Catalytic chain-breaking pyridinol antioxidants

Article abstract of DOI:10.1021/jo902226t

(Chemical Equation Presented) The synthesis of 3-pyridinols carrying alkyltelluro, alkylseleno, and alkylthio groups is described together with a detailed kinetic, thermodynamic, and mechanistic study of their antioxidant activity. When assayed for their capacity to inhibit azo-initiated peroxidation of linoleic acid in a water/chlorobenzene two-phase system, tellurium-containing 3-pyridinols were readily regenerable by N-acetylcysteine contained in the aqueous phase. The best inhibitors quenched peroxyl radicals more efficiently than α-tocopherol, and the duration of inhibition was limited only by the availability of the thiol reducing agent. In homogeneous phase, inhibition of styrene autoxidation absolute rate constants k_inh for quenching of peroxyl radical were as large as 1 x 10⁷ M^-1 s ^-1, thus outperforming the best phenolic antioxidants including α-tocopherol. Tellurium-containing 3-pyridinols could be quantitatively regenerated in homogeneous phase by N-tert-butoxycarbonyl cysteine methyl ester, a lipid-soluble analogue of N-acetylcysteine. In the presence of an excess of the thiol, a catalytic mode of action was observed, similar to the one in the two-phase system. Overall, compounds bearing the alkyltelluro moiety ortho to the OH group were much more effective antioxidants than the corresponding para isomers. The origin of the high reactivity of these compounds was explored using pulse-radiolysis thermodynamic measurements, and a mechanism for their unusual antioxidant activity was proposed. The tellurium-containing 3-pyridinols were also found to catalyze reduction of hydrogen peroxide in the presence of thiol reducing agents, thereby acting as multifunctional (preventive and chain-breaking) catalytic antioxidants.

Full text of DOI:10.1021/jo902226t

pubs.acs.org/joc
Catalytic Chain-Breaking Pyridinol Antioxidants
†
†
†
,†
‡
€
Sangit Kumar, Henrik Johansson, Takahiro Kanda, Lars Engman,* Thomas Muller,
Helena Bergenudd,‡ Mats Jonsson,‡ Gian Franco Pedulli,§ Riccardo Amorati,§ and
Luca Valgimigli*,§
†Department of Biochemistry and Organic Chemistry, Uppsala University, Box 576, SE-751 23 Uppsala,
Sweden, ‡School of Chemical Science and Engineering, Nuclear Chemistry, Royal Institute of Technology,
10044 Stockholm, Sweden, and Department of Organic Chemistry “A. Mangini”. University of Bologna, via
S. Giacomo 11, 40126 Bologna, Italy
§
lars.engman@biorg.uu.se
Received October 16, 2009
The synthesis of 3-pyridinols carrying alkyltelluro, alkylseleno, and alkylthio groups is described
together with a detailed kinetic, thermodynamic, and mechanistic study of their antioxidant activity.
When assayed for their capacity to inhibit azo-initiated peroxidation of linoleic acid in a water/
chlorobenzene two-phase system, tellurium-containing 3-pyridinols were readily regenerable by
N-acetylcysteine contained in the aqueous phase. The best inhibitors quenched peroxyl radicals more
efficiently than R-tocopherol, and the duration of inhibition was limited only by the availability of the
thiol reducing agent. In homogeneous phase, inhibition of styrene autoxidation absolute rate
constants kinh for quenching of peroxyl radical were as large as 1 ꢀ 107 M-1 s-1, thus outperforming
the best phenolic antioxidants including R-tocopherol. Tellurium-containing 3-pyridinols could be
quantitatively regenerated in homogeneous phase by N-tert-butoxycarbonyl cysteine methyl ester, a
lipid-soluble analogue of N-acetylcysteine. In the presence of an excess of the thiol, a catalytic mode
of action was observed, similar to the one in the two-phase system. Overall, compounds bearing the
alkyltelluro moiety ortho to the OH group were much more effective antioxidants than the
corresponding para isomers. The origin of the high reactivity of these compounds was explored
using pulse-radiolysis thermodynamic measurements, and a mechanism for their unusual antioxi-
dant activity was proposed. The tellurium-containing 3-pyridinols were also found to catalyze
reduction of hydrogen peroxide in the presence of thiol reducing agents, thereby acting as multi-
functional (preventive and chain-breaking) catalytic antioxidants.
Introduction
radicals much faster than the chain-propagating H-atom
transfer step of lipid peroxidation. It is well-known that
electron-donating ortho and para substituents in the phenolic
moiety weaken the O-H bond3 and thus increase the rate of
The majority of chain-breaking antioxidants, both in
Nature1 and in man-made materials,2 are phenolic. The
antioxidant activity of these compounds stems from their
ability to transfer the phenolic hydrogen to lipidperoxyl
ꢀ
(3) (a) Lind, J.; Shen, X.; Eriksen, T. E.; Merenyi, G. J. Am. Chem. Soc.
1990, 112, 479–482. (b) Jonsson, M.; Lind, J.; Eriksen, T. E.; Merenyi, G.
ꢀ
(1) Burton, G. W.; Ingold, K. U. Acc. Chem. Res. 1986, 19, 194–201.
(2) Zweifel, H. Stabilization of Polymeric Materials; Springer: Berlin
1997.
J. Chem. Soc., Perkin Trans. 2 1993, 1567–1568. (c) Lucarini, M.; Pedrielli,
P.; Pedulli, G. F. J. Org. Chem. 1996, 61, 9259–9263. (d) Wright, J. S.;
Johnson, E. R.; DiLabio, G. A. J. Am. Chem. Soc. 2001, 123, 1173–1183.
716 J. Org. Chem. 2010, 75, 716–725
Published on Web 01/14/2010
DOI: 10.1021/jo902226t
r
2010 American Chemical Society

Kumar et al.
JOCArticle
hydrogen atom transfer. However, improving the antioxidant
activity of phenolic compounds along these lines will be
successful only until the ionization potential of the compounds
(which is also decreased by introduction of electron-donating
groups) becomes so low4 that the material will be consumed in
a spontaneous electron-transfer reaction with molecular oxy-
gen. Recently, by incorporating one or two nitrogens into the
hydroxyaromatic ring, the groups of Pratt and Valgimigli
found a way to circumvent the problem with the low ionization
potential. Thus, whereas the O-H bond dissociation enthal-
pies (BDEs) of substituted 5-pyrimidinols and 3-pyridinols
differed only marginally from those of the corresponding
phenols, the ionization potentials were substantially higher.
