.D., 1993. Density-functional thermochemistry. 3. The role of exact exchange. J. Chem. Phys. 98, 75648–75652. Chermette, H., 1999. Chemical reactivity indexes in density functional theory. J. Comput. Chem. 20, 129–154. Eberhardt, M.K., Yoshida, M., 1973. Radiation-induced homolytic aromatic substitution. I. Hydroxylation of nitrobenzene, chlorobenzene, and toluene. J. Phys. Chem. 77 (5), 589–597. Feilberg, A., Kamens, R.M., Strommen, M.R., Nielsen, T., 1999. Photochemistry and partitioning of semivolatile nitro-PAH in the atmosphere. Polycycl. Aromat. Comp. 14, 151–160. Foresman, J.B., Frisch, A., 1996. Exploring Chemistry with Electronic Structure Methods, second ed. Gaussian, Inc. Pittsburgh, PA. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Montgomery Jr., J.A., Vreven, T., Kudin, K.N., Burant, J.C., Millam, J.M., Iyengar, S.S., Tomasi, J., Barone, V., Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G.A., Nakatsuji, H., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Klene, M., Li, X., Knox, J.E., Hratchian, H.P., Cross, J.B., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Ayala, P.Y., Morokuma, K., Voth, G.A., Salvador, P., Dannenberg, J.J., Zakrzewski, V.G., Dapprich, S., Daniels, A.D., Strain, M.C., Farkas, O., Malick, D.K., Rabuck, A.D., Raghavachari, K., Foresman, J.B., Ortiz, J.V., Cui, Q., Baboul, A.G., Clifford, S., Cioslowski, J., Stefanov, B.B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Martin, R.L., Fox, D.J., Keith, T., Al-Laham, M.A., Peng, C.Y., Nanayakkara, A., Challacombe, M., Gill, P.M.W., Johnson, B., Chen, W., Wong, M.W., Gonzalez, C., Pople, J.A., 2003. Gaussian 03, B01. Gaussian Inc., Pittsburgh, PA. Geerlings, P., De Proft, F., Langenaeker, W., 2003. Conceptual density functional theory. Chem. Rev. 103, 1793–1873. Gerasimov, G., 2007. Modeling study of electron-beam polycyclic and nitropolycyclic aromatic hydrocarbons treatment. Radiat. Phys. Chem. 76, 27–36. Gramatica, P., Pilutti, P., Papa, E., 2007. Approaches for externally validated QSAR modeling of nitrated polycyclic aromatic hydrocarbon mutagenicity. SAR QSAR Environ. Res. 18 (1-2), 169–178. Hemelsoet, K., Speybroeck, V.V., Marin, G.B., De Proft, F., Geerlings, P., Waroquier, M., 2004. Reactivity indices for radical reactions involving polyaromatics. J. Phys. Chem. A 108, 7281–7290. Hemelsoet, K., Speybroeck, V.V., Waroquier, M., 2007. How useful are reactivity indicators for the description of hydrogen abstraction reactions on polycyclic aromatic hydrocarbons? Chem. Phys. Lett. 444, 17–22. Hien, T.T., Thanh, L.T., Kameda, T., Takenaka, N., Bandow, H., 2007. Nitro-polycyclic aromatic hydrocarbons and polycyclic aromatic hydrocarbons in particulate matter in an urban area of a tropical region: Ho Chi Minh City, Vietnam. Atmos. Environ. 41, 7715–7725. Jonker, M.T.O., 2008. Absorption of polycyclic aromatic hydrocarbons to cellulose. Chemosphere 70, 778–782. Lee, C., Yang, W., Parr, R.G., 1988. Development of the Colle–Salvetti correlationenergy formula into a functional of the electron-density. Phys. Rev. B 37 (2), 785–789. Maurino, V., Minero, C., Pelizzetti, E., Piccinini, P., Serpone, N., Hidaka, H., 1997. The fate of organic nitrogen under photocatalyticconditions: degradation of nitrophenols and aminophenols on irradiated TiO2. J. Photochem. Photobiol. A 109, 171–176. Mills, A., Hunte, S.L., 1997. An overview of semiconductor photocatalysis. J. Photochem. Photobiol. A 108, 1–35. Ohura, T., Amagai, T., Makino, M., 2008. Behavior and prediction of photochemical degradation of chlorinated polycyclic aromatic hydrocarbons in cyclohexane. Chemosphere 70 (11), 2110–2117. Pelizzetti, E., Minero, C., Maurino, V., Sclafani, A., Hidaka, H., Serpone, N., 1989. Photocatalytic degradation of nonylphenol ethoxylated surfactants. Environ. Sci. Technol. 23, 1380–1385. Pelizzetti, E., Minero, C., 1999. Role of oxidative and reductive pathways in the photocatalytic degradation of organic compounds. Colloid. Surface A 151, 321– 327. Piccinini, P., Minero, C., Vincenti, M., Pelizzetti, E., 1997a. Photocatalytic interconversion of nitrogen-containing benzene derivatives. J. Chem. Soc. Faraday Trans. 93, 1993–2000. Piccinini, P., Minero, C., Vincenti, M., Pelizzetti, E., 1997b. Photocatalytic mineralization of nitrogen-containing benzene derivatives. Catal. Today 39, 187–195. Pichat, P., 1997. In: Ertl, G., Knozinger, H., Weitkamp, J. (Eds.), Handbook of Heterogeneous Photo-Catalysis, vol. 4. VCH, Weinheim, p. 2111. Shemer, H., Linden, K.G., 2007a. Aqueous photodegradation and toxicity of the polycyclic aromatic hydrocarbons fluorine, dibenzofuran and dibenzothiophene. Water Res. 41, 853–861. Shemer, H., Linden, K.G., 2007b. Photolysis, oxidation and subsequent toxicity of a mixture of polycyclic aromatic hydrocarbons in natural waters. J. Photochem. Photobiol. A 187, 186–195.
