Phenol transformation photosensitised by quinoid compounds

Article abstract of DOI:10.1039/c1cp20355j

The phototransformation of phenol in aqueous solution was studied with different quinoid compounds, which are usually detected on atmospheric particulate matter: 2-ethylanthraquinone (EtAQ), benzanthracene-7,12-dione (BAD), 5,12-naphthacenequinone (NQ), 9,10-anthraquinone (AQ), and 2,6-dihydroxyanthraquinone (DAQ). All the studied quinones were able to sensitise the phototransformation of phenol. Under blue-light irradiation the approximated, polychromatic quantum yields for phenol photodegradation were in the order AQ > BAD > EtAQ > NQ > DAQ. Quantum mechanical calculations showed that AQ and DAQ have a very different spin distribution in the triplet state (largely located on the carbonyl oxygen and delocalised over the aromatic ring, respectively) that could account for the difference in reactivity. The spin distribution of EtAQ is similar to that of AQ. Under simulated sunlight, EtAQ induced the highest rate of phenol degradation. Radiation-excited EtAQ would oxidise both ground-state EtAQ and phenol; a kinetic model that excludes the OH radical and singlet oxygen as reactive species is supported by the experimental data. Quinones were also able to oxidise nitrite to nitrogen dioxide, thereby inducing phenol nitration. Such a process is a potential source of nitrogen dioxide and nitrophenols in the atmospheric aerosols. the Owner Societies.

Full text of DOI:10.1039/c1cp20355j

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
PCCP
Cite this: Phys. Chem. Chem. Phys., 2011, 13, 11213–11221
www.rsc.org/pccp
PAPER
Phenol transformation photosensitised by quinoid compoundsw
Valter Maurino,a Andrea Bedini,a Daniele Borghesi,a Davide Vione*ab and
Claudio Minero*a
Received 9th February 2011, Accepted 13th April 2011
DOI: 10.1039/c1cp20355j
The phototransformation of phenol in aqueous solution was studied with different quinoid
compounds, which are usually detected on atmospheric particulate matter: 2-ethylanthraquinone
(EtAQ), benzanthracene-7,12-dione (BAD), 5,12-naphthacenequinone (NQ), 9,10-anthraquinone
(AQ), and 2,6-dihydroxyanthraquinone (DAQ). All the studied quinones were able to sensitise the
phototransformation of phenol. Under blue-light irradiation the approximated, polychromatic
quantum yields for phenol photodegradation were in the order AQ 4 BAD 4 EtAQ 4 NQ 4
DAQ. Quantum mechanical calculations showed that AQ and DAQ have a very different spin
distribution in the triplet state (largely located on the carbonyl oxygen and delocalised over the
aromatic ring, respectively) that could account for the difference in reactivity. The spin
distribution of EtAQ is similar to that of AQ. Under simulated sunlight, EtAQ induced the
highest rate of phenol degradation. Radiation-excited EtAQ would oxidise both ground-state
ꢀ
EtAQ and phenol; a kinetic model that excludes the OH radical and singlet oxygen as reactive
species is supported by the experimental data. Quinones were also able to oxidise nitrite to
nitrogen dioxide, thereby inducing phenol nitration. Such a process is a potential source of
nitrogen dioxide and nitrophenols in the atmospheric aerosols.
to the first singlet state (PS1), followed by deactivation back to
Introduction
the ground state or by inter-system crossing (ISC) to the first
triplet state (PT1). The triplet state could undergo deactivation,
react with ground-state molecular oxygen (3O2) to yield singlet
oxygen (1O2), or react with an oxidisable substrate S via
energy, electron or hydrogen transfer.7,8
Photochemical processes that take place on the surface of
atmospheric particulate matter are potentially important
pathways for the transformation of organic and inorganic
compounds, for the generation of reactive species and for
the production of oligomers such as the humic-like
substances.1–3 Such processes include direct photolysis of
sunlight-absorbing molecules, transformations sensitised by
photoactive compounds via singlet oxygen or the excited
triplet states, and reactions with atmospheric oxidants aꢁnd
P + hn - PS1
PS1 - P
(1)
(2)
(3)
(4)
(5)
(6)
(7)
P
S1–(ISC) - PT1
PT1 - P
ꢀ
photogenerated radical species (e.g. ꢀOH, ꢀNO2, Cl2
,
Br2 ꢀ).4 Atmospheric photosensitisers, among which are
aromatic carbonyls, quinones and furans, can enhance the
phototransformation of other organic molecules.5,6 The
transformation process involves absorption of sunlight by
the photosensitiser P, with electron transition from the ground
ꢁ
3
1
PT1 + O2 - P + O2
1O2 + S - Products
PT1 + S - Products
a
`
Dipartimento di Chimica Analitica, Universita di Torino,
via P. Giuria 5, 10125 Torino, Italy.
Quinoid compounds are common components of atmospheric
particulate matter, where they mainly occur as primary or
secondary pollutants.9 Fig. S1 (ESIw) shows the oxygenated
aromatic compounds, including the quinones, detected on
atmospheric particulate matter next to a high-traffic road near
the town of Torino (NW Italy).10 Primary emission sources of
these compounds to the atmosphere are traffic,11 industrial
activities12 and house heating.13 Quinones can also be formed
as secondary pollutants upon atmospheric oxidation of
E-mail: claudio.minero@unito.it, davide.vione@unito.it;
Web: http://www.chimicadellambiente.unito.it;
Fax: +39-011-6707615
b Centro Interdipartimentale NatRisk, Universita` di Torino,
via Leonardo Da Vinci 44, 10095 Grugliasco (TO), Italy.
Web: http://www.natrisk.org
w Electronic supplementary information (ESI) available: Quinone
levels on airborne particulate, structures of the studied compounds
and of the detected intermediates, optimised structures of AQ and
DAQ. See DOI: 10.1039/c1cp20355j
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 11213–11221 11213

View Online
polycyclic aromatic hydrocarbons (PAHs), in the presence of
14
reactive species such as OH and NO3.
representative of the water layer that is usually found on the
particle surface.29 To achieve deposition, each target quinone
was dissolved in CH2Cl2 and the glass spheres were added.