The O-H bond of 5-pyrimidinol 1 (78.2 kcal mol-1) has
almost the same strength as that of R-tocopherol (2; 78.3 kcal
mol-1), but the calculated ionization potential is 7.7 kcal
mol-1 higher (167.0 kcal mol-1).5 Furthermore, as judged by
the capacity to inhibit autoxidation of styrene, pyrimidinol 1
(kinh = 8.6 ꢀ 106 M-1 s-1) transfers its phenolic hydrogen
atom to peroxyl radicals almost three times as fast as
R-tocopherol (kinh = 3.2 ꢀ 106 M-1 s-1).4 It has been hypo-
thesized that this rate enhancement is due to polar effects in the
transition state of the atom-transfer reaction.6 Incorporation
of an electron-donating amino substituent para to the hydroxyl
group in the 3-pyridinol scaffold in a fused five-membered ring
(compound 3) resulted in the most effective phenolic chain-
breaking antioxidant reported to date. Its ionization potential
is 7.0 kcal mol-1 lower than calculated for R-tocopherol, and
therefore it slowly decomposes when exposed to atmospheric
oxygen. As a result of the 2.8 kcal mol-1 weaker O-H BDE of
compound 3 compared to that of R-tocopherol, the reactivity
eration to allow for a catalytic mode of action of the antioxidant.
In biological membranes this process is thought to occur by do-
nation of a hydrogen atom from ascorbate (AscH-) to the R-to-
copheroxyl radical at the lipid-aqueous interphase (eq 1).11-14
R-TO• þ AscH- f R-TOH þ Asc• -
ð1Þ
The tripeptide glutathione (GSH) is present in much
higher concentrations than ascorbate in human plasma. It
is known to act as a biological antioxidant and reducing
agent both by one-electron (hydrogen atom) and two-elec-
tron (e.g., as a cofactor for the glutathione peroxidase
enzymes) chemistry. However, early studies by Barclay15
showed that GSH is incapable of regenerating R-tocopherol
from the R-tocopheroxyl radical in simple model systems. In
our search for other chain-breaking antioxidants that could
perform in a catalytic fashion in the presence of thiols, we
recently found that 2,3-dihydrobenzo[b]selenophene-5-ols
516 and ethoxyquins 617 were regenerable by N-acetylcys-
teine when assayed for their capacity to inhibit azo-initiated
peroxidation of linoleic acid in a two-phase system. Con-
sidering that none of these antioxidants quenched peroxyl
radicals as efficiently as R-tocopherol, we thought it would
be interesting to try to modify the efficient pyridinol anti-
oxidants in such a way that they could also act in a catalytic
fashion in the presence of stoichiometric amounts of a thiol
reducing agent. In this paper we report on the synthesis of
3-pyridinols carrying organosulfur organoselenium, and
organotellurium substituents. Their inhibition of autoxida-
tion in a two-phase system and in a purely lipid environment
is described, and a mechanism is proposed for the observed
catalytic mode of action.18
toward peroxyl radicals is an impressive 88-fold higher (kinh
=
280 ꢀ 106 M-1 s-1).7,8 The synthesis and antioxidative proper-
ties of more vitamin E-like derivatives of this kind,9 such as the
naphthyridinol 4, have recently been described.10
Results and Discussion
Synthesis. For introduction of chalcogens into the
3-pyridinol scaffold, it was thought that halogenated deri-
vatives, which are commercially available with some variety,
could serve as suitable starting materials. On treatment with
3 equiv of t-BuLi in dry THF at -78 °C, 6-bromo-3-pyri-
dinol produced a solution of the corresponding 6,O-dili-
thiated species (Scheme 1). For preparation of the octyltel-
luro derivative 7a, it was found most convenient to add finely
ground elemental tellurium to the organolithium and then
allow the insertion product to air-oxidize to the correspond-
ing ditelluride. Subsequent borohydride reduction of crude
ditelluride and alkylation with octyl bromide afforded com-
pound 7a in 66% overall yield. Selenide 7b and sulfide 7c
R-Tocopherol, the most reactive component of vitamin E, is
known to trap two peroxyl radicals before it is converted into
nonradical products. Nature has therefore arranged for its regen-
(11) Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and
Medicine, 4th ed.;Oxford University Press: Oxford, 2007; p 168.
˚
(12) Kamal-Eldin, A.; Appelquist, L.-A. Lipids 1996, 31, 671–701. ; also
see refs 266-286. See also: Sato, K.; Niki, E.; Shimasaki, H. Arch. Biochem.
Biophys. 1990, 279, 402–405. Bisby, R.; Parker, A. W. Arch. Biochem.
Biophys. 1995, 317, 170–178. Noguchi, N.; Niki, E. Free Radical Res. 1998,
28, 561–572. Niki, E.; Noguchi, N. Acc. Chem. Res. 2004, 37, 45–51.
(13) Niki, E.; Saito, T.; Kawakami, A; Kamiya, Y. J. Biol. Chem. 1984,
259, 4177–4182.
(14) Niki, E.; Kawakami, A.; Yamamoto, Y.; Kamiya, Y. Bull. Chem.
Soc. Jpn. 1985, 58, 1971–1975. Doba, T.; Burton, G. W.; Ingold, K. U.
Biochem. Biophys. Acta 1985, 835, 298–303.
(15) Barclay, L. R. C. J. Biol. Chem. 1988, 263, 16138–16142.
(16) Kumar, S.; Johansson, H.; Engman, L.; Valgimigli, L.; Amorati, R.;
Fumo, M. G.; Pedulli, G. F. J. Org. Chem. 2007, 72, 2583–2595.
(17) Kumar, S.; Engman, L.; Valgimigli, L.; Amorati, R.; Fumo, M. G.;
Pedulli, G. F. J. Org. Chem. 2007, 72, 6046–6055.
(4) 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.
(5) Pratt, D. A.; DiLabio, G. A.; Brigati, G.; Pedulli, G. F.; Valgimigli, L.
J. Am. Chem. Soc. 2001, 123, 4624–4626.
(6) Valgimigli, L.; Brigati, G.; Pedulli, G. F.; DiLabio, G. A.; Mastra-
gostino, M.; Arbizzani, C.; Pratt, D. A. Chem.;Eur. J. 2003, 9, 4997–5010.
(7) Wijtmans, M.; Pratt, D. A.; Valgimigli, L.; DiLabio, G. A.; Pedulli, G.
F.; Porter, N. A. Angew. Chem., Int. Ed. 2003, 42, 4370–4373.
