Fate of monoterpenes in near-critical water and supercritical alcohols assisted by microwave irradiation

Article abstract of DOI:10.1039/b924748c

The rearrangement of α- and β-pinene was studied under microwave irradiation in near-critical water and supercritical lower aliphatic alcohols, with the aim of identifying the pathway of α- and β-pinene isomerization. Generally, two pathways occur, pyroly

Full text of DOI:10.1039/b924748c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Fate of monoterpenes in near-critical water and supercritical alcohols assisted
by microwave irradiation†‡
Tony Szuppa, Achim Stolle* and Bernd Ondruschka
Received 5th November 2009, Accepted 9th January 2010
First published as an Advance Article on the web 1st February 2010
DOI: 10.1039/b924748c
The rearrangement of a- and b-pinene was studied under microwave irradiation in near-critical water
and supercritical lower aliphatic alcohols, with the aim of identifying the pathway of a- and b-pinene
isomerization. Generally, two pathways occur, pyrolysis on the one and acidolysis on the other hand,
whereby acidolysis is predominant in the case of near-critical water and the second pathway is favored
for experiments employing supercritical alcohols. The different behavior of these two structurally related
solvents is attributed to the increased availability of protons if water is heated to 270 ◦C and 80 bar, thus
enhancing the autoprotolysis of water. The application of alcohols instead furnished rearrangement
products clearly attributed to thermal pyrolysis route by the formation of radical reaction intermediates.
limonene (3), alloocimene (4) as well as pyronenes (5) resulting
Introduction
from consecutive reactions of 4.2,3 Contrarily, on pyrolysis, 2
A small but heterogeneous class of natural products are the ter-
penes and terpenoids, which often are employed as major building
blocks in the synthesis of high-valued chemicals.1 In addition to the
modification of the carbon skeleton by introducing heteroatoms
(oxidation, epoxidation) or other functionalities (selective hy-
drogenation), terpenes and terpenoids offer the opportunity for
performing interesting rearrangement reactions in the gaseous,
liquid and supercritical phases.2–5 Often the rearrangement of
terpenes with constrained rings, e.g. a-pinene (1) or b-pinene (2,
Chart 1), is focused on thermal (pyrolysis) and acid-catalyzed
isomerizations (acidolysis). Without any catalyst under a high-
temperature regime it is reported that 1 can easily be converted into
furnishes besides 3 and y-limonene (7), the acyclic hydrocarbon
myrcene (6) as the main product in up to 85% yield.2,3c,f ,4,5a Due
to the formation of these products it is supposed that 1, and
respectively 2, undergoes ring opening by forming a biradical
intermediate, which undergoes fast rearrangement (e.g. retro-ene
reaction) forming isomerization products.2
In presence of acidic or basic catalysts other types of isomeriza-
tion are predominant. In case of 1 and 2, the acid-catalyzed rear-
rangement results in an enlarged variety of products.6,7 Performing
the reaction under ambient conditions, compounds related to the
p-menthadiene-structure are found: 3, terpinolene (8), g- (9) and
a-terpinene (10). For the reactions, metal oxides, mineralic acids,
metal salts or (ion-exchanged) zeolites are common catalysts.1a,b,7
The formation of a-phellandrene (11), p-cymene (12), camphene
(13), and b-phellandrene (14) is observed under special prerequi-
sites concerning catalyst or reaction condition.7f ,h,l,8–11
Commonly, pyrolysis and acidolysis reactions of terpenes are
performed in the gas phase in order to control the residence
time and to prevent bimolecular reactions furnishing undesirable
polymers or dimers. In both cases, these gas phase reactions
are accomplished under reduced- or high-pressure conditions,
Institute for Technical Chemistry and Environmental Chemistry, Friedrich-
Schiller University Jena, Lessingstraße 12, D-07743, Jena, Germany.
E-mail: Achim.Stolle@uni-jena.de; Fax: 49 3641 948402; Tel: 49 3641
948413
† This paper is part of an Organic & Biomolecular Chemistry web theme
issue on enabling technologies for organic synthesis.
‡ Electronic supplementary information (ESI) available: Details for the
generation of near-critical water and supercritical alcohols. See DOI:
10.1039/b924748c
Chart 1
1560 | Org. Biomol. Chem., 2010, 8, 1560–1567
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Dissipation factor tan d,15a critical data (pc, Tc), vapor pressure
pv,15b of water and different lower aliphatic alcohols as well as their
Table 2 Results of the rearrangement of a-pinene (1) and b-pinene (2) in
near-critical watera
15c,d
equilibrium constants for autoprotolysis pKauto
a-Pinene (1)
b-Pinene (2)
a
Alcohol
tan d
pc/bar Tc/K pv/mbara
pKauto
Compound
Yield (%)
Originb
Yield (%)
Originb
A
Water
Methanol
Ethanol
Propan-1-ol
Propan-2-ol
Butan-1-ol
0.132
0.659
0.941
0.757
0.799
0.571
220.6
80.84
61.37
51.69
47.64
44.14
42.95
647.1 32
512.5 129
514.0 54
536.8 20
508.3 43
563.0 6.7
547.8 17
14.995
17.20
18.88
19.43
Limonene (3)
Alloocimene (4)
Pyronene (5)
Terpinolene (8)
g-Terpinene (9)
a-Terpinene (10)
a-Phellandrene (11)
p-Cymene (12)
Camphene (13)
Unknown
14
9
4
20
24
12
2
5
3
7
P, (A)
P
P
A
A
A
A
A
A
19
—
—
23
24
14
2
2
5
10
A
A
A
A
A
A
?
21.56
2-Methylpropan-1-ol 0.522
a At 25 ◦C.
