A laser flash photolysis and theoretical study of hydrogen abstraction from phenols by triplet α-naphthoflavone

Article abstract of DOI:10.1039/c002738c

The hydrogen abstraction (HA) reaction by the triplet of α-naphthoflavone (1) has been investigated experimentally by the use of laser flash photolysis (LFP) and theoretically with density functional theory (DFT) and atoms in molecules (AIM). The triplet

Full text of DOI:10.1039/c002738c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
A laser flash photolysis and theoretical study of hydrogen abstraction
from phenols by triplet a-naphthoflavonew
a
b
c
a
ˆ
Nanci C. de Lucas,* Helena S. Fraga, Cristiane P. Cardoso, Rodrigo J. Correa,
Simon J. Garden*a and J. C. Netto-Ferreirabc
Received 9th February 2010, Accepted 12th May 2010
DOI: 10.1039/c002738c
The hydrogen abstraction (HA) reaction by the triplet of a-naphthoflavone (1) has been
investigated experimentally by the use of laser flash photolysis (LFP) and theoretically with
density functional theory (DFT) and atoms in molecules (AIM). The triplet excited state of 1,
in acetonitrile, has an absorption maximum at 430 nm and lifetime of 10 ms. The quenching rate
constants for the triplet of 1 with 1,4-cyclohexadiene, substituted phenols and amines were
determined. The low reactivity of this ketone with respect to HA from 1,4-cyclohexadiene is in
accord with a p,p* excited state. HA from phenols in acetonitrile is proposed to occur in a
diffusion controlled reaction from free phenol based upon the determination of the Abraham
bH value for acetonitrile and correction of the quenching rate constants for hydrogen bonding of
2
the phenols to acetonitrile. A molecular orbital analysis of the triplet (SOMO and SOMO-1) of 1
reveals contributions from the carbonyl oxygen atom, but principally from the a-carbon and the
associated p-bond network, consistent with a p,p* excited state. From a thermodynamic point of
view, the triplet HA from phenol to oxygen of the carbonyl group is 17 kcal molꢀ1 less demanding
than the transfer to the a-carbon, consistent with the acidic nature of the phenolic hydrogen atom.
DFT and AIM analysis of the hydrogen abstraction reaction reveals that the transition state (TS) is
pseudo-symmetrically polarized and that HA in the hydrogen bonded exciplex occurs in a concerted
manner but not necessarily by simultaneous electron and proton transfer.
flavone in 2-propanol.5 Time-resolved spectroscopy experiments
suggest that flavone photochemistry is mainly mediated by the
triplet excited state,6,7 as confirmed by the high intersystem
crossing (ISC) quantum yield (B0.9).8 The transient absorption
spectrum of triplet flavone shows bands at 365–370 nm and
640–650 nm, and this transient has been shown to be reactive
towards hydrogen donors.6–9 Transient spectroscopic investigation
of flavone, 7-methoxyflavone and 3-methoxyflavone revealed
the quantitative formation of the triplet states for the first two
flavones, but reduced triplet formation for 3-methoxyflavone
as a result of formation of a 1,4-biradical due to intra-
molecular hydrogen abstraction (HA).10
Introduction
Flavones are a large class of natural pigments with antioxidant
activity. They posses a 2-phenyl-4-benzopyrone unit and a
large number of derivatives result from combinations of
multiple hydroxyl and/or methoxyl substituents. These
compounds are usually found in the tissues of higher plants,
mainly in tropical regions.1 Photocontrol of their production
is well documented and seems to operate at many different
points along the biosynthetic pathway.2 Their efficient absorbance
of UV irradiation has implied their possible involvement in
protection against UV-induced damage in plants.3,4
Photophysical and photochemical studies on flavonoids
have revealed important properties of these substances, and
photodimerization is the principal photochemical reaction of
The phosphorescence spectra and lifetime measured for
flavone at 77 K indicates a mixed n,p*–p,p* character for
the lowest triplet excited state irrespective of the solvent
polarity.11 Sakaguchi et al.12 have shown that when flavones
are irradiated in 2-propanol in the presence of hydrogen
donors, two different radical species can be generated as the
result of hydrogen transfer to oxygen or to carbon.
a
´
ˆ
Instituto de Quımica, Departamento de Quimica Organica,
Universidade Federal do Rio de Janeiro, Ilha do Funda˜o, CT,
Bloco A, Rio de Janeiro, 21949-900 RJ, Brazil.
E-mail: nancicl@iq.ufrj.br, garden@iq.ufrj.br;
Fax: +55-21-2562-7106; Tel: +55-21-2562-7278
Departamento de Quımica Geral e Inorganica, Universidade Federal
da Bahia, Rua Bara˜o de Geremoabo, Campus de Ondina,
40170-290 BA, Brazil
Photoinduced HA by triplet ketones has been the subject of
considerable interest in molecular photochemistry.13–28 The
nature of the lowest triplet excited state (n,p* or p,p*) of a
ketone is known to influence the rate of photoinduced HA
from a donor species, where the n,p* triplet is the more
reactive state. In the case of p,p* triplets, HA reactions were
postulated to be the result of thermal population of the higher
energy n,p* triplet.15,29
b
c
´
ˆ
´
Departamento de Quımica, Universidade Federal Rural do Rio de
Janeiro, Antiga-Rio-Sa˜o Paulo km 47, Serope´dica 23850-250 RJ,
Brazil
w Electronic supplementary information (ESI) available: A detailed
analysis of the kinetic data in terms of correlations with Hammett
parameters and the use of Abraham’s aH2 and bH2 parameters, and the
calculated transition state structure for hydrogen atom transfer to the
keto a-carbon. See DOI: 10.1039/c002738c
The mechanisms for HA are influenced by the identity of the
atoms involved in the transfer of the ‘‘hydrogen atom’’, the
ꢁc
This journal is the Owner Societies 2010
10746 | Phys. Chem. Chem. Phys., 2010, 12, 10746–10753

View Article Online
bond energy, the oxidation potential of the hydrogen donor
and by the solvent. Mechanisms spanning the extremes of
alkoxyl-radical-like hydrogen atom transfer (HAT), as with
alkanes or alcohols,30,31 and full electron transfer (as with
amines) have been identified where intermediate mechanisms
may have a proton coupled electron transfer (PCET) char-
acter.32–39 In the case of excited state ketones abstracting
hydrogen from phenols it has been suggested that a mechan-
ism involving initial formation of a hydrogen-bonded triplet
exciplex followed by coupled electron/proton transfer occurs
where the nature and the reduction potential of the triplet state
may influence the HA mechanism.13,20,23,24,27,40,41
Additionally, DFT and AIM calculations of the reaction of
the triplet excited state of 1 with phenol were performed in
order to investigate the possible mechanism for phenolic HA.
