Photo-oxidative degradation of toluene in aqueous media by hydroxyl radicals

Article abstract of DOI:10.1016/j.jphotochem.2010.07.021

This study has the goal of determining the most probable reaction path and the product distribution for the photo-oxidative degradation of toluene in aqueous media. A combination of experimental and quantum mechanical methods were used to elucidate the effect of water solvent on the reaction rate and on the subsequent formation of the primary intermediates. In the experimental part of the study, the formation yields of hydroxylated intermediates and of benzaldehyde were measured by HPLC, in the presence of nitrate under UVB irradiation as the OH source. Modeling of the reaction paths was performed with density functional theory (DFT) calculations, to investigate the most plausible mechanism for the initial OH attack and to determine the identities of the primary intermediates. Rate coefficients for all the reaction paths were computed by the Transition State Theory (TST) to obtain the product distribution. The effect of solvent water was investigated by using COSMO as the solvation model. The experimental results combined with DFT calculations indicate that ortho-addition to finally give o-cresol is the dominant reaction path for gas and aqueous media. The presence of a dielectric medium such as water has a stabilizing effect that decreases the overall energy for this mechanism. Finally, the significance for surface waters of the reaction between toluene and OH was studied by use of a recently developed photochemical model that foresees the lifetime of a compound upon reaction with OH, as a function of the reaction rate constant, the chemical composition of the surface water layer, and the water column depth.

Full text of DOI:10.1016/j.jphotochem.2010.07.021

Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Photo-oxidative degradation of toluene in aqueous media by hydroxyl radicals
Arzu Hatipoglua, Davide Vioneb, Yelda Yalc¸ ına, Claudio Minerob, Zekiye C¸ ınara,∗
a Yıldız Technical University, Department of Chemistry, 34220 Istanbul, Turkey
b Università di Torino, Dipartimento di Chimica Analitica, Via P. Giuria 5, 10125 Torino, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 May 2010
Received in revised form 24 June 2010
Accepted 20 July 2010
Available online 29 July 2010
This study has the goal of determining the most probable reaction path and the product distribution
for the photo-oxidative degradation of toluene in aqueous media. A combination of experimental and
quantum mechanical methods were used to elucidate the effect of water solvent on the reaction rate
and on the subsequent formation of the primary intermediates. In the experimental part of the study,
the formation yields of hydroxylated intermediates and of benzaldehyde were measured by HPLC, in
the presence of nitrate under UVB irradiation as the •OH source. Modeling of the reaction paths was
performed with density functional theory (DFT) calculations, to investigate the most plausible mechanism
for the initial •OH attack and to determine the identities of the primary intermediates. Rate coefficients
for all the reaction paths were computed by the Transition State Theory (TST) to obtain the product
distribution. The effect of solvent water was investigated by using COSMO as the solvation model. The
experimental results combined with DFT calculations indicate that ortho-addition to finally give o-cresol
is the dominant reaction path for gas and aqueous media. The presence of a dielectric medium such as
water has a stabilizing effect that decreases the overall energy for this mechanism. Finally, the significance
for surface waters of the reaction between toluene and •OH was studied by use of a recently developed
photochemical model that foresees the lifetime of a compound upon reaction with •OH, as a function of
the reaction rate constant, the chemical composition of the surface water layer, and the water column
depth.
Keywords:
Toluene
Hydroxyl radical
Photo-oxidative degradation
DFT calculation
COSMO
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
reach some 10 ppbv levels in polluted air [6] and even 1 ppmv in or
near refueling stations [7]. The most important removal pathway of
Volatile aromatic compounds constitute an important class of
water and air contaminants. They are mainly emitted into the
environment from anthropogenic sources such as combustion pro-
cesses, vehicle emissions and industrial sources, as well as from
biogenic processes [1]. Recent calculations have shown that in
urban areas up to 40% of photochemically produced ozone can be
attributed to emissions of aromatics [2]. Furthermore, the degrada-
tion reactions of aromatic hydrocarbons yield non-volatile organic
compounds that contribute to the formation of secondary organic
aerosols [1]. The latter have serious effects on human health and
the global climate [3]. The cited atmospheric reactions are also
involved in the formation of polycyclic aromatic hydrocarbons, par-
ticulate matter and soot [4,5]. Therefore, the oxidative degradation
mechanism of aromatic compounds is a current hot research topic.
Toluene is the most abundant volatile aromatic hydrocarbon
present in the urban atmosphere, being a rather common con-
stituent of crude oil and a major component of gasoline. It can
toluene from the atmosphere is the gas-phase reaction with •OH,
which mainly yields the hydroxylated derivatives (cresols) [1,8].
The latter are 7–11 times more reactive than toluene toward •OH
and, also being more water-soluble, can be more easily removed
from the atmosphere [8].
Toluene is also present in surface and groundwater, usually
because of fuel spills or petroleum production [9–15]. The con-
centration of toluene in surface waters is usually a fraction of
␮g L−1, but values as high as ∼10 ␮g L−1 have been described [14].
As a volatile organic compound, toluene can undergo partition-
ing between surface waters and the atmosphere, but additional
processes are operational that account for its transformation [16],
e.g. biodegradation [15]. Photochemistry is another important
transformation route for water-dissolved pollutants [17], and the
reaction with the hydroxyl radical (•OH) can play a key role in
the photosensitized degradation of hardly oxidized aromatic com-
pounds [18]. In recent years, advanced oxidation processes (AOPs)
are being used in order to destroy toluene and other aromatic
hydrocarbons in water or air. In all AOPs, including UV/ozone,
UV/hydrogen peroxide and near-UV/titanium dioxide, •OH radicals
are generated by irradiation. These radicals initiate the degradation
of all the aromatic hydrocarbons, but the transformation of pollu-
∗
Corresponding author. Tel.: +90 212 383 4179; fax: +90 212 383 4134.
E-mail address: cinarz@yildiz.edu.tr (Z. C¸ ınar).
1010-6030/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jphotochem.2010.07.021

60
A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
tants may yield harmful intermediates that are more hazardous
than the parent compounds. Therefore, knowledge on the inter-
mediates and the reaction products is a necessity in the application
of water and air treatment processes. Although the toluene + •OH
reaction has great importance in various fields of chemistry such as
environmental chemistry, atmospheric chemistry and combustion
chemistry, its degradation mechanism and product distribution is
still uncertain.
The most important experimental result reported in the liter-
ature is that at room temperature the activation energy for the
toluene + •OH reaction is positive, whereas at high temperature it is
negative [3]. The curved Arrhenius plots for this reaction have been
attributed to the competition between different possible reaction
paths. The reaction of toluene with •OH radicals proceeds by three
pathways: H-atom abstraction either from the side-chain or from
the ring and addition to the aromatic ring. In the addition reactions,
hydroxymethylcyclohexadienyl radical is formed. Of the hydroxy-
lated adducts, o-cresol has been reported to be the most abundant
[19]. On the other hand, in the H-abstraction paths from the methyl
group and the ring, a benzyl and a methylphenyl radical are formed,
respectively. The benzyl radical yields benzaldehyde through the
reaction with O2.