?
allowing for the use of glass flasks or autoclaves made of stainless
steel. Although it was shown recently that microwaves can be used
as a tool for pyrolysis reactions,12 the utilization of microwave
irradiation to perform pyrolysis or acidolysis reactions with
terpenes has not been reported yet. Within this work, a suitable
method was investigated for the rearrangement of 1 and 2 using
microwave irradiation13 under high-temperature (<300 ◦C) and
high-pressure (<80 bar) conditions. Thus, it was possible to
generate near-critical water14 and supercritical alcohols as reaction
media for the transformation of the mentioned monoterpenes.
a 250 ml pinene, 15 ml 0.03 M NaCl solution, 80 ml quartz vessel, heating:
10 min, reaction time: 60 min, cooling: 20 min, Pmax: 1.2 kW, microwave:
Synthos 3000. b A: acidolysis, P: pyrolysis.
aqueous solution of sodium chloride (15 ml, 0.03 M). This, under
ambient conditions, bi-phasic mixture was exposed to microwave
irradiation in quartz vessels, and heated up to 270 ◦C and 80 bar.
After a heat-up period of 10 min, the conditions were kept constant
for 60 min. After cooling the mixture to ambient temperature and
work-up of the reaction mixture, the conversion (X) of both 1
and 2 was found to be quantitative. In the case of 1, the resulting
products indicate that both pyrolysis (4, 5) and acidolysis reaction
(8–13) took place (cf. Table 2), whereby the reaction of 2 under
similar reaction conditions furnished the latter only.
Results and discussion
Comparison of the critical data for water and lower aliphatic
alcohols15 listed in Table 1 reveals great differences for both
the critical pressure pc and temperature Tc between these two
structurally related class of molecules. Because of the experimental
boundary conditions determined◦by the microwave apparatus used
for the experiments (Tlim = 300 C, plim = 80 bar; Synthos 3000;
Anton Paar GmbH, Graz, Austria), it was not possible to perform
experiments with supercritical water. In accordance to the work
of Kremsner and Kappe in the field of microwave-assisted near-
critical water generation, basic experiments approve the necessity
of working with saline aqueous solutions instead of pure water,
to achieve stable and reproducible reaction conditions as well as
improved heat-up behavior (cf. ESI‡).14a Thus, experiments were
conducted employing an 0.03 M NaCl solution.
Using lower aliphatic alcohols listed in Table 1 as solvents under
the same experimental prerequisites (Tlim = 300 ◦C, plim = 80 bar)
also, special experimental conditions have to be set up to avoid
pressure and temperature instabilities when heating to the limiting
parameters. In particular, when the alcohols undergo phase-
transformation from liquid to gaseous or supercritical state, the
energy uptake is distorted resulting in process instabilities. Facing
and overcoming this problem was possible by employing passive
heating elements made from silicon carbide into the reaction
vessels (cf. ESI‡).16 The application of alcoholic saline solutions
is inappropriate since the saline residue (after evaporation of the
solvent) might absorb the reactants and therefore adulterating the
experimental outcomes.
In both cases, the majority of obtained product originates from
the acidolysis reaction pathway. This is due to the fact that the
activation energy for pyrolytic ring cleavage is higher than for the
competitive reaction via carbocations (induced by addition of a
proton to the double bond), resulting in a strong preference for
the latter. This hypothesis is supported by the fact that thermal
rearrangement normally required high temperatures (>300 ◦C)
and short contact times in the case of flow systems or T of 200 ◦C
and reaction times of >2 h if the reactions are carried out in batch
system either under vacuum or high-pressure conditions.2–5 In this
◦
case, 270 C is not high enough to pyrolyze great amounts of 1
or 2. Pyrolysis of 1 runs through a biradical reaction intermediate
(I; Scheme 1), which, upon degenerative rearrangement, afforded
racemic limonene (3; dipentene) and 3Z-ocimene (15) as primary
pyrolysis products. The latter undergoes fast sigmatropic [1,5]H
shift yielding alloocimene 4, which cyclizes to form cyclohexadiene
isomers 5.2b,3,17 The pyrolysis of 1 initially starts with the opening
of the cyclobutane ring, forming the biradical intermediate I
(Scheme 1). The radical is stabilized by the allylic system and
furnishes the monocyclic retro-ene product 3 by a formal 1,5H
shift in racemic form (dipentene). Dipentene formation is due to
the mesomeric delocalization of one radical position as an allyl-
type radical.2b,3c,e,g Furthermore, it is supposed that intermediate I
yields the formal retro-[2 + 2]-cycloaddition product 3Z-ocimene
(15) by complete ring opening. Due to selective formation of the Z-
isomer, the molecule comprises an ideal structure for undergoing
a sigmatropic [1,5]H shift to afford 4.3d,g,17 Due to this entropic
and enthalpic favored rearrangement, the reaction from 15 to 4
is very fast, making it impossible to detect 15 in the reaction
mixture yielding from pyrolysis of 1.2,3a–e,5 Conjugated triene 4 is
able to undergo various electrocyclizations and sigmatropic group
Rearrangement of a-pinene and b-pinene in near-critical water
Experiments were conducted to study the behavior of a- (1)
and b-pinene (2) under near-critical water conditions, whereby
a discrete portion of the substrate (250 ml) was added to an
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1560–1567 | 1561
©

View Article Online
Scheme 1 Formation of pyrolysis products from a-pinene (1).
Table 3 Results of the rearrangement of selected p-menthadienes in near-
To identify the origin of the resulting products, 3 and 8–10 were
treated separately in near-critical water under the same conditions
as applied for 1 and 2. The data compiled in Table 3 indicate that
3 is the most reactive p-menthadiene isomer, since it underwent
isomerization to 8–10. Contrarily, these compounds seem to be
relatively stable under the conditions the reactions are carried
out. The formation of p-cymene (12) as side reaction for the
endocyclic isomers 9 and 10 is due to their tendency to form the
thermodynamic stable aromatic. Important is also the fact that
the isomerization of the monocyclic isomers yielded no traces of
bicyclic products (1, 2, and 13).
critical watera
Yield (%)
3
7
8
9
10 11 12 13 Unknown
a-Pinene (incl. 3)
a-Pinene (excl. 3)
Limonene (3)
Terpinolene (8)
g-Terpinene (9)
a-Terpinene (10)
16
—
21
8
4
2
23 28 14
27 33 16
22 28 18
62 11 10
2
3
3
1
3
1
6
7
2
3
11
19
3
4
2
0
0
0
8
10
2
3
8
2
2
5
3
7
8
57
18 44
5
5
a 250 ml a-pinene (1), 15 ml 0.03 M NaCl solution, 80 ml quartz vessel,
heating: 10 min, reaction time: 60 min, cooling: 20 min, Pmax: 1.2 kW,
microwave: Synthos 3000.
