Isomerization of α-pinene oxide over cerium and tin catalysts: Selective synthesis of trans-carveol and trans-sobrerol

Article abstract of DOI:10.1016/j.molcata.2011.05.020

A remarkable effect of the solvent nature on the acid catalyzed transformation of α-pinene oxide allowed direction of the reaction to either trans-carveol or trans-sobrerol. Each of these highly valuable compounds was obtained in nearly 70% yield using an appropriate polar solvent, whose basicity affected strongly the product distribution. In acetone, a weakly basic solvent, the reaction over heterogeneous sol-gel Sn/SiO₂ or Ce/SiO₂ catalysts gave mainly trans-sobrerol. No leaching of active components occurs under the reaction conditions and the catalysts can be recovered and reused. On the other hand, in more basic solvent, i.e., dimethylacetamide, the reaction was essentially directed to trans-carveol. Due to the leaching problems with Sn/SiO₂ and Ce/SiO₂ materials, the synthesis of trans-carveol was performed under homogeneous conditions using CeCl₃ or SnCl₂ as catalysts with a catalyst turnover number up to ca. 1200. The method represents one of the few examples of the synthesis of isomers from α-pinene oxide, other than campholenic aldehyde, with a sufficient for practical usage selectivity.

Full text of DOI:10.1016/j.molcata.2011.05.020

Journal of Molecular Catalysis A: Chemical 345 (2011) 69–74
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Isomerization of ␣-pinene oxide over cerium and tin catalysts: Selective
synthesis of trans-carveol and trans-sobrerol
a
b
b
b
Vinícius V. Costa , Kelly A. da Silva Rocha , Líniker F. de Sousa , Patricia A. Robles-Dutenhefner ,
a,∗
Elena V. Gusevskaya
Departamento de Química, Universidade Federal de Minas Gerais, 31270-901, Belo Horizonte, MG, Brazil
a
b
Departamento de Química, Universidade Federal de Ouro Preto, 35400-000, Ouro Preto, MG, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
A remarkable effect of the solvent nature on the acid catalyzed transformation of ␣-pinene oxide allowed
direction of the reaction to either trans-carveol or trans-sobrerol. Each of these highly valuable com-
pounds was obtained in nearly 70% yield using an appropriate polar solvent, whose basicity affected
strongly the product distribution. In acetone, a weakly basic solvent, the reaction over heterogeneous
sol–gel Sn/SiO2 or Ce/SiO2 catalysts gave mainly trans-sobrerol. No leaching of active components occurs
under the reaction conditions and the catalysts can be recovered and reused. On the other hand, in more
basic solvent, i.e., dimethylacetamide, the reaction was essentially directed to trans-carveol. Due to the
leaching problems with Sn/SiO2 and Ce/SiO2 materials, the synthesis of trans-carveol was performed
under homogeneous conditions using CeCl3 or SnCl2 as catalysts with a catalyst turnover number up
to ca. 1200. The method represents one of the few examples of the synthesis of isomers from ␣-pinene
oxide, other than campholenic aldehyde, with a sufficient for practical usage selectivity.
Received 3 March 2011
Received in revised form 9 May 2011
Accepted 22 May 2011
Available online 27 May 2011
Keywords:
Isomerization
␣
-Pinene oxide
Solvent effect
Tin catalysts
Cerium catalysts
©
2011 Elsevier B.V. All rights reserved.
1
. Introduction
of sandalwood-like fragrances [4]. Campholenic aldehyde can be
obtained from ␣-pinene oxide with a sufficient for practical appli-
cations selectivity [5–7], whereas all other compounds are usually
produced as minor products with less than 25% selectivities.
We have recently studied the application of heteropoly acid
Terpenic compounds, in general, are an important renewable
feedstock for the flavor and fragrance industry. Their oxygenated
derivatives usually show interesting organoleptic properties and
form one of the most important groups of fragrance ingredients
H PW12O40 (PW) for the liquid-phase isomerization of ␣-pinene
3
[
1,2]. ␣-Pinene, a major constituent of turpentine oils obtained
oxide in various solvents [8,9]. A remarkable effect of solvent
polarity and basicity on the chemoselectivity of this reaction was
found, which allowed a selective synthesis of various valuable fra-
grance compounds with reasonable selectivities. In cyclohexane,
a non-polar solvent, 70% selectivity to campholenic aldehyde was
obtained on a heterogeneous silica-supported PW catalyst, with a
combined yield for campholenic aldehyde and trans-carveol being
nearly quantitative. On the other hand, the use of polar solvents
allowed switching the reaction pathway to the formation of com-
pounds with a para-menthenic skeleton, such as trans-carveol and
trans-sobrerol. Polar basic solvents, such as dimethylacetamide,
favored the formation of trans-carveol; whereas polar weakly basic
solvents, such as acetone, promoted the formation of trans-sobrerol
[9].
