Absolute rate constants and transient intermediates in the free-radical-induced peroxidation of γ-terpinene, an unusual hydrocarbon antioxidant

Article abstract of DOI:10.1039/b305631g

The peroxidation of γ-terpinene (TH) initiated by tert-butoxyl radicals has been investigated by means of nanosecond laser flash photolysis techniques in acetonitrile at room temperature. The absolute rate constants for the reactions implicated in the peroxidation process have been determined and the main short-lived intermediates, terpenyl and terpenylperoxyl radicals, have been detected. The rate constant for H-atom abstraction from TH by tert-butoxyl radicals, K₃, has been found to be 1.4 × 10⁸ M ^-1 s^-1 while the values for the self-quenching of T? radicals, 2k₄, and the addition of O₂ to T? k ₉, are, respectively, 1.2 × 10¹⁰ M^-1 s^-1 and 1.3 × 10⁹ M^-1 s^-1. The overall results strongly substantiate the peroxidation mechanism recently proposed by Foti and Ingold, J. Agric. Food Chem., 2003, 51, 2758, and provide a solid basis for a better understanding of the unusual antioxidant activity exhibited by this hydrocarbon.

Full text of DOI:10.1039/b305631g

View Article Online / Journal Homepage / Table of Contents for this issue
Absolute rate constants and transient intermediates in the
free-radical-induced peroxidation of c-terpinene, an unusual
hydrocarbon antioxidanty
a
Salvatore Sortino,* Salvatore Petralia and Mario C. Foti*
a
b
a
Dipartimento di Scienze Chimiche, Universit a` degli Studi di Catania, Viale Andrea Doria 8,
5125, Catania, Italy. E-mail: ssortino@mbox.unict.it
Istituto di Chimica Biomolecolare del CNR – Sez. di Catania, Via del Santuario 110, I-95028,
Valverde (CT), Italy. E-mail: foti@issn.ct.cnr.it
9
b
Received (in Montpellier, France) 20th May 2003, Accepted 27th June 2003
First published as an Advance Article on the web 2nd September 2003
The peroxidation of g-terpinene (TH) initiated by tert-butoxyl radicals has been investigated by means of
nanosecond laser flash photolysis techniques in acetonitrile at room temperature. The absolute rate constants
for the reactions implicated in the peroxidation process have been determined and the main short-lived
intermediates, terpenyl and terpenylperoxyl radicals, have been detected. The rate constant for H-atom
8
ꢁ1 ꢁ1
abstraction from TH by tert-butoxyl radicals, k , has been found to be 1.4 ꢀ 10 M
s
while the values for
10
3
ꢂ
ꢂ
ꢁ1 ꢁ1
the self-quenching of T radicals, 2k
9
4
, and the addition of O
2
to T , k
9
, are, respectively, 1.2 ꢀ 10
M s and
s . The overall results strongly substantiate the peroxidation mechanism recently proposed by
ꢁ1 ꢁ1
1
Foti and Ingold, J. Agric. Food Chem., 2003, 51, 2758, and provide a solid basis for a better understanding of
.3 ꢀ 10 M
the unusual antioxidant activity exhibited by this hydrocarbon.
ꢂ
tion of two LOO ’s. Therefore, in the presence of TH, the
steady-state concentration of LOO is lower and (conse-
Introduction
ꢂ
g-Terpinene (TH) is a monoterpene hydrocarbon present in
variable amounts in the essential oils (EOs) of many medicinal
quently) the rate of peroxidation of LH decreases.