Based on the experimental outcomes of the reactions of 1–3
and 8–10, a general reaction scheme can be drawn (Scheme 2).
With respect to Markovnikov’s rule, a proton is added to the
double bond of 1 or 2, respectively, to form a triple-substituted
carbocation (carbenium ion II; Scheme 2), which is stabilized by
the neighboring methyl group by hyperconjugation. Interestingly,
the present result supports the hypothesis that the formation
of carbocation II (pinanyl cation) is an irreversible reaction,
since neither in the case of the reaction of 1 nor 2 was an
isomerization into the other observed.7j This is in contrast to
previous studies, which indicate the existence of an equilibrium
between 1 and 2 on treatment with acids or bases.7e,9,20,21 The
pinanyl cation is in equilibrium with the isobornyl cation (IV),
whereby as intermediate structure the non-classical carbenium
ion III (carbonium ion) can be formulated. Calculations on the
B3LYP/6-31G(d) level of theory indicate a small energy difference
between II and IV, accounting for these being real transition states
and III is the global maximum on the reaction coordinate from
1 to 13.9 Despite the fact that calculations on the same level
of theory indicate an increased stability of the bicyclic cations
II–IV compared to the monocyclic menthenyl cation V,9 the
reaction preferably runs through this reaction pathway, probably
due to entropic and enthalpic preference for an unsaturated less
strained monocyclic ring system. Starting from V, an exocyclic
deprotonation afforded limonene (3), whereas an endocylic loss of
the proton furnishes terpinolene (8; Scheme 2). It has to be pointed
out that the outcome of isomerization experiments with 3 and 8
indicate that the rearrangement of 3 to 8 runs through intermediate
V, which is in contrast to the work of Comelli et al., assuming 3
as an intermediate for the formation of 8–11.7j By reprotonation
of 8, carbocation VI (terpinyl cation; Scheme 2) is formed, which
is able to undergo two consecutive reactions: (i) g-terpinene (9)
is formed containing two isolate endocyclic double bonds (1,4-
cyclohexadiene subsystem) and (ii) a-terpinene (10) is formed with
transformations forming different tetramethylated cyclohexadi-
enes 5.2,3a–e,5,18 The equality of the amounts of 3 (14%) formed
through rearrangement of 1 with the summarized amounts of 4
and 5 (13%; Table 3) is an indicator that 3 mainly originates from
pyrolysis and not from acidolysis. Kinetic pyrolysis experiments
of 1 in the gas phase strongly indicate that the rate constants
for competitive reactions yielding 3 and 15 (and its consecutive
products) are almost equal.2b,3d,e For pyrolysis of 2 the ratio
between pathways leading to 3, acyclic 6, and monocyclic 7 is
approximately 2 : 17 : 1.2,4e Without the formation of 6 as major
pyrolysis product, it can be concluded that pyrolysis does not take
place for treatment of 2 under the herein presented conditions.§
Referring to 1, the ratio of pyrolysis to acidolysis is 0.2,
neglecting 3, since this compound can arise from both reaction
pathways. No pyrolysis occurs using 2 as the starting material.
At elevated temperature, the ionic product of water is shifted to a
higher value,19 which leads to a higher concentration of protons
available for chemical reactions, superposing the beneficial effect
of higher T on the rate constant for thermal isomerization. Due
to this behavior of water, it is not necessary to add any further
catalyst for the acidic rearrangement of 1 or 2 in the present case.
Another small but important fact is, that except for 12 the reaction
almost exclusively yielded isomerization products. The formation
of polymers or the hydration of 1 or 2 furnishing alcohols was not
observed in all experiments. Additionally, no reaction with the
reaction atmosphere (nitrogen, oxygen, argon, air) was detectable.
§ The lower activity of b-pinene 2 in the pyrolysis reaction compared to
a-pinene 1 was also validated by comparative pyrolysis studies in the gas
phase.2,3b,g,4e This effect is accounted to the higher thermodynamic stability
of 2 due to the less strained bicyclic ring system in comparison with 1.
1562 | Org. Biomol. Chem., 2010, 8, 1560–1567
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 2 Acid-catalyzed rearrangement of a-pinene (1) and b-pinene (2) in near-critical water.
a conjugated double bond system (1,3-cyclohexadiene subsystem).
Addition of a proton to 9 forms an intermediate carbocation
VII, which undergoes deprotonation resulting in a-phellandrene
(11). In the case of 9–11, both double bonds are located in the
ring system. Within the dehydrogenation of these compounds,
p-cymene (12) is formed (Scheme 2). This reaction is known to
proceed very effectively in the gas phase under the presence of acid
catalysts like zeolites making this reaction an important industrial
process, even when starting from 1 as substrate.22
Variation of reaction parameters for near-critical water
isomerization of a-pinene
The time of the overall process was summarized by adding the
time for heating, cooling, and the actual reaction time (eqn (1)).
Within 10 min, the reaction mixture was heated up to 270 ◦C: tH.
The reaction time (tR) itself is defined as the time the bulk reaction
parameters T and p remain constant, in practice the time between
up- and downstream heating or cooling processes, respectively (cf.
ESI‡). After the reaction time follows a cool-down phase of 20 min
duration (tC).
