Hydrogen-atom transfer reactions from ortho-alkoxy-substituted phenols:An experimental approach

Article abstract of DOI:10.1002/chem.200802454

The role of intramolecular hydrogen bonding (HB) on the bond-dissociation enthalpy (BDE) of the phenolic O-H and on the kinetics of H-atom transfer to peroxyl radicals (k_inb) of several 2-alkoxyphenols was experimentally quantified by the EPR e

Full text of DOI:10.1002/chem.200802454

DOI: 10.1002/chem.200802454
Hydrogen-Atom Transfer Reactions from ortho-Alkoxy-Substituted Phenols:
An Experimental Approach
Riccardo Amorati,*[a] Stefano Menichetti,[b] Elisabetta Mileo,[a]
Gian Franco Pedulli,*[a] and Caterina Viglianisi[b]
Abstract: The role of intramolecular
hydrogen bonding (HB) on the bond-
dissociation enthalpy (BDE) of the
the case of polymethoxyphenols, signif-
both conformational isomers in which
the OH group points toward or away
from the oxygen in position 2 were de-
tected by FTIR spectroscopy and the
intramolecular HB strength was thus
estimated. The contribution to the
ꢀ
ꢀ
icant deviations from the BDE
N
values predicted by the additive effects
of the substituents were found. The
logarithms of the kinh constants in
cumene were inversely related to the
ꢀ
ꢀ
phenolic O H and on the kinetics of
H-atom transfer to peroxyl radicals
(kinh) of several 2-alkoxyphenols was
experimentally quantified by the EPR
equilibration technique and by inhibit-
ed autoxidation studies. These com-
pounds can be regarded as useful
models for studying the H-atom ab-
straction from 2-OR phenols, such as
many lignans, reduced coenzyme Q
and curcumin. The effects of the vari-
ꢀ
BDE
N
BDEACTHNGUTRENNU(G O H) of the ortho-OR substitu-
Evans–Polanyi plot with the same
slope of other substituted phenols and
a y-axis intercept slightly smaller than
that of 2,6-dimethyl phenols. In the
cases of phenols having the 2-OR sub-
stituent included in a five-membered
condensed ring (i.e, compounds 9–11),
ent in 9, corrected for intramolecular
HB formation, was calculated as
ꢀ5.6 kcalmolꢀ1. The similar behaviour
of cyclic and non-cyclic ortho-alkoxy
derivatives clearly showed that the pre-
ferred conformation of the OMe group
in ortho-methoxyphenoxyl radicals is
that in which the methyl group points
away from the phenoxyl oxygen, in
contrast to the geometries predicted by
DFT calculations.
ous substituents on the BDEACTHNUTRGNE(NUG O H) of
2-methoxy, 2-methoxy-4-methyl, 2,4-di-
methoxyphenols versus phenol were
measured in benzene solution as ꢀ1.8;
ꢀ3.7; ꢀ5.4 kcalmolꢀ1, respectively. In
Keywords:
antioxidants
·
bond energy · EPR spectroscopy ·
hydrogen transfer · phenols
Introduction
anti-inflammatory, anti-angiogenic, wound healing and anti-
cancer effects.[1] The antioxidant and radical scavenging ac-
tivity of curcumin is deeply connected to the presence of
two phenolic moieties ortho-substituted by OMe groups,[2]
although in certain conditions the initial deprotonation of
the central cheto-enolic functionality may trigger the scav-
enging reaction.[3] The formation of 2-OMe phenoxyl radi-
cals is also the first step of the radical–radical coupling reac-
tions of coniferyl alcohol (2) to form lignans[4] such as (+)-
pinoresinol (3), which are gaining increasing interest thanks
to their biological properties.[5] For instance, podophyllotox-
in and its semisynthetic derivative etoposide are well known
Phenols bearing a methoxy group in the ortho-position rep-
resent an important class of natural compounds. Curcumin
(1) has attracted the attention of many scientists due to its
[a] Dr. R. Amorati, Dr. E. Mileo, Prof. G. F. Pedulli
Dipartimento di Chimica Organica “A. Mangini”
Universitꢀ di Bologna, Via S. Giacomo 11
40126 Bologna (Italy)
Fax : (+39)051-209-5688
E-mail: Riccardo.amorati@unibo.it
Gianfranco.pedulli@unibo.it
for their antiviral and antitumoral activity, respectively.[6]
A
[b] Prof. S. Menichetti, Dr. C. Viglianisi
Dipartimento di Chimica Organica e
Laboratorio di Progettazione Sintesi e
Studio di Eterocicli Bioattivi (HeteroBioLab)
Polo Scientifico e Tecnologico Universitꢀ di Firenze
Via della Lastruccia 13, 50019 Sesto Fiorentino (Italy)
deeper knowledge of the elementary reactions leading to
the formation of 2-OMe phenoxyl radicals is therefore a
prerequisite for the rational biomimetic synthesis of lignans
and related compounds and for the prediction of their anti-
oxidant properties.[5]
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802454.