This paper has the purpose of determining the most probable
reaction paths and the product distribution for the photo-oxidative
degradation of toluene in aqueous media, to elucidate the effect
of the water solvent on the reaction rate and on the subse-
quent formation of the primary intermediates. The experimental
formation yields of the hydroxylated intermediates and of ben-
zaldehyde were measured, and the most plausible reaction paths
were modeled with density functional theory (DFT) calculations.
These results were coupled with a recently developed photochem-
ical model for surface waters, in order to assess the importance of
the toluene + •OH reaction in the aqueous environment [20,21].
Fig. 1. (a) Emission spectrum of the UVB lamp (Philips TL 01) adopted for the irra-
diation experiments. Molar absorption coefficient of nitrate. (b) Time evolution of
1.4 mM toluene and of the detected transformation intermediates (cresols and ben-
zaldehyde) upon UVB irradiation of 0.10 M NaNO3, pH ∼ 6. The time evolution curve
of toluene is plotted against the left-hand Y-axis, while those of the intermedi-
ates (cresols and benzaldehyde) are plotted against the right-hand Y-axis (note the
different scales). The arrows point toward the axis relevant to each curve.
2. Experimental details
2.1. Reagents and materials
account. The Fig. also reports the absorption spectrum of nitrate,
measured with a Varian Cary 100 Scan UV–vis spectrophotometer.
Toluene (for residue analysis) was purchased from Fluka, ben-
zaldehyde (>99.5%) and H3PO4 (85%) from Aldrich, o-cresol (99%),
p-cresol (99%) and benzoic acid (98%) from Carlo Erba, m-cresol
(98%) and NaNO3 (99%) from Merck, acetonitrile (Supergradi-
ent grade) from Scharlau. All reagents were used as received,
without further purification. Water used was of Milli-Q quality
(>18 Mꢀ cm−1, TOC ≈ 2 ppb).
2.3. Analytical determinations
After irradiation, the solutions were analyzed with a Merck-
Hitachi High-Pressure Liquid Chromatograph equipped with
AS2000A autosampler, L-6200 and L-6000 pumps for high-pressure
gradients, L-4200 UV–vis detector, and a RP-C18 LichroCART
column (Merck, length 125 mm, diameter 4 mm) packed with
LiChrospher 100 RP-18 (5 ␮m diameter). Elution was carried
out with a 45:55 mixture of acetonitrile: aqueous H3PO4 (pH
3), at a flow rate of 1.0 mL min−1. Injection volume was set at
100 ␮L, detection wavelength at 210 nm. Under these conditions
the retention times were (min): benzoic acid (2.60), m- and p-
cresol (3.65), o-cresol (4.00), benzaldehyde (4.15), toluene (14.40).
The column dead time was 0.90 min. The perfect co-elution of
m- and p-cresol caused some problems, but they showed similar
slopes in the calibration curve. For this reason, the sum of the
concentrations of m- and p-cresol is reported in the parts that
follow.
2.2. Irradiation experiments
Solutions to be irradiated (5 mL total volume), containing 1 mM
toluene and 0.1 M NaNO3 (pH ≈ 6.5, measured with a Metrohm
713 pH meter equipped with a combined glass electrode), were
placed into cylindrical Pyrex glass cells (4.0 cm diameter, 2.3 cm
height). The solutions were magnetically stirred during irra-
diation, which was carried out under a 100 W Philips TL 01
lamp with emission maximum at 313 nm. The lamp had an
irradiance of 5.0 0.2 W m−2, measured with a CO.FO.ME.GRA.
(Milan, Italy) power meter and corresponding to a photon flux of
(3.3 0.1) × 10−6 Einstein L−1 s−1 in the solutions, actinometrically
determined with the ferrioxalate method [22]. Nitrate under UVB
irradiation was used as the •OH source to induce the transforma-
tion of toluene [23]; the lamp was chosen to achieve excitation of
nitrate that has an absorption maximum at around 305 nm [17].
Fig. 1(a) reports the lamp emission spectrum, taken with an Ocean
Optics SD 2000 CCD spectrophotometer and normalized to the acti-
nometry data, also taking the transmittance of the Pyrex glass into
2.4. Kinetic data treatment
The concentration vs. time data of toluene and its transforma-
tion intermediates are the average results of a run carried out in
triplicate. The time evolution data of toluene were fitted with Eq.

A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
61
(1):
Ground-state and transition-state structures were confirmed by
frequency analyses at the same level. Transition structures were
characterized by having one imaginary frequency that belonged
to the reaction coordinate, corresponding to a first-order saddle
point. Zero-point vibrational energies (ZPEs) were calculated at the
B3LYP/6-31G(d) level. The same ZPEs were used for the B3LYP/6-
311 + G(d,p)//B3LYP/6-31G(d) calculations. IRC calculations were
performed for all of the transition geometries, and the correspond-
ing maxima were confirmed.
Ct = Co exp(−ktdol t)
(1)
where Ct is the concentration of toluene at the time t, Co its ini-
tial concentration, and ktdol the pseudo-first order degradation rate
constant. The initial transformation rate of toluene is Rtol = ktdolCo.
The time evolution of each transformation intermediate was fitted
with Eq. (2):
kifntCo[exp(−ktdolt) − exp(−kidntt)]
Cint,t
=
(2)
kidnt − ktdol
3.3. Solvent effect model
where Co, kd and t are as above, Cint,t is the concentration of the
In aqueous media, water molecules affect the energetics of
the degradation reactions of all organic compounds. Moreover,
H2O induces geometry relaxation on the solutes. The latter effect
becomes more important when hydrogen-bonded complexes are
present. However, the results obtained in earlier studies indicate
that geometry changes have a negligible effect on the energy of
the solute in water for both open- and closed-shell structures
[27,28]. In this study, to take into account the effect of H2O on
the energetics and the kinetics of the toluene + •OH reactions,
DFT/B3LYP/6-311 + G(d,p) calculations were carried out for the
optimized structures of the reactants, the pre-reactive and the tran-
sition state complexes and the product radicals, by using COSMO
(conductor-like screening solvation model) [28] as the solvation
model, implemented in the Gaussian 03 package. The solvent was
water at 25 ◦C, with dielectric constant ε = 78.39 [29].
tol
transformation intermediate at the time t, and kifnt, kidnt are the
pseudo-first order formation and transformation rate constants of
the intermediate, respectively. The initial formation rate of each
intermediate is Rint = kifntCo, which is also the slope of the tan-
gent to the relevant time evolution curve (Cint,t vs. t) at t → 0. The
errors associated to the rates (ꢁ ꢂ) represent the scatter of the
experimental data around the fitting curve. The yields of the dif-
ferent intermediates were derived from the ratio of their initial
rates to that of toluene, as ꢃint = RintRt−o1l . In the yield calculations
we adopted the initial formation rates of the primary intermedi-
ates and of toluene (t → 0), because they are affected to a limited
or negligible extent by the subsequent transformation of the inter-
mediates by •OH and/or direct photochemistry.