Fig. 1 Changing of selectivity of pyrolysis (Spyr; cf. eqn (2)) and acidolysis
(Sacid; cf. eqn (3)) and conversion of a-pinene (X1) depending on reaction
time tR (eqn (1); 250 ml 1, 15 ml 0.03 M NaCl solution, 80 ml quartz vessel,
heating: 10 min, cooling: 20 min, Pmax: 1.2 kW, microwave: Synthos 3000).
the case of treatment of 1, the selectivities of compounds 4 and 5
were summarized and defined as pyrolysis products (Spyr), whereas
the selectivities for products 8–13 were summarized as acidolysis
products (Sacid; eqn (2) and (3)). The selectivities of 2 and of the
compounds which are unknown were not considered because the
origin whether from pyrolysis or acidolysis is not possible to assign.
Therefore the yield of 3 (Y3) and unknown compounds (Yu) were
t = tH + tR +tC
(1)
To obtain an overview of the reaction progress at different
reaction times, the changes of selectivity S were investigated
(Fig. 1), while varying the reaction time tR from 0–60 min. In
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1560–1567 | 1563
©

View Article Online
Table 5 Conversion of a-pinene (1) depending on the volume of 1 added
subtracted from the conversion of 1 (X1) leading to the respective
equations for Spyr and Sacid
to the solventa
.
Volume of 1/ml
(2)
0.25
0.5
1
2
Ratio: reactant : solvent
max/◦Cb
X1 (%)
0.017
267
85
0.033
266
72
0.067
268
68
0.133
266
68
T
(3)
a 15 ml 0.03 M NaCl solution, 80 ml quartz vessel, heating: 10 min, reaction
time: 10 min, cooling: 20 min, Pmax: 1.2 kW, microwave: Synthos 3000.
b Maximal temperature of the reactor from recorded data.
Within the period of heating and cooling 10% of 1 were
converted while only acidolysis products appeared (tr: 0 min).
Increasing of reaction time resulted in changing selectivity ratio
of 5.7 in favor of Sacid, whereby the ratio itself remains constant
after 10 min. The conversion of 1 was found to be quantitative
after 15 min. The obtained results show that acidolysis is strongly
preferred independently from the conditions the experiments
have been performed at. As reported,3,6,7e–h,j,l pyrolysis reactions
generally require much higher temperatures than acid-catalyzed
rearrangement, due to the higher activation energies necessitated
for the initiation of homolytic bond cleavage. Therefore, more
acidolysis products were formed because temperatures of around
conversion of 1 and also Tmax remains constant. The conversion of
2 increases to nearly quantitative values by adding another amount
of NaCl. These results show that the presence of small amounts of
salt is essential for generating near-critical water under microwave
radiation and therewith reaching high conversion of 1. Instead of
the 0.03 M NaCl solution, a lower concentrated NaCl solution is
also practicable to receive adequate conversion.
Because the ratio of quartz vessel volume (80 ml) to reaction
volume (15 ml) has to be constant in order to guarantee constant
reaction conditions, it is not possible to vary the ratio by simply
decreasing the amount of solvent. Nevertheless, increasing the
amount of 1 during one reaction cycle would help to reduce waste
and therefore improve the performance of the reaction considering
terms of sustainability or green chemistry. Experiments were
conducted performing the reaction with different amounts of
1 ranging from 0.25 to 2 ml. As demonstrated in Table 5, the
increased amount of 1 shows only a small effect on the conversion
of 1. By doubling the amount of reactant (from 0.25 to 0.5 ml)
the conversion of 1 decreases by about 15%. Four (1 ml) and
eight times (2 ml) more reactant as used in former experiments is
also practicable to receive moderate conversions of 1, whereby it
has to be mentioned that the product distribution remains constant
for all experiments and no polymers are formed. Experiments have
demonstrated that it is possible to reduce the ratio of solvent to
substrate.
◦
270 C are not high enough for initial homolytic rupture of the
bond between carbon atoms C1 and C6 (Scheme 2) initiating
pyrolysis reactions of 1.2,3 Another fact counting against thermal
isomerization via biradical intermediates is the fact that radical
reactions are strongly disfavored if bimolecular reactions are able
to take place. Due to high-pressure conditions, the probability
for collisions between two reactant molecules or between cationic
species (protons or proton-related species) and 1 or its products is
much higher compared to thermal processes in the gas phase.
Related to the investigations of Kremsner and Kappe, a 0.03 M
NaCl solution for the generation of near-critical water was applied
so far.14a In order to investigate the effect of NaCl concentration
on the conversion of 1 experiments with varied concentrations
(0–0.05 mol l-1 NaCl) have been performed (Table 4). Without
any amounts of salt, water becomes transparent for microwave
irradiation at higher temperature further affecting the energy
uptake and therefore the maximal temperature Tmax (220 ◦C) that
can be reached under salt-free reaction conditions, thus resulting
in conversions of 7 and 35% for 1 and 2, respectively. By adding
small amounts of NaCl (0.006 mol l-1) to the reaction mixture
the microwave absorption capacity of the bulk solvent can be
improved significantly pointed out by higher Tmax of 265 ◦C
and therewith increased conversion of both pinenes. Further
increase of NaCl concentration revealed only small effects on the
Rearrangement of a-pinene in supercritical alcohols
It has to be pointed out that the maximal temperature which can
be achieved under the experimental prerequisites (Tlim = 300 ◦C,
plim = 80 bar) in general depends on the applied alcohol (cf. Table 1
and ESI‡). Within the experiments it has been shown that the
limiting factor is neither the temperature nor the power input,
rather the pressure limit is reached first. According to theory
of ideal gases (law of Boyle–Mariotte) volume and pressure are
constrained variables. Increasing the free reaction volume (by
decreasing the volume of heated alcohol) would consequently
allow for the establishment of higher temperatures. However, basic
experiments in this direction revealed that this interrelation is very
complex. Therefore, the following experiments have been carried
out using a constant reaction volume.
