Ring-Opening Reactions of α- And β-Pinenes in Pressurized Hot Water in the Absence of Any Additive

Article abstract of DOI:10.1021/op4000863

Reactions of α- and β-pinenes in pressurized hot water were examined in a batch reactor made of a SS316 1/2-in. tube at temperatures of 250-400 C, pressures of 4-30 MPa, and reaction times of 1-30 min in the absence of any additive under an argon atmosphere. The maximum yields of limonene from α-pinene were ca. 70% in 20 min at 300 C or 1 min at 400 C. Limonene was obtained from β-pinene in ca. 16% yield for 30 min at 300 C and 1 min at 400 C. Reversible production of myrcene in 14% yield and formation of unidentified C₂₀ dimer fractions were noted for 1 min at 370 C from β-pinene. The conversion of α-pinene to limonene took place under anhydrous conditions, albeit at slightly lower yield of 65% compared to processes conducted in the presence of water, where increased limonene yield of 70% was observed for 1 min at 400 C. The conversion of β-pinene to limonene under anhydrous conditions was limited to 6.1% in contrast to 11.9% in the presence of water for 7 min at 370 C. In the presence of oxygen, p-cymene was formed in 23% and 24% yield at the expense of limonene from α- and β-pinenes, respectively, for 30 min at 400 C.

Full text of DOI:10.1021/op4000863

Article
pubs.acs.org/OPRD
Ring-Opening Reactions of α- and β‑Pinenes in Pressurized Hot
Water in the Absence of Any Additive
Tomomi Kawahara,† Yui Henmi,† Natsu Onda,† Toshiyuki Sato,† Akiko Kawai-Nakamura,† Kiwamu Sue,§
Hiizu Iwamura,‡,* and Toshihiko Hiaki*,†
†College of Industrial Technology, Nihon University, 1-2-1 Izumi-cho, Narashino, Chiba 275-8575, Japan
‡College of Science and Technology, Nihon University, 1-8-14 Kanda-Surugadai, Chiyoda-ku, Tokyo 101-8308, Japan
§Nanosystem Research Institute, National Institute of Advanced Industrial Science and Technology, Tsukuba Central 5, 1-1-1 Higashi
Tsukuba, Ibaraki 305-8565, Japan
S
* Supporting Information
ABSTRACT: Reactions of α- and β-pinenes in pressurized hot water were examined in a batch reactor made of a SS316 1/2-in.
tube at temperatures of 250−400 °C, pressures of 4−30 MPa, and reaction times of 1−30 min in the absence of any additive
under an argon atmosphere. The maximum yields of limonene from α-pinene were ca. 70% in 20 min at 300 °C or 1 min at 400
°C. Limonene was obtained from β-pinene in ca. 16% yield for 30 min at 300 °C and 1 min at 400 °C. Reversible production of
myrcene in 14% yield and formation of unidentified C20 dimer fractions were noted for 1 min at 370 °C from β-pinene. The
conversion of α-pinene to limonene took place under anhydrous conditions, albeit at slightly lower yield of 65% compared to
processes conducted in the presence of water, where increased limonene yield of 70% was observed for 1 min at 400 °C. The
conversion of β-pinene to limonene under anhydrous conditions was limited to 6.1% in contrast to 11.9% in the presence of
water for 7 min at 370 °C. In the presence of oxygen, p-cymene was formed in 23% and 24% yield at the expense of limonene
from α- and β-pinenes, respectively, for 30 min at 400 °C.
substances, primarily clays, minerals, and inorganic salts, have
been studied as catalysts, the objectives having been to increase
the yield of camphene and reduce polymer formation.12,13 The
yield of limonene decreases and that of allo-ocimene increases
with increasing temperature.7,8 Upon heterogeneous catalytic
reactions in the liquid phase over a calcinated clay or cobalt
molybdite, α-pinene decomposes rapidly to camphene and
limonene in nearly equal amounts, followed by slower reaction
to other monoterpenes, dimers, and polymers at the reflux
temperature of 156−172 °C. α-Terpineol is synthesized by
treating α- and β-pinenes with dilute acids such as aqueous
sulfuric acid,14 formic acid, or phosphoric acid15 to give terpin
hydrate, which in turn is dehydrated with more concentrated
acids.
In contrast to the acid- or base-catalyzed treatment of the
bicyclic hydrocarbons, we became interested in their behavior
in hot pressurized water to see if the development of such a
conversion process without addition of any catalyst might be
possible. Physicochemical properties of water change drastically
near or above its critical points (TC = 374.15 °C, PC = 22.12
MPa) and provide water with various merits as alternative
reaction media. Its high specific dielectric constant (εr) of 78 at
room temperature drops continuously to about 28 at 250 °C
(cf. εr = 28 of ethanol and 33 of methanol at 25 °C) and ca. 2 at
450 °C (cf. εr = 2.3 for benzene at 25 °C under atmospheric
pressure) as the hydrogen bond is broken at higher
temperature.16,17 Thus, near the critical point, as vapor and
INTRODUCTION
■
α- and β-Pinenes are typical monoterpenes distributed in large
amounts in pine and other coniferous trees and released to the
atmosphere, by logging operations, and from kraft pulp mills.
They occur in wood turpentine obtained by the steam-
distillation of chopped tree trunks and dead wood, crude sulfate
turpentine (CST) obtained from wood pulp, as a waste product
in the manufacture of cellulose via the sulfate process, black
liquor usually burned to help save the fuel, and finally in
wastewater. Worldwide annual productions of turpentine in
which pinenes are the major components have been estimated
at around 330,000 tons.1
In view of its wide distribution in nature, chemical
conversions of pinenes have been studied as early as in the
nineteenth century.2,3 No significant pyrolysis of α- and β-
pinenes takes place for a short contact time at 250−300 °C.4
However, myrcene is formed from the latter with a maximal
overall yield of 77% in gas phase reactions with a contact time
of 0.5−2.5 s at 300−600 °C.5,6 This method of the pyrolysis of
β-pinene has been used for the commercial production of
myrcene at 550−600 °C. Effective pyrolysis in the gas phase is
usually carried out over solid catalysts such as copper
chromite,7,8 amorphous zirconium phosphate,4 and Pd powder
in the presence of hydrogen.9,10 Limonene and dipentene are
always formed with various other products, and the selectivity is
low. In the last example, high yield (65%) of p-cymene is
achieved when gaseous pinenes are passed through a tube (i.d.
6 mm) filled with Pd at 300 °C.9,10 Since the liquid phase
isomerization and polymerization of α-pinene by heteroge-
neous catalysis was reported in 1915,11 a large variety of acidic
Received: April 4, 2013
Published: November 12, 2013
© 2013 American Chemical Society
1485
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491