Materials and methods
Materials
The solvents (VETEC spectrograde) were used as received.
3-Methoxyphenol,
4-methoxyphenol,
4-hydroxyphenol,
3-hydroxyphenol, 4-chlorophenol, 3-chlorophenol, 4-bromo-
phenol, 4-methylphenol, 3-methylphenol, 4-cyanophenol,
4-fluorophenol, 1,3-cyclohexadiene, 1,4-cyclohexadiene and
DABCO were used as received from Aldrich. a-Naphthoflavone
(Aldrich) was recrystallized from ethanol.
The hydrogen bonding ability of solvents has been known to
influence the quenching rates of triplet ketones by phenols
since the early work of Scaiano and co-workers.13 More recent
systematic studies of kinetic solvent effects on HA from
phenols has resulted in the development of a better under-
standing of how solvents can influence HA from a hydrogen
donor and have resulted in the identification of mechanisms
for HA that involve the solvent.28,42–55
Laser flash photolysis
The laser flash photolysis experiments were carried out on a
LuzChem Instrument model mLFP122. Samples were contained
in a 10 mm ꢃ 10 mm cell made from Suprasil tubing and
were deaerated by bubbling with oxygen-free nitrogen for
about 30 min. The samples were irradiated with the third
harmonic of a Nd/YAG Surelite laser (l = 355 nm, B4 to
6 ns, Z 40 mJ/pulse).
The a-naphthoflavone (1) concentration (B1 mM) was
chosen in order to give an absorption in the wavelength of
excitation (355 nm) of 0.3–0.5. Stock solutions of quenchers
were prepared so that it was only necessary to add microlitre
volumes to the sample cell in order to obtain appropriate
concentrations of the quencher.
In addition to the extensive literature of experimental results
on hydrogen atom abstraction, theoretical calculations have
made a considerable impact upon our understanding of the
underlying reaction mechanisms involved in these reactions,
particularly in the case of ground state alkoxyl or peroxyl
radicals abstracting hydrogen from phenol.56–61 With specific
reference to differences between HAT and PCET mechanisms,
work by Mayer et al.36 that investigated self exchange reac-
tions between R–H/Rꢂ radical (toluene/benzyl, phenol/
phenoxyl and methanol/methoxy) and the more recent study
by Tishchenko et al.39 that investigated topographic features
of the ground and excited potential energy surfaces for the
reactions between phenol and N- and O- centered radicals, are
notable advances in the theoretical interpretation of HA
reaction mechanisms.
In a recent study, Singh et al.62 studied the mechanistic
aspects of HA by a hydroperoxyl radical from substituted
phenols using AIM theory. The analysis of the transferring
hydrogen atom showed that HA followed a PCET mechanism.
With respect to the study of excited state ketones, we have
recently employed DFT and AIM calculations to study the
HA mechanism from phenol and 2-propanol by ortho-quinone
triplets.20,21 Additionally, DFT calculations have been used to
understand an analogous mechanism, the photoreduction of
2-benzoylthiophene by phenol.25,63
Computational methods
Geometrical, energetic and AIM calculations were performed
at (U)B3LYP/6-311++G**//(U)B3LYP/6-31G* for all
structures. The Gaussian 98 package of programs was used
to calculate the geometries and energies.64 The geometries
were optimized using standard techniques,65 and, after
geometry optimization, vibrational analysis was performed
to confirm that the geometries were true minima on the
potential energy surface, as shown by the absence of imaginary
frequencies. Transition states were confirmed by the presence
of a single imaginary vibration that corresponded to vibration
along the reaction coordinate. The DFT method was used as it
reasonably estimates spin contamination. Values for hS2i were
typically consistent with two unpaired electrons for triplets.
AIM (atoms in molecules) calculations were performed using
AIM 2000, v2.0.66,67 The AIM theory was used to investigate
the atomic properties of the atoms involved in the triplet HA
reaction.
The present study details the results of a laser flash
photolysis study of the photochemistry of a-naphthoflavone
(1) towards hydrogen and electron donors in organic solvents.
Results and discussion
Triplet characterization
Irradiation of a degassed acetonitrile solution of 1 led to the
formation of a detectable transient with absorption at 440 nm
(Fig. 1). This transient decays by mixed first and second order
kinetics with a lifetime of 10 ms. This transient was quenched
by oxygen and by 1,3-cyclohexadiene with an essentially
diffusion-controlled rate constant 9.9(ꢄ0.2) ꢃ 109 L molꢀ1 sꢀ1
.