Fig. 2 pictures the results from the thermal isomerization of 1
using lower supercritical aliphatic alcohols as solvents (Table 1).
Experimental outcomes revealed that besides the classical reac-
tions expected for the thermal isomerization,3,5 neither intermolec-
ular reactions between reactants or rectant and solvent occurred,
nor was the presence of products resulting from acidolysis route
Table 4 Effect of NaCl-concentration on the conversion of a-pinene (X1)
and b-pinene (X2)a
Concentration of NaCl solution/mol l-1
0
0.006
0.015
0.03
0.05
T
max/◦Cb
222
7
35
262
80
65
262
86
92
265
88
98
261
92
98
X1 (%)c
X2 (%)c
a 250 ml pinene, 15 ml NaCl solution, 80 ml quartz vessel, heating: 10 min,
reaction time: 15 min, cooling: 20 min, Pmax: 1.2 kW, microwave: Synthos
3000. b Maximal temperature of the reactor from recorded data. c Product
distribution is independent from conversion.
1564 | Org. Biomol. Chem., 2010, 8, 1560–1567
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
and therefore ruling out an involvement of alcohol molecules or
fragments in the rearrangement process.
As indicated in Fig. 2, the conversion of 1 strongly depends
on the alcohol applied as solvent for the experiment: the higher
the molar mass of the alcohol the higher the conversion is after
1 h of reaction time. Supercritical data for the alcohols listed in
Table 1 revealed opposite trends: the higher the molar weight
(i) the lower the critical pressure pc is, and (ii) the higher the
critical temperature Tc is, the higher the vapor pressure pv is.15b
Due to the lower pv of the C3- and C4-alcohols, the external
pressure limit (plim: 80 bar) is reached at a higher temperature level
and also the existence of the supercritical state occurred earlier
(reaction volume = constant). Due to the pressure–temperature
dependency, the maximal temperatures Tmax measured inside the
reaction vessels are higher, which directly effects the reaction rate
and therefore the conversion (Fig. 3). With the exception of butan-
1-ol there is a linear correlation between Tmax and conversion
of 1.
The alcohols used for the experiments proved to be inert under
the high temperature and pressure regime. Prior to use, the alcohols
were dried over molecular sieve before the reaction to avoid
catalytic amounts of water. The absence of water was proved
by determining residual water content by Karl–Fischer–Titration.
Water is known to react as strong base, and as an acid, under near-
and supercritical conditions, respectively, instead of a higher state
of autoprotolysis.14a,16a,19 Similar to water, alcohols may undergo
autoprotolysis themself (Table 2). However, the equilibrium con-
stants for these reactions are considerably low due to the disfavored
formation of protonated alcohol species. By the loss of water, the
protonated alcohols may form carbocations, which themselves are
strong electrophiles and can undergo various reactions. Besides
the formation of ethers also reaction with double bonds of 1 or
its pyrolysis products 2–5 may occur. Both the formation of ethers
and the formation of addition products can be ruled out, since the
expected products have not been detected in the reaction mixtures
(GC-MS, carbon mass balance for monoterpenes). This behavior
is in contrast to the studies of Kamitanaka et al. and Horikawa
et al. who reported addition of supercritical alcohols to styrene and
aliphatic n-alkenes.24 On the other hand, in a proton-rich medium,
acid-catalyzed rearrangement products of 1 could be found, e.g.
Fig. 2 Conversion of 1 and yield of 3–5 depending on the used alcohol
(500 ml 1, 30 ml alcohol, 80 ml quartz vessel; Tlim = 300 ◦C, plim = 80 bar,
Plim = 1.2 kW; heating: 10 min, reaction time: 60 min, cooling: 20 min,
microwave: Synthos 3000).
(Scheme 2) observed. Pyrolysis in supercritical alcohols yielded the
rearrangement products classified to the reaction pathway from
biradical 1 in a 3 : 7 ratio in favor of the formation of 3 (Scheme 1).
This is in clear contrast to the results observed when near-critical
water was employed as the solvent (Table 2). Additionally the
higher 3 : 4 ratio is contrary to the results from classical liquid
or gas phase pyrolysis.2,3d–g,5 The increase of limonene-selectivity
through pyrolysis in supercritical media was recently described by
Anikeev and coworkers for the use of water–ethanol mixtures (p:
120 bar).23 The beneficial effect was accounted to the increased
acidity of water due to change of its ion product.19 However,
in the present case, the presence of water can be ruled out,
hence the higher limonene ratio has to be attributed to other
circumstances. The independency of the product selectivity from
the solvent applied is traced back to the fact that the reaction
runs through a similar reaction intermediate, whose formation
does not depend on the solvent properties/reaction environment
Fig. 3 Correlation between critical pressure pc15b conversion of 1 (X1; 500 ml 1, 30 ml alcohol) and maximal temperature measured in the reaction vessel
(Tmax; left) and correlation between X1 and Tmax (right, squares = n-alcohols; 80 ml quartz vessel; Tlim = 300 ◦C, plim = 80 bar, Plim = 1.2 kW; heating:
10 min, reaction time: 60 min, cooling: 20 min, microwave: Synthos 3000).
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1560–1567 | 1565
©

View Article Online
menthadienes.7l,8,9,20,25 In the present case, only pyrolysis product
of 1 can be found. Both, the absence of water and of acid-catalyzed
rearrangement products of 1 imply that no protons are generated
during the reaction. Thus, it can be concluded that the alcohols
used as solvents for the isomerization of 1 under closed-vessel
conditions are practically inert towards inter- and intramolecular
reactions.
retention times and/or mass spectra of pure reference compounds.
GC-FID: HP 5, 30 m ¥ 0.32 mm ¥ 0.25 mm, 5 psi H2; program:
35 ◦C (hold 1 min), 4◦K min-1 up to 80 ◦C, 4.5 K min-1 up to 90 ◦◦C,
35 K min-1 up to 280 C (hold 3 min); injector temperature: 250 C;
detector temperature: 280 ◦C. GC-MS: HP 5, 30 m ¥ 0.32 mm ¥
0.25◦mm, 7 psi He; program: 55 ◦C (hold 1 min), 5 K min-1 up to
150 C, 20 K min-1 up to 280 ◦C (hold 5 min); injector temperature:
280 ◦C, EI (70 eV).
