Naphthalene diols: a new class of antioxidants intramolecular hydrogen bonding in catechols, naphthalene diols, and their aryloxyl radicals.

Article abstract of DOI:10.1021/jo020184v

1,8-Naphthalenediol, 5, and its 4-methoxy derivative, 6, were found to be potent H-atom transfer (HAT) compounds on the basis of their rate constants for H-atom transfer to the 2,2-di(4-t-octylphenyl)-1-picrylhydrazyl radical (DOPPH*), k(ArOH/DOPPH)*, or as antioxidants during inhibited styrene autoxidation, k(ArOH/ROO)*, initiated with AIBN. The rate constants showed that 5 and 6 are more active HAT compounds than the ortho-diols, catechol, 1, 2,3-naphthalenediol, 2, and 3,5-di-tert-butylcatechol, 3. Compound 6 has almost twice the antioxidant activity, k(ArOH/ROO)* = 6.0 x 10(6) M(-)(1) s(-1), of that of the vitamin E model compound, 2,2,5,7,8-pentamethyl-6-chromanol, 4. Calculations of the O-H bond dissociation enthalpies compared to those of phenols, (deltaBDEs), of 1-6 predict a HAT order of reactivity of 2 < 1 < 3 approximately 4 < 5 < 6 in general agreement with kinetic results. Calculations on the diols show that intramolecular H-bonding stabilizes the radicals formed on H-atom transfer more than it does the parent diols, and this effect contributes to the increased HAT activity of 5 and 6 compared to the activities of the catechols. For example, the increased stabilization due to the intramolecular H-bond of 5 radical over 5 parent of 8.6 kcal/mol was about double that of 2 radical over 2 parent of 4.6 kcal/mol. Linear free energy plots of log k(ArOH/DOPPH)* and log k(ArOH/ROO)* versus deltaBDEs for compounds 1-6 along with available literature values for nonsterically hindered monophenols placed the compounds on common scales. The derived Evans-Polanyi constants from the plots for the two reactions, alpha(DOPPH)* = 0.48 > alpha(ROO)* = 0.32, gave the expected order, since the ROO* reaction is more exothermic than the DOPPH* reaction. Compound 6 is sufficiently reactive to react directly with oxygen, and it lies off the log k(ArOH/ROO)* versus deltaBDE plot.

Full text of DOI:10.1021/jo020184v

Na p h th a len e Diols: A New Cla ss of An tioxid a n ts In tr a m olecu la r
Hyd r ogen Bon d in g in Ca tech ols, Na p h th a len e Diols, a n d Th eir
Ar yloxyl Ra d ica ls
Mario C. Foti,† Erin R. J ohnson,‡ Melinda R.Vinqvist,§ J ames S. Wright,‡
L. Ross C. Barclay,* and K. U. Ingold|
Department of Chemistry, Carleton University, 1125 Colonel By Drive, Ottawa,
Ontario, K1S 5B6 Canada, Department of Chemistry, Mount Allison University,
Sackville, N.B., E4L 1G8 Canada, and National Research Council, Ottawa, Ontario, K1A 0R6 Canada
rbarclay@mta.ca
Received March 15, 2002
1,8-Naphthalenediol, 5, and its 4-methoxy derivative, 6, were found to be potent H-atom transfer
(HAT) compounds on the basis of their rate constants for H-atom transfer to the 2,2-di(4-t-
octylphenyl)-1-picrylhydrazyl radical (DOPPH•), kArOH/DOPPH , or as antioxidants during inhibited
•
•
styrene autoxidation, kArOH/ROO , initiated with AIBN. The rate constants showed that 5 and 6 are
more active HAT compounds than the ortho-diols, catechol, 1, 2,3-naphthalenediol, 2, and 3,5-di-
6
•
tert-butylcatechol, 3. Compound 6 has almost twice the antioxidant activity, kArOH/ROO ) 6.0 × 10
M-1 s-1, of that of the vitamin E model compound, 2,2,5,7,8-pentamethyl-6-chromanol, 4.
Calculations of the O-H bond dissociation enthalpies compared to those of phenols, (∆BDEs), of
1-6 predict a HAT order of reactivity of 2 < 1 < 3 ≈ 4 < 5 < 6 in general agreement with kinetic
results. Calculations on the diols show that intramolecular H-bonding stabilizes the radicals formed
on H-atom transfer more than it does the parent diols, and this effect contributes to the increased
HAT activity of 5 and 6 compared to the activities of the catechols. For example, the increased
stabilization due to the intramolecular H-bond of 5 radical over 5 parent of 8.6 kcal/mol was about
•
double that of 2 radical over 2 parent of 4.6 kcal/mol. Linear free energy plots of log kArOH/DOPPH
•
and log kArOH/ROO versus ∆BDEs for compounds 1-6 along with available literature values for
nonsterically hindered monophenols placed the compounds on common scales. The derived Evans-
•
•
Polanyi constants from the plots for the two reactions, RDOPPH ) 0.48 > RROO ) 0.32, gave the
expected order, since the ROO• reaction is more exothermic than the DOPPH• reaction. Compound
•
6 is sufficiently reactive to react directly with oxygen, and it lies off the log kArOH/ROO versus ∆BDE
plot.
