Long-lasting antioxidant protection: A regenerable BHA analogue

Article abstract of DOI:10.1021/jo101239c

Introduction of an octyltelluro group ortho to the phenolic moiety in 3-tert-butyl-4-hydroxyanisole (BHA) was found to significantly improve the antioxidant characteristics of the material. In contrast to BHA and the corresponding ortho-substituted octylt

Full text of DOI:10.1021/jo101239c

pubs.acs.org/joc
Long-Lasting Antioxidant Protection: A Regenerable BHA Analogue
Henrik Johansson,† David Shanks,† Lars Engman,*,† Riccardo Amorati,‡
Gian Franco Pedulli,‡ and Luca Valgimigli*,‡
†Department of Biochemistry and Organic Chemistry, Uppsala University, Box 576, SE-751 23 Uppsala,
Sweden, and ‡Department of Organic Chemistry “A. Mangini”, University of Bologna, via S. Giacomo 11,
40126 Bologna, Italy
lars.engman@biorg.uu.se; luca.valgimigli@unibo.it
Received June 24, 2010
Introduction of an octyltelluro group ortho to the phenolic moiety in 3-tert-butyl-4-hydroxyanisole
(BHA) was found to significantly improve the antioxidant characteristics of the material. In contrast
to BHA and the corresponding ortho-substituted octylthio- (9c) and octylseleno (9b) derivatives, the
organotellurium 9a was regenerable when assayed for its capacity to inhibit azo-initiated peroxida-
tion of linoleic acid in a chlorobenzene/water two-phase system containing N-acetylcysteine as a
stoichiometric reducing agent, and peroxyl radicals were quenched more efficiently than with
R-tocopherol. In the homogeneous phase, inhibition of styrene autoxidation occurred with a rate
constant kinh as large as 1 ꢀ 107 M-1 s-1 but with a low (n = 0.4) stoichiometric factor. Evans-
Polanij plots of log (kinh) versus BDE(O-H), which are usually linear for phenols with similar steric
crowding reacting by H-atom transfer, revealed that compound 9a was more than 2 orders of
magnitude more reactive than expected. Although further mechanistic investigations are needed, it
seems that the ortho-arrangement of an alkyltelluro group and hydroxyl should be considered a
privileged structure for phenolic antioxidants.
Introduction
3-tert-butyl-4-hydroxyanisole (1; >85%) and 2-tert-butyl-4-
hydroxyanisole (2;<15%).3
Butylated hydroxyanisole (BHA) has been used for more
than 60 years1 as a food additive (E320) to preserve prod-
ucts containing fats and oils2 and as a stabilizer for cos-
metics such as lipsticks and eye shadow. The commercially
available antioxidant is prepared by alkylation (tert-butyl-
ation) of 4-hydroxyanisole and is obtained as a mixture of
(1) IARC monograph on the evaluation of the carcinogenic risks to
humans. Some Naturally Occurring and Synthetic Food Components, Furo-
coumarines and Ultraviolet Radiation; Lyon, France, 1986; Vol. 40, p 123,
ISBN 92832 1240 1.
As a mixture of phenolic compounds, BHA acts as a chain-
breaking donating antioxidant.4 It donates hydrogen atoms
(2) For example, the maximum level in frying oil is 200 mg/kg. European
Parliament and Council Directive No 95/1/EC of 20 February, 1995 on food
additives other than colors and sweeteners.
(3) Verhagen, H.; Schilderman, P. A. E. L.; Kleinjans, J. C. S. Chem. Biol.
Interactions 1991, 80, 109–134.
DOI: 10.1021/jo101239c
r
Published on Web 10/25/2010
J. Org. Chem. 2010, 75, 7535–7541 7535
2010 American Chemical Society

Johansson et al.
JOCArticle
to peroxyl radicals at a rate much higher than chain propa-
gation of lipid peroxidation. Since the resulting phenoxyl
radical is well stabilized by resonance, it does not get in-
volved in further hydrogen abstraction chemistry and per-
oxidation of the lipid is efficiently interrupted.
regenerability was excellent in the two-phase system as well
as under homogeneous phase conditions using a lipid-solu-
ble thiol-reducing agent.13
The effects of BHA in biological systems has been looked
into in some detail.5 At very high doses, forestomach squa-
mous cell carcinomas was observed in rodents.6 However, in
humans there was no significant association with stomach
cancer risk found for normal (ca. 100 μg/day) intake of BHA
in the diet.7 The present acceptable daily intake for BHA was
set by the joint FAO/WHO expert committee to 0-0.5 mg/
kg bodyweight.5b Largely, this value is rarely exceeded,8 and
there is no reason to believe that low-level intake of BHA
should pose any cancer hazard.9
Antioxidants such as vitamins E and C cannot be bio-
synthesized in man. Because of their importance, it is not sur-
prising that nature has evolved a mechanism for the regen-
eration of vitamin E in biological membranes in which it is
the only antioxidant. It is thought that ascorbate is serving as
the stoichiometric reducing agent by transferring a hydrogen
atom to the tocopheroxyl radical before it is converted to
nonradical products.10
We have for some time tried to design and synthesize
regenerable antioxidants which could perform in a similar
catalytic fashion in the presence of suitable stoichiometric
reducing agents. Since -SH is arguably the most abundant
reducing function available in biological systems, we focused
on catalytic antioxidants regenerable by thiols. The primitive
R-tocopherol analogue 3, containing a dihydroselenophene
moiety, performed in a truly catalytic fashion in a two-phase
lipid peroxidation model containing N-acetylcysteine as a
coreductant in the aqueous phase.11 Although compound 3
did not quench peroxyl radicals as efficiently as R-tocopherol (4),
it outperformed the natural antioxidant when it came to
duration of protection. Peroxidation was inhibited as long as
there remained some thiol in the aqueous phase. Ethoxyquin (5)
was similarly reactive toward peroxyl radicals as R-tocopherol
but considerably more regenerable.12 Pyridinols such as
telluride 6 showed still improved antioxidant characteristics.