Conclusion
Microwave system
By using near-critical water or supercritical lower aliphatic
alcohols generated under the influence of microwave irradiation
in closed vessel reactors, it is possible to investigate pyrolysis
and acidolysis reactions of terpenes. Comparison of the two
solvent systems revealed significant differences in the experimental
outcome regarding the isomerization of a- (1) and b-pinene (2).
Since the improved availability of protons using near-critical water
the reaction afforded mainly products of the p-menthane series,
attributed to proton-catalyzed rearrangement mechanism, as the
addition of other acid-catalyzing reactants like mineral acids,
zeolites or heteropolyacids is not necessary. Within a reaction
time of 15 min a quantitative conversion of 1 and 2 is achieved
at temperatures of about 270 ◦C and a system pressure of 80
bar. Independently of both inert and non-inert conditions, the
formation of polymers or oxidized products was not observed in
any experiment. However, NaCl used for stable reaction condition
shows no catalytic effect on the reaction mixture. Rearrangement
of 1 afforded products resulting from acid-catalyzed rearrange-
ment passing through cationic intermediates and thermal induced
isomerization via biradical transition states, strongly disfavoring
the latter route (5 : 1).
Regarding the pyrolysis of a-pinene (1) in supercritical alcohols
typical compounds were also identified that were reported to
yield from pyrolysis under classical conditions: limonene (3),
alloocimenes (4), and pyronenes (5). Contrarily to earlier reports
on the pyrolysis of 1, the reaction favors that reaction route
furnishing 2 over that one yielding 3 and 4. Within the microwave-
assisted heating of alcohols up to their critical parameters, the
established pressure inside the quartz vessel is auto-generated
depending on the applied solvent. In the case of methanol and
ethanol, the pressure reaches its limitation (80 bar) with the
consequence that the critical temperature can not be reached
leading to low conversions of 1. The higher alcohols are able to
reach the supercritical state and resulting in the total conversion
of 1. The solvent itself is stable in the supercritical state and does
not dehydrate or oxidize.
The experiments were carried out in the microwave system Synthos
3000 (Anton Paar GmbH, Graz, Austria). The microwave oven is
equipped with two magnetrons within a frequency of 2.45 GHz
and an overall possible energy-input ranging from 0 to 1.4 kW. In
the middle of the reaction space (w-d-h: 450 ¥ 420 ¥ 350 mm) a
rotor is placed, which contains eight positions for quartz vessels.
The 80 ml quartz vessels have a thickness of 0.8 cm and can
be filled with maximal 60 ml of reaction solution. The vessels
were closed by caps made of PTFE. The so prepared vessels were
place in a security jacket made of polyetheretherketone PEEK.
Then the vessels were placed into the rotor and this was closed
with a security cap. The quartz vessels are approvable for pressure
up to 80 bar and temperatures up to 300 ◦C. The pressure was
determined by a hydraulic system while the cap of the vessels press
against a stamp installed in the rotor. Temperature was determined
by a gas-thermometer in only one quartz vessel. Additionally
the temperature of all vessels was controlled by an IR-sensor.
The power of the microwave irradiation could be controlled via a
pressure-, temperature- or power-regulated program. In this case
the power input was limited by temperature Tlim = 300 ◦C, pressure
plim = 80 bar, and power Plim = 1.2 kW. For heating up to Tlim 10 min
and for cooling down 20 min (pressurized air) were used.
Details concerning the heating behavior of near-critical water
and supercritical alcohols under the experimental prerequisites are
provided within the ESI.‡
Experimental procedure for conversion in near-critical water
To an aqueous solution of NaCl (15 ml, 0.03 M) saturated with
nitrogen, a-pinene (1.57 mmol, 215 mg, 250 ml) was added into
the quartz vessel equipped with a magnetic stirring bar. Four as-
prepared vessels were placed into the rotor. The first reactor was
equipped with a gas-thermometer. After closing the rotor with the
security cap it was placed into the microwave oven. The predicted
microwave program was started. After the reaction was finished,
the mixture was extracted with ethyl acetate (15 ml). The organic
phase was separated and analyzed by FID- and GC-MSD.
Experimental section
Experimental procedure for conversion in supercritical alcohol
General
To degassed and water-free alcohol (30 ml; cf. Table 1) a-pinene
(3.15 mmol, 430 mg, 500 ml) was added into the quartz vessel
equipped with a magnetic stirring bar and three passive heating
elements (PHE, SiC, cylindrical r: 2.5 mm, h: 10 mm; Anton Paar
GmbH). Four as-prepared vessels were placed into the rotor. The
first reactor was equipped with a gas-thermometer. After closing
the rotor with the security cap was placed into the microwave oven.
The predicted microwave-program was started. After the reaction
1 ml of the reaction mixture was analyzed by FID- and GC-MSD.
a-Pinene (1, 98%), b-pinene (2, 98%), limonene (3, 97%), ter-
pinolene (8, 98%), g-terpinene (9, 97%), and a-terpinene (10,
95%) were purchased from Sigma-Aldrich and used without
further purification. Purity was determined by capillary gas
chromatography. Water used for reaction has a purity of 99.9999%
and a conductivity of ≤ 2 mS cm-1. Analyses were carried out with
a 6890 Series GC-MSD and a 6890 Series II GC-FID from Agilent
Technologies. Products were identified by comparison with either
1566 | Org. Biomol. Chem., 2010, 8, 1560–1567
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
10 A. I. Allahverdiev, G. Gu¨ndu¨z and D. Yu Murzin, Ind. Eng. Chem. Res.,
1998, 37, 2373–2377.
