Influence of Remote intramolecular hydrogen bonds on the stabilities of phenoxyl radicals and benzyl cations

Article abstract of DOI:10.1021/jo100491a

(Figure presented) Remote intramolecular hydrogen bonds (HBs) in phenols and benzylammonium cations influence the dissociation enthalpies of their O-H and C-N bonds, respectively. The direction of these intramolecular HBs, para → meta or meta → para, dete

Full text of DOI:10.1021/jo100491a

pubs.acs.org/joc
Influence of “Remote” Intramolecular Hydrogen Bonds on the Stabilities
of Phenoxyl Radicals and Benzyl Cations
Mario C. Foti,*,† Riccardo Amorati,*,‡ Gian Franco Pedulli,‡ Carmelo Daquino,†
Derek A. Pratt,§ and K. U. Ingold^
†Istituto di Chimica Biomolecolare del CNR, Via Paolo Gaifami 18, I-95126 Catania, Italy,
‡Dipartimento di Chimica Organica “A. Mangini”, Via San Giacomo 11, Universitaꢀ di Bologna,
Bologna, Italy, §Department of Chemistry, Queen’s University, 90 Bader Lane, Kingston, ON, Canada K7L
3N6, and ^National Research Council of Canada, 100 Sussex Drive, Ottawa, ON, Canada K1A 0R6
mario.foti@icb.cnr.it; riccardo.amorati@unibo.it
Received March 16, 2010
Remote intramolecular hydrogen bonds (HBs) in phenols and benzylammonium cations influence
the dissociation enthalpies of their O-H and C-N bonds, respectively. The direction of these
intramolecular HBs, para f meta or meta f para, determines the sign of the variation with respect
to molecules lacking remote intramolecular HBs. For example, the O-H bond dissociation
enthalpy of 3-methoxy-4-hydroxyphenol, 4, is about 2.5 kcal/mol lower than that of its isomer
3-hydroxy-4-methoxyphenol, 5, although group additivity rules would predict nearly identical
values. In the case of 3-methoxy-4-hydroxybenzylammonium and 3-hydroxy-4-methoxybenzyl-
ammonium ions, the CBS-QB3 level calculated C-N eterolytic dissociation enthalpy is about
3.7 kcal/mol lower in the former ion. These effects are caused by the strong electron-withdrawing
character of the -O• and -CH2þ groups in the phenoxyl radical and benzyl cation, respectively,
þ
which modulates the strength of the HB. An O-H group in the para position of ArO• or ArCH2
becomes more acidic than in the parent molecules and hence forms stronger HBs with hydro-
gen bond acceptors (HBAs) in the meta position. Conversely, HBAs, such as OCH3, in the
para position become weaker HBAs in phenoxyl radicals and benzyl cations than in the
parent molecules. These product thermochemistries are reflected in the transition states for,
and hence in the kinetics of, hydrogen atom abstraction from phenols by free radicals (dpph•
and ROO•). For example, the 298 K rate constant for the 4 þ dpph• reaction is 22 times greater
than that for the 5 þ dpph• reaction. Fragmentation of ring-substituted benzylammonium ions,
generated by ESI-MS, to form the benzyl cations reflects similar remote intramolecular HB
effects.
4434 J. Org. Chem. 2010, 75, 4434–4440
Published on Web 06/08/2010
DOI: 10.1021/jo100491a
2010 American Chemical Society
r

Foti et al.
JOCArticle
Introduction
SCHEME 1. Rate Constants for Reactions of Phenols 1-3 with
dpph• in Alkane Solvents at 298 K from Reference 13
The effects of ring substituents, Y, on YC6H4O-H bond
dissociation enthalpies (BDEs) are rather well-established.1
Electron-withdrawing (EW) Y increases and electron-donating
(ED) Y decreases YC6H4O-H BDEs.2 These BDEs are
very well-correlated by Brown’s3 electrophilic substituent
constants, σþ(Y).1a-c,e The thermodynamics of YC6H4O-H
BDEs have kinetic consequences for hydrogen atom abstrac-
tions from phenols (ArOH):
YC6H4O-H þ X• f YC6H4O• þ X-H
ð1Þ
Two of these consequences are that activation energies for
reaction 1 decrease and rate constants increase for all X•
radicals as Y becomes a stronger ED group. Further conse-
quences are that Ea,1 and log(k1) correlate with σþ(Y), as has
been shown for peroxyl radicals, X• = ROO•,4 X• = 2,2-
diphenyl-1-picrylhydrazyl (dpph•),5 and tert-butoxyl radicals.6
These facts become particularly important in situations where
phenols are employed (on purpose or by chance) as anti-
oxidants, whether in vitro or in vivo, because the faster reaction
1 (X• = ROO•) becomes, the more effective (generally) the
phenol is as an antioxidant.7
ubiquinol-0 (1), was more than 10-fold as reactive toward
dpph• in alkane solvents14 as its two monomethyl ethers, 2
and 3 (see Scheme 1). It was later discovered16 that 2-methoxy-
hydroquinone, 4, was much more reactive toward dpph• than
4-methoxyresorcinol, 5,17 a result which implies that the “free”
O-H BDE in 4 is smaller than in 5, an implication that
contrasts with the prediction that would be made using group
additivity rules10 that these two compounds would have the
same free O-H BDEs. The similarly disubstituted phenols, 3,4-
dimethoxyphenol, 6, and sesamol, 7, both of which lack an
intramolecular HB, were of intermediate reactivity (see Scheme
2 and Table 1).
