Regenerable chain-breaking 2,3-dihydrobenzo[b]selenophene-5-ol antioxidants

Article abstract of DOI:10.1021/jo0700023

A series of 2,3-dihydrobenzo[b]selenophene-5-ol antioxidants was prepared by subjecting suitably substituted allyl 4-methoxyphenyl selenides to microwave-induced seleno-Claisen rearrangement/intramolecular Markovnikov hydroselenation followed by boron tri

Full text of DOI:10.1021/jo0700023

Regenerable Chain-Breaking 2,3-Dihydrobenzo[b]selenophene-5-ol
Antioxidants
Sangit Kumar,† Henrik Johansson,† Lars Engman,*,† Luca Valgimigli,‡ Riccardo Amorati,‡
Maria Grazia Fumo,‡ and Gian Franco Pedulli‡
Department of Biochemistry and Organic Chemistry, Uppsala UniVersity,
Box 576, SE-751 23 Uppsala, Sweden, and Dipartimento di Chimica Organica “A. Mangini”,
UniVersita` di Bologna, Via S. Giacomo 11, 40126 Bologna, Italy
lars.engman@biorg.uu.se
ReceiVed January 2, 2007
A series of 2,3-dihydrobenzo[b]selenophene-5-ol antioxidants was prepared by subjecting suitably
substituted allyl 4-methoxyphenyl selenides to microwave-induced seleno-Claisen rearrangement/
intramolecular Markovnikov hydroselenation followed by boron tribromide-induced O-demethylation.
The novel antioxidants were assayed for their capacity to inhibit azo-initiated peroxidation of linoleic
acid in a water/chlorobenzene two-phase system containing N-acetylcysteine as a thiol reducing agent in
the aqueous phase. Antioxidant efficiency as determined by the inhibited rate of peroxidation, Rinh, increased
with increasing methyl substitution (Rinh ) 46-26 µM/h), but none of the compounds could match
R-tocopherol (Rinh ) 22 µM/h). Regenerability as determined by the inhibition time, Tinh, in the presence
of the thiol regenerating agent decreased with increasing methyl substitution. Thus, under conditions
where the unsubstituted compound 5a inhibited peroxidation for more than 320 min, R-tocopherol worked
for 90 min and the trimethylated antioxidant 5g for 60 min only. Sampling of the aqueous phase at
intervals during peroxidation using antioxidant 5a showed that N-acetylcysteine was continuously oxidized
with time to the corresponding disulfide. In the absence of the regenerating agent, compounds 5 inhibited
peroxidation for 50-60 min only. A (RO)B3LYP/LANL2DZdp//B3LYP/LANL2DZ model was used
for the calculation of homolytic O-H bond dissociation enthalpies (BDE) and adiabatic ionization
potentials (IP) of phenolic antioxidants 5. Both BDE (80.6-76.3 kcal/mol) and IP (163.2-156.0 kcal/
mol) decrease with increasing methyl substitution. The phenoxyl radical corresponding to phenol 5g
gave an intense ESR signal centered at g ) 2.0099. The H-O bond dissociation enthalpy of the phenol
was determined by a radical equilibration method using BHA as an equilibration partner. The observed
BDE (77.6 ( 0.5 kcal/mol) is in reasonable agreement with calculations (76.3 kcal/mol). As judged by
calculated log P values, the lipophilicity of compounds 5 increased slightly when methyl groups were
introduced into the phenolic moiety (2.9 > C log P < 4.2). The capacity of compounds 5a (kinh ) 3.8
× 105 M-1 s-1) and 5g (kinh ) 1.5 × 106 M-1 s-1) to inhibit azo-initiated autoxidation of styrene in the
homogeneous phase (chlorobenzene) was also studied. More efficient regeneration at the lipid-aqueous
interphase is the most likely explanation why the intrinsically poorest antioxidant 5a can outperform its
analogues as well as R-TOC in the two-phase system. Possible mechanisms of regeneration are discussed
and evaluated.
10.1021/jo0700023 CCC: $37.00 © 2007 American Chemical Society
Published on Web 03/03/2007
J. Org. Chem. 2007, 72, 2583-2595
2583

Kumar et al.
SCHEME 1. Antioxidant Mechanisms
shows different ways of initiation of the autoxidation process.
Preventive antioxidants keep initiation of new chains to a
minimum either by destroying hydroperoxides (PD ) peroxide
decomposer) by absorbing ultraviolet light (UVA ) UV
absorbers) or by complexing transition metals which would
otherwise give rise to radicals via Fenton-like chemistry of
hydroperoxides (MD ) metal deactivator).
In vivo, peroxide decomposition is brought about by enzy-
matic systems: Catalase containing ferric heme groups catalyze
decomposition of hydrogen peroxide into oxygen and water. A
family of closely related selenium-containing glutathione per-
oxidases (GPxs) catalyzes the reduction of hydrogen peroxide
and organic hydroperoxides into water and alcohols, respec-
tively.7 The stoichiometric reducing agent in these processes is
the tripeptide glutathione, GSH (γ-glutamyl cysteinyl glycine)
which is oxidized to the corresponding disulfide, GSSG (eq 1).8
Introduction
All organic materials exposed to air undergo oxidative
degradation. Reducing the rate of such processes by utilizing
low concentrations of “antioxidants” is of paramount importance
for aerobic organisms as well as for producers of most kinds of
commercial products. In medicine, the term “oxidative stress”
has been introduced to describe a situation characterized by an
elevation in the cellular steady-state concentration of reactive
oxygen-derived species. This condition occurs if the balance
between oxidants and various antioxidant defenses is impaired.
The recent interest in pharmaceutical antioxidants1 and “anti-
oxidant pharmacotherapy”2 has emerged as a remedy for
pathological conditions characterized by oxidative stress (chronic
inflammatory disorders, atherosclerosis, cataract, etc.). Food
technologists use antioxidants to inhibit lipid peroxidation and,
thus, rancidity in food materials.3 Polymer chemists use anti-
oxidants to control polymerization in the manufacture of rubber,
plastics, and paint and for the stabilization of polymeric materials
during processing and use.4 The oil industry makes extensive
use of antioxidants in the design of better automobile fuels and
lubricating oils.5 Provided toxicity/environmental problems
associated with increasing antioxidant use can be coped with,
disease prevention/extending the lifetime of various materials
is obviously of significant benefit to man and society.
Mechanistically, the chemical processes responsible for
oxidative damage to biological material and the deterioration
of the mechanical properties of a polymer during processing
and use are very similar (Scheme 16). The catalytic cycle in
the upper right of the scheme is a free radical chain process
known as autoxidation. If not intercepted, this will in short time
convert organic material (RH) to the corresponding alkylhy-
droperoxide (ROOH). Chain-breaking accepting antioxidants
(CB-A) trap carbon-centered radicals in competition with
dioxygen. Chain-breaking donating antioxidants (CB-D) donate
hydrogen atoms to peroxyl radicals before they can propagate
the peroxidation reaction.6 The cycle to the left in Scheme 1
ROOH/HOOH + 2 GSH GPxs8 ROH/H2O + GSSG + H2O
(1)
With drug and other antioxidant applications in mind, some of
us9 and others10 have developed simple organochalcogen
compounds that could mimic the properties of the GPxs and
catalyze the decomposition of hydroperoxides in the presence
of suitable stoichiometric reductants.
Vitamin E is the most important lipophilic, chain-breaking
antioxidant in vivo.11 Thus, it serves to protect DNA, proteins,
lipids, carbohydrates, and other important biomolecules from
becoming oxidatively damaged by radical processes involving
reactive oxygen and nitrogen species. R-Tocopherol (R-TOH),
the most reactive component of vitamin E, is known to trap
two peroxyl radicals before it is converted into nonradical
products. It would therefore seem wise for Nature to provide a
system for its regeneration and catalytic mode of action,
especially within biological membranes where it is the only
protective agent. Although it has been hard to establish
conclusively that this regeneration process occurs in vivo,12 it
is generally accepted13 that ascorbate as a co-antioxidant (CoAH)
could donate a hydrogen atom to the R-tocopheryl radical,
(7) (a) Spallholz, J. E.; Boylan, L. M. In Peroxidases in Chemistry and
Biology; Everse, J., Everse, K. E., Grisham, M. B., Eds.; CRC Press: Boca
Raton, FL, 1991; Vol. 1, Chapter 12. (b) Sunde, R. A. In Handbook of
Nutritionally Essential Mineral Elements; O’Dell, B., Sunde, R. A., Eds.;
Marcel Dekker: New York, 1997; Chapter 18. (c) Birringer, M.; Pilawa,
S.; Flohe´, L. Nat. Prod. Rep. 2002, 19, 693-718.
(8) See: Prabhakar, R.; Vreven, T.; Morokuma, K.; Musaev, D. G.
Biochemistry 2005, 44, 11864-11871 and refs cited therein.
(9) (a) Engman, L.; Stern, D.; Cotgreave, I. A.; Andersson, C. M. J.
Am. Chem. Soc. 1992, 114, 9737-9743. (b) Andersson, C.-M.; Hallberg,
A.; Brattsand, R.; Cotgreave, I. A.; Engman, L.; Persson, J. Bioorg. Med.
Chem. Lett. 1993, 3, 2553-2558. (c) Engman, L.; Stern, D.; Pelcman, M.
Andersson, C.-M. J. Org. Chem. 1994, 59, 1973-1979. (d) Kanda, T.;
Engman, L.; Cotgreave, I. A.; Powis, G. J. Org. Chem. 1999, 64, 8161-
8169. (e) McNaughton, M.; Engman, L.; Birmingham, A.; Powis, G.,
Cotgreave, I. A. J. Med. Chem. 2004, 47, 233-239.
† Uppsala University.
‡ Universita` di Bologna.
(1) Andersson, C. M.; Hallberg, A.; Ho¨gberg, T. AdV. Drug Res. 1996,
28, 65-180.
(2) (a) Rice-Evans, C. A.; Diplock, A. T. Free Radical Biol. Med. 1993,
15, 77-96. (b) Bast, A. Drug News Persp. 1994, 7, 465.
(3) Pokorny, J.; Yanishlieva, N.; Gordon, M. Antioxidants in Foods;
Woodhead Publishing: Cambridge, 2001.
(4) Zweifel, H. Stabilization of Polymeric Materials; Springer: Berlin,
1997.
(5) Mortier, R. M., Orszulik, S. T., Eds. Chemistry and Technology of
Lubricants; Blackie Academic & Professional: London 1997.
(6) Scott, G. In Free Radicals: Chemistry, Pathology and Medicine;
Rice-Evans, C., Dormandy, T., Eds; Richelieu Press: London, 1988;
p 103.
(10) (a) Mugesh, G.; Singh, H. B. Chem. Soc. ReV. 2000, 29, 347-357.
(b) Mugesh, G.; du Mont, W.-W.; Sies, H. Chem. ReV. 2001, 101, 2125-
2179.
(11) Burton, G. W.; Ingold, K. U. Acc. Chem. Res. 1986, 19, 194-201.
(12) Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and
Medicine, 3rd ed.; Oxford University Press: Oxford, 1999; p 215.
(13) (a) Kamal-Eldin, A.; Appelquist, L.-Å. Lipids 1996, 31, 671-701.
See refs 266-286. See also: (b) Sato, K.; Niki, E.; Shimasaki, H. Arch.
Biochem. Biophys. 1990, 279, 402-405. (c) Bisby, R.; Parker, A. W. Arch.
Biochem. Biophys. 1995, 317, 170-178. (d) Noguchi, N.; Niki, E. Free
Radical Res. 1998, 28, 561-572. (e) Niki, E.; Noguchi, N. Acc. Chem.
Res. 2004, 37, 45-51.
2584 J. Org. Chem., Vol. 72, No. 7, 2007