ꢀ
ꢀ
Atmospheric quinones are generally able to absorb
sunlight,15,16 which induces a singlet–singlet transition (n–p* or
p–p*) that can be followed by inter-system crossing to the first
triplet state, responsible for the photochemical activity.17,18
Photoinduced reactions could be intra- or inter-molecular
electron transfer, hydrogen abstraction and cycloaddition.19–21
The majority of previous studies of quinone reactivity have
been carried out in organic solvents or in water–acetonitrile
mixtures, but it has been demonstrated that quinones can also
act as photosensitisers in aqueous solution.22 In a study of the
transformation of phenolic compounds, photosensitised by the
water-soluble anthraquinone-2-sulfonate we have detected
potential secondary pollutants such as biphenyls and dibenzo-
furans, which were formed by dimerisation of phenol.23
The present study is focused on phenol phototransformation
processes sensitised by quinones that are actually found on
atmospheric particulate matter: 2-ethylanthraquinone (EtAQ),
benzanthracene-7,12-dione (BAD), 5,12-naphthacenequinone
(NQ), 9,10-anthraquinone (AQ), and 2,6-dihydroxyanthraquinone
(DAQ).8,10 The structures of the investigated compounds
(and of the detected transformation intermediates, vide infra)
are reproduced in Scheme S2 (ESIw). Phenol was chosen
as a substrate because of its atmospheric occurrence as a
transformation intermediate of benzene.24
Dichloromethane was then evaporated using
a Buchi
011 Rotavapor device equipped with a Buchi 461 heating
bath. The quinone-loaded glass spheres (10% w/w) were
stored under refrigeration and suspended in water before
use. The sphere loading was adjusted so as to obtain an initial
quinone concentration of 1 mM in the irradiated systems.
Irradiation experiments
Irradiation was carried out in cylindrical Pyrex glass cells
(4.0 cm diameter, 2.3 cm height, tightly closed with a lateral
screw cap), containing 5 mL of aqueous phenol solution in
which the glass spheres were dispersed. The suspensions were
magnetically stirred during irradiation. The adopted light
sources were: (i) a solar simulator (Solarbox, CO.FO.ME.GRA.,
Milan, Italy) equipped with a 1500 W Philips Xenon lamp and
a 310 nm cut-off filter. The radiation intensity incident on the
cells was 25 W mꢁ2, measured with a CO.FO.ME.GRA.
power meter in the wavelength interval of 290–400 nm; (ii) a
40 Watt Philips TL K03 lamp with an emission maximum at
435 nm (blue light). The total incident photon flux in the cells
was Qo = 1.8 ꢂ 10ꢁ5 Einstein Lꢁ1
s
ꢁ1, actinometrically
determined using the ferrioxalate method.30 The calculations
to obtain Qo have taken into account the lamp spectrum, the
radiation absorption of 0.15 M ferrioxalate and its photolysis
quantum yield.31 Fig. 1 reports the emission spectrum of the
blue lamp (spectral photon flux density, q0n,p(l)), measured
with an Ocean Optics SD2000 CCD spectrophotometer and
normalised to the actinometry data.
The ability of the studied quinones to induce the nitration of
phenol in the presence of nitrite under irradiation was also
investigated. Indeed, nitration and photonitration processes of
phenolic compounds are important pathways that yield highly
phytotoxic nitrophenols in the atmosphere. Such compounds are
strongly suspected to substantially contribute to forest damage in
polluted areas and in their downwind regions.25 A combination of
field,26 laboratory27 and modelling data28 suggests that aqueous-
phase nitration and photonitration can play a substantial role in
the occurrence of nitrated phenols in the atmosphere.
In the irradiation experiments under N2 atmosphere, air was
eliminated from the cells by bubbling into the suspension a
gentle flow of high-purity N2 (99.998%, SIAD, Bergamo,
Italy), for 20 min prior to irradiation.
Fig. 1 also reports the absorption spectra of the quinones
under study (molar absorption coefficients e(l)), taken in a
90 : 10 water : acetonitrile mixture with a Varian Cary 100 Scan
UV-Vis spectrophotometer. The absorbed photon flux Pa
Experimental
(Einstein Lꢁ1 sꢁ1
) of the studied quinones under the
Reagents and materials
2-Ethylanthraquinone (EtAQ), benzanthracene-7,12-dione
(BAD), 5,12-naphthacenequinone (NQ), 9,10-anthraquinone
(AQ), 2,6-dihydroxyanthraquinone (DAQ) (all purity grade
Z 97%), phenol, catechol, 4-phenoxyphenol, 2,20-dihydroxy-
biphenyl, 4,40-dihydroxybiphenyl (all purity grade Z 99%),
4-aminoantipyrine (98%), Na2SO4 (499.5%), H2SO4 (95–97%)
and horseradish peroxidase were purchased from Aldrich, H2O2
(35%) from VWR Int. Acetonitrile was LiChrosolv gradient
grade and dichloromethane was Suprasolv grade from VWR
Int. All reagents were used as received, without further
purification. The quinones under study were supported on
hollow glass spheres (9–13 mm average diameter) from Aldrich.
Quinone deposition on glass spheres
The studied quinones were not sufficiently soluble in water to
enable irradiation in solution. They were thus deposited on
glass spheres that would also simulate the adsorption of the
compounds on particulate matter. The aqueous solution is
Fig. 1 Spectral photon flux q0n,p(l) of the Philips TL K03 lamp. Molar
absorption coefficients e(l) of the quinones under study, taken in a
90 : 10 H2O : CH3CN mixture.
c
11214 Phys. Chem. Chem. Phys., 2011, 13, 11213–11221
This journal is the Owner Societies 2011

View Online
blue lamp can be approximately derived as the integral over
wavelength of the product between the spectral photon flux of
the lamp and the fraction of light absorbed by each compound
(absorption factor f(l)).32 It is f(l) = 1 ꢁ 10ꢁe(l)bc, where c is
the quinone concentration (1 mM) and b is the optical path
length of the solution (0.4 cm).