This scenario is quite unusual because: 1) only very rarely
11
have peroxidizable hydrocarbons such as triphenylmethane,
1
2
and aromatic plants which has recently come to the limelight
,3
4
because of its unexpected antioxidant properties. A recent
12
13
b-carotene and tetralin been shown to have antioxidant
activities; 2) the observed antioxidant mechanism for TH is
completely different from the usual mechanism of action of
5
investigation, in fact, demonstrated that the antioxidant activ-
6
ity of EOs is due not only to their contents of phenols (which
1
7
are well-known inhibitors of lipid peroxidation ) but, surpris-
7a
phenols, such as vitamin E.
ingly, also to the activity of simple hydrocarbons like g-
terpinene. EOs are also known to have antibacterial and
The present contribution aims to provide an insight into the
free-radical induced peroxidation of TH by means of laser
flash photolysis (LFP) techniques. The detection of some
short-lived intermediates appearing in Scheme 2 and the deter-
mination of a few relevant absolute rate constants can indeed
substantiate the mechanism and provide a solid basis for a bet-
ter understanding of the unusual antioxidant activity exhibited
by this hydrocarbon.
8
antimycotic activities which, together with the antioxidant
properties, make them appealing candidates as preservatives
for food and cosmetics, at least where ‘‘their use is not in con-
trast with their aroma’’.
5
These unexpected properties of TH spurred some of us to
investigate the kinetics of its autoxidation and the mechanism
by which it retarded the thermal peroxidation of linoleic
4
acid (LH). The main results of this study are therefore briefly
summarized below for sake of clarity.
Experimental section
9
g-Terpinene in the presence of free radicals and molecular
oxygen yields p-cymene (Cy) as the only organic product
Chemicals
(Scheme 1) in a radical-chain reaction. This reaction resembles
the peroxidation of 1,4-cyclohexadiene which yields benzene in
a chain reaction having HOO as the chain-carrying species.
g-Terpinene ( > 99%) was purchased from Fluka and used as
received. Diphenylmethanol and di-tert-butyl peroxide were
purchased from Aldrich and this latter was passed through alu-
mina before use. Acetonitrile was of spectrophotometric grade.
ꢂ
10
It is safe, therefore, to assume that the peroxidation of TH
involves similar steps and has the same chain carrier (see
Scheme 2).
ꢂ
The hydroperoxyl radical, HOO , plays a key role in the
antioxidant activity of TH because, on mechanistic grounds,
4
ꢂ
it has been suggested that the cross-termination between HOO
and LOO (the chain carrier for the peroxidation of LH) might
be some 2–3 orders of magnitude faster than the self-termina-
ꢂ
y Dedicated to Professor Giuseppe Condorelli on the occasion of his
rd
3
7
birthday.
Scheme 1
DOI: 10.1039/b305631g
New J. Chem., 2003, 27, 1563–1567
1563
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
LFP experiments in the absence of oxygen
ꢂ
The photogenerated t-butoxyl radicals, t-BuO , were expected
both to react with TH (by H-atom abstraction) and decay with
first (b-cleavage) or pseudo-first order (reaction with the sol-
1
5
vent) processes, eqn. 2. The experiments were therefore car-
ried out at relatively high concentrations of TH (0.02 M) so
ꢂ
that the loss of t-BuO was minimized. Under such conditions,
most t-BuO radicals abstracted a hydrogen atom from TH
ꢂ
ꢂ
and produced the corresponding terpenyl radicals, T , eqn. 3
(
vide infra).
k2
ꢂ
t ꢁ BuO ƒƒƒ! decay
ð2Þ
ð3Þ
k3
ꢂ
ꢂ
t ꢁ BuO þ TH ƒƒƒ! T þ t ꢁ BuOH
Fig. 1 shows the transient absorption spectrum taken 2 ms after
the laser pulse. The spectrum is characterized by a maximum in
the 315–325 nm region and by a clear drop of the absorption at
around 350 nm. Apart from a small red shift, this spectrum is
very similar to the reported spectra of cyclohexadienyl and
Scheme 2
1
6,17
methylcyclohexadienyl radicals.
attribute the transient spectrum of Fig. 1 to the terpenyl radi-
Therefore, we could safely
ꢂ
cal, T , generated by eqn. 3.