They were considerably more reactive than R-tocopherol
(kinh=1 ꢀ 107 M-1 s-1 as compared to kinh=3ꢀ106 M-1 s-1
for R-tocopherol for quenching of peroxyl radicals), and
In search for novel regenerable antioxidants, it occurred to
us that introduction of chalcogens into the BHA-scaffold
may improve both the chain-breaking antioxidant capacity
and regenerability of the compound. We have recently dis-
cussed the thermodynamic consequences of such a molec-
ular arrangement on the phenolic O-H bond dissociation
enthalpy.14 In the following, we report on the synthesis
and antioxidant characteristics of BHA-analogues carrying
alkylthio, alkylseleno, and alkyltelluro substituents.
Results and Discussion
Synthesis. To introduce chalcogens ortho to the hydroxyl
group in 3-tert-butyl-4-hydroxyanisole, 2-bromo-6-tert-bu-
tyl-4-methoxyphenol (7) was considered a suitable starting
material. The bromoaromatic was prepared in 81% yield by
bromination of 1 in carbon tetrachloride as previously
described.15 By treatment of phenol 7 with 3 equiv of t-BuLi
in dry THF at -78 °C, a solution of the 2,O-dilithiated
species 8 was formed in situ (Scheme 1).
SCHEME 1. Preparation of Compounds 9a-c
(4) Scott, G. In Free Radicals: Chemistry, Pathology and Medicine; Rice-
Evans, C., Dormandy, T., Eds; Richelieu Press: London, 1988; p 103.
(5) (a) Joint FAO/WHO Expert Committee on Food Additives (JECFA)
Monographs and Evaluations: Butylated Hydroxyanisole (BHA) (WHO
Food Additives Series 15). (b) Joint FAO/WHO Expert Committee on Food
Additives (JECFA) Monographs and Evaluations: Butylated Hydroxyani-
sole (BHA) (WHO Food Additives Series 18).
Addition of di-n-octyl ditelluride, diselenide, and disul-
fide, respectively, afforded the chalcogen containing BHA-
analogues 9a, 9b, and 9c after aqueous workup in yields
ranging from 30 to 79% (Scheme 1). The yield’s variability
on changing the dichalcogen used in otherwise identical
procedures was similarly recorded in our previous experience
with this reaction; however, it did not seem to follow any
clear trend. In order to study the antioxidative properties of a
compound similar to BHA analogue 9a, but lacking the
(6) Williams, G. M.; Iatropoulos, M. J.; Whysner, J. Food Chem. Toxicol.
1999, 37, 1027–1038.
(7) Botterweck, A. A. M.; Verhagen, H.; Goldbohm, R. A.; Kleinjans, J.;
van den Brandt, P. A. Food Chem. Toxicol. 2000, 38, 599–605.
(8) Joint FAO/WHO Expert Committee on Food Additives (JECFA)
Monographs and Evaluations: Butylated Hydroxyanisole (BHA) (WHO
Food Additives Series 42, 1999). ISBN-13 9789241660426.
(9) Soubra, L.; Sarkis, D.; Hilan, C.; Verger, Ph. Reg. Toxicol. Pharma-
col. 2007, 47, 68–77.
€
(13) (a) Kumar, S.; Johansson, H.; Kanda, T.; Engman, L.; Muller, T.;
Jonsson, M.; Pedulli, G. F.; Petrucci, S.; Valgimigli, L. Org. Lett. 2008, 10,
(10) Halliwell, B.; Gutteridge, J. M. C. Free radicals in biology and
medicine, 4th ed.; Oxford University Press: Oxford, 2007; p 168.
(11) (a) Kumar, S.; Johansson, H.; Engman, L.; Valgimigli, L.; Amorati,
R.; Fumo, M. G.; Pedulli, G. F. J. Org. Chem. 2007, 72, 2583–2595.
€
4895–4898. (b) Kumar, S.; Johansson, H.; Kanda, T.; Engman, L.; Muller,
T.; Bergenudd, H.; Jonsson, M.; Pedulli, G. F.; Amorati, R.; Valgimigli, L.
