Foti et al.
TABLE 2. Va lu es of B,a B′,a ,b a n d m a Rela tive to Eq 6 Deter m in ed in a Ra n ge c of Con cen tr a tion s (µM) for 1-5 a t 25 °C
methanol
ethanol
B′
b
b
phenol
B
m
B′
c
B
m
c
1
3.475
7.470c
0.022
0.100
1.996
0.903
0.495
0.645c
0.341
0.661
0.710
0.493
2.93
10-1190
18-1370c
500-4900
100-2300
5-1700
3.780
0.588
11.50
20-2000
39.57c
0.005
0.70
2
3
4
5
0.035
0.149
13.03
0.392
0.325
0.687
0.549
0.668
0.005
1.10
26.94
2.94
400-4900
90-2500
5-1650
4-560
18.50
0.85
29-3300
a
b
Error has been estimated to be ca. (20%. In all cases, R2 was g0.95. Values B′ have been obtained by setting m ) 0.5, k1 ) B′[1-
5]-0.5, where k1 is given in units of M-1 s-1 c In CH3OD.
.
may not be representative of the rate of H-atom abstrac-
tion from ArOH by DPPH• (or ROO•24).
reaction with respect to 1-5 is ca. 0.5 since m ≈ 0.5
-d[DPPH]/dt ) nB[1-5]10-m[DPPH]0 ≈
Our current results indicate that in these solvents, the
main mechanism by which reaction 1 occurs does not
involve a HAT mechanism. Generally, reaction 1 with
ArOH ) 1-10 was, in fact, fast both in methanol and in
ethanol (see Table 1), and this was rather surprising
because a reduction of the rate by kinetic solvent effect
(KSE) was expected.7,8 This situation is particularly
evident for catechol 10. The H-atom transfer from 10 to
nB[1-5]00.5[DPPH]0 (7)
and this indicates that the reactive species is not the
neutral form, ArOH, because in this case the reaction
order would be unity with respect to both reactants.
However, an order of reaction equal to 1 is certainly not
sufficient for demonstrating a HAT mechanism. We
noticed, in fact, that phenols 6-10, at low concentration,
had an order of reaction of ca. 1; however, even in this
case, the HAT mechanism seems to be rather unlikely
because the methyl esters were 3-5 times more reactive
than the corresponding free acids (see Table 1), and this
different reactivity is not explainable for a reaction of
H-atom transfer to DPPH•.
DPPH• is, in n-hexane, very fast, k1 ) 2.1 × 104 M-1 s-1 8
.
However, this rate constant decreases as the hydrogen
bond acceptor ability of the solvent (i.e., the âH2 param-
eter)15 increases, e.g., k1 (in M-1 s-1) is8 1400 (1-chlorobu-
tane), 64 (acetonitrile), 52 (ethyl acetate), 70 (n-propanol),
44 (tert-butyl alcohol), and 12 (acetone). Our results show
instead that this rate constant is comparatively large in
ethanol since k1 ) 2100 M-1 s-1 (see Table 1). Thus, the
rate decreases with respect to that in the hydrocarbon
solvent by a factor of 21 000/2100 ) 10 (28 in methanol),
while in the other solvents with comparable âH2 values,
the KSEs are responsible for a ca. 400-fold reduction of
the rate! This very large difference suggests that in
ethanol and methanol the HAT mechanism has been
replaced by a more efficient path that makes reaction 1
occur at a higher rate.
The foregoing kinetic observations imply that the
actual mechanism by which reaction 1 occurs is influ-
enced by the carboxylic group of 1-5, as has also been
confirmed by two more experiments. First, when 30 µM
caffeic acid 1 was allowed to react with DPPH• (90 µM)
in methanol both in the absence and presence of 30 µM
KOH,27 the observed rate constant k1 increased, respec-
tively, from ca. 600 to 3310 M-1 s-1. Second, substituting
the base with 30 µM acetic acid27 caused the rate of
reaction 1 to decrease dramatically, and k1 became ca.
As reported in Table 1, caffeic acid 1, its methyl ester
7, and surprisingly25 sinapic acids 4 and its methyl ester
8 are also very efficient DPPH• scavengers since the
values of k1, at lower concentrations of scavenger, ranged
from 1300 to 20 000 M-1 s-1 (the esters being the most
active compounds). We also observed that the rate
constants k1 for 1-5 decline as expressed by eq 6 and
shown in Figure 1. Equation 6 shows that the order of
32 M-1 s-1
.
The entire picture described above led us to understand
how the multifaceted aspects of reaction 1 are easily
explained by taking into account the dissociation of the
phenolic hydroxyl of 1-10, equilibrium 8, and the
subsequent cascade of reactions 9-11. The direction of
the ET step from ArO- to DPPH•, reaction 9, may be
justified by the favorable pKa ) 8.5 of H-DPPH17 with
respect to that of phenols, pKa ≈ 8.7-11.
(22) In apolar media, it is even possible to calculate the value of
k0
at 30 °C (for ROO• ) polyperoxylstyryl radicals), with good
ArOH/ROO
ArOH a ArO- + H+
KOH
(8)
0
approximation, from the experimental value of k1 at 25 °C. It turns
out, in fact, that for nonhindered phenols, both the log k0
and
ArOH/ROO
0
the log k1 are linearly dependent on the bond dissociation enthalpy
ArO- + DPPH• f ArO• + DPPH-
DPPH- + H+ f H-DPPH
(9)
(10)
(11)
of the phenolic O-H23 involved in the reactions with ROO• and DPPH•,
and this leads to the following empirical equation:23 log k0
)
ArOH/ROO
0
3.61 + 0.66log k1
.
(23) Foti, M. C.; J ohnson, E. R.; Vinqvist, M. R.; Wright, J . S.;
Barclay, L. R. C.; Ingold, K. U. J . Org. Chem. 2002, 67, 5190-5196.
(24) Peroxyl radicals ROO• are also able to abstract an H-atom from
methanol or ethanol.
(25) Our surprise arose from the fact that our measurement of k1-
(8) ) 130 ( 20 M-1 s-1 in n-hexane at 20 °C indicates that the methyl
ester 8 (and the corresponding acid 4) is not an efficient scavenger of
DPPH• by HAT. This is because the hydroxyl group is strongly
hydrogen-bonded to one of the methoxyls in the ortho position. If the
HAT mechanism were prevalent also in methanol or ethanol, this rate
constant would be further decreased by a factor of ca. 6.26 The observed
value of k1(8) in methanol and ethanol is instead 2.0 × 104 M-1 s-1
(Table 1), i.e., ca. 150 times larger than that in the hydrocarbon solvent.
ArO• + DPPH• f products
In the case of 1-5 (which are relatively strong acids),
the acidity of the medium is substantially determined by
the dissociation of the carboxylic group, reaction 12,
(26) de Heer, M. I.; Mulder, P.; Korth, H.-G.; Ingold, K. U., Lusztyk,
J . J . Am. Chem. Soc. 2000, 122, 2355-2360.
(27) Addition of small quantities of KOH or acetic acid did not cause
any appreciable decay of DPPH• within the time of reaction 1.
2312 J . Org. Chem., Vol. 69, No. 7, 2004