Consistent with eqns. 1–3, we observed that the growth
traces (following excitation) of T radicals followed pseudo-
first order kinetics, see Fig. 2A. Furthermore, the decay of
T radicals followed remarkably clean second-order kinetics,
Nanosecond laser flash photolysis setup
ꢂ
The samples were excited with either the fourth or the third
harmonic of a Nd-YAG. Continuum Surelite II-10 laser sys-
tem (pulse width 6 ns FWHM) and the excited solution were
analyzed at a right angle geometry using a mini mLFP-111
apparatus developed by Luzchem Research. The monitoring
beam was supplied by a ceramic xenon lamp and delivered
through quartz fiber optical cables. The laser pulse was probed
by a fiber that synchronized the mLFP system with a Tektro-
nix TDS 3032 digitizer operating in the pre-trigger mode. The
signals from a compact Hamamatsu photomultiplier were
initially captured by the digitizer and then transferred to a per-
sonal computer that controlled the experiments with a Luz-
chem software developed, in the LabView 5.1 environment,
from National Instruments. The energy of laser pulse was mea-
sured at each laser shot by a SPHD25 Scientech pyroelectric
energy monitor. Acetonitrile solutions for the experiments in
the absence of oxygen were deoxygenated by bubbling with a
vigorous and constant flux of pure argon (previously saturated
with acetonitrile). The experiments at variable oxygen con-
centration were carried out by using mixtures of oxygen and
nitrogen prepared with a flow gas controller-mixer system
ꢂ
in the whole spectral range explored, in agreement with the
self-quenching 4 (see Fig. 2B). No new transient species was
detected as T decayed.
ꢂ
k4
ꢂ
ꢂ
T þ T ƒƒƒ! non radical products
ð4Þ
The second-order fit of the trace reported in Fig. 2B gave a
6
ꢁ1
value of 2k
of the terpenyl and cyclohexadienyl radicals, we may use the
4
/e ¼ 2.9 ꢀ 10 cm s . In view of the similarities
ꢁ1
ꢁ1
18
value of emax ¼ 4400 M cm for cyclohexadienyl radical
10
ꢁ1 ꢁ1
s for the self-
to estimate a value of 2k ꢄ 1.2 ꢀ 10
M
4
ꢂ
termination of T radicals. Since this rate constant is close to
the diffusion-limit, we can therefore assert that the alkyl
substituents do not significantly hinder eqn. 4.
In order to determine the absolute rate constant k we used a
3
low laser pulse energy generating therefore low initial concen-
trations of t-BuO radicals. Under such conditions, the loss of
radicals by the reverse of reaction 1 was negligible and the
ꢂ
ꢂ
observed rate constant, kobs , for the growth of T was thereby
related to the concentration of TH via the simple linear eqn. 5:
(
accuracy, ꢃ0.1%). In all the experiments the solutions were
renewed (in a flow cell of 1 cm optical path) after each laser
shot. The sample temperature was 295 ꢃ 2 K.
kobs ¼ k þ k ½THꢅ
2
3
ð5Þ
GC-MS analyses were done on a Hewlett-Packard 5890
interfaced to a Hewlett-Packard 5971A Mass Selective Detec-
tor (DB-5 capillary columns. 30 m ꢀ 0.25 mm, film thickness
0
.25 mm).
Results and discussion
The peroxidation of TH was studied in acetonitrile at room
temperature and initiated by tert-butoxyl radicals generated
in situ by fast photolysis of di-tert-butyl peroxide
1
4
hn
ꢂ
t ꢁ BuOOBu ꢁ t ƒƒƒ! 2t ꢁ BuO
ð1Þ
Unless otherwise stated, the fourth harmonic of a Nd-YAG
laser (l ¼ 266 nm, ꢄ5 mJ) was used as excitation wavelength.
At this wavelength, TH did not show any significant absorp-
tion (at the concentrations used in the experiments). This
offered the advantage of requiring a relatively low concentra-
tion of di-tert-butyl peroxide (A266 ꢄ 0.5) which allowed
exploring the UV-spectral changes down to ca. 270 nm (other-
wise precluded by its absorption).