J. Org. Chem. 2010, 75, 716–725.
(14) Amorati, R.; Pedulli, G. F.; Valgimigli, L.; Johansson, H.; Engman,
L. Org. Lett. 2010, 12, 2326–2329.
(15) Tashiro, M.; Yoshiya, H. Heterocycles 1983, 20, 653–660.
€
(b) Malmstrom, J.; Jonsson, M.; Cotgreave, I. A.; Hammarstrom, L.; Sjodin,
M.; Engman, L. J. Am. Chem. Soc. 2001, 123, 3434–3440.
€
€
(12) Kumar, S.; Engman, L.; Valgimigli, L.; Amorati, R.; Fumo, M. G.;
Pedulli, G. F. J. Org. Chem. 2007, 72, 6046–6055.
7536 J. Org. Chem. Vol. 75, No. 22, 2010

Johansson et al.
JOCArticle
capacity to act as a hydrogen atom donor, compound 7 was
O-methylated before introduction of the octyltelluro group.
The alkylation product 10 could only be obtained if
methyl iodide was present when the base (K2CO3) was added
(Scheme 2). Inversion of the order of addition caused a rapid
degradation of the bromophenol. Further treatment of
compound 10 with 2 equiv of t-BuLi and subsequent addi-
tion of finely ground elemental tellurium, followed by air
oxidation, gave the corresponding ditelluride. This was then
reduced with NaBH4 and the tellurolate finally alkylated
with n-octyl bromide to give compound 11 in 33% yield
(Scheme 2).
In the two-phase system, 3-tert-butyl-4-hydroxyanisole
1 inhibited peroxidation for ca. 110 min whether or not
N-acetylcysteine was present (Table 1). It was slightly infer-
ior to R-tocopherol when it came to rate of inhibition (Rinh
=
33 ( 5 μM/h in the presence of NAC). Introduction of
bromine next to the hydroxyl (compound 7) did not affect
regenerability, but the inhibition capacity was much poorer
(Rinh ca. 200 μM/h). The commercially available antioxidant
3,5-di-tert-butyl-4-hydroxyanisole 13 was also not regener-
able in the two-phase system (Tinh = 112 ( 10 min). As
compared with compound 1, the additional tert-butyl group
significantly lowered the antioxidant capacity (Rinh = 93 (
6 μM/h). All chalcogen-containing BHA analogues 9 retarded
peroxidation for 80 min in the absence of N-acetylcysteine,
but they were all much poorer antioxidants than R-tocopherol
(249<Rinh <295 μM/h). In the presence of thiol, the per-
formance of the organoselenium compound 9b and organo-
sulfur compound 9c was essentially unchanged, whereas
organotellurium catalyst 9a clearly outperformed R-tocoph-
erol (Table 1 and Figure 1) both when it came to rate of
SCHEME 2. Preparation of Organotellurium Derivative 11
inhibition (Rinh = 14 ( 2 μM/h) and inhibition time (Tinh
=
In order to extend the investigation to include a phenol
with a p-alkyltelluro group, we also prepared 3,5-di-tert-
butyl-4-hydroxyphenyl 3-phenoxypropyl telluride (12). This
compound was obtained in 90% overall yield from commer-
cially available 2,6-di-tert-butylphenol by reaction with tel-
lurium tetrachloride, sodium borohydride reduction of the
resulting aryltellurium trichloride to a tellurolate, and alky-
lation with 3-phenoxypropyl bromide (Scheme 3).
365 ( 12 min). Since residual linoleic acid hydroperoxide
present in commercial samples of the acid can rapidly
oxidize organotelluriums, the retardation of peroxidation
observed in the absence of thiol (Rinh = 252 and Tinh = 80
min) is likely to be due to the poor antioxidant activity of
the corresponding telluroxide formed initially in the ex-
periment. Compound 11, the O-methylated telluride 9a,
did not inhibit peroxidation at all in the absence of thiol.
NAC is obviously needed to keep the organotellurium
compound in its reduced divalent state. Also, as hydrogen
atom transfer (HAT) is impossible, the antioxidant activ-
ity (Rinh = 19 μM/h; Tinh = 90 min) presumably is due to
electron transfer (ET) from tellurium to peroxyl radical.
The resulting radical cation is then not efficiently regen-
erated by the thiol. Compound 12, carrying an alkyltelluro
group para to the phenolic moiety, turned out to be a good
inhibitor (Rinh = 23 μM/h) of lipid peroxidation in the
presence of NAC in the aqueous phase, but regenerability
was poor (Tinh = 100 min).
SCHEME 3. Preparation of Compound 12
Inhibition Studies in the Two-Phase System. We have
previously designed a two-phase system which allows the
study of antioxidant regeneration by water-soluble coantioxi-
dants during azo-initiated peroxidation of linoleic acid.16 In
the experimental setup, linoleic acid (L-H) and the antiox-
idant (40 μM) to be evaluated were stirred in chlorobenzene
at 42 °C with an aqueous solution of N-acetylcysteine (NAC).