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10746–10753 | 10747

View Article Online
In the presence of b-carotene, a frequently used triplet state
quencher due to its low triplet energy, the formation of a
520 nm band due to the b-carotene triplet could be easily
observed.68 These results led to the assignment of the transient
in Fig. 1 as the triplet excited state of 1.
where k0 is the decay rate constant of the triplet in the absence
of the quencher, and [Q] the quencher concentration. Plots
based on this equation for the triplet of 1 being quenched by
1,4-cyclohexadiene, phenols, triethylamine and DABCO were
found to be linear, from which the respective values of kq were
determined (Table 1). Fig. 2 shows representative quenching
plots for 1 by various phenols.
Quenching of triplet 1 by 1-methylnaphthalene (ET
=
60.6 kcal molꢀ1) was observed in acetonitrile (kq = 5.1 ꢃ
109 mol Lꢀ1), with the concomitant formation of the triplet of
1-methylnaphthalene at 420 nm.68 Since the quenching of
triplet 1 by 1,3-cyclohexadiene (ET = 52.4 kcal molꢀ1) is
diffusion-controlled, one can infer that the triplet energy of 1
has a value greater than 52.4 and close to 60.6 kcal molꢀ1. For
Addition of a typical hydrogen donor, such as 1,4-cyclo-
hexadiene, can lead to the formation of the cyclohexadienyl
radical, which does not absorb strongly above 350 nm, and
the ketone-derived ketyl radical. The transient absorption
spectrum registered in the presence of 1,4-cyclohexadiene
(0.35 mol Lꢀ1) clearly shows the formation of a new transient
around 470 nm that can be attributed to the a-naphthoflavone
ketyl radical (Fig. 3A). To confirm this the normalized spectra
for the transient obtained upon laser flash photolysis of 1 was
subtracted from the normalized spectrum of the transients
obtained in the presence of 1,4-cyclohexadiene thus
clearly revealing that the absorption maximum for this new
transient is at 470 nm (Fig. 3B). The small value for the
quenching rate constant of triplet 1 by 1,4-cyclohexadiene
comparison, flavone has a triplet energy of 62 kcal molꢀ1 8
.
The lifetime of triplet 1 is very concentration dependent as a
result of self-quenching, a rather common process with long
lived triplet ketones.41 A plot of the triplet decay rate constant
as a function of the concentration of 1 yields ksq, the
self-quenching rate constant from the slope. For acetonitrile
we were able to measure a self-quenching rate constant of
2.1 ꢃ 109 L molꢀ1
s
ꢀ1. This large value is similar to the one
obtained for dinaphthyl ketones and are quite typical for p,p*
triplets.41,69
(6.9 ꢃ 106 L molꢀ1
s
ꢀ1) is in agreement with the expected
low reactivity for an HA reaction by an aromatic p,p* triplet.
In comparison, a typical n,p* triplet, for example benzo-
phenone,70 reacts with 1,4-cyclohexadiene with a rate constant
of 2.9 ꢃ 108 L molꢀ1 sꢀ1 whilst the quenching rate constant for
the dinaphthyl ketone triplet, which has p,p* character, is on
Photoreduction of triplet 1
The photoreduction of triplet 1 by 1,4-cyclohexadiene, phenols
and DABCO was studied in acetonitrile. Addition of these
quenchers led to a shortening of the triplet lifetime of 1 with
the triplet decay following pseudo-first order kinetics in the
presence of these quenchers. The experimentally observed rate
constant, kobs, is related to the quenching rate constant, kq,
according to eqn (1).
the order 105 L molꢀ1 sꢀ1 41
.
Quenching of the triplet of 1 with DABCO occurred with a
rate constant approaching diffusion control, i.e. kq = 1.9 ꢃ
109 L molꢀ1 ꢀ1, a value typical for an electron transfer
s
reaction.71 Fig. 4 shows the transient absorption spectra
obtained after laser excitation of 1 in the presence of
1.9 ꢃ 10ꢀ3 mol Lꢀ1 of DABCO, in an acetonitrile solution.
These spectra show the disappearance of the 440 nm band,
corresponding to triplet 1 and the formation of new bands at
390 and 500 nm. On a short time scale the formation of this
transient was followed at 390 nm (Fig. 4, insert). This new
kobs = k0 + kq[Q]
(1)
Fig. 1 Transient absorption spectra recorded after 355 nm laser
excitation of 1 in acetonitrile solution at the indicated intervals of
time after the laser pulse. Insert: an example of a decay (430 nm), from
which values of DOD were determined at specific time intervals after
the pulse (approx. 5 ms) for each wavelength.
Fig. 2 Representative quenching plots for the triplet of 1 by various
phenols. lexc = 355 nm; lmon = 430 nm; solv. = acetonitrile.
ꢁc
This journal is the Owner Societies 2010
10748 | Phys. Chem. Chem. Phys., 2010, 12, 10746–10753

View Article Online
Table 1 Second-order rate constants (kq) for the quenching of the
triplet of 1 in acetonitrile and corrected values (kabs) for hydrogen
bonding with the solvent
Quencher
kqa/L molꢀ1 sꢀ1
kabsb/L molꢀ1 sꢀ1
1,4-Cyclohexadiene
1,4-Cyclohexadiene
Phenol
6.9 ꢃ 106
—
—
4.0 ꢃ 107c
2.3 ꢃ 108
2.79 ꢃ 1010
4.86 ꢃ 1010
2.84 ꢃ 1010
4.68 ꢃ 1010
2.30 ꢃ 1010
1.63 ꢃ 1010
3.44 ꢃ 1010
1.49 ꢃ 1010
4.20 ꢃ 1010
8.48 ꢃ 109c
—
4-Cyanophenol
3-Fluorophenol
3-Chlorophenol
4-Chlorophenol
4-Bromophenol
3-Methylphenol
4-Methylphenol
4-Methoxyphenol
4-Methoxyphenol
4-Methoxyphenol
4-Methoxyphenol
4-Hydroxyphenol
3-Hydroxyphenol
DABCO
9.9 ꢃ 107
1.3 ꢃ 108
1.9 ꢃ 108
1.1 ꢃ 108
7.6 ꢃ 107
3.4 ꢃ 108
1.5 ꢃ 108
4.1 ꢃ 108
3.7 ꢃ 109c
8.7 ꢃ 108 (ACN : H2O)
9.1 ꢃ 108 (ACN : D2O)
1.0 ꢃ 109
—
1.01 ꢃ 1011
3.45 ꢃ 1010
—
2.7 ꢃ 108
1.9 ꢃ 109
a
b
Estimated to be accurate to ꢄ10%. kabs = kq corrected for
hydrogen bonding of the phenol to acetonitrile, assuming that only
free phenol reacts with the triplet ketone (see the main text and the
c
ESIw). In chlorobenzene.