Fig. 1 Transient absorption spectrum taken 2 ms after a 266 nm
laser excitation of an Ar-saturated solution of di-tert-butyl peroxide
in acetonitrile containing 0.02 M TH.
1
564
New J. Chem., 2003, 27, 1563–1567

View Article Online
In the presence of a second substrate (in our case, TH) which
reacts in competition for t-BuO radicals, the signal of the
ketyl radical becomes weaker but its growth faster. At constant
ꢂ
2
Ph CHOH concentration, the observed rate constant for the
growth of the ketyl radical is therefore expected to be depen-
dent upon [TH], eqn. 7. Except for the intercept,
k
to eqn. 5, (see eqn. 8).
0
¼ k
2 6 2
+ k [Ph CHOH], this equation is formally identical
kobs ¼ k2 þ k6½Ph2CHOHꢅ þ k3½THꢅ
kobs ¼ k0 þ k3½THꢅ
Because of the strong absorption of Ph
ð7Þ
ð8Þ
2
CHOH at 266 nm, this
time we used the third harmonic of a Nd-YAG laser (l ¼ 355
nm, ꢄ5 mJ, absorbance of di-tert-butyl peroxide at 355 ca. 0.5)
as excitation light source. The results of these experiments are
shown in Fig. 4 for two different concentrations of Ph
As expected, the two straight lines are parallel (within the [TH]
2
CHOH.
8
ꢁ1 ꢁ1
s in
explored) and afford a value of k
3
¼ 1.1 ꢀ 10
M
excellent agreement with that obtained by a direct detection
of the terpenyl radical.
It is worth observing that the value of k is ca. 2-fold larger
3
than the corresponding value reported for the H-atom abstrac-
ꢂ
22
tion from 1,4-cyclohexadiene by t-BuO radicals. This ratio is
in pretty good agreement with the previously reported ratio for
the H-atom abstraction from TH and 1,4-cyclohexadiene by
ꢂ
4
linoleoylperoxyl radicals, LOO . This greater hydrogen
donating ability of TH was proposed to be at the basis of its
better antioxidant activity in comparison with that of 1,4-
4
cyclohexadiene. That the order of reactivity of these two
hydrocarbons is independent of the abstracting radical is
further proof that it can only be attributed to a better stabiliza-
tion of the terpenyl radical by the two alkyl substituents, i.e.
the C–H bond dissociation enthalpy for the ring methylene
groups is slightly lower in TH than in 1,4-cyclohexadiene.
Fig. 2 Representative experimental traces showing (A) the buildup
and (B) the decay of the terpenyl radical monitored at 320 nm. They
were monitored at different time-scales after a 266 nm laser excitation
of an Ar-saturated solution of di-tert-butyl peroxide in acetonitrile
containing 0.02 M TH.
LFP experiments in the presence of oxygen
Fig. 3 reports a typical plot kobs versus [TH] which is in excel-
lent agreement with eqn. 5. The slope of the straight-line yields
ꢂ
In the presence of oxygen, the radical T is quickly oxidized to
the corresponding peroxyl radicals (vide infra):
8
ꢁ1 ꢁ1
s .
The value so obtained for k has also been confirmed with
a value of k
3
¼ 1.4 ꢀ 10 M
3
k9
ꢂ
ꢂ
an alternative kinetic approach which makes use of diphenyl-
methanol (Ph CHOH). This method was introduced for the
T þ O ƒƒƒ! TOO
ð9Þ
The decay of T at variable concentrations of oxygen afforded
2
2
ꢂ
first time by Scaiano in the late seventies and it has been
proved to give good results for a large variety of substrates,
especially for those which do not produce easily detectable
the value k . We used a relatively high concentration of TH
9
ꢂ
(
0.06 M) at which the generation of T occurred in ca. 100
1
4,19
ns (ꢄ1/k [TH]) and relatively low O concentrations (ꢆ1.8
3
2
radicals.