2,20-Azobis(2,4-dimethylvaleronitrile) (AMVN) was used as
an initiator in the organic phase, and the progress of peroxi-
dation (formation of conjugated diene hydroperoxide) (eq 1)
was monitored by HPLC. From the inhibited rate of peroxi-
dation, Rinh, determined by least-squares methods from
absorbance/time plots and the duration of the inhibited
phase, Tinh, determined graphically as the cross-point for the
inhibited and the uninhibited lines, comparison of antioxi-
dant and catalyst efficiency could be done. As a reference
control, R-tocopherol was used. Under the conditions used,
it inhibited peroxidation with a Rinh of 24 ( 2 μM/h. It was
not regenerable by NAC.
AMVN,O2,anti-
L- H
f L- OOH ð1Þ
oxidant in chlorobenzene=N-acetylcycteine in H2O
FIGURE 1. Peroxidation traces (linoleic acid hydroperoxide con-
centrations vs time) recorded using compound 9a and R-tocopherol
(40 μM) as antioxidants in the chlorobenzene layer in the presence of
NAC (1 mM) in the aqueous phase.
€
(16) Vessman, K.; Ekstrom, M.; Berglund, M.; Andersson, C.-M.;
Engman, L. J. Org. Chem. 1995, 60, 4461–4467.
J. Org. Chem. Vol. 75, No. 22, 2010 7537

Johansson et al.
JOCArticle
TABLE 1. Inhibited Rates of Linoleic acid Peroxidation (Rinh) and Inhibition Times for Antioxidants and Reference Compounds Tested in the
Two-Phase Model
aAverage rate of peroxidation during the inhibited phase (uninhibited rate ca. 650 μM h-1). Errors correspond to ( SD for triplicates. Where not
indicated, errors are estimated as (20% from available duplicates. bAverage duration of the inhibited phase of peroxidation. Reactions were monitored
for 400 min. Estimated errors are (7%, unless differently indicated.
R• þ O2 f ROO•
ð3Þ
Inhibition Studies in the Homogeneous Phase. To achieve
a deeper insight into the antioxidant activity of com-
pound 9a we investigated the homogeneous-phase autoxi-
dation of styrene or cumene in chlorobenzene, initiated
by thermal decomposition of 2,20-azobisisobutyronitrile
(AIBN) at 303 K and inhibited by 1, 9a-c, 11, and 12
(eqs 2-7).17
Kp
s
ROO• þ RH
f ROOH þ R•
ð4Þ
ð5Þ
2Kt
sf nonradical products
ROO• þ ROO•
ROO• þ ArOH
Ri
s
initiator
f R•
ð2Þ
kinh
s
f ROOH þ ArO•
ð6Þ
ð7Þ
(17) 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.
ROO• þ ArO• f nonradical products
7538 J. Org. Chem. Vol. 75, No. 22, 2010

Johansson et al.
JOCArticle
FIGURE 2. Typical oxygen uptake traces during homogeneous-phase autoxidation of cumene (7.1 M, panel A) or styrene (4.3 M, panel B) in
chlorobenzene, initiated by AIBN (0.05 M) at 303 K, in the absence of inhibitors (dashed line) and inhibited by 6.3 ꢀ 10-6 M of 9a-c or by 2.5 ꢀ
10-5 M of 9a.
TABLE 2. Rate Constants kinh, Stoichiometric Factors n, and
Bond Dissociation Enthalpy (BDE) of the Phenolic OH Bond for
Compound 14, was also included as its O-H bond dissociation
enthalpy (BDE) had been studied in our previous works.14,18
the Investigated Compoundsa
kinh (M-1 s-1
)
n
BDEb (kcal/mol)
1
9a
9b
9c
11
12
(6.4 ( 0.5) ꢀ 105
(1.0 ( 0.3) ꢀ 107
(1.7 ( 0.2) ꢀ 104
(8.2 ( 0.4) ꢀ 103
<1ꢀ 103
1.8 ( 0.2
0.4 ( 0.1
2.1 ( 0.2
1.9 ( 0.2
80.3
78.9
79.8
80.6
(9 ( 2) ꢀ 103
0.7 ( 0.2
2.0 ( 0.2
2d
78.6
78.1
77.1e
Autoxidations were followed by measuring oxygen consump-
tion (see Figure 2) by a custom-built differential oxygen-uptake
apparatus that has been previously described in detail.19 Rate
constants for reaction of the title compounds with peroxyl radi-
cals (kinh), reported in Table 2, were obtained from the slopes
during the inhibited periods.19 The choice of oxidizable substrate
was based on the reactivity of the antioxidant: less oxidizable
cumene was first employed for all test compounds, and then the
most effective antioxidants were further investigated in styrene,
which due to the higher oxidizability (kp = 41 M-1 s-1) allowed
for chain reaction of length g8 even during the inhibited
period.17,19,20a The number of peroxyl radicals trapped by
each antioxidant molecule (n) is also reported in Table 2. For
phenolic antioxidants, n = 2 is usually observed.17,19 Homo-
geneous-phase autoxidations confirm the high reactivity
of 9a toward peroxyl radicals as compared with the other
chalcogen-containing derivatives 9b and 9c or the reference
BHA (1). Remarkably, compound 9a is also more reactive
than R-tocopherol, the most active natural lipophilic
antioxidant,17 commonly used as a benchmark for evaluat-
ing new inhibitors.20
14
R-tocopherol
(3.0 ( 0.2) ꢀ 104c
3.2 ꢀ 106d
aData are mean ( SD, each measurement in triplicate. bFrom ref 14.