The differences in the reactivities of phenols 1-3 and 4-7
were recognized to arise from the very strong EW character
of the incipient O• atom in the transition state during
phenoxyl radical formation (vide infra)18 and to depend
not only on the presence or absence of a “remote” intramo-
lecular hydrogen bond (HB) but also on the direction of any
such HB. However, this was not specifically commented
upon in either report13,16 because the mechanism of reaction
of dpph• with phenols is not straightforward. Computations
indicate that the transition state (TS) cannot be described as
a “clean” hydrogen atom transfer (HAT, involving primarily
3 electrons and the proton) nor as a “clean” proton-coupled
electron transfer (PCET, involving primarily 5 electrons and
the proton) but rather as some mixture of these two
mechanisms.16,19 Furthermore, the TSs for these reactions
are extremely congested.16 The possibility that the measured
rates for these highly substituted phenols were confounded
by interactions of the dpph• with the phenol’s substituents
(e.g., HB formation, dipole-dipole interactions, etc.) could not
be ignored. We therefore withheld comment on the kinetic (and
thermodynamic) effects of remote intramolecular HBs until
additional kinetic measurements on H-atom abstractions from
these phenols could be made using a radical for which the
potentially confounding problems in the dpph• reactions would
Both experiment8 and theory9,10 conclude that the electro-
nic effects of para-methoxy and para-hydroxy11 on phenolic
O-H BDEs are nearly identical.12 It therefore came as a
surprise when two of us13 discovered that the hydroquinone,
(1) (a) Mulder, P.; Saastad, O. W.; Griller, D. J. Am. Chem. Soc. 1988,
110, 4090–4092. (b) Jonsson, M.; Lind, J.; Eriksen, T. E.; Merenyi, G. J.
Chem. Soc., Perkin Trans. 2 1993, 1567–1568. (c) Wayner, D. D. M.; Lusztyk,
E.; Ingold, K. U.; Mulder, P. J. Org. Chem. 1996, 61, 6430–6433. (d)
Lucarini, M.; Pedrielli, P.; Pedulli, G. F.; Cabiddu, S.; Fattuoni, C. J. Org.
Chem. 1996, 61, 9259–9263. (e) Pratt, D. A.; DiLabio, G. A.; Mulder, P.;
Ingold, K. U. Acc. Chem. Res. 2004, 37, 334–340.
(2) For example,1b,c,e ΔBDE{(4-NO2C6H4O-H) - (4-CH3OC6H4O-H)} =
∼10 kcal/mol.
(3) Brown, H. C.; Okamoto, Y. J. Am. Chem. Soc. 1958, 80, 4979–4987.
(4) (a) Howard, J. A.; Ingold, K. U. Can. J. Chem. 1963, 41, 1744–1751.
(b) Howard, J. A.; Ingold, K. U. Can. J. Chem. 1963, 41, 2800–2806.
(5) Snelgrove, D. W.; Lusztyk, J.; Banks, J. T.; Mulder, P.; Ingold, K. U.
J. Am. Chem. Soc. 2001, 123, 460–477.
(6) Ingold, K. U. Can. J. Chem. 1963, 41, 2816–2825.
(7) See for example: (a) Burton, G. W.; Ingold, K. U. J. Am. Chem. Soc.
1981, 103, 6475–6477. (b) Burton, G. W.; Ingold, K. U. Acc. Chem. Res. 1986,
19, 194–201. (c) Foti, M. C. J. Pharm. Pharmacol. 2007, 59, 1673–1685. (d)
Foti, M. C.; Amorati, R. J. Pharm. Pharmacol. 2009, 61, 1435–1448.
(8) Lucarini, M.; Mugnaini, V.; Pedulli, G. F. J. Org. Chem. 2002, 67,
928–931.
(9) Pratt, D. A.; de Heer, M. I.; Mulder, P.; Ingold, K. U. J. Am. Chem.
Soc. 2001, 123, 5518–5526.
(10) Wright, J. S.; Johnson, E. R.; DiLabio, G. A. J. Am. Chem. Soc.
2001, 123, 1173–1183.
(11) These are two of the strongest ED groups. Although dialkylamino
groups are even stronger EDs, they lower the ionization potential of
4-R2NC6H4OH to such an extent that these aminophenols react directly
with O2 and cannot be used as antioxidants. See: (a) Burton, G. W.; Doba,
T.; Gabe, E. J.; Hughes, L.; Lee, F. L.; Prasad, L.; Ingold, K. U. J. Am. Chem.
Soc. 1985, 107, 7053–7065. (b) Wright, J. S.; Pratt, D. A.; DiLabio, G. A.;
Bender, T. P.; Ingold, K. U. Cancer Detþect. Prev. 1998, 22, 204.
(12) It has been concluded that σp (HO) is not -0.92, as originally
proposed,3 but rather that σpþ(HO) ≈ σpþ(CH3O) = -0.78. (a) See footnote
25 in ref 9. (b) See footnote g to Table 1 and footnote 42 in: Pratt, D. A.; DiLabio,
G. A.; Valgimigli, L.; Pedulli, G. F.; Ingold, K. U. J. Am. Chem. Soc. 2002, 124,
11085–11092.