1014
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
Stewart, J.P., 1989. Optimization of parameters for semiempirical methods II. Applications. J. Comput. Chem. 10, 221–264. Torrent-Sucarrat, M., De Proft, F., Geerlings, P., 2005. Stiffness and Raman intensity: a conceptual and computational DFT study. J. Phys. Chem. A 109, 6071–6076. Tunesi, S., Anderson, M., 1991. Influence of chemisorption on the photodecomposition of salicylic acid and related compounds using suspended titania ceramic membranes. J. Phys. Chem.-US 95 (8), 3399–3405. Turchi, C.S., Ollis, D.F., 1990. Photocatalytic degradation of organic water contaminants: mechanisms involving hydroxyl radical attack. J. Catal., 122– 178.
Vos, A.M., Nulens, K.H.L., De Proft, F., Schoonheydt, R.A., Geerlings, P., 2002. Reactivity descriptors and rate constants for electrophilic aromatic substitution: acid zeolite catalyzed methylation of benzene and toluene. J. Phys. Chem. B 106, 2026–2034. Waki, K., Zhao, J., Horikoshi, S., Watanabe, N., Hidaka, H., 2000. Photooxidation mechanism of nitrogen-containing compounds at TiO2/H2 O interfaces: an experimental and theoretical examination of hydrazine derivatives. Chemosphere 41, 337–343.
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Photocatalytic oxidation of dinitronaphthalenes: Theory and experiment Miray Bekbolet a, Zekiye Çınar b,*, Murat Kılıç b, Ceyda Senem Uyguner a, Claudio Minero c, Ezio Pelizzetti c a b c
Bogazici University, Institute of Environmental Sciences, 34342 Bebek, Istanbul, Turkey Yıldız Technical University, Department of Chemistry, 34220 Istanbul, Turkey Universita di Torino, Dipartimento di Chimica Analitica, Via P. Giuria 5, 10125 Torino, Italy
a r t i c l e
i n f o
Article history: Received 25 July 2008 Received in revised form 16 January 2009 Accepted 16 January 2009 Available online 20 February 2009 Keywords: Dinitronaphthalenes Photocatalytic oxidation DFT descriptors Reactivity indices Adsorption model
a b s t r a c t A combination of photocatalytic oxidation experiments and quantum mechanical calculations was used in order to describe the mechanism and the nature of the photocatalytic oxidation reactions of dinitronaphthalane isomers and interprete their reactivities within the framework of the Density Functional Theory (DFT). The photocatalytic oxidation reactions of three dinitronaphthalene isomers, 1,3-dinitronaphthalene, 1,5-dinitronaphthalene and 1,8-dinitronaphthalene in the presence of TiO2 Degussa P-25 grade were investigated experimentally. The reactions were carried out in a Solarbox photoreactor equipped with a Xenon lamp. The removal of the individual substrates was followed by means of a gas chromatographic method. Nonpurgable organic carbon contents of the samples were determined by means of the catalytic oxidation method using Total Organic Carbon analyzer. With the intention of determining the best reactivity descriptors to explain the differences in the photocatalytic oxidation rates in terms of the molecular properties, geometry optimizations of the compounds were performed with the Density Functional Theory DFT at B3LYP/6-31G* level. In order to take the effect of adsorption on the oxidation rate, a cluster Ti9O18 cut from the anatase bulk structure was modeled. The binding energies for the compounds were calculated by using the double-zeta basis set. Global hardness, softness, Fukui functions, local hardness–softness and local softness differences were calculated. The results show that the reactions investigated are orbital-controlled and electrophilic in nature. Local DFT descriptors reflect the reactivities of the dinitronaphthalene isomers better than the global ones, due to the differences in their adsorptive capacities. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction Policyclic aromatic hydrocarbons (PAHs) and their derivatives have become of enormous concern in the last few years (Gramatica et al., 2007; Hien et al., 2007; Hemelsoet et al., 2007; Gerasimov, 2007; Shemer and Linden, 2007a; Jonker, 2008). PAHs have been recognized as priority pollutants, due to their toxicity, carcinogenicity and mutagenicity. Although they are naturally originating in volcano eruptions, forest fires and fossilization processes, anthropogenic combustion processes are the main source of PAHs. Nitrated polyaromatic hydrocarbons (nitro-PAHs) are derivatives of PAHs that are ubiquitous environmental pollutants found in the exhaust fumes of gasoline and diesel combustion engines, in certain food products as a result of incomplete combustion and in combustion source emissions. They are also formed in the atmosphere through photochemical reactions of the parent PAHs with OH and NO3 radicals. These compounds exhibit higher mutagenicity (2 105 times) and carcinogenicity (10 times) than their parent PAHs (Hien et al., 2007). * Corresponding author. Tel.: +90 212 383 4179; fax: +90 212 383 4106. E-mail address: [email protected] (Z. Çınar). 0045-6535/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2009.01.051
The vast use of nitroderivatives of naphthalenes as explosives during the World Wars I and II received a recent interest on the environmental fate of these compounds (Bausinger et al., 2004). The knowledge of the environmental behavior and the dispersal of nitronaphthalenes at explosives-contaminated sites is very limited. PAHs are also formed in steam cracking units used in the petroleum industry for the production of light olefines such as ethylene and propylene. So far, most of the studies on PAHs and nitro-PAHs are on such growth processes of these chemicals, in order to eliminate some possible reaction paths (Hemelsoet et al., 2004, 2007). In all PAH growth processes various classes of elementary reactions such as hydrogen abstraction, addition, cyclization and dehydrogenation reactions can be distinguished. Among these reaction paths, hydrogen abstractions by methyl radicals and intra- or intermolecular additions to double bonds are the most widely and theoretically studied. Correlations between the mutagenicity of PAHs and their molecular structure have also been developed by the QSAR (quantitative structure–activity relation) regression (Gramatica et al., 2007; Ohura et al., 2008). The majority of the reports were focused on the air pollution perspectives rather than aquatic systems as well as soil media (Feilberg et al., 1999).