(8) Wijtmans, M.; Pratt, D. A.; Brinkhorst, J.; Serwa, R.; Valgimigli, L.;
Pedulli, G. F.; Porter, N. A. J. Org. Chem. 2004, 69, 9215–9223.
(9) Kim, H.-Y.; Pratt, D. A; Seal, J. R.; Wijtmans, M.; Porter, N. A.
J. Med. Chem. 2005, 48, 6787–6789. Nam, T.; Wijtmans, M.; Pratt, D. A.;
Porter, N. A. Synthesis 2005, 1397–1404.
(10) Nam, T.; Rector, C. L.; Kim, H.; Sonnen, A. F.-P.; Meyer, R.; Nau,
W. M.; Atkinson, J.; Rintoul, J.; Pratt, D. A.; Porter, N. A. J. Am. Chem. Soc.
2007, 129, 10211–10219.
(18) For a preliminary account on this work, see: Kumar, S.; Johansson,
€
H.; Kanda, T.; Engman, L.; Muller, T.; Jonsson, M.; Pedulli, G. F.; Petrucci,
S.; Valgimigli, L. Org. Lett. 2008, 10, 4895–4898.
J. Org. Chem. Vol. 75, No. 3, 2010 717

Kumar et al.
JOCArticle
SCHEME 1. Organolithium Approach to Compounds 7
SCHEME 2. Preparation of Compound 16a
were prepared in 65% and 84% yields, respectively, by
treatment of the organolithium with readily available dioctyl
diselenide and dioctyl disulfide. Compounds 8a-8c were
analogously prepared from 2-bromo-3-pyridinol and tell-
uride 9a from 2-iodo-6-methyl-3-pyridinol.
a p-methoxy group, 3-cyano-5-methoxy-4,6-dimethyl-2-pyr-
idinol (17; which is present as the 2-pyridone tautomer) was
prepared by cyclocondensation of 3-methoxypentane-2,4-
dione with 2-cyanoacetamide21 and taken through the steps
described in Scheme 2 to provide telluride 18a.
Selenide 9b and sulfide 9c were conveniently prepared by
using a microwave-induced, copper-catalyzed coupling of
aryl halides with diselenides and disulfides we developed
some time ago (eq 2).19
Yield-wise the two syntheses did not differ very much. However,
it was noted that compound 18a had a tendency to decompose
more readily than compound 16a during handling, purification,
and storage. Selenide 18b was prepared in 64% yield from the
corresponding 2-bromopyridine in analogy with the procedure
for compound 7b (Scheme 1). To find the optimal positioning
of the chalcogen moiety (ortho or para with respect to the
pyridinol), compounds 21a-c were prepared as outlined in
Scheme 3. 3-Aminopyridine 19 was an intermediate in the
conversion of nitrile 17 to compounds 18. Diazotization and
hypophosphorous acid reduction (37%) followed by boron
tribromide induced O-demethylation (78%) afforded bromo-
pyridinol 20, which was converted into telluride 21a and selenide
21b using lithiation methodology. As a control for the mecha-
nistic studies, we wanted access to a chalcogen-containing
pyridine that could no longer act as a hydrogen atom donor.
Bromopyridinol 15 was therefore first O-methylated with methyl
iodide and the octyltelluro group installed following lithiation
and reaction with dioctyl ditelluride (eq 3, compound 23).
Compounds 10a-c carrying an arylchalcogeno group next
to the hydroxyl were all described in our report on the
coupling reaction.19 Introduction of additional methyl
groups into the 3-pyridinol scaffold was more demanding
from a synthetic point of view. It occurred to us that a cyano
group in the 3-position of the pyridine could serve as a source
of a pyridinol via a short series of transformations involving
hydrolysis to an amide, Hofmann rearrangement, and dia-
zotization/aromatic nucleophilic substitution. Shown in
Scheme 2 is the preparation of a 2,6-dimethylated derivative
16a from the commercially available 3-cyano-2-pyridinol 11.
2-Pyridinol to 2-bromopyridine conversion was induced in
high yield by P2O5 and tetrabutylammonium bromide in
refluxing toluene.20 Hydrolysis of nitrile 12 in concentrated
sulfuric acid provided the corresponding amide 13 in 68%
yield. Hofmann rearrangement was induced in 84% yield by
treatment with bromine in an alkaline solution, and the
resulting amine 14 was diazotized in water to provide
3-pyridinol 15 in a modest yield (40%). Then, by using the
methodologies outlined in Scheme 1, octyltelluro (16a 44%
yield), octylseleno (16b, 60%), and octylthio groups (16c,
35%) were introduced into the 2-position. In an attempt to
increase the electron density on tellurium by introduction of
To study the antioxidative capacity of telluroxides, some
of the tellurium-containing pyridinols (8a and 16b) were
(19) Kumar, S.; Engman, L. J. Org. Chem. 2006, 71, 5400–5403.
(20) Kato, Y.; Okada, S.; Tomimoto, K.; Mase, T. Tetrahedron Lett.
2001, 42, 4849–4851.
(21) Tang, H.; Chen, S.; Zhang, P. Acta Chim. Sinica (Huaxue Xuebao)
1985, 43, 72–78. Yu, Y.; Chen, G.; Zhu, J.; Zhang, X.; Chen, S.; Tang, H.;
Zhang, P. J. Chem. Soc., Perkin Trans. 1 1990, 2239–2243.
718 J. Org. Chem. Vol. 75, No. 3, 2010

Kumar et al.
JOCArticle
SCHEME 3. Preparation of Compounds 21
treated with a slight excess of tert-butyl hydroperoxide in
methanol. The polar compounds 24 and 25, isolated after
evaporation of all volatiles under reduced pressure, were
difficult to purify and characterize. For example, peaks in the
1H NMR spectrum were extremely broad, which could be
indicative of ongoing exchange processes or oligomerization.