from Ph
radical, eqn. 6, which can easily be detected at 545 nm due
tert-Butoxyl radicals quickly abstract the H-atom
CHOH and produce the resonance-stabilized ketyl
2
0
2
2
1
to its typical and intense absorption.
k6
ꢂ
_
t ꢁ BuO þ Ph2CHOH ƒƒƒ! t ꢁ BuOH þ Ph2COH
ð6Þ
Fig. 4 Observed pseudo-first-order rate constants for the buildup of
the ketyl radical as a function of the concentration of TH for two dif-
ferent concentrations of diphenylmethanol: (S) 0.1 M; (G) 0.2 M.
Each point was obtained by signal average of 15 traces. The inset
shows the buildup of the ketyl radical at 545 nm observed after
355 nm laser excitation of an Ar-saturated solution of di-tert-butyl
peroxide in acetonitrile containing 0.2 M diphenylmethanol.
Fig. 3 Plot of the experimental pseudo-first order rate constant for
the formation of T as a function of the concentration of TH. Each
point was obtained by signal average of 15 traces.
ꢂ
New J. Chem., 2003, 27, 1563–1567
1565

View Article Online
ꢂ
mM) at which the lifetime of T was substantially longer than
1
tion was always sufficient to ‘‘capture’’ all the T radicals (eqn.
00 ns (ca. 1 ms). Nevertheless, the quantity of oxygen in solu-
ꢂ
9
). In other words, under our experimental conditions, the loss
ꢂ
ꢂ
of T by eqn. 4 was negligible. Consequently, the decay of T
radicals can simply be described by the following equation:
0
kobs ¼ k 0 þ k9½O2ꢅ
ð10Þ
Fig. 5 shows a typical plot of kobs vs. [O ] together with a repre-
2
ꢂ
sentative trace for the decay of T radicals (see inset). The slope
of the straight-line yields a value of k ¼ 1.3 ꢀ 10 M s in
excellent agreement with the expected rate constants for the
addition of oxygen to carbon-centered radicals.
The low value of the intercept of the plot in Fig. 5 demon-
strates that apart from the reaction with oxygen, all the other
possible first-order processes, e.g. the reverse of eqn. 9
see Scheme 2 step ii), are (at room temperature) relatively
9
ꢁ1 ꢁ1
9
2
3
2
3b
Fig. 6 Transient absorption spectra taken (S) 0.1 ms and (X) 1 ms
(
unimportant.
3
after 266 nm laser excitation of an air-equilibrated CH CN solution
of di-tert-butyl peroxide in the presence of 0.06 M TH. The inset shows
(
tively.
S) the decay and (X) the buildup observed at 320 and 270 nm, respec-
We went further and investigated the spectral evolution of
eqn. 9. Fig. 6 shows, for instance, the time evolution of the
transient absorbance observed after 266 nm laser excitation
of di-tert-butyl peroxide in the presence of TH in air-equili-
brated solution. It can easily be noted that a new absorption
which extends to below the 270 nm cut off appears after the
double bonds (corresponding in our case to 1, 2, 4, 5) could
abstract the adjacent H-atom and eliminate HOO to yield
benzene (in our case, Cy, see Scheme 2 step iii). On the con-
trary, the peroxyl radical with non-conjugated double bonds
ꢂ
ꢂ
decay of the T radicals.
As shown in the inset of Fig. 6, the decay of T monitored at
ꢂ
24–27
(3 and 6 in our case) led to oxygenated products.
With
3
tored at 270 nm. This clearly indicates that the decaying tran-
20 nm matches well the growth of the new transient moni-
TH, the GC-MS analysis of a photolyzed solution did not
reveal any significant presence of oxygenated products, under
the peroxidation conditions employed, and Cy was the only
ꢂ
sient, T , is the precursor of the grow-in. On this basis it is
quite reasonable to assign this new transient absorption to
2
7
detected organic product. In addition, Porter and co-work-
ꢂ
28–30
the peroxyl radicals TOO generated via eqn. 9. The similar
absorption spectrum exhibited by peroxyl radicals produced
ers
conjugated diene moieties, like 3 and 6, undergo extremely
have recently shown that peroxyl radicals with non-
2
4
upon pulse radiolysis of 1,4-cyclohexadiene validates further
our assignment.