band was attributed to the anion radical derived from
a-naphthoflavone, as the radical cation of DABCO is ruled
out as it absorbs weakly at 460–480 nm.72–75
The HA reaction of triplet 1 from phenols leads to the
formation of the corresponding phenoxyl radicals. These
radicals have strong absorption bands in the 370–505 nm
region, that can be easily detected by LFP experiments.13
For example, the transient absorption spectrum obtained in
the presence of 3.8 ꢃ 10ꢀ3 mol Lꢀ1 of 4-methoxyphenol shows
an absorption band at 405 nm, that can be attributed to the
4-methoxyphenoxyl radical (Fig. 5).
The reactivity of triplet 1 with phenols (Table 1) is, as
expected, greater than the reactivity with 1,4-cyclohexadiene,
and can be attributed to the low bond dissociation energy and
the ease of oxidation of the phenols.13,17,20,50,76 The reactivity
is observed to be dependent upon the nature of the phenolic
substituent, i.e. phenols with an electron donating substituent
react faster than phenols with an electron withdrawing
substituent. A similar reactivity pattern has been observed in
previous studies of n,p* triplets of benzophenone77 and
acenaphthone.17 In these studies excellent correlations with
the Hammett parameter s+ were obtained with a reaction
constant (r) of ꢀ1.5 suggesting that the transition states are
electrophilic in nature, consistent with the known electrophilic
character of the carbonyl triplets. The observed rate constants
in this study reveal a similar tendency to the previous studies
but only a moderate correlation with s+ (R2 = 0.64,
r = ꢀ0.54) was observed. The respective correlations with
Fig. 3 (A) Transient absorption spectra obtained 0.06 ms after the
laser pulse excitation (355 nm) of 1 in acetonitrile and in 0.35 mol Lꢀ1
of 1,4-cyclohexadiene in an acetonitrile solution. Insert: quenching
plot of the triplet by 1,4-cyclohexadiene. lmon = 430 nm. (B) Transient
absorption spectrum obtained by subtracting the normalized spectrum
in the presence of 1,4-cylohexadiene from that in the absence of the
quencher.
this reaction is not important, and suggests that a PCET
mechanism operates.27
Ingold and co-workers have performed extensive studies of
solvent effects on HA from phenolic compounds.42–45,47–49
They have identified and quantified the importance of hydrogen
bonding by the solvent and expressed this in terms of
78,79
Abraham’s thermodynamic parameters aH and bH
,
2
2
which, respectively, represent the acidity of a proton donor
other substituent parameter constants (s[R2
r = ꢀ0.87], sjj[R2 = 0.12]) were less satisfactory.
= 0.60,
and the basicity of a proton acceptor, as given by a theoretically
derived eqn (2). The experimentally determined equation had a
45,80
The rate of the reaction of the triplet of 1 with 4-methoxy-
phenol was investigated in 9 : 1 acetonitrile–H2O and 9 : 1
acetonitrile–D2O. The results are given in Table 1. An isotope
effect of 0.95 was observed. The value for the isotope effect
indicates that stretching of the hydrogen bond in the TS for
gradient of ꢀ8.3aH2bH
.
2
ꢀ1
ꢀ1
s
log(kS
/M
s
ꢀ1) = log(ko
ꢀ1) ꢀ 7.4aH2bH
ꢂ
ꢂ
/M
XH/Y
XH/Y
2
(2)
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10746–10753 | 10749

View Article Online
constant for hydrogen bond formation between the phenol
and the solvent and that this dependence is, more or less,
proportionally constant in the series of phenols. On the basis
that HA may be occurring exclusively from free phenol, it is
possible to calculate corrected rate constants. Abraham’s
values of logKH can be used to determine the equilibrium
A
constants for hydrogen bonding between phenol and
acetonitrile.78 When such a correction is made to the quenching
rate constants (kq), the corrected quenching rate constants
(kabs) are found to be remarkably similar to the diffusion
constant (1.9 ꢃ 1010
M
ꢀ1 sꢀ1) in acetonitrile.81 Almost all the
values (kabs) fall in the range 3.0 ꢄ 1.5 ꢃ 1010 L molꢀ1 sꢀ1 with
a few outlying exceptions (Table 1). Therefore the HA reactions
from phenols by triplet 1 are diffusion controlled in acetonitrile.
The apparent differences between values of kq in acetonitrile or
a less polar ‘‘non-hydrogen bonding’’ solvent (for example
chlorobenzene) are therefore non-existent.