COSMO is one of the polarizable continuum methods (PCMs).
In PCMs, the solute molecule is placed in a cavity surrounded by a
polarizable continuum, whose reaction field modifies the energy
and the properties of the solute [30]. The geometry of the cav-
ity is determined by the shape of the solute. The reaction field
is described in terms of apparent polarization charges or reaction
field factors included in the solute Hamiltonian, so that it is possi-
ble to perform iterative procedures leading to the self-consistence
between the solute wave-function and the solvent polarization.
The COSMO method describes the solvent reaction field by means
of apparent polarization charges distributed on the cavity sur-
face, which are determined by imposing that the total electrostatic
potential cancels out at the surface. This condition can describe the
solvation in polar liquids. Hence, it is the method of choice in this
study.
3. Computational set-up and methodology
3.1. Computational models
The molecular models were created by using the mean bond dis-
tances, the geometric parameters of the benzene ring, tetrahedral
angles for sp3-hybridized carbon and oxygen atoms, and 120◦ for
sp2-hybridized carbon atoms. In the calculation of the hydroxylated
radicals, the aromatic ring was left planar except for the position of
attack. The attacking •OH radical was assumed to form a tetrahe-
dral angle with the C–H bond due to the change in the hybridization
state of the carbon at the addition center from sp2 to sp3.
3.2. Methodology
The reaction system under consideration consists of •OH radi-
cals that are open-shell species. It is well-known that open-shell
molecules pose severe problems in quantum mechanical calcu-
lations. Therefore, geometry optimization of the reactants, the
product radicals, pre-reactive and transition state complexes were
performed with the DFT method within the Gaussian 03 pack-
age [24]. DFT methods use the exact electron density to calculate
molecular properties and energies, taking electron correlation into
account. They do not suffer from spin contamination and this fea-
ture makes them suitable for calculations involving open-shell
systems. The DFT calculations were carried out by the hybrid B3LYP
functional, which combines HF and Becke exchange terms with the
Lee–Yang–Parr correlation functional.
Choice of the basis set is very important in such calculations. In
previous studies, radical addition reactions have been studied to
determine the level of theory necessary for predicting energy bar-
riers, and B3LYP/6-311 + G(d,p)//B3LYP/6-31G(d) has been found
to be the most suitable [25,26]. Based on these results, optimiza-
tions in the present study were performed at the B3LYP/6-31G(d)
level followed by single point energy calculations at the B3LYP/6-
311 + G(d,p) level. The forming C–O bonds in the addition paths
and the H–O bond in the abstraction path were chosen as the
reaction coordinates in the determination of the transition states.
4. Results and discussion
4.1. Photo-oxidation upon nitrate irradiation
Fig. 1(b) reports the time evolution of 1.4 mM toluene and of its
identified transformation intermediates (o- and m- + p-cresol, ben-
zaldehyde), upon UVB irradiation of 0.10 M NaNO3. Interestingly,
no formation of benzoic acid took place in the studied system at
the adopted irradiation time scale (up to 4 h). The relevant values
of Rtol, Rint and ꢃint are presented in Table 1. Note that negligible
direct phototransformation of toluene was observed in the absence
of nitrate, under otherwise identical conditions.
Under the approximation of monochromatic irradia-
tion that is reasonable for the 313-nm line of the adopted
lamp, the photon−A flux absorbed by nitrate would be
-
3
NO
PaNO3− = P◦ × (1 − 10
),
where
ANO3− = εNO3−
b
[NO3−].
At 313 nm (the lamp emission maximum in the UVB) it is
εNO3− = 5.7 M−1 cm−1. It is also b = 0.40 cm, [NO3−] = 0.10 M and
P◦ = (3.3 0.1) × 10−6 Einstein−L−1 1 s−1. From these data one gets
PaNO3− = 1.3 × 10−6 Einstein L s−1
.
The observed degrada-
tion rate of toluene was Rtol = (9.85 0.79) × 10−9 M s−1 (see
Table 1). A lower limit for the quantum yield of •OH production

62
A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
Table 1
mined. The first three paths, ortho-addition (o-add), meta-addition
(m-add) and para-addition (p-add) to the ring are shown below:
Initial degradation rate of 1.4 mM toluene and initial formation rates of the detected
intermediates.
Initial rate (M s−1
)
Yields (%)
(3)
(4)
(9.85 0.79) × 10−9
94.5 19.4a
(1.67 0.19) × 10−9
17.0 3.3
(3.00 0.35) × 10−9
30.5 6.0
(5)
The •OH radical attacks a ring carbon with its unpaired electron and
upon contact forms a C–O bond, while a ␲–bond of the aromatic
system is broken and a hydroxymethylcyclohexadienyl radical is
formed. The possibility of ipso-addition to the methyl-substituted
site has been determined to be of minor importance [3]. The fourth
reaction path:
(4.63 0.62) × 10−9
47.0 10.1
a
The yield shown for toluene is the sum of the yields of the intermediates and of
the associated errors.
−1
−1
−
−
−
upon nitrate photolysis, ꢄNO = R OH(PNO
)
≥ Rtol(PNO
)
=
3
3
3
•
•
a
a
OH
(7.3 0.6) × 10−3, can be obtained under the hypothesis that all
(6)
photogenerated •OH reacts with toluene. Such a result compares
−
consists in H-abstraction (H-abs) from the methyl group to produce
a benzyl radical and a water molecule. The process of H-abstraction
by the •OH radical is a simple atom-transfer reaction in which
the bond to the carbon atom in the –CH3 group is broken and a
new bond to the oxygen atom of the •OH radical is formed. It has
been suggested that ring abstraction paths are negligible at room
temperature due to their larger energy barriers than side-chain
abstraction paths [34,35].
relatively well with the literature value ꢄNO ≈ 9 × 10−3 [23].
3
•
OH
All the detected intermediates are compatible with a reaction
between toluene and •OH. The radical would mainly attack the
aromatic ring of toluene to produce the cresols (hydroxymethyl-
benzenes) with a total yield of (78 16)%, while benzaldehyde
would derive from the attack of •OH on the methyl group, with
a yield of about 17 3%. Evidence that •OH is involved in the for-
mation of benzaldehyde from toluene has been obtained both in
the gas phase [8] and in aqueous solution [31].