In tr od u ction
standard for comparison with the antioxidant efficiencies
of synthetic phenolic antioxidants. The antioxidant ac-
tivities of phenols are best evaluated by determination
of the rate constants for H-atom abstraction from the
phenolic groups by peroxyl radicals.4
The reactions of oxygen-centered radicals, such as
peroxyls R-O-O•, with biological molecules in vivo are
implicated in various degenerative diseases,1 such as
cancer, heart disease, inflammation, and the aging
process.2 Consequently, the mechanism and activities of
natural and synthetic antioxidants continue to receive a
great deal of attention. R-Tocopherol, the natural phenolic
antioxidant which is the main lipid-soluble chain-break-
ing antioxidant in human blood,3 is commonly used as a
Active antioxidants of the polyalkylchromanol class
(e.g., vitamin E) owe their activity to a combination of
electronic, steric, and stereoelectronic effects which lower
the bond dissociation enthalpy (BDE) of the O-H bond
of the phenol and consequently increase the rate of its
reaction with peroxyl radicals.4 Interest in “maximizing”
the antioxidant activity of phenols led to the development
of a pentamethylfuran analogue of R-tocopherol which
increased the stereoelectronic effect by providing better
overlap of the para ether oxygen’s p-orbital with the
aromatic ring in the intermediate phenoxyl radical
formed by reaction with peroxyls. This increased the
antioxidant activity by a factor of 1.8 over those of
chromanols of the R-tocopherol (vitamin E) class.4a Fur-
* Author for correspondence: L. R. C. Barclay, Department of
Chemistry, Mount Allison University, 63C York Street, Sackville,
N.B. E4L 1G8, Canada.
† On leave from the CNR Istituto per Io Studio delle Sostanze
Naturali, Via del Santuario 110, 95028 Valverde (CT), Italy.
‡ Carleton University.
§ Mount Allison University.
| National Research Council.
(1) Halliwell, B.; Gutteridge, J . M. C. Free Radicals in Biology and
Medicine, 3rd ed.; Oxford University Press: New York, 1999.
(2) Free Radicals in Aging; Yu, B. P., Ed.; CRC Press: Boca Raton,
FL, 1993.
(3) Burton, G. W.; J oyce, A.; Ingold, K. U. Arch. Biochem. Biophys.
1982, 221, 281-290.
(4) (a) Burton, G. W.; Ingold, K. U. Acc. Chem. Res. 1986, 19, 194-
201. (b) Burton, G. W.; Doba, T.; Gabe, E. J .; Hughes, I.; Lee, F. L.;
Prasad, L.; Ingold, K. U. J . Am. Chem. Soc. 1985, 107, 7053-7065.
10.1021/jo020184v CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/22/2002
5190
J . Org. Chem. 2002, 67, 5190-5196

Naphthalene Diols: A New Class of Antioxidants
SCHEME 1
ther flattening of the furan ring did not produce any
additional effect.5 However, increased stabilization of the
incipient ArO• radical by incorporating the furan ring into
the R-naphthol aromatic system produced an antioxidant
10 times more active than vitamin E.6
The search for new structural features which increase
the antioxidant activities of phenols is of significance to
biomedicine because of the potential development of new
protective antioxidants and to chemistry because of
continuing interest in the relationship between structure
and reactivity. In this connection, there is considerable
evidence that intramolecular H-bonding in the catechol
ring system has a pronounced effect on antioxidant
activity. Earlier, we found that catechols are very active
antioxidants, 3,5-di-tert-butylcatechol having an antioxi-
7
•
dant activity (kArOH/ROO ) one-half that of R-tocopherol.
the formation of superoxide,16 and quinone methides
derived from 4-alkylcatechols cause damage through
alkylation of cellular proteins and DNA.17
The key structural feature is the intramolecular hydro-
gen bond, since recent calculations8 and experiments9
have both shown that the H-bond stabilizes the o-
semiquinone radical by about 4 kcal/mol more than it
does the parent catechol. A very large class of natural
compounds known as flavonoids contain the catechol
structure, and there are several reviews emphasizing the
antioxidant properties of these polyphenols.10 Further-
more, there is theoretical11 and experimental12 evidence
linking the antioxidant activities of flavonoids to their
catechol systems. This kind of structure also occurs in
catechol amines such as L-dopa and dopamine, which are
reported to have both toxic and antioxidant effects,13 in
catechol steroids, for which effective antioxidant activity
was found in lipoproteins,14 and in rat liver microsomes.15
However, both catechols and 1,4-hydroquinones have
associated toxic properties in biological systems. The
cytotoxicity appears to be due to two processes. Redox
cycling between a semiquinone and quinone results in
As part of a program to design and test new antioxi-
dants, we considered the 1,8-naphthalenediol system to
be a promising starting point for three reasons. First, the
corresponding aryloxyl radical is expected to be stabilized
by hydrogen bonding in a manner similar to that in the
semiquinone radical from catechol, although the ring size
containing the H bond differs (six-member versus five-
member). Second, the naphthalene ring system is ex-
pected to provide additional stabilization of the H-bonded
intermediate, and third, this system cannot form a
quinone in the same way as catechols, so the toxicity
shown by some quinones should be absent. The relation-
ships between the structures are shown in Scheme 1. It
is postulated that the increased stabilization of the
semiquinone radical, 1′, compared to that of the parent,
1, is due to the increase in strength of the intramolecular
H-bond induced by dipolar a and keto-enol b contribu-
tions to the radical’s structure. Similar interactions are
expected to be at least as strong in structures c and d
from the 8-hydroxy-1-naphthyloxyl radical, 5′, formed
from parent compound 5.
(5) Gilbert, J . C.; Pinto, M. J . Org. Chem. 1992, 57, 5271-5276.
(6) Barclay, L. R. C.; Vinqvist, M. R.; Mukai, K.; Itoh, S.; Morimoto,
H. J . Org. Chem. 1993, 58, 7416-7420.
(7) Xi, F.; Barclay, L. R. C. Can. J . Chem. 1998, 76, 171-182.
(8) (a) Wright, J . S.; J ohnson, E. R.; DiLabio, G. A. J . Am. Chem.
Soc. 2001, 123, 1173-1183. (b) Wright, J . S.; Carpenter, D. J .; McKay,
D. J .; Ingold, K. U. J . Am. Chem. Soc. 1997, 119, 4245-4252.
(9) Lucarini, M.; Mugnaini, V.; Pedulli, G. F. J . Org. Chem. 2002,
67, 928-931.