from coniferous trees, is a particularly valuable raw material in
the production of synthetic substitutes for natural aromas. The
epoxidation of ␣-pinene provides ␣-pinene oxide, which can be
converted by acid-catalyzed transformations in various expensive
ingredients for the flavor industry, such as campholenic aldehyde,
trans-carveol, trans-sobrerol and trans-pinocarveol.
␣
-Pinene oxide is highly reactive in the presence of acid and can
result in a great variety of products. In one of the previous works,
over 200 compounds formed from ␣-pinene oxide at temperatures
above 100 C have been found [3]. Therefore, the development of
the reaction which is selective for one particular product is a very
challenging task. Most of the studies have been focused on the syn-
thesis of campholenic aldehyde, which is used for the manufacture
◦
The highest previously reported yield of trans-carveol (45%),
an expensive constituent of the Valencia orange essence oil used
in perfume bases and food flavor compositions, was obtained
using acidic molecularly imprinted polymers as protic catalysts in
equimolar to ␣-pinene oxide amounts [10]. trans-Sobrerol, a well-
known mucolytic drug, presently attracts increasing attention for
∗ Corresponding author. Tel.: +55 31 34095741; fax: +55 31 34095700.
E-mail address: elena@ufmg.br (E.V. Gusevskaya).
1
381-1169/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2011.05.020

7
0
V.V. Costa et al. / Journal of Molecular Catalysis A: Chemical 345 (2011) 69–74
O
OH
OH
OH
+
+
+
O
OH
1
2
3
4
5
α- pinene oxide
campholenic aldehyde
trans-carveol
trans-pinocarveol
trans-sobrerol
Scheme 1. Some products of the acid-catalyzed transformations of ␣-pinene-oxide.
its other important biological activities [11]. The highest previously
reported yield of trans-sobrerol (30%) was obtained in the reflux
dichloroethane solutions of ␣-pinene oxide and catalytic amounts
2.2. Catalytic experiments
The reactions were carried out in a glass reactor equipped
with a magnetic stirrer at 5–140 C. In a typical run, a mixture of
◦
of TiCl [12]. Although we found that the PW catalyst could promote
4
the formation of trans-carveol and trans-sobrerol in higher than 70%
yield each using an appropriate polar solvent [9], the reactions were
performed under homogeneous conditions due to high solubility of
PW in polar solvents. Nevertheless, these results encouraged us to
work further with the isomerization of ␣-pinene oxide in order to
develop heterogeneous catalytic processes for the selective synthe-
sis of the products other than campholenic aldehyde, in particular,
trans-carveol and trans-sobrerol.
Within our program aimed at adding value to natural ingre-
dients of renewable essential oils [13–16] we decided to prepare
cerium and tin containing molecular sieves and used them as acid
catalysts for the isomerization of ␣-pinene oxide in polar solvents.
The catalysts have been synthesized through a direct hydrothermal
sol–gel technique which usually affords more stable to leaching
materials than conventional impregnation methods due to higher
levels of metal incorporation in a silica framework.
In the present work, we report a simple and efficient synthesis
of several highly valuable fragrance and/or pharmaceutical com-
pounds through the liquid-phase isomerization of ␣-pinene oxide
in the presence of cerium and tin catalysts. A remarkable effect of
solvent basicity on chemoselectivity allowed direction of the reac-
tion to either trans-carveol or trans-sobrerol, which were obtained
in nearly 70% yields.
␣-pinene oxide (0.122–0.365 g, 0.8–2.4 mmol), dodecane (110 ␮L,
0.5 mmol, internal standard) and the solid catalyst (0.03–0.12 g of
Ce/SiO , Sn/SiO2 or SiO ) or soluble metal chloride (5–20 ␮mol of
2
2
CeCl × 7H O or SnCl × 2H O) in a specified solvent (5 mL) was
3
2
2
2
intensively stirred under air at a specified temperature. The reac-
tion progress was followed by gas chromatography (GC) using a
Shimadzu 17 instrument fitted with a Carbowax 20 M capillary
column and a flame ionization detector. At appropriate time inter-
vals, aliquots were taken and analyzed by GC. The GC mass balance,
product selectivities and yields were calculated based on the sub-
strate charged using dodecane as internal standard. The difference
was attributed to the formation of oligomers, which were not GC
determinable.