(13) Foti, M. C.; Daquino, C. Chem. Commun. 2006, 3252–3254.
(14) Alkanes are neither HB acceptors nor HB donors, and therefore,
kinetic solvent effects due to HB formation between the phenol and solvent
do not occur.5 Furthermore, phenols do not ionize in alkanes, and therefore,
the sequential proton-loss, electron-transfer (SPLET) mechanism cannot
occur.15
(16) Foti, M. C.; Daquino, C.; Mackie, I. D.; DiLabio, G. A.; Ingold,
K. U. J. Org. Chem. 2008, 73, 9270–9282.
(17) Hydrogen-atom abstractions from 4 and 5 primarily involve the
hydroxyl group that is not involved in an intramolecular HB. This is indi-
cated by the kdpph values for phenol, 2-methoxy, 3-methoxy, and 4-methoxy-
phenol, which are, respectively, 0.1, 0.9, 1.4, and 238 M-1 s-1 in alkane
solvents at 25 ꢀC, and by the A faþctors for these reactions;16 see also below.
(18) The value of σp(O•) = σp (O•) has been estimated to be as large as
2.0.12b This vaþlue implies that O• is a more powerful EW moiety than nitro,
σp(NO2) = σp (NO2) = þ0.78. We are also aware that the possible out-of-
plane rotation of the para-OMe group in 3 (see Scheme 1) may contribute to
making it less reactive than 2.
(19) For much more detailed descriptions of the HAT and PCET reac-
tions mechanisms see: (a) Mayer, J. M.; Hrovat, D. A.; Thomas, J. L.;
Borden, W. T. J. Am. Chem. Soc. 2002, 124, 11142–11147. (b) Mayer, J. M.
Annu. Rev. Phys. Chem. 2004, 55, 363–390. (c) DiLabio, G. A.; Johnson,
E. R. J. Am. Chem. Soc. 2007, 129, 6190–6203.
(15) (a) Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2003, 68, 3433–
3438. (b) Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69, 5888–5896.
(c) Foti, M. C.; Daquino, C.; Geraci, C. J. Org. Chem. 2004, 69, 2309–2314.
J. Org. Chem. Vol. 75, No. 13, 2010 4435

Foti et al.
JOCArticle
SCHEME 2. Structures of Phenols 4-15 and Benzylamines 16-28 Used in the Present Study
TABLE 1. Arrhenius Parameters (A, M-1 s-1, and Ea, kcal/mol) and Rate
kinetics of its reactions with these same phenols, 4-7, are
reported herein. The peroxyl radical was also favored because
computations have shown that peroxyl radical þ phenol
reactions proceed by clean PCET mechanisms, albeit with a
cisoid TS being favored over the less congested transoid
TS;19c,20 see also below. In fact, cisoid TSs are surprisingly
common in PCET processes.19c,21
Because the effects of remote intramolecular HBs on
phenoxyl radical stabilities (as reflected by phenol reactiv-
ities toward attacking radicals and O-H BDEs) are a
consequence of the strong EW character of O• in YC6H4O•,
it was predicted that similar, or even greater, effects
should be observed with benzyl cations, YC6H4CH2þ. This
prediction has been confirmed in the present work both by
mass spectrometric measurements22 and by theoretical
calculations.
Constants kdpph (M-1 s-1) for the Reaction of Phenols 4-15 with dpph• at
298 K and Rate Constants kROO (M-1 s-1) for the Reaction with ROO•
Radicals at 303 K. Experimental (dpph• and EPR Methods, 1 M Standard
State) and Calculated (1 atm Standard State) O-H Bond Dissociation
Enthalpies (kcal/mol)a
O-H BDE (kcal/mol)
no.
ArOH A/105 b Ea kdpph
kROO
/
105 d dpphe EPRf calcdg additivityh
b
c
4
5
11
12
3.25 4400
5.1 200
3.8 1800
4.2 935
2.0 9400i
7.2
2.5
4.7
3.4
80.0
82.1
80.4
80.6
-
-
-
80.8
80.1
82.6
80.3
80.8
79.7
80.9
-
82.6
82.5
90.4
85.2
88.4
80.9
81.5
80.6
-
80.7
81.3
80.4
82.3
84.1
86.7
-
6
12
12
7
8
2.3
2.3
2.3
4.1
12
4.4i e78.7 78.9
9
3.2
-
4.7
5.6
999i
1700
155
97
1.7i
3.5
0.48
1.0
-
80.0
79.3
81.5
82.6
88.3
84.1
87.1
79.8
79.1
81.1
-
-
-
10
11
12
13
14
15
12
12
10.8 0.015
Results
6.9
9.7
11
0.09
0.32
12
-
-
-
Scheme 2 contains the structures of phenols 4-7 (with
which we are most concerned) and 8-15, and benzylamines
16-28 used in the current work. The rate constants kdpph
measured in the present work together with some previously
reported values16 and the rate constants kROO are gathered in
Table 1. Also included are values of the free O-H BDEs.