At this time, we report on the following: (a) synthesis
of 4-methoxy-1,8-naphthalenediol, 6, a new powerful
hydrogen atom transfer agent (HAT); (b) the H-atom
(10) (a) Rice-Evans, C. A.; Miller, N. J .; Paganga, G. Trends Plant
Sci. 1997, 2, 152-159. (b) Kandaswami, C.; Middleton, E. In Natural
Antioxidants, Chemistry, Health Effects, and Applications; Shahidi, F.,
Ed.; AOAC Press: Champaign, IL, 1997; Chapter 10, pp 174-203. (c)
Li, W.; Sun, G. Y. In Biological Oxidants and Antioxidants. Molecular
Mechanisms and Health Effects; Packer, L., Ong, A. S. H., Eds.; AOAC
Press: Champaign, IL, 1998; Chapter 12, pp 90-103. (d) Cook, N. C.;
Samman, S. J . Nutr. Biochem. 1996, 7, 66-76. (e) Formica, J . V.;
Regelson, W. Food Chem. Toxicol. 1995, 33, 1061-1080.
•
)
donating ability of antioxidants 5 and 6 (kArOH/DOPPH
toward the 2,2-di(4-t-octylphenyl)-1-picrylhydrazyl (DOP-
PH•) radical and the antioxidant activities (kArOH/ROO ) of
•
these naphthalenediols compared to catechols and typical
monophenolic antioxidants; and (c) calculations of bond
dissociation enthalpies (BDEs) of naphthalenediol sys-
tems, some selected catechols, and typical phenolic
antioxidants.
(11) van Acker, S. A. B. E.; de Groot, M. J .; van den Berg, D. J .;
Tromp, M. N. J . L.; den Kelder, G. D. O.; van der Vijh, W. J . F.; Bast,
A. Chem. Res. Toxicol. 1996, 9, 1305-1312.
(12) (a) J ovanovic, S. V.; Steenken, S.; Hara, Y.; Simic, M. G. J .
Chem. Soc., Perkin Trans. 2 1996, 2497-2504. (b) Arora, A.; Nair, M.
G.; Strasburgh, G. M. Free Radical Biol. Med. 1998, 24, 1355-1363.
(c) J ovanovic, S. V.; Steenken, S.; Tosic, M.; Marjanovic, B.; Simic, M.
J . Am. Chem. Soc. 1994, 116, 4846-4952. (d) Foti, M.; Piattelli, M.;
Baratta, M. T.; Ruberto, G. J . Agric. Food Chem. 1996, 44, 497-501.
(13) (a) Hastings, T. G.; Lewis, D. A.; Zigmond, M. J . Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 1956-1961. (b) Mytilineou, C.; Han, S. K.;
Cohen, G. J . Neurochem. 1993, 61, 1470-1478.
(14) (a) Tang, M.; Abplanalp, W.; Ayers, S.; Subbiah, M. T. R. Metab.,
Clin. Exp. 1996, 45, 411-414. (b) Taniguchi, S.; Yanase, T.; Kobayashi,
K.; Takayanagi, R.; Haji, M.; Umeda, F.; Nawata, H. Endocrinol. J pn.
1994, 41, 605-611.
(15) (a) Lacort, M.; Leal, A. M.; Liza, A. M.; Martin, C.; Matiinez,
R.; Ruiz-Larrea, M. B. Lipids 1995, 30, 141-146. (b) Takanashi, K.;
Watanabe, K.; Yoshizawa, I. Biol. Pharm. Bull. 1995, 18, 1120-1125.
Resu lts
1. Syn th eses. Synthetic schemes for the target mol-
ecules 5 and 6 are given in Scheme 2. The 1,8-naphtha-
lenediol, 5, a known compound, was readily prepared
(16) (a) Inbaraj, J . J .; Gandhidasan, R.; Murugesan, R. Free Radical
Biol. Med. 1999, 26, 1072-1078. (b) Inbaraj, J . J .; Krishna, M. C.;
Gandhidasan, R.; Murugesan, R. Biochim. Biophys. Acta 1999, 1472,
462-470.
(17) (a) Bolton, J . L.; Pisha, E.; Shen, L.; Krol, E. S.; Iverson, S. L.;
Huang, Z.; van Breemen, R. B.; Pezzuto, J . M. Chem.-Biol. Interact.
1997, 106, 133-148. (b) Krol, E. S.; Bolton, J . L. Chem.-Biol. Interact.
1997, 104, 11-27.
J . Org. Chem, Vol. 67, No. 15, 2002 5191

Foti et al.
by following the decay of DOPPH• at 519 nm in the
presence of excess antioxidant at various concentrations
using a stopped-flow spectrophotometer at room temper-
ature (eq 1).
SCHEME 2
kArOH/DOPPH
ArO-H + DOPPH•
•8 ArO• + DOPPH-H
(1)
Under these pseudo-first-order conditions, eq 2 applies:
ks ) k0 + ksArOH/DOPPH•[ArOH]
(2)
exptl
Excellent correlation coefficients were obtained in all
cases (r2 g 0.98) from plots of ks
versus [ArOH] from
exptl
which ks
values were calculated (see Table 1).
ArOH/DOPPH•
The naphthalenediols, 5 and 6, are very active H-atom
donors compared to catechols, such as 1-3, and more
active than the vitamin E model compound, 4.19 The
activity of 2,3-dihydroxynaphthalene, 2, is somewhat
smaller than might have been expected, since the second
aromatic ring might have been expected to lower the
O-H bond dissociation enthalpy (BDE) in 2 compared
to that in catechols 1 or 3.