Catalyst recycling experiments were performed as follows: after
the reaction, the catalyst was centrifuged, washed with acetonitrile
and then with cyclohexane and reused. To control metal leach-
ing, the catalyst was removed at the reaction temperature and the
solution was allowed to react further.
The structures of products 2–5 were confirmed by GC/MS
(Shimadzu QP2010-PLUS instrument, 70 eV) by comparison with
authentic samples. Major products were isolated by column chro-
1
13
matography (silica) and identified by H and C NMR spectroscopy
(Bruker DRX-400, tetramethylsilane, CDCl3, COSY, HMQC, DEPT and
NOESY experiments).
Spectroscopic data for campholenic aldehyde (2) and trans-carveol
3) have been reported in our previous work [8].
2
. Experimental
(
Spectroscopic data for trans-pinocarveol (4): MS (m/z/rel.int.):
2
.1. Catalyst preparation and characterization
+
1
9
34/27 [M −H O], 119/53, 109/33, 105/20, 95/20, 93/32, 92/100,
2
1/80, 83/78, 81/38, 79/32, 77/20, 70/92, 69/50, 67/22, 55/90, 53/20.
The Sn/SiO2 and Ce/SiO2 catalysts were prepared by
a
Spectroscopic data for trans-sobrerol (5) have been reported pre-
viously [9].
sol–gel method using tetraethoxysilane (14.90 g, TEOS, 99%,
Sigma–Aldrich) and SnCl × 2H O (0.90 g, Sigma–Aldrich) or
2
2
CeCl × 7H O (1.44 g, Sigma–Aldrich), respectively, as precursors.
3
2
The sol was obtained from a TEOS/ethanol/water mixture in a
3. Results and discussion
1
/3/10 molar ratio with the addition of HCl and HF (up to pH 2.0)
◦
as catalysts. The samples were dried at 110 C for 24 h and ther-
mally treated for 2 h at 500 C in air. A pure SiO2 to be tested in
The results of the elemental analysis of the prepared solid
materials, their BET surface areas, average pore sizes, and pore vol-
umes are presented in Table 1. Both Ce/SiO2 and Sn/SiO2 samples
exhibited the isotherms characteristic of mesoporous materials
◦
blank reactions was prepared by the same procedure without the
addition of cerium or tin chlorides. The determination of total tin
and cerium contents was performed by inductively coupled plasma
atomic emission spectrometry (ICP-AES) on a Spectro Ciros CCD
instrument.
2
−1
,
and relatively higher specific surface areas (297 and 366 m g
respectively). The materials were tested as heterogeneous acid
catalysts in the liquid-phase isomerization of ␣-pinene oxide in
various solvents. Transformations of ␣-pinene oxide (1) under
acidic conditions can result in campholenic aldehyde (2), trans-
carveol (3), trans-pinocarveol (4) and trans-sobrerol (5) shown in
Scheme 1, as well as in numerous other products. In the present
work, our efforts were directed to achieve high selectivity to any
other than campholenic aldehyde product and to perform the pro-
cess under heterogeneous conditions, i.e., in the absence of leaching
The textural characteristics of the catalysts were determined
from nitrogen adsorption isotherms (Autosorb-Quantachrome
◦
NOVA-1200 instrument, nitrogen, −196 C). Samples were out-
◦
gassed for 2 h at 300 C before analysis. Specific surface
areas were determined by the Brunauer–Emmett–Teller (BET)
equation. Average pore diameters were determined by the
Barrett–Joyner–Halenda (BJH) method.

V.V. Costa et al. / Journal of Molecular Catalysis A: Chemical 345 (2011) 69–74
71
Table 1
Elemental analysis data and textural properties of the Ce/SiO2 and Sn/SiO2 catalysts.