2,3-Dihydrobenzo[b]selenophene-5-ol Antioxidants
R-TO•, which results from the interaction of R-TOH with
peroxyl radicals (eqs 2 and 3).
ROO• + R-TOH f ROOH + R-TO•
R-TO• + CoAH f R-TOH + CoA•
(2)
(3)
Niki and co-workers showed that vitamin E, although it is
intrinsically a better scavenger of peroxyl radicals, is spared by
vitamin C when both antioxidants were present in the homo-
geneous phase.14 Also, when a lipid-soluble radical initiator was
used in liposomes, both Niki and Ingold found that ascorbate
in the aqueous phase was able to spare lipid-soluble R-TOH in
the lipid phase.15 R-Tocopherol has also been shown to catalyze
the reduction of peroxyl radicals (eq 4)
FIGURE 1. Two-phase model used for studying regeneration of chain-
breaking antioxidants.
properties of synthetic and natural compounds.23 Some time ago,
we designed a two-phase variant of this system which allows
the study of antioxidant regeneration by water-soluble co-
antioxidants.24 In the experimental setup used (schematically
shown in Figure 1), linoleic acid and the antioxidant to be
evaluated were vigorously stirred in chlorobenzene at 42 °C
with an aqueous solution of N-acetylcysteine (NAC). 2,2′-
Azobis(2,4-dimethylvaleronitrile) (AMVN) was added as an
initiator in the organic phase and the progress of peroxidation
monitored by HPLC (conjugated diene hydroperoxide forma-
tion). For comparison of catalyst efficiency, the inhibited rate
of peroxidation, Rinh, was determined by least-squares methods
from absorbance/time plots. The progress of peroxidation was
followed for 320 min and the duration of the inhibited phase,
Tinh, determined graphically as the cross-point for the inhibited
and the uninhibited lines. We previously reported25 that, whether
(Tinh ) 90 min) or not (Tinh ) 80 min) NAC (1 mM) was present
in the aqueous phase, R-tocopherol (40 µM) inhibited peroxi-
dation of linoleic acid for almost the same time in the lipid
phase. This is in line with Barclay’s22 results from liposome
studies which indicated that water-soluble glutathione could not
regenerate R-TOH. In the same study, we compared the
antioxidant profile of 2,3-dihydrobenzo[b]furan-5-ol (1a) and
its 1-thio (1b), 1-seleno (1c), and 1-telluro (1d) analogues.
Interestingly, the inhibition time for all compounds was
significantly prolonged if NAC was present in the aqueous
phase. The effect of the thiol reducing agent was particularly
noteworthy with the selenium analogue 1c (Tinh ) 50 and >300
min, respectively, in the presence and absence of NAC).
ROO• + CoAH R-TOH8 ROOH + CoA•
(4)
in the presence of lipid-soluble co-antioxidants such as
ubiquinols,16,17 R-tocopheryl hydroquinone,17 flavonoids,18 phe-
nothiazines,19 catechols,19 and phenols20 with sufficiently low
O-H bond dissociation energies. A kinetic model for the
recycling of R-TOH by co-antioxidants in homogeneous solution
has been proposed by Pedulli and co-workers.19,20
Other water-soluble co-antioxidants than ascorbate have also
been considered for the regeneration of R-TOH across the lipid
aqueous interphase. Although both urate and glutathione are
often present in much higher concentrations than ascorbate in
human plasma, it seems they are incapable21,22 of regenerating
R-tocopherol from tocopheryl radicals produced in a lipid
environment.
In the present study, we have systematically investigated the
regeneration of a series of 2,3-dihydrobenzo[b]selenophene-5-
ol derivatives by water-soluble N-acetylcysteine in a two-phase
model system for lipid peroxidation. As it turns out, regeneration
and catalytic performance is largely dictated by the number and
position of methyl substituents contained in the phenolic ring
of the antioxidant.
Results
Azo-initiated peroxidation of linoleic acid or derivatives
thereof has frequently been used for studying the antioxidative
(14) Niki, E.; Saito, T.; Kawakami, A.; Kamiya, Y. J. Biol. Chem. 1984,
259, 4177-4182.
(15) (a) Niki, E.; Kawakami, A.; Yamamoto, Y.; Kamiya, Y. Bull. Chem.
Soc. Jpn. 1985, 58, 1971-1975. (b) Doba, T.; Burton, G. W.; Ingold, K.
U. Biochem. Biophys. Acta 1985, 835, 298-303.
(16) Frei, B.; Kim, M. C.; Ames, B. N. Proc. Natl. Acad. Sci. U.S.A.
1990, 87, 4879-4883.
(17) Shi, H.; Noguchi, N.; Niki, E. Free Radical Biol. Med. 1999, 27,
334-346.
Encouraged by this result, but still largely ignorant about the
parameters relevant for catalyst regeneration, we recently
embarked on a total synthesis of all-rac-R-selenotocopherol (2),
the selenium analogue of TOH.26 However, we were rather
(18) (a) Jia, Z.-S.; Zhou, B.; Yang, L.; Wu, L.-M.; Liu, Z.-L. J. Chem.
Soc., Perkin Trans. 2 1998, 911-915. (b) Zhou, B.; Jia, Z.-S.; Chen, Z.-
H.; Yang, L.; Wu, L.-M.; Liu, Z.-L. J. Chem. Soc., Perkin Trans. 2 2000,
785-791. (c) Chen, Z.-H.; Zhou, B.; Yang, L.; Wu, L.-M.; Liu, Z.-L. J.
Chem. Soc., Perkin Trans. 2 2001, 1835-1839. (d) Pedrielli, P.; Skibsted,
L. H. J. Agric. Food Chem. 2002, 50, 7138-7144.
(23) (a) Cosgrove, J. P.; Church, D. F.; Pryor, W. A. Lipids 1987, 22,
299-304. (b) Braughler, J. M.; Pregenzer, J. F. Free Radical Biol. Med.
1989, 7, 125-130. (c) Pryor, W. A.; Cornicelli, J. A.; Devall, L. J.; Tait,
B.; Trivedi, B. K.; Witiak, D. T.; Wu, M. J. Org. Chem. 1993, 58, 3521-
3532.
(24) Vessman, K.; Ekstro¨m, M.; Berglund, M.; Andersson, C.-M.;
Engman, L. J. Org. Chem. 1995, 60, 4461-4467.
(19) Amorati, R.; Ferroni, F.; Lucarini, M.; Pedulli, G. F.; Valgimigli,
L. J. Org. Chem. 2002, 67, 9295-9303.
(20) Amorati, R.; Ferroni, F.; Pedulli, G. F.; Valgimigli, L. J. Org. Chem.
2003, 68, 9654-9658.
(21) Niki, E.; Saito, M.; Yoshikawa, Y.; Yamamoto, Y.; Kamiya, Y.
Bull. Chem. Soc. Jpn. 1986, 59, 471-477.
(22) (a) Barclay, L. R. C. J. Biol. Chem. 1988, 263, 16138-16142. (b)
However, there is esr evidence that glutathione may regenerate R-TOH from
the R-tocopheryl radical, R-TO•:22a Niki, E.; Tsuchiya, J.; Tanimura, R.;
Kamiya, Y. Chem. Lett. 1982, 789-792.
(25) Malmstro¨m, J.; Jonsson, M.; Cotgreave, I. A.; Hammarstro¨m, L.;
Sjo¨din, M.; Engman, L. J. Am. Chem. Soc. 2001, 123, 3434-3440.
(26) Shanks, D.; Amorati, R.; Fumo, M. G.; Pedulli, G. F.; Valgimigli,
L.; Engman, L. J. Org. Chem. 2006, 71, 1033-1038.
J. Org. Chem, Vol. 72, No. 7, 2007 2585