cFrom ref 18 dFrom ref 17. eFrom ref 21.
decrease in the antioxidant activity (as was also observed in the
two-phase system).13b From Figure 2, it is noteworthy that, at
the end of the inhibited period using antioxidant 9a, the rate of
oxygen consumption is still retarded as compared with an
uninhibited autoxidation under identical conditions. This effect,
which depends on the concentration of 9a, is related to residual
antioxidant activity of the products formed after oxidation and/
or detelluration of 9a.13b
Rate constants measured in the present work can be discussed
in light of linear Evans-Polanij relationships between log(kinh
)
and BDE(O-H), as illustrated in Figure 3.22,23 During H-atom
transfer (HAT) from phenols to peroxyl radicals, the energy
of activation is expected to be linearly dependent on the
exothermicity of the reaction,22 that is, inversely propor-
tional to the phenolic BDE(O-H). Figure 3 clearly shows
that such a correlation holds for series of phenols having the
However, in the homogeneous phase, the number of per-
oxyl radicals trapped by 9a was quite low (n = 0.4), as was
also observed for 12 and other Te-containing phenols.13b We
previously explained these low n values by assuming that inhibi-
tion (eqs 6 and 7) is going on in competition with the reaction of
the Te-alkyl moieties with peroxyl radicals to afford the corre-
sponding telluroxides and alkoxyl radicals (RO•),13b which in
turn propagate the oxidative chain. A previous investigation
showed that telluroxide-containing phenols are so unreactive
toward peroxyl radicals that their formation causes a dramatic
(18) Amorati, R.; Fumo, M. G.; Menichetti, S.; Mugnaini, V.; Pedulli,
G. F. J. Org. Chem. 2006, 71, 6325–6332.
(19) 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.
(20) (a) Valgimigli, L.; Brigati, G.; Pedulli, G. F.; DiLabio, G. A.;
Mastragostino, M.; Arbizzani, C.; Pratt, D. A. Chem.;Eur. J. 2003, 9,
4997–5010. (b) Wijtmans, M.; Pratt, D. A.; Valgimigli, L.; DiLabio, G. A.;
Pedulli, G. F.; Porter, N. A. Angew. Chem., Int. Ed. 2003, 42, 4370–4373.
FIGURE 3. Evans-Polanij plot of log(kinh) vs BDE(O-H) for inves-
tigated compounds (b) and some reference phenols:23 (0) 4-X-C6-
H4OH; (4) 4-X-2,6-Me2C6H2OH, (O) 4-X-2,6-t-Bu2-C6H2OH.
J. Org. Chem. Vol. 75, No. 22, 2010 7539

Johansson et al.
JOCArticle
same steric crowding around the OH group. Hence, different
sets of substituents in the 2,6-positions belong to parallel
correlations having different intercept on the y-axis.23
Compound 1 fits on the regression line for unhindered
phenols, indicating that the OH group points away from the
ortho tert-butyl group and thus there is no steric barrier to
the approaching peroxyl radical.
disappears when the substituent is placed in the (conjugated)
para position. Indeed, compound 12, bearing the Te-alkyl
group in the para position, did not show any enhanced anti-
oxidant behavior with respect to the kinh value expected from
the Evans-Polanij correlation (Figure 3). Actually, com-
pound 12, even more so than analogous 14 bearing an S-alkyl
substituent in the para position, has lower reactivity than
that expected for its BDE(OH).28
Derivatives 9b and 9c, on the other hand, have much
smaller kinh values than 1, despite their similar BDE(O-H).
The BDE(O-H) values of 9b and 9c are the result of two
opposing effects: the BDE-lowering electron donation from
the chalcogen atoms and the BDE-rising effect due to for-
mation of an intramolecular hydrogen bond between the
reactive OH and the ortho substituents.14 Since it is known
that intramolecularly H-bonded phenolic groups are still
reactive toward peroxyl radicals,23,24 we envisage two possi-
ble reasons for the low kinh values for 9b and 9c as compared
to 1. First, steric crowding of the ortho chalcogen groups is
not negligible but similar to that of a methyl group (the van
We therefore feel that the good inhibitory activity of 9a
must be somehow related to the transfer of the phenoxylic
H-atom, although the exact mechanism remains to be clarified.