Sample
Ce or Sn
content (wt%)
BET surface
Total pore
volume (cm
BJH average pore
diameter (nm)
2
−1
3
−1
g )
area (m
g
)
SiO2
Ce/SiO2
Sn/SiO2
–
4.1
4.6
272
297
366
0.91
0.82
0.56
10.2
5.2
6.7
of active components from the solid. As the best results in the
synthesis of trans-carveol and trans-sobrerol from ␣-pinene oxide
were previously obtained in dimethylacetamide (DMA) and ace-
tone solutions, respectively [9], we initiated our study with the
not decrease further, the rate of the reaction with the spent catalyst
was nearly the same as that with pure silica. Whereas the fresh cat-
alyst resulted in a virtually complete conversion for 3 h (Table 2, run
3), in recycling experiments as well as with pure silica, only ca. 60%
of the substrate was converted for the same reaction time (Table 2,
runs 2, 5 and 6). Moreover, when the catalyst was filtered off at the
reaction temperature, the filtrate continued to convert ␣-pinene
oxide. Thus, the Ce/SiO2 catalyst releases active components to the
DMA solution and the process is not truly heterogeneous.
Ce/SiO2 and Sn/SiO2 catalysts using these compounds as solvents.
◦
␣
-Pinene oxide was relatively stable in DMA at 140 C. In a blank
reaction, only a 5% conversion was observed for 20 h (Table 2, run
). To our surprise, pure silica, which was prepared by the same
procedure as Ce/SiO2 and Sn/SiO , promoted the isomerization of
1
2
␣
-pinene oxide at a reasonable rate with high selectivity for trans-
The sol–gel silica-included tin catalyst, Sn/SiO , also promoted
2
carveol. In the presence of only 2 wt% of SiO2 in DMA, a nearly
complete conversion of ␣-pinene oxide was attained for 10 h result-
ing in trans-carveol in 70% selectivity (Table 2, run 2). It is important
that only two minor products have been observed in detectable
amounts, i.e., campholenic aldehyde and trans-pinocarveol, which
are also valuable compounds. A combined selectivity for camp-
holenic aldehyde, trans-carveol and trans-pinocarveol was as high
as 97%.
the isomerization of ␣-pinene oxide, with the reaction being even
faster than that over Ce/SiO2 (Table 2, cf. runs 3 and 7). The main
reaction product was also trans-carveol obtained in a 70% yield.
The combined yield for trans-carveol and campholenic aldehyde
reached 90% in this reaction. However, leaching experiments with
the tin catalyst also showed disappointing results. In run 8 (Table 2),
which was performed under the same conditions as run 7, the
catalyst was removed by filtration after 0.5 h at the reaction tem-
perature and the filtrate was allowed to react further. As it can be
seen, the reaction keeps going in the absence of the solid, albeit at
a lower rate. Thus, both the Ce/SiO2 and Sn/SiO2 materials are not
stable in DMA solutions under the reaction conditions and release
active components to the solution.
The sol–gel silica-included cerium catalyst, Ce/SiO , promoted
2
much faster reaction than pure silica giving a similar product distri-
bution (Table 2, run 3 and run 2). In run 4, trans-carveol was formed
in nearly 70% yield. However, in a further study it has been shown
that the reaction is not truly heterogeneous and the catalyst loses
the activity due to leaching of active components. The Ce/SiO2 cat-
alyst recovered from the reaction mixture after run 3 by filtration
and reused in run 5 showed much lower activity (52% vs. 98% con-
version in 3 h) than during the first use (Table 2). Although in the
third reaction cycle (Table 2, run 6) the activity of the catalyst did
In a further study, we decided to use cerium and tin chlo-
rides as soluble homogeneous catalysts for the isomerization of
␣-pinene oxide in DMA. These compounds could represent cheaper
alternatives for the homogeneous PW catalyst for the synthesis of
trans-carveol from ␣-pinene oxide developed in our previous work
Table 2
Isomerization of ␣-pinene oxide (1) in dimethylacetamide (DMA) solutions at 140 C.a
◦
Run
Catalyst
Ce or Sn (␮mol)
Substrate (mmol)
Time (h)
Conversion (%)
Product selectivity (%)
TONb
TOFb (h−1
)
2
3
4
1
2
–
–
–
0.8
0.8
20
3
5
58
95
98
98
52
58
48
SiO2 (0.12 g)
10
16
19
18
19
20
70
72
73
69
70
11
5
5
7
6
3
4
5
Ce/SiO2
Ce/SiO2
Ce/SiO2
Ce/SiO2
Sn/SiO2
40
0.8
2.4
0.8
0.8
0.8
3
8
3
3
40
40
40
40
c
d
6
7
0.5
1
0.5
3
2
4
0.5
1
1
1.5
1.5
2.5
4
.5
100
52
92
100
100
100
100
100
100
98
20
70
5
8e
Sn/SiO2
40
0.8
20
17
16
16
15
19
17
17
19
18
71
42
41
70
68
62
65
62
64
60
6
36
38
10
12
15
15
17
15
19
9
CeCl3
CeCl3
SnCl2
SnCl2
SnCl2
SnCl2
SnCl2
20
20
20
20
20
5
0.8
0.8
0.8
0.8
1.6
1.6
1.6
40
40
40
40
80
320
634
938
1226
20
10
80
40
80
f
1
1
1
1
1
1
0
1
2
3
4
5
f
213
g
5
210
122
72
1
1
.6
.6
95
90
a
b
c
d
e
f
DMA 5 mL, solid catalyst ca. 2.0–2.5 wt%. Conversion and selectivity were determined by GC.