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
PAHs and nitro-PAHs are removed from the environment by photo- and chemical oxidation and biodegradation. Chemical oxidation methods are expensive and nitro-PAHs are known to withstand biodegradation. Photolytic oxidation processes, such as UV/ ozone, UV/H2O2 or UV/perchlorate have been proven to be successful in destroying many PAHs in wastewater. However, these processes result in the formation of products of higher polarity, and solubility in water which are more toxic than the parent compound (Shemer and Linden, 2007a,b). Therefore, in recent years attention has been directed towards heterogeneous photocatalysis, an advanced oxidation process (AOP) in order to avoid these highly toxic pollutants. Heterogeneous photocatalysis is based on the combined use of low energy UV-A light and semiconductor photocatalysts. The anatase form of TiO2 is the current preferred semiconductor, since it is inexpensive, non-toxic, chemically and biologically inert and commercially available. The most important feature of this advanced oxidation process is the generation of the .OH radicals upon irradiation. Band-gap excitation of TiO2 (3.2 eV) generates electron– hole pairs that can initiate redox reactions on the surface of TiO2 particles. The oxidation reactions of the adsorbed OH ions or H2O molecules with the photogenerated holes yield .OH radicals which can degrade a great variety of organic compounds (Bahnemann et al., 1994; Mills and Hunte, 1997; Pichat, 1997). The photocatalytic degradation of nitrogenous organic compounds has been investigated extensively (Maurino et al., 1997; Piccinini et al., 1997a,b; Waki et al., 2000). In light of the relevant studies on the concurrent oxidative and reductive processes occurring during photocatalysis, an evaluation of the system has been illustrated by Pelizzetti and Minero, (1999). In order to use heterogeneous photocatalysis as an effective water treatment technique, and to eliminate certain reaction paths in the degradation process yielding more hazardous compounds than the original pollutant, the mechanism and the nature of the photocatalytic degradation reaction of the pollutant under consideration should be known. As for the mixtures of different pollutants, the reactivities of the individual molecules are also needed. A detailed understanding of the reactivities of the pollutant molecules and the mechanism of the photocatalytic degradation reactions can only be achieved through the use of experimental techniques together with theoretical calculations. The aim of this study was to investigate the photocatalytic oxidation mechanism of three selective dinitronaphthalene isomers namely 1,3-dinitronaphthalene (1,3-diNN), 1,5-dinitronaphthalene (1,5-diNN) and 1,8-dinitronaphthalene (1,8-diNN). This study was conducted to investigate the photocatalytic oxidation of diNNs in combination with the quantum mechanical calculations in order to describe the mechanism and the nature of the photocatalytic oxidation reactions of diNNs and interpret their reactivities within the framework of the Density Functional Theory (DFT). With this aim, the photocatalytic oxidation reactions of the three selected dinitronaphthalene isomers as 1,3-diNN, 1,5-diNN, 1,8diNN were modeled and DFT calculations were performed in order to obtain their electronic properties. Then by the application of the Conceptual DFT (Geerlings et al., 2003), reactivity descriptors providing information on the mechanism of the reactions and the reactivities of the diNN molecules were calculated. In order to compare with the calculated results, photocatalytic oxidation experiments were carried out and the apparent rate constants were determined by following the removal of the individual substrates as well as the corresponding organic carbon contents. And finally, the DFT-based reactivity descriptors were assessed whether they are capable of providing reliable information about the experimentally obtained reactivity sequence of the three diNN molecules.