Theoretical calculations
Fig. 4 Transient absorption spectra obtained after laser excitation
(355 nm) of 1 in an acetonitrile solution of DABCO (1.9 ꢃ 10ꢀ3 mol Lꢀ1
)
Given that the experimental analysis of the HA from phenols
by the triplet of 1 indicates that the reaction may occur via
HAT or PCET type mechanisms, we have theoretically
investigated the reaction mechanism using DFT calculations
and the AIM (atoms in molecules) theory. All computed
geometries of the S0 and T1 excited states were performed
without solvent effects.
at the indicated intervals of time after the laser pulse. Insert: an
example of a growth (390 nm), from which values of DOD were
determined at specific time intervals after the pulse (approx. 0.6 ms) for
each wavelength.
Fig. 6 shows the computed molecular orbitals at (U)HF/
6-31G*//(U)B3LYP/6-31G*. As can be seen, for the S0 state
both HOMO and LUMO show p character. The same is found
for the SOMO and SOMO-1 of the T1 state. Although
qualitative, this analysis reinforces the experimental results
which indicated a p,p* triplet state. Further, it is notable that
there is a very large contribution from the a-keto carbon of the
keto-vinyl ether to the SOMO of the T1 state.
Sakaguchi et al.12 have shown that flavones can abstract
hydrogen either by the oxygen atom of the keto group or by
the a-keto carbon. In order to test these possibilities with
phenol, transition states for hydrogen transfer to oxygen and
to carbon were localized. In the case of hydrogen transfer to
the a-keto carbon, the respective TS was found to be
17 kcal molꢀ1 higher in energy. Therefore, in the present study
we have rationalized the HA from phenol (PhOH) by the
triplet ketone, yielding a triplet radical pair, in a four step
mechanism. Initially, a ground state S0 complex (ketone/
phenol) involving a carbonyl oxygen interaction with the
phenolic hydrogen was located. This complex was then used
as a starting point to fully optimize the geometry of the T1
excited state complex. From this, the triplet TS for HA and the
triplet radical pair were also fully optimized. Fig. 7 shows the
calculated structures for the S0 ketone/PhOH complex (a), the
T1 ketone/PhOH complex (b), the triplet TS (c), and the triplet
radical pair (d).
Fig. 5 Transient absorption spectra obtained 2 ms after laser excitation
(355 nm) of 1: (A) in the absence of and (B) in the presence of
4-methoxyphenol (3.8 ꢃ 10ꢀ3 mol Lꢀ1) in an acetonitrile solution.
By building upon the work of Ingold and colleagues, we have
analyzed eqn (2) by an alternative means. We questioned as to
whether it was possible to determine a value of bH2 for a given
hydrogen bond acceptor by varying the hydrogen donor (aH2).
Using the LFP rate data, a value approximating to 0.42 for
bH2 for acetonitrile was determined from the gradient of a plot
Table 2 shows the computed geometric parameters for the
four computed geometries. Initially, in the S0 state of 1 the
exocyclic phenyl group is twisted at an angle of 241 relative to
the plane of the a-naphthoflavone ring system. However after
excitation to T1, and in all following computed structures, the
of the log(kq) against aH (full details are given in the ESIw).
2
Our experimentally determined value of bH is in excellent
2
agreement with the literature value of 0.44.79 This indicates
that HA from the phenols is dependent upon the equilibrium
ꢁc
This journal is the Owner Societies 2010
10750 | Phys. Chem. Chem. Phys., 2010, 12, 10746–10753

View Article Online
the TS. Further, from Table 2 one can follow the total charge
on the ketone oxygen through the reaction path. In a previous
study,20 the total charge on four triplet diketones reacting with
phenol along their respective reaction coordinates was
compared. In that study, n,p* and p,p* excited states revealed
large differences in charge distribution along the respective
reaction coordinates. The n,p* molecules revealed significant
charge transfer in the triplet complex whilst the p,p* triplet
only revealed charge transfer at the TS.20 Furthermore, the
n,p* excited state molecules revealed a greater extent of charge
transfer at the TS when compared to the respective p,p* triplet
TS. In this study, the calculations reveal that triplet 1 is only
reduced at the TS, thus resembling the previous result found
for 1,2-aceanthrylenedione, a diketone with a p,p* triplet.20
Fig. 8 details the relative energy data for the computed
reaction path. The theoretical value for the triplet energy
(61.3 kcal molꢀ1) of 1 is in good agreement with the experi-
mental results obtained using triplet quenchers and the
literature value for flavone (62 kcal molꢀ1).8 The T1 complex
energy is 1.60 kcal molꢀ1 lower than that for the respective
triplet TS. The small positive activation barrier found for HA
in the 1/phenol pair may also be indicative of a p,p* excited
state nature of 1. It is interesting to compare these results with
those from a study by Nicodem et al. for phenanthrenequinone.
In that case, no activation barrier for HA by the ortho-
quinone, even when the hydrogen donor was 2-propanol,
was observed in CCl4.82 Whilst in acetonitrile the experimentally
measured activation barrier was ca. 2 kcal molꢀ1, which lead
to a kq one order of magnitude lower than that in CCl4. These
results were interpreted as a change in the nature of the
reactive triplet state, from n,p* in CCl4 to p,p* in acetonitrile.
To further investigate the triplet reactivity of 1, a comple-
mentary study using AIM theory was performed. Leigh et al.23
have shown that phenolic HA reactions are consistent with a
continuum of two different mechanisms, which varies from a
simple electron transfer (which applies to the n,p* triplet
ketones and those p,p* triplets with low reduction potential)
to a PCET mechanism, involving the participation of a
hydrogen-bonded exciplex for p,p* triplets. de Lucas et al.,20
following the AIM methodology used by Singh et al.62 for
studying the mechanistic aspects of HA from phenolic
antioxidants, showed that the HA reactions for n,p* excited
state diketones follow a PCET like mechanism, and not a
HAT mechanism. These results were consistent with the highly
accentuated electrophilic nature of the excited state triplets of
both quinones. Following the same methodology, the trans-
ferring hydrogen atom (Htransferring) for the transition states
involving HA from phenol by 1, t-butoxyl radical and methyl
radical were evaluated by use of the AIM theory. Table 3
details three atomic properties obtained for the hydrogen atom
abstractions: the atomic volume and charge depicted are
related to the actual basin size and charge, respectively. The
atomic moment Ml is sensitive to the electron density far from
the nucleus.