4402
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4402 – 4410

FULL PAPER
ꢀ
Table 1. Literature data for the differences between the O H bond-dis-
sociation enthalpies of some ortho-methoxyphenols and that of phenol,
measured by different techniques [kcalmolꢀ1].[a]
ArOH
PAC[b]
EPR[c]
UV/Vis[d]
DFT[e]
2-OMe
+0.4
ꢀ6.3
ꢀ4.7
ꢀ9.3
ꢀ2.2
ꢀ5.8
ꢀ7.0
ꢀ9.4
2,6-(OMe)2
2,4-(OMe)2
2,4,6-(OMe)3
ꢀ4.4
ꢀ7.6
ꢀ4.5
[a] Including DHintra-HB
. [b] Photoacoustic calorimetry, gas phase (see
Table 1 in ref. [13]). [c] EPR equilibration, in benzene.[15] [d] UV/Vis
C
study of the equilibration reaction with the dpph (1,1-diphenyl-2-picryl-
hydrazyl) radical, in hexane.[16] [e] DFT calculations at the B3LYP/6-
31G* level, gas phase.[13]
The oxidation of phenols to phenoxyl radicals is indeed a
fundamental reaction in life science. Some enzymes,[7] the
respiratory[8] and the photosynthetic systems[9] exert their
redox chemistry thanks to the formation of tyrosyl radicals
or to the presence of phenolic co-factors, such as ubiquinol
in the case of the respiratory chain, which possesses two
ortho-methoxyphenolic functionalities.[8]
The occurrence of H-atom transfer reactions from 2-OMe
phenols to free radicals is quite surprising, considering that
the formation of a hydrogen bond (HB) between a phenolic
OH group and a HB-accepting solvent is known to suppress
phenol reactivity, causing the so called “kinetic solvent
effect” (KSE).[10,11] However, it has been shown that the
suppression of reactivity occurs only in the case of intermo-
lecular HB formation, because the linear geometry of such
interactions prevents the approach of the attacking radi-
cal.[10] Actually, in 2-OMe phenols the intramolecular HB is
bent so that the OH group is still available for interacting
with the approaching radical, as well as with the solvent, as
demonstrated by the detection of some KSE on these com-
pounds.[12]
Table 1),[13,15,16] but also by the lack of reliable experimental
determinations of the strength of the intramolecular hydro-
gen bond (DHintra-HB) in the parent phenol. In the majority
of ortho-methoxyphenols the equilibrium between the
“away” (4a) and “toward” (4b) conformations of the phe-
nolic OH is completely shifted to the right, preventing any
easy quantification of the two isomers by IR or other spec-
troscopic techniques.[17] Actually, the DFT-calculated value
of ꢁꢀ4.4 kcalmolꢀ1 for the equilibrium 4b/4a has been
adopted by several authors,[13,14,20] because of the scattering
(from ꢀ2.0 to ꢀ4.1 kcalmolꢀ1) of the DHintra-HB values re-
ported in the literature.[21]
One important parameter controlling the homolytic re-
With the aim to better clarify the factors controlling the
A
reactivity of ortho-methoxyphenols toward free radicals, we
ꢀ
of the phenolic OH. In the case of ortho-methoxyphenols,
investigated phenols 4–11 by measuring their BDE
N
ꢀ
the variation of the BDE
N
values and the rate constants for their reaction with peroxyl
radicals. In the case of cyclic phenols 9–11, it was possible to
detect both conformational isomers of the OH group and to
determine the strength of intramolecular HB by IR spec-
troscopy, thus allowing the first experimental determination
of the electronic effect of an ortho-alkoxy group.
phenol is due to: 1) a decrease induced by the electron-do-
nating effect of the ortho-OMe group, and 2) an increase
due to the formation of an intramolecular HB between the
OH and OMe groups.[13,14] The latter effect constitutes the
main reason why, in apolar solvents, guaiacol (4) is charac-
ꢀ
terised by a higher BDEACTHNUTRGNE(UNG O H) and a lower reactivity
toward free radicals than 4-methoxyphenol, despite the pres-
ence in both phenols of the strongly electron-donating OMe
group in positions conjugated with the reactive centre.[12] On
the other hand, in polar solvents, the presence of an ortho-
OMe group partially protects the phenolic OH from inter-
molecular hydrogen bonding with the solvent molecules.
Thus, it can be expected that in biological environments the
difference between the reactivities of 4 and 4-methoxyphe-
nol is less.[12]
A detailed investigation of the relative importance of the
ꢀ
two contributions to the BDEACTHNUTRGNEUGN(O H) of ortho-methoxyphe-
nols, that is, the strength of the intramolecular HB and the
radical stabilisation by the OMe substituent, is made diffi-
cult not only by the differences between the reported BDE
values obtained by various techniques (see, for instance,
Chem. Eur. J. 2009, 15, 4402 – 4410
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4403

R. Amorati, G. F. Pedulli et al.
Results and Discussion
FTIR spectra: To evaluate the strength of the intramolecu-
lar HB, the IR spectra of phenols 4–11 were recorded in
CCl4 solution, using diluted concentrations to avoid auto-as-
sociation. The data, reported in Table 2, show as expected
ꢀ
Table 2. Frequencies of the O H stretching vibration and strength of the
intramolecular hydrogen bond measured by IR spectroscopy in CCl4 at
room temperature.
[a]
[a]
[b]
ꢀ
ꢀ
ArOH
n
G
n
(O H)
DHintra-HB
toward
away
[c]
[c]
[c]
[c]
[c]
4
5
6
7
8
9
10
11
3557
3561
3567
3533
3554
3582
3590
3589
–
–
–
–
–
–
3607
3606
3607
–
Figure 1. IR spectra of guaiacol (4) and of the cyclic compounds 9 and 11
(6ꢃ10ꢀ3 m in CCl4) in the O H bond stretching region. Arrows indicate
ꢀ
the shoulders due to the non-HB conformers and the peaks due to the
HB species.
ꢀ2.6
ꢀ1.6
ꢀ1.6
–
to the toward conformation, while keeping blocked the
other substituents (such as from 4a to 4b),[13,14] the DHintra-HB
values were obtained from the equilibrium constants assum-
ing DSꢁ0 calmolꢀ1 Kꢀ1 (see Table 2 and Supporting Infor-
mation).[13,24] Note that the DHintra-HB values for 9–11 are
much smaller than that reported for 4 (ꢀ4.4 kcalmolꢀ1).[13]
In addition, the value for 9 is 1 kcalmolꢀ1 larger than those
ꢀ
phenol
4-OMePhOH
3611
3617
–
–
–
[a] In cmꢀ1. [b] In kcalmolꢀ1, error: ꢂ0.2 kcalmolꢀ1. [c] Not calculable by
IR spectroscopy.
ꢀ
that the O H stretching frequencies [nACTHNUGTRENNU(G OCAHTUNTGERNHNUGN )] of ortho-
alkoxy phenols are lower than those of ortho-unsubstituted
phenols, due to the formation of an intramolecular HB.
Notice that the presence of electron-donating groups para
to the phenolic OH (i.e., compounds 5 and 6 vs. 4) causes
ꢀ
for 10 and 11, consistent with the different n
of their toward conformers (Table 2).
ACHTUNGTRENUN(NG O H) values
The smaller intramolecular HB strength in phenols 9–11
can be explained by remembering that DHintra-HB is defined
as the enthalpy difference for equilibrium 4b/4a. The
DHintra-HB value is therefore the sum of two contributions,
one depending on the attraction between the OH and the
lone pair of the ortho-OR groups in 4b, and the other one
on the repulsion between the two oxygen atoms in the away
conformation 4a.[14] Korth and co-workers estimated that in
4 the latter term may contribute up to 2 kcalmolꢀ1 to the
overall HB strength.[14] In phenols 9–11, the straining effect
only slight changes in the nACTHNURGTNE(UNG O H). The substituted phenol 7
ꢀ1
ꢀ
represents an exception, as the nACTHNUTRGNE(NUG O H) is 24 cm lower
than that of guaiacol (4). This peculiarity was explained by
Schaefer and co-workers in terms of mutual repulsion be-
tween the two methoxy substituents, which has the effect of
pushing the 2-OMe group out of the molecular plane. It was
1
also estimated, by H NMR experiments, that this change of
geometry of 7 increases the intramolecular HB strength by
about 0.5 kcalmolꢀ1 with respect to planar ortho-methoxy-
phenols.[22]
ꢀ
of the penta-atomic condensed ring implies longer O
ꢀ
The most interesting results were obtained in the case of
the cyclic compounds 9–11. As shown in Table 2 and
ꢀ
H···OR and H O···OR distances, so both contributions are
expected to be lower, thus causing a relatively large de-
Figure 1, their n
(O H) values are shifted to higher frequen-
ACHTUGNNERTNcUNG rease in DHintra-HB values.