1009
2. Theoretical background In any chemical reaction, the reactivity of a molecule depends upon its response to the perturbations caused by the attacking chemical species, in our case OH radicals. The typical perturbations for a chemical reaction are changes in external potential v(r) and the number of electrons N. The Conceptual DFT provides a framework to discuss reactions in terms of these changing properties. This approach leads to a series of DFT-based reactivity descriptors, such as; the electronic chemical potential, hardness, softness and Fukui function. These descriptors then may be connected to the different reactivity principals to be used in the determination of the reactivities of the compounds under investigation (Torrent-Sucarrat et al., 2005). The calculations based on DFT-based reactivity descriptors are computationally less intensive because all information is obtained through study of the reactants only. DFT-based reactivity descriptors are defined as derivatives of the electronic energy with N and v(r). Using the finite-difference approach together with Koopman’s theorem, global hardness g and global softness S can be written in terms of the first ionization potential I and the electron affinity A of the molecule (Chermette, 1999; Vos et al., 2002; Torrent-Sucarrat et al., 2005). Whereas, in the frozen-core approximation g equals the gap between the frontier orbitals HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital);
g¼
I A ELUMO EHOMO 1 ¼ ¼ 2 2S 2
ð1Þ
Global hardness is a measure of the stability of the molecule. It is also a measure of the resistance of a chemical species to change its electronic configuration. According to the ‘‘Maximum Hardness Principle (MHP)” molecules arrange themselves so as to be as hard as possible. Therefore, stable molecules are likely to be harder than less stable molecules and thus they have low reactivities. On the other hand; global softness is related with the polarizability of the molecule. Soft molecules have a high polarizability, which can allow a large deformation of the electron cloud. Thus, these two DFT descriptors can be used to discuss the reactivities of the molecules by means of ‘‘Hard and Soft Acid Base Principle (HSAB)” which states that hard acids prefer to react with hard bases and soft acids prefer to react with soft bases (Chermette, 1999). Generally, local properties are used in the determination of the reactivities of different sites of a molecule. Fukui function f ðrÞ is the most important local DFT descriptor. It is based on the frontier molecular orbital (FMO) Theory and defined as the mixed second derivative of the energy of the molecule with respect to N and v(r). The Fukui function reflects the reactivity of a site of the molecule and it is the change in the electron density driven by a change in the number of electrons. The Fukui function is the reactivity index for orbital-controlled reactions, the larger the value of the Fukui function, the higher the reactivity. Fukui functions per atom i in a molecule are defined as (Vos et al., 2002);
fi ¼ ½qi ðNÞ qi ðN 1Þ
ð2Þ
fio
ð3Þ
¼
½fiþ
þ
fi =2
where qi is the electron population of atom i in the molecule. f ðrÞ is used when the system undergoes an electrophilic attack, whereas f o ðrÞ is valid when the system undergoes a radical attack. The Fukui function contains relative information about different sites of a molecule, but the local softness sðrÞ is more important when comparing different sites in different molecules. It is related to the Fukui function through sðrÞ ¼ S f ðrÞ. This equation indicates that f ðrÞ redistributes the global softness among different parts of the molecule.
1010
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
3. Experimental and computational details
used for the preparation of the potassium hydrogen phthalate standard solutions for the respective calibration curves.
3.1. Materials 3.4. Computational set-up Dinitronaphthalene isomers as 1,3-dinitronaphthalene (1,3diNN, purity 97%, kmax = 340 nm), 1,5-dinitronaphthalene (1,5diNN purity 95.4%, kmax = 331 nm) and 1,8-dinitronaphthalene (1,8-diNN purity 99.5%, k max = 328 nm) were used as substrates. TiO2 Degussa P-25 was used as a photocatalyst and in order to avoid interferences from adsorbed organic and inorganic species, the powder was suspended in water (about 10 g L1) preliminarily irradiated for 24 h using the below described light source, washed several times with bidistilled water and dried at 80 °C for 1 h. Predeposition method was employed using 10 mg of each of the individual substrates and 40 mg of TiO2 in dichloromethane (DCM) followed by evaporation to dryness and stored under dark conditions in a glass bottle. Working aqueous solutions were prepared by dispersing 12.5 mg of the TiO2 powder loaded with the substrate in 250 mL of bidistilled water, mixing was provided by sonication followed by a continuous magnetic stirring. The sample volume taken for photocatalytic experiments was 5 mL. Under all conditions the slurries were kept in dark by covering with Al-foil. 3.2. Photoreactor Solarbox, CO.FO.Megra, (Milan, Italy) equipped with a 1500 W Xenon Lamp was used as the photocatalytic reactor. 340 nm cutoff filter was employed to avoid the possibility of direct photolytic action on the substrates (kmax 6 340 nm). Cylindrical Pyrex cells with flat parallel optical windows (u: 40 mm and h: 30 mm) and a lateral aperture (i.d.:10 mm) near the upper window which was closed with a cap were used (Pelizzetti et al., 1989). Total photonic fluence was determined by potassium ferrioxalate actinometer as 5.8 105 Einstein min1 cm2.
Geometry optimizations of the dinitronaphthalane isomers under investigation were performed with the Density Functional Theory DFT method within the GAUSSIAN 03 package (Frisch et al., 2003) due to the fact that it takes electron correlation into account. The DFT calculations were performed by the hybrid B3LYP functional which combines Hartree–Fock HF and Becke exchange terms with the Lee–Yang–Parr correlation functional, in combination with the 6-31G* basis set (Lee et al., 1988; Becke, 1993). Vibrational frequencies were calculated for the determination of the diNN structures as stationary points and true minima on the potential energy surfaces. All the possible stationary geometries located as minima were generated by free rotation around single bonds (Stewart, 1989). The geometric parameters, the frontier orbital energies EHOMO and ELUMO were calculated. With the intent to determine the role of adsorption at TiO2 particles on the photocatalytic reactivities of the dinitronaphthalenes, an adsorption model for each of the compounds was developed. For this purpose, a cluster Ti9O18, cut from the anatase bulk structure was modeled by quantum mechanical techniques. In the developed cluster model, all the distances are fixed at the bulk values. The total energies of the bare and diNNs adsorbed cluster models were calculated by means of the DFT/B3LYP method with the double-zeta LanL2DZ effective core potential, in order to take the relativistic effects into account. Throughout the calculations, the cluster geometry was kept frozen and the adsorption processes of the diNN molecules on TiO2 surface were examined. The orientations and the positions of the molecules on the cluster surface were determined by the molecular mechanics MMFF (molecular mechanics force field) method (Foresman and Frisch, 1996), followed by the geometry optimizations of the adsorbed molecules through the utilization of the DFT/B3LYP method.