4.2.2. Reactants and product radicals
To make a comparison, in the gas phase the benzaldehyde yield
from toluene + •OH is reported to be 5–10%, while the cresols are
formed in 20–30% yield (with o-cresol strongly prevailing over the
m- and the p-isomers). Other intermediates detected in the gas
phase include benzyl nitrate and m-nitrotoluene, but their for-
mation would mainly be a consequence of the atmospherically
relevant conditions chosen in the experiments (e.g. presence of •NO
and significant gas-phase nitration with •OH + •NO2) [8]. If only the
common intermediates of the reactions in the gas phase and in
water are considered, it can be seen that in both cases the •OH
attack on the ring prevails over that on the methyl group.
Four different radicals were determined as the products of the
reaction between toluene and the •OH radical. The electronic and
thermodynamic properties for the most stable conformers were
also calculated. Fig. 2 represents the potential energy profile of the
DFT-modeled mechanism for the initial attack of the •OH radical to
toluene.
The total energies for each of the reactants and the product rad-
icals obtained for both gas and aqueous media (Fig. 2) indicate that
the radicals, o-R, m-R and p-R produced in the addition paths are
more stable than the H-R produced in the abstraction path. The
reason may be attributed to the fact that all the radicals produced
in the addition paths have hydrogen-bond-like stabilizations. The
interaction distance between the original hydrogen atom at the
addition center and the oxygen atom of the added •OH was calcu-
4.2. Computational modeling
˚
˚
4.2.1. Reaction paths
lated to be 1.997 A for m-R and p-R, while it is 1.999 A for o-R. For
the gas phase, among the three radicals produced in the addition
paths, o-R has the lowest energy and is, therefore, the most stable
one. The p-R is around 0.3 and m-R around 1.8 kcal mol−1 less sta-
ble than o-R. The energy of H-R is too high to be compared with
those of the addition radicals. For the aqueous phase, the sequence
remains the same. The most and the least stable radicals are again
the o-R and the H-R, respectively. However, all the product radi-
The hydroxyl radical is a very active species and has a strong
electrophilic character [32]. Once formed, it can readily attack the
toluene molecule and produce the reaction intermediates. •OH rad-
ical reactions with aromatic compounds proceed mainly by two
reaction pathways: H-atom abstraction from C–H bonds and addi-
tion to aromatic rings [33]. Based on previous results [3,4,19], four
different paths for the reaction of toluene with •OH were deter-

A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
63
Depending on the direction of approach of the •OH radicals, four
different chemically activated pre-reactive complexes, o-PC, m-PC,
p-PC and H-PC with shallow wells, ca. 3 kcal mol−1 were deter-
mined for both gas and aqueous phases. The optimized structures
of the pre-reactive complexes are shown in Fig. 3. The •OH radi-
cal is seen to approach the aromatic ring from above, lying almost
parallel to the ring plane with a deviation of about 5–10◦. In the
addition complexes, the hydrogen atom was found to be facing
toward the center of the ring, while the ring carbons at the addition
centers were found to deviate from the ring plane by around 2◦. The
main difference in the optimized geometries occurs in the distance
between the oxygen atom of •OH and the ring carbon. m-PC has
˚
˚
much longer distance as compared with 1.9 A in o-PC and 1.4 A in p-
PC. Energetically the o-PC and p-PC are the most stable complexes
followed by H-PC, lying 3.27, 3.22 and 2.95 kcal mol−1 below the
isolated reactants, respectively. The least stable pre-reactive com-
plex is the m-PC, which is around 1.39 kcal mol−1 less stable than
o-PC. It is observed that the stability of the pre-reactive complexes
is higher in the aqueous than in the gas phase, because the pres-
ence of water molecules, hydrogen-bonded to the hydrogen atoms
of the complexes lowers the total energies. The reduction in energy
was calculated to be ca. 6.65 kcal mol−1 for the three addition com-
plexes and ca. 8.24 kcal mol−1 for the H-PC. Although the energy
gain of the reactants by complex formation is much less in water,
the newly formed hydrogen-bonds between the complex and the
water molecules make the structures more stable.
4.2.4. Transition state complexes
Four transition state complexes o-TS, m-TS, p-TS and H-TS,
one for each of the possible reaction paths were identified. The
optimized structures of the TS are displayed in Fig. 4. In all the
complexes the •OH radical is almost parallel to the ring plane, with
slight deviations of ca. 0.5◦ in p-TS and H-TS, and 7◦ in o-TS and
m-TS. The oxygen atom points to the carbon atom at the reaction
center, while the hydrogen atom points to the center of the aro-
matic ring. The hydrogen atom at the addition center rotates out of
the ring plane to the opposite side of •OH, by 13.9◦, 15.5◦ and 3.3◦
in o-TS, m-TS and p-TS, respectively.
Fig. 2. (a) The potential energy profile of the DFT-modeled mechanism for the
toluene + •OH reaction in gas phase. (b) The potential energy profile of the DFT-
modeled mechanism for the toluene + •OH reaction in aqueous phase.
cals are more stable in the aqueous phase, due to the newly formed
hydrogen-bonds between their H atoms and the O atoms of the
water molecules. The stabilities of the addition radicals increase by
around 8.6 kcal mol−1 in the aqueous solution. On the other hand,
the difference in the gas and aqueous-phase energies for H-R was
much lower (4.2 kcal mol−1). This is consistent with the fact that
H-R has fewer hydrogen atoms, which causes weaker hydrogen-
bonding compared to the hydroxylated adducts.
Adopting the localization approach of the Wheland’s approx-
imation [36], it may be predicted that •OH-addition is the most
probable reaction path and that the positions of attack of the •OH
radical are the ortho and para ones for both gas and aqueous phases.
As seen in Fig. 2, the products of the three addition paths lie ca.
15 kcal mol−1 below the reactants. It may be concluded that the
reactions will proceed to the corresponding products, once the
reactants are sufficiently close to each other. The thermodynami-
cally most favored product is 1-hydroxy-2-methylcyclohexadienyl
radical, followed by 1-hydroxy-4-methylcyclohexadienyl radical
for both gas and aqueous phases.