cies than those of simple ortho-methoxyphenols, and there is
a shoulder at about 3610 cmꢀ1, which is particularly evident
for dioxoles 10 and 11. These results clearly indicate that in
9–11 the intramolecular HB is weaker than in the other in-
ꢀ
EPR spectra and determination of BDEACHTNUTRGNEUNG(O H) values: The
ꢀ
O H bond-dissociation enthalpies in phenols 4–11 were de-
termined by measuring, by means of EPR spectroscopy, the
equilibrium constant, K1, for the hydrogen-atom transfer re-
action between the investigated molecule (ArOH), a refer-
ence phenol (Ar’OH) and the corresponding phenoxyl radi-
cals generated under continuous photolysis (see [Eq. (1)]
and Figure 2).[25] The EPR spectral parameters of the arylox-
yl radicals obtained by H-atom abstraction from 4–11 are re-
ported in the Experimental Section:
ꢀ
vestigated compounds, and that the O H absorption con-
tains a small contribution from the isomer with the OH
group in the non-HB away conformation.
By analysing the spectral patterns by numerical simulation
(see Experimental Section), we obtained the O H stretch-
ing frequencies reported in Table 2 and we could also esti-
mate the concentration of the away conformers by compari-
son of the areas of the higher-frequency peaks with that of a
reference compound at the same concentration (i.e., 4-me-
thoxyphenol).[23] Following Mulderꢂs suggestion to define
the hydrogen-bond strength in ortho-substituted phenols as
the energy gained by rotating the OH group from the away
ꢀ
K1
0
0
C
C
ArOH þ Ar O
ArO þ Ar OH
ð1Þ
G
H
The BDEs for the species ArOH (Table 3) were calculat-
ed, on the assumption that the entropic term can be neglect-
4404
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4402 – 4410

Alkoxy-Substituted Phenols
FULL PAPER
were corrected for association with the solvent mixture[27] by
downscaling the experimental results by 0.7 kcalmolꢀ1 (see
Supporting Information for details).
ꢀ
The experimental BDE
N
the sum of the bond energies of the toward and away con-
formers, corrected for their relative concentrations. From
the DHintra-HB values available from IR studies for 9–11 and
from DFT calculations for 4–6 and 8,[13] the contribution of
ꢀ
the away species to the experimental BDEACTHNUTRGENU(GN O H) was found
ꢀ
ꢀ
to be negligible (that is, BDE
N
ACHTUNGTRENNUNG(O
ꢀ
H)experimental). The DBDE
A
for the away species represent the contribution of the sub-
stituents to the phenolic BDE in the absence of the intramo-
lecular HB, that is, the step 9b!9’ in Scheme 1.
Figure 2. Experimental (top) and computer-simulated (bottom) EPR
spectrum observed at room temperature upon photolysing a deoxygenat-
ed benzene solution of 9 and 2,6-di-tert-butyl-4-cyanophenol in a concen-
tration ratio of 1:4 in the presence of 5% di-tert-butylperoxide. The outer
doublets are due to the phenoxyl radical from 9.
ꢀ
Scheme 1. Contribution of the substituents to the bond-dissociation en-
thalpy (kcalmolꢀ1) for compound 9, with respect to phenol. The energies
relative to the steps 9a!9b and 9a!9’ were experimentally obtained by
IR spectroscopy and by the EPR equilibration technique, respectively.
Table 3. Dissociation enthalpies of the O H bond with respect to phenol
[a]
(DBDE, kcalmolꢀ1
)
and reactivity toward peroxyl radicals (kinh
,
[b]
mꢀ1 sꢀ1
)
for the toward and away conformers of ortho-alkoxy phenols.
ArOH
toward
ꢀ
(O H)
away
kinh/104
DBDEACTHNGUETRNNU(G O H)
kinh/104
ꢀ
DBDE
N
4
5
6
7
8
9
10
11
ꢀ1.8
ꢀ3.7
ꢀ5.4
–
0.47
1.2
4.8
–
–
–
–
–
–
–
–
–
–
In the case of compound 9 the contribution of the fused
ꢀ1
ꢀ
ring to the DBDE
G
0.18
1.8
Scheme 1), can be considered as the sum of the effects of
the ortho-OR and the meta-alkyl substituent (ꢀ0.5 kcal
molꢀ1).[15] The ortho-effect of the non-HB alkoxy group is
ꢀ4.4[c]
ꢀ3.5
1.4[d]
0.6[d]
0.6[d]
ꢀ6.1
86[d]
32[d]
38[d]
ꢀ2.1 (ꢀ2.8)[e]
ꢀ3.7 (ꢀ4.4)[e]
ꢀ2.3 (ꢀ3.0)[e]
ꢀ3.9 (ꢀ4.6)[e]
therefore ꢀ6.1ꢀ
(ꢀ0.5)=ꢀ5.6 kcalmolꢀ1. Interestingly,
[a] Solvent: benzene, 298 K, error: ꢂ0.2 kcalmolꢀ1. BDE
AHCTUNGTRENNG(UN O H) phenol=
similar result (ꢀ6.2 kcalmolꢀ1) can also be obtained for 4 if
ꢀ
86.5 kcalmolꢀ1
.