The antioxidant activities of the same six compounds,
1-6, were determined by measuring absolute rate con-
stants for reactions with peroxyl radicals, R-O-O•, using
the inhibition of oxygen uptake (IOU) method. We used
styrene, which is known to have several advantages as
a substrate for quantitative studies of antioxidants,4b and
controlled the rate of free radical initiation, Ri, using the
azo initiator, azo-bis-isobutyrylnitrile (AIBN). The rel-
evant reactions are given in eqs 3 and 4, and the
expression for oxygen uptake is given by eq 5.
from naphthosultone, as shown. The 4-methoxy-1,8-
naphthalenediol, 6, is a new compound, and the natural
product juglone is a convenient starting material for its
synthesis. The latter was readily reduced to the triol
intermediate, I, as shown, but because of the susceptibil-
ity of this compound to air oxidation, it was converted to
its triacetate, II, by reaction with acetic anhydride in
pyridine. The triacetate is stable in air. It was selectively
deacetylated by enzymolysis at the 4-position to give III.
The diacetate, III, was methylated to yield IV which was
then hydrolyzed to produce the desired methoxy deriva-
tive, 6.18
kArOH/ROO
ArO-H + R-O-O•
•8 Ar-O• + R-O-OH
(3)
Followed by Ar-O• + R-O-O• fast8
nonradical products (4)
2. H-Atom Don a tin g P r op er ties a n d An tioxid a n t
Activities. To determine the H-atom donating ability of
the diols 5 and 6, some catechols, and a typical hydroxy-
chroman of the vitamin E class, we measured their
reactivity toward the DOPPH• radical in hexane. The
-d[O2]/dt ) kp/kArOH/ROO•[styrene]Ri/n[Ar-OH] (5)
The propagation rate constant, kp, for the autoxidation
of styrene is taken from the literature to be 41 M-1s-1 at
30 °C.4b The antioxidant activities are determined by
•
second-order rate constants, kArOH/DOPPH , were obtained
evaluating the absolute rate constant for inhibition,
k
(18) The synthesis of the target molecule 6 was accomplished in the
laboratory of G. Nicolosi and C. Rocco, CNR Instituto per lo Studio
delle Sostanze Naturali (Italy) as follows: The naphthalene triol I on
acetylation by treatment with acetic anhydride in pyridine yielded a
triacetate derivative II which gave the expected 13C and 1H NMR
spectra in CDCl3 [δ 2.42 (6H, 2 × CH3COO, positions 4 and 5); 2.49
(3H, 1 × CH3COO, position 1); 7.14-7.33 (aromatic Hs at positions 2
and 3, a and b quartet); 7.19, 7.23, 7.85, and 7.89 (Hs at 6 and 8, split
doublets); 7.54 (H at 7, triplet)]. This triacetate, which was stable in
air, was selectively hydrolyzed at the 4 position by treatment in
solution with a lipase enzyme at 40 °C. The resulting monophenol III
was methylated with trimethylsilyldiazomethane to give the 4-meth-
oxydiacetate derivative IV which gave the expected 1H NMR spectrum
[δ 4.03 (OCH3); 2.41 and 2.42 (2 × CH3COO)]. Further lipase-catalyzed
hydrolysis of the latter yielded the required 6 in 70% overall yield from
the triacetate. Compound 6 prepared in this way showed a prominent
M+ peak (60%) at m/e ) 190 with a base peak at 175 for loss of CH3.
1H NMR spectrum in deuterated acetone (TMS): δ 3.90 (3H, OCH3);
five distinct multiplets between 6.68 and 7.66 (each corresponding to
ArOH/ROO . This was done as usual4b,6 by measuring oxygen
•
consumed during the induction period. The rate constant
•
for inhibition, kArOH/ROO , was calculated using the inte-
grated form of the equation for inhibited oxidation, eq 6,
which in each case gave a linear plot of ∆[O2]t versus -
ln(1 - t/τ), where τ is the duration of the induction period.
∆[O2]t ) -kp/kArOH/ROO•[styrene] ln(1 - t/τ) (6)
The number of peroxyl radicals trapped per molecule of
antioxidant, the stoichiometric factor, n, was determined
via eq 7, and the rate of initiation by peroxyl radicals,
Ri, was calculated from the induction period obtained
with pentamethylhydroxychroman, 4, an antioxidant for
one H, aromatic hydrogens); 9.69 and 10.06 (hydroxyl hydrogens). 13
C
spectrum: 55.23 (OCH3 removed on DEPT spectrum); 10 resolved
peaks for aromatic carbons. DEPT spectra: CH groups at 104.7, 107.6,
109.8, 113.3, 126.4; quaternary carbons at 115.4, 128.4, 147.2, 148.9,
and 154.4 δ.