In the addition paths the major structural changes in the ring
geometry relative to the parent toluene molecule are all localized
around the carbon atom at the reaction center. The two C–C bonds
˚
connecting the addition center lengthen by 0.025–0.029 A, whereas
the change in the bond lengths of the subsequent bonds is lower,
˚
around 0.005 A. This finding indicates that the two C–C bonds con-
necting the addition center have single rather than double bond
character. The bonds away from the addition center have a more
pronounced double bond character. In the H-TS all the geometric
changes are located around the methyl group and the carbon atom
connecting it. Since a new bond is being formed between one of the
hydrogen atoms of the methyl group and the oxygen atom of the
•OH radical, H–C–H angles in the methyl group become narrower
by 4◦. There is also a slight elongation in the breaking bond, around
˚
0.2 A.
In the light of the potential energy profiles reported in Fig. 2,
4.2.3. Pre-reactive complexes
it may be suggested that the forming bond length is a sensitive
measure of the formation of the transition state complex along the
reaction coordinate. As displayed in Fig. 4, the addition complexes
Our calculations indicated the formation of weakly bound com-
plexes between the •OH radical and toluene as the precursors of
the reaction. Due to the existence of these pre-reactive van der
Waals complexes, the reaction paths under investigation proceed
over barriers that are lower in energy than the reactants. The pre-
reactive complexes exert strong influence over the kinetics of the
reactions by altering barrier heights and by affecting the energy
partitioning of the reaction products. Furthermore, the presence of
such complexes affects the reaction dynamics by spatially directing
the site of the reaction, either by steric direction or by providing a
low potential well that favors a reaction site.
˚
have much longer C–O bonds, 2.062, 2.032 and 2.431 A in o-TS, m-
TS and p-TS, respectively, as compared to the C–O bonds in the
˚
corresponding radicals, 1.45 A. This suggests that the addition com-
plexes are early transition states, whereas the abstraction complex
is formed late along the reaction coordinate. Among the three addi-
tion complexes the longest C–O bond belongs to the p-TS, while the
m-TS and o-TS have much shorter C–O bonds. This indicates that
they are formed late as compared to p-TS and that, therefore, p-TS
is the earliest transition state. Furthermore, the p-TS was found to

64
A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
Fig. 3. The optimized structures of the pre-reactive complexes.
have the lowest total energy among all the possible transition states
(Fig. 2), indicating that it is the most thermodynamically stable one.
The transition-state structures, p-TS, o-TS, H-TS and m-TS lie 3.23,
2.89, 1.88 and 1.46 kcal mol−1, respectively, below the reactants.
Thus, it may be concluded that in the photo-oxidative degradation
of toluene, p-TS occurs early along the reaction coordinate, giving
rise to a long C–O bond and a low total energy. As a consequence,
it is the most probable transition-state structure. It was observed
that the stability order changes in the aqueous phase, where the
H-TS has lower energy than o-TS. However, the stability of all the
transition-state structures increases due to weak interactions with
the water molecules. Although the energies of all the transition
state complexes decrease by around 7 kcal mol−1, the p-TS is the
most thermodynamically stable TS for the aqueous phase as well.
the aqueous phase, the barrier for H-abstraction also decreases by
ca. 0.06 kcal mol−1. The reason may be attributed to the additional
stabilizing effect of the newly formed hydrogen-bonds with the
water molecules. Moreover, para- and ortho-addition paths were
found to have the highest exothermicities in the gas phase, whereas
the ortho-addition path has the highest exothermicity in the aque-
ous phase. The result is consistent with Hammond’s postulate [37],
which states that early transition states have low energy barriers
and high exothermicities. So, it may be concluded that p-TS and
o-TS are the most probable transition states for gas and aqueous
phases, respectively.
4.2.6. Reaction rates and product distribution
The rate constant k for each reaction path was calculated by
using the Transition State Theory for 300 K. The classical rate con-
stant k in the Transition State Theory is given by Eq. (7):
4.2.5. Energetics of the reaction paths
kBT qTS
/RT
a
The activation energies Ea for the four possible reaction paths
were calculated and presented in Table 2 along with the reaction
energies ꢅEr, for both gas and aqueous phases. The values in Table 2
show that the activation energies for the three •OH-addition paths
are lower than for H-abstraction. The lowest activation energy in
the gas phase belongs to the para-addition path. The energy barrier
for the ortho-addition path is 0.37 kcal mol−1 higher than for para-
addition. In the aqueous phase, the stability order of the transition
state complexes changes in favor of ortho-addition. Although water
has a stabilizing effect in terms of total energies, the energy barri-
ers in aqueous solution for the para- and meta-addition paths are
slightly higher than the barriers in the gas phase. They increase by
around 0.49 and 0.13 kcal mol−1, respectively, because of the pre-
vention of the formation of the nice intramolecular hydrogen-bond
network observed in the two transition state complexes. In con-
trast, the activation energy for the ortho-addition path decreases
by around 0.12 kcal mol−1, causing the relevant reaction path to
have the lowest activation energy among all the possible ones. In
k =
e−E
(7)
h
qT qOH
where kB is Boltzmann’s constant, T is temperature, h is Planck’s
constant, q’s are molecular partition functions for TS and the reac-
tant species, toluene and •OH, and Ea is the activation energy. Each
of the molecular partition functions was assumed to be the prod-
uct of translational, rotational, vibrational and electronic partition
functions of the corresponding species.
The calculated rate constants in Table 2 show that in the gas
phase the highest rate constant belongs to para-addition, fol-
lowed by ortho- and meta-addition, whereas H-abstraction is the
slowest reaction path, consistent with its high energy barrier.
The branching ratio for each of the reaction paths was calcu-
lated by dividing the corresponding rate constant by the sum of
the rate constants, taking the number of similar addition cen-
ters into account. The overall rate constant for the gas phase
was calculated to be 3.79 × 10−13 cm3 molecule−1 s−1 at 300 K
which is in perfect agreement with the experimental value of

A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
65
Fig. 4. The optimized structures of the transition state complexes.
4.33 × 10−13 cm3 molecule−1 s−1 reported by Knispel et al. [38].
By using the branching ratios, the relative concentrations of the
primary intermediates were also calculated and are presented in
Table 2. The results indicate that ortho-addition is the prevailing
reaction path for both gas and aqueous phases, followed by para-
addition in the gas phase and by meta-addition in the aqueous
phase. Moreover, the calculated rate constants show that ortho-
addition and H-abstraction proceed faster in the aqueous compared
to the gas phase. In contrast, the rates of the remaining paths
decrease in aqueous solution, coherently with the increase in the
energy barriers due to the weakening effect of water molecules on
the original hydrogen-bonds in the pre-reactive complexes.