[b] Error: ꢂ10%; stoichiometric coefficients n are in
[27]
combining its experimental DBDE
(O H) value found
the range 2.1ꢂ0.2. [c] From [15]. [d] Calculated by using Equations (11)
and (12) from the experimental kinh values (2.4, 2.7, 3.1ꢃ104 mꢀ1 sꢀ1 for 9,
10 and 11, respectively). [e] Values corrected for association with compo-
nents of the reaction mixture, see Supporting Information.
herein (ꢀ1.8 kcalmolꢀ1) with the intramolecular HB
strength (ꢀ4.4 kcalmolꢀ1) obtained by DFT calculations.[13]
That the contribution of an ortho-OMe group to the BDE-
ꢀ
AHCTUNGTREGUN(NN O H) is larger than that of one in the para-position
(ꢀ4.8 kcalmolꢀ1),[15] contrary to what is usually observed
with other substituents,[15] can be explained by considering
that the energy difference between species 9b and 9’ does
not only depend on the electronic effect of the methoxy
group, but also on the repulsion difference between the adja-
cent oxygen atoms.[28] A repulsion decrease passing from 9b
to 9’ may therefore account for the ꢁ1-kcalmolꢀ1 larger
effect for a non-HB OR substituent in ortho-position.[29]
ꢀ
ed,[25] by means of Equation (2) from K2 and the known
ꢀ
BDE
G
either 2,6-di-tert-butyl-4-cyanophenol, or, in the case of 6,
2,6-di-tert-butyl-4-methylphenol (BHT), whose recently re-
ꢀ1
ꢀ
vised BDE
G
spectively:[15,26,27]
BDEðArOꢀHÞ ¼ BDEðAr0OꢀHÞ ꢀ RT lnK1
ð2Þ
Upon subtracting from the DBDEACTHNUTRGNEUNG(O H) values of com-
pound 10 the additive contribution of a meta alkoxy group
(ꢀ1.0 kcalmolꢀ1)[30] and from 11 those of a meta alkoxy and
of a meta methyl group (ꢀ0.5 kcalmolꢀ1), the effects of the
non-HB ortho-OR substituents were calculated as ꢀ3.4 and
As preliminary studies showed that 10 and 11 give rise to
rather strong intermolecular HB interactions with HB-do-
ꢀ
nating solvents, differently to 9, their BDEACHTNUTRGNE(UNG O H) values
Chem. Eur. J. 2009, 15, 4402 – 4410
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4405

R. Amorati, G. F. Pedulli et al.
ꢀ3.1 kcalmolꢀ1, respectively, that is, approximately 2 kcal
molꢀ1 smaller than that found for 9. Unfortunately, an esti-
mate of the electronic effect of two OMe substituents in
ortho- and meta-positions on the phenolic BDE is not possi-
ble, as the open analogue 7 did not afford sufficiently per-
number of peroxyl radicals trapped by each antioxidant mol-
ecule, n, was experimentally determined from Equation (10),
in which Ri is the rate of free-radical initiation obtained
using 2,2,5,7,8-pentamethyl-6-chromanol (PMHC) as refer-
ence antioxidant, for which n=2:[28,31]
ACHTUNGTRENNUNGsistACHTUNGTRENNUNGent phenoxyl radicals. The smaller effect of the non-HB
ortho-OR substituents in 10 and 11 with respect to 9 could
be ascribed to the electron-withdrawing effect exerted by
the b-ester groups. However, a more convincing explanation
is that the contribution of the meta substituents on the
ꢀ
kp ½cumeneꢄ
ð9Þ
ꢀD½O2ꢄt ¼
lnð1ꢀt=tÞ
kinh
n½ArOHꢄ
ð10Þ
Ri ¼
BDEACHTUNGTRENNUNG(O H) of 10 and 11 is smaller than that predicted by
t
the additive rules. Actually, the meta effect of the OMe
group has been reported to decrease dramatically as the
substitution on the aromatic ring increases.[30]
In the case of compound 7, which did not afford a clear
induction period, the kinh value was estimated by the slope
of the oxygen consumption by using a different kinetic treat-
ment previously reported,[31b] assuming n=2 (see Supporting
Information).
Kinetics with peroxyl radicals: The rate constants for the re-
action of the investigated phenols with peroxyl radicals, kinh
(see Figure 3 and Table 3) were measured by studying the
The kinh value obtained in the present study for 5 is about
one order of magnitude smaller than that reported by Bar-
clay and co-workers (1.4ꢃ105 mꢀ1 sꢀ1)[33] at 308C using sty-
rene as an oxidizable substrate and Equation (9). We believe
that this high kinh value depends on the fact that ortho-me-
thoxyphenols are very weak inhibitors of styrene autoxida-
tion, showing only a slight retarding effect on the radical
chain reaction. Under these conditions, Equation (9) is no
longer valid, as it assumes that peroxyl radicals disappear
only by reacting with the antioxidant [Eq. (7) and (8)].[28,31a]
C
In the case of weak inhibition, ROO radicals disappear also
through the reaction shown in Equation (6), thus Equa-
tion (9) is expected to give overestimates of the antioxidant
activity.[34] This point of view is also supported by the recent-
ly measured rate constant for the reaction between an a-
aminoalkylperoxyl radical and 8, 2.4ꢃ104 mꢀ1 sꢀ1,[35] in rea-
sonable agreement with the value obtained herein (1.8ꢃ
104 mꢀ1 sꢀ1).
The low kinh value for compound 7 can be explained by
considering the higher strength of the intramolecular HB
detected by IR spectroscopy and the lack of planarity of the
ortho-OMe group,[22] so that the transition state of the reac-
tion of H-atom transfer can not be efficiently stabilised by
resonance.
Figure 3. Oxygen-consumption traces observed during AIBN (0.05m)-ini-
tiated cumene (6.2m) autoxidation in chlorobenzene solution at 308C, in
the absence (U) and presence of phenols 4, 6, 7 and 8 (6.3ꢃ10ꢀ6 m).
inhibited autoxidation of cumene, initiated by azobis(isobu-
tyronitrile) (AIBN) at 308C, in the presence of small
amounts of compounds 4–11 [Eq. (3–8)].[28,31] The reaction
was followed by monitoring the oxygen consumption with
an automatic recording gas-absorption apparatus previously
described:[31b]
To evaluate the reactivity of the toward conformer of the
investigated phenols, the contribution given by the away one
is to be subtracted to the experimentally determined kinh
values. This was done by using Equation (11), in which Xt,
kinh(t) and Xa, kinh(a) represent the molar fractions and the re-
activities of the toward and away conformers, respectively:
Ri
C
initiator
R
ð3Þ
ð4Þ
ð5Þ
ð6Þ
ð7Þ
ð8Þ
ƒ!
C
C
R þ O2 ! ROO
kp
C
C
ROO þ RH
ROOH þ R
2kt
ƒ!
C
C
ROO þ ROO
non-radical products
ƒ!
kinhðexpÞ ¼ Xt ꢅ kinhðtÞ þ Xa ꢅ kinhðaÞ
The reactivity of the away species was estimated from the
ð11Þ
kinh
C
C
ROO þ AzOH
ROOH þ ArO
ƒ!
C
C
ROO þ ArO ! non-radical products
linear Evans–Polanyi relationship between logACHTUNRGTNEUNG(kinh) versus
ꢀ
During the inhibition period, the oxygen consumption is
given by Equation (9), in which kp is the propagation rate
constant of the oxidizable substrate (0.32mꢀ1 sꢀ1 for
cumene)[32] and t is the length of the induction period. The
BDE
G
assuming that this conformer behaves as a phenol without
ortho-substituents [see Eq. (12), obtained from Figure 3].