•
(19) Literature values for kArOH/DPPH for 3 and 4 in hexane are 20 ×
103 and 6.8 × 103 M-1 s-1, respectively: Barclay, L. R. C.; Edwards,
C. E.; Vinqvist, M. R. J . Am. Chem. Soc. 1999, 121, 6226-6231.
5192 J . Org. Chem., Vol. 67, No. 15, 2002

Naphthalene Diols: A New Class of Antioxidants
•
•
TABLE 1. Hyd r ogen Atom Don a tin g Abilities, kAr OH/DOP P H , a n d An tioxid a n t Activities, kAr OH/ROO , of Na p h th a len e Diols,
Ca tech ols, a n d a Vita m in E Mod el Com p ou n d
a
b
•
•
kArOH/DOPPH
kArOH/ROO
compound
(×10-3 M-1 s-1
)
(×10-6 M-1 s-1
)
nc
catechol, 1
2,3-naphthalenediol, 2
1.8
1.3
21
7.4
310
2000
0.55
2.3
2.4
0.02-0.04d
3,5-di-tert-butylcatechol, 3
1.5
3.3
4.3
6.0
2.3
2,2,5,7,8-pentamethyl-6-hydroxychroman, 4
1,8-naphthalenediol, 5
(2.0)
1.5
4-methoxy-1,8-naphthalenediol, 6
1.1
a
b
H-atom donating ability to the DOPPH• radical in hexane at 25 °C under argon. Antioxidant activities determined by the inhibited
oxygen uptake method during oxidation of styrene at 30 °C, initiated by AIBN. Values for 1 and 3 are taken from ref 7. c Stoichiometric
factors for the ArOH/ROOC reaction relative to a value of 2.0 for 4. For compound 2, n was determined in cumene relative to a value of
2.0 for 2,6-di-tert-butyl-4-methoxyphenol. The kArOH/ROO value was estimated from the initial rate of oxygen uptake.24
d
•
•
•
which the stoichiometric factor, n, is known to be ap-
nation of kArOH/ROO for 2. The kArOH/ROO in styrene was
proximately 2.4b
estimated from the reduced oxygen uptake.24
3. Ca lcu la tion s. (a ) Th eor etica l Meth od s. The
Gaussian 98 program25 was used for calculations of the
bond dissociation enthalpies (BDEs) using the Medium
Level Model 2 (MLM2).26 In the MLM2 method, the
electronic energy of the H-atom was set to its exact value,
-0.500 00 hartree (1 hartree ) 627.51 kcal mol-1). The
geometry was optimized for the parent and radical using
B3LYP/6-31G(d). Applying the temperature correction of
5/2RT, the enthalpy of the H-atom at 298 K is -0.497 64
hartree. Frequencies for ArOH and ArO• were determined
by the same method using B3LYP/6-31G(d), and zero-
point energy and enthalpy corrections were scaled by the
factor 0.9806. At the potential minimum, the single-point
energy was determined using (RO)B3LYP/6-311+G(2d,-
2p), where RO means restricted open-shell. The complete
method for ArOH and ArO• can be represented as
(RO)B3LYP/6-311+G(2d,2p)//B3LYP/6-31G(d)/B3LYP/6-
31G(d). In the cases of PMHC (compound 4) and of
compounds 7 and 8, the locally dense basis set (LDBS)
approach8,27,28 was used to obtain the single-point ener-
gies of the parent and radical because of the large size
of these species. The LDBS approach may be represented
as (RO)B3LYP/LDBS//B3LYP/6-31G(d)/B3LYP/6-31G(d).
This has been shown to give results which are of
comparable accuracy to the balanced basis set approach,
with agreement to within (1 kcal/mol. For example, the
BDE of phenol has been calculated to be 87.5 kcal/mol
with both the balanced basis set and LDBS methods.
(b) Resu lts. Calculated O-H bond dissociation en-
thalpies (BDEs) for O-H bonds are listed in Table 2. For
Ri ) n[Ar-OH]/τ
(7)
The n factors for the compounds studied were determined
from the induction periods measured under the same
conditions as had been used for the measurement of Ri.
•
The antioxidant activities, kArOH/ROO , of the naphtha-
lenediols 5 and 6 and the other antioxidants studied in
this work are given in Table 1. The value for the
R-tocopherol model, 4, is in agreement with the literature
value (3.8 × 106 M-1 s-1).4b The kArOH/ROO values for the
•
naphthalenediols 5 and 6 show that they both are VERY
active peroxyl radical traps, more active than either of
the catechols and more active than the vitamin E model,
4. The increased activity of 6 over that of the parent diol,
5, can be attributed to the electron-supplying 4-methoxy
group which is expected to weaken the phenolic O-H
BDE and hence accelerate the reaction with peroxyl
radicals. Unfortunately, 6 reacts directly with dioxygen,
its solutions turning dark green on standing in air,
yielding at least four products (TLC analysis). This
“wasting” reaction with oxygen and possibly chain trans-
fer reactions6 are undoubtedly responsible for the unusu-
ally low stoichiometric factor found for 6 (viz. n ) 1.1).
The 2,3-naphthalenediol, 2, is a surprisingly poor trap
for peroxyl radicals and behaves only as a “retarder”
under these conditions. As explained earlier,20,21 the
behavior of a retarder is quite different from that of an
efficient radical trapping antioxidant. Retarders react
relatively slowly with peroxyl radicals, only slightly
reduce the rate of oxygen uptake, and do not give well-
defined induction periods. This was the case with 2 in
styrene. The stoichiometric factor for 2 had therefore to
be measured in cumene which has a kp of only 0.18 M-1
s-1 (ref 22) at 30 °C. In this substrate, even very poor
peroxyl radical trapping antioxidants give well-defined
induction periods.23 The stoichiometric factor determined
in this way for 2 was 2.4 (relative to 2.0 for 2,6-di-tert-
butyl-4-methoxyphenol).23 There was insufficient oxygen
uptake in cumene to use the IOU method for determi-
(24) The kArOH/ROO was estimated to be (2-4) × 104 M-1 s-1 from
•
the initial reduced rate of oxygen uptake using the method of: Foti,
M.; Roberto, G. J . Agric. Food Chem. 2001, 49, 342-348.
(25) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA,
1998.
(20) Waters, W. A.; Wickham-J ones, C. J . Chem. Soc. 1951, 51, 812-
823.
(21) Howard, J . A.; Ingold, K. U. Can. J . Chem. 1964, 42, 2324-
2333.
(26) DiLabio, G. A.; Pratt, D. A.; LoFaro, A. D.; Wright, J . S. J . Phys.
Chem. A 1999, 103, 1653-1661.
(22) Howard, J . A.; Ingold, K. U.; Symonds, M. Can. J . Chem. 1968,
46, 1017-1022.
(27) DiLabio, G. A.; Pratt, D. A.; Wright, J . S. Chem. Phys. Lett.
1998, 297, 181-186.