(1 mL) for 30 s, injector temperature 300 1C, oven temperature
40 1C (3 min), then to 300 1C at 15 1C minꢁ1, and stay at
300 1C for 10 min. Quantification was carried out with
the external standard method. In the irradiated systems
containing phenol and EtAQ it was possible to identify the
following transformation intermediates (retention times are in
brackets (min), structures are reported in Scheme S2, ESIw):
2-phenoxyphenol (12.85), 2,20-dihydroxybiphenyl (13.76),
4-phenoxyphenol (14.24), 2,40-dihydroxybiphenyl (15.18),
9,10-anthraquinone (16.06), 4,40-dihydroxybiphenyl (16.54),
2-(1-hydroxyethyl)anthraquinone (18.05) and 2-(1-oxoethyl)-
anthraquinone (18.42). The retention times of phenol and EtAQ
were 7.08 and 17.49 min, respectively.
Z
Pa ¼ qn0;pðlÞð1 ꢁ 10ꢁeðlÞbcÞdl
ð8Þ
l
The described approach is approximated for a number of
reasons: (i) the spectra (Fig. 1) have been recorded in a 90: 10
water : acetonitrile mixture, while the quinones have been
deposited in the solid form on the glass spheres and irradiated
in the presence of water. Some solvatochromic effects can be
expected and the actual absorption spectra in the irradiated
systems could be slightly different; (ii) the Lambert–Beer
approach of eqn (8) is not an exact representation of the
irradiated systems, which were heterogeneous because the
quinones were present in the solid form over the glass spheres.
HPLC-UV analysis
The time evolution of phenol present in the dissolved phase
was monitored by HPLC-UV, using
a Merck-Hitachi
chromatograph equipped with RP-C18 LichroCART
a
column (Merck, length 125 mm, diameter 4 mm) packed with
LiChrospher 100 RP-18 (5 mm diameter). The samples were
isocratically eluted with a 50 : 50 mixture of acetonitrile : water
(1.0 mL minꢁ1 flow rate), and detected at 210 nm. Under these
conditions the column dead time was 0.90 min. The retention
times of the different compounds were (min): phenol (2.25),
4-nitrophenol (2.95), 2-nitrophenol (4.20), and 2-EtAQ (6.65).
Sample treatment after irradiation
It was necessary to characterise both the dissolved compounds
and those present on the glass spheres. After irradiation, 4 mL
of the quinone suspension were filtered on Millipore Millex
HV syringe filters. The filtered aqueous phase was divided into
two parts: one was directly analysed by liquid chromatography
(HPLC-UV, quantification of phenol), the other underwent
solvent extraction with CH2Cl2 followed by analysis in gas-
chromatography coupled with mass spectrometry (GC-MS) to
characterise the water-dissolved transformation intermediates.
Prior to solvent extraction, the aqueous phase was acidified
with 0.9 M H2SO4 and salted with Na2SO4. Each sample was
extracted three times with 1 mL CH2Cl2 each, and the three
extracts were put together. Water traces were eliminated with
anhydrous Na2SO4, the extract was concentrated under a
gentle flow of high-purity nitrogen and analysed by Gas
Chromatography–Mass Spectrometry (GC-MS).
H2O2 determination
The analysis of hydrogen peroxide in the aqueous phase
after filtration was carried out in dedicated experiments with
horseradish peroxidase.33 The adopted procedure was as
follows: 1 mg of enzyme was dissolved in 100 mL aqueous
solution containing 25 mM phenol, 4.9 mM 4-aminoantipyrine,
1.0 mM phosphate buffer (pH 6.9), and 2.5 mM H2O2 to
preserve the enzyme, but with negligible effects on the detection
limit of the method. Aliquots of the solutions after irradiation
(0.3 mL) were added to 2 mL enzyme solution and diluted with
water to 5 mL. The absorbance was spectrophotometrically
measured at 505 nm. The detection limit of H2O2 was
around 3 mM.
The quinone-coated glass spheres remaining on the HV filter
were extracted twice with 1 mL CH2Cl2 each, and the two
extracts were put together. Water traces were eliminated with
Na2SO4 and the extracts analysed by GC-MS. Evaporative
concentration of the extract was not carried out at this stage,
which aimed at quinone quantification.
Quantum mechanical calculations
A study of the excited-state properties of AQ, DAQ and EtAQ
was carried out by employing the density functional theory
(DFT). DFT techniques have already been used to predict the
absorption spectra of AQ and related compounds.34,35 The
present work aims at exploring the features of the triplet states
of AQ, DAQ and EtAQ, with particular interest into the spin
density distribution over the molecular orbitals and the
geometry of the molecules in the excited state.
Some dedicated irradiation experiments were carried out for
the identification of the transformation intermediates present
on the glass spheres. In this case the suspension after irradiation
was filtered on HV membranes, each filter was extracted three
times with 1 mL CH2Cl2 each, and the extracts were put
together. It was followed by water elimination with Na2SO4,
concentration under a gentle N2 flow, and analysis by GC-MS.
Geometry optimisations in the ground state (S0) of AQ,
DAQ and EtAQ were carried out in the gas phase and
were confirmed by analytical calculation of frequencies. An
optimisation of the lowest-lying triplet state (T1) was also
performed. For each molecular species, the gas-phase optimised
structures were determined by gradient procedures, within the
Density Functional Theory (DFT) and the B3LYP hybrid
functional.36 The latter is a well-performing functional for
polycyclic aromatic systems and their functionalised derivatives
GC-MS analysis
The dichloromethane extracts were analysed with a Finnigan
Thermoquest Trace GC Series connected to an ion trap GCQ
Plus mass analyser. The instrument was equipped with a
phenyl-methylsilicone capillary column (HP 5 MS, 30 m
length, 0.25 mm ID, 0.25 mm film thickness). Conditions
employed were: carried flow (He) 1 mL minꢁ1, splitless injection
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 11213–11221 11215

View Online
such as the anthraquinones.37 The polarised 6-311G(d,p) basis
set was used in all the calculations, and the nature of the
stationary points was confirmed by analytical calculation of
the vibrational frequencies. All calculations were carried out
with the GAUSSIAN 03W suite of programs.38 The figures
were generated with the GaussView program.