3.3. Analytical methods 4. Results and discussion Nine samples for each irradiation period were collected (total 45 mL) and subjected to extraction with DCM using a continuous liquid extractor for 40 min. Following the liquid–liquid extraction step, samples were evaporated using a vacuum rotary evaporator and stored refrigerated in closed cap vials. The same procedure was also employed for the determination of the initial concentrations of the substrates. Duplicate experiments were performed unless otherwise stated. The disappearance of the substrates; i.e. dinitronaphthalenes isomers were followed by means of a gas chromatographic method using Gas Chromotograph Varian 3400CX equipped with a FID detector. A methylphenyl silicone 5–10%ph DB5 column was used. Appropriate amounts of the parent compounds were dissolved in dichloromethane in order to prepare the calibration curves; 9-nitroantracene was used as the internal standard. The following conditions were employed during analysis: detector temperature: 350 °C, injection temperature: 300 °C, oven temperature, temperature programme: 50 °C 3 min., 20 °C min1, 300 °C, injection mode: splitless, carrier gas: nitrogen, flow rate of 20 mL min1. The retention times (min) for the substrates were 14.3, 14.2 and 15.5 for 1,3diNN, 1,5-diNN and 1,8-diNN, respectively. Nonpurgable organic carbon (NPOC) contents of the samples were determined by means of the catalytic oxidation method (Pt/ 680 °C) using Total Organic Carbon analyzer, Shimadzu Model TOC-5000. After appropriate irradiation periods, the suspensions were passed through 0.45 lm cellulose acetate filters (Millipore HA). Duplicate analyses were performed. Deionized water was
4.1. Disappearance of 1,3-, 1,5-, and 1,8-dinitronaphthalenes The parent substrate removal experiments were carried out for the assessment of the reaction kinetics. Although the reactivities of the isomers in the complex redox medium could be visualized as specific to the substitution positions of the nitro groups on the aromatic moieties, dinitronaphthalene isomers followed a rather basic trend of pseudo first order removal kinetics as depicted in Fig. 1. Thus the disappearance of the parent substrate diNN isomer could be described as:
Fig. 1. Removal of dinitronaphthalene isomers via photocatalysis during the course of irradiation (() 1,3-diNN, (j) 1,5-diNN, (N) 1,8-diNN).
1011
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
rate ¼ ðd½diNN=dtÞ ¼ kapp ½diNN
ð4Þ
and, the normalized concentration (C) term as [diNN]t/[diNN]0 with respect to irradiation time t and 0 would yield the apparent rate constant upon linear regression analysis. The first order removal rate constants kapp (min1) were 5.70 102, 3.51 102 and 3.92 102 for 1,3-dinitronaphthalene, 1,5-dinitronaphthalene and 1,8-dinitronaphthalene respectively (R2 > 0.92). The corresponding half-life values were 12 min, 20 min, and 18 min. Relatively higher removal efficiency was achieved for 1,3-dinitronaphthalene isomer followed by considerably similar rates for 1,5-and 1,8-isomers. 4.2. Removal of Total Organic Carbon In the course of photocatalysis, degradation of the parent substrates would not necessarily lead to the removal of the total organic compounds that are formed via radical attacks. Therefore, removals of NPOC were monitored throughout the irradiation periods even for longer periods than 1 half-life (Fig. 2). During the early stages of the photocatalytic oxidation of the dinitronaphthalene isomers, the removal of organic carbon from the reaction medium could be assessed as a rather fast process. On the other hand, due to the predeposition of the substrates on the TiO2 surface followed by the desorption and readsorption of the substrates as well as the products, the measured NPOC contents in solutions could reveal a questionable removal efficiency of the photocatalytic oxidation. The release of organic carbon from the surface in addition to the presence of the formed organic products revealed a considerable amount of the organic carbon in solution even up to the irradiation periods of 60 min. Considering the initial loading amount of the dinitronaphthalene isomers, approximately 45%, 60% and 40% removals were obtained for 1,3-dinitronaphthalene, 1,5-dinitronaphthalene and 1,8dinitronaphthalene respectively under irradiation periods of up to 3 h. No further irradiation periods were performed since substrate removal was followed by GC measurements. 4.3. Reaction model Fig. 3 shows the optimized structures of the three diNN molecules under investigation. Due to its symmetric structure, 1,5-diNN molecule is planar. In 1,3-diNN, the naphthalene ring and the nitrogen atoms of the nitro groups lie on the same plane. But, the oxygen atoms deviate around 10o from the ring plane. 1,8-diNN molecule is characterized by a nonplanar geometry, due to steric interactions between the two nitro groups. The angle between the two benzene rings was calculated to be 11°. The two nitrogen atoms deviate from the ring planes by around 10°, each one facing
Fig. 3. Optimized structures of the dinitronaphthalene isomers.