Mass spectral analysis was also inconclusive. In fact, the
strongest indication of the identity of the compounds comes
from reduction of the crude telluroxide with sodium disulfite,
which always cleanly returned the corresponding alkyl pyri-
dyl telluride.
experience, this approach provides the most evident indica-
tion of regenerability and catalytic behavior, the two-phase
system served to screen all of the compounds and identify the
best performing structural architectures, which were subse-
quently subjected to more detailed kinetic investigation. In
the absence of N-acetylcysteine, all organotellurium com-
pounds failed to inhibit peroxidation of linoleic acid. This is
because organotelluriums are rapidly oxidized by residual
linoleic acid hydroperoxide present in commercial samples of
the acid, and the resulting telluroxides are poor chain-break-
ing antioxidants (vide infra). Pyridinols carrying octylseleno
substituents often show a longer inhibition time if NAC is
present in the aqueous phase (up to a 2-fold increase;
compounds 7b, 8b, 9b, 10b, 16b, 18b, 21b). Thus, they are
regenerable to some extent and notably more so than the
corresponding organosulfur compounds. However, even if
electron-donating substituents are introduced in order to
lower the O-H bond dissociation enthalpy, the organosele-
nium and organosulfur compounds often turn out to be poor
quenchers of peroxyl radicals with Rinh values in the range of
Inhibition Studies in a Two-Phase System. Azo-initiated
peroxidation of linoleic acid or derivatives thereof has fre-
quently been used for antioxidant studies. Some time ago we
designed a two-phase variant that allows the study of lipid-
soluble antioxidant regeneration by water-soluble co-anti-
oxidants.22 In the experimental setup (see Supporting In-
formation, Figure S1), linoleic acid and the antioxidant to be
evaluated were vigorously stirred in chlorobenzene at 42 °C
with an aqueous solution of N-acetylcysteine (NAC). 2,20-
Azobis(2,4-dimethylvaleronitrile) (AMVN) was added as an
initiator in the organic phase, and the progress of peroxida-
tion was monitored by HPLC by following the conjugated
diene hydroperoxide formation. The inhibited rate of perox-
idation, Rinh, was determined by least-squares methods from
absorbance/time plots, while the duration of the inhibited
phase, Tinh, was determined graphically as the cross-point
for the inhibited and the uninhibited lines (see Figure 1).
Whether NAC was present in the aqueous phase or not,
R-tocopherol, which we used as a control, inhibited peroxi-
dation of linoleic acid for almost the same time as in the lipid
phase (Table 1, bottom). Thus, it is essentially not regenerable
under the conditions used. The capacity of the chalcogen-
containing pyridinol antioxidants to inhibit peroxidation is
summarized in Table 1. Since, according to our previous
45-328 μM h-1
.
The two 3-pyridinols 7a and 8a, carrying octyltelluro
groups para and ortho to the hydroxyl, respectively, inhibited
peroxidation with similar rates (32 vs 27 mM h-1), but the
former was clearly inferior in terms of inhibition time (70 vs
200 min) in the presence of NAC. Organotellurium catalyst
9a inhibited peroxidation as efficiently as R-tocopherol with
an inhibition time exceeding 360 min. Thus, regenerability
was also improved by introduction of the methyl group.
Phenyltelluro derivative 10a performed poorer (Tinh = 280
min) than its octyltelluro counterpart in this respect.
4,6-Dimethylated 3-pyridinol derivative 16a, carrying an
octyltelluro group in position 2, quenched peroxyl radicals
three times as efficiently as R-tocopherol (Rinh = 8 μM h-1
)
in the presence of N-acetylcysteine. Regenerability was also
excellent (Tinh > 400 min). Again, chain-breaking capacity
and regenerability of the corresponding organoselenium and
sulfur compounds 16b and 16c could not match those of the
organotellurium derivative. Organotellurium 18a, prepared
in analogy with 16a, showed similar antioxidant characteri-
stics and clearly outperformed its selenium analogue 18b.
Compounds 21 are isomeric to compounds 16, but the
organochalcogen and 6-methyl groups were swapped. As
noted above with compounds 7a and 8a, a close (ortho)
arrangement of octyltelluro and hydroxyl groups in the
€
(22) Vessman, K.; Ekstrom, M.; Berglund, M.; Andersson, C.-M.; Eng-
man, L. J. Org. Chem. 1995, 60, 4461–4467.
J. Org. Chem. Vol. 75, No. 3, 2010 719

Kumar et al.
JOCArticle
TABLE 1. Inhibited Rate of Linoleic Acid Peroxidation (Rinh) and
Inhibition Times (Tinh) for Antioxidants and Reference Compounds Tested
in the Two-Phase Model
FIGURE 1. Peroxidation traces (linoleic acid hydroperoxide con-
centration vs time) recorded using compound 16a (40 μM) or
R-TOH (40 μM) as antioxidants in the chlorobenzene layer in the
presence of NAC (1 mM) in the aqueous phase.
pyridinol is much preferred over the distant when it comes to
regenerability. Bromopyridinol 15, which was tested as a
chalcogen-free reference antioxidant, did not inhibit peroxi-
dation at all under the conditions of the assay. Obviously an
octyltelluro group in the ortho position has a considerable
effect on the reactivity. Compound 23, the O-metylated
analogue of our most potent regenerable antioxidant 16a,
showed only a limited inhibiting effect indicating that hydro-
xylic hydrogen atom transfer to peroxyl radicals is the
prevailing antioxidant mechanism. NAC is needed to keep
the compound in the reduced divalent state where it may
quench peroxyl radicals either by hydrogen atom transfer
(HAT) or by electron transfer (ET). Whereas HAT would be
impaired by O-methylation, ET would still be possible;
however, the resulting radical cation does not undergo
effective reduction by aqueous thiol.
The peroxidation traces in Figure 1 show that compound
16a can clearly outperform R-tocopherol when it comes to
the duration of the inhibited phase. In fact, the limitation is
the availability of the thiol reducing agent, and Tinh decreases
in a more or less linear fashion to the limiting value under
nonregenerating conditions as the NAC concentration is
lowered (see Supporting Information, Figure S2). While
keeping the concentration of N-acetylcysteine constant
(1 mM), the concentration of antioxidant 16a was lowered
from the 40 μM used in the standard assay. The inhibition of
peroxidation did not change at the 20 μM level, decreased at
10 μM (Tinh = 310 min), and could not be observed at the
5 μM level. Sampling of the aqueous (instead of the chloro-
benzene) phase at intervals during peroxidation and analysis
by reversed phase HPLC showed that the thiol/disulfide
ratio decreased linearly with time until complete consump-
tion of N-acetylcysteine after ca. 400 min (Figure 2).
Inhibition Studies in Homogeneous Phase. To achieve a
deeper insight into the antioxidant behavior of alkyltelluro-
substituted 3-pyridinols and into the mechanisms underlying
both their surprising reactivity toward peroxyl radicals and
their catalytic behavior, we investigated the inhibition of
homogeneous-phase autoxidation of styrene in chloroben-
zene (or acetonitrile), initiated by thermal decomposition of
2,20-azobisisobutyronitrile (AIBN) at 303 K. Under such
robust and well established conditions described by the usual
eqs 4-9,4 kinetics of oxygen consumption was monitored in
aRate of peroxidation during the inhibited phase (uninhibited rate ca.