Kumar et al.
SCHEME 2
antioxidants inhibit peroxidation for considerably longer times
in the presence of the thiol reducing agent in the aqueous phase
(regeneration conditions). The inhibition time in the presence
of NAC is highly dependent on the number and position of
methyl groups contained in the aromatic moiety of the antioxi-
dant. The best regenerability is found with the unsubstituted
disappointed to find that compound 2, just like R-TOH, was
not regenerable in our two-phase model for lipid peroxidation
(Tinh ) 58 min whether NAC was present or not).
compound 5a (Tinh > 320 min) and fluoro derivative 6 (Tinh )
300 min). Antioxidants 5b and 5c carrying one methyl group
in positions 6 and 7, respectively, showed markedly poorer
catalytic performance (Tinh ) 230 and 190 min). Regenerability
continued to decrease when two methyl groups were introduced
into the phenolic ring (compounds 5d, 5e, and 5f). The effect
was particularly noteworthy in the case of 4,6-disubstitution (Tinh
) 80 min for compound 5d). The fully ring-methylated
antioxidant 5g, just like R-TOH, could not be regenerated by
NAC under the experimental conditions used. The peroxidation
traces recorded with the most regenerable antioxidant 5a and
R-TOH are shown in Figure 2. The peroxyl radical quenching
ability of the antioxidant is reflected in the inhibited rate of
peroxidation. Although R-TOH (Rinh ) 22 µM/h in the absence
of thiol reducing agent) is intrinsically a better chain-breaking
antioxidant than any of the synthetic compounds, it is clearly
outperformed by compound 5a (Rinh ) 46 µM/h in the absence
of thiol) after ca. 100 min. In fact, antioxidant 5a is the least
efficient peroxyl radical quencher in the series of new orga-
noseleniums prepared. As could be expected,28 introduction of
methyl groups (especially into the ortho- and para-positions)
of phenolic antioxidants would serve to increase the antioxidant
activity. This trend is reflected in the observed inhibited rates
of peroxidation for compounds 5 in the absence of NAC. Thus,
compound 5b carrying a meta-methyl group is only slightly
more efficient than compound 5a. Compounds 5b (ortho-
methyl) and 5e and 5f (ortho,meta-dimethyl) are clearly more
efficient, and compound 5g (ortho, ortho,meta-trimethyl) is the
most efficient peroxyl radical quencher in the series. The trend
seen in the inhibition data without NAC in the aqueous phase
is largely reflected in the data obtained under regeneration
conditions. It might be noted that all the antioxidants tested
appear to be slightly more efficient when evaluated in the
absence of thiol. This could be due to differences in pH (neutral
in purely aqueous solution as compared with pH ca. 3.8 in a 1
mM aqueous solution of NAC) or ionic strength in the aqueous
phase. More likely, however, this is due to different kinetic
pathways of inhibition in the presence and absence of the co-
antioxidant.29
Synthesis. In order to carry out a more systematic investiga-
tion of catalyst regenerability, we needed ready access to
selenochromanol or 2,3-dihydrobenzo[b]selenophene-5-ol de-
rivatives substituted in the aromatic moiety with a variable
number of methyl groups. It occurred to us that the seleno-
Claisen rearrangement reported by Stefani and co-workers27
some time ago could provide the desired organoselenium
antioxidants with relatively little effort (Scheme 2). The allyl
aryl selenides 3 required for this approach were prepared from
commercially available 4-bromoanisoles via conversion to a
Grignard reagent, insertion of elemental selenium into the
carbon-magnesium bond, air oxidation, borohydride reduction
of the crude resulting diaryl diselenide, and, finally, arenese-
lenolate allylation. Compounds 3 were isolated in yields ranging
from 32 to 96%. According to the literature procedure for
seleno-Claisen rearrangement,27 2,3-dihydrobenzo[b]selenophenes
unsubstituted in the aromatic moiety were obtained in low yields
by heating for 2-3 h with quinoline at 210 °C. Obviously, the
2-allylbenzeneselenol product of the seleno-Claisen rearrange-
ment undergoes intramolecular hydroselenation in a Markovni-
kov fashion under the harsh reaction conditions. For the
preparation of compounds 4, we found it optimal to perform
the reaction in a microwave reactor at 220-230 °C for ca. 45
min. Still, isolated yields were only modest (13-59%). The
higher reaction temperature was crucial for obtaining the
sterically encumbered compounds 4 carrying methyl groups in
either or both positions 4 and 7 of the dihydrobenzo[b]-
selenophene. O-Demethylation to give the desired phenolic
compounds 5 proceeded in good yields (47-98%) by treatment
with boron tribromide in dichloromethane. Compounds prepared
carry none (5a), one (5b, 5c), two (5d, 5e, 5f), or three (5g)
methyl substituents in the aromatic moiety (for structures, see
Table 1). In addition, mostly in analogy with the chemistry
shown in Scheme 2, we have also prepared dihydrobenzo[b]-
selenophene-5-ol derivatives carrying a fluorine in the aromatic
ring (compound 6) and a few compounds (7, 8, and 9) where
the substitution in the 2-position of the dihydrobenzo[b]-
selenophene system has been varied.
Inhibition times for organoselenium antioxidants 5 and 6 in
the absence of NAC in the aqueous phase were all shorter (50-
70 min) than that recorded for R-TOH (80 min). R-TOH has a
capacity to trap two peroxyl radicals before it is converted into
nonradical products (the stoichiometric factor n equals 2). Thus,
the stoichiometric factor for most of the synthetic antioxidants
(5a, 5b, 5e, 5f, 5g) is only 1.25. Only fluoro derivative 6 showed
a significantly higher value (n ) 1.75).
Inhibition Studies. Inhibition studies with newly prepared
antioxidants were performed in the model system described in
Figure 1. The inhibition time, Tinh, and the inhibited rate of
peroxidation, Rinh, in the presence/absence of NAC in the
aqueous phase are shown in Table 1 for compounds studied.
The results with R-tocopherol are included for comparison.
There are several interesting trends in the data: Some of the
Compound 7, unsubstituted in the phenolic ring just like
compound 5a, but carrying an additional butyl group in position
(28) (a) Lucarini, M.; Pedrielli, P.; Pedulli, G. F. J. Org. Chem. 1996,
61, 9259-9263. (b) Wright, J. S.; Johnson, E. R.; DiLabio, G. A. J. Am.
Chem. Soc. 2001, 123, 1173-1183.
(29) In the absence of thiol and under conditions where n ) 2, the slope
of the inhibited period is proportional to 1/2kinh, unlike when regeneration
takes place. In this case, the phenoxyl radical is recycled before trapping
the second peroxyl radical, and the slope of the inhibited trace is proportional
(27) Stefani, H. A.; Petragnani, N.; Ascenso, M. F. C.; Zeni, G. Synth.
Commun. 2003, 33, 2161-2166.
to 1/kinh
.
2586 J. Org. Chem., Vol. 72, No. 7, 2007

2,3-Dihydrobenzo[b]selenophene-5-ol Antioxidants
TABLE 1. Inhibition Times (Tinh) and Inhibited Rates of Peroxidation (Rinh) in the Presence and Absence of 1 mM N-Acetylcysteine in the
Aqueous Phase, Calculated O-H Bond Dissociation Enthalpies, Ionization Potentials, and Lipophilicities for Compounds 5-9 and r-Tocopherol
a Rate of peroxidation during the inhibited phase (uninhibited rate ) ca. 650 µM/h). b Duration of the inhibited phase of peroxidation. Reactions were
monitored for e320 min. c Calculated O-H bond dissociation enthalpies. d Calculated ionization potential. e Calculated log P values.
2, inhibited peroxidation for a similarly long time (Table 1; Tinh
> 320 min) under regeneration conditions. In the absence of
NAC, the compound showed similar antioxidant efficiency (Rinh
) 45 µM/h) but slightly longer inhibition time (Tinh ) 70 min).
As compared with antioxidant 5g, variation in the 2-substitution
(compounds 8 and 9) did not affect the inhibition characteristics
much (Tinh ) 80 and 60 min, respectively, with and without
NAC for both compounds; Rinh ) 30 and 28 µM/h, respectively,
without NAC in the aqueous phase).
only chemical process occurring in the aqueous phase during
peroxidation is oxidation of NAC to the corresponding disulfide.
In the evaluation of the most regenerable antioxidant 5a, the
aqueous phase was sampled at intervals and the thiol/disulfide
concentrations were monitored by reversed phase HPLC. As
shown in Figure 3, the concentration of NAC drops more or
less linearly with time and can be extrapolated to reach zero
after ca. 390 min. Inhibition times were also recorded in the
presence of less (0.5-0.0625 mM) than the standard (1 mM)
concentration of NAC. As shown in Figure 4, Tinh decreases in
a more or less linear fashion to the limiting value under non-
regenerating conditions as the NAC concentration is lowered.
The consumption of thiol in the aqueous phase was not
routinely monitored with time for experiments carried out in
the two-phase system. However, we have confirmed that the
J. Org. Chem, Vol. 72, No. 7, 2007 2587

Kumar et al.
TABLE 2. Rate Constants, kinh, for the Reaction of Compounds 2,
5a, and 5g with Peroxyl Radicals in Styrene/Chlorobenzene at 303
K, Number of Radicals Trapped by Each Antioxidant Molecule, n,
and Experimental Bond Dissociation Enthalpies, BDEOH, Measured
by Radical Equilibration EPR in Benzene at 298 Ka
kinh
BDEOH
(kcal/mol)
compound
(M-1s-1
)
n
2b
5a
5g
BHA
R-TOH
1.2 × 106
1.9
2.0
1.7
2c
78.1 ( 0.3
81.6d
77.6 ( 0.5
77.2 ( 0.2e
77.15 ( 0.3e
3.8 ( 0.4 × 105
1.5 ( 0.3 × 106
1.2 × 105c
3.2 × 106c
2c
a Data obtained under identical experimental settings with reference
antioxidants BHA (2,6-di-tert-butyl-4-methoxyphenol), and R-tocopherol
are shown for comparison. b Data from ref 26. c Reference values from ref
20 re-measured here to calibrate the instrumental settings. d Estimated value.
e Data from ref 28a, corrected for the revised BDE of 2,4,6-tri-tert-
butylphenol.31
FIGURE 2. Peroxidation traces (linoleic acid hydroperoxide concen-
tration vs time) recorded using compound 5a or R-TOH as antioxidants
in the chlorobenzene layer in the presence of NAC (1 mM) in the
aqueous phase.
based on a Validyne DP15 differential pressure transducer,
which has been previously described.30 The observed kinetics
are in accord with eqs 5-10. Both compounds 5a and 5g
showed in styrene a neat inhibition period, Tinh, whose length
provided the stoichiometric factor, which is the number of
peroxyl radicals trapped by each molecule of antioxidant. This
is given by n ) RiTinh/[AH], where Ri is the initiation rate,
measured in a preliminary set of experiments using R-TOH,
and [AH] is the initial concentration of antioxidant.
Ri
9
initiator
8 R•
(5)
(6)
R• + O2 f ROO•
kp
9
ROO• + RH
8 ROOH + R•
(7)
(8)
2kt
ROO• + ROO• 98 products
FIGURE 3. Concentration of N-acetylcysteine in the aqueous phase
with time during a normal peroxidation experiment using antioxidant
5a.
kinh
ROO• + ArOH 8 ROOH + ArO•
(9)
ROO• + ArO• f products
(10)
Integration of the oxygen consumption trace affords the rate of
reaction with peroxyl radicals, kinh, provided kp is known (eq
11).
∆[O2]t ) -kp/kinh[RH]ln(1 - t/Tinh)
(11)
As can be seen from the inhibition data in Table 2, compound
5a is much less reactive toward peroxyl radicals than the fully
methylated analogue 5g. The stoichiometric factor n equals the
theoretical value for a phenol. The peroxyl radical quenching
ability of compound 5g is very similar to that previously reported
for selenotocopherol 2. This is not surprising if the small
difference in BDEOH (vide infra) is considered not significant.
However, the stoichiometric factor obtained experimentally for
a freshly prepared solution of compound 5g (n ) 1.7) was lower
than that recorded for the more tocopherol-like analogue 2 (n
) 1.9). Furthermore, if the solution of 5g was left for some
time at room temperature before the peroxidation experiment,
the recorded stoichiometric factor was even lower (n ) 1.3-
FIGURE 4. Inhibition time recorded with compound 5a in the presence
of various amounts of N-acetylcysteine (1, 0.5, 0.25, 0.125, 0.0625,
and 0 mM) in the aqueous phase.
Inhibition studies with two of the novel antioxidants (5a and
5g) and R-TOC as a reference have also been performed in the
homogeneous phase (inhibited autoxidation of styrene in chlo-
robenzene at 303 K with AIBN initiation). The autoxidation
was followed by monitoring the oxygen consumption in an
oxygen uptake apparatus built in one of our laboratories and
(30) 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.
2588 J. Org. Chem., Vol. 72, No. 7, 2007