(23) Horswill, E. C.; Howard, J . A.; Ingold, K. U. Can. J . Chem. 1966,
44, 985-991.
(28) DiLabio, G. A. J . Phys. Chem. A 1999, 103, 11414-11424.
J . Org. Chem, Vol. 67, No. 15, 2002 5193

Foti et al.
TABLE 2. Ca lcu la ted Bon d Dissocia tion En th a lp ies of
O-H Bon d s a n d Hyd r ogen Bon d En th a lp ies of P h en ols
1-8 in k ca l/m ol
As already pointed out, intramolecular hydrogen bond-
ing is important in determining the O-H BDEs of these
diols and, therefore, their antioxidant activities. The
intramolecular H-bonds in the naphthalenediols are
stronger than those in the catechols for two reasons.
First, our calculations indicate that the length of the
hydrogen bond is shorter in the naphthalenediols, viz.
1.79 Å in 5 and 2.12 Å in 1. Second, the O-HsO angle
is more nearly linear in the naphthalenediols, viz. 143°
in 5 and 115° in 1. Thus, the greater HAT activities of
the naphthalenediols 5 and 6 relative to the activities of
1 and 2 can be attributed, at least in part, to the greater
intramolecular H-bond-induced stabilization of radicals
5′ and 6′, relative to their parent molecules 5 and 6 (viz.,
for 5′/5, 14.8 kcal/mol - 6.2 kcal/mol ) 8.6 kcal/mol and,
for 6′/6, 15.2 kcal/mol - 6.6 kcal/mol ) 8.6 kcal/mol)
compared with the 1′/1 couple (9.1 kcal/mol - 3.8 kcal/
mol ) 5.3 kcal/mol) and the 2′/2 couple (8.2 kcal/mol -
3.6 kcal/mol ) 4.6 kcal/mol).
compounda
-∆BDEb
parent H-bond
radical H-bond
1
2
3
4
5
6
7
8
10.0
9.2
3.8
3.6
c
9.1
8.2
c
12.6
12.6
15.7
20.1
15.8
16.8
6.2
6.6
14.8
15.2
a
Compounds 7 and 8 were described previously.6
The data from our Tables 1 and 2 and related literature
data for a wide range of nonhindered monophenolic
antioxidants8,29-31 are summarized in the form of Evans-
b
Differences in bond dissociation enthalpies: BDE(ArOH) -
BDE(PhOH), where BDE(PhOH) ) 87.5 kcal/mol. For compounds
with two different “active sites” for hydrogen atom transfer (e.g.
non H-bonded and intramolecularly H-bonded O-H), 3 and 6, the
value is for the nonbonded O-H. c See text.
Polanyi32 plots of -∆BDE versus log kArOH/DOPPH (in
•
•
hexane at 25 °C) and log kArOH/ROO (in styrene at 30 °C)
(see Figure 1). According to Evans and Polanyi, for a
series of similar reactions, such as Y• + H-OAr, the
activation enthalpy, Ea, will be proportional to the
enthalpy of reaction, ∆Hreact. That is, for the same radical
Y•,
the diols with their intramolecular hydrogen bonds, BDE
values are are given for the “free”, nonbonded hydroxyl
group. Since the intramolecular hydrogen bonds in both
the parent molecule and its radical play an important
role in the behavior of these compounds, hydrogen bond
strengths in both the parents and their radicals were
calculated for catechol, 1, 2,3-naphthalenediol, 2, and the
naphthalenediols 5 and 6. For these calculations, the
hydrogen bond strength is defined as the enthalpy
difference for an OH group coplanar with the aromatic
ring pointing away and toward the hydrogen bond
accepting group. This calculation was not performed on
3 because the “away” conformer would point toward an
adjacent tert-butyl group, and our method of calculation
is known to overestimate the steric strain in such
molecules.8a For compound 6, the conformation with the
hydroxyl group at the 1-position as the H-bond acceptor
is the more stable by 0.4 kcal/mol. The results are
summarized in Table 2. Optimized structures of the
parent compounds 1-8 and the corresponding radicals
are given as Supporting Information.
∆Ea ) R∆∆Hreact ) R∆BDE, 0 < R < 1
(8)
where ∆BDE ) BDE(ArO-H) - BDE(PhO-H) and
BDE(PhO-H) was calculated to be 87.5 kcal/mol. For the
same Y•, the Arrhenius pre-exponential factors will be
similar or even idential33 and therefore eq 8 can be
written as the following:
kArOH/kPhOH ) exp(-R∆BDE/RT)
(9a)
or log kArOH ) log kPhOH - R∆BDE/2.303RT (9b)
The value R depends on the thermochemistry of the
overall H-atom transfer reactions.34 That is, R will be
small (even zero) for highly exothermic reactions and will
(29) Table S1 of ∆BDEs and rate constants for reactions of phenols
with DPPH•, DOPPH, and ROO• used in Figure 1 is available as
Supporting Information.
(30) Howard, J . A.; Ingold, K. U. Can. J . Chem. 1963, 41, 1744-
1751.
(31) Snelgrove, D. W.; Lusztyk, J .; Banks, J . T.; Mulder, P.; Ingold,
K. U. J . Am. Chem. Soc. 2001, 123, 469-477.
Discu ssion
(32) Evans, M. G.; Polanyi, M. Trans. Faraday Soc. 1938, 34, 11-
It is of interest to compare the kinetic data for H-atom
donating abilities and antioxidant activities of compounds
1-6 with the theoretical calculations of BDEs which are
expected to predict the order of “activity”. If one assumes
29.
(33) Foti, M.; Ingold, K. U.; Lusztyk, J . J . Am. Chem. Soc. 1994,
116, 9440-9447.