The formation of very unstable peroxyl radicals, TOO ,
rapid reversion to their carbon-centered radicals (in our case
T ) and O (eqn. 9). For all these reasons, we exclude, there-
ꢂ
2
ꢂ
fore, the existence of the radicals 3 and 6.
As shown in Fig. 7, the decay of TOO monitored at 270 nm
followed quite a clean first order kinetics with a rate constant
ꢂ
ꢂ
which rapidly eliminate hydroperoxyl radicals, HOO (see
Scheme 2 step iii) is the keystone of the antioxidant mechanism
4
of TH. Since t-BuO radicals can abstract an H-atom from
ꢂ
4
ꢁ1
k ꢄ 5 ꢀ 10 s . This value is considerably smaller than that
both the ring methylene groups with formation of two distinct
T radicals, the addition of oxygen could lead (in principle) to
reported by von Sonntag and coworkers for the elimination
of HOO radicals from cyclohexadienylperoxyl radical in
ꢂ
ꢂ
5
ꢁ1 24
all the peroxyl radicals 1–6 shown in Scheme 3.
Although all these radicals might potentially be responsible
for the transient absorption observed (see Fig. 6), we believe
that 2 and 5 (in roughly 1 : 1 ratio) are the most likely candi-
dates, as discussed below.
water (k ꢇ 8 ꢀ 10 s ). Apart from possible solvent effects
on these reactions, another explanation for this discrepancy
might be found by taking into account the structures of the
peroxyl radicals involved. Actually, the radicals 1 and 4 do
not show any significant difference with respect to cyclohexa-
dienylperoxyl radical and, therefore, the intramolecular H-
atom abstraction is expected to occur much at the same rate.
On the other hand, the radicals 2 and 5 are tertiary alkylper-
oxyl radicals whereas the cyclohexadienylperoxyl radical (as
1
0a
24
Ingold
first and von Sonntag later proposed and
demonstrated, respectively, that in the peroxidation of 1,4-
2
cyclohexadiene, only the peroxyl radical with conjugated
3
ꢂ
Fig. 5 Determination of the rate constant for oxygen addition to T
radical, reaction 9, according to eqn. 10 ([TH] ¼ 0.06 M). Each
point was obtained by signal average of 15 traces. The inset shows a
] ¼ 1.8 mM.
ꢂ
representative decay trace of T for [O
2
Scheme 3
1
566
New J. Chem., 2003, 27, 1563–1567

View Article Online
gratefully acknowledge Dr. Keith Ingold (National Research
Council, Ottawa, Canada) for his helpful suggestions.
References
1
Essential oils are complex mixtures containing mono- and sesqui-
terpenes, oxygenated terpenes, a few free phenols and other com-
pounds, see for instance refs. 2 and 3.
2
V. Formacek and K. H. Kubeczka, in Essential oils analysis by
capillary gas chromatography and carbon-13 spectroscopy, Wiley,
New York, 1985.
3
4
F. Senatore, in Oli Essenziali, EMSI, Roma, 2000.
M. C. Foti and K. U. Ingold, J. Agric. Food Chem., 2003, 51,
2
758.
Fig. 7 Decay trace observed at 270 nm after 266 nm laser excitation
of an air-equilibrated solution of di-tert-butyl peroxide in CH CN con-
3
taining 0.06 M TH.
5
6
G. Ruberto and M. T. Baratta, Food Chem., 2000, 69, 167.
(a) H. J. D. Dorman, S. G. Deans, R. C. Noble and P. Surai, J.
Essent. Oil Res., 1995, 7, 645; (b) K. P. Svoboda and S. G. Deans,
Flav. Fragr. J., 1992, 7, 81; (c) M. Charai, M. Faid and A.