2,3-Dihydrobenzo[b]selenophene-5-ol Antioxidants
1.4 after 4 h). This lower value of n possibly indicates that the
antioxidant is partly degraded by air in solution.
peroxidation for compounds 5 showed a good correlation both
with calculated bond dissociation enthalpies (R2 ) 0.87) and
ionization potentials (R2 ) 0.87).
In order to evaluate the possible synergism between com-
pounds 5a and 5g with thiol under homogeneous phase
conditions, we have studied the inhibited autoxidation of styrene
(4.3 M in chlorobenzene) under identical experimental condi-
tions as described above using a fixed concentration of 6.25 ×
10-6 M of 5a or 5g and variable amounts of 1-octylmercaptan.
The use of styrene at such concentration cancelled out the minor
antioxidant activity of the mercaptan up to 1 mM. Therefore,
we could safely use the co-antioxidant in concentrations up to
160 times those of the organoselenium. Interestingly, as
compared with compound 5a or 5g alone (no thiol added), no
difference in the oxygen consumption traces was ever observed
with any concentration of co-antioxidant used. This shows that
1-octylmercaptan cannot regenerate compounds 5a and 5g from
the corresponding phenoxyl radicals under homogeneous phase
conditions.
Calculations. Recently, the introduction of easily applicable
DFT models for the prediction of relative bond dissociation
energies (BDEs) and ionization potentials (IPs) by Wright and
co-workers28b has greatly facilitated the design of novel anti-
oxidants. However, these DFT methods are not directly ap-
plicable to antioxidants containing heavy elements such as
selenium. This is because relativistic effects upon the core
electrons become too great to ignore for these atoms, precluding
the use of the standard all-electron basis set these models are
based upon. Therefore, we recently developed DFT models
closely resembling those by Wright but utilizing a basis set
incorporating relativistic effective core potentials for the core
electrons.35 These inherently take care of the relativism problem
and drastically reduce the number of electrons to be calculated
upon and, thus, the cost of calculation. This (RO)B3LYP/
LANL2DZdp//B3LYP/LANL2DZ model was used for the
calculation of homolytic OH bond dissociation enthalpies and
adiabatic ionization potentials of phenolic antioxidants 5 and
6, and the results are shown in Table 1. Not unexpectedly, both
BDE and IP for compounds 5 decrease with increasing methyl
substitution as one traverses the table from top to bottom (BDE
) 80.8 kcal/mol and IP ) 163.2. kcal/mol for compound 5a;
BDE ) 76.3 kcal/mol and IP ) 156.0 kcal/mol for compound
5g). Introduction of fluorine into the 4-position of the dihy-
drobenzo[b]selenophene-5-ol system (compound 6) also has a
weakening effect on the OH bond (in line with the predicted
effects of ortho-chlorine substitution28b in phenolic compounds).
On the other hand, the electron-withdrawing fluorine caused a
significant increase in the ionization potential of compound 6
(IP ) 169.5 kcal/mol). Interestingly, the inhibited rates of
Depending on the number and position of aromatic methyl
groups, the organoselenium phenolic antioxidants 5-9 are more
or less readily regenerated at the aqueous lipid interphase. We
therefore thought it would be interesting to see how structural
modifications in the series affected lipophilicities of the
compounds. Shown in Table 1 are calculated log P values. As
expected, lipophilicity in the series of compounds 5 increases
steadily as one, two, or three aromatic methyl groups are
introduced. The introduction of additional alkyl groups in the
2-position of the dihydrobenzo[b]selenophene moiety results in
a further increase in lipophilicity (compounds 7-9).
EPR and Experimental Bond Dissociation Enthalpy Data.
Out of the various methods available for determination of
phenolic O-H bond strengths, the electron paramagnetic
resonance (EPR) equilibration technique presently seems to
afford the most accurate values.28a,32 In this method, the
equilibrium constant, Ke, for the hydrogen atom transfer reaction
between a reference phenol (Ar′OH) and the compound under
investigation (ArOH) is measured (eq 12). In order to check
the accuracy of our H-O BDE calculations, we tried to obtain
experimental values for the two extremes of compounds 5 (5a
and 5g).
ArOH + Ar′O• a ArO• + Ar′OH
(12)
Whereas compound 5a produced only a weak signal when
reacted at room temperature inside the cavity of an EPR
spectrometer with alkoxyl radicals generated photolytically from
di-tert-butyl peroxide in deoxygenated benzene, compound 5g
produced an intense spectrum centered at g ) 2.0099 (referenced
to the DPPH radical36 g ) 2.00364). This spectrum is shown in
Figure 5 together with a computer simulation and was inter-
preted in terms of the following hyperfine splitting constants
of 4.61 G (3H), 6.11 G (3H), 1.25 G (3H), 1.00 G (2H), and
1.60 G (1H). For obtaining the H-O BDE of compound 5g,
butylated hydroxyanisole (BHA; BDE ) 77.2 kcal/mol in
benzene31) was used as a reference. In the calculation of Ke,
the initial concentrations of 5g and BHA were used, while the
relative radical concentrations were determined by means of
EPR spectroscopy. The BDE for the species ArOH was
calculated, in the assumption that the entropic term can be
neglected,28a by means of eq 13 from Ke and the BDE value of
the reference phenol. From these measurements, repeated under
different light intensities in order to check the constancy of Ke,
a BDE value of 77.6 ( 0.5 kcal/mol was obtained, a value
reasonably well reproduced by calculations (1.3 kcal/mol
higher).
(31) All the BDE values determined in benzene solution by means of
the EPR radical equilibration technique,32 based on the O-H BDE of 2,4,6-
tri-tert-butylphenol determined many years earlier by Mahoney et al.33 using
calorimetric measurements, must be downscaled by 1.1 kcal/mol due to
the revision of the heat of formation of (E)-azobenzene.34
(32) (a) Brigati, G.; Lucarini, M.; Mugnaini, V.; Pedulli, G. F. J. Org.
Chem. 2002, 67, 4828-4832. (b) Pratt, A. D.; DiLabio, G. A.; Valgimigli,
L.; Pedulli, G. F.; Ingold, K. U. J. Am. Chem. Soc. 2002, 124, 11085-
11092. (c) Valgimigli, L.; Brigati, G.; Pedulli, G. F.; DiLabio, G.;
Mastragostino, M.; Arbizzani, C.; Pratt, D. A. Chem.sEur. J. 2003, 9,
4997-5010.
(33) Mahoney, L. R.; Ferris, F. C.; DaRooge, M. A. J. Am. Chem. Soc.
1969, 91, 3883-3889.
(34) Mulder, P.; Korth, H.-G.; Pratt, D. A.; DiLabio, G. A.; Valgimigli,
L.; Pedulli, G. F.; Ingold, K. U. J. Phys. Chem. A 2005, 109, 2647-2655.
(35) Shanks, D.; Frisell, H.; Ottosson, H.; Engman, L. Org. Biomol.
Chem. 2006, 4, 846-852.
BDE(ArO-H) = BDE(Ar′O-H) - RT ln(Ke) (13)
Due to the short persistence of the phenoxyl radical correspond-
ing to phenol 5a (which would introduce a large error in the
radical equilibration measurement), the BDEOH value was
instead obtained from the measured value for 5g by considering
the additive contribution of ring substituents.28a,32 Indeed, two
ortho-methyls are known to lower the BDEOH of a phenol by
3.5 kcal/mol, while a methyl in the meta-position contributes
(36) Valgimigli, L.; Ingold, K. U.; Lusztyk, J. J. Org. Chem. 1996, 61,
7947-7950.
J. Org. Chem, Vol. 72, No. 7, 2007 2589

Kumar et al.
FIGURE 5. Experimental (A) and computer-simulated (B) spectra obtained by irradiating compound 5g in benzene containing 10% v/v di-tert-
butyl peroxide at 298 K.
-0.5 kcal/mol. Thus, the resulting estimated BDEOH of com-
pound 5a is 81.6 kcal/mol, which is in good agreement with
the calculated value of 80.8 kcal/mol (Table 1). Measured and
estimated BDEOH data are shown in Table 2.
transfer,37 or proton tunneling,38 the process is expected to be
facilitated by increasing methyl substitution. This is what we
observe experimentally. More efficient regeneration at the lipid-
aqueous interphase is the most likely explanation why the
intrinsically poorest antioxidant 5a can outperform its analogues
as well as R-TOC in the two-phase system. Some speculation
as to the mechanisms of antioxidant action and regeneration
are presented in the following: The synthetic precursor of the
efficient antioxidant 5a, the O-methylated compound 4a, did
not show any antioxidant activity when evaluated in the two-
phase model.
Discussion
Nature is using thiols for regeneration of some of its most
efficient antioxidant enzymes. Our interest in regenerable chain-
breaking organoseleniums stems from the observation that
organoselenium antioxidant 1c was active for a much longer
time than the corresponding oxygen (1a) and sulfur (1b)
analogues in the presence of a water-soluble thiol when
evaluated in the two-phase model for lipid peroxidation.
However, since neither R-TOC nor selenotocopherol (2) are
regenerable by thiol, replacement of oxygen for selenium in
general is not enough, at least in the chromanol series, for
imposing regenerability on the catalyst. The microwave-assisted
seleno-Claisen rearrangement offered a facile route to a variety
of dihydrobenzo[b]selenophene-5-ol derivatives 5 where the
number and position of methyl groups in the aromatic moiety
could be varied. Among compounds prepared and evaluated,
the unsubstituted derivative 5a was clearly the more easily
regenerable, whereas the fully methylated compound 5g acted
only on a slightly more than stoichiometric basis (n ) 1.25).
Introduction of methyl groups is likely to both weaken the
phenolic OH bond and lower the oxidation potential. Thus,
whether the peroxyl radical quenching by these antioxidants
involves hydrogen atom transfer, proton-coupled electron
This suggests that the phenolic organoselenium antioxidants are
transformed into the corresponding phenoxyl radicals by one
of the mechanisms mentioned above. A mechanism involving
electron transfer only, followed by disproportionation of the
resulting (selenium-centered) radical cation into selenium(II)/
(IV) compounds, followed by thiol-mediated reduction of the
tetravalent species (selenoxide), appears unlikely considering
the lower calculated (by 3.8 kcal/mol) oxidation potential of
compound 4a as compared with that of 5a.
(37) Costentin, C.; Robert, M.; Save´ant, J.-M. J. Am. Chem. Soc. 2006,
128, 4552-4553.
(38) Nagaoka, S.; Inoue, M.; Nishioka, C.; Nishioku, Y.; Tsunoda, S.;
Ohguchi, C.; Ohara, K.; Mukai, K.; Nagashima, U. J. Phys. Chem. B 2000,
104, 856-862.
2590 J. Org. Chem., Vol. 72, No. 7, 2007

