THERMAL TRANSFORMATION OF MONOTERPENES
17
Table 1. Transformation of monocyclic monoterpenes within
thionin/Na-Y
supported Na-Y, do not occur in Na-Y as obtained from
commercial sources. Upon heating the zeolite at 120°C
under vacuum for 1–2 h, a slight formation (<4%) of the
isomeric products is observed in 30 min, whereas after
heating for at least 6–10 h, Na-Y has almost the same
properties as thionin/Na-Y, and gives very similar results
to those presented in Table 1. Prolonged drying for 24 h
does not improve the efficiency of Na-Y in performing
faster isomerization–dehydrogenation reactions. These
results indicate that thionin, a well known electron
acceptor,8 does not participate in these intrazeolite
reactions.
From the first point of view, the isomerization
reactions seem to proceed by acid catalysis, attributable
to the few Brønsted acid3 sites present in Na-Y. However,
this explanation is questionable. The acid-catalyzed
skeletal isomerization of monoterpenes having the mol-
ecular formula C10H16 within porous materials and
zeolites is well known and affords not only various
amounts of p-cymene, but also reduced products with the
molecular formula C10H18, such as p-menthenes.9 This
disproportionation reaction occurs by hydride transfer
from a neutral monoterpene molecule to the carbocation
generated by addition of a proton to another monoterpene
molecule. The low percentage of C10H18 terpenes, or
even their absence, indicates that p-cymene is formed by
an alternative pathway. In addition, thionin/Na-Y neither
dehydrates nor racemizes optically active sec-pheny-
lethanol. [In a competing reaction, (Æ)-sec-phenyletha-
nol and several monoterpenes used this study were
treated in cyclohexane with the acidic catalyst Amberlyst
15. The alcohol was dehydrated at rates significantly
higher to the isomerization of the monoterpenes.] Only
the highly unstable a,a-dimethylbenzyl alcohol dehy-
drates to a-methylstyrene in 82% yield after 30 min of
stirring with thionin/Na-Y.
Terpene
Conversion (%)a 1 (%) 2 (%) 6 (%) 7 (%)
a-Terpinene, 1
g-Terpinene, 2
Limonene, 3
Terpinolene, 4
a-Phellandrene, 5
75
40
>97
>97
>97
25
10
20
25
36
4
60
22
10
3
32
10
25
33
30
36
17
29
28
27
a
Intrazeolite reaction time of 30 min at room temperature. If stirring is
continued for a further 2 h, p-cymene is by far the major product. The C10
alcohols (2–4%) and other isomeric monoterpenes with less than 2%
contribution are not shown. For details on the intrazeolite experiments, see
the Experimental section.
used in the experiments presented in Table 1 does not
improve the rate of the transformations. Approximately
2–4% of alcohols with the molecular formula C10H18O
are also formed under the reaction conditions, via
hydration of the monoterpenes. The mass balance for
all reactions, as measured by using n-nonane or n-
dodecane as internal standards, was always >80%. The
reactions can be performed efficiently in the open air with
the formation, in addition to p-cymene and the isomeric
terpenes, of trace amounts (ꢀ1%) of the endoperoxide of
a-terpinene (ascaridole). The open-air reactions were
carried out strictly in the dark to avoid the formation of
singlet oxygen ene-type adducts1 or ascaridole (up to
15%), which are produced slowly even in the ambient
light of the laboratory.
The dehydrogenation of monoterpenes such as 1–6 to
p-cymene is identical with their transformation within
MV2/Na-Y,6 in photosensitized electron-transfer reac-
tions in solution,10,11 on metal oxides surfaces upon
irradiation,12 or by catalytic amounts of the mixed-
Note that in our previous work within MV2/Na-Y6 we
reported that terpinolene (4) is formed as one of the
intermediate terpenes that finally lead to p-cymene.
Careful re-examination of the products revealed that
isoterpinolene (6) is formed instead. Unfortunately,
terpinolene and isoterpinolene have identical retention
times in GC, even on a 50 m capillary column, and that
led to the wrong product assignment. Compound 6 is
formed within thionin/Na-Y in appreciable amounts (see
Table 1), and we were able to characterize it properly,
after isolation by preparative GC. [Isoterpinolene (6) has
the following 1H NMR data in CDCl3: 6.41 (dd,
J1 = 10 Hz, J2 = 2.2 Hz, 1H), 5.55 (d, J = 10 Hz, 1H),
2.51 (m, 1H), 2.13 (m, 2H), 1.83 (m, 2H), 1.77 (s, 3H),
1.73 (s, 3H), 1.01 (d, J = 7.2 Hz, 3H). Its fragmentation
pattern in the mass spectrum is identical with that
reported in the literature.] Nevertheless, 6 also ends up as
p-cymene after prolonged intrazeolite reaction time.
The above-mentioned transformations of 1–6, or
isomerizations of other simple alkenes within thionin-
addenda heteropolyanion PV2Mo10O40 .
5À 13 These oxida-
tion reactions have been attributed to proceed via
formation of the radical cations of the monoterpenes.
We postulate that during the thermal dehydration
treatment of Na-Y, acidic sites within the cages14
and/or extra-lattice aluminum species15 are activated
and catalyze these rearrangements. For example, Lewis
acid sites within zeolites are known to act as ‘electron
holes’16 and accept one electron from substrates with low
ionization potential. For the thionin/Na-Y system we
propose the mechanistic rationale shown in Scheme 1.
The terpenes may isomerize due to the acidic
environment [Eqn. (1)]. On the other hand, the ‘electron
holes’ within Na-Y could also accept an electron from the
monoterpene [Eqn. (2)], whose radical cations can
isomerize to the more stable radical cation of a-terpinene
Copyright 2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 16–20