Results and discussion
Fig. 2 reports the time evolution of 20 mM phenol upon
irradiation under simulated sunlight, in the presence of the
glass spheres alone or the different quinones supported on the
spheres (1 mM quinone concentration in each 5 mL system). It
is apparent that all the quinones enhanced phenol degradation
compared to the naked glass spheres. The reactivity order was
EtAQ 4 BAD 4 AQ E NQ 4 DAQ. The most reactive
quinone under simulated sunlight (EtAQ) was selected for the
study of the reaction pathways.
Fig. 3 Time evolution of (i) 0.1 mM phenol in the presence of the
glass spheres, not loaded with EtAQ. (ii) 0.1 mM phenol + 1 mM
EtAQ loaded on the glass spheres, under N2 atmosphere. (iii) 0.1 mM
phenol + 1 mM EtAQ loaded on the glass spheres, in aerated
solution. Irradiation under simulated sunlight.
Irradiation of phenol and EtAQ
Within the studied systems, EtAQ was detected both in the
aqueous phase and on the spheres, but its amount was
considerably higher on the latter. The concentrations of EtAQ
in the two phases were summed up and the total one is here
reported. Phenol was exclusively present in solution.
Fig. 3 shows the time evolution of 0.1 mM phenol and
1 mM EtAQ upon irradiation under simulated sunlight, in
aerated solution and under N2 atmosphere. Insignificant
phenol transformation was observed upon irradiation in the
presence of the glass spheres only, without EtAQ. In this way
it is possible to exclude the direct photolysis of phenol or a
significant photoactivity of the glass spheres under the
adopted conditions. EtAQ in aerated solution was able to
photosensitise the degradation of phenol, and to undergo
degradation as well. Both phenol and EtAQ underwent faster
degradation in aerated solution than under N2 atmosphere.
Fig. 4 reports the initial transformation rates of phenol
(RPhenol) and of 1 mM EtAQ (REtAQ), calculated from the
Fig. 4 Initial transformation rates of phenol and of 1 mM EtAQ
in aerated solution as a function of phenol initial concentration.
Irradiation under simulated sunlight. Data fit with eqn (19) and
(20). The dotted curves represent the 95% confidence bands of the fit.
initial slope of the time evolution curves, as a function of phenol
concentration. Note the different scales of the plot in Fig. 4: the
degradation rates of phenol (around 10ꢁ8 M sꢁ1) were
considerably lower than those of EtAQ (about 10ꢁ7 M sꢁ1).
Interestingly, the degradation rates of EtAQ and phenol show
opposite trends with phenol concentration. Such a behaviour
suggests that the two compounds could compete for the same
reactive species. In the case of the quinones, this is usually the
excited triplet state.17,18,39 It can thus be hypothesised that
EtAQT1 reacts with both ground-state EtAQ and phenol.
GC-MS analysis enabled the identification of the following
transformation intermediates of EtAQ: 2-(1-oxoethyl)anthra-
quinone (main intermediate), 2-(1-hydroxyethyl)anthraquinone,
and 9,10-anthraquinone (see Scheme S2 (ESIw) for the
structures). All the compounds were detected in the aqueous
Fig. 2 Time evolution of phenol (initial concentration 20 mM) in the
presence of the glass spheres alone or the quinones (1 mM total
concentration) supported on the spheres. Irradiation under simulated
sunlight.
c
11216 Phys. Chem. Chem. Phys., 2011, 13, 11213–11221
This journal is the Owner Societies 2011

View Online
k13
EtAQred þ EtAQox ꢁ! 2EtAQ
k14
phase, and 2-(1-hydroxyethyl)anthraquinone was also detected
on the glass spheres. Trace amounts of a compound that could be
the hydroquinone derivative of EtAQ (2-ethyl-9,10-dihydroxy-
anthracene) were also detected on the glass spheres. The detected
intermediates suggest that the main oxidation pathway of EtAQ
would involve the alkyl lateral chain. This datum is compatible
with a reaction where EtAQT1 extracts an hydrogen atom from
EtAQ, a very common reaction pathway for light-excited
quinones.19 Moreover, the reduction of EtAQT1 by EtAQ
would yield the semiquinone radical EtAQ-Hꢀ. The latter
could yield the corresponding hydroquinone by further
reduction. Interestingly, the hydroquinone was tentatively
detected only at trace levels by GC-MS. In aerated solution
the most likely explanation is that EtAQ-ꢀHꢀ reacts promptly
with dissolved O2 to give EtAQ and HO2 . In the absence of
oxygen, EtAQ-Hꢀ could react with oxidised EtAQ or oxidised
phenol to yield the initial substrates. The result would be a null
cycle that would slow down the transformation kinetics,
coherently with the experimental data obtained under N2
atmosphere. A similar scenario has already been observed in
the case of anthraquinone-2-sulfonate, AQ2S.23
ð13Þ
ð14Þ
ð15Þ
ð16Þ
ð17Þ
ð18Þ
EtAQred þ Phenox ꢁ! EtAQ þ Phenol
k15
EtAQred þ O2 ꢁ! ðEtAQredO2Þ
k16
ðEtAQredO2Þ ꢁ! EtAQ þ HOꢀ2=O2ꢁꢀ
k17
Phenox ꢁ! ProductsPhen
k18
EtAQox ꢁ! ProductsEtAQ
In the absence of O2 the reactions (9)–(14) would constitute
a null cycle with negligible transformation of the substrates.
The fact that, under N2 atmosphere, the transformation of
EtAQ and phenol was slow but not nil (Fig. 3) suggests that
reactions (17) and (18) would not be completely negligible
compared to (13) and (14) when oxygen is not available.
Manageable solutions of the kinetic system can be obtained
for aerated media under the hypothesis that reactions (13) and
(14) are negligible compared to (15), and by application of the
steady-state approximation to EtAQT1, EtAQred, (EtAQredO2),
EtAQox, and Phenox. Under this hypothesis one gets the
following expressions for the initial transformation rates of
EtAQ and phenol (REtAQ and RPhenol, respectively):
GC-MS analysis of the aqueous phase yielded 2,20-,
2,40- and 4,40-dihydroxybiphenyl and 2- and 4-phenoxyphenol
as transformation intermediates of phenol (see Scheme S2
(ESIw) for the structures). These compounds can be formed
upon reaction between phenol and the phenoxyl radical,40
which suggests the formation of phenoxyl upon phenol
ak12kꢁ111½EtAQðtÞꢃ
REtAQ
¼
ð19Þ
b þ ½PhenolðtÞꢃ
oxidation by EtAQT1
.