The obtained product distribution indicates that the major
primary intermediate of the photo-oxidative degradation of
toluene is the o-R (1-hydroxy-2-methylcyclohexadienyl radical)
which then forms o-cresol through abstraction of the redun-
dant ring
H by molecular oxygen. The reaction also yields
p-R (1-hydroxy-4-methylcyclohexadienyl radical) and m-R (1-
hydroxy-3-methylcyclohexadienyl radical), in lower amounts in
water compared to the gas phase. The relative concentration of
the benzyl radical that would then be converted to benzaldehyde
through the reaction with O2 is much less than all the other reaction
products. The formation of benzaldehyde could be more favored in
solution compared to the gas phase, but the solvent effect is rel-
Table 2
Activation energies Ea, reaction energies Er, rate constants and product distributions for the possible reaction paths.
o-Addition
m-Addition
p-Addition
H-Abstraction
Gas phase
Ea (kcal mol−1
)
0.38
−15.50
7.61 × 10−14
43.5
0.41
−13.73
3.43 × 10−14
19.6
0.01
−15.20
9.80 × 10−14
28.0
1.07
Er (kcal mol−1
)
Endothermic
3.11 × 10−14
8.9
k (cm3 molecule−1 s−1
)
)
Branching ratio (%)
Aqueous phase
Ea (kcal mol−1
)
0.26
0.54
0.50
1.01
Er (kcal mol−1
)
−16.16
−13.61
−13.65
Endothermic
3.41 × 10−14
7.95 × 106
10.7
k (cm3 molecule−1 s−1
9.34 × 10−14
5.62 × 107a
58.6
2.74 × 10−14
1.65 × 107
17.2
4.31 × 10−14
2.59 × 107
13.5
Branching ratio (%)
The unit for the values in italic is M−1 s−1
.
a

66
A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
atively small and it is lower than the errors of the experimental
data [8, and this work]. The computational results indicate that
the products of the photo-oxidative degradation of toluene are o-
cresol (o-CR), p-cresol (p-CR), m-cresol (m-CR) and benzaldehyde
(B), with [o-CR] > [p-CR] > [m-CR] > [B] in the gas phase. The concen-
tration order is [o-CR] > [m-CR] > [p-CR] > [B] in the aqueous phase,
which is consistent with the experimental findings of this study.
4.3. Environmental importance of toluene degradation
It is possible to foresee the half-life time of a water-dissolved
SSD
compound CP because of reaction with •OH ( CP, OH), as a function
•
of its second-order reaction rate constant with •OH, kCP, OH, the
•
chemical composition of the surface water layer (and in particular
the concentration values of the main sources and sinks of •OH),
and the water column depth d. Note that d would be the average
depth of thoroughly mixed water bodies or the mixing layer depth
of large, stratified lakes. The half-life time units are summer sunny
days (SSD), equivalent to a fair-weather 15 July at 45◦N latitude.
SSD
One gets
as follows [21]:
•
CP, OH
ꢀ
kSi[Si]
i
= 1.9 × 10−5
(8)
SSD
•
CP, OH
Rtot kCP, OH
•
•
OH
ꢁ
where
ikSi[Si] is the rate constant of the natural •OH scavengers
and Rtot is the formation rate of •OH, which depends on the con-
•
OH
centration of the main •OH sources (nitrate, nitrite and colored
dissolved organic matter) and on the depth d of the water column. In
the case of toluene it is kCP, OH = 3.0 × 109 M−1 s−1 [39]. A more com-
•
plete description of the adopted photochemical model is reported
as Supplementary Material (hereafter SM).
SSD
Fig. 5(a) reports
for toluene in the presence of 2.1 mM
•
bicarbonate and 26 ␮CMP, OcHarbonate, for d = 1 m and different con-
centration values of the dissolved organic matter (expressed as
NPOC, Non-Purgeable Organic Carbon) and nitrate/nitrite. Nitrate
was set to vary in the range of 1–30 ␮M, and it was hypothesized
that [NO3−]/[NO2−] = 100. Such a concentration ratio is reasonable
in surface waters [40,41]. The half-life time for reaction with •OH
is lower at higher nitrate and nitrite that are major •OH sources
(nitrite is also a very minor •OH sink, but its role as scavenger is
Fig. 5. (a) Half-life time of toluene in the presence of 2.1 mM bicarbonate and 26 ␮M
carbonate, for d = 1 m, as a function of [NO3−]/[NO2−] = 100 and of the NPOC. The
vertical line represents the estimated lifetime for biological degradation (1 to 6
months [16]). (b) Time trend of toluene and of its main transformation interme-
diates upon reaction with •OH, in the presence of 30 ␮M nitrate, 0.3 ␮M nitrite,
1 mg C L−1 NPOC, 2.1 mM bicarbonate and 26 ␮M carbonate, for d = 1 m, under the
hypothesis that all the compounds undergo transformation with •OH alone. The
time evolution curve of toluene is plotted against the left-hand Y-axis, while those
of the intermediates (cresols and benzaldehyde) are plotted against the right-hand
Y-axis (note the different scales). The arrows point toward the axis relevant to each
curve.
negligible compared to organic and inorganic carbon). Note the
SSD
minimum of
for NPOC ∼ 1 mg C L−1 in the presence of low
•
CP, OH
nitrate and nitrite. The reason for the minimum is that, under such
conditions, colored dissolved organic matter (CDOM) could prevail
over nitrate/nitrite as •OH source and, at the same time, carbonate
and bicarbonate would scavenge •OH to a comparable extent as
organic matter itself. Therefore, for NPOC < 1 mg C L−1, the produc-
tion of •OH by dissolved organic compounds would prevail over the
scavenging. At elevated NPOC, the general increase of the half-life
times is accounted for by the combination of •OH scavenging and
radiation screening that are carried out by dissolved organic matter.
Fig. 5(a) also reports the time interval that is hypothesized for the
biological degradation of toluene in surface waters (1–6 months;
[16]). It can be observed that the lifetime with •OH can be com-
parable to or even lower than that for the biological degradation,
provided that nitrate and nitrite are present in sufficiently high
concentration and that NPOC is not too high.
Fig. 5(b) reports the expected time trends of 10 nM toluene
(∼1 ␮g L−1), cresols and benzaldehyde in the presence of 30 ␮M
nitrate, 0.3 ␮M nitrite, 1 mg C L−1 NPOC, 2.1 mM bicarbonate and
26 ␮M carbonate, for d = 1 m, under the hypothesis that all the
compounds are formed and transformed upon reaction with •OH
(the formation yields of the intermediates as reported in Table 1
were considered). The reaction rate constants with •OH of o-
cresol, p-cresol and benzaldehyde are 1.1 × 1010, 1.2 × 1010 and
4.4 × 109 M−1 s−1, respectively [39]. The reaction rate constant of
m-cresol is not reported but, given the similar values for the other
two isomers, a reasonable approximation is to use the litera-
ture rate constant of p-cresol. Interestingly, all the transformation
intermediates would be more reactive than toluene toward •OH.
By application of the model equation (8) and considering that
•
kCdP = 0.693 (
)
−1, under the hypothesized conditions it is
CP, OH
tol, OH = 47 SSD (kd = 1.5 × 10−2SSD−1), o-cr, OH = 13 SSD (kd
=
•
•
tol
5.4 × 10−2SSD−1),
=
p-cr, OH = 12 SSD (km/p-cr = o5-.c9r ×
d
•
•
m-cr, OH
10−2SSD−1), and benzald, OH = 32 SSD (kd
= 2.2 × 10−2SSD−1).