650 μM h-1). Data for potent catalysts are mean ( SD for triplicates.
Estimated errors for poorer catalysts are (20%. bInhibited phase of
peroxidation. Reactions were monitored for 400 min. Estimated errors
are (5%.
a custom-built differential oxygen-uptake apparatus that has
been previously described in detail.23 Compounds 7a, 8a, and
(23) 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.
720 J. Org. Chem. Vol. 75, No. 3, 2010

Kumar et al.
JOCArticle
TABLE 2. Absolute Rate Constants kinh and Stoichiometric Factors n
for Inhibition of Homogeneous-Phase Autoxidation of 4.3 M Styrene at
303 K by Selected Pyridinols and Reference Compoundsa
chlorobenzene
acetonitrile
kinh (M-1 s-1
)
n
kinh (M-1 s-1
)
n
7a
8a
16a
24
25
26
27
28
29
(7.5 ( 0.9) ꢀ 105 0.3 ( 0.1 (1.9 ( 0.3) ꢀ 105 0.3 ( 0.1
(1.0 ( 0.3) ꢀ 107 0.5 ( 0.1 (2.9 ( 0.3) ꢀ 106 0.5 ( 0.1
(9.2 ( 2.0) ꢀ 106 0.4 ( 0.1
<100
<100
<100
<100
<1000b
2.2 ꢀ 104b
2.9 ꢀ 105b
1.6 ꢀ 107c
2.0
2.0
2.0
r-TOC 3.2 ꢀ 106d
6.5 ꢀ 105
2.0
aData are mean ( SD, each measurement in triplicate. bReference
value from ref 8. cReference value from ref 7. dReference value from ref 4.
FIGURE 2. Concentration of N-acetylcysteine in the aqueous
phase with time during a normal peroxidation experiment using
antioxidant 16a.
To confirm this last assumption, we investigated the anti-
oxidant capacity of telluroxides 24 and 25. No inhibition or
retardation of styrene autoxidation was observed, allowing
us to estimate kinh < 103 M-1 s-1 for these compounds. In
our preliminary communication, we observed that during
inhibition of styrene autoxidation by 16a, at the end of the
inhibited period, the rate of oxygen consumption was still
retarded as compared with an uninhibited autoxidation
under identical conditions.18 We tentatively attribute this
effect to styryl (alkyl) radical trapping by quinonoid pro-
ducts (eq 10) formed by detelluration of the antioxidant.
16a and the telluroxides 24 and 25 were chosen as represen-
tative of the structural diversity contained in the type of
pyridinols of this study.
Ri
initiator f R•
ð4Þ
ð5Þ
R• þ O2 f ROO•
kp
ROO• þ RH f ROOH þ R•
ð6Þ
ð7Þ
2kt
ROO• þ ROO• f non-radical products
kinh
ROO• þ ArOH f ROOH þ ArO•
ROO• þ ArO• f non-radical products
ð8Þ
ð9Þ
The very high reactivity of compounds 7a, 8a, and 16a was
indeed surprising and can be compared to those previously
measured for pyridinols 26-29.Telluro-pyridinol 16a is
approximately 400 times more reactive than structurally
related compound 27, carrying a methyl in place of the
o-octyltelluro group. Since the reactivity of pyridinols to-
Each of compounds 7a, 8a, and 16a produced a neat
inhibited period in the autoxidation of 4.3 M styrene
in chlorobenzene from which slope the absolute kinh for
reaction with peroxyl radicals was obtained as detailed in
Table 2. Both compounds with the octyltelluro moiety ortho
to the -OH function (8a and 16a) displayed a remarkable
antioxidant behavior, being able to quench peroxyl radicals
approximately 3 times faster than the reference antioxidant
R-tocopherol, confirming and extending our preliminary
results. Interestingly, compound 7a, carrying the octyltelluro
moiety para to the -OH, resulted in a kinh of “only” ∼0.8 ꢀ
106 M-1 s-1, i.e., one order of magnitude lower than the
value recorded for the isomeric 8a.
ward peroxyl radicals is dictated by the capacity of substit-
5,7
uents to lower the BDEOH
,
the high reactivity of alkyl-
telluro-pyridinols would imply either that they react by a
conceptually different mechanism related to the presence of
the chalcogen (e.g., by electron transfer from tellurium lone
pairs to peroxyl radical) or that the alkyltelluro group has a
strong electron-donating character, lowering the BDEOH
.
Despite the high efficiency in trapping peroxyl radicals,
the stoichiometric factor n, i.e. the number of peroxyl
radicals trapped by 1 molecule of antioxidant, as observed
from the length of the inhibition period, was in the range
0.3-0.5 for the three pyridinols, largely below the ideal value
n = 2 produced by R-tocopherol and other phenolic anti-
oxidants. This suggests that either the reaction with peroxyl
radicals follows a mechanism substantially different from the
well-known behavior of conventional phenolic and pyri(mi)-
dinolic antixidants or the alkyltelluro-pyridinol is rapidly
consumed during the course of autoxidation in a side reac-
tion, to produce the corresponding telluroxides unable to
inhibit the autoxidation or with limited antioxidant activity.
The drop in reactivity observed for all investigated com-
pounds on changing the solvent from apolar chlorobenzene
to polar and H-bonding acetonitrile (Table 2) is exactly what
was expected for regular phenolic or pyridinolic compounds
reacting by HAT or proton coupled electron transfer (PCET)
to the peroxyl radicals.24
(24) Valgimigli, L.; Banks, J. T.; Lusztyk, J.; Ingold, K. U. J. Org. Chem.
1999, 64, 3381–3383.
J. Org. Chem. Vol. 75, No. 3, 2010 721

Kumar et al.
JOCArticle
Thermodynamic Properties. The scarce data on the elec-
tronic properties of alkyltelluride substituents prompted us
to measure the BDEOH of compounds 7a and 8a in compari-
son with unsubstituted 3-pyridinol 26. Unfortunately, the
telluro-pyridinols were unstable to UV irradiation, and the
low persistency of the corresponding 3-pyridinoxyl radicals
prevented the use of the radical equilibration EPR-ap-
proach.6 We therefore turned to electrochemical cycles in
water. According to this approach the BDEOH (kJ/mol) is
obtained from eq 11, where E° is the standard reduction
potential of the pyridinoxyl radical, pKa is the O-H pKa of
the pyridinol, and C is a constant that depends on the family
of compounds and the solvent, accounting for the entropy
and solvation terms.