TON – moles of the substrate converted/moles of Ce or Sn. TOF – the average turnover frequency.
The catalyst was re-used after run 3.
The catalyst was re-used after run 5.
The catalyst was removed by filtration after 0.5 h at the reaction temperature and the filtrate was allowed to react further.
◦
1
20 C.
g
After run 14, three fresh portions of the substrate (1.6 mmol each) were added consequently, each one after the nearly complete conversion of the previous portion. Total
TON is given.

7
2
V.V. Costa et al. / Journal of Molecular Catalysis A: Chemical 345 (2011) 69–74
Table 3
Isomerization of ␣-pinene oxide (1) in acetone solutions.
a
◦
Run
Catalyst
Ce or Sn (␮mol)
Substrate (mmol)
T ( C)
Time (min)
Conversion (%)
Product selectivity (%)
2
3
5
1
2
3
4
5
6
7
8
9
–
–
–
–
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
1.6
0.8
25
25
40
25
40
5
25
25
25
25
25
300
180
180
30
30
90
180
15
15
60
30
5
84
97
96
100
100
98
100
100
95
SiO2 (0.12 g)
SiO2 (0.12 g)
Ce/SiO2
Ce/SiO2
Ce/SiO2
Ce/SiO2
Sn/SiO2
Sn/SiO2
Sn/SiO2
Sn/SiO2
19
16
14
21
18
16
17
16
17
10
13
12
11
10
10
11
10
11
69
70
72
65
70
72
70
72
72
40
40
40
20
20
10
10
10
b
1
1
0
1
b
43
180
80
20
14
63
a
Conditions: acetone 5 mL, solid catalyst 0.5–3.0 wt%. Conversion and selectivity were determined by GC.
Water was added (1 vol%).
b
[
9]. Representative results are shown in Table 2. In the presence of
with pure silica, a nearly complete conversion was achieved for 3 h
CeCl (0.025 eq), the reaction occurred smoothly resulting in a com-
giving trans-sobrerol in ca. 70% yield (Table 3, run 3).
3
plete conversion for 2–4 h depending on the reaction temperature
With the Ce/SiO catalyst, the reaction showed a similar product
2
(
Table 2, runs 9 and 10). However, the product distribution was dif-
distribution; however, it was much faster than that with pure silica
(Table 3, runs 2 and 3 vs. runs 4 and 5). For the comparison purpose,
the same mass amounts of SiO2 and Ce/SiO2 were used in these
runs. In an attempt to change the reaction selectivity, we decreased
the reaction temperature; however, it did not affect significantly
the reaction pathways (Table 3, run 6). Then, we have added water
(1 vol%) to the system trying to increase the yield of the main prod-
uct, trans-sobrerol, whose formation requires water (Table 3, run
7). However, we did not succeed: the best obtained yield of trans-
sobrerol was nearly 70%. It is important to note that only two
minor products have been observed in detectable amounts along
with trans-sobrerol, i.e., campholenic aldehyde and trans-carveol,
with a total yield for all three valuable compounds being almost
quantitative.
ferent from that observed in the heterogeneous system, in which
trans-pinocarveol was detected only as a minor product (ca. 5%).