to the opposite sides of the naphthalene ring. The oxygen atoms deviate more ca. 35° than the nitrogen atoms of the two nitro groups. During photocatalysis, the governing role of active species leading to the initial photoreaction process is still a matter of active controversy. The interfacial transfer of conduction band electrons to dissolved dioxygen present in aqueous reaction matrix acting as primary electron acceptor, would probably be accepted as the rate determining step of the whole photocatalytic reaction. Hydroxyl radicals either adsorbed or present in the bulk solution are generally considered to be the principle reactive species responsible for the photocatalytic reaction. Therefore, the reaction model used in the theoretical part of this study is based on the primary attack of the photogenerated OH radicals (Turchi and Ollis, 1990). The most plausible reaction pathway of hydroxyl radical expressing a strong electrophilic character would be a direct attack on one of the carbon atoms of the naphthalene ring, generally the one with the highest electron density, forming a C–O bond while a p-bond of the aromatic ring is broken. In order to elucidate qualitative data on the nature of the photocatalytic oxidation reaction, the energies of the frontier
Table 1 DFT-based reactivity descriptors, frontier orbital and binding energies for dinitronaphthalene isomers. 1,3-diNN Frontier orbital energies (eV) EHOMO ELUMO
0.2608 0.1151
1,5-diNN 0.2610 0.1140
1,8-diNN 0.2582 0.1007
DFT-based reactivity descriptors Global descriptors
g S Local descriptors f fo s so Ds Fig. 2. NPOC removal via photocatalysis during the course of irradiation. (() 1,3diNN, (j) 1,5-diNN, (N) 1,8-diNN).
Binding energy (eV)
0.07286 6.862
0.07349 6.804
0.07878 6.347
0.187 0.231 1.283 1.585 0.723
0.167 0.201 1.136 1.368 0.870
0.182 0.199 1.155 1.263 0.851
4092.03
4091.49
4097.44
1012
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
molecular orbitals (FMO) of the reactants were calculated and the overlapping orbitals were determined. Due to its highly electrophilic character, OH radical has a very low-lying SOMO (singly occupied molecular orbital), the energy of which was calculated to be 0.324 eV. The calculated EHOMO and ELUMO energies for the diNN molecules investigated are presented in Table 1. As can be seen, EHOMO values are in the range (0.258 eV)–(0.261 eV), whereas ELUMO values are much higher in the range (0.115 eV)– (0.101 eV). Therefore, in the hydroxylation of the diNNs, the SOMO of the OH radical interacts with the HOMOs of the molecules, as displayed in Fig. 4. By the application of the FMO Theory (Eberhardt and Yoshida, 1973), it may be concluded that the reactions under consideration are orbital-controlled reactions. 4.4. Global DFT-based reactivity descriptors The values for the global hardness and global softness values of the three diNN isomers calculated by using Eq. (1) are presented in Table 1. The experimentally determined photocatalytic oxidation rates of the three diNN isomers are in the order 1,3-diNN > 1,8diNN > 1,5-diNN. As seen in Table 1, the highest S value belongs
to 1,3-diNN which is consistent with its highest oxidation rate constant. Soft molecules are more reactive, thus their oxidation rates are faster than the hard molecules. As for the remaining two compounds, softness values exhibit a different order than the experimental one. 1,8-diNN undergoes oxidation faster than 1,5-diNN, however it has a lower S value. The reason may be attributed to the different adsorptive capacities of the diNNs under investigation which will be explained in the following section. 4.5. Adsorptive capacities The efficiency of the photocatalytic oxidation processes of organic compounds in the presence of TiO2 depends upon their adsorption behavior. The TiO2 surface consists of titanium atoms whose coordination is incomplete. These atoms are fourfold coordinated to oxygen and have two unfilled orbitals. Therefore, they can accept two lone electron pairs from electron donors to complete the octahedral coordination (Tunesi and Anderson, 1991). As the active species, OH radicals are formed at the surface of TiO2 particles; adsorption process enhances the photocatalytic oxidation rate of the organic contaminant. In the case of diNNs, the oxygen atoms of the nitro groups may form chelates on TiO2 by forming bonds to titanium cations. The adsorption model for the diNN molecules developed in this study is displayed in Fig. 5. Binding energies Ebind defined as the difference in total energies ET at the optimized geometries, were obtained using the following equation:
Ebind ¼ ET ðTi9 O18 Þ þ ET ðdiNNÞ ET ðTi9 O18 diNNÞ
Fig. 4. Frontier molecular orbitals of the OH radical and 1,3-dinitronaphthalene.
ð5Þ
The calculated binding energies for the three diNN isomers, presented in Table 1, are in the order 1,8-diNN > 1,3-diNN > 1,5-diNN which is different than that of the experimentally determined rate constants. 1,8-diNN is the most strongly bound molecule on to the cluster surface among the three molecules investigated in this study. But, due to its small global softness value, it undergoes oxidation more slowly than 1,3-diNN which is the most reactive molecule with its highest S value. On the other hand, 1,5-diNN undergoes oxidation with the slowest rate in contrast to its high global softness, however it is weakly bound to the TiO2 surface. Consequently, the fact that the global DFT descriptors do not describe the photocatalytic reactivities of the diNN isomers is mainly
Fig. 5. Adsorption model for 1,3-dinitronaphthalene.