From the results reported in Table 3, it is evident that the
4406
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4402 – 4410

Alkoxy-Substituted Phenols
FULL PAPER
reactivities of 9–11 are almost entirely due to the away con-
former, despite its molar fraction being less than 10%. On
the other hand, in the case of compounds 4–6, the away spe-
cies contributes negligibly to the overall kinh value,[12] on the
basis of the Xa molar fraction calculated from the reported
stant of the medium by the PCM continuum model,[20]
showed that the preferred conformation in polar solvents
(i.e., MeCN and acetone) is 4’a, whereas in apolar solvents
(n-heptane and benzene) it is 4’b.
DHintra-HB of ꢀ4.4 kcalmolꢀ1 [13]
:
log ðkinhðaÞÞ ¼ ꢀ0:35 BDEðOꢀHÞaway þ 34:24
ð12Þ
Upon plotting log(kinh(t)) values, resulting from this proce-
AHCTUNGTRENNUNG
ꢀ
dure, against the BDEACHTNUTGRNE(UNG O H)toward as shown in Figure 4, a
This kind of interaction cannot be a priori discarded, be-
cause in many biological molecules (such as DNA and pro-
ꢀ
teins) stabilising C H···O “hydrogen bonds” have been pro-
posed.[36] Moreover, in the present case the OMe hydrogen
atoms are activated by the strong electron-withdrawing
effect of the phenoxyl oxygen. However, some lines of evi-
dence suggest that such interaction, if present, is very weak
and does not influence the behaviour of the studied com-
pounds. Actually, Figure 4 shows that the values for all the
investigated phenols lie on the same line, independent of
the conformational freedom of the ortho-alkoxyl substituent.
If the preferred conformation of the phenoxyl radicals from
4–6 and 8 was that of 4’b, their kinh values should be lower
than those expected on the basis of their experimental
ꢀ
ꢀ
Figure 4. Plot of BDE
G
N
*
compounds ( ) and for some reference phenols: ( ) 2,6-tBu-4-XPhOH;
~
(
)
2,6-Me-4-XPhOH; (&) 4-XPhOH.[15, 25,26,28] The kinh value for 4-
XPhOHꢂs and for 2,6-tBu-4-CN phenol were measured in the present
work, see Experimental Section.
good linear relationship was obtained ([Eq. (13)], r2 =0.92,
excluding 4), characterised by about the same slope of 2,6-
alkyl substituted phenols:
BDEACTHUNGETRNNU(G O H), because the ortho-OMe group can rotate only
after the H-atom is removed.[37] Therefore, we should ob-
serve two distinct, parallel regression lines, one for the cyclic
compounds 9–11 and one for 4–8, the latter having a lower
intercept on the y axis.
log ðkinhðtÞÞ ¼ ꢀ0:31 BDEðOꢀHÞtoward þ 29:31
We also compared the experimental hydrogen isotropic
hyperfine splitting constants (hfsc) for 4’ with those calculat-
ed at the B3LYP/6-311++GACTHNUTRGNE(NUG d,p) level of theory for the two
isomers. As shown in Table 4, hfsc calculated for 4’a are in
much better agreement with experimental values than those
relating to the isomer 4’b. In particular, the hfsc at the me-
thoxy hydrogen atoms seems to be strongly dependent on
the OMe group conformation.
The intercept on the y axis suggests that the ortho-OR
group exerts a slightly larger steric shielding on the phenolic
OH than an ortho-methyl group. This is in line with the re-
cently reported Arrhenius parameters for the reaction of
phenols with dpph (diphenylpicrylhydrazyl) radicals, which
show that ortho-methoxyphenols are characterised by pre-
exponential A-factors lower than those of unsubstituted
phenols, and similar to 2,6-dimethylphenols.[16b]
Do substituent additivity rules hold for polymethoxyphe-
Conformation of ortho-methoxyphenoxyl radicals: Calcula-
tions performed at the DFT level predict that the more
stable conformation of ortho-methoxyphenoxyl radicals is
that with the CH3 group pointing toward the phenoxyl
oxygen, 4’b.[13,20] The DH value of this equilibrium, approxi-
mately ꢀ1.6 kcalmolꢀ1 in the gas phase, seems to suggest
nols? Mulder and co-workers suggested that there is linear
Table 4. Isotropic hyperfine splitting constants [Gauss, absolute values]
of the hydrogen atom bonded to the various carbons of the phenoxyl rad-
ical from 4, experimentally measured in benzene and calculated for 4’a
and 4’b at the B3LYP/6-311++GACTHNUGRTENUNG(d,p) level.
C
the existence of some strong OCH3···O interaction, able to
Position
Exptl
4’a
4’b
overcome the steric repulsion between the two groups. For
comparison, the energy of ortho-dimethoxybenzene in the
planar conformation 12b is computed to be +2.8 kcalmolꢀ1
larger than that of 12a.[18] Interestingly, DFT calculations
performed on 4’ by taking into account the dielectric con-
C3-H
C4-H
C5-H
C6-H
OCH3
1.96
8.73
0.94
5.23
1.58
2.1
7.9
1.4
5.4
1.6
1.9
6.4
0.6
3.9
3.3
Chem. Eur. J. 2009, 15, 4402 – 4410
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4407

R. Amorati, G. F. Pedulli et al.
ꢀ
correlation between the BDEACTHNUGTRNEUNG(O H) values measured by
photoacustic calorimetry of several ortho-methoxyphenols
and the sum of the Brown substituents constant s+, assum-
ing that sortho+ =0.66 spara+, after correcting BDE values for
intramolecular HB formation.[13] This would suggest that the
electronic effect of the methoxy substituents is additive, that
ꢀ
is, a OMe group in para position decreases the BDEACTHUNTGRNEUNG(O H)
by the same quantity in phenol, 2-methoxyphenol and 2,6-
dimethoxyphenol. From the data reported in Table 3 it is
evident that this is not the case, the effects being ꢀ4.8, ꢀ3.6
and ꢀ3.2 kcalmolꢀ1, respectively. Moreover, the lack of ad-
ditive contributions is striking if the effect of the non-HB
ortho-OR group measured for 9 (ꢀ5.6 kcalmolꢀ1) is com-
ꢀ
ꢀ
Figure 5. Relationship between the DBDE
(O H) of 1–3, 5 and 2,4,6-tri-
methoxyphenol,[15] and the hyperfine splitting constant of their phenoxyl
radicals with the ortho-OCH3 protons; r2 =0.96.
pared with the BDEACTHNUGRTENUNG(O H) difference between 8 and 4
(ꢀ2.6 kcalmolꢀ1). Actually, the effect of a second non-HB
ortho-OMe group is 3 kcalmolꢀ1 smaller than that of the
first one.