In the cerium catalyzed homogeneous reaction, trans-carveol and
trans-pinocarveol were formed in comparable amounts in ca. 40%
selectivity each, with campholenic aldehyde being a main minor
product (ca. 15%). The system is synthetically useful considering
a high combined yield for three expensive fragrance compounds
(
95%), low catalyst to substrate ratio and possibility to re-use the
catalyst solution in DMA after the extraction of the products with
hydrocarbon solvent, e.g., hexane. In most of the reported systems,
the main product of the ␣-pinene oxide isomerization (usually
campholenic aldehyde) was obtained along with numerous side
products [3,7].
The conversion of ␣-pinene oxide in the solutions of SnCl2
In the presence of the Sn/SiO2 material, the reaction was very
fast as the analysis of the first aliquot taken after 15 min showed a
complete conversion (Table 3, runs 8 and 9). Further, the amount
of the catalyst was decreased whereas the substrate concentra-
tion was increased (Table 3, run 10). In this run with only 0.5 wt%
of the catalyst, trans-sobrerol was obtained in ca. 70% yield. Con-
sidering that for the formation of trans-sobrerol water is needed
we have added 1 vol% of water to the system (Table 3, run 11).
However, the presence of extra amounts of water resulted in no
changes in the product distribution, with the reaction rate being
decreased significantly (Table 3, run 11 vs. run 9). It should be
mentioned that, generally, water was not added to the reaction
system. The amount of hydration water present in the solid cata-
lysts and commercial ␣-pinene oxide and acetone was sufficient
for the formation of sobrerol. In addition, water could be present
inside SiO2 pores of the catalysts which had a pore volume of
(
0.025 eq) was very fast, much faster than with CeCl . The tin
3
catalyzed reactions showed four times higher average turnover
frequencies (TOFs) than those catalyzed by cerium (Table 2, runs
1 and 12 vs. runs 9 and 10) and gave ca. 95% combined yield for
1
campholenic aldehyde, trans-carveol and trans-pinocarveol, with
trans-carveol accounting for ca. 70% of the mass balance. Aiming to
improve the catalyst efficiency in terms of turnover number (TON)
we increased the substrate concentration and then decreased the
catalyst amounts (Table 2, runs 13 and 14). In run 14, TON of 320
per mol of Sn and 65% selectivity to trans-carveol were obtained.
After run 14, three fresh portions of the substrate (1.6 mmol each)
were added consequently, each one after the nearly complete con-
version of the previous portion. The total TON of 1226 per mol of
Sn was obtained in this reaction, which gave campholenic alde-
hyde, trans-carveol and trans-pinocarveol in a nearly quantitative
combined yield (Table 2, run 15).
3
−1
.
0.56–0.91 cm g
Thus, in the DMA solutions, tin and cerium catalysts as well
as pure silica promote the isomerization of ␣-pinene oxide giving
mainly trans-carveol, with only two minor products: campholenic
aldehyde and trans-pinocarveol.
The leaching of active components from the Ce/SiO and Sn/SiO2
2
materials under the reaction conditions was verified in special
experiments. After runs 5 and 9 (Table 3), the catalysts were filtered
off at the reaction temperature to avoid re-adsorption of leached
metal ions onto the solid support. Then, the filtrates were recharged
with fresh substrate and allowed to react further. No conversion
of ␣-pinene oxide was observed after catalyst removing, provid-
ing strong evidence in support of heterogeneous catalysis. Thus,
the reaction solutions contained no significant amounts of active
species and the activity of both sol–gel silica-included catalysts,
In a further study, we have tested the prepared Ce/SiO2 and
Sn/SiO2 materials for the isomerization of ␣-pinene oxide in ace-
tone solutions. Representative data are collected in Table 3. In this
case, much more promising results in terms of catalyst leaching
were obtained, which allowed the development of truly heteroge-
neous processes for the synthesis of highly valuable trans-sobrerol.
In a blank reaction with no catalyst added, only 5% conversion
was observed for 6 h at room temperature (Table 3, run 1). In the
presence of pure silica, ␣-pinene oxide undergoes a relatively fast
Ce/SiO2 and Sn/SiO , was due to the active species immobilized in
2
the mesoporous silica framework. The behavior of both spent cat-
alysts recovered after runs 5 and 9 (Table 3) with fresh substrate
was nearly the same as in the original reactions.