11 E. L. Foletto, A. Valentini, L. F. D. Probst and L. M. Porto, Lat. Am.
Appl. Res., 2002, 32, 141–144.
12 H. Yeon Cho, A. Ajaz, D. Himali, P. A. Waske and R. P. Johnson,
J. Org. Chem., 2009, 74, 4137–4142.
13 Recent reviews on microwave-assisted organic chemistry: (a) C. O.
Kappe and D. Dallinger, Mol. Diversity, 2009, 13, 71–193; (b) C. O.
Kappe, Chem. Soc. Rev., 2008, 37, 1127–1139; (c) V. Polshettiwar and
R. S. Varma, Acc. Chem. Res., 2008, 41, 629–639; (d) U. R. Pillai, E.
Sahle-Demessi and R. S. Varma, J. Mater. Chem., 2002, 12, 3199–3207;
(e) G. S. Vanier, Synlett, 2007, 131–135.
Notes and references
1 (a) M. Eggersdorfer, Terpenes, in Ullmann’s Encyclopedia of Industrial
Chemistry, ed. W. Gerhartz, C. Mayer, R. Pfefferkorn and H. Buchholz-
Meisenheimer, Wiley-VCH, Weinheim, 6th edn, 2002, vol. 35 pp. 653–
671; (b) J. O. Bledsoe Jr., Terpenoids, in Kirk-Othmer Encyclopedia of
Chemical Technology, ed. J. I. Kroschwitz and M. Howe-Grant, John
Wiley & Sons, New York, 4th edn 1997, vol. 23 pp. 833–882; (c) E.
Breitmeier, Terpene, Wiley-VCH, Weinheim, 2nd edn, 2005; (d) H.
Surburg and J. Panten, Common Fragrance and Flavour Materials,
Wiley-VCH, Weinheim, 5th edn, 2006.
2 (a) A. Stolle, B. Ondruschka and H. Hopf, Helv. Chim. Acta, 2009, 92,
1673–1719; (b) A. Stolle and B. Ondruschka, J. Anal. Appl. Pyrolysis,
2009, 85, 252–259.
14 (a) J. M. Kremsner and C. O. Kappe, Eur. J. Org. Chem., 2005, 3672–
3679; (b) C. R. Strauss and R. W. Trainor, Aust. J. Chem., 1995, 48,
3 (a) L. A. Goldblatt and S. Palkin, J. Am. Chem. Soc., 1941, 63, 3517–
3522; (b) T. R. Savich and L. A. Goldblatt, J. Am. Chem. Soc., 1945, 67,
2027–2031; (c) R. L. Burwell, J. Am. Chem. Soc., 1951, 73, 4461–4462;
(d) K. Riistama and O. Harva, Finn. Chem. Lett., 1974, 4, 132–138;
(e) J. J. Gajewski, I. Kuchuk, C. Hawkins and R. Stine, Tetrahedron,
2002, 58, 6943–6950; (f) A. Stolle, B. Ondruschka and W. Bonrath,
Eur. J. Org. Chem., 2007, 2310–2317; (g) A. Stolle, B. Ondruschka and
M. Findeisen, J. Org. Chem., 2008, 73, 8228–8235.
4 (a) J. E. Hawkins and J. W. Vogh, J. Phys. Chem., 1953, 57, 902–
905; (b) A. Stolle, C. Brauns, M. Nu¨chter, B. Ondruschka, W.
Bonrath and M. Findeisen, Eur. J. Org. Chem., 2006, 3317–3325;
(c) M. B. Kolicheski, L. C. Cocco, D. A. Mitchell and M. Kaminski,
J. Anal. Appl. Pyrolysis, 2007, 80, 92–100; (d) A. Stolle and B.
Ondruschka, J. Anal. Appl. Pyrolysis, 2008, 81, 136–138; (e) A. Stolle,
B. Ondruschka and W. Bonrath, J. Anal. Appl. Pyrolysis, 2008, 83, 26–
36.
5 (a) A. M. Chibiryaev, A. Yermakova, I. V. Kozhevnikov, O. I.
Sal’nikova and V. I. Anikeev, Russ. Chem. Bull., 2007, 56, 1234–1238;
(b) A. M. Chibiryaev, V. I. Anikeev, A. Yermakova, P. E. Mikenin, I. V.
Kozhevnikov and O. I. Sal’nikova, Russ. Chem. Bull., 2006, 55, 987–
992; (c) A. Yermakova, A. M. Chibiryaev, I. V. Kozhevnikov and V. I.
Anikeev, Chem. Eng. Sci., 2008, 62, 2414–2421; (d) V. I. Anikeev, A.
Ermakova, A. M. Chibiryaev, I. V. Kozhevnikov and P. E. Mikenin,
Russ. J. Phys. Chem. A, 2007, 81, 711–716.
¨
1665–1692; (c) A. A. Peterson, F. Vogel, R. P. Lachance, M. Froling,
M. J. Antal Jr. and J. W. Tester, Energy Environ. Sci., 2008, 1, 32–
65; (d) Y. Matsumura, T. Minowa, B. Potic, S. R. A. Kersten, W.
Prins, W. P. M. van Swaaij, B. van de Beld, D. C. Elliott, G. G.
Neuenschwander, A. Kruse and M. J. Antal Jr., Biomass Bioenergy,
2005, 29, 269–292; (e) A. Loppinet-Serani, C. Aymonier and F. Cansell,
ChemSusChem, 2008, 1, 486–503; (f) Y. Matsumura, M. Sasaki, K.
Okuda, S. Takami, S. Ohara, M. Umetsu and T. Adschiri, Combust.
Sci. Technol., 2006, 178, 509–536.