Catechol, hydroquinone and 1,4-benzoquinone were not
detected as phenol transformation intermediates, by either
GC or HPLC. All these compounds are formed by phenol
and OH, and 1,4-benzoquinone also by phenol and O2.42
In analogy with previous results obtained with AQ2S,23 one
can exclude an important role of either ꢀOH or 1O2 in the
transformation of phenol or EtAQ.
a½PhenolðtÞꢃ
RPhenol
¼
ð20Þ
b þ ½PhenolðtÞꢃ
where a = FPa and b = k11ꢁ1 (k10 + k12[EtAQ(t)]). The fit of
the initial rate data (t = 0) of EtAQ and phenol with eqn (19)
and (20) is reported in Fig. 4, with reasonable agreement
considering the heterogeneous nature of the studied systems.
The fit yielded a = (6.95 ꢄ 1.86) ꢂ 10ꢁ8, k12k11ꢁ1 = 3.7 ꢄ 1.0
and b = (3.9 ꢄ 1.1) ꢂ 10ꢁ4 (error bounds represent ꢄs).
41
1
ꢀ
It is possible to propose a kinetic model to describe the
transformation of EtAQ and phenol. The model involves
irradiation of EtAQ (with absorbed photon flux Pa) to yield
the excited singlet state EtAQS1. It follows either deactivation,
or inter-system crossing (ISC) to give EtAQT1 with quantum
yield F (reaction (9)). The triplet state could undergo deacti-
vation (reaction (10)) or oxidise the ground states of both
phenol and EtAQ (reactions (11) and (12)). The radical
ꢁ
ꢀ
Finally, the formation of HO2ꢀ/O2 in reaction (16) would
be followed by disproportionation and production of H2O2.23
Indeed, around 0.1 mM H2O2 was formed upon 3 h irradiation
of 20 mM phenol and 1 mM sphere-supported EtAQ under
simulated sunlight.
Quantum yields of phenol phototransformation under blue light
intermediates are indicated as EtAQred, EtAQox and Phenox
.
EtAQred could react with oxygen (reaction (15)), EtAQox
(reaction (13)), or Phenox (reaction (14)). Reaction of EtAQred
with EtAQox or Phenox would give back the initial substrates
and account for the low reaction rates observed under N2
atmosphere. Reaction (15) between EtAQred and O2 would
Irradiation under the blue lamp (Philips TL K03) was carried
out to enable an (approximate) assessment of the photon flux
absorbed by the quinones and, as a consequence, of the
quantum yields of phenol phototransformation. Fig. 5 reports
the time trend of 20 mM phenol with the different quinones. In
all the cases the sphere-supported quinones enhanced phenol
transformation compared to the naked glass spheres, and the
reactivity order was similar to that observed with the solar
simulator (Fig. 2). A notable exception is EtAQ, which showed
limited reactivity under blue light. The most likely reason is
that EtAQ has an absorption maximum around 340 nm and
absorbs only a limited fraction of the radiation emitted by the
blue lamp (see Fig. 1).
ꢁ
ꢀ
yield EtAQ and HO2ꢀ/O2 (reaction (16)).
hn
ISC
EtAQ ꢁ! EtAQS1 ꢁ! EtAQT1
ð9Þ
ð10Þ
ð11Þ
ð12Þ
k10
EtAQT1 ꢁ! EtAQ
k11
EtAQT1 þ Phenol ꢁ! EtAQred þ Phenox
Table 1 reports the photon fluxes absorbed by the different
quinones (Pa, calculated by eqn (8)), the initial rates of phenol
k12
EtAQT1 þ EtAQ ꢁ! EtAQred þ EtAQox
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 11213–11221 11217

View Online
Fig. 5 Time evolution of phenol (initial concentration 20 mM) in the
presence of the glass spheres alone or the quinones (1 mM total
concentration) supported on the spheres, under blue light (Philips
TL K03 lamp). Note the zoomed Y-axis, which starts from 0.8.
Table 1 For each studied quinone it is reported: (i) the (approximate)
absorbed photon flux (Pa) under the adopted irradiation conditions,
calculated with eqn (8) under the Lambert–Beer approximation and
upon numerical integration; (ii) the initial degradation rate of phenol
(RP) upon quinone irradiation under blue light, calculated as the initial
slope of the time evolution curves of phenol in Fig. 5; (iii) the
polycꢁh1romatic quantum yield of phenol photodegradation, F =
RPPa
Fig. 6 (a) Spin density surface of AQ in its T1 state (isovalue =
0.005) at right angled view (faded). (b) Spin density surface (faded) of
CO in its T1 state (isovalue = 0.01).
Quinone
Pa/Einstein Lꢁ1 sꢁ1
RP/M sꢁ1
F
AQ skeleton. The reason is the favourable position of the
oxygen p orbital that is parallel to the aromatic p system.
Hydrogen atoms exhibit negligible spin density. Considering
that the photochemical processes involving AQT1 would
probably lead to its reduction to a semiquinone radical species
(in analogy with EtAQ, see reactions (9)–(18)), one can expect
that the spin density on the carbonylic oxygen is connected
with the photochemical reactivity of AQ. In contrast, the spin
density delocalised over the aromatic skeleton would not be
linked with reactivity, because the addition of an electron or a
hydrogen atom there would destabilise the aromaticity of the
molecule and generate unstable photoproducts.