2,3-Dihydrobenzo[b]selenophene-5-ol Antioxidants
If regeneration is occurring in the aqueous phase or at the
aqueous lipid interphase, one may argue that regenerability
would simply depend on the relative lipophilicities of the
antioxidants investigated. Admittedly, as shown in Table 1,
compounds 5 become more lipid soluble as the number of
aromatic methyl groups increase. However, since the two most
readily regenerable antioxidants 5a (C log P ) 2.9) and 7 (C
log P ) 5.0) show a more than 2 orders of magnitude difference
in water solubility, lipophilicity cannot explain the observed
differences in regenerability. Rather, it seems that the presence/
absence of methyl groups flanking the phenolic moiety has a
much stronger influence on regenerability. In support of this
view, we found that the selenochromanol derivative 10, prepared
by us some time ago,39 showed a significantly better regener-
ability (Tinh ) 150 min in the presence and Tinh ) 70 min in
the absence of N-acetylcysteine) than selenotocopherol (2).
In the absence of N-acetylcysteine, organoselenium antioxi-
dants 1c, 2, and 5-10 inhibit peroxidation for shorter times
than R-tocopherol in the two-phase model for lipid peroxidation.
Thus, the stoichiometric factors for all these compounds are
lower than the theoretical value n ) 2 expected for chain-
breaking phenolic antioxidants. The reason for this may be that
a fraction of the antioxidant is converted to an inactive form,
probably a selenoxide. The small amount of linoleic acid
hydroperoxide which is always present in the commercially
available linoleic acid is likely to convert the organoseleniums
to the corresponding selenoxides before and during the experi-
ment. Also, peroxyl radicals may transfer an oxygen atom to
the organoseleniums to produce an alkoxyl radical and sele-
noxide. Such oxygen transfer processes have previously
been observed with organotelluriums40 as well as other orga-
nochalcogen compounds. Since the alkoxyl radical is also
chain-carrying, this reaction will not affect the rate of per-
oxidation of the substrate. Under regeneration conditions, the
two selenoxide forming reactions will be unproblematic. As
indicated in eq 14, selenoxides are likely to be rapidly reduced
by aqueous thiol to give the active selenide form of the
antioxidant.
As one traverses group 16 of the periodic table from top to
bottom, the chalcogens become more readily oxidized. This
trend is also seen when these elements are incorporated into
organic molecules. Thus, the oxidation potential of compound
1c, as determined by cyclic voltammetry, was 0.22 V lower
than that recorded for the sulfur and oxygen analogues.25 Sulfur
and selenium can easily expand their valence shells to accom-
modate 10 or 12 electrons. Tetravalent sulfur and selenium
derivatives such as sulfoxides and selenoxides are therefore
perfectly stable compounds. In contrast to sulfoxides, selenox-
ides are readily reduced to the divalent state by a variety of
mild reducing agents such as thiols and ascorbate. It is therefore
tempting to try to involve a selenoxide in the mechanism
responsible for regeneration of compound 5a. Such a species
(11) could form (eq 14) if the resonance-stabilized phenoxyl
radical is further oxidized and the resulting selenonium ion reacts
with water with loss of a proton. Regeneration is then induced
by thiol and accompanied by disulfide formation. Although the
mechanism proposed in eq 14 nicely explains why selenium
compounds would be expected to be superior to their sulfur
and oxygen counterparts when it comes to regeneration, it does
not seem to explain the trends seen in Table 1.
The simplest process for regeneration of antioxidants 5 would
involve hydrogen atom transfer from thiol contained in the
aqueous phase to phenoxyl radicals formed in the chlorobenzene
layer. The decrease in regenerability on increasing methyl
substitution, especially in the 4- and 6-positions, could then be
due to steric hindrance around the phenoxyl radical. This
mechanism is likely to be operative only if the bond dissociation
enthalpy of the thiol is lower or at least comparable to that of
the phenolic O-H bond and if the rate of H-atom transfer is
significant.20 Experimental S-H bond dissociation enthalpy data
are not abundant in the literature, and the methodology used in
calculating them seems to be critical for obtaining reliable
results.41 Although we have not been able to find a useful value
for N-acetylcysteine, it seems reasonable to assume that the S-H
BDE would be in the range of 85-90 kcal/mol.41,42 The
predicted value for cysteine in proteins is 87.8 ( 2.4 kcal/mol.43
Thus, considering that the H-O BDEs of compounds 5 are in
the range of 76-81 kcal/mol, NAC is not likely to act as a
regenerating agent by donation of a thiol-derived hydrogen atom.
However, due to captodative stabilization of the resulting
R
carbon-centered radical, the C-H bond of NAC is probably
weaker than the S-H bond. The corresponding BDE for cysteine
in proteins was estimated to 82.7 ( 2.4 kcal/mol.43 The
calculated H-O BDEs of compounds 2a-d are all very similar
(81.3, 80.5, 80.3, and 80.5, respectively25). Yet, the organose-
lenium compound 2c was found to be considerably more
regenerable than its analogues in the two-phase model. This
also suggests that regeneration occurs by some other mechanism
than hydrogen atom transfer to a phenoxyl radical.
In the light of these thermodynamic considerations, the failure
of 1-octylmercaptan to regenerate compounds 5a and 5g from
their corresponding phenoxyl radicals under homogeneous phase
conditions is not surprising. The S-H BDE of 1-octylmercaptan
Thus, an increasing number of electron-donating alkyl groups
in the aromatic moiety is likely to facilitate one-electron
oxidation of the phenoxyl radical. Also, since compound 5d
shows such poor regenerability, steric hindrance around the
selenoxide (which could be a reason why this mechanism does
not come into play with 7-substituted derivatives) does not seem
to be an important issue.
(40) Engman, L.; Persson, J.; Mere´nyi, G.; Lind, J. Organometallics 1995,
14, 3641-3648.
(41) Cabral do Couto, P.; Costa Cabral, B. J.; Martinho Simoes, J. A.
Chem. Phys. Lett. 2006, 421, 504-507.
(42) Escoubet, S.; Gastaldi, S.; Vanthuyne, N.; Gil, G.; Siri, D.; Bertrand,
M. P. Eur. J. Org. Chem. 2006, 3242-3250.
(39) Al-Maharik, N.; Engman, L.; Malmstro¨m, J.; Schiesser, C. H. J.
Org. Chem. 2001, 66, 6286-6290.
(43) Rauk, A.; Yu, D.; Armstrong, D. A. J. Am. Chem. Soc. 1998, 120,
8848-8855.
J. Org. Chem, Vol. 72, No. 7, 2007 2591

Kumar et al.
SCHEME 3. Reactions Involved in the Antioxidant Activity
of Compounds 5 in the Two-Phase System, Including (Box)
the Regeneration of Phenoxyl Radical at the
Organic-Aqueous Interfacea
can be considered similar to that of ethanethiol for which the
values of 88.1 and 88.7 can be found in the literature.44,45 These
values are ∼7 and ∼ 11 kcal/mol, respectively, higher than the
H-O BDE of compounds 5a and 5g, making regeneration by
hydrogen atom transfer practically impossible. However, in
water or very polar solvents such as acetonitrile or in two-phase
systems, regeneration by thiols may still be possible if it occurs
by electron transfer rather than hydrogen atom transfer. To get
a rough idea about the feasibility of such a process (eq 15),
one could compare the standard reduction potential of the thiol
in water with those for the redox couple ArO•/ArO-.
2 RSH + 2 ArO• f RSSR + 2 ArOH
(15)
The reduction potential for NAC has recently been reported to
be slightly more positive (∆E° ) + 63 mV) than that for
glutathione,46 whose standard potential in water at 25 °C and
pH 7 is known as E° ) -240 mV versus NHE.47,48 This
corresponds to E° ) -145 mV versus NHE at pH 3.848 (which
is the pH in the aqueous phase under our experimental
conditions). This redox potential for the thiol/disulfide couple
should be matched with the potential for the couple ArO•/ArO-,
which for most phenoxyl radicals fall in the range 0.1 < E° <
1.4 V versus NHE, depending on the substituents in the aromatic
ring.49Although standard reduction potentials for aryloxyl
radicals corresponding to phenols 5-9 are not available, a value
for the closely related compound 1c is known as E° ) +0.49
V (vs NHE at pH 1225) and, considering the reported pKa )
9.9,25 it is expected to increase by about 0.36 V at pH 3.8.50 E°
is also expected to vary linearly with Hammett σ+ parameter51
of ring substituents.25,49 A quantitative estimation of the EMF
(∆E) for the cell arising from eq 15, according to Nernst’s
equation at 25 °C (eq 16), would clearly require several
parameters which are presently not known (e.g., the actual
concentration of transient species at the aqueous lipid interface
to be inserted in the reaction quotient Q). However, on a
qualitative basis, regeneration of all compounds 5 from their
corresponding phenoxyl radicals in the aqueous phase by
electron transfer from NAC is expected to be thermodynamically
favored.
a Stoichiometric factors n (the number of peroxyl radicals that each
antioxidant molecule can destroy) are indicated when appropriate.
and other processes as summarized in Scheme 3. Considering
that the purported electron transfer is occurring at the lipid
aqueous interphase, it may be essential that phenoxyl radicals
are sterically unhindered for the process to be efficient.
Experimental Section
Typical Procedure for the Preparation of Allyl Aryl Selenides.
Allyl 4-Methoxyphenyl Selenide (3a). To a dry 100 mL three-
necked flask containing Mg turnings (720 mg, 30 mmol) and
anhydrous tetrahydrofuran (50 mL) was added 4-bromoanisole (3.9
mL, 30 mmol) dropwise with stirring. The reaction mixture was
heated at reflux, and stirring continued until all magnesium had
disappeared. Selenium powder (2.40 g, 30 mmol) was then added
in portions over a period of 30 min. After another 2 h, the resulting
solution of magnesium areneselenolate was poured into a beaker
and kept in the open air overnight. Diaryl diselenide formed was
extracted with Et2O (4 × 25 mL). The ether solution was washed
with water, and the organic layer was separated, dried over Na2-
SO4, and evaporated under reduced pressure to give a brown solid.
To this crude product in EtOH (30 mL) was added NaBH4 (0.945
g, 25 mmol) at 0 °C under N2. The reaction mixture was warmed
to room temperature and stirred for 30 min. During this process,
the yellow color of the reaction mixture had faded away. Allyl
bromide (2.11 mL, 25 mmol) was added to the cooled (0 °C)
reaction mixture, and stirring continued for an additional 30 min
at room temperature. After aqueous workup, ether extraction (3 ×
20 mL), drying over Na2SO4, and evaporation, column chroma-
tography using pentane/CH2Cl2 (85:15) as an eluent afforded the
pure title compound (4.85 g, 71%): 1H NMR δ 7.48 (d, J ) 8.8
Hz, 2H), 6.83 (d, J ) 8.8 Hz, 2H), 5.92 (m, 1H), 4.90 (t, J ) 15.1
Hz, 2H), 3.80 (s, 3H), 3.44 (d, J ) 7.6 Hz, 2H); 13C NMR δ 136.4,
134.8, 132.3, 116.6, 115.8, 114.7, 55.3, 31.8.
∆E ) ∆E° - (59.1 mV/2) log Q
(16)
The success of the regeneration process would also depend on
the rate of electron transfer as compared with those of other
reactions that the phenoxyl radical could undergo (such as
trapping of a second peroxyl radical). Overall, the successful
regeneration of phenolic antioxidants such as compound 5a is
due to the favorable interplay and balance between several redox
The following compounds were analogously prepared starting
from the appropriately substituted 4-bromoanisole.
(44) Orlov, Y. D.; Lebedev, Y. A. Bull. Acad Sci. USSR DiV. Chem.
Sci. (Engl. Transl.) 1984, 33, 1227-1230.
Allyl 4-Methoxy-3-methylphenyl Selenide (3b): Yield 32%;
1H NMR δ 7.31-7.36 (several peaks, 2H), 6.73 (d, J ) 8.0 Hz,
1H), 5.93 (m, 1H), 4.92-4.88 (several peaks, 2H), 3.82 (s, 3H),
3.43 (dm, J ) 8.0 Hz, 2H), 2.19 (s, 3H); 13C NMR δ 157.8, 137.2,
134.9, 133.6, 127.6, 119.3, 116.5, 110.6, 55.4, 31.8, 16.2.
Allyl 4-Methoxy-2-methylphenyl Selenide (3c): Yield 96%; 1H
NMR δ 7.46 (dm, J ) 8.5 Hz, 1H), 6.80 (d, J ) 2.8 Hz, 1H), 6.66
(dm, J ) 8.5 Hz, 1H), 5.93 (m, 1H), 4.94-4.89 (several peaks,
2H), 3.79 (s, 3H), 3.41 (d, J ) 7.6 Hz, 2H), 2.47 (s, 3H); 13C NMR
δ 159.7, 142.8, 136.8, 134.7, 121.1, 116.6, 115.8, 112.0, 55.3, 30.8,
23.4.
(45) Janousek, B. K.; Reed, K. J.; Brauman, J. I. J. Am. Chem. Soc.
1980, 102, 3125-3129.
(46) Nosza´l, B.; Visky, D.; Kraszni, M. J. Med. Chem. 2000, 43, 2176-
2182.
(47) Scott, E. M.; Duncan, I. W.; Ekstrand, V. J. Biol. Chem. 1963, 238,
3928-3933.
(48) (a) Schafer, F. Q.; Buettner, G. R. Free Radical Biol. Med. 2001,
30, 1191-1212. (b) Shen, D.; Dalton, T. P.; Nebert, D. W.; Shertzer, H.
G. J. Biol. Chem. 2005, 280, 25305-25312.
(49) (a) Jonsson, M.; Lind, J.; Mere´nyi, G. J. Phys. Chem. A 2002, 106,
4758-4762. (b) Lind, J.; Shen, X.; Eriksen, I. E.; Mere´nyi, G. J. Am. Chem.
Soc. 1990, 112, 479-482. (c) Armstrong, D. A.; Sun, Q.; Schuler, R. H. J.
Phys. Chem. 1996, 100, 9892-9899.
(50) Li, C.; Hoffman, M. Z. J. Phys. Chem. B 1999, 103, 6653-6656.
(51) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165-195.
Allyl 3,5-Dimethyl-4-methoxyphenyl Selenide (3d): Yield 76%;
1H NMR δ 7.20 (s, 2H), 5.97 (m, 1H), 4.98 (m, 2H), 3.73 (s, 3H),
2592 J. Org. Chem., Vol. 72, No. 7, 2007