TABLE 3. Thermodynamic Data for Selected Pyridinols from Pulse
Radiolysis Equilibration Measurements in Water at pH 12 (E°) and Cyclic
Voltammetry in Acetonitrile (Ep) at 298 K
a
compound pKa E° [V] vs NHE BDEOH [kJ/mol] Ep [V] vs NHE
7a
7b
7c
8a
8b
8c
26
9.1
8.8
8.7
9.3
9.0
9.7
9.1
0.80
0.81
0.79
0.78
0.82
0.80
1.06b
361
360
358
360
362
364
386
1.05
1.31
1.40
1.05
1.25
1.32
1.58
aAcidity of the OH group in water. The pKa values for protonation of
the pyridine ring are as follows: 7a, 2.6; 7b, 2.5; 7c, 2.6; 8a, 3.0; 8b, 2.5; 8c,
e2; 26, 5.3. bReference value from ref 26.
amounts of LipCys. The lengths of τ varied linearly with the
concentration of LipCys (Figure 3B) as observed previously
in our two-phase system, and the limiting factor for the
duration of inhibition was the availability of thiol. However,
in order to discriminate the antioxidant effect of 16a from
that of LipCys itself, we could not increase the amount of
thiol much more. Conversely, in the presence of LipCys, τ
was almost independent of the initial concentration of 16a
(see Supporting Information, Figure S3), confirming a true
catalytic behavior of 16a.
While the inhibition time for R-tocopherol was unaffected
by addition of LipCys within experimental error,18 also the
simpler pyridinol 8a, carrying the octyltelluro moiety in the
ortho position, could be effectively regenerated by the thiol,
as judged from the extension of τ. Again, τ was proportional
to the initial concentration of LipCys (Figure 4) and there
was no apparent upper limit, provided a sufficient amount of
LipCys was added to the system. Interestingly, its corre-
sponding telluroxide 24, which was unable to inhibit styrene
autoxidation in the absence of thiol, showed almost identical
inhibition characteristics as 8a in the presence of LipCys
(Figure 4), clearly suggesting that the role of the thiol
includes reduction of telluroxide to the parent telluride.
Conversely, only modest catalytic behavior was observed
for the para isomer 7a and significantly larger amounts of
LipCys were needed to extend its inhibition time up to the
value corresponding to n = 2. Qualitatively, the behavior in
homogeneous chlorobenzene solution paralleled our experi-
ence in the two-phase system.
BDEOH ¼ 96:48 E° þ 5:70 pKa þ C
ð11Þ
In this work, the standard reduction potentials of the
anions from 7a-c and 8a-c were measured in water by
pulse radiolysis from equilibration with a reference reduc-
tant (4-iodophenol E° = 0.82 V vs NHE and phenol E° = 0.
79 V vs NHE) at pH 12 (where the compound is mostly in its
dissociated form), and the constant C previously found for
phenols (232 kJ/mol)25 was used. Results reported in Table 3
show that the para alkyltelluro group has a strong electron-
donating character, inducing a BDEOH lowering in water
(-6.6 kcal mol-1) higher than that of para MeO (-5.6 kcal
mol-1 in water3a), but lower than Me2N- (-14.0 kcal mol-1
in water3a). These measurements provide a satisfying expla-
nation for the reactivity of compound 7a. However, they do
not explain the much higher reactivity of the isomeric 8a (not
significantly different in reactivity than 16a). Indeed the
BDEOH values for ortho (8a) and para (7a) alkyltellurium
pyridinols were almost identical.
It can be noticed that while in water at pH 12 the standard
potentials of all investigated compounds (in their dissociated
form) are approximately the same, in acetonitrile the ease of
oxidation of the parent chalcogen-containing pyridinols
follows the trend Te > Se > S. This indicates that the ortho
and para alkyl-chalcogenide substituent effects are virtually
identical with respect to the redox properties of dissociated
pyridinols. However, for the nondissociated pyridinols, the
chalcogen-dependent trend indicates that the chalcogen is
directly involved in the redox process rather than just being a
remote substituent.
To further test the ability of octyltelluro-pyridinols and
their corresponding oxides to act in a catalytic fashion and to
be fully regenerated by LipCys, we performed a series of
styrene autoxidations inhibited initially only by the organo-
telluriums, followed by the injection of LipCys in the auto-
xidating mixture after the end of inhibition, i.e., when the
pyridinol antioxidant had already been consumed and the
autoxidation was running uninhibited. As exemplified in
Figure 5, any of the three pyridinols was fully regenerated
by LipCys and the pyridinol acted in catalytic fashion, exactly
as it would have done by adding the thiol at the beginning of
the autoxidation experiment. By this approach, with com-
pounds 8a and 16a, it was possible to extend indefinitely the
inhibition time, simply by adding aliquots of LypCys at
regular time intervals (see Supporting Information, Figure
S4). The catalytic efficiency of the tellurium-containing pyri-
dinols, n(SH), can be expressed as the number of radical
chains interrupted by each molecule of thiol, given by eq 12,
where τ is the length of the inhibition period after subtracting
Catalytic Activity in Homogenous Phase. In our prelimin-
ary report18 we indicated that addition of 1-2 equiv of
alkylmercaptans to the autoxidation of styrene (at 303 K in
chlorobenzene) inhibited by 16a had little effect on the length
of the inhibition period. In contrast, an inhibition period
corresponding to n = 2 was observed by adding to 16a
N-tert-butoxycarbonyl cysteine methyl ester (LipCys), a
lipid-soluble analogue of N-acetylcysteine. We attributed
this to complete suppression of oxidation at tellurium and
partial regeneration of the pyridinol from the corresponding
aryloxyl radical; in other words, we attributed it to catalytic
antioxidant activity of 16a even in homogeneous solution. As
depicted in Figure 3, the inhibition time τ for compound 16a
could be extended well beyond the value corresponding to
n = 2 recorded for R-tocopherol simply by adding increasing
€
€
(25) Malmstrom, J.; Jonsson, M.; Cotgreave, I. A.; Hammarstrom, L.;
€
Sjodin, M.; Engman, L. J. Am. Chem. Soc. 2001, 123, 3434–3440.
722 J. Org. Chem. Vol. 75, No. 3, 2010

Kumar et al.