•
benzald
The time trend of toluene in Fig. 5(b) was plotted with Eq. (1), that of
the intermediates with Eq. (2). It can be seen that the concentrations
of the intermediates would reach a maximum in approximately 50
SSD, and that the sum of their concentrations would never exceed
20% of the initial toluene. Note that benzaldehyde is the inter-
mediate formed in lower yield but it is the most stable toward
•OH. The outcome is that benzaldehyde would be the most concen-
trated intermediate at long transformation times. Also note that the
cresols could be transformed faster than foreseen by the adopted
model, because they also undergo important reactions with the
excited triplet states of CDOM [42].

A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
67
5. Conclusions
[10] D. Holdway, D.T. Heggie, Direct hydrocarbon detection of produced formation
water discharge on the Northwest Shelf, Australia, Estuar. Coast. Shelf Sci. 50
(2000) 387–402.
Hydroxyl radicals, photogenerated by UVB irradiation of nitrate
oxidize toluene by pseudo-first-order reaction rates. The experi-
mental findings indicate that •OH mainly attacks the aromatic ring
of toluene to produce the cresols (hydroxymethylbenzenes) with
a total yield of (78 16)%, and benzaldehyde through the abstrac-
tion of a hydrogen atom from the methyl group, with a yield of
about (17 3)%. Based on the quantum mechanical calculations,
four different reaction paths have been determined for the reac-
tion of toluene with •OH. DFT calculations indicate the formation of
pre-reactive complexes between the •OH radical and toluene as the
precursors to the reaction. The addition complexes are early tran-
sition states, whereas the abstraction complex is formed late along
the reaction coordinate. The activation energies for the addition
paths are lower than that for the abstraction path due to hydrogen-
bonding. The presence of a dielectric continuum such as water
increases the stability of all the species but it also alters the energy
barriers. In both the gas and aqueous phases, the •OH attack on
the ring prevails over that on the methyl group. Based on aque-
ous photochemistry modeling, it is suggested that the •OH reaction
could be a competitive sink for toluene in surface waters compared
to microbial processes, in nitrate/nitrite-rich and DOM-poor water
bodies.
[11] I. Arambarri, M. Lasa, R. Garcia, E. Millan, Determination of fuel dialkyl ethers
and BTEX in water using headspace solid-phase microextraction and gas
chromatography-flame ionization detection, J. Chromatogr. A 1033 (2004)
193–203.
[12] B. Karaconji, L. Skender, V. Karacic, Benzene, toluene, ethylbenzene, and iso-
meric xylenes in various water samples in Canada, Bull. Environ. Contam.
Toxicol. 76 (2006) 458–463.
[13] S.C. Soh, M.P. Abdullah, Determination of volatile organic compounds pollu-
tion sources in Malaysian drinking water using multivariate analysis, Environ.
Monit. Assess. 124 (2007) 39–50.
[14] D. Fatta, C. Michael, S. Canna-Michaehdou, M. Christoulidou, N. Kythreotou, M.
Vasquez, Pesticides, volatile and semivolatile organic compounds in the inland
surface waters of Cyprus, Desalination 215 (2007) 223–236.
[15] C. Robinson, A. Brovelli, D.A. Barry, L. Li, Tidal influence on BTEX biodegradation
in sandy coastal aquifers, Adv. Water Resour. 32 (2009) 16–28.
[16] P.C. Heald, S.G. Schladow, J.E. Reuter, B.C. Allen, Modeling MTBE and BTEX in
lakes and reservoirs used for recreational boating, Environ. Sci. Technol. 39
(2005) 1111–1118.
[17] J. Hoigné, Formulation and calibration of environmental reaction kinetics: oxi-
dations by aqueous photooxidants as an example, in: W. Stumm (Ed.), Aquatic
Chemical Kinetics, John Wiley and Sons, New York, 1990, pp. 43–70, Chapter 2.
[18] P.L. Brezonik, J. Fulkerson-Brekken, Nitrate-induced photolysis in natural
waters: controls on concentrations of hydroxyl radical photo-intermediates
by natural scavenging agents, Environ. Sci. Technol. 32 (1998) 3004–3010.
[19] V.H. Uc, I.G. Cruz, A.H. Laguna, A.V. Bunge, New channels in the reaction mech-
anism of the atmospheric oxidation of toluene, J. Phys. Chem. A 104 (2000)
7847–7855.
[20] D. Vione, S. Khanra, S. Cucu Man, P.R. Maddigapu, R. Das, C. Arsene, R.I. Olariu,
V. Maurino, C. Minero, Inhibition vs. enhancement of the nitrate-induced pho-
totransformation of organic substrates by the •OH scavengers bicarbonate and
carbonate, Water Res. 43 (2009) 4718–4728.
[21] D. Vione, R. Das, F. Rubertelli, V. Maurino, C. Minero, S. Barbati, S. Chiron, Mod-
elling the occurrence and reactivity of hydroxyl radicals in surface waters:
implications for the fate of selected pesticides, Int. J. Environ. Anal. Chem. 90
(2010) 258–273.
Interestingly, the •OH radical has been found to undergo accu-
mulation at the air–water interfaces compared to the solution bulk
[43]. Also toluene could be accumulated at the interface, in a similar
way as benzene [44], and the combination of the two phenom-
ena could greatly enhance the interface reaction compared to the
bulk. The overall effect would hardly be significant in surface water
bodies where the interface is a negligible fraction of the total vol-
ume. However, it could be important in atmospheric droplets that
have radii in the micrometer range and, therefore, possess a very
favorable surface-to-volume ratio [44,45].
[22] J.H. Kuhn, S.E. Braslavsky, R. Schmidt, Chemical actinometry, Pure Appl. Chem.
76 (2004) 2105–2146.
[23] J. Mack, J.R. Bolton, Photochemistry of nitrite and nitrate in aqueous solution:
a review, J. Photochem. Photobiol. A: Chem. 128 (1999) 1–13.
[24] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian,
J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J.
Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A.
Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels,
M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman,
J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham,
C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen,
M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision B.04, Gaussian, Inc.,
Pittsburgh, PA, 2003.
Acknowledgements
The authors express their thanks to Yıldız Technical University
Research Foundation (Project No: 24-01-02-15), PNRA-Progetto
Antartide and MIUR-PRIN 2007 (2007L8Y4NB, Area 02, Project No:
36) for financial support.
Appendix A. Supplementary data
[25] M.W. Wong, L. Radom, Radical additions to alkenes: further assessment of
theoretical procedures, J. Phys. Chem. A 102 (1998) 2237–2245.