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
related with their different adsorptive capacities, which can be explained by the binding energies. 4.6. Local DFT-based reactivity descriptors Since the carbon atom with the highest fi is the site where the electrophilic attack will occur, the calculated fi values are used in the determination of the point of attack of the OH radical for each of the diNN molecules studied. Then, the highest fi values were used to calculate the local softnesses. The calculated local softness and Fukui functions are presented in Table 1. The local softness s and the Fukui function f governing the electrophilic attack are in the same order as the experimental one 1,3-diNN > 1,8-diNN > 1,5diNN, whereas f o and so values show a different trend. This finding indicates that the reactions of the diNN isomers with the photogenerated OH radicals are orbital-controlled reactions rather than radical attack, which is in agreement with our former prediction explained in Section 4.3. In order to obtain the best reactivity descriptor and the nature of the photocatalytic oxidation reactions of the diNN isomers, ‘‘softness-matching principle” (Vos et al., 2002) was used. This principle is based on the HSAB principle. According to the local HSAB principle, soft atoms react preferentially with other soft atoms and hard atoms with other hard atoms. The reactions investigated in this study are orbital-controlled reactions in which softsoft interactions dominate. For these reactions, Ds between the reacting atoms must be as small as possible. Therefore, Ds ¼ sþ ðOÞ s ðCÞ, the difference between the oxygen atom of the attacking OH radical and the carbon atom of the naphthalene ring that undergoes the electrophilic attack was calculated for each of the diNN molecules. The calculated Ds values agree well with the experimental oxidation rates. Therefore, it may be concluded that local DFT descriptors describe the reactivities of the diNN molecules in their photocatalytic oxidation reactions better than the global ones. 5. Conclusions The principal conclusions of the present study can be summarized as follows: 1. The photocatalytic oxidation of dinitronaphthalenes can be described by the simple first order kinetic model. The experimentally determined photocatalytic oxidation rates of the three diNN isomers are in the order 1,3-diNN > 1,8-diNN > 1,5-diNN. 2. The release of organic carbon from the surface in addition to the presence of the formed organic products revealed a considerable amount of the organic carbon in solution. 3. Electronic properties of the dinitronaphthalenes affect their photocatalytic reactivities in aqueous TiO2 suspensions. 4. The photocatalytic oxidation reactions of dinitronaphthalenes are orbital-controlled and electrophilic in nature. 5. The DFT-based reactivity descriptors can well describe the mechanisms of the reactions and the reactivities of the dinitronaphthalene isomers investigated. 6. Local DFT descriptors reflect the reactivities of the dinitronaphthalene isomers better than the global ones, due to the differences in their adsorptive capacities. The local softness values showing the same order as the reaction rates are the best reactivity descriptors to be used for these reactions. 7. The softness difference between the oxygen atom of the attacking OH radical and the carbon atom of the naphthalene ring that undergoes the electrophilic attack indicates that soft–soft interactions are more dominant than radical attack for the photocatalytic oxidation reactions of dinitronaphthalenes.
1013
References Bahnemann, D., Cunningham, J., Fox, M.A., Pelizzetti, E., Pichat, P., Serpone, N., 1994. In: Helz, G.R., Zepp, R.G., Crosby, D.G. (Eds.), Aquatic and Surface Photochemistry. Lewis, Baco Raton, FL, p. 261. Bausinger, T., Dehner, U., Preuß, J., 2004. Determination of mono-, di- and trinitronaphthalanes in soil samples contaminated by explosives. Chemosphere 57, 821–829. Becke, A.D., 1993. Density-functional thermochemistry. 3. The role of exact exchange. J. Chem. Phys. 98, 75648–75652. Chermette, H., 1999. Chemical reactivity indexes in density functional theory. J. Comput. Chem. 20, 129–154. Eberhardt, M.K., Yoshida, M., 1973. Radiation-induced homolytic aromatic substitution. I. Hydroxylation of nitrobenzene, chlorobenzene, and toluene. J. Phys. Chem. 77 (5), 589–597. Feilberg, A., Kamens, R.M., Strommen, M.R., Nielsen, T., 1999. Photochemistry and partitioning of semivolatile nitro-PAH in the atmosphere. Polycycl. Aromat. Comp. 14, 151–160. Foresman, J.B., Frisch, A., 1996. Exploring Chemistry with Electronic Structure Methods, second ed. Gaussian, Inc. Pittsburgh, PA. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Montgomery Jr., J.A., Vreven, T., Kudin, K.N., Burant, J.C., Millam, J.M., Iyengar, S.S., Tomasi, J., Barone, V., Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G.A., Nakatsuji, H., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Klene, M., Li, X., Knox, J.E., Hratchian, H.P., Cross, J.B., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Ayala, P.Y., Morokuma, K., Voth, G.A., Salvador, P., Dannenberg, J.J., Zakrzewski, V.G., Dapprich, S., Daniels, A.D., Strain, M.C., Farkas, O., Malick, D.K., Rabuck, A.D., Raghavachari, K., Foresman, J.B., Ortiz, J.V., Cui, Q., Baboul, A.G., Clifford, S., Cioslowski, J., Stefanov, B.B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Martin, R.L., Fox, D.J., Keith, T., Al-Laham, M.A., Peng, C.Y., Nanayakkara, A., Challacombe, M., Gill, P.M.W., Johnson, B., Chen, W., Wong, M.W., Gonzalez, C., Pople, J.A., 2003. Gaussian 03, B01. Gaussian Inc., Pittsburgh, PA. Geerlings, P., De Proft, F., Langenaeker, W., 2003. Conceptual density functional theory. Chem. Rev. 103, 1793–1873. Gerasimov, G., 2007. Modeling study of electron-beam polycyclic and nitropolycyclic aromatic hydrocarbons treatment. Radiat. Phys. Chem. 76, 27–36. Gramatica, P., Pilutti, P., Papa, E., 2007. Approaches for externally validated QSAR modeling of nitrated polycyclic aromatic hydrocarbon mutagenicity. SAR QSAR Environ. Res. 18 (1-2), 169–178. Hemelsoet, K., Speybroeck, V.V., Marin, G.B., De Proft, F., Geerlings, P., Waroquier, M., 2004. Reactivity indices for radical reactions involving polyaromatics. J. Phys. Chem. A 108, 7281–7290. Hemelsoet, K., Speybroeck, V.V., Waroquier, M., 2007. How useful are reactivity indicators for the description of hydrogen abstraction reactions on polycyclic aromatic hydrocarbons? Chem. Phys. Lett. 444, 17–22. Hien, T.T., Thanh, L.T., Kameda, T., Takenaka, N., Bandow, H., 2007. Nitro-polycyclic aromatic hydrocarbons and polycyclic aromatic hydrocarbons in particulate matter in an urban area of a tropical region: Ho Chi Minh City, Vietnam. Atmos. Environ. 41, 7715–7725. Jonker, M.T.O., 2008. Absorption of polycyclic aromatic hydrocarbons to cellulose. Chemosphere 70, 778–782. Lee, C., Yang, W., Parr, R.G., 1988. Development of the Colle–Salvetti correlationenergy formula into a functional of the electron-density. Phys. Rev. B 37 (2), 785–789. Maurino, V., Minero, C., Pelizzetti, E., Piccinini, P., Serpone, N., Hidaka, H., 1997. The fate of organic nitrogen under photocatalyticconditions: degradation of nitrophenols and aminophenols on irradiated TiO2. J. Photochem. Photobiol. A 109, 171–176. Mills, A., Hunte, S.L., 1997. An overview of semiconductor photocatalysis. J. Photochem. Photobiol. A 108, 1–35. Ohura, T., Amagai, T., Makino, M., 2008. Behavior and prediction of photochemical degradation of chlorinated polycyclic aromatic hydrocarbons in cyclohexane. Chemosphere 70 (11), 2110–2117. Pelizzetti, E., Minero, C., Maurino, V., Sclafani, A., Hidaka, H., Serpone, N., 1989. Photocatalytic degradation of nonylphenol ethoxylated surfactants. Environ. Sci. Technol. 23, 1380–1385. Pelizzetti, E., Minero, C., 1999. Role of oxidative and reductive pathways in the photocatalytic degradation of organic compounds. Colloid. Surface A 151, 321– 327. Piccinini, P., Minero, C., Vincenti, M., Pelizzetti, E., 1997a. Photocatalytic interconversion of nitrogen-containing benzene derivatives. J. Chem. Soc. Faraday Trans. 93, 1993–2000. Piccinini, P., Minero, C., Vincenti, M., Pelizzetti, E., 1997b. Photocatalytic mineralization of nitrogen-containing benzene derivatives. Catal. Today 39, 187–195. Pichat, P., 1997. In: Ertl, G., Knozinger, H., Weitkamp, J. (Eds.), Handbook of Heterogeneous Photo-Catalysis, vol. 4. VCH, Weinheim, p. 2111. Shemer, H., Linden, K.G., 2007a. Aqueous photodegradation and toxicity of the polycyclic aromatic hydrocarbons fluorine, dibenzofuran and dibenzothiophene. Water Res. 41, 853–861. Shemer, H., Linden, K.G., 2007b. Photolysis, oxidation and subsequent toxicity of a mixture of polycyclic aromatic hydrocarbons in natural waters. J. Photochem. Photobiol. A 187, 186–195.
1014
M. Bekbolet et al. / Chemosphere 75 (2009) 1008–1014
Stewart, J.P., 1989. Optimization of parameters for semiempirical methods II. Applications. J. Comput. Chem. 10, 221–264. Torrent-Sucarrat, M., De Proft, F., Geerlings, P., 2005. Stiffness and Raman intensity: a conceptual and computational DFT study. J. Phys. Chem. A 109, 6071–6076. Tunesi, S., Anderson, M., 1991. Influence of chemisorption on the photodecomposition of salicylic acid and related compounds using suspended titania ceramic membranes. J. Phys. Chem.-US 95 (8), 3399–3405. Turchi, C.S., Ollis, D.F., 1990. Photocatalytic degradation of organic water contaminants: mechanisms involving hydroxyl radical attack. J. Catal., 122– 178.
Vos, A.M., Nulens, K.H.L., De Proft, F., Schoonheydt, R.A., Geerlings, P., 2002. Reactivity descriptors and rate constants for electrophilic aromatic substitution: acid zeolite catalyzed methylation of benzene and toluene. J. Phys. Chem. B 106, 2026–2034. Waki, K., Zhao, J., Horikoshi, S., Watanabe, N., Hidaka, H., 2000. Photooxidation mechanism of nitrogen-containing compounds at TiO2/H2 O interfaces: an experimental and theoretical examination of hydrazine derivatives. Chemosphere 41, 337–343.