Conclusions
Both in homogeneous and biphasic systems, the tellurium
analogue 9a of BHA showed an extraordinary reactivity
toward peroxyl radicals which is due to the Te-alkyl group
positioned ortho to the phenolic OH. In addition, compound
9a showed excellent catalytic behavior when evaluated in a
two-phase model for lipid peroxidation in the presence of
stoichiometric amounts of N-acetylcysteine as a coantioxi-
dant in the aqueous phase. Although the mechanisms for
quenching of peroxyl radicals and for regeneration by thiol
across a lipid aqueous interphase are presently not fully
understood, it seems clear that the ortho arrangement of
an alkyltelluro group and hydroxyl should be considered a
privileged structure both for phenolic and isosteric pyri-
dinolic13 antioxidants.
˚
der Waals radii are 1.8, 1.9, 2.1, and ∼2 A for S, Se, Te, and
a methyl group respectively).25 Second, stabilization of the
incipient phenoxyl radical during the transition state of
H-atom abstraction may be smaller than expected for geo-
metrical reasons. It was previously noted, based on theore-
tical calculations, that in the phenols 9b and 9c the dihedral
angle between the alkylchalcogeno substituents and the
aromatic ring is approximately 90°, while the corresponding
phenoxyl radicals have a planar geometry.14 Therefore, it is
conceivable that the planar conformation, necessary for full
stabilization of the phenoxyl radical, is not attained during
the TS, with a consequent increase in the barrier for hydro-
gen atom transfer.26
Experimental Section
2-Bromo-6-tert-butyl-4-methoxyphenol (7). Compound 7 pre-
pared as described showed NMR data in good agreement with
literature.15 Purification by column chromatography gave the
title compound in 81% yield.
The high kinh of 9a is therefore surprising, as a kinh of about
3 ꢀ 104 M-1 s-1 would be expected if 9a-c were on the same
Evans-Polanji plot (see the X in Figure 3). A possible ex-
planation may be that this reaction is triggered by an electron
transfer (ET) between 9a and peroxyl radicals, followed by
proton transfer (PT), that is, by an ET-PT mechanism.27 To
test this hypothesis, we performed the same measurements
using a more polar solvent which should stabilize charged
species. The rate constant kinh for 9a measured in MeCN
was (4 ( 1) ꢀ 106 M-1 s-1, is somewhat smaller than
obtained in chlorobenzene, suggesting that ET plays a minor
role in the reaction with peroxyl radicals. In line with this
interpretation is also the observed poor inhibiting activity in
chlorobenzene and MeCN of 11, the methylated analogue of
9a, which should behave as a similarly good electron donor.
Interestingly, the enhanced reactivity toward peroxyl radi-
cals is observed only when the Te-alkyl substituent is ortho-
positioned with respect to the phenol (like in 9a), while it
2-tert-Butyl-4-methoxy-6-(octyltelluro)phenol (9a). Compound
9a was prepared according to the procedure for compound 9c
using di-n-octyl ditelluride instead of di-n-dioctyl disulfide: yield
30%; 1HNMRδ7.19 (d, J= 3.1 Hz, 1H), 6.92 (d, J= 3.1 Hz, 1H),
6.25 (s, 1H), 3.76 (s, 3H), 2.72 (t, J = 7.4 Hz, 2H), 1.71
(m, J = 7.4 Hz, 2H), 1.44-1.19 (several signals, 19H), 0.87
(t, J = 6.9 Hz, 3H); 13C NMR δ 152.7, 150.6, 135.6, 123.1, 117.2,
103.0, 56.0, 35.6, 32.0, 31.8, 31.7, 29.4, 29.3, 29.0, 22.8, 14.3,
10.4. Anal. Calcd for C19H32O2Te: C, 54.33; H, 7.68. Found: C,
54.51; H, 7.74.
2-tert-Butyl-4-methoxy-6-(octylseleno)phenol (9b). Compound
9b was prepared according to the procedure for compound
9c using di-n-octyl diselenide instead of di-n-octyl disulfide:
yield 79%; 1H NMR δ 6.99 (d, J = 3.1 Hz, 1H), 6.89 (d, J =
3.1 Hz, 1H), 6.68 (s, 1H), 3.76 (s, 3H), 2.72 (t, J = 7.4 Hz, 2H),
1.63 (m, J = 7.4 Hz, 2H), 1.43-1.20 (several signals, 19 H), 0.87
(t, J = 6.9 Hz, 3H); 13C NMR δ 152.3, 149.5, 136.7, 118.2, 116.5,
116.5, 56.0, 35.4, 31.9, 30.4, 30.3, 29.7, 29.4, 29.3, 29.2, 22.8,
14.2. Anal. Calcd for C19H32O2Se: C, 61.44; H, 8.68. Found: C,
61.37; H 8.70.
(21) (a) Lucarini, M.; Pedrielli, P.; Pedulli, G. F. J. Org. Chem. 1996, 61,
9259–9263. (b) Mulder, P.; Korth, H. G.; Pratt, D. A.; Di Labio, G. A.;
Valgimigli, L.; Pedulli, G. F.; Ingold, K. U. J. Phys. Chem. A 2005, 109, 2647–
2655.
(22) Foti, M.; Daquino, C.; Mackie, I. D.; DiLabio, G. A.; Ingold, K. U.