JOCArticle
FIGURE 3. (A) Typical oxygen uptake kinetic traces during homogeneous phase autoxidation of styrene (4.3 M) in chlorobenzene, initiated
by AIBN (0.05 M) at 303 K, in the absence of inhibitors (slashed line) and inhibited by (1) N-tert-butoxycarbonyl cysteine methyl ester (LipCys)
9.0 ꢀ 10-5M; (2) 16a 1.3 ꢀ 10-5 M; (3) R-TOH 6.3 ꢀ 10-6 M; (4) 16a 6.3 ꢀ 10-6 M þ LipCys 9.0 ꢀ 10-5 M; (5) 16a 6.3 ꢀ 10-6 M þ LipCys 1.8 ꢀ
10-4M. (B) Inhibition times τ obtained with 16a 6.3 ꢀ 10-6 M, as a function of the initial concentration of LipCys.
FIGURE 4. Length of the inhibited period observed during styrene
autoxidation in the presence of organotelluriums 7a, 8a, and 24
(6.3 ꢀ 10-6 M) with increasing amounts of N-tert-butoxycarbonyl
cysteine methyl ester (LipCys).
the intrinsic τ of the pyridinol alone, and Ri is the initiation
rate, measured as reported in the experimental section (8.9 ꢀ
10-9 M s-1). The values of n(SH) recorded in homogeneous
phase were 0.11 ( 0.03, 0.25 ( 0.08, 0.26 ( 0.06, 0.27 ( 0.07,
and 0.37 ( 0.03 for compounds 7a, 8a, 16a, 24, and 25,
respectively. Efficiencies lower than unity clearly indicate that
a fraction of the thiol is consumed in reactions not leading to
chain termination.
nðSHÞ ¼ ðRiτÞ=½SHꢁ
ð12Þ
For comparison, the catalytic efficiency in the two-phase
system can be calculated by comparing the inhibition time
recorded with R-tocopherol (Tinh = 80 min at 40 μM and
known to quench two peroxyl radicals) with those using
octyltelluro-pyridinols in the presence of 1 mM NAC. By
using Tinh = 420 min for compound 16a, one could calculate
the catalytic efficiency as 0.42, which is just slightly higher
than that recorded under homogeneous phase conditions.
This specific aspect would be of relevance not only to under-
stand the mechanism of regeneration but also to balance the
antioxidant versus pro-oxidant (thiol depletion) behavior in
biological systems.
FIGURE 5. Oxygen consumption recorded during the autoxida-
tion of styrene without inhibitors and after the injection of the
organotellurium pyridinols (AH, 1.0 ꢀ 10-4 M, first arrow),
followed by the injection of LipCys (9.5 ꢀ 10-5 M, second arrow).
The pyridinols were 8a (a), 24 (b), and 7a (c).
and these species are readily reduced by thiol to regenerate
the parent telluride. This facile redox cycling allows for their
use as peroxide decomposers. In fact, they mimic the action
of the selenium-containing glutathione peroxidase enzymes
(GPx). To compare the efficiency of this catalysis, we
measured the initial rate of reduction of hydrogen peroxide
(ν0) by monitoring the formation of diphenyl disulfide from
thiophenol by UV spectroscopy at 305 nm as described
Catalytic Decomposition of Hydroperoxides. From our
inhibition studies in homogeneous phase and in the two-
phase system it is clear that pyridinols carrying alkyltelluro
groups are readily oxidized to the corresponding telluroxides
(26) Das, T. N.; Neta, P. J. Phys. Chem. A 1998, 102, 7081–7085.
J. Org. Chem. Vol. 75, No. 3, 2010 723

Kumar et al.
JOCArticle
TABLE 4. Thiol Peroxidase Activity of Tellurium-Containing 3-Pyr-
idinols
SCHEME 4. Reactions Involved in the Antioxidant Behaviour
of Alkyltelluro-pyridinols in the Absence of NAC, LipCys, or
Other Thiol Co-antioxidants
thiol peroxidase
activitya (μM min-1
)
antioxidant
7a
8a
9a
10a
16a
18a
4.8 ( 1.1
7.8 ( 1.0
6.5 ( 0.5
1.5 ( 0.1
15.4 ( 2.2
17.8 ( 3.0
9.8 ( 1.8
0.7 ( 0.2
21a
diphenyl diselenide
(PhSeSePh)
aInitial rate of hydrogen peroxide reduction in the presence of
thiophenol and antioxidant. Data are mean ( SD for triplicates.
by Tomoda.27 Initial rates recorded for the reduction of
hydrogen peroxide (3.75 mM) in methanol by thiophenol
(1 mM) in the presence of several organotellurium com-
pounds (0.01 mM) as recorded by UV spectroscopy at
305 nm are shown in Table 4. For comparison, the thiol
peroxidase activity of diphenyl diselenide under these con-
ditions was only 0.67 μM min-1. Reduction of hydrogen per-
oxide (or alkyl-hyroperoxides) involves, in the rate-deter-
mining step, nucleophilic attack of tellurium on oxygen.28
Both electronic and steric effects are therefore expected to
influence the catalytic performance. Considering that all
compounds in Table 2 contain an electron-withdrawing
pyridyl group, it is not surprising that the observed thiol
peroxidase activities cannot match those recorded with the
most efficient organotellurium catalysts studied in this sys-
tem (a value of 199 μM min-1 was recorded for the alkyl
4-aminophenyl telluride obtained by replacing oxygen for
tellurium in ethoxyquin).17
The most active pyridinol compound 18, carrying a meth-
oxy group para to tellurium, was roughly an order of
magnitude less active than this compound. However, it is
noteworthy that all tellurium compounds investigated
showed a thiol peroxidase activity higher than that recorded
for diphenyl diselenide, which has often been used as a
standard in measurements of this kind.
Mechanistic Considerations. Pyridinols have been shown
to express their chain-breaking antioxidant activity follow-
ing the well-known mechanism of quenching chain-carrying
peroxyl radicals by HAT or PCET to yield a phenoxyl radical
that would quickly trap a second peroxyl radical, thereby
breaking two oxidation chains (eqs 8 and 9). The rate-
determining inhibition (eq 8) is characterized by a rate
constant kinh that would depend on the BDE of the O-H
moiety and would be decreased by inter- and intramolecular
H-bonding to any hydrogen bond acceptor. However, our
kinetic measurements indicated that pyridinols bearing the
alkyltelluro substituent ortho to the OH are largely more
reactive than their para isomers, despite the identical value
of BDEOH (in water).
dant by a co-antioxidant have been rationalized for phenols,
both in organic solution29 and in heterogeneous aqueous
environments.13,14 Coherently, we suggested it to occur
either by HAT from the thiol to the pyridinoxyl radical
(in chlorobenzene) or via electron transfer from thiol to
phenoxyl radical at or near the aqueous-lipid interphase,
accompanied by transfer of a proton to produce disulfide.