These were estimated, as described previously,16 from the
activation energies of the dpph• reactions. There is obviously
some ambiguity when the molecule in question contains two
aSome data are from ref 16. To convert from one standard state to the
other: O-H BDE (1 atm) = O-H BDE (1 M) þ 0.4 kcal/mol; see ref 16.
bThe A factors (in italics) for phenols 6, 7, 10, and 12-15 were assumed
to be equal to the expected value for free phenolic O-H groups (see ref
16), and the corresponding Ea values (in italics) were calculated from the
experimental rate constants. cRate constants determined in cyclohexane
and (in italics) rate constants calculated for cyclohexane solvent from
measurements made in CH2Cl2 (see ref 16); experimental error (10%.
dRate constants determined in chlorobenzene/styrene or cumene; error
within (15%. eO-H BDEs obtained from the activation energies of the
reaction with dpph• (error < ( 1 kcal/mol, ref 16); the O-H BDE of
phenol was found to be 86.3 kcal/mol. fO-H BDEs obtained by the EPR
equilibration technique (error (0.2 kcal/mol). gO-H BDE calculated
using the CBS-QB3 approach; the O-H BDE of phenol was calculated to
be 87.1 kcal/mol. hCalculated using the additive contributions of the
individual substituents on the O-H BDE; see the Supporting Information.
iNot statistically corrected.
(20) Singh, N.; O’Malley, P. J.; Popelier, P. L. A. Phys. Chem. Chem.
Phys. 2005, 7, 614–619.
(21) See for example: (a) DiLabio, G. A.; Ingold, K. U. J. Am. Chem. Soc.
2005, 127, 6693–6699. (b) Vaidya, V.; Ingold, K. U.; Pratt, D. A. Angew.
Chem., Int. Ed. 2009, 48, 157–160.
(22) For a recent study on carbocation stabilities by mass spectrometry,
see: Ustinov, A. V.; Shmanai, V. V.; Patel, K.; Stepanova, I. A.; Prokhorenko,
I. A.; Astakhova, I. V.; Malakhov, A. D.; Skorobogatyi, M. V.; Bernad, P. L.,
Jr.; Khan, S.; Shahgholi, M.; Southern, E. M.; Korshun, V. A.; Shchepinov,
M. S. Org. Biomol. Chem. 2008, 6, 4593–4608.
be expected to be of much less importance. The peroxyl radical
ROO• was chosen for these additional experiments, and the
4436 J. Org. Chem. Vol. 75, No. 13, 2010

Foti et al.
JOCArticle
(or more) phenolic OH groups in dissimilar environments.
When the molecule contains both a free OH group and an
OH group involved in an intramolecular HB, it was con-
cluded that the dpph• reacted primarily with the free OH.16,17
It is probably worthwhile exploring this matter again for the
phenol structural types relevant to the present paper by
comparing the disubstituted phenols, 6, 7, and 11. Com-
pounds 6 and 7 have only a free OH group with alkoxyl
groups at the 3 and 4 positions. Compound 11 has only an
OH group involved in an intramolecular HB with alkoxyl
substitution at the 2 and 4 positions. The ratios of rate con-
stants, 100 ꢀ k11/k6, are 8.6 (dpph•) and 10.2 (ROO•), while
the 100 ꢀ k11/k7 ratios are 16.6 (dpph•) and 14.1 (ROO•). This
indicates that, provided electron donation to the aromatic
ring by the substituents is approximately equal, an OH group
involved in an intramolecular HB with a methoxyl is only ca.
12% as reactive as a free OH. In phenol 4 that contains both
types of OH groups, it certainly would be possible to parse
the measured rate constants (and dpph• derived O-H BDEs)
using this 12% figure. However, such corrections were
deemed unnecessary because the changes would probably
be less than errors from other sources.
The EPR equilibrium method for measuring O-H
BDEs23 was also employed with phenols 7-11, including
the pair of isomeric phenols 8 and 9, which have the same
“remote” substituents as 4 and 5. Unfortunately, the other
phenols yielded phenoxyls that were insufficiently persistent
for application of this technique. The EPR-derived O-H
BDEs are essentially identical to the values obtained by the
dpph• method (see Table 1). The O-H BDEs were also
calculated for phenols 4-9 and 11-15 for comparison with
the experimental data. The CBS-QB3 methodology, which
we have previously shown to accurately predict O-H BDEs,
was used for these calculations. The agreement, where
comparison is possible, is quite good. The calculated BDEs
could be compared directly with those obtained for the
additive contributions of each individual substituent, which
were derived from separate calculations on the monosubsti-
tuted phenols at the same level of theory (see Supporting
Information).
SCHEME 3. Fragmentation of Substituted Benzylamines in the
Electrospray Ion Source
of 16-28 provided a pseudomolecular ion, [M þ 1]þ, and a
[M - 16]þ benzyl ion due to NH3 loss from the [M þ 1]þ ion
(see Scheme 3 and Figure 1). The relative intensity of these two
peaks varies enormously among the differently substituted
benzylamines 16-28 (see Table 2). The fragmentation ratio
(FR) (i.e., the ratio of the peak intensities [M - 16]þ/[M þ 1]þ)
obtained under identical experimental conditions increased
with increasing ED ability of the substituents (see Table 2
and Figure 2).