[26] A.S. Ozen, V. Aviyente, R.A. Klein, Modeling the oxidative degradation of azo
dyes: a density functional theory study, J. Phys. Chem. A 107 (2003) 4898–4907.
[27] J. Andzelm, C. Kölmel, A. Klamt, Incorporation of solvent effects into density-
functional calculations of molecular-energies and geometries, J. Chem. Phys.
103 (1995) 9312–9320.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jphotochem.2010.07.021.
References
[28] V. Barone, M. Cossi, Quantum calculation of molecular energies and energy
gradients in solution by a conductor solvent model, J. Phys. Chem. A 102–11
(1998) 1995–2001.
[29] J.B. Foresman, Æ. Frisch, Exploring Chemistry with Electronic Structure Meth-
ods, Gaussian Inc., Pittsburgh, PA, 1996.
[30] N.S. Hush, J. Schamberger, G.B. Bacskay, A quantum chemical computational
study of the relative stabilities of cis- and transplatinum dichloride in aqueous
solution, Coord. Chem. Rev. 249 (2005) 299–311.
[31] X.Q. Wang, J.P. Wu, M.W. Zhao, Y.F. Lv, G.Y. Li, C.W. Hu, Partial oxidation of
toluene in CH3COOH by H2O2 in the presence of VO(acac)2 catalyst, J. Phys.
Chem. C 113 (2009) 14270–14278.
[32] V. Brezová, M. Ceppan, E. Brandsteterova, M. Breza, L. Lapcik, Photocatalytic
hydroxylation of benzoic acid in aqueous titanium dioxide suspension, J. Pho-
tochem. Photobiol. A: Chem. 59 (3) (1991) 385–391.
[33] M. Kılıc¸ , G. Kocturk, N. San, Z. C¸ ınar, A model for product distributions for the
reactions of phenol derivatives with hydroxyl radicals, Chemosphere 69 (9)
(2007) 1396–1408.
[34] F.P. Tully, A.R. Ravishankara, R.L. Thompson, J.M. Nicovich, R.C. Shah, N.M.
Kreutter, P.H. Wine, Kinetics of the reactions of hydroxyl radical with benzene
and toluene, J. Phys. Chem. 85 (1981) 2262–2269.
[35] V.H. Uc, J.R. Alvarez-Idaboy, A. Galano, I.G. Cruz, A.V. Bunge, Theoretical deter-
mination of the rate constant for oh hydrogen abstraction from toluene, J. Phys.
Chem. A 110 (2006) 10155–10162.
[1] B.J. Finlayson-Pitts, J.N. Pitts Jr., Chemistry of the Upper and Lower Atmosphere,
Academic Press, San Diego, 2000.
[2] B. Bohn, Formation of peroxy radicals from OH–toluene adducts and O2, J. Phys.
Chem. 105 (2001) 6092–6101.
[3] I. Suh, D. Zhang, R. Zhang, L.T. Molina, M.J. Molina, Theoretical study of OH
addition reaction to toluene, Chem. Phys. Lett. 364 (2002) 454–462.
[4] T. Seta, M. Nakajima, A. Miyoshi, High-temperature reactions of OH
radicals with benzene and toluene, J. Phys. Chem.
5081–5090.
[5] M. Bekbolet, Z. C¸ ınar, M. Kılıc¸ , C.S. Uyguner, C. Minero, E. Pelizzetti, Photocat-
alytic oxidation of dinitronaphthalenes: theory and experiment, Chemosphere
75–8 (2009) 1008–1014.
[6] P. Bruno, M. Caselli, G. De Gennaro, L. Scolletta, L. Trizio, M. Tutino, Assessment
of the impact produced by traffic source on VOC level in the urban area of
Canosa di Puglia (Italy), Water Air Soil Pollut. 193 (2008) 37–50.
[7] R.N. Wixtrom, S.L. Brown, Individual and population exposures to gasoline, J.
Expo. Anal. Environ. Epidemiol. 2 (1992) 23–78.
[8] R. Atkinson, Gas-phase tropospheric chemistry of organic compounds, in: J.W.
Gallagher (Ed.), J. Phys. Chem. Ref. Data, Monograph n. 2, 1994.
[9] P.T. Katsumata, W.E. Kastenberg, Fate and transport of methanol fuel from spills
and leaks, Hazard. Waste Hazard. Mater. 13 (1996) 495–498.
A 110 (2006)

68
A. Hatipoglu et al. / Journal of Photochemistry and Photobiology A: Chemistry 215 (2010) 59–68
[36] M.K. Eberhardt, M. Yoshida, Radiation-induced homolytic aromatic substitu-
tion. I. Hydroxylation of nitrobenzene, chlorobenzene, and toluene, J. Phys.
Chem. 77–5 (1973) 589–597.
[41] C. Minero, S. Chiron, G. Falletti, V. Maurino, E. Pelizzetti, R. Ajassa, M.E. Carlotti,
D. Vione, Photochemical processes involving nitrite in surface water samples,
Aquat. Sci. 69 (2007) 71–85.
[37] W.J. Hehre, L. Radom, P.R. Schleyer, J.A. Pople, Ab initio Molecular Orbital The-
ory, Wiley, New York, 1986.
[42] S. Canonica, M. Freiburghaus, Electron-rich phenols for probing the pho-
tochemical reactivity of freshwaters, Environ. Sci. Technol. 35 (2001)
690–695.
[43] M. Roeselova, J. Vieceli, L.X. Dang, B.C. Garrett, D.J. Tobias, Hydroxyl
radical at the air–water interface, J. Am. Chem. Soc. 126 (2004)
16308–16309.
[44] D. Vione, C. Minero, A. Hamraoui, M. Privat, Modelling photochemical reactions
in atmospheric water droplets: an assessment of the importance of surface
processes, Atmos. Environ. 41 (2007) 3303–3314.
[38] R. Knispel, R. Koch, M. Siese, C. Zetzsch, Adduct formation of OH radicals with
benzene, toluene and phenol and consecutive reactions of the adducts with
NOx and O2, Ber. Bunsenges. Phys. Chem. 94 (1990) 1375–1379.
[39] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical review of rate
constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl
radicals (•OH/•O−) in aqueous solution, J. Phys. Chem. Ref. Data 17 (1988)
1027–1284.
[40] K. Takeda, H. Takedoi, S. Yamaji, K. Ohta, H. Sakugawa, Determination of
hydroxyl radical photoproduction rates in natural waters, Anal. Sci. 20 (2004)
153–158.
[45] P. Nissenson, C.J.H. Knox, B.J. Finlayson-Pitts, L.F. Phillips, D. Dabdub, Enhanced
photolysis in aerosols: evidence for important surface effects, Phys. Chem.
Chem. Phys. 8 (2004) 4700–4710.