However, no higher efficiency would be justified for ortho
versus para tellurides: clearly both the antioxidant and the
co-antioxidant activities of these compounds must involve a
distinct mechanism for ortho tellurides that is not accessible
to the para isomers.
Our studies with telluroxides suggested that they form
during the inhibited phase of peroxidation both in homo-
geneous phase and in the two-phase system and that they are
effectively reduced to the corresponding tellurides by thiols
present in the systems. Regarding their formation, we sug-
gest it to occur by two distinct pathways: oxidation by
hydroperoxides present in the oxidizable substrate already
at the beginning of autoxidation and oxygen atom transfer
from chain-carrying peroxyl radical during the inhibited
time of the autoxidation (Scheme 4). The former reaction is
mainly responsible for the negligible inhibition time ob-
served in hydroperoxide-rich substrates such as linoleic acid
(vide supra); however, it is also the basis for an additional
mode of antioxidant behavior: the preventive, GPx-like
activity. The second pathway, oxygen atom transfer from
peroxyl radicals, could in fact provide an explanation for the
unusually high antioxidant activity of pyridinols carrying an
alkyltelluro group in the ortho position. Oxygen transfer
from peroxyl radicals to tellurides is a fast reaction known
to proceed with a rate constant as large as 108 M-1 s-1 with
diphenyltelluride.30 We suggest that this reaction competes
Interestingly, they were also more regenerable by NAC
in the two-phase system or by LipCys in chlorobenzene.
Mechanisms leading to regeneration of the parent antioxi-
(27) Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1994, 116, 2557–2561.
(28) Engman, L.; Stern, D.; Pelcman, M. J. Org. Chem. 1994, 59, 1973–
1979.
(29) (a) Amorati, R.; Ferroni, F.; Lucarini, M.; Pedulli, G. F.; Valgimigli,
L. J. Org. Chem. 2002, 67, 9295–9303. (b) Amorati, R.; Ferroni, F.; Pedulli,
G. F; Valgimigli, L. J. Org. Chem. 2003, 68, 9654–9658.
(30) Engman, L.; Persson, J.; Merenyi, G.; Lind, J. Organometallics 1995,
14, 3641–3648.
724 J. Org. Chem. Vol. 75, No. 3, 2010

Kumar et al.
JOCArticle
SCHEME 5. Reactions Involved in the Antioxidant Behaviour
of Alkyltelluro-pyridinols in the Presence of NAC or LipCys as
Co-antioxidants
cage. Escape of the alkoxyl radical into the bulk solution
(path C) would however start a new oxidative chain, thereby
reducing the stoichiometric factor in the absence of thiol co-
antioxidant (Scheme 4) or decreasing the catalytic efficiency
in the presence of the thiol (Scheme 5). The observed
stoichiometric factors n of 0.3-0.5 for the investigated
pyridinols and their catalytic efficiency n(SH) of 0.2-0.4 is
indicative of the relative relevance of path C.
As indicated, path B would not be available for p-alkyl-
telluro-pyridinols, which would react by a combination of
path A (effective antioxidant behavior) and path C (unpro-
ductive consumption of antioxidant and co-antioxidant).
Conclusion
The above results demonstrate that 3-pyridinols suitably
substituted with organyltelluro groups can act as catalysts
not only for the reduction of peroxyl radicals (chain-break-
ing antioxidant activity) but also for decomposition of
hydroperoxides (preventive antioxidant activity) in the pre-
sence of stoichiometric amounts of thiol reducing agents.
Regarding their reaction with peroxyl radicals, compounds
carrying the organyltelluro group ortho to the hydroxyl show
a significantly higher reactivity than expected on the basis of
the usual mechanism involved in peroxyl-radical quenching
by phenols or pyridinols. This has been explained by a novel
peculiar mechanism that deserves further investigation. At
present, the catalytic efficiency in terms of thiol consumption
was lower than unity both in homogeneous solution and in
the two-phase system. However, our ongoing research sug-
gests that it depends on the experimental conditions, and this
issue will be addressed in detail in future work. We feel that
our catalytic multifunctional antioxidants would be useful
tools in the research related to disorders involving free-
radical-mediated damage (after appropriate toxicological
evaluation) and in the development of novel stabilizing
agents for man-made and natural materials during process-
ing and use.
with hydrogen atom transfer from the pyridinol to peroxyl
radicals and might actually be the key difference between
pyridinols carrying the alkyltelluro group ortho and para to
the OH as illustrated in Schemes 4 and 5. In the presence of
LipCys, phenoxyl radicals and telluroxides are reduced to
the parent phenols and tellurides respsctively (Scheme 5).
Whereas path A in Schemes 4 and 5 describes the usual
mechanism for the antioxidant activity of phenol-like com-
pounds (including pyridinols), path B (in competition with
path C) would be available only for pyridinols carrying an
o-alkyltelluro group. Oxidation of Te to TedO would in-
crease the BDEOH by about 10 kcal mol-1 31
significantly
,
reducing the reactivity of the pyridinol; however, the peroxyl
radical would be turned into a much more reactive alkoxyl
radical (BDEROO-H ∼88 kcal mol-1 vs BDERO-H ∼105 kcal
mol-1), able to abstract the phenolic H-atom in the solvent
Acknowledgment. The Swedish Research Council and
MIUR (Rome, Italy) are acknowledged for financial support.
(31) This value was estimated by assuming that the strength of the
intramolecular H-bond between OH and TedO groups is similar to that
involving the SdO group (7-11 kcal mol-1), reported in: Korth, H.-G.; de
Heer, M. I.; Mulder, P. J. Phys. Chem. A 2002, 106, 8779-8789. The SdO
group also increases the BDEOH by ∼0.4 kcal mol-1 because of its electronic
effects: Amorati, R.; Fumo, M. G.; Menichetti, S.; Mugnaini, V.; Pedulli, G.
F. J. Org. Chem. 2006, 71, 6325-6332.
Supporting Information Available: All experimental details
including 1H and 13C NMR spectra of compounds prepared;
results of autoxidation experiments cited in the text. This
material is available free of charge via the Internet at http://
pubs.acs.org.
J. Org. Chem. Vol. 75, No. 3, 2010 725