2,3-Dihydrobenzo[b]selenophene-5-ol Antioxidants
3.50 (d, J ) 7.4 Hz, 2H), 2.28 (s, 6H); 13C NMR δ 156.7, 134.7,
134.3, 131.7, 123.9, 116.7, 59.8, 31.1, 16.0.
was then treated with HCl (300 mL of 4 M aqueous solution) and
the organic phase extracted into diethyl ether (3 × 30 mL). After
drying (Na2SO4), evaporation, and purification by column chro-
matography using pentane/CH2Cl2 (80:25) as an eluent, the pure
title compound was isolated as a light yellow oil (2.74 g, 59%):
1H NMR δ 7.20 (d, J ) 8.4 Hz, 2H), 6.80 (d, J ) 2.6 Hz, 1H),
6.71 (dd, J ) 2.6, 8.4 Hz, 1H), 4.17 (m, 1H), 3.78 (s, 3H), 3.41
(dd, J ) 6.8, 15.2 Hz, 1H), 3.03 (dd, J ) 6.6, 15.2 Hz, 1H), 1.61
(d, J ) 6.8 Hz, 3H); 13C NMR δ 158.0, 143.9, 137.3, 126.3, 113.0,
111.7, 55.3, 47.4, 41.5, 22.6.
Allyl 2,3-Dimethyl-4-methoxyphenyl Selenide (3e): Yield 62%;
1H NMR δ 7.45 (d, J ) 8.6 Hz, 1H), 6.68 (d, J ) 8.7 Hz, 1H),
5.97 (m, 1H), 5.00-4.93 (several peaks, 2H), 3.84 (s, 3H), 3.42
(d, J ) 7.4 Hz, 2H), 2.51 (s, 3H), 2.24 (s, 3H); 13C NMR δ 157.5,
140.7, 134.7, 133.7, 125.8, 121.8, 116.4, 108.4, 55.5, 31.0, 20.1,
12.7.
Allyl 2,5-Dimethyl-4-methoxyphenyl Selenide (3f): Yield 65%;
1H NMR δ 7.34 (s, 1H), 6.72 (s, 1H), 5.95 (m, 1H), 4.99 (several
peaks, 2H), 3.83 (s, 3H), 3.43 (d, J ) 7.6, 2H), 2.49 (s, 3H), 2.20
(s, 3H). 13C NMR δ 157.7, 139.9, 137.6, 134.8, 124.6, 120.2, 116.4,
111.7, 55.3, 30.8, 23.1, 15.6.
The following compounds were analogously prepared starting
from the appropriately substituted allyl phenyl selenides.
2,6-Dimethyl-5-methoxy-2,3-dihydrobenzo[b]selenophene (4b):
1
Yield 35%; H NMR δ 7.05 (s, 1H), 6.70 (s, 1H), 4.15 (m, 1H),
Allyl 4-Methoxy-2,3,5-trimethylphenyl Selenide (3g): Yield
1
3.76 (s, 3H), 3.40 (dd, J ) 6.6, 15.2 Hz, 1H), 3.00 (dd, J ) 6.8,
15.2 Hz, 1H), 2.16 (s, 3H), 1.59 (d, J ) 6.8 Hz, 3H); 13C NMR δ
156.2, 140.9, 127.8, 126.5, 126.3, 108.2, 55.7, 47.7, 41.7, 22.8,
16.2.
70%; H NMR δ 7.39 (s, 1H), 5.98 (m, 1H), 4.95-5.07 (several
peaks, 2H), 3.71 (s, 3H), 3.47 (d, J ) 7.6 Hz, 2H), 2.42 (s, 3H),
2.28 (s, 3H), 2.27 (s, 3H); 13C NMR δ 156.6, 138.0, 134.6, 134.0,
130.3, 128.5, 125.8, 116.8, 60.0, 30.6, 19.8, 16.0, 13.4.
2,7-Dimethyl-5-methoxy-2,3-dihydrobenzo[b]selenophene (4c):
Yield 13%; 1H NMR δ 6.22 (d, J ) 2.5 Hz, 1H), 6.56 (d, J ) 2.5
Hz, 1H), 4.11 (m, 1H), 3.78 (s, 3H), 3.43 (dd, J ) 6.8, 15.4 Hz,
1H), 3.03 (dd, J ) 6.8, 15.4 Hz, 1H), 2.22 (s, 3H), 1.58 (d, J ) 6.8
Hz, 3H); 13C NMR δ 158.5, 143.4, 135.5, 128.6, 113.6, 109.1, 55.6,
47.9, 40.7, 23.2, 22.9.
Allyl 3-Fluoro-4-hydroxyphenyl Selenide. To a solution of
3-fluoro-4-hydroxybromobenzene (0.91 mL, 8.33 mmol) in THF
(20 mL) was added t-BuLi (15.9 mL of a 1.7 M solution in pentane,
27 mmol) dropwise at -78 °C under N2. The resulting yellowish
reaction mixture was stirred for 15 min when selenium powder
(2.16 g, 27 mmol) was added in two portions under a brisk flow of
N2. After another 30 min at -78 °C, the reaction was brought to
room temperature and stirred for 2 h. All selenium had reacted to
give a brownish-red reaction mixture which was poured over ice
(25 g) and kept for 1 h before addition of aqueous HCl (4 M).
After ether extraction (4 × 30 mL), drying over Na2SO4, and
concentration in vacuo, a yellowish liquid (2.25 g) was obtained.
Crystallization from pentane/ether gave almost pure, light yellow
crystals (1.457 g) of bis(2-fluoro-4-hydroxyphenyl)diselenide. This
diselenide (1.14 g) was dissolved in EtOH (30 mL), and NaBH4
(230 mg, 6 mmol) was added under N2 at 0 °C. After additional
stirring for 40 min at this temperature and for 30 min at room
temperature, allyl bromide (0.58 mL, 6.7 mmol) was added
dropwise. After stirring at 0 °C for 5 min and at room temperature
for 1 h, workup including ether extraction (3 × 15 mL), drying
over Na2SO4, and concentration in vacuo gave a colorless liquid
which was purified by column chromatography using pentane/ethyl
acetate (95:5) as an eluent. The yield of the title compound was
787 mg (57%): 1H NMR δ 7.26 (ddd, J ) 0.9, 2.0, 10.2 Hz, 1H),
7.20 (dm, J ) 8.3 Hz, 1H), 6.91 (td, J ) 8.3 Hz, 0.9, 1H), 5.90
(m, 1H), 5.41 (br s, 1H), 4.91 (m, 2H), 3.46 (dm, J ) 7.6 Hz, 2H);
13C NMR δ 150.6 (d, JC-F ) 240.5 Hz), 143.5 (d, JC-F ) 14.3
Hz), 134.4, 131.7 (d, JC-F ) 3.5 Hz), 121.9 (d, JC-F ) 18.2 Hz),
119.9 (d, JC-F ) 5.6 Hz), 117.8 (d, JC-F ) 2.0 Hz) 117.1, 31.9.
3-Methyl-2-propen-1-yl 4-Methoxy-2,3,5-trimethylphenyl Se-
lenide (E/Z ca. 3/1) was prepared according to the typical procedure
using (E/Z)-3-methyl-2-propen-1-yl bromide instead of allyl bro-
mide: Yield 97%; 1H NMR δ 7.34, 7.30 (s, 1H), 5.73-5.45 (m, 2
H), 3.77 (s, 3H), 3.58, 3.51 (d, J ) 7.6 Hz, 2H), 2.50, 2.47 (s, 3H),
2.34 (s, 3H), 2.32 (s, 3H), 1.72, 1.62 (d, J ) 6.3 Hz, 3 H); 13C
NMR δ 156.5, 156.4, 137.9, 137.8, 134.1, 134.0, 130.0, 128.3,
128.3, 128.0, 127.0, 126.7, 126.2, 126.1, 126.0, 59.8, 30.0, 24.1,
19.7, 19.6, 17.7, 15.8, 15.8, 13.2, 12.3.
5-Methoxy-2,4,6-trimethyl-2,3-dihydrobenzo[b]selenophene
1
(4d): Yield 46%; H NMR δ 6.98 (s, 1H), 4.18 (m, 1H), 3.70 (s,
3H), 3.38 (dd, J ) 7.0, 15.3 Hz, 1H), 3.00 (dd, J ) 6.6, 15.3 Hz,
1H), 2.26 (s, 3H), 2.24 (s, 3H), 1.63 (d, J ) 6.8 Hz, 3H); 13C NMR
δ 154.8, 140.4, 131.0, 130.0, 127.9, 125.1, 60.0, 45.8, 40.6, 23.1,
16.2, 13.8.
5-Methoxy-2,6,7-trimethyl-2,3-dihydrobenzo[b]selenophene (4e):
1
Yield 39%; H NMR δ 6.63 (s, 1H), 4.09 (m, 1H), 3.78 (s, 3H),
3.45 (dd, J ) 6.9, 15.2 Hz, 1H), 3.05 (dd, J ) 6.5, 15.2 Hz, 1H),
2.19 (s, 3H), 2.13 (s, 3H), 1.58 (d, J ) 6.8 Hz, 3H); 13C NMR δ
156.3, 139.2, 134.0, 129.1, 124.0, 106.2, 56.0, 48.2, 40.3, 22.9,
21.2, 12.1.
5-Methoxy-2,4,7-trimethyl-2,3-dihydrobenzo[b]selenophene (4f):
1
Yield 26%; H NMR δ 6.54 (s, 1H), 4.12 (m, 1H), 3.78 (s, 3H),
3.46 (dd, J ) 7.4, 15.8 Hz, 1H), 3.01 (dd, J ) 6.9, 15.8 Hz, 1H),
2.22 (s, 3H), 2.14 (s, 3H), 1.60 (d, J ) 6.8 Hz, 3H); 13C NMR δ
156.5, 142.2, 131.9, 128.2, 121.1, 110.7, 56.0, 46.1, 39.6, 23.3,
23.2, 13.3.
5-Methoxy-2,4,6,7-tetramethyl-2,3-dihydrobenzo[b]sele-
nophene (4g): Yield 26%; 1H NMR δ 7.39 (s, 1H), 4.08 (m, 1H),
3.64 (s, 3H), 3.43 (dd, J ) 7.1, 15.5 Hz, 1H), 3.00 (dd, J ) 7.0,
15.5 Hz, 1H), 2.19 (s, 3H), 2.18 (s, 3H), 2.16 (s, 3H), 1.59 (d, J )
6.8 Hz, 1H); 13C NMR δ 155.2, 139.01, 133.2, 131.2, 128.2, 125.2,
60.3, 46.2, 39.2, 23.3, 21.1, 13.9, 12.8.
2-Ethyl-5-methoxy-4,6,7-trimethyl-2,3-dihydrobenzo[b]sele-
nophene: Yield 19%; 1H NMR δ 3.95 (m, 1H), 3.65 (s, 3H), 3.43
(dd, J ) 7.7, 15.4 Hz, 1H), 3.05 (dd, J ) 7.7, 15.4 Hz, 1H), 2.20
(s, 3H), 2.19 (s, 3H), 2.18 (s, 3H), 1.87 (m, 2H), 1.04 (t, J ) 6.6
Hz, 3H); 13C NMR δ 155.3, 139.6, 139.4, 128.2, 125.2, 114.6, 60.5,
48.1, 44.0, 30.6, 21.2, 14.2, 14.0, 12.9.
5-Methoxy-2,2,4,6,7-pentamethyl-2,3-dihydrobenzo[b]sele-
1
nophene: Yield 21%; H NMR δ 3.64 (s, 3H), 3.12 (s, 2H), 2.18
2-Methyl-2-propen-1-yl 4-Methoxy-2,3,5-trimethylphenyl Se-
lenide was prepared according to the typical procedure using
2-methyl-2-propen-1-yl bromide instead of allyl bromide: Yield
(s, 6H), 2.13 (s, 3H), 1.68 (s, 6H); 13C NMR δ 155.2, 139.1, 134.1,
131.3, 128.1, 125.4, 60.3, 53.3, 52.5, 31.7, 21.0, 13.9, 12.8.
Typical Procedure for O-Demethylation. 2-Methyl-2,3-dihy-
drobenzo[b]selenophene-5-ol (5a). To a cooled (-78 °C), stirred
solution of dihydrobenzo[b]selenophene 4a (1.14 g, 5 mmol) in
CH2Cl2 (10 mL) kept under N2 was added BBr3 (6 mL of a 1 M
solution in CH2Cl2; 6 mmol) dropwise. The mixture was then
allowed to warm to room temperature and stirred for another 20 h
when it was poured into water and extracted with additional CH2-
Cl2 (3 × 20 mL). The organic phase was washed with brine and
dried over Na2SO4, and the solvent was removed under reduced
pressure. Purification by column chromatography using pentane/
ethyl acetate (85:15) as an eluent afforded the title compound in
1
60%; H NMR δ 7.24 (s, 1H), 4.71 (several peaks, 2H), 3.68 (s,
3H), 3.43 (m, 2H), 2.39 (s, 3H), 2.24 (s, 3H), 2.23 (s, 3H), 1.88
(m, 3H); 13C NMR δ 156.6, 142.1, 138.2, 134.1, 130.2, 128.6,
126.4, 113.4, 60.1, 35.9, 21.7, 19.8, 16.0, 13.4.
Typical Procedure for Microwave-Assisted Seleno-Claisen
Rearrangement/Intramolecular Hydroselenation. 5-Methoxy-
2-methyl-2,3-dihydrobenzo[b]selenophene (4a). Freshly prepared
allylic selenide 3a (4.62 g, 20.3 mmol) in quinoline (10 mL) was
charged into a glass tube (20 mL), and the sealed vial was heated
in a microwave reactor for 1 h at 220 °C. The reaction mixture
J. Org. Chem, Vol. 72, No. 7, 2007 2593