Fig. 6 Molecular orbitals: (a) S0 HOMO, (b) S0 LUMO, (c) T1
SOMO-1 and (d) T1 SOMO.
exocyclic phenyl group is co-planar with the a-naphthoflavone
ring system.
When phenol is the hydrogen donor, the four computed
structures reveal a hydrogen bonded complex between both
reactants or both products (Fig. 7). As expected, the hydrogen
bond is longer in the T1 complex (r2 = 1.785 A) than in the S0
complex (r2 = 1.779 A) as the nature of the excited state of the
ketone is less basic. After reaching the triplet TS the Htransferring
atom is partially transferred to the ketone (r2 = 1.341 A) and
there is a minimum distance between the ketone carbonyl
oxygen atom and the phenol oxygen atom (2.444 A). However,
it is interesting to note that the phenolic O–H bond has
elongated by only 11% in comparison with the S0 complex,
whilst the CQOꢅ ꢅ ꢅHtransferring bond (r2) has shortened by
0.438 A (25% change) on going from the S0 complex to the
TS. Therefore, despite the reactants being at their closest
approach, hydrogen atom transfer is not very significant in
By comparison of the atomic properties for the Htransferring
atom in HA reactions with either a methyl or t-butoxyl radical
it is clearly seen that a similar polar effect observed with HA
by the t-butoxyl radical is operative in the photochemical HA
reaction of 1.14,21
Fig. 7 Calculated structures: (a) S0 complex, (b) T1 complex, (c) T1
transition state and (d) T1 radical pair.
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10746–10753 | 10751

View Article Online
Table 2 Computed geometric parameters for the hydrogen abstraction mechanism at (U)B3LYP/6-31G*
Bond distances/A
S0 complex
T1 complex
Transition state
T1 radical pair
Bond
r1 CQO
1.246
1.779
0.989
1.361
0.000
1.263
1.785
0.988
1.361
ꢀ0.031
1.295
1.341
1.103
1.314
ꢀ0.299
1.358
0.987
1.795
1.265
ꢀ1.000
r2 COꢅ ꢅ ꢅHOphenol
r3 H–Ophenol
r4 O–Cphenol
Ketone oxygen chargea
a
ChelpG ((U)B3LYP/6-311++G**) charges on the oxygen atom of the ketone.
the Htransferring atom reveals the HA reaction of 1 to be similar
to that observed by the electrophilic t-butoxyl radical,
consistent with a concerted, but not necessarily simultaneous,
PCET mechanism.
Acknowledgements
The authors wish to thank the Brazilian institutions FAPERJ,
CAPES and CNPq for financial assistance.
References
1 M. M. Caldwell, ed. A. C. Giese, Academic Press, New York,
1971, vol. VI, pp. 131–177.
2 J. B. Harborne and T. W. Goodwin, Plant Pigments, Academic
Press, Oval Road, London, 1987.
Fig. 8 Relative computed energies (not to scale) for the hydrogen
3 H. A. Stafford, Plant Physiol., 1991, 96, 680–685.
4 M. I. Wilson and B. M. Greenberg, Photochem. Photobiol., 1993,
57, 556–563.
abstraction at (U)B3LYP/6-311++G**//(U)B3LYP/6-31G*.
5 I. Yokoe, M. Taguchi, Y. Shirataki and M. Komatsu, J. Chem.
Soc., Chem. Commun., 1979, 333–334.
6 K. Hamanoue, T. Nakayama, T. Miyake and H. Teranishi, Chem.
Lett., 1981, 39–42.
In conclusion, a value for the hydrogen bond basicity
parameter bH was determined from the quenching rate
2
constant data of a series of phenols with the triplet of 1. This
value (0.42) was found to be in excellent agreement with the
literature value of bH2 (0.44) for acetonitrile. Therefore, it was
possible to correct the quenching rate constants for hydrogen
bonding of the phenols to acetonitrile. The resulting absolute
values for the quenching rate constants were found to be
remarkably similar to the diffusion constant for acetonitrile.
HA from phenols by the triplet of 1 can therefore occur
exclusively from free phenol in a diffusion controlled reaction
or from a pre-associated ground state ketone–phenol hydrogen
bonded complex (a pseudo-unimolecular process). DFT
calculations of geometric and charge parameters (where at
the TS for HA the PhO–H bond has elongated by only 0.114 A
and the total charge transfer from phenol to 1 is 0.299e)
indicate that the mechanism for the hydrogen atom abstraction
by 1 has substantial PCET character but that the degree of
electron transfer preceding the proton transfer is less than that
observed for more reactive n,p* ketones. The AIM analysis of
7 H. Kawata, T. Kumagai, E. Suzuki and S. Niizuma, J. Photochem.
Photobiol., A, 1996, 101, 201–204.
8 K. Bhattacharyya, D. Ramaiah, P. K. Das and M. V. George,
J. Phys. Chem., 1986, 90, 5984–5989.
9 V. Avila and C. M. Previtali, J. Chem. Soc., Perkin Trans. 2, 1995,
2281–2285.
10 M. Christoff, V. G. Toscano and W. J. Baader, J. Photochem.
Photobiol., A, 1996, 101, 11–20.