Chaouch, J. Essent. Oil Res., 1999, 11, 517; (d ) K. A. Youdim,
H. J. D. Dorman and S. G. Deans, J. Essent. Oil Res., 1999, 11,
643; (e) H. J. D. Dorman, P. Surai and S. G. Deans, J. Essent.
Oil Res., 2000, 12, 241.
(a) G. W. Burton and K. U. Ingold, J. Am. Chem. Soc., 1981, 103,
6472; (b) G. W. Burton, T. Doba, E. J. Gabe, L. Hughes, F. L.
Lee, L. Prasad and K. U. Ingold, J. Am. Chem. Soc., 1985, 107,
well as 1 and 4) is a sec-alkylperoxyl radical. It has been
demonstrated that tert-alkylperoxyls are roughly one order
of magnitude less reactive than sec-alkylperoxyls in intermole-
7
3
1
cular hydrogen atom abstraction from organic substrates.
We therefore attribute the absorption at 270 nm and its slower
ꢂ
7
1
053; (c) G. W. Burton and K. U. Ingold, Acc. Chem. Res.,
986, 19, 194.
decay to the radicals 2 and 5 which eliminate the HOO radical
to a comparatively slower rate than 1 and 4. This assumption
does not exclude, in fact, the co-formation of the radicals 1 and
8
9
0
(a) Q. Delespaul, V. G. de Billerbeck, C. G. Roques, G. Michel, C.
Marquier-Vinuales and J. M. Bessiere, J. Essent. Oil Res., 2000,
12, 256 and cited references; (b) S. G. Deans and K. P. Svoboda,
Flav. Fragr. J., 1990, 5, 187.
4. Their decay, according to ref. 24, would be as fast as their
formation preventing therefore their spectral detection.
Free carbon radicals were generated in organic solutions by
0
thermolysis of 2,2 -azobis(isobutyronitrile) (AIBN). These initial
carbon radicals, R , quickly reacted with oxygen dissolved in
ꢂ
Conclusions
ꢂ
solution and produced peroxyl radicals, ROO .
(a) J. A. Howard and K. U. Ingold, Can. J. Chem., 1967, 45, 785;
1
The present investigation has provided a direct spectroscopic
evidence for the formation of the main transient intermediates,
terpenyl and terpenylperoxyl radicals, involved in the free-
radical induced peroxidation of g-terpinene in acetonitrile at
room temperature. From a comparison with 1,4-cyclohexa-
diene, it emerges that the alkyl groups in g-terpinene seem to
have an opposite effect in the initial and final steps of the per-
oxidation route. Indeed, the intermolecular H-atom abstrac-
(
7123.
b) D. G. Hendry and D. Shueltzle, J. Am. Chem. Soc., 1975, 97,
11 (a) G. A. Russell, J. Am. Chem. Soc., 1956, 78, 1047; (b) D. G.
Hendry and G. A. Russell, J. Am. Chem. Soc., 1964, 86, 2371;
(
1
c) see also J. A. Howard and K. U. Ingold, Can. J. Chem.,
968, 46, 2655.
1
1
2
3
G. W. Burton and K. U. Ingold, Science, 1984, 224, 569.
(a) G. A. Russell, J. Am. Chem. Soc., 1955, 77, 4583; (b) See also
J. R. Thomas and C. A. Tolman, J. Am. Chem. Soc., 1962, 84,
2079.
ꢂ
tion by tert-butoxy radicals (or HOO radicals) from the
1
1
1
4
5
6
H. Paul, R. D. Small and J. C. Scaiano, J. Am. Chem. Soc., 1978,
1
methylene groups of g-terpinene occurs ca. 2-fold as fast as
that in 1,4-cyclohexadiene because of the greater stabilization,
due to the alkyl groups, of the terpenyl radical. On the other
hand, the intramolecular H-atom abstraction by the terpenyl-
00, 4520.