NQ
1.6 ꢂ 10ꢁ5
9.6 ꢂ 10ꢁ6
5.6 ꢂ 10ꢁ6
1.2 ꢂ 10ꢁ5
3.6 ꢂ 10ꢁ6
2.1 ꢂ 10ꢁ10
7.0 ꢂ 10ꢁ11
1.3 ꢂ 10ꢁ10
6.4 ꢂ 10ꢁ10
3.0 ꢂ 10ꢁ10
1.3 ꢂ 10ꢁ5
7.3 ꢂ 10ꢁ6
2.3 ꢂ 10ꢁ5
5.3 ꢂ 10ꢁ5
8.3 ꢂ 10ꢁ5
DAQ
EtAQ
BAD
AQ
transformation (RP) and the (approximate) polychromatic
photodegradation quantum yields (F = RPPaꢁ1). The latter
were in the order AQ 4 BAD 4 EtAQ 4 NQ 4 DAQ.
To get further insight into the different reactivity of the
studied compounds, a DFT quantum mechanical study was
carried out for the ground (S0) and excited triplet states (T1) of
the quinones that showed the highest (AQ) and the lowest
(DAQ) quantum yields for phenol degradation. The geometry
optimisation of the structures is reported in ESI.w To
understand the nature of the excited state, a total spin density
surface of AQ T1 has been created and is reported in Fig. 6a.
There are two main areas where the spin density is the highest:
the strongest one is centred on the carbonyl group, the other
one is on the atoms that allow the resonance of unpaired
electrons over the aromatic skeleton of the molecule. In
particular, the spin density distribution of the carbonyl
group that is more involved in the S0–T1 transition (and that
undergoes the highest bond lengthening, see ESIw for details)
is localised on the p* system, unoccupied in the S0 state. This is
in reasonable analogy with the distribution shown by the CO
molecule in its lower-lying triplet state (Fig. 6b).
If the assumption reported above is correct, one should
expect to find a similar spin distribution also in the case of
EtAQT1, which is expected to oxidise phenol and undergo
reduction to a semiquinone radical. Fig. 7 reports the spin
density surface of the EtAQ molecule in its T1 state. It is very
similar to that observed for AQ. Also in this case there are two
main areas where the spin density is concentrated: the main
one is centred on the carbonyl group, the other over the
aromatic skeleton of the molecule. When considering the
DFT results and the likely reactivity of EtAQ, there appears
to be a link between spin distribution on the carbonyl group
and the ability of that group to be involved in redox reactions.
For the DAQ molecule in its T1 state, the interpretation of
the spin density surface reported in Fig. 8 is more complex.
The molecule has two –OH substituents, which leads to a
change in symmetry from D2h of AQ to C2h of DAQ
(EtAQ has CS symmetry). Interestingly and differently from
In contrast, the spin density present on the oxygen atom of
the other carbonyl group is itself the result of resonance on the
c
11218 Phys. Chem. Chem. Phys., 2011, 13, 11213–11221
This journal is the Owner Societies 2011

View Online
causes the ring itself to evolve from a phenolic-type into a
quinoid-type structure, with consequent readjustment of the
bond lengths (see ESIw for the details of geometry optimisation).
Fig. 9 gives a pictorial scheme of the cited effects, highlighting
the changes in bond features that occur from S0 to T1.
Adiabatic and vertical transition energies for AQ, EtAQ
and DAQ are reported in Table S9 (ESIw).
Table 2 reports spin density values, expressed as percentage,
for some molecular regions that could be meaningful for
anthraquinone photochemistry. Looking at the values
reported in the table, it can be observed that the spin density
of AQ is mainly located on the carbonylic oxygen (57%), while
the remaining is spread over the carbon skeleton of the
molecule. This means that a substantial part of the unpaired
electrons is placed on the photoreactive carbonyl group, which
could justify the high reactivity of AQ in terms of quantum
yields of phenol transformation. The spin distribution of
EtAQ is practically equal to that of AQ, because the ethyl
group appears not to affect the spin resonance over the
anthraquinone skeleton of the molecule. In contrast, in the
case of DAQ most of the spin density (65%) is distributed on
the carbon skeleton that would not be involved in photo-
chemical reactions. Only 27% of the DAQ T1 spin density is
located on the carbonylic oxygen, which would result in lower
photoactivity compared to AQ.
Fig. 7 Spin density surface of EtAQ in its T1 state (isovalue = 0.005)
at right angled view (faded).
Fig. 8 Spin density surface of DAQ in its T1 state (isovalue = 0.005)
Phenol photonitration upon irradiation of the studied quinones
and nitrite
at right angled view (faded).
Fig. 10 reports the time evolution of 2-nitrophenol (2NP) and
4-nitrophenol (4NP) upon irradiation of 1 mM phenol and
0.10 M nitrite under the blue lamp. Irradiation was carried out
with nitrite alone and in the presence of the quinones
supported on the glass spheres (1 mM total quinone loading
in all the cases). The most reactive quinones under blue light
(AQ, BAD and NQ) were selected for this study. It can be
observed that all the adopted quinones enhanced the formation
of 2NP and 4NP compared to nitrite alone. The reactivity order
was NQ 4 BAD 4 AQ.
The photonitration of phenol is accounted for by reaction
with ꢀNO2.43 In the presence of nitrite alone, nitrogen dioxide
ꢁ
is formed upon oxidation of NO2 by the photogenerated
ꢀOH radicals (reactions (21) and (22)):44
ꢁ
NO2 + hn + H+ - OH + NO
(21)
(22)
ꢀ
ꢀ
ꢁ
ꢁ
ꢀ
ꢀ
NO2 + OH - NO2 + OH
Phenol photonitration in neutral solution is a useful probe of
ꢀNO2,45 thus the enhancement of nitrophenol formation by
irradiated quinones suggests a higher availability of nitrogen
dioxide. It can be hypothesised that the adopted quinones
are able to oxidise nitrite under photochemical conditions,
according to the following process (Q = quinone):
Fig. 9 Molecular structures of DAQ showing through bond type the
transition effects from S0 (above) to T1 (below) states, to underline the
change in bond types. The molecular portion remarkably involved in
spin resonance is highlighted through balls and stick representation,
while the rest of the molecule is depicted through wireframe.