J. Org. Chem. 2008, 73, 9270–9282.
2-tert-Butyl-4-methoxy-6-octylthiophenol (9c). To 2-bromo-
6-tert-butyl-4-methoxyphenol (0.3 g, 1.16 mmol) in THF (10 mL)
under N2 at -78 °C was added dropwise t-BuLi (2.1 mL 1.7 M
solution in pentane; 3.6 mmol). After the mixture was stirred for
(23) Amorati, R.; Menichetti, S.; Mileo, E.; Pedulli, G. F.; Viglianisi, C.
Chem.;Eur. J. 2009, 15, 4402–4410.
(24) de Heer, M.; Korth, H.-G.; Mulder, P. J. Org. Chem. 1999, 64, 6969–
6975.
(25) Bondi, A. J. Phys. Chem. 1964, 68, 441–451. Pauling, L.; Carpenter,
D. C. J. Am. Chem. Soc. 1936, 58, 1274–1278.
(26) Amorati, R.; Catarzi, F.; Menichetti, S.; Pedulli, G. F.; Viglianisi, C.
J. Am. Chem. Soc. 2008, 130, 237–244.
(27) Mayer, J. M. Annu. Rev. Phys. Chem. 2004, 55, 363–390.
(28) The fact that kinh of both 12 and 14 is somewhat smaller than
expected from BDE(O-H) values and the steric crowding around the OH
group (see Figure 3) may derive from the fact that also in these compounds
the optimal conformation of the alkylchalcogeno group in the parent phenol
is different from that of the phenoxyl radical (90° and 0°, respectively)14 and
the planar conformation, necessary for the full stabilization of the phenoxyl
radical, is not reached during the TS.26
7540 J. Org. Chem. Vol. 75, No. 22, 2010

Johansson et al.
JOCArticle
60 min at -78 °C, di-n-octyl disulfide (1.04 g, 3.6 mmol) was
added, and the reaction was stirred for an additional 2 h at
room temperature. After the reaction mixture was poured into
NaHCO3 (5% aq), extracted with Et2O, dried over Na2SO4, and
concentrated in vacuo, the residue was purified by column
chromatography using pentane/ethyl acetate (95/5) to give the
crude aryltellurium trichloride, 16.6 g (100%), which was used
without further purification in the next step. Under inert atmo-
sphere, solid sodium borohydride (large excess) was added
portionwise to a solution of the aryltellurium trichloride
(0.88 g, 2.0 mmol) in ethanol (10 mL) until the mixture remained
colorless. Neat 3-phenoxypropyl bromide (0.52 g, 2.4 mmol)
was then added, and the mixture stirred for 16 h. Water was added
to the reaction, and the aqueous phase was extracted with diethyl
ether, washed with brine, dried over MgSO4, and evaporated to
give crude product. This was subjected to column chromatography
(pentane/ethyl acetate 8:2) to give 0.84 g (90%) of the pure title
1
title compound as a yellow oil in 30% yield: H NMR δ 6.90
(s,1H), 6.89 (d, J = 3.0 Hz, 1H), 6.88 (d, J = 3.0 Hz, 1H), 3.76
(s, 3H), 2.69 (t, J = 7.3 Hz, 2H), 1.55 (m, J = 7.3 Hz, 2H),
C
1.44-1.18 (several signals, 19H), 0.88 (t, J = 6.9 Hz, 3H); 13
NMR δ 152.2, 150.0, 137.2, 119.7, 116.4, 116.3, 55.9, 37.2, 35.3,
31.9, 29.7, 29.4, 29.3, 29.2, 28.7, 22.8, 14.2. Anal. Calcd for
C19H32O2S: C, 70.32; H,9.94. Found: C, 70.58; H, 9.90.
2-Bromo-6-tert-butyl-1,4-dimethoxybenzene (10). To a solu-
tion of methyl iodide (0.5 mL, 8.03 mmol) in DMF (7.5 mL) was
added 2-bromo-6-tert-butyl-4-methoxyphenol (0.80 g, 3.1 mmol)
followed by K2CO3 (440 mg, 3.2 mmol), and the reaction was
stirred under N2 for 3 h and poured into water. After extraction
with Et2O, the organic phase was dried with Na2SO4, filtered,
and evaporated in vacuo to give the title compound as a yellow
oil in a quantitative yield: 1H NMR δ 6.95 (d, J = 2.9 Hz, 1H),
6.84, (d, J = 2.9 Hz, 1H), 3.88 (s, 3H), 3.76 (s, 3H), 1.38 (s, 9H);
13C NMR δ 155.4, 150.8, 145.9, 118.2, 115.5, 113.9, 61.7, 55.8,
35.8, 30.9; EI-MS 272.01/273.97 [Mþ].
1
compound as beige crystals: mp 62-64 °C; H NMR δ 7.57 (s,
2H), 7.27 (m, 2H), 6.94 (m, 1H), 6.87 (m, 2H), 5.25 (bs, 1H), 4.01 (t,
J = 6.0 Hz, 2H), 3.00 (t, J = 7.4 Hz, 2H), 2.29 (m, 2H), 1.43 (s,
18H); 13C NMR δ 159.0, 154.3, 137.2, 136.5, 129.6, 120.8, 114.7,
100.6, 68.9, 34.4, 31.6, 30.4, 4.5. Anal. Calcd for C23H32O2Te: C,
59.02; H, 6.89. Found: C, 58.83; H, 6.86.