◦
isomerization even at room temperature (Table 3, run 2). At 40 C

V.V. Costa et al. / Journal of Molecular Catalysis A: Chemical 345 (2011) 69–74
73
O
O
La
-
La
route B
route C
O
1
0
B
2
1
La
O
1
6
2
3
5
4
8
7
10
10
La
9
1
1
OH
O
La
A
6
9
6
5
2
3
2
3
- La
5
La
8
7
O
- La
route A
4
4
7
C
9
8
3
OH
- La
+
H O
2
10
OH
1
6
5
2
3
La = Lewis acid
4
4
OH
7
8
9
5
Scheme 2. Schematic representation of the Lewis acid-catalyzed transformations of ␣-pinene-oxide.
A schematic representation of the Lewis acid-catalyzed trans-
formations of 1 into products 2–5 is given in Scheme 2. In this
scheme, “La” represents Lewis acid active sites on the catalyst sur-
DMA solutions but not in acetone. Both, DMA and acetone are highly
polar solvents; however, DMA has much higher basicity (pKa ≈ −0.5
for DMA H+ vs. pKa ≈ −7 for (CH ) COH [17]). Therefore, it seems
+
3
2
3+
face (cationic cerium and tin species) or in the solution (Ce and
reasonable that carbocation A is more prompted to undergo a pro-
ton shift (route A) before its isomerization (routes B and C) in DMA,
which has relatively strong proton acceptor properties, than in less
basic acetone.
2+
Sn ions). The activation of the epoxide moiety by the acidic sites
induces epoxy ring opening and initially gives carbenium ion A. The
latter can undergo several competing processes. First, it can give
directly, without any isomerization, trans-pinocarveol 4 through a
hydrogen shift from the C-10 methyl group to the oxygen atom.
The hydrogen abstraction and transfer could be assisted in our
On the other hand, the ratio between aldehyde 2 and para-
menthenic compounds 3 and 5 is influenced by the balance
between two different rearrangements of carbeniun ion A: into
cyclopentanic cation B (route B) and into para-menthenic cation
C (route C). Tertiary carbeniun ion C is more thermodynamically
stable than secondary carbenium ion B; therefore, the formation of
aldehyde 2 should be kinetically controlled. Carbenium ion C can
give trans-carveol 3 or undergo a nucleophilic attack by water to
form trans-sobrerol 5.
In the previous work [9], we have found that the increase in
both solvent basicity and polarity strongly prejudices route B to
aldehyde 2, whereas it favors route C to para-menthenic prod-
ucts. For example, in cyclohexane, a non-polar non-basic solvent,
the main reaction product was campholenic aldehyde 2. A high
ion-solvating ability of polar solvents could result in stabilization
of cation A favoring its rearrangement into more stable tertiary
carbenium ion C. On the other hand, the distribution between para-
menthenic products 3 and 5 depends on solvent basicity rather than
on its polarity. In basic and polar DMA, where the proton transfer
in C can be assisted by the solvent, the formation of trans-carveol
3 occurs faster than the formation of trans-sobrerol 5 and becomes
the preferable reaction pathway.
system by the basic solvent, DMA. In the CeCl /DMA system, trans-
3
pinocarveol accounts for ca. 40% of the mass balance. In addition,
carbenium ion A can rearrange into carbenium ions B or C through
the movement of the pair of electrons from the same carbon-carbon
␴
-bond to either C-7 or C-6, respectively. Carbocation B gives cam-
pholenic aldehyde 2 through the C-1–C-2 bond cleavage, whereas
carbocation C forms trans-carveol 3 through the proton shift from
the C-9 methyl group. The latter process could also be assisted by
the basic solvent (in DMA) and/or by the oxygen atoms from the
SiO2 support acting as Lewis basic sites (in both acetone and DMA
solvents).
The results obtained in the present work showed that a solvent
nature can profoundly affect the reaction pathways and product
composition. This observation allowed us to develop efficient syn-
theses of various compounds from ␣-pinene oxide through the
choice of the appropriate solvent. As it can be seen in Scheme 2,
relative amounts of trans-pinocarveol 4, the product with a bicyclic
non-isomerized pinane skeleton, and monocyclic compounds 2, 3
and 5 depend on the balance between the hydrogen transfer in
cation A (route A) and its isomerization in cations B and C (routes
B and C). Trans-pinocarveol was formed in significant amounts in
It should be mentioned that the results obtained with differ-
ent catalysts (PW, Ce, Sn and SiO ) clearly show that the balance
2

7
4
V.V. Costa et al. / Journal of Molecular Catalysis A: Chemical 345 (2011) 69–74
between the reaction pathways leading to different products from
-pinene oxide is mostly determined by the nature of the solvent
Acknowledgments
␣
rather than by the nature of the catalyst or other reaction variables.