11 T. Nakayama, T. Shimizu, Y. Torii, S. Miki and K. Hamanoue,
J. Photochem. Photobiol., A, 1997, 111, 35–39.
12 Y. Sakaguchi, H. Hayashi, H. Murai and Y. J. Lhaya, J. Phys.
Chem., 1986, 90, 6416–6420.
13 P. K. Das, M. V. Encinas and J. C. Scaiano, J. Am. Chem. Soc.,
1981, 103, 4154–4162.
14 P. K. Das, M. V. Encimas, S. Steenken and J. C. Scaiano, J. Am.
Chem. Soc., 1981, 103, 4162–4166.
15 P. J. Wagner and B.-S. Park, in Organic Photochemistry, ed.
A. Padwa, Marcel Dekker, New York, 1991, pp. 227–365.
16 N. J. Turro, Modern Molecular Photochemistry, University Science
Books edn, Mill Valley, California, 1991, pp. 362–385.
17 N. C. de Lucas and J. C. Netto-Ferreira, J. Photochem. Photobiol.,
A, 1998, 116, 203–208.
18 N. C. de Lucas, M. T. Silva, C. Gege and J. C. Netto-Ferreira,
J. Chem. Soc., Perkin Trans. 2, 1999, 2795–2801.
19 A. C. Serra, N. C. Lucas and J. C. Netto-Ferreira, J. Braz. Chem.
Soc., 2004, 15, 481–486.
20 N. C. de Lucas, R. J. Correa, A. C. Albuquerque, C. L. Firme,
S. J. Garden, A. R. Bertoti and J. C. Netto-Ferreira, J. Phys.
Chem. A, 2007, 111, 1117–1122.
Table 3 Htransferring atomic properties in the transition state for
different reactant/phenol pairs compared with the phenolic hydrogen
atom
Reactive species
Volume (O)
Moment (O)
Charge (O)
21 N. C. de Lucas, M. M. Elias, C. L. Firme, R. J. Correa,
S. J. Garden, J. C. Netto-Ferreira and D. E. Nicodem,
J. Photochem. Photobiol., A, 2009, 201, 1–7.
22 C. Evans, J. C. Scaiano and K. U. Ingold, J. Am. Chem. Soc., 1992,
114, 4589–4593.
1
9.6
9.5
16.7
23.9
0.046
0.049
0.101
0.168
0.621
0.613
0.414
0.558
t-Butoxyl radical
Methyl radical
Phenol
ꢁc
This journal is the Owner Societies 2010
10752 | Phys. Chem. Chem. Phys., 2010, 12, 10746–10753

View Article Online
23 W. J. Leigh, E. C. Lathioor and M. J. S. Pierre, J. Am. Chem. Soc.,
1996, 118, 12339–12348.
56 K. Tanaka, S. Sakai, S. Tomiyama, T. Nishiyama and F. Yamada,
Bull. Chem. Soc. Jpn., 1991, 64, 2677–2680.
24 L. Biczo
119, 11071–11077.
25 J. Perez-Prieto, F. Bosca
and M. A. Miranda, J. Org. Chem., 2003, 68, 5104–5113.
26 J. Perez-Prieto, R. E. Galian, M. C. Morant-Minana and
´
k, T. Be
´
rces and H. Linschitz, J. Am. Chem. Soc., 1997,
57 S. Tomiyama, S. Sakai, T. Nishiyama and F. Yamada, Bull. Chem.
Soc. Jpn., 1993, 66, 299–304.
´
´
, R. E. Galian, A. Lahoz, L. R. Domingo
58 M. I. de Heer, P. Mulder, H. G. Korth, K. U. Ingold and
J. Lusztyk, J. Am. Chem. Soc., 2000, 122, 2355–2360.
59 P. J. O’Malley, Chem. Phys. Lett., 2002, 364, 318–322.
60 O. Tishchenko, E. S. Kryachko and M. T. Nguyen, J. Mol. Struct.,
2002, 615, 247–250.
61 Y.-M. Sun and C.-B. Liu, Eur. J. Org. Chem., 2004, 120.
62 N. Singh, P. J. O’Malley and P. L. A. Popelier, Phys. Chem. Chem.
Phys., 2005, 7, 614–619.
´
M. A. Miranda, Chem. Commun., 2005, 3180–3182.
27 E. C. Lathioor and W. J. Leigh, Photochem. Photobiol., 2006, 82,
291–300.
28 R. E. Galian, G. Litwinienko, J. Perez-Prieto and K. U. Ingold,
J. Am. Chem. Soc., 2007, 129, 9280–9281.
29 P. J. Wagner, Top. Curr. Chem., 1976, 66, 1–52.
30 P. J. Wagner, R. J. Truman and J. C. Scaiano, J. Am. Chem. Soc.,
1985, 107, 7093–7097.
63 J. Perez-Prieto, S.-E. Stiriba, F. Bosca, A. Lahoz, L. R. Domingo,
´ ´
F. Mourabit, S. Monti and M. A. Miranda, J. Org. Chem., 2004,
69, 8618–8625.
31 C. Walling and M. Gibian, J. Am. Chem. Soc., 1965, 87,
3361–3364.
32 S. G. Cohen, A. Parola and G. H. J. Parson, Chem. Rev., 1973, 73,
141–161.
33 S. Inbar, H. Linschitz and S. G. Cohen, J. Am. Chem. Soc., 1981,
103, 1048–1054.
34 J. D. Simon and K. S. Peters, J. Am. Chem. Soc., 1982, 104,
6542–6547.
35 H. Miyasaka, K. Morita, K. Kamada and N. Mataga, Chem. Phys.
Lett., 1991, 178, 504–510.