(34) For a series of substituted phenols in a nonalkane solvent, the
overall thermochemistry will, of course, contain a contribution from
intermolecular hydrogen bonding. This contribution will not be a
constant along the series but will be larger for phenols with electron-
withdrawing (EW) substituents than for phenols with electron-
donating (ED) substituents. Since EW substituents reduce and ED
substituents enhance the H-atom donating abilities of phenols, an
•
•
that the relative kArOH/DOPPH and the relative kArOH/ROO
are determined primarily by the O-H BDE, then the
following order of reactivity, from the least active to most
active, would be predicted: 2 < 1 < 3 ≈ 4 < 5 < 6.
Evans-Polanyi plot for any specific reaction in
a hydrogen bond
•
Experimentally, kArOH/DOPPH gives 2 < 1 < 4 < 3 < 5 < 6
accepting (HBA) solvent will have a larger slope, i.e, a larger R value,
than that for the same reaction in an alkane or other non-HBA solvent.
This effect will not be large for the present reactions, but it does mean
that the R value for the ArOH/ROO• reaction in styrene (a weak HBA)
will be slightly enhanced over the value which would have been in a
non-HBA solvent.
•
and kArOH/ROO gives 2 < 1 < 3 < 4 < 5 < 6, in general
agreement with the calculated O-H BDEs. This agree-
ment supports a hydrogen atom transfer (HAT) mecha-
nism as the rate-determining step for both reactions.
5194 J . Org. Chem., Vol. 67, No. 15, 2002

Naphthalene Diols: A New Class of Antioxidants
F IGURE 1. Linear free energy plots of H-atom transfer rate constants versus calculated bond dissociation energies compared to
those of phenol (-∆BDEs). Plots of log kArOH/DOPPHC (]) and log kArOH/ROOC (2) versus calculated -∆BDEs, reference PhOH ) 87.5
kcal/mol, for compounds 1-6 and literature data log kArOH/DPPHC values for I ) 4-CF3-C6H4OH, II ) 3,5-Cl2-C6H3OH, III )
4-CN-C6H4OH, IV ) 3-Cl-C6H4OH, V ) C6H5OH (set ) 0), VI ) 3-Me-C6H4OH, VII ) 3-MeO-C6H4OH, VIII ) 4-Cl-C6H4OH,
IX ) 4-Me-C6H4OH, X ) 4-t-pentyl-C6H4OH, XI ) 4-OH-C6H4OH, XII ) 4-MeO-C6H4OH, 7 ) 2,5-dimethyl-2-phytyl-6-hydroxy-
7,8-benzochroman, and 8 ) 2,2,4-trimethyl-5-hydroxynaphtho[1,2-b]-2,3-dihydrofuran. Calculations for I, III-IX, XI, and XII
are from ref 8a and were performed using the LLM method, which may be denoted as (RO)B3LYP/6-311+G(2d,2p)//AM1/AM1.
For compound II, the BDE was estimated using the activity scheme from ref 8a. For compound X, the calculated BDE of 4-tert-
butylphenol from ref 8a was used. The ArOH/DPPH• rate constants for I-XII are from ref 31. The ArOH/ROO• rate constants for
I-XII are from ref 30 but have been multiplied by the correction factor 2.24, as in: Burton, G. W.; Ingold, K. U. J . Am. Chem.
Soc. 1981, 103, 6472-6477. The value for XI was divided by 2 for statistical reasons. To convert rate constants from 65to 30 °C,
an average Ea of 2.5 kcal/mol was used as an approximation. The ArOH/ROO• rates for 1 and 3 are from ref 7, and those for 7 and
8 are from ref 6.
increase up to a limit (presumably) of 1.0 for highly
endothermic reactions. Since the N-H bond energy in
DPPH-H is ∼79.635 kcal/mol and the O-H BDE in
fall below the ROO•/ArOH line. We suggest that this
lower than expected reactivity of 6 toward ROO• radicals
and lower stoichiometric factor (i.e. n ) 1.1) are likely a
consequence of the “wastage” of 6 in its observed direct
reaction with oxygen.
•
ROO-H is ∼88 kcal/mol, RDPPH is expected to be larger
•
than RROO . The plots can be described by the following:
In any case, our search for new antioxidants based on
the 1,8-naphthalenediol system which would be more
active than R-tocopherol and 4 has been successful with
the parent diol, 5. Although compound 6 with its methoxy
substituent is a very active HAT agent, this compound’s
sensitivity to oxygen results in a lower than “expected”
antioxidant efficiency. Substituents that are less electron
rich than methoxy may provide derivatives of these diols
which are not sensitive to direct reaction with oxygen.
Such derivatives may prove to be efficient antioxidants
in natural biological media where lower oxygen partial
pressures prevail.37,38
log kArOH/DOPPH• ) -0.33 + 0.35(-∆BDE),
r2 ) 0.99; yielding RDOPPH• ) 0.48 (10)
and log kArOH/ROOC ) 3.39 + 0.23(-∆BDE),
r2 ) 0.98; yielding RROO• ) 0.32 (11)
•
•
As expected, RDOPPH > RROO
.
The only serious outlier in Figure 1 is the point for the
reaction of ROO• radicals with compound 6 which falls
well below the best line through all of the ROO•/ArOH
data. Since the point for the DOPPH• + 6 reaction also
falls below the best line through the DOPPH•/ArOH data,
it seems probable that ∆BDE for 6 has been overesti-
mated,36 perhaps by as much as 2 kcal/mol; that is, the
O-H BDE should probably be ∼69 kcal/mol rather than
the calculated 67 kcal/mol. While such a 2 kcal/mol
“adjustment” to the ∆BDE would put the DOPPH•/6 point
on the DOPPH•/ArOH line, the ROO•/6 point would still
Exp er im en ta l Section
Ma ter ia ls a n d Gen er a l Meth od s. All solvents used were
of highest purity, dried over molecular sieves, and distilled
just before use. The chemicals, 1,8-naphthosultone, 5-hydroxy-
1,4-naphthoquinone (juglone), 2,2,5,7,8-pentamethyl-6-chro-
manol (PMHC, 4), 2,3-naphthalenediol (2), catechol (1), 3,5-
di-tert-butylcatechol (3), styrene, and cumene were purchased
from a commercial supplier. The 2,2-di(4-t-octylphenyl)-1-
picrylhydrazyl radical (DOPPH•) was obtained from Northern
Sources, Inc.