However, further studies should be targeted toward clarifying the
effect of solvent polarity and bacisity on the acidity of the cata-
lyst surface, which should also be considered for understanding
the reaction selectivity.
Financial support from the CNPq, CAPES, FAPEMIG and INCT-
Catálise (Brazil) is gratefully acknowledged.
References
[
[
[
1] H. Mimoun, Chimia 50 (1996) 620–625.
2] C. Chapuis, D. Jacoby, Appl. Catal. A 221 (2001) 93–117.
3] W.F. Hölderich, J. Röseler, G. Heitmann, A.T. Liebens, Catal. Today 37 (1997)
4
. Conclusions
353–366.
[
[
[
[
[
[
4] G. Orloff, B. Winter, C. Fehr, in: P.M. Muller, D. Lamparsky (Eds.), Perfumes, Art,
Science & Technology, Elsevier, New York, 1991, p. 287.
5] P.J. Kunkeler, J.C. van der Waal, J. Bremmer, B.J. Zuurdeeg, R.S. Downing, H. van
Bekkum, Catal. Lett. 53 (1998) 135–138.
6] L. Alaerts, E. Séguin, H. Poelman, F. Thibault-Starzyk, P.A. Jacobs, D.E. De Vos,
Chem. Eur. J. 12 (2006) 7353–7363.
Chemoselectivity of the isomerization of ␣-pinene oxide can
be controlled through the choice of the solvent, whose polarity
and basicity determine the reaction pathways. The heterogeneous
process for the synthesis of highly valuable trans-sobrerol has
been developed using weakly basic acetone as a solvent and
silica-included tin and cerium materials as easily recoverable cat-
alysts. Even pure silica alone can be used as a heterogeneous
catalyst in this reaction, albeit less effective in terms of cat-
alytic activity. On the other hand, in dimethylacetamide, which
is much more basic, the reaction is essentially directed to trans-
carveol, regardless of the catalyst. The sol–gel Sn/SiO2 and Ce/SiO2
materials showed high activity and selectivity for trans-carveol;
however, the filtrate solutions promoted the isomerization of ␣-
pinene oxide after the catalyst removal. Due to these leaching
problems, the synthesis of trans-carveol was performed using dis-
solved CeCl3 or SnCl2 as homogeneous catalysts or pure silica as
a heterogeneous catalyst. The developed methods are among few
examples of the selective synthesis of other than campholenic alde-
hyde compounds from ␣-pinene oxide through its acid-catalyzed
isomerization.
7] N. Ravasio, F. Zaccheria, A. Gervasini, C. Messi, Catal. Commun. 9 (2008)
1125–1127.
8] K.A. da Silva Rocha, I.V. Kozhevnikov, E.V. Gusevskaya, Appl. Catal. A 294 (2005)
106–110.
9] K.A. da Silva Rocha, J.L. Hoehne, E.V. Gusevskaya, Chem. Eur. J. 14 (2008)
6166–6172.
[
10] W.B. Motherwell, M.J. Bingham, J. Pothier, Y. Six, Tetrahedron 60 (2004)
3231–3241.
[11] Q. Wang, Y. Li, Q. Chen, Synth. Commun. 33 (2003) 2125–2134.
[
[
[
12] J. Kaminska, M.A. Schwegler, A. Hoefnagel, H. van Bekkum, Recl. Trav. Chim.
Pays-Bas 111 (1992) 432–437.
13] L.A. Parreira, L. Menini, J.C. da Cruz Santos, E.V. Gusevskaya, Adv. Synth. Catal.
352 (2010) 1533–1538.
14] K.A. da Silva Rocha, N.V.S. Rodrigues, I.V. Kozhevnikov, E.V. Gusevskaya, Appl.
Catal. A 374 (2010) 87–94.
[15] M.G.
Speziali, V.V. Costa, P.A. Robles-Dutenhefner, E.V. Gusevskaya,
Organometallics 28 (2009) 3186–3192.
[
[
16] P.A. Robles-Dutenhefner, K.A. da Silva Rocha, E.M.B. Sousa, E.V. Gusevskaya, J.
Catal. 265 (2009) 72–79.
17] J. March, Advanced Organic Chemistry: Reactions, Mechanisms and Structures,
fourth ed., Wiley, New York, 1992, p. 250.