phenolic moiety, a quantitative evaluation of the opposite
effects of the ortho-methoxy substituent on the phenolic
ꢀ
We therefore suggest that, at least in the case of the
strongly p-conjugating OMe group, the substituent effect on
ꢀ
BDE
(O H), that is, the BDE-lowering contribution given
by the electron-donation and the BDE-increasing effect due
to the formation of an intramolecular hydrogen bond, is still
lacking. Among the ortho-alkoxy phenols studied in the
present work, those having the 2-OR substituent included in
a five-membered ring (i.e., compounds 9–11) showed a
weak intramolecular HB, which enabled the direct determi-
nation by IR spectroscopy of the relative amounts of the
toward and away conformations of the OH group and thus
the strength of the intramolecular HB. With the exception
of 2,3-dimethoxyphenol, all the investigated compounds
gave sufficiently persistent EPR spectra, so that their BDE-
ꢀ
the BDEACHTUNGTRENNUNG(O H) is not additive, but it experiences some
“saturation”.[38] As the OMe substituent lowers the BDE-
ꢀ
(O H) by donating electron density to the strongly elec-
C
tron-withdrawing O group, through resonance structures
with charge separation (see Scheme 2),[26] a second OMe
group is expected to have a smaller effect because electron
C
density has already been transferred to the O group by the
first OMe substituent.[38]
AHCTUNGTREG(NNUN O H) could be determined by the EPR equilibration tech-
nique. These data, in conjunction with the strengths of intra-
molecular HB, were used for a quantitative evaluation of
ꢀ
the factors determining the BDE
phenols.
(O H) in ortho-alkoxy
Scheme 2. Polar resonance structures in the ortho-methoxyphenoxyl radi-
cal.
The results obtained in the case of 9 showed that the elec-
tron-donating effect of the ortho-OR group accounts for a
ꢀ1
ꢀ
lowering of the BDE
G
To find experimental evidence for this effect, we exploited
the fact that polar resonance structures are characterised by
positive spin density on the methoxy oxygen, which in turn
increases the splitting constant with the methoxy protons
ꢀ
larger effect than that of a methoxy group in para position
(4.8 kcalmolꢀ1). This result was explained by taking into ac-
count the reduction of the repulsive interaction between the
two oxygen atoms when going from the non-HB parent
phenol to the phenoxyl radical.
(see Scheme 2). If the BDEACTHNUTRGNE(NUG O H) values are plotted against
the hfsc of the OMe protons, a linear relationship is found
(Figure 5), showing that the non-additive behaviour of poly
OMe substitution depends mainly on electron distribution in
the radical. It should be stressed that this effect is particular-
ly evident only in polymethoxyphenols, whereas in those
phenols with alkyl substituents and one OMe group, devia-
tions from additive rules are less important.[15,23]
By comparing the results obtained with several polyme-
thoxyphenols, it was evidenced that the additive contribu-
tion of the OMe group to the BDE(OH) is not additive, but
experiences a strong reduction if other OMe groups are al-
ready present on the aromatic ring.
The similar behaviour of cyclic and non-cyclic ortho-
C
alkoxy derivatives clearly showed that the OCH3···O inter-
action, predicted by DFT calculations in all ortho-methoxy-
phenoxyl radicals in the gas phase,[13] is not relevant under
the present experimental conditions. This drawback of DFT
calculations should be taken into account, particularly if
considering the popularity of DFT methods in the determi-
Conclusion
H-atom transfer from 2-OMe phenols is a key reaction in
the biosynthesis of lignans, in the antioxidant action of cur-
cumin and phenolic acids and in the radical chemistry of
coenzyme Q. Despite the simple structure of the 2-OMe
[39]
ꢀ
nation of the BDE
U
4408
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4402 – 4410

Alkoxy-Substituted Phenols
FULL PAPER
Acknowledgements
Experimental Section
Financial support from MIUR (Research projects “Radicals and Radical
Ions: Basic Aspects and Role in Chemistry, Biology, and Material and
Environmental Sciences”, contract 2006033539 and “Stereoselection in
Organic Synthesis: Methodology and Application”, contract
2007FJC4SF) is gratefully acknowledged. Ente Cassa di Risparmio di
Firenze is acknowledged for a grant to C.V. We also thank Dr. Guerra
for access to computational facilities.
Materials: Compounds 4–9 were purchased and used as received. Com-
pound 10 was prepared as previously reported.[40] Derivative 11 was simi-
larly obtained, purification by flash chromatography using petroleum
ether/ethyl acetate 3:1 allowed the isolation of 11 as a white solid
(0.130 g, 54%). 1H NMR (200 MHz, CDCl3): d=1.27 (t, J=7.1 Hz, 3H),
2.20 (s, 3H), 2.97 (d, J=5.2 Hz, 2H), 4.21 (q, J=7.1 Hz, 2H), 5.92 (brs,
OH), 6.25 (brs, 1H), 6.30 (brs, 1H), 6.43 ppm (t, J=5.2 Hz, 1H);
13C NMR (50 MHz, CDCl3): d=14.1 (q, 1C), 21.3 (q, 1C), 40.1 (t, 1C),
61.3 (t, 1C), 120.5 (d, 1C), 108.0 (d, 1C), 111.2 (d, 1C), 131.3 (s, C), 132.2
(s, 1C), 138.7 (s, 1C), 147.8 (s, 1C), 168.5 ppm (s, 1C); elemental analysis
calcd (%) for C12H14O5: C 60.50, H 5.92; found: C 60.31, H 6.00.
[1] a) R. K. Maheshwari, A. K. Singh, J. Gaddipati, R. C. Srimal, Life
Sci. 2006, 78, 2081–2087; b) J. J. Johnson, H. Mukhtar, Cancer Lett.
2007, 255, 170–181; c) G. Chadalapaka, I. Jutooru, S. Chintharlapal-
li, S. Papineni, R. Smith, X. Li, S. Safe, Cancer Res. 2008, 68, 5345–
5354.
All other compounds used in the present investigation were commercial-
ly available.
EPR and thermochemical measurements: Deoxygenated benzene solu-
tions containing compounds 4–11 (0.05m) and di-tert-butyl peroxide
(10% v/v) were sealed under nitrogen in a suprasil quartz EPR tube. The
sample was inserted in the cavity of an EPR spectrometer and photo-
lysed with the unfiltered light from a 500-W high-pressure mercury lamp
at RT. The EPR spectra were recorded by using a spectrometer equipped
with a microwave frequency counter for determination of the g-factors,
which were corrected with respect to that of the perylene radical cation
in concentrated H2SO4 (g=2.00258). The BDE values were determined
by photolysing concentrated solutions of BHT or 2,6-di-tert-butyl-4-cya-
nophenol and 4–11 in the presence of di-tert-butyl peroxide (10% v/v).