(35) Mahoney, L. R.; Mendenhall, G. D.; Ingold, K. U. J . Am. Chem.
Soc. 1973, 95, 8610-8614.
Syn th eses of Com p ou n d s 5 a n d 6. Compound 5 was
(36) It has been observed before that calculated O-H BDEs in very
electron rich phenols are smaller than actual values (Pratt, D. A.;
DiLabio, G. A.; Brigati, G.; Pedulli, G. F.; Valgimigli, L. J . Am. Chem.
Soc. 2001, 123, 4625-4626) but in other electron rich phenols there
was agreement between calculated and measured values (de Heer, M.
I.; Korth, H.-G.; Mulder, P. J . Org. Chem. 1999, 64, 6969-6975).
synthesized by a modification of the literature procedure.39 For
(37) Burton, G. W.; Ingold, K. U. Science 1984, 224, 569-573.
(38) Barclay, L. R. C.; Vinqvist, M. R. Free Radical Biol. Med. 2000,
28, 1079-1090.
J . Org. Chem, Vol. 67, No. 15, 2002 5195

Foti et al.
convenience and safety, the reaction was carried out in a deep
(12 cm × 4 cm) stainless steel cylinder surrounded by an
electrically heated jacket and fitted with a thermocouple for
temperature control. A mechanically stirred mixture of potas-
sium hydroxide (15 g) and sodium hydroxide (15 g) was melted
at 300 °C in an argon atmosphere. The naphthosultone, 5 g,
was added in small portions, and the resulting dark liquid was
stirred for about 30 min. This product was then poured onto
aluminum foil and allowed to solidify. The solid was ground
into small pieces and then stirred with 200 mL of hydrochloric
acid/water (1:2 v/v). The organic material was extracted with
ethyl acetate (3 times, 100 mL). This solution was dried over
anhydrous magnesium sulfate, and the solvent was removed
on a rotatory evaporator to leave a dark liquid. Purification
was effected in two stages. The crude material was dissolved
in the minimum volume of hexane and chromatographed on
silica gel. A dark brown band remained on the column while
the product eluted with hexane/ethyl acetate (9:1 v/v). Distil-
lation of the solvent left a pale yellow crystalline product.
Recrystallization of this from hexane containing a small
portion of ethanol yielded white needles (40% yield): mp 143-
antioxidants of varying concentrations. For the more soluble
compounds, the concentration range was 7 × 10-4 to 6 × 10-3
M, and for less soluble compounds (e.g., 5), it was 1 × 10-4 to
9 × 10-4 M. The separate solutions were deoxygenated by
bubbling with argon before each experiment. Decay of the
DOPPH absorption at 519 nm was monitored over time to
obtain kexptl results, and from the linear plots of kexptl versus
concentrations, the second-order rate constants were calcu-
lated.
(b) Au toxid a tion /In h ibition P r oced u r es. Autoxidations
were carried out at 30 °C under 760 Torr oxygen in a dual
channel oxygen uptake apparatus equipped with a sensitive
pressure transducer, as described previously.6,41 The styrene
used was separated from a commercial inhibitor by rapid bulb-
to-bulb distillation on the vacuum line and passed through a
short column of alumina just before use. As emphasized
earlier,6 it is necessary to ensure that there is an appreciable
kinetic chain length during the inhibition periods.
Ack n ow led gm en t. This research was supported by
a Strategic Projects Grant from the Natural Science and
Engineering Research Council of Canada. We thank the
College of Natural Science of Carleton University, Bill
J ack for the use of their computational resources, and
Dr. G. A. DiLabio for helpful discussions.
1
13
144°, lit.38 mp 139-140°. The H and C NMR spectra were
consistent with the literature.38 The mass spectrum gave a
parent mass m/e 160 which was the base peak.
The reduction of juglone was readily carried out in either
ethyl acetate or ether by rapidly stirring a solution of it with
an aqueous solution of sodium hydrosulfite under argon, and
the triol 1,4,8-naphthalenetriol was isolated as described
previously.40 Compound 6 was synthesized from the triol in
another laboratory.18 Details will be reported separately.
Kin et ic Met h od s. (a ) R ea ct ion s of P h en ols w it h
DOP P H•. The stopped-flow experiments were carried out
using DOPPH• on a spectrometer with a xenon 150 W arc light
source. In a typical procedure, a solution of DOPPH• (∼5 ×
10-5 M) in hexane was mixed 1:1 with hexane solutions of the
Su p p or tin g In for m a tion Ava ila ble: A table, S1, of
∆BDEs and rate constants of phenols with DPPH•, DOPPH•,
and ROO• used in Figure 1 and Cartesian coordinates for
structures of compounds 1-8 and their radicals optimized with
B3LYP/6-31G(d). This material is available free of charge via
the Internet at http://pubs.acs.org.
J O020184V
(39) Poirier, M.; Simard, M.; Wuest, J . D. Organometallics 1996,
15, 1296-1300.
(41) Wayner, D. D. M.; Burton, G. W. Handbook of Free Radicals
and Antioxidants in Biomedicine; CRC Press: Boca Raton, FL, 1989;
Vol III, pp 223-232.
(40) Malesani, G.; Ferlin, M. G. J . Heterocycl. Chem. 1987, 24, 513-
517.
5196 J . Org. Chem., Vol. 67, No. 15, 2002