The dissociation energies of the benzylammonium ions
were also calculated using the same CBS-QB3 approach27
that was used to compute the O-H BDEs in Table 1. These
energies yielded an excellent linear correlation with the log of
the intensity ratio [M - 16]þ/[M þ 1]þ for all benzylamines,
16-28 (see Figure 2). The excellence of this correlation may
be due to the fact that there are no potentially confounding
media effects since the experiments were conducted in the gas
phase. The inset in Figure 2 shows a plot of log(FR) versus
Σσþ of the substituents and yields a reasonable straight
line. A similar plot of ΔDE versus Σσþ gives a line with Fþ =
13.4 kcal/mol (see Supporting Information).
Benzyl cations, appropriately substituted for studying the
effects of remote intramolecular HBs on their stabilities,
were produced in an ESI-MS spectrometer by in-source
fragmentation (“in-source collision-induced dissociation”)24
of benzylamines 16-28 (see Scheme 2). Extensive studies25
on benzylpyridinium cations have shown that ion fragmen-
tation depends on several experimental factors, such as
pressure, acceleration voltage, nature of the solution, and
the gas phase. The initial benzyl cations can isomerize to
tropylium ions25 (see Scheme 3). However, this rearrange-
ment is very unlikely under our conditions because we used
the “soft” ESI technique at a low cone voltage.25,26 In the
positive-ion mode with a low cone voltage, the ionization
Discussion
Intermolecular HB formation between phenols and HB
acceptor (HBA) solvent molecules has long been known to
reduce the ability of the phenol to donate a H atom to an
attacking radical.29 These reductions in the rates of H-atom
abstraction in HBA solvents have been quantified using
Abraham’s parameters for HBA activity of the solvent
and HB donor (HBD) activity of the phenol.5 With few
exceptions (see below), intermolecular HBs between phenol
and solvent reduce the phenol’s reactivity.5 In contrast,
“proximate” intramolecular HBs in aryl-1,2-diols and
naphthalene-1,8-diols cause these diols to be better H-atom
donors than most phenols.30 This is because the O• atom in a
(23) Lucarini, M.; Pedrielli, P.; Pedulli, G. F.; Cabiddu, S.; Fattuoni, C.
J. Org. Chem. 1996, 61, 9259–9263.
(27) Montgomery, J. A., Jr.; Ochterski, J. W.; Petersson, G. A. J. Chem.
Phys. 1994, 101, 5900.
(28) Foti, M. C.; Daquino, C.; DiLabio, G. A.; Ingold, K. U. J. Org.
Chem. 2008, 73, 2408–2411.
(29) Howard, J. A.; Ingold, K. U. Can. J. Chem. 1964, 42, 1044–1056.
(30) (a) Foti, M. C.; Johnson, E. R.; Vinqvist, M. R.; Wright, J. S.;
Barclay, L. R. C.; Ingold, K. U. J. Org. Chem. 2002, 67, 5190–5196. (b) Foti,
M.; Ruberto, G. J. Agric. Food Chem. 2001, 49, 342–348. (c) Lucarini, M.;
Pedulli, G. F.; Guerra, M. Chem.;Eur. J. 2004, 10, 933–939.
(24) Gabelica, V.; De Pauw, E. Mass Spectrom. Rev. 2005, 24, 566–587.
(25) (a) Katritzky, A. R.; Watson, C. H.; Dega-Szafran, Z.; Eyler, J. R.
J. Am. Chem. Soc. 1990, 112, 2471–2478. (b) Collette, C.; De Pauw, E. Rapid
Commun. Mass Spectrom. 1998, 12, 165–170. (c) Gabelica, V.; De Pauw;
Karas, M. Int. J. Mass Spectrom. 2004, 231, 189–195.
(26) Isomerization of the benzyl cation to the tropylium cation has a
computed activation energy of 32.7 kcal/mol. See: Cone, C.; Dewar, M. J. S.;
Landman, D. J. Am. Chem. Soc. 1977, 99, 372–376.
J. Org. Chem. Vol. 75, No. 13, 2010 4437

Foti et al.
JOCArticle
FIGURE 2. Relation between the logarithm of the fragmentation
ratio (FR = [M - 16]þ/[M þ 1]þ) and the calculated dissociation
enthalpies of substituted benzylammonium ions. The equation of
the best-fit line is log(FR) = -0.11 ꢀ DE (kcal/mol) þ 3.87 (R2 =
0.99). Inset: Correlation of log(FR) with the known values of Σσþ:
log(FR) = -1.40Σσþ - 0.45 (R2 = 0.87).
parent diol, and hence, the O-H BDE of the HBA OH
group (i.e., the free OH) is lower than in simple phenols.8,30
Of more relevance in the present context, some of us
reported that the addition of small amounts of CH3CN or
DMSO (two strong HBAs) to a CCl4 solution of 2,5-di-tert-
butylhydroquinone, increased its rate of reaction with dpph•
and ROO• radicals,31 while larger quantities of either HBA
produced the expected rate decrease.31 These results sug-
gested that the O-H BDE in this hydroquinone is weakened,
and its reactivity toward radicals is increased, when the other
(remote) OH group forms an intermolecular HB to an added
HBA molecule. This decrease in O-H BDE can be assigned
to reinforcement, relative to the hydroquinone, of the inter-
molecular HB in the semiquinone radical and, hence, also in
the TS leading to hydroquinone formation. This reinforce-
ment arises because the EW O• atom makes the para-OH
group more acidic than in the hydroquinone and, hence, a
stronger HBD.