36 J. M. Mayer, D. A. Hrovat, J. L. Thomas and W. T. Borden,
J. Am. Chem. Soc., 2002, 124, 11142–11147.
37 J. M. Mayer, Annu. Rev. Phys. Chem., 2004, 55, 363–390.
38 C. Isborn, D. A. Hrovat, W. T. Borden, J. M. Mayer and
B. K. Carpenter, J. Am. Chem. Soc., 2005, 127, 5794–5795.
39 O. Tishchenko, D. G. Truhlar, A. Ceulemans and M. T. Nguyen,
J. Am. Chem. Soc., 2008, 130, 7000–7010.
64 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, A. G. Baboul,
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. G. Johnson, W. Chen, M. W. Wong, J. L. Andres,
M. Head-Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 98
(Revision A.7), Gaussian, Inc., Pittsburgh, PA, 1998.
65 R. Fletcher, Practical Methods of Optimization, Wiley, New York,
1980.
40 N. J. Turro and R. Engel, J. Am. Chem. Soc., 1969, 91, 7113–7121.
41 S. V. Jovanovic, D. G. Morris, C. N. Pliva and J. C. Scaiano,
J. Photochem. Photobiol., A, 1997, 107, 153–158.
42 D. V. Avila, K. U. Ingold, J. Lusztyk, W. H. Green and
D. R. Procopio, J. Am. Chem. Soc., 1995, 117, 2929–2930.
43 L. Valgimigli, J. T. Banks, K. U. Ingold and J. Lusztyk, J. Am.
Chem. Soc., 1995, 117, 9966–9971.
66 R. F. W. Bader, Chem. Rev., 1991, 91, 893–928.
67 R. F. W. Bader, Acc. Chem. Res., 1985, 18, 9–15.
68 S. L. Murov, I. Carmichael and G. L. Hug, Handbook of
Photochemistry, Marcel Dekler, Inc., New York, 1993.
69 J. C. Scaiano, J. Am. Chem. Soc., 1980, 102, 7747–7753.
70 M. V. Encinas and J. C. Scaiano, J. Am. Chem. Soc., 1981, 103,
6393–6397.
44 J. T. Banks, K. U. Ingold and J. Lusztyk, J. Am. Chem. Soc., 1996,
118, 6790–6791.
45 D. W. Snelgrove, J. Lusztyk, J. T. Banks, P. Mulder and
K. U. Ingold, J. Am. Chem. Soc., 2001, 123, 469–477.
46 G. Litwinienko and K. U. Ingold, J. Org. Chem., 2003, 68,
3433–3438.
47 G. Litwinienko and K. U. Ingold, J. Org. Chem., 2004, 69,
5888–5896.
48 G. Litwinienko and K. U. Ingold, J. Org. Chem., 2005, 70,
8982–8990.
71 D. Rehm and A. Weller, Isr. J. Chem., 1970, 8, 259.
72 G. T. Davis, M. M. Demek and D. H. Rosenblatt, J. Am. Chem.
Soc., 1972, 94, 3321–3325.
73 C. Devadoss and R. W. Fessenden, J. Phys. Chem., 1990, 94,
4540–4549.
74 D. A. Dunn and D. I. Schuster, J. Am. Chem. Soc., 1985, 107,
2802–2804.
75 G. Balakrishnan, T. Keszthelyi, R. Wilbrandt, J. M. Zwier,
A. M. Brouwer and W. J. Buma, J. Phys. Chem. A, 2000, 104,
1834–1841.
49 G. Litwinienko and K. U. Ingold, Acc. Chem. Res., 2007, 40, 222–230.
50 M. C. Foti, C. Daquino, I. D. Mackie, G. A. DiLabio and
K. U. Ingold, J. Org. Chem., 2008, 73, 9270–9282.
51 G. Horner, G. L. Hug, A. Lewandowska, F. Kazmierczak and
¨
B. Marciniak, Chem.–Eur. J., 2009, 15, 3061–3064.
52 G. Horner, G. L. Hug, D. Pogocki, P. Filipiak, W. Bauer,
¨
A. Grohmann, A. Laemmermann, T. Pedzinski and
B. Marciniak, Chem.–Eur. J., 2008, 14, 7913–7929.
53 G. Horner, A. Lewandowska, G. L. Hug and B. Marciniak,
¨
J. Phys. Chem. C, 2009, 113, 11695–11703.
54 P. A. MacFaul, K. U. Ingold and J. Lusztyk, J. Org. Chem., 1996,
61, 1316–1321.
55 A. L. Koner, U. Pischel and W. M. Nau, Org. Lett., 2007, 9,
2899–2902.
76 D. A. Pratt, G. A. Dilabio, P. Mulder and K. U. Ingold, Acc.
Chem. Res., 2004, 37, 334–340.
77 P. K. Das and S. N. Bhattacharyya, J. Phys. Chem., 1981, 85, 1391.
78 M. H. Abraham, P. L. Grellier, D. V. Prior, P. P. Duce,
J. J. Morris and P. Taylor, J. Chem. Soc., Perkin Trans. 2, 1989,
699–711.
79 M. H. Abraham, P. L. Grellier, D. V. Prior, J. J. Morris and
P. J. Taylor, J. Chem. Soc., Perkin Trans. 2, 1990, 521–529.
80 M. F. Nielsen and K. U. Ingold, J. Am. Chem. Soc., 2006, 128,
1172–1182.
81 N. J. Turro, V. Ramamurthy and J. C. Sciano, Principals of
Molecular Photochemistry: An Introduction, 2009.
82 D. E. Nicodem, R. S. Silva, D. M. Togashi and M. F. V. Cunha,
J. Photochem. Photobiol., A, 2005, 175, 154–158.
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10746–10753 | 10753