15 (a) B. Hayes, Microwave Synthesis, CEM Publishing, Mattews, 2002,
chapter 2; (b) D. R. Lide (chief editor), Handbook of Chemistry and
Physics 83rd edition, CRC Press, Boca Raton, 2002, pp. 6-49–6-62
and 6-4 (critical data), pp. 6-66–6-93 and 6-10–6-11 (vapor pressure);
(c) D. R. Lide (chief editor), Handbook of Chemistry and Physics 83rd
edition, CRC Press, Boca Raton, 2002, p. 8-85 (water); (d) C. Reichardt,
Solvents and Solvent Effects in Organic Chemistry, VCH, Weinheim,
1990, pp. 431–433.
16 (a) J. M. Kremsner and C. O. Kappe, J. Org. Chem., 2006, 71, 4651–
4658; (b) Properties of Silicon Carbide, ed. G. L. Harris, Institute of
Electrical Engineers, London, 1995.
17 T. Sasaki, S. Eguchi and H. Yamada, Tetrahedron Lett., 1971, 12, 99–
103.
18 (a) K. J. Crowley and S. G. Traynor, Tetrahedron Lett., 1975, 16, 3555–
3558; (b) K. J. Crowley and S. G. Traynor, Tetrahedron, 1978, 34, 2783–
2789.
6 K. P. Volcho and N. F. Salakhutdinov, Mini-Rev. Org. Chem., 2008, 5,
345–354.
19 H. Weinga¨rtner and E. U. Franck, Angew. Chem., 2005, 117, 2730–
7 (a) C. M. Williams and D. Whittaker, J. Chem. Soc. B, 1971, 668–677;
(b) K. L. Stevens, L. Jurd and G. Manners, Tetrahedron, 1972, 28, 1939–
1944; (c) N. Bluthe, J. Ecoto, M. Fetizon and S. Lazare, J. Chem. Soc.,
Perkin Trans. 1, 1980, 1747–1751; (d) R. Muneyuki, Y. Yoshimura, K.
Tori, Y. Terui and J. N. Shoolery, J. Org. Chem., 1988, 53, 358–366;
(e) M. C. C. Costa, R. A. W. Johnstone and D. Whittaker, J. Mol.
Catal. A: Chem., 1996, 104, 251–259; (f) S. Findik and G. Gu¨ndu¨z,
J. Am. Oil Chem. Soc., 1997, 74, 1145–1151; (g) A. I. Allahverdiev,
G. Gu¨ndu¨z and D. Y. Murzin, Ind. Eng. Chem. Res., 1998, 37, 2373–
2377; (h) A. I. Allahverdiev, S. Irandoust and D. Y. Murzin, J. Catal.,
1999, 185, 352–362; (i) Y. P. Bazhenov, L. Z. Kas’yanova, A. I. Bokin,
B. I. Kutepov, A. N. Khazipova, E. A. Travkin, N. A. Shchadneva,
R. I. Khusnutdinov and U. M. Dzhemilev, Russ. J. Appl. Chem., 2003,
76, 234–237; (j) N. A. Comelli, E. N. Ponzi and M. I. Ponzi, J. Am.
Oil Chem. Soc., 2005, 82, 531–535; (k) C. Fernandes, C. Catrinescu, P.
Castilho, P. A. Russo, M. R. Carrott and C. Breen, Appl. Catal., A,
2007, 318, 108–120; (l) R. Rachwalik, Z. Olejniczak, J. Jiao, J. Huang,
M. Hunger and B. Sulikowski, J. Catal., 2007, 252, 161–170; (m) I. V.
Il’ina, K. P. Volcho and N. F. Salakhutdinov, Russ. J. Org. Chem., 2008,
44, 1–23.
2752, (Angew. Chem., Int. Ed., 2005, 44, 2672–2692).
20 N. Flores-Holguın, A. Aguilar-Elguezabal, L. M. Rodrıguez-Valdez
and D. Glossman-Mitnik, THEOCHEM, 2008, 854, 81–88.
21 W. R. Rocha, H. M. S. Milagre and W. B. De Almeida, THEOCHEM,
2001, 544, 213–220.
´
´
´
22 (a) A. Stanislaus and L. M. Yeddanapalli, Can. J. Chem., 1972, 50,
113–118; (b) W. F. Ho¨lderich, Stud. Surf. Sci. Catal., 1993, 75, 127–
163; (c) P. A. Weyrich and W. F. Ho¨lderich, Appl. Catal., A, 1997, 158,
145–162; (d) D. Buhl, D. M. Roberge and W. F. Holderich, Appl. Catal.,
¨
A, 1999, 188, 287–299.
23 A. Ermakova, A. M. Chibiryaev, P. E. Mikenin, O. I. Sal’nikova and
V. I. Anikeev, Russ. J. Phys. Chem. A, 2008, 82, 62–67; V. I. Anikeev,
A. Ermakova, P. E. Mikenin, A. M. Chibiryaev, and O. I. Sal’nikova,
RU Pat. 2 320 630, 2008.
24 (a) T. Kamitanaka, T. Hikida, S. Hayashi, N. Kishida, T. Matsuda and
T. Harada, Tetrahedron Lett., 2007, 48, 8460–8463; (b) Y. Horikawa, Y.
Uchino and T. Saku, Chem. Lett., 2003, 32, 232–233.
25 (a) R. Muneyuki, Y. Yoshimura, K. Tori, Y. Terui and J. N. Shoolery,
J. Org. Chem., 1988, 53, 358–366; (b) M. Roma´n-Aguirre, Y. P. Gochi,
A. R. Sanchez, L. de la Torre and A. Aguilar-Elguezabal, Appl. Catal.,
A, 2008, 334, 59–64; (c) K. A. da Silva Rocha, P. A. Robles-Dutenhefner,
I. V. Kozhevnikov and E. V. Gusevskaya, Appl. Catal., A, 2009, 352,
188–192.
´
8 C. M. Lo´pez, F. J. Machado, K. Rodr´ıguez, D. Arias, B. Me´ndez and
M. Hasegawa, Catal. Lett., 1999, 62, 221–226.
9 F. Ebmeyer, THEOCHEM, 2002, 582, 251–255.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1560–1567 | 1567
©