R. D. Small, J. C. Scaiano and L. K. Patterson, Photochem.
Photobiol., 1979, 29, 49.
J. E. Jordan, D. W. Pratt and D. E. Wood, J. Am. Chem. Soc.,
1974, 96, 5588.
17 M. C. Sauer and B. Ward, J. Phys. Chem., 1967, 71, 3971.
ꢂ
peroxyl radical with elimination of HOO radicals seems to
occur more than one order of magnitude slower than in the
corresponding cyclohexadienylperoxyl radical. Such a differ-
ence is in agreement with the lower reactivity of tert-alkyl per-
oxyls relative to sec-alkyl peroxyls in intermolecular H-atom
abstraction. The overall kinetic and spectroscopic results pre-
sented herein provide a strong support to the antioxidant
mechanism recently proposed for TH. This mechanism is unu-
1
8
X. M. Pan, E. Bastian and C. von Sonntag, Z. Naturforsch., Teil
B, 1988, 43, 1201.
R. D. Small and J. C. Scaiano, J. Am. Chem. Soc., 1978, 100, 296.
J. C. Scaiano, J. Photochem., 1973, 2, 81.
(a) M. R. Topp, Chem. Phys. Lett., 1975, 31, 144; (b) A. Beckett
and G. Porter, Trans. Faraday Soc., 1963, 59, 2038.
19
20
21
22 A. Effio, D. Griller, K. U. Ingold, J. C. Scaiano and S. J. Sheng,
J. Am. Chem. Soc., 1980, 102, 6063.
23 (a) B. Maillard, K. U. Ingold and J. C. Scaiano, J. Am. Chem.
Soc., 1983, 105, 5095; (b) Z. B. Alfassi, in Peroxyl Radicals, ed.
Z. B. Alfassi, Wiley, New York, 1997, Chapter 1, p. 1–17.
ꢂ
sual because involves the HOO species (formed by elimination
from the TOO radicals) which is typically, from a biological
ꢂ
standpoint, a harmful radical species (it is able to initiate a per-
oxidation process).
2
4
X. M. Pan, M. N. Schuchmann and C. von Sonntag, J. Chem.
Soc., Perkin Trans. 2, 1993, 1021.
The use of vitamin E and other phenolic antioxidants as
food preservatives may be limited by the pro-oxidant effect
that, under particular conditions, these compounds may
25 R. Bausch, H. P. Schuchmann, C. von Sonntag, R. Benn and H.
Dreeskamp, J. Chem. Soc. Chem. Commun., 1976, 418.
26 R. Benn, H. Dreeskamp, H. P. Schuchmann and C. von Sonntag,
Z. Naturforsch., Teil B, 1979, 34, 1002.
3
2
exert. In this context, TH might instead provide an alterna-
tive or supplementary strategy since this hydrocarbon does
not present any pro-oxidant effect.
2
7
p-Cymene was essentially the sole organic product detected by
GC-MS and UV spectroscopy also during the AIBN-induced per-
4
oxidation of a dilute solution of TH .
K. A. Tallman, D. A. Pratt and N. A. Porter, J. Am. Chem. Soc.,
2
8
9
2
001, 123, 11 827.
D. A. Pratt, J. H. Mills and N. A. Porter, J. Am. Chem. Soc.,
003, 125, 5801.
Acknowledgements
2
2
Financial supports from MIUR ‘‘Cofinanziamento di Pro-
grammi di Ricerca di Rilevante Interesse Nazionale’’ and
INCA (Consorzio Interuniversitario per la Chimica dell’
Ambiente) are gratefully acknowledged. The authors also
3
3
0
1
D. A. Pratt and N. A. Porter, Org. Lett., 2003, 5, 387.
S. Korcek, J. H. B. Chenier, J. A. Howard and K. U. Ingold, Can.
J. Chem., 1972, 50, 2285.
32 V. W. Bowry and K. U. Ingold, Acc. Chem. Res., 1999, 32, 27–34.
New J. Chem., 2003, 27, 1563–1567
1567