Kumar et al.
1
the form of colorless crystals (1.00 g, 94%): mp 89-90 °C; H
2-Ethyl-4,6,7-trimethyl-2,3-dihydrobenzo[b]selenophene-5-
ol (8): Yield 98%; mp 121-123 °C; 1H NMR δ 4.44 (s, 1H), 3.92
(m, 1H), 3.44 (dd, J ) 7.8, 15.6 Hz, 1H), 3.07 (dd, J ) 7.8, 15.6
Hz, 1H), 2.18 (s, 3H), 2.17 (s, 3H), 2.15 (s, 3H), 1.86 (m, 2H),
1.03 (t, J ) 7.6 Hz, 3H); 13C NMR δ 150.3, 138.8, 130.7, 128.2,
120.9, 118.1, 47.8, 44.1, 30.5, 21.2, 14.2, 13.6, 12.4; 77Se NMR δ
363. Anal. Calcd for C12H16OSe: C, 58.01; H, 6.741. Found: C,
57.84; H, 6.67.
NMR δ 7.11 (d, J ) 8.2 Hz, 1H), 6.70 (d, J ) 2.6 Hz, 1H), 6.60
(dd, J ) 2.6, 8.2 Hz, 1H), 4.89 (s, 1H), 4.15 (m, 1H), 3.37 (dd, J
) 7.0, 15.4 Hz, 1H), 2.98 (dd, J ) 6.8, 15.4 Hz, 1H), 1.58 (d, J )
6.9 Hz, 3H); 13C NMR δ 153.8, 144.4, 127.2, 126.6, 114,7, 113.1,
47.4, 41.8, 22.7; 77Se NMR δ 415. Anal. Calcd for C9H10OSe: C,
50.74; H, 4.73. Found: C, 50.52; H, 4.69.
The following compounds were analogously prepared starting
from the appropriately substituted 5-methoxy-2,3-dihydrobenzo[b]-
selenophenes.
2,2,4,6,7-Pentamethyl-2,3-dihydrobenzo[b]selenophene-5-ol (9):
Yield 69%; mp 130-132 °C; H NMR δ 4.42 (s, 1H), 3.14 (s,
1
2H), 2.16 (s, 3H), 2.15 (s, 6H), 1.68 (s, 6H); 13C NMR δ 150.3,
138.7, 131.0, 129.9, 120.8, 118.5, 53.5, 52.3, 31.6, 21.2, 13.7, 12.4;
77Se NMR δ 489. MS m/z 271 (M + 1, 100), 256 (39), 175 (21).
The precursor of compound 7, 2-butyl-5-methoxy-2-methyl-2,3-
dihydrobenzo[b]selenophene, was prepared from 2-bromo-5-meth-
oxybenzyl bromide as shown below. The key steps in the synthesis
involve lithium tellurium exchange for benzyllithium generation
and homolytic intramolecular substitution at selenium for hetero-
cycle formation.
2,6-Dimethyl-2,3-dihydrobenzo[b]selenophene-5-ol (5b): Yield
65%; mp 106-107 °C; 1H NMR δ 7.02 (s, 1H), 6.64 (s, 1H), 4.42
(br s, 1H), 4.14 (m, 1H), 3.36 (dd, J ) 6.7, 15.3 Hz, 1H), 2.96 (dd,
J ) 6.7, 15.3 Hz, 1H), 2.19 (s, 3H), 1.57 (d, J ) 6.8 Hz, 3H); 13
C
NMR δ 151.9, 141.6, 127.9, 127.1, 123.2, 112.6, 47.2, 41.8, 22.8,
15.8; 77Se NMR δ 416. Anal. Calcd for C10H11OSe: C, 52.89; H,
5.32. Found: C, 52.59; H, 5.20.
2,7-Dimethyl-2,3-dihydrobenzo[b]selenophene-5-ol (5c): Yield
47%; mp 81-82 °C; 1H NMR δ 6.54 (s, 1H), 6.49 (s, 1H), 4.49 (s,
1H), 4.11 (m, 1H), 3.4 (dd, J ) 6.9, 15.4 Hz, 1H), 3.01 (dd, J )
6.9, 15.4 Hz, 1H), 2.19 (s, 3H), 1.59 (d, J ) 6.8 Hz, 3H); 13C
NMR δ 154.1, 143.7, 135.8, 128.6, 115.0, 110.3, 47.7, 40.7, 23.0,
22.9; 77Se NMR δ 390. Anal. Calcd for C10H11OSe: C, 52.87; H,
5.32. Found: C, 52.90; H, 5.21.
2,4,6-Trimethyl-2,3-dihydrobenzo[b]selenophene-5-ol (5d):
1
Yield 98%; mp 121-122 °C; H NMR δ 6.91 (s, 1H), 4.53 (s,
Reagents and conditions: (a) (i) n-BuTeLi, (ii) n-BuLi, -78 °C,
(iii) 2-hexanone, -115 °C; (b) (i) t-BuLi, -100 °C, (ii) Se, -78
°C, (iii) air oxidation, (iv) NaBH4, (v) BnBr; (c) (i) oxalyl chloride,
(ii) 2-mercaptopyridine-N-oxide Na salt/DMAP reflux.
1H), 4.15 (m, 1H), 3.39 (dd, J ) 7.2, 15.6 Hz, 1H), 2.98 (dd, J )
7.0, 15.6 Hz, 1H), 2.20 (s, 3H), 2.19 (s, 3H), 1.60 (d, J ) 6.9 Hz,
3H); 13C NMR δ 150.0, 140.2, 126.4, 124.8, 122.5, 121.1, 45.9,
40.5, 23.1, 16.1, 13.5; 77Se NMR δ 416. Anal. Calcd for C11H14-
OSe: C, 54.78; H, 5.85. Found: C, 54.92; H, 5.88.
1-(2-Bromo-5-methoxyphenyl)-2-methyl-2-hexanol. To a solu-
tion of n-BuLi (3.12 mL of 1.6 M solution in hexane, 5 mmol) in
THF (5 mL) was added tellurium powder (625 mg, 5 mmol) in
portions, and the mixture was stirred for 15 min at room temperature
under N2. To the resulting solution of lithium butanetellurolate was
added 2-bromo-5-methoxybenzylbromide (1.43 g, 5 mmol) in THF
(2 mL) dropwise and stirring continued for 10 min. The red solution
obtained was evaporated in vacuo, and dry Et2O (15 mL) was added
under N2. To this solution, cooled to -98 °C, was added n-BuLi
(3.25 mL 1.6 M, 5.2 mmol) dropwise. The resulting light yellow
solution was stirred for 20 min at -78 °C and then cooled to -115
°C. 2-Hexanone (1.23 mL, 10 mmol) was added to the reaction
mixture, and stirring continued for an additional 30 min at this
temperature and 30 min at room temperature. After aqueous workup,
extraction with CH2Cl2 (4 × 20 mL), drying (Na2SO4), evaporation,
and purification by column chromatography using pentane/ethy-
lacetate (9:1) as an eluent, the title compound was isolated as a
pale yellow liquid (1.27 g, 85%): 1H NMR δ 7.44 (d, J ) 8.8 Hz,
1H), 6.92 (d, J ) 3.1 Hz, 1H), 6.67 (dd, J ) 3.2, 8.8 Hz, 1H), 3.78
(s, 3H), 2.97 (d, J ) 13.6 Hz, 1H), 2.92 (d, J ) 13.6 Hz, 1H),
1.56-1.33 (several peaks, 7H), 1.17 (s, 3H), 0.92 (t, J ) 7.2 Hz,
3H); 13C NMR δ 158.6, 138.7, 133.5, 117.9, 116.8, 114.0, 73.7,
55.6, 46.8, 42.6, 26.4, 26.2, 23.4, 14.3.
Benzyl 2-(2-Hydroxy-2-methylhexyl)-4-methoxyphenyl Se-
lenide. To a solution of 1-(2-bromo-5-methoxyphenyl)-2-methyl-
2-hexanol (602 mg, 2 mmol) in THF (10 mL) was added t-BuLi
(3.75 mL of 1.6 M solution in pentane, 6 mmol) dropwise at -100
°C under N2. Stirring was continued for an additional 1 h at this
temperature and then 20 min at room temperature. After cooling
to -78 °C, selenium powder (480 mg, 6 mmol) was added, and
the reaction mixture was allowed to warm to room temperature.
After 5 h, the resulting dark red reaction mixture was poured onto
crushed ice and kept in the open air for 2 h. After aqueous workup,
extraction with CH2Cl2 (4 × 20 mL), drying (Na2SO4), and
evaporation, the crude mixture was purified by column chroma-
tography using pentane/ethyl acetate (9:2) as an eluent to give 402
mg of a diselenide/triselenide mixture. Without further characteriza-
tion, 450 mg of this product was dissolved in EtOH (20 mL), and
NaBH4 (378 mg, 10 mmol) was added at 0 °C under nitrogen. After
2,6,7-Trimethyl-2,3-dihydrobenzo[b]selenophene-5-ol (5e): Yield
92%; mp 133-134 °C; 1H NMR δ 6.52 (s, 1H), 4.66 (s, 1H), 4.07
(m, 1H), 3.38 (dd, J ) 7.1, 15.1 Hz, 1H), 2.98 (dd, J ) 6.7, 15.1
Hz, 1H), 2.19 (s, 3H), 2.14 (s, 3H), 1.58 (d, J ) 6.8 Hz, 3H); 13C
NMR δ 152.0, 139.7, 134.2, 129.2, 121.4, 110.3, 47.7, 40.4, 22.8,
21.2, 12.0; 77Se NMR δ 398. Anal. Calcd for C11H14OSe: C, 54.78;
H, 5.85. Found: C, 54.58; H, 5.72.
2,4,7-Trimethyl-2,3-dihydrobenzo[b]selenophene-5-ol (5f): Yield
81%; mp 122-124 °C; 1H NMR δ 6.46 (s, 1H), 4.58 (s, 1H), 4.12
(m, 1H), 3.40 (dd, J ) 7.0, 15.6 Hz, 1H), 3.00 (dd, J ) 6.9, 15.6
Hz, 1H), 2.16 (s, 6H), 1.60 (d, J ) 6.8 Hz, 3H); 13C NMR δ 152.2,
77
142.4, 132.4, 128.4, 118.7, 115.0, 46.2, 39.7, 23.3, 22.7, 13.2;
-
Se NMR δ 392. Anal. Calcd for C11H14OSe: C, 54.78; H, 5.85.
Found: C, 54.65; H, 5.92.
2,4,6,7-Tetramethyl-2,3-dihydrobenzo[b]selenophene-5-ol (5g):
1
Yield 78%; mp 124-125 °C; H NMR δ 4.44 (s, 1H), 4.06 (m,
1H), 3.44 (dd, J ) 6.9, 15.3 Hz, 1H), 3.02 (dd, J ) 6.9, 15.3 Hz,
1H), 2.18 (s, 3H), 2.17 (s, 3H), 2.15 (s, 3H), 1.59 (d, J ) 6.8 Hz,
3H); 13C NMR δ 150.4, 138.5, 130.9, 128.8, 121.0, 118.3, 46.5,
39.0, 23.3, 21.3, 13.7, 12.4; 77Se NMR δ 398. Anal. Calcd for
C11H14OSe: C, 56.49; H, 6.27. Found: C, 56.27; H, 6.14.
4-Fluoro-2-methyl-2,3-dihydrobenzo[b]selenophene-5-ol (6)
was prepared from allyl 3-fluoro-4-hydroxyphenyl selenide (247
mg, 1.07 mmol) according to the typical procedure for preparation
1
of compounds 4: Yield 18%; H NMR δ 6.90 (dm, J ) 8.2 Hz,
1H), 6.79 (tm, J ) 8.2 Hz, 1H), 4.88 (br s, 1H), 4.23 (m, 1H), 3.46
(dd, J ) 6.9, 15.7 Hz, 1H), 3.08 (dd, J ) 6.7, 15.7 Hz, 1H), 1.61
(d, J ) 6.9 Hz, 3H); MS m/z 232 (M+, 100), 150 (50). Anal. Calcd
for C9H9OFSe: C, 46.77; H, 3.92. Found: C, 46.55; H, 4.02.
2-Butyl-2-methyl-2,3-dihydrobenzo[b]selenophene-5-ol (7):
Yield 91%; 1H NMR δ 7.09 (d, J ) 8.2 Hz, 1H), 6.68 (d, J ) 2.6
Hz, 1H), 6.59 (dd, J ) 2.6, 8.2 Hz, 1H), 4.71 (s, 1H), 3.17 (d, J )
15.3 Hz, 1H), 3.01 (d, J ) 15.3 Hz, 1H), 1.92-1.79 (several peaks,
2H), 1.62 (s, 3H), 1.29-1.50 (several peaks, 4H), 0.92 (t, J ) 6.8
Hz, 3H); 13C NMR δ 153.6, 144.5, 127.8, 126.8, 114.6, 113.4, 61.2,
53.3, 43.2, 30.1, 28.5, 23.2, 14.2; 77Se NMR δ 483. Anal. Calcd
for C13H18OSe: C, 57.99; H, 6.74. Found: C, 58.11; H, 6.77.
2594 J. Org. Chem., Vol. 72, No. 7, 2007