HPLC Peroxidation Assay. In the experimental setup, linoleic
acid, and the antioxidant to be evaluated were vigorously stirred
in chlorobenzene at 42 °C with an aqueous solution of N-acetyl-
cysteine (NAC). 2,20-Azobis(2,4-dimethylvaleronitrile) (AMVN)
was added as an initiator in the organic phase and the progress
of peroxidation monitored by HPLC (conjugated diene hydro-
peroxide formation). For comparison of catalyst efficiency,
2-tert-Butyl-1,4-dimethoxy-6-(octyltelluro)benzene (11). To
2-bromo-6-tert-butyl-1,4-dimethoxybenzene (500 mg, 1.8 mmol)
in THF (15 mL) was added dropwise t-BuLi (2.1 mL 1.7 M in
pentane; 3.7 mmol) at -78 °C. The solution was stirred for
30 min at this temperature, and finely ground elemental tel-
lurium (750 mg, 5.9 mmol) was added. The cooling bath was then
removed and stirring continued for an additional 3 h at room
temperature. The reaction was then poured onto crushed ice and
kept in the open air overnight. After evaporation of the solvent
in vacuo, the residue was dissolved in EtOH (10 mL) and fil-
tered. To the solution of crude bis-(3-tert-butyl-2,5-dimethoxy-
phenyl)ditelluride under N2, NaBH4 (0.18 g, 5 mmol) was added
in portions until the reaction mixture turned colorless. Octyl
bromide (1.0 g, 5 mmol) was added and stirring continued for
1 h when the contents of the reaction flask was poured into
brine and extracted with Et2O (15 mL ꢀ 3). After drying over
Na2SO4 and evaporation in vacuo, the product was purified by
column chromatography using pentane/ethyl acetate (95/5) as
an eluent. The title compound was obtained as a yellow oil (259
mg, 33%): 1H NMR δ 6.90 (d, J = 3.0, 1H), 6.80 (d, J = 3.0,
1H), 3.80 (s, 3H), 3.77 (s, 3H), 2.90 (t, J = 7.4, 2H), 1.82 (m, J =
7.4, 2H), 1.45-1.20 (several signals, 19H), 0.87 (t, J = 6.7, 3H);
13C NMR δ 155.8, 154.8, 143.5, 118.3, 113.6, 110.7, 62.1,
55.7, 35.5, 32.2, 32.0, 31.6, 31.2, 29.3, 29.1, 22.8, 14.2, 7.8. Anal.
Calcd for C20H34O2Te: C, 55.34; H, 7.89. Found: C, 55.61; H,
7.80.
inhibition times (Tinh) and inhibited rates of peroxidation, Rinh,
were determined by least-squares methods from absorbance/
time plots as previously described.11b,29
Homogenous Phase Autoxidations. Kinetic measurements
with peroxyl radicals were performed by studying the inhibited
autoxidation of styrene or cumene in chlorobenzene or MeCN
at 303 K, initiated by AIBN (0.05 M), in the presence of variable
amounts ((1-20) ꢀ 10-6 M) of the investigated phenols and of
2,2,5,7,8-pentamethyl-6-chromanol (PMHC), a synthetic ana-
logue of R-tocopherol, as reference antioxidants. The autoxida-
tion was followed by monitoring the oxygen consumption in an
oxygen uptake apparatus built in our laboratory and based on a
differential pressure transducer, which have been previously
described.19 The rate of initiation Ri was measured in a prelimi-
nary set of experiments from the length of the inhibition period
T
T
inh, using PMHC as a reference antioxidant: Ri = 2[PMHC]/
inh. Integration of the oxygen consumption trace afforded
the rate of reaction with peroxyl radicals kinh, according to the
equation Δ[O2]t = -kp/kinh [RH] ln(1 - t/Tinh). The kp values
of styrene and cumene at 303° are 41 and 0.32 M-1 s-1
,
respectively.19
Acknowledgment. Financial support by MIUR (Rome),
the University of Bologna, and the Swedish Research Coun-
cil is gratefully acknowledged.
3,5-Di-tert-butyl-4-hydroxyphenyl 3-Phenoxypropyl Telluride
(12). A solution of 2,6-di-tert-butylphenol (20.6 g, 100 mmol) in
CCl4 (20 mL) was slowly added to a slurry of finely crushed
TeCl4 (10.2 g, 38 mmol) in CCl4 (10 mL) at 0 °C. The mixture
was brought to room temperature and stirred for a further 2 h.
The formed precipitate was then collected by filtration, washed
with dichloromethane, and dried under vacuum to give the
Supporting Information Available: General experimental
details and 1H and 13C NMR spectra of compounds prepared.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(29) 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. 75, No. 22, 2010 7541