FIGURE 1. Spectra obtained by direct injection of a methanolic
solution of benzylamines (2 ꢀ 10-5 M) (from the top) 16, 17, and 25
in an electrospray mass spectrometer recorded at a 15 V cone voltage.
TABLE 2. Fragmentation Ratio, FR = [M - 16]þ/[M þ 1]þ, Brown’s
Electrophilic Substituent Constants, σþ, and CBS-QB3 Calculated Dis-
sociation Enthalpies (DE) of Benzylammonium Ions
DEb
kcal/mol
ΔDEc
kcal/mol
no.
substituent(s)
log FR
Σσþ a
16
17
18
19
20
21
22
23
4-OH, 3-OMe
4-OMe, 3-OH
3,4-di-OMe
3,4-OCH2O-
4-OMe
4-OH
4-Me
3-OMe
1.11
0.68
1.30
0.67
0.93
0.52
-0.18
-0.50
26.6
30.3
24.8 (BA)
28.8
27.7
30.7
37.3
39.5 (tw)
42.0 (aw)
40.7 (BA)
39.3
-16.0
-12.3
-17.8
-13.8
-14.9
-11.9
-5.3
-3.1
-0.6
-1.9
-3.3
0.0
-1.3
0.6
0.3
4.6
6.9
-0.92d
The kinetics for the reactions of dpph• with 1 versus 2 and 3
(Scheme 1), 4 versus 6 and 7, and 8 versus 10 (Scheme 2 and
Table 1) are congruent with the effects of low concentrations
of HBAs on the kinetics of H-atom abstraction from 2,5-di-
tert-butylhydroquinone,31 except that it is now the remote
intramolecular HB in hydroquinones 1, 4, and 8 that is
strengthened in the TSs of the reactions leading to formation of
the semiquinone radicals (see Scheme 4). Conversely, the higher
reactivities of 6 and 7 compared with 5 and of 10 compared with
9 (Table 1) indicate that the remote intramolecular HBs in
resorcinols 5 and 9 are weakened in their phenoxyl radicals.
This is because the electron density on the methoxy group in
4-methoxyresorcinol, for example, is reduced in its phenoxyl
radical, which makes for a weaker intramolecular HBA in
the radical compared with its parent molecule and also in the
TS for formation of the radical. This raises the BDE of the
free OH (Scheme 4). Thus, the direction of a remote intra-
molecular HB can either lower the free O-H BDE (para f
meta intramolecular HB, i.e., hydroquinones with an HBA
-0.78
-0.78e
-0.31
-0.14 f
24
25
26
4-Cl
H
3-OH
-0.51
-0.78
-0.76
þ0.11
0
42.6
-0.14g
41.3 (tw)
43.2 (aw)
42.9 (BA)
47.2
27
28
4-CN
4-NO2
-1.08
-1.32
þ0.66
þ0.79
49.5
aFrom ref 3 unless otherwise noted. bWhere conformations of different
energy were found, they are indicated as follows: tw (toward), i.e., with the
OMe or OH group pointing toward the -CH2þ; aw (away), OMe or OH
group pointing away from the -CH2þ; BA, Boltzmann averaged. cCalcu-
lated with respect to the unsubstituted benzyl cation 25. dBy additivity,
i.e, -0.78 þ (-0.14). eAssumed = σþ (4-OMe), see ref 12. fFrom ref 28.
gAssumed = σþ (3-OMe).
phenoxyl is a much stronger HBA than the OH group that it
replaces in the diol. As a consequence, the intramolecular HB
in the diol’s semiquinone radical is stronger than in the
(31) Amorati, R.; Franchi, P.; Pedulli, G. F. Angew. Chem., Int. Ed. 2007,
46, 6336–6338.
4438 J. Org. Chem. Vol. 75, No. 13, 2010

Foti et al.
JOCArticle
SCHEME 4. Resonance Structures with Charge Separation in
the Phenoxyl Radicals Derived from 4 and 5
group at the 2 position) or raise the free O-H BDE (meta f
para intramolecular HB, i.e., resorcinols with an HBA group
at the 4 position). Both results are due to the strong EW
character of the O• substituent. Similar intramolecular HB
effects explain the decreased reactivity of 14 in comparison to
12 and of 13 in comparison to 15. These compounds can
serve as models for studying intramolecular HB effects in
many natural polyphenols, such as hesperitin and quercetin
(the structures of which are presented in the Supporting
Information).
Phenols are considerably more reactive toward peroxyls
than toward dpph• for both thermodynamic, viz. BDE-
(ROO-H) ≈ 88 and BDE(dpph-H) = 78.9 kcal/mol, and
steric reasons.16 Importantly, the four phenols of primary
concern exhibit the same pattern of reactivities toward both
of these radicals, that is, 4 > 6 > 7 > 5 (see Table 1).