Q + hn - QS1–(ISC) - QT1
(23)
(24)
AQ, the spin distribution over the most elongated carbonyl
group does not show the carbon monoxide-like pattern over
the p* system. In contrast, a significant fraction of the spin
density is delocalised on one of the lateral aromatic rings that
bear the OH substituents. Interestingly, the conjugation
between the OH group and the dienic portion of the ring
ꢁ
QT1 + NO2 - Qꢁ + NO2
ꢀ
ꢀ
Interestingly, the production of nitrogen dioxide by nitrate
photolysis and by nitrite oxidation by ꢀOH is an important
ꢀNO2 source in atmospheric waters, possibly prevailing over
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 11213–11221 11219

View Online
Table 2 Spin density distribution over AQ, EtAQ and DAQ molecules, expressed in terms of percentage (the sum of Mulliken atomic spin density
is 2.00). n/a = not applicable
AQ (%)
EtAQ (%)
DAQ (%)
Ring A
Ring B
O of CQO
C of CQO
O of OH
14
14
57
15
n/a
14
13
57
16
n/a
55
10
27
0
8
particle surface. Similarly, it has been shown that nitrite
oxidation by photoexcited chromophoric dissolved organic
matter can be a significant ꢀNO2 source in organic-rich surface
waters, and that quinoid compounds are probably involved in
the process.45
Conclusions
This work shows that quinones supported on glass spheres are
able to sensitise the transformation of phenol in irradiated
aqueous solution. Such a process is relevant to the photo-
chemistry of quinones present on airborne particles, the
surface of which is usually covered by a water layer.29 Among
the studied photosensitisers, the order of the quantum yields
for phenol phototransformation under blue light was AQ 4
BAD 4 EtAQ 4 NQ 4 DAQ. For AQ and DAQ, DFT
calculations showed very different spin density distributions in
the T1 state. In the case of AQ (and of EtAQ), most of the spin
density is centred over the oxygen atom of the carbonyl group,
which is reasonable considering that AQT1 would oxidise
phenol and undergo reduction to a semiquinone radical. In
the case of DAQ most of the spin density is delocalised over
the unreactive aromatic skeleton, which could account for the
lower photoactivity.
EtAQ was the most reactive quinone toward phenol
transformation under simulated sunlight. The investigation
of phenol transformation by EtAQ supported on glass spheres
led to a kinetic model made up of reactions (9)–(18), able to
account for the experimental data. According to the model,
radiation-excited EtAQ (EtAQT1) would oxidise the ground
states of both EtAQ and phenol. In aerated solution the
oxidised radicals would evolve into lateral-chain oxidation
compounds (from EtAQ), dihydroxybiphenyls and phenoxy-
phenols (from phenol). EtAQT1 would undergo reduction,
and in aerated solution the reduced radical (most likely a
semiquinone) would react with O2 to yield back EtAQ. In the
absence of oxygen, reduced and oxidised species would
generate back the starting compounds, in a null cycle that
would keep the reaction rates very low.
Fig. 10 (A) Time evolution of 2-nitrophenol (2NP) upon irradiation
of 1 mM phenol + 0.10 M NaNO2, alone or in the presence of the
quinones supported on the glass spheres (1 mM total quinone loading).
Irradiation under the Philips TL K03 lamp (blue light). (B) Time
evolution of 4-nitrophenol (4NP) under the same conditions as above.
dissolution of ꢀNO2 from the gas phase.46 Moreover, nitrite
photochemistry is a major source of radical species in irradiated
fog and rain water.47,48 The results reported here suggest that
the oxidation of nitrite by light-excited quinones might play a
role as a source of ꢀNO2 and nitroaromatic compounds on the
Phenol transformation intermediates are compatible with
the formation of a phenoxyl radical in the primary reaction
step(s), and it is also suggested that environmentally concerning
secondary pollutants can be formed in the presence of
phenolic compounds and quinones under irradiation. Phenol
c
11220 Phys. Chem. Chem. Phys., 2011, 13, 11213–11221
This journal is the Owner Societies 2011

View Online
intermediates are also not compatible with the occurrence of
23 V. Maurino, D. Borghesi, D. Vione and C. Minero, Photochem.
Photobiol. Sci., 2008, 7, 321–327.
24 R. Volkamer, B. Klotz, I. Barnes, T. Imamura, K. Wirtz,
N. Washida, K. H. Becker and U. Platt, Phys. Chem. Chem. Phys.,
2002, 4, 1598–1610.
25 M. A. J. Harrison, S. Barra, D. Borghesi, D. Vione, C. Arsene and
R. I. Olariu, Atmos. Environ., 2005, 39, 231–248.
26 J. Luttke, V. Schher, K. Levsen, G. Wounsch, J. N. Cape,
K. J. Hargreaves, R. L. Storeton-West, K. Acker, W. Wieprecht
and B. Jones, Atmos. Environ., 1997, 31, 2637–2648.
27 D. Vione, V. Maurino, C. Minero and E. Pelizzetti, Environ. Sci.
Technol., 2005, 39, 7921–7931.
28 M. A. J. Harrison, M. R. Heal and J. N. Cape, Atmos. Chem.
Phys., 2005, 5, 1679–1695.
29 V. Samburova, T. Didenko, E. Kunenkov, C. Emmenegger,
R. Zenobi and M. Kalberer, Atmos. Environ., 2007, 41, 4703–4710.
30 H. J. Kuhn, S. E. Braslavsky and R. Schmidt, Pure Appl. Chem.,
2004, 76, 2105–2146.
1
ꢀ
reactive species such as OH or O2. Furthermore, formation
of dihydroxybiphenyls and phenoxyphenols can potentially be
the first steps of an oligomerisation process, similar to the
reactions that yield the humic-like substances in the
atmosphere.49,50
Photoexcited quinones can oxidise nitrite to nitrogen
dioxide, which enhances phenol nitration compared to the
irradiation of nitrite alone. Such a process could lead to
significant generation of phytotoxic nitrophenols in the
atmospheric aqueous phase, which could contribute to vegetation
damage. Because of the potential importance of quinone-
enhanced nitration, further investigation will be required to
assess its significance in the atmospheric aerosol phase.
31 J. G. Calvert and J. N. Pitts, Photochemistry, Wiley, NY, 1966.
32 S. E. Braslavsky, Pure Appl. Chem., 2007, 79, 293–465.
33 X. Q. Yang and K. Ma, Anal. Biochem., 2005, 344, 130–134.
34 L. Shen, H.-F. Ji and H.-Y. Zhang, J. Mol. Struc. (THEOCHEM),
2008, 851, 220–224.