2,3-Dihydrobenzo[b]selenophene-5-ol Antioxidants
30 min, benzyl bromide (1.72 g, 10 mmol) was added to the
resulting colorless reaction mixture and stirring continued for 1 h.
After aqueous workup, extraction with diethyl ether (3 × 15 mL),
drying (Na2SO4), and evaporation, the crude mixture was purified
by column chromatography using pentane/ethyl acetate (9:1) as an
eluent to give the title compound as a colorless oil (374 mg, 43%
over two steps): 1H NMR δ 7.49 (d, J ) 8.5 Hz, 1H), 7.25-7.13
(several peaks, 5H), 6.88 (d, J ) 2.9 Hz, 1H), 6.70 (dd, J ) 2.9,
8.5 Hz, 1H), 3.97 (s, 2H), 3.82 (s, 3H), 2.89 (s, 2H), 1.55-1.34
(several peaks, 7H), 1.11 (s, 3H), 0.95 (t, J ) 7.0 Hz, 3H); 13C
NMR δ 159.4, 142.7, 138.8, 137.6, 128.9, 128.4, 126.8, 123.5,
116.9, 113.0, 73.4, 55.4, 47.1, 42.7, 33.7, 26.4, 26.3, 23.4, 14.3.
2-Butyl-5-methoxy-2-methyl-2,3-dihydrobenzo[b]seleno-
phene. Benzyl 2-(2-hydroxy-2-methylhexyl)-4-methoxyphenyl se-
lenide (98 mg, 0.25 mmol) was stirred with oxalyl chloride (0.3
mL) in benzene (2 mL) for 18 h at ambient temperature under an
atmosphere of dry N2. After evaporation to dryness under reduced
pressure, the residue was taken into benzene (3 mL) and added
over 15 min to a well stirred suspension of the sodium salt of
2-mercaptopyridine N-oxide (53 mg, 0.35 mmol) and 4-N,N-
dimethylaminopyridine (3.6 mg, 0.03 mmol) in benzene (3 mL) at
reflux under N2. After 2.5 h, the reaction mixture was filtered
through a plug of Celite, evaporated, and the residue purified by
flash chromatography using pentane/CH2Cl2 (9:1) as an eluent to
give almost pure title compound (24 mg, 34%): 1H NMR δ 7.15
(d, J ) 8.3 Hz, 1H), 6.74 (d, J ) 2.6 Hz, 1H), 6.67 (dd, J ) 2.6,
8.3 Hz, 1H), 3.77 (s, 3H), 3.19 (d, J ) 15.2 Hz, 1H), 3.03 (d, J )
15.2 Hz, 1H), 1.98-1.74 (several peaks, 2H), 1.62 (s, 3H), 1.32-
1.54 (several peaks, 4H), 0.92 (m, 3H); 13C NMR δ 158.0, 144.2,
127.7, 126.7, 112.9, 112.2, 61.1, 55.6, 53.4, 43.3, 30.0, 28.5, 23.2,
14.2.
Initiation rates, Ri, were determined for each condition in prelimi-
nary experiments by the inhibitor method using R-tocopherol as
52
reference antioxidant: Ri ) 2[R-TOH]/Tinh
.
Calculations. Homolytic O-H bond dissociation enthalpies and
adiabatic ionization potentials of phenolic antioxidants 5 and 6 were
calculated as previously described.35 The C log P values for
compounds 5-9 were calculated by using software from BioByte
Corporation together with SYbYL, version 7.2.4.
EPR and Thermochemical Measurements. Deoxygenated
benzene solutions containing the phenol (0.01-0.001 M) and di-
tert-butyl peroxide (10% v/v) were sealed under nitrogen in a
suprasil quartz EPR tube. The sample was inserted at room
temperature in the cavity of an EPR spectrometer and photolyzed
with the unfiltered light from a 500 W high-pressure mercury lamp.
The temperature was controlled with a standard variable temperature
accessory and was monitored before and after each run with a
copper-constantan thermocouple. The EPR spectra were recorded
on a Bruker ESP 300 spectrometer equipped with a Hewlett-Packard
5350B microwave frequency counter for the determination of the
g factors, which were corrected with respect to that of DPPH radical
(g ) 2.00364). When using mixtures of BHA and compound 5g,
the molar ratio of the two equilibrating radicals was obtained from
the EPR spectra and used to determine the equilibrium constant,
Ke. Spectra were recorded a few seconds after starting to irradiate
in order to avoid significant consumption of the phenols during
the course of the experiment. Relative radical concentrations were
determined by comparison of the digitized experimental spectra
with computer-simulated ones as previously described.28a
Acknowledgment. The Swedish Research Council and
MIUR (Rome, Italy) are gratefully acknowledged for financial
support.
HPLC Peroxidation Assay. The experimental setup has been
recently described.26 Inhibition times, Tinh
,
26 and inhibited rates of
25
peroxidation, Rinh
,
were determined as previously described.
Supporting Information Available: General synthetic details
and 1H, 13C, and 77Se NMR spectra of compounds prepared.
Absolute energies, energy corrections, and Cartesian coordinates
for compounds 4a, 5, and 6 and their corresponding radical cations
and phenoxyl radicals. This material is available free of charge via
the Internet at http://pubs.acs.org.
Kinetic Measurements. The rate constants for the reaction of
the antioxidants with peroxyl radicals presented in Table 2 have
been measured by following the autoxidation of styrene (4.3 M) in
chlorobenzene, at 303 K using as initiator AIBN (5 × 10-2 M).
The antioxidant concentration was varied in the range of 2.5 ×
10-6 to 5.0 × 10-5 M. In order to evaluate the regenerability of 5a
and 5g in homogeneous solution, 1-octanethiol (1 × 10-5 to 1 ×
10-3 M) was added to the autoxidation, while the concentration of
the antioxidant (5a or 5g) was kept constant for all measurements
(6.25 × 10-6 M) in order to compare more easily their behavior.
JO0700023
(52) 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.
J. Org. Chem, Vol. 72, No. 7, 2007 2595