Consistently, phenols 8-10, analogues of 4-6, follow the
same pattern of reactivity toward dpph• and ROO•, that is,
8 > 10 > 9 (see Table 1). The order of reactivity of these
phenols to both dpph• and peroxyl radicals are, therefore,
unlikely to be due to any “special” effects related to either
radical. Instead, they simply reflect the differences in the
ArO-H BDEs (and thus in the Ea of the transition states;16
see Table 1), which we suggest are due to the absence or
presence and the direction of remote intramolecular HBs.32
We can see this directly in the calculated structures and
activation enthalpies for the reactions of phenols 4 and 5 with
methylperoxyl (a model peroxyl), as is shown in Figure 3.
While both 4 and 5 have preferred transition state structures
for the reaction that features the expected (vide supra) cisoid
geometry to facilitate the PCET reaction, the calculated
activation enthalpy is lower by 1.7 kcal/mol for 4 relative to 5.
Thermodynamic effects of remote intramolecular HBs on
the O-H BDEs in phenols underlie the more readily mea-
FIGURE 3. Calculated (UB3LYP/CBSB7) minimum energy tran-
sition state structures for the reactions of phenols 4 and 5 with
methylperoxyl radical. The given calculated enthalpies of activation
were determined at the same level and are Boltzmann-averaged (to
include the contributions of higher energy conformers) relative to
H-bonded pre-reaction complexes.
sured kinetic effects (kdpph and kROO). The O-H BDEs
obtained by EPR agreed with the values obtained by the
dpph• method16 to within (0.4 kcal/mol (see Table 1). A
similar agreement was also found between calculated and
dpph• derived O-H BDEs (see Table 1), confirming the
reliability, that is, the “authenticity”, of O-H BDEs deter-
mined by the dpph• method. Of course, remote intramolecular
HBs produce small, but significant, differences between mea-
sured or calculated O-H BDEs and values estimated using
group additivity principles,10 for example, (BDEcalculated
-
BDEadditivity)/kcal/mol = -0.8 (4), þ1.1 (5), butonly-0.3 for
6 which has no remote intramolecular HB (see Table 1).33
The origin of the above-described effects on the kinetics
and thermodynamics of H-atom abstractions from phenols
lies in the strong EW nature of O• in XC6H4O•. Similar, but
larger, effects were therefore predicted to be present in
benzyl cations, XC6H4CH2þ. These were generated from
benzylamines (see Results and Scheme 3). Formation of the
[M - 16]þ benzyl ion by loss of NH3 from the [M þ 1]þ ion
(see Scheme 3 and Figure 1) was favored by ED substituents
that stabilize the benzyl cation (see inset in Figure 2). The
plot of the logarithm of the intensity ratio [M - 16]þ/[M þ 1]þ
against the calculated dissociation enthalpies (DE) of the
benzylammonium ions gave an excellent linear relation (see
Figure 2). Remote HB effects are quite apparent. For
example, the substituent pattern for the benzylammonium
ion pair, 16 and 17, is the same as the pattern seen for the
phenol pair, 4 and 5. As predicted, the remote HB-induced
differences between the DE values for formation of the two
benzyl cations is slightly greater than the differences in O-H
BDEs between the correspondingly substituted phenols,
viz. ΔDEcalculated/kcal/mol: (17 - 16) = þ3.7, versus ΔOH
BDEcalculated/kcal/mol: (5 - 4) = þ2.5 kcal/mol. Surprisingly,
although the appropriate 3,4-disubstituted phenols that lack
remote intramolecular HBs, such as 6 and 7, exhibit reactivities
toward radicals that are intermediate between those of 4 and 5
(see Table 1), the analogously substituted benzylammonium
ions lacking remote intramolecular HBs, such as 18 and 19, do
not show the same pattern with the 18 ion fragmenting slightly
more readily than the 16 ion and fragmentation of the 19 ion
occurring as readily as for the 17 ion (see Table 2). We prefer
not to speculate on the origin of these differences between the
(32) Though not directly related to the main thrust of the present paper,
there is a sometimes an unrecognized factor that plays an important role in
determining both the absolute magnitudes and the patterns of reactivity of
phenols toward free radicals. Comparison of the disubstituted phenols, 4, 5,
and 6 (all of which have one “free” OH group that is presumed to be the main
site of reaction16) with the “comparable” trisubstituted phenols, viz. 8, 9, and
10, respectively, shows that the trisubstituted phenols are as reactive as their
disubstituted counterparts despite their lack of a free OH group. In more
detail, toward dpph•, the pairwise relative reactivities are 8 > 4; 10 ∼ 6; 9 > 5;
and toward ROO•, they are reversed, viz. 8 < 4; 10 < 6; 9 < 5; these orders
are essentially maintained whether or not a statistical correction is applied to
8 and 9. Despite the absence of a free OH, the trisubstituted phenols tend to
be more active towards dpph• than the disubstituted phenols, but this is not
the case for ROO• radicals. These patterns suggest that polar effects are
important in stabilizing the TSs in these reactions, being more important for
the very electron deficient dpph• than for ROO• and being more important for
the phenols substituted with three electron-donating substituents than for
those with only two such substituents, such as charge-separated contribu-
tions to the TS of the type:
‡
½X•ArOHTX- ArOH• þ ꢁ
are more important when X• = dpph• than when X• = ROO•.
(33) Relevant isodesmic reactions are given in the Supporting Information.
J. Org. Chem. Vol. 75, No. 13, 2010 4439

Foti et al.