35 J. Preat, D. Jacquemin, J. Burton, D. P. Vercauteren and
E. A. Perpete, Dyes Pigm., 2009, 81, 97–102.
36 J. B. Foresman and Æ. Frisch, Exploring Chemistry with Electronic
Structure Methods, Gaussian, Inc., Pittsburgh, PA, 1996,
pp. 166–168.
Acknowledgements
Financial support from PNRA-Progetto Antartide is gratefully
acknowledged. We are thankful to Prof. Eliano Diana for
useful discussion.
Notes and references
37 D. Jacquemin, J. Preat, M. Charlot, V. Wathelet, J. M. Andre and
E. A. Perpete, J. Chem. Phys., 2004, 121, 1736–1743.
38 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery Jr., R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui,
K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts,
R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-
Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 03 (Revision
B.05), Gaussian, Inc., Pittsburgh, PA, 2003.
39 P. R. Maddigapu, A. Bedini, C. Minero, V. Maurino, D. Vione,
M. Brigante, G. Mailhot and M. Sarakha, Photochem. Photobiol.
Sci., 2010, 9, 323–330.
40 J. Platz, O. J. Nielsen, T. J. Wallington, J. C. Ball, M. D. Hurley,
A. M. Straccia, W. F. Schneider and J. Sehested, J. Phys. Chem. A,
1998, 102, 7964–7974.
41 F. Machado and P. Boule, J. Photochem. Photobiol., A, 1995, 86,
73–80.
42 M. Nowakowska and M. Kepczynski, J. Photochem. Photobiol., A,
1998, 116, 251–256.
1 A. Jammoul, S. Gligorovski, C. George and B. D’Anna, J. Phys.
Chem. A, 2008, 112, 1268–1276.
2 S. Net, S. Gligorovski, S. Pietri and H. Wortham, Phys. Chem.
Chem. Phys., 2010, 12, 7603–7611.
3 I. Grgic, L. I. Nieto-Gligorovski, S. Net, B. Temime-Roussel,
S. Gligorovski and H. Wortham, Phys. Chem. Chem. Phys.,
2010, 12, 698–707.
4 D. Vione, V. Maurino, C. Minero, E. Pelizzetti, M. A. J. Harrison,
R. I. Olariu and C. Arsene, Chem. Soc. Rev., 2006, 35, 441–453.
5 C. Anastasio, B. C. Faust and C. J. Rao, Environ. Sci. Technol.,
1997, 31, 218–232.
6 S. Canonica, B. Hellrung, P. Muller and J. Wirz, Environ. Sci.
Technol., 2006, 40, 6636–6641.
7 M. Jang and S. R. McDow, Environ. Sci. Technol., 1995, 29,
2654–2660.
8 M. Jang and S. R. McDow, Environ. Sci. Technol., 1997, 31,
1046–1053.
9 J. O. Allen, N. M. Dookeran, K. Taghizadeh, A. L. Lafleur,
K. A. Smith and A. F. Sarofim, Environ. Sci. Technol., 1997, 31,
2064–2070.
10 C. Minero and V. Maurino, Study of the space-time distribution of
compounds dangerous for human health in high-traffic areas,
Technical report to the Province of Torino, Italy, 2000.
11 T. E. Jensen and R. A. Hites, Anal. Chem., 1983, 55, 594–599.
12 Y. A. Levendis, A. Atal and J. B. Carlson, Environ. Sci. Technol.,
1998, 32, 3767–3777.
13 W. F. Rogge, L. M. Hildemann, M. A. Mazurek, G. R. Carr and
B. R. T. Simoneit, Environ. Sci. Technol., 1998, 32, 13–22.
14 A. H. Neilson, PAHs and Related Compounds, Springer, Berlin,
1998.
15 T. Corrales, F. Catalina, C. Peinado and N. S. Allen,
J. Photochem. Photobiol., A, 2003, 159, 103–114.
16 R. G. Brinson, S. C. Hubbard, D. R. Ziudema and P. B. Jones,
J. Photochem. Photobiol., A, 2005, 175, 118–128.
17 T. Del Giacco, L. Latterini and F. Elisei, Photochem. Photobiol.
Sci., 2003, 2, 681–687.
18 H. Gorner, Photochem. Photobiol. Sci., 2006, 5, 1052–1058.
19 S. Patai, The Chemistry of the Quinoid Compounds, Wiley, NY, 1974.
20 H. Gorner, Photochem. Photobiol. Sci., 2004, 3, 933–938.
21 V. Latour, T. Pigot, M. Simon, H. Cardy and S. Lacombe,
Photochem. Photobiol. Sci., 2005, 4, 221–229.
43 P. R. Maddigapu, D. Vione, B. Ravizzoli, C. Minero, V. Maurino,
L. Comoretto and S. Chiron, Environ. Sci. Pollut. Res., 2010, 17,
1063–1069.
44 J. Mack and J. R. Bolton, J. Photochem. Photobiol., A, 1999, 128,
1–13.
45 P. R. Maddigapu, C. Minero, V. Maurino, D. Vione, M. Brigante
and G. Mailhot, Chemosphere, 2010, 81, 1401–1406.
46 C. Minero, V. Maurino, E. Pelizzetti and D. Vione, Environ. Sci.
Pollut. Res., 2007, 14, 241–243.
47 C. Anastasio and K. G. McGregor, Atmos. Environ., 2001, 35,
1079–1089.
48 A. Albinet, C. Minero and D. Vione, Sci. Total Environ., 2010, 408,
3367–3373.
49 A. Hoffer, G. Kiss, M. Blazso and A. Gelencser, Geophys. Res.
Lett., 2004, 31, L06115.
22 A. E. Alegrıa, A. Ferrer, G. Santiago, E. Sepulveda and W. Flores,
J. Photochem. Photobiol., A, 1999, 127, 57–65.
50 B. J. Holmes and G. A. Petrucci, J. Atmos. Chem., 2007, 58,
151–166.
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 11213–11221 11221