The molar ratio of the two equilibrating radicals, obtained from the EPR
spectra, was used to calculate the equilibrium constant, K1. At least three
different concentration ratios of starting phenols were used to check if
the equilibrium was reached. Spectra were recorded a few seconds after
starting to irradiate in order to avoid significant consumption of the phe-
nols during the course of the experiment. Relative radical concentrations
were determined by comparison of the digitised experimental spectra
with computer simulated data as previously described.[15]
[2] a) S. V. Jovanovic, C. W. Boone, S. Steenken, M. Trinoga, R. B.
Kaskey, J. Am. Chem. Soc. 2001, 123, 3064–3068; b) W. Bors, S. P.
Kazazic, C. Michel, V. D. Kortenska, K. Stettmaier, L. Klansinc, Int.
J. Quantum Chem. 2002, 90, 969–979.
[3] G. Litwinienko, K. U. Ingold, J. Org. Chem. 2004, 69, 5888–5896.
[4] L. B. Davin, N. G. Lewis, Curr. Opin. Biotechnol. 2005, 16, 398–406.
[5] a) Y.-W. Chin, H.-B. Chai, W. J. Keller, A. D. Kinghorn, J. Agric.
Food Chem. 2008, 56, 7759–7764; b) C. Hua, Y. V. Yuanb, D. D.
Kittsa, Food Chem. Toxicol. 2007, 45, 2219–2227.
[6] C. Canela, R. M. Moraesb, F. E. Dayana, D. Ferriera, Phytochemis-
try 2000, 54, 115–120.
[7] J. Stubbe, W. A. Donk, Chem. Rev. 1998, 98, 705–762.
[8] a) B. E. Schultz, S. I. Chan, Annu. Rev. Biophys. Biomol. Struct.
2001, 30, 23–65; b) R. Amorati, P. Franchi, G. F. Pedulli, Angew.
Chem. 2007, 119, 6452–6454; Angew. Chem. Int. Ed. 2007, 46, 6336–
6338.
[9] H. Dau, M. Haumann, Biochim. Biophys. Acta, J. Bioenerg. 2007,
1767, 472–483.
[10] L. Valgimigli, J. T. Banks, K. U. Ingold, J. Lusztyk, J. Am. Chem.
Soc. 1995, 117, 9966–9971.
[11] P. Franchi, M. Lucarini, G. F. Pedulli, L. Valgimigli, B. Lunelli, J.
Am. Chem. Soc. 1999, 121, 507–514.
[12] M. I. de Heer, P. Mulder, H.-G. Korth, K. U. Ingold, J. Lusztyk, J.
Am. Chem. Soc. 2000, 122, 2355–2360.
[13] M. I. de Heer, H. G. Korth, P. Mulder, J. Org. Chem. 1999, 64, 6969–
6975.
[14] H.-G. Korth, M. I. de Heer, P. Mulder, J. Phys. Chem. A 2002, 106,
8779–8789.
[15] M. Lucarini, P. Pedrielli, G. F. Pedulli, S. Cabiddu, C. Fattuoni, J.
Org. Chem. 1996, 61, 9259–9263.
[16] a) M. C. Foti, C. Daquino, Chem. Commun. 2006, 3252–3254;
b) M. C. Foti, C. Daquino, I. D. Mackie, G. A. DiLabio, K. U.
Ingold, J. Org. Chem. 2008, 73, 9270–9282.
[17] The exact structure of the away conformer 4a deserves some com-
ment. Various theoretical studies have found that in the gas phase a
“perpendicular conformation” of the OCH3 group, allowing some
EPR spectral parameters of phenoxyl radicals: Phenoxyl from 4: a
5.23 G, aoA(OMe) 1.58 G, amA(1H) 1.96 G, amA(1H) 0.94 G, apA(1H) 8.73 G,
g=2.0051; 5: aoA(1H) 5.19 G, aoA(OMe) 1.43 G, amA(1H) 1.91 G, amA(1H)
0.64 G, ap(Me) 10.03 G, g=2.0050; 6: aoA(1H) 5.11 G, aoA(OMe) 1.19 G, am-
(1H) 1.47 G, mA(1H) 0.09 G, pA(OMe) 1.40 G, g=2.0050; 9: oA(1H)
4.05 G, amA(1H) 0.37 G, amA(2H) 2.22 G, apA(1H) 7.66 G, g=2.0052; 10: ao-
(1H) 3.71 G, aoA(1H) 2.54 G; amA(1H) 0.56 G, apA(1H) 7.35 G, g=2.0052;
11: aoA(1H) 2.86 G, aoA(1H) 2.86 G, am(Me) 0.28 G, apA(1H) 7.35 G, g=
2.0051.
oACHTUNGERTN(NUNG 1H)
C
N
G
CHTUNGTRENNUNG
C
N
G
CHTUNGTRENNUNG
C
CHTUNGTRENNUNG
A
a
E
a
N
a CHTUNGTRENNUNG
C
E
CHTUNGTRENNUNG
A
E
N
CHTUNGTRENNUNG
C
E
CHTUNGTRENNUNG
Autoxidation experiments: The rate constants for the reaction of the title
compounds, 2,6-di-tert-butyl-4-cyanophenol (kinh =9.9ꢃ103 mꢀ1 sꢀ1), phenol
(kinh =6.0ꢃ103 mꢀ1 sꢀ1) and 2,6-dimethylphenol (kinh =1.5ꢃ104 mꢀ1 sꢀ1) with
peroxyl radicals were measured by monitoring the autoxidation of
cumene (7.1m) in chlorobenzene at 308C using AIBN (0.05m) as initia-
tor. The kinh value for 4-methoxyphenol (2.7ꢃ105 mꢀ1 sꢀ1) was measured
by studying the inhibited autoxidation of styrene (4.3m) initiated by
AIBN (0.05m) at 308C.[28,31] The reaction was performed in an oxygen-
uptake apparatus built in our laboratory using a procedure previously de-
scribed.[31b]
ꢀ
C H···O interaction to take place, has an energy very similar to the
planar one (refs. [13,18]). On the other hand, ortho-dimethoxyben-
zene, which can be considered a methylated analogue of 4, was
found to be planar and (reasonably) in the away conformation by
NMR experimental investigations (ref. [19]). The effective structure
of the non-HB isomer 4a, however, is not a critical point in the pres-
ent work, as the DHintra-HB is measured on phenols whose ortho-OR
group is blocked in a planar conformation.