JOCArticle
properties of phenols, 4, 5, 6, and 7, and the properties of the
corresponding benzylammonim ions, 16, 17, 18, and 19.33
In summary, remote intramolecular HBs, O-H f OCH3,
have appreciable thermodynamic effects on the O-H BDEs
of phenols and the DE values of with benzylammonium
ions and hence on the kinetics of many of their reactions.
These effects are due to the strong EW ability of O• and
CH2þ groups. Intramolecular para-OH f meta-OCH3 HBs
weaken, while intramolecular meta-OH f para-OCH3 HBs
strengthen both the O-H BDEs and the benzylammonium
ion DEs compared with similarly substituted (e.g., 3,4-
dimethoxy) molecules lacking an intramolecular HB.
argon, followed by the addition of 23.6 mg of 60% pure NaH
(0.59 mmol). This solution was left under stirring until efferves-
cence ceased (ca. 15 min), after which about 6 mmol of CH3I was
added and left to react under argon for about 1 h at room
temperature. After this, the solution was acidified with 2 N HCl
and extracted with ethyl acetate (3 ꢀ 60 mL). The organic phase
was washed in succession with NaHCO3 solution, water, and
brine and then dried over Na2SO4, filtered, and the solvent
removed. The residue was purified by chromatography on silica
gel. The initial eluent was a mixture of hexane, CH2Cl2, and
CH3OH in a volume ratio of 50:50:0.5, respectively. This was
gradually changed to 100% CH2Cl2. The final yield of 10 was
45%: 1H NMR (400.13 MHz in acetone-d6) three resolved
singlets at δ 3.74, 3.75, and 3.79 (3H each, three nonequivalent
OCH3), two singlets at δ 6.56 (1H) and 6.71 (1H), and a singlet at
Experimental Section
δ 7.18 (1H, OH, which disappeared upon addition of D2O); 13
C
General. The procedures (and some kinetic results) for the
dpph• reactions in saturated hydrocarbon solvents (cyclohexane
or n-hexane) at 298 K have been presented previously.16 The rate
constants for the ROO• þ ArOH reactions were determined
from rates of azo-initiated, phenol-inhibited autoxidation of styr-
ene (in chlorobenzene) or of cumene at 30 ꢀC.4a,b,11a,34 The O-H
BDEs of phenols were experimentally determined from the activa-
tion energies16 of ArOH þ dpph• reactions and, in the case of
persistent phenoxyl radicals, by the EPR radical equilibration
technique.23 All solvents and compounds (except for 8,16 9,16
and 10) were purchased at high purity from commercial suppliers.
ESI-MS Measurements. Mass spectra of the benzylamines (2 ꢀ
10-5 M in MeOH) were obtained by direct infusion with a
microsyringe pump (15 μL/min) into a Micromass ZMD ESI-
MS spectrometer using the following instrumental settings:
positive ions; desolvation gas (N2), 250 L/h; cone gas (skimmer),
22 L/h; desolvation temperature, 100 ꢀC; capillary voltage, 3.0 kV;
cone voltage, 10-40 V; hexapole extractor, 3 V; RF lens, 0.3 V.
Theoretical Calculations. All calculations were carried out
using the CBS-QB3 method of Petersson and co-workers27 as it
is implemented in the Gaussian 0335 suite of programs, compiled to
run on Sun Microsystems SunFire 25000 or Enterprise M9000
servers with UltraSPARC-IVþ or Sparc64 VII CPUs, respectively.
Synthesis of 10. Hydroquinone, 816 (100 mg, 0.59 mmol), was
dissolved in 2 mL of DMSO, and the solution was degassed with
NMR (100.62 MHz in acetone-d6) δ 55.9, 56.5, and 56.7 ppm
(OCH3); 101.7 and 102.3 ppm (CH); 140.6, 140.7, 142.3, 144.3
ppm (quaternary Cs); ESI-MS in the negative ion-mode gave a
base peak at m/z 183 [M - H]-.
Acknowledgment. We sincerely thank two anonymous
reviewers for their insightful and helpful comments. D.A.P.
thanks the Natural Sciences and Engineering Research
Council of Canada and the Canada Research Chairs pro-
gram for financial support.
Supporting Information Available: Computational data,
plot of ΔDE vs Σσþ; calculated enthalpies of isodesmic reac-
tions. This material is available free of charge via the Internet at
http://pubs.acs.org.
(35) All calculations were performed using Gaussian 03: Frisch, M. J.;
Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman,
J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.;
Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala,
P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz,
J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu,
G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.;
Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Pittsburgh PA, 2003.
(34) (a) Howard, J. A.; Ingold, K. U. Can. J. Chem. 1962, 40, 1851–1864.
(b) Howard, J. A. In Free Radicals; Kochi, J. K., Ed.; Wiley-Interscience:
New York, 1975; Vol. 2, Chapter 12. The oxidations were followed by monitoring
the oxygen uptake using the automatic recording gas absorption apparatus
previously described. See: (c) Amorati, R.; Pedulli, G. F.; Valgimigli, L.;
Attanasi, O. A.; Filippone, P.; Fiorucci, C.; Saladino, R. J. Chem. Soc., Perkin
Trans. 2 2001, 2142–2146.
4440 J. Org. Chem. Vol. 75, No. 13, 2010