IR measurements: FTIR spectra of compounds 4–11, phenol and 4-me-
thoxyphenol were measured in diluted tetrachloromethane solutions
(0.003–0.01m) in a sealed KBr cell with 0.5-mm optical path, taking care
ꢀ
to avoid phenol auto-association. The O H stretching signals of 9–11
were analysed by numerical simulation of the experimental peaks, consid-
ering them as the sum of Gaussian–Lorentzian functions.
[18] S. Tsuzuki, H. Houjou, Y. Nagawa, K. Hiratani, J. Chem. Soc.
Perkin Trans. 2 2002, 1271–1273.
Computational details: The isotropic hfsc for ortho-methoxyphenoxyl
radical in the two planar conformations[13] were computed by performing
DFT calculations with the B3LYP functional, using the Gaussian03
system of programs.[41] Geometries were optimised at the UB3LYP/6-
[19] A. Makriyannis, S. Fesik, J. Am. Chem. Soc. 1982, 104, 6462–6463.
[20] A. T. Lithoxoidou, E. G. Bakalbassis, J. Phys. Chem. A 2005, 109,
366–377.
[21] a) G. L. Carlson, W. G. Fateley, J. Phys. Chem. 1973, 77, 1157–1163;
b) J. N. Spencer, K. S. Robertson, E. E. Quick, J. Phys. Chem. 1974,
78, 2236–2240; c) I. M. Skvortsov, O. E. Makarov, E. E. Fedorov, J.
Chromatogr. A 2006, 1132, 248–255.
31G
lations. Single-point energies and hfsc were computed at the UB3LYP/6-
311++G(d,p) level.
ACHTUNGTRENNUNG(d,p) level and stationary points were confirmed by frequency calcu-
ACHTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 4402 – 4410
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4409

R. Amorati, G. F. Pedulli et al.
[22] T. Schaefer, D. M. McKinnon, R. Sebastian, J. Peeling, G. H.
Penner, R. P. Veregin, Can. J. Chem. 1987, 65, 908–913.
[23] a) K. U. Ingold, Can. J. Chem. 1962, 40, 111–121; b) K. U. Ingold,
D. R. Taylor, Can. J. Chem. 1961, 39, 471–480.
[24] H. H. Jaffꢄ, J. Am. Chem. Soc. 1957, 79, 2373–2375.
[25] M. Lucarini, G. F. Pedulli, M. Cipolline, J. Org. Chem. 1994, 59,
5063–5070.
[35] J. Lalevꢅe, X. Allonas, J.-P. Fouassier, K. U. Ingold, J. Org. Chem.
2008, 73, 6489–6496.
[36] a) M. Wahl, M. Sundaralingam, Trends Biochem. Sci. 1997, 22, 97–
102; b) A. C. Pierce, K. L. Sandretto, G. W. Bemis, Proteins 2002, 49,
567–576; c) S. Sarkhel, G. R. Desiraju, Proteins 2004, 54, 247–259;
d) G. R. Desiraju, Chem. Commun. 2005, 2995–3001.
[37] R. Amorati, F. Catarzi, S. Menichetti, G. F. Pedulli, C. Viglianisi, J.
Am. Chem. Soc. 2008, 130, 237–244.
[26] G. Brigati, M. Lucarini, V. Mugnaini, G. F. Pedulli, J. Org. Chem.
2002, 67, 4828–4832.
[38] F. G. Bordwell, G. J. McCollum, J. Org. Chem. 1976, 41, 2391–2395.
[39] Research performed by using Google Scholar (http://scholar.google.-
com) of the papers containing the terms “DFT polyphenols” or
“DFT phenol BDE” showed that in the last seven years the number
of publications containing these words increased by ten and five
times, respectively (see Supporting Information for complete re-
sults).
[40] R. Amorati, A. Cavalli, M. G. Fumo, M. Masetti, S. Menichetti, C.
Pagliuca, G. F. Pedulli, C. Viglianisi, Chem. Eur. J. 2007, 13, 8223–
8230.
[41] Gaussian 03, Revision B.05, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, 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, V. Bakken, 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. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, 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, Inc., Wall-
ingford, CT, 2004.
[27] P. Mulder, H.-G. Korth, D. A. Pratt, G. A. DiLabio, L. Valgimigli,
G. F. Pedulli, K. U. Ingold, J. Phys. Chem. A 2005, 109, 2647–2655.
[28] The effect of the ortho-OR substituent in 9 can be compared to that
of a para-OCH3, because in both cases the preferred geometry of
the oxygenated group is planar. For stereoelectronic effects on the
ꢀ
phenolic BDEACHTUNGTRENNUNG(O H), see: G. W. Burton, T. Doba, E. J. Gabe, L.
Hughes, F. L. Lee, L. Prasad, K. U. Ingold, J. Am. Chem. Soc. 1985,
107, 7053–7065.
[29] Also geometric straining caused by the fused five-membered ring
could give some extra BDE-lowering effect as suggested in: D.
Shanks, H. Frisell, H. Ottosson, L. Engman, Org. Biomol. Chem.
2006, 4, 846–852.
[30] M. C. Foti, C. Daquino, G. A. DiLabio, K. U. Ingold, J. Org. Chem.
2008, 73, 2408–2411; M. C. Foti, C. Daquino, G. A. DiLabio, K. U.
Ingold, J. Org. Chem. 2008, 73, 7440.
[31] This autoxidation scheme is applicable only at low conversions for
styrene and cumene. See: a) G. W. Burton, K. U. Ingold, J. Am.
Chem. Soc. 1981, 103, 6472–6477; b) R. Amorati, G. F. Pedulli, L.
Valgimigli, O. A. Attanasi, P. Filippone, C. Fiorucci, R. Saladino, J.
Chem. Soc. Perkin Trans. 2 2001, 2142–2146.
[32] R. Amorati, M. G. Fumo, S. Menichetti, V. Mugnaini, G. F. Pedulli,
J. Org. Chem. 2006, 71, 6325–6332.
[33] L. R. C. Barclay, M. R. Vinqvist, K. Mukai, H. Goto, Y. Hashimoto,
A. Tokunaga, H. Uno, Org. Lett. 2000, 2, 2841–2843.
[34] We also performed a styrene autoxidation inhibited by 5 following
the same conditions reported in ref. [33]. Analysis of the oxygen-
consumption curve by using Equation (11) gave a kinh value of 2.3ꢃ
105 mꢀ1 sꢀ1, whereas by using the kinetic treatment described in the
Supporting Information and in ref. [31c], the 10-fold smaller kinh
value of 3.0ꢃ104 mꢀ1 sꢀ1 was obtained.
Received: November 24, 2008
Published online: March 13, 2009
4410
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4402 – 4410