Organic Process Research & Development
Article
liquid phases become indistinguishable, organic compounds are
soluble in water and inorganic compounds such as salt become
insoluble. The ionic dissociation of water is endothermic and
therefore enhanced as the temperature is raised in accordance
with Le Chatelier’s principle. Since the dielectric constant
decreases as the temperature approaches the critical point, the
ion product of water reaches a maximum of pKW = 11.2 at ca.
200 °C under its saturated vapor pressure,18,19 thereby
facilitating various acid- or base-catalyzed organic reactions.
All these properties are temperature-dependent and can be
tuned to optimize the reaction conditions. If successful, it might
be possible to attain acidic or basic conditions without adding a
strong acid or base, and yet require no additional workup
processes of neutralization with base or acid, respectively, after
the reactions. Dielectric properties would allow us to carry out
the reactions under homogeneous condition and yet the
organic layer and water separate after the reaction under
ambient conditions. Other than the above factors favoring
reactions in hot water, the rate of any reaction with a positive
activation enthalpy should be accelerated at the expense of the
selectivity at higher temperature. However, the manufacturing
of the high-pressure reactors and the electricity required for
carrying out such high-temperature reactions have to be taken
into account if the merits of the above potential economic and
environmental improvements are claimed seriously.
There are a number of examples for the application of
pressurized hot water to the treatment of biomass. They are
mostly focused on decomposition of woods and hydrolysis/
gassification of lignocellulose.20,21 The resulting glucose is
reported to undergo retro-aldol condensation, dehydration,
tautomerization, and hydration. The formation of furfural in
13% yield for 0.8 min at 400 °C and 80 MPa is an example.22
Our attention was paid to the processing of fine chemicals
under these conditions as generally reviewed in references.23−27
Limonene most probably produced efficiently from α- and/or
β-pinenes should be used as fragrance and for the synthesis of
d-carvone, a fragrant present in spearmint oil. Myrcene will
serve as a starting material for the industrial production of l-
menthol. p-Cymene is a mild antiseptic and used for the
synthesis of thymol and other chemicals.
yield of ca. 70% was reached (Figure 1). It is clearly seen that
the yield decreased with time at temperatures higher than 300
Figure 1. Effect of temperature (250−400 °C) and reaction time (1−
30 min) on limonene yield from α-pinene under argon.
°C. Secondary decomposition/fragmentation reactions of
limonene seemed to set in sooner at higher temperature as
suggested by a number of minor peaks in GC-FID chromato-
gram.
Allo-ocimene (= 2,6-dimethylocta-2,4,6-triene) was obtained
in 16.2% in 10 min at 300 °C (Table 1, Scheme 1). A
Table 1. Effect of Reaction Time (1−30 min) on α-Pinene
Conversion and Limonene and Allo-Ocimene Yields at 300
a
°C under Argon
reaction time
/min
unconverted α-
pinene /%
limonene
produced /%
allo-ocimene
produced /%
1
64.8
0.63
0.15
0.14
18.9
59.3
70.0
65.3
8.1
16.2
13.5
8.6
10
20
30
a
See also Table S1.
Scheme 1. Reaction Course of α-Pinene
RESULTS AND DISCUSSION
■
Batch reactions in pressurized hot water were carried out in a
reactor made of a SS316 1/2-in. tube of an effective inner
volume of ca. 10 cm3 that was dipped into a preheated salt bath
as detailed in the Experimental Section.
I. Reactions of α-Pinene, β-Pinene, and Myrcene in
Near-Critical and Supercritical Water under Argon. α-
Pinene disappeared gradually in pressurized hot water with half-
lives of ca. 20 and 1.5 min at 250 and 300 °C, respectively. No
unconverted α-pinene was detected after 1 min at 400 °C (see
Table S1). Limonene was the main product and its maximum
yield was ca. 70% in 20 min at 300 °C and in 1 min at 400 °C.
Limonene had never been obtained in yields as high as 70%
from α-pinene without using a catalyst, and this yield of
limonene under near-critical and supercritical water is
comparable to the reported highest yields of limonene from
α-pinene using acidic catalyst in the literature.4,7,8 The lack of
formation of significant amounts of dimers and polymeric
materials from α-pinene as judged from GC-FID chromato-
gram (Figure S2) appears to be unique to the pressurized hot
water reaction. As the temperature was decreased from 400 °C,
it took longer reaction times before the maximum limonene
chromatographic peak that appeared at ca. 0.7 min later than
that of allo-ocimene was negligible at the reaction time of 3 min
and grew in at 20 min (see Figure S2). The latter peak was
tentatively assigned to the formation 2,6-dimethyl-1,3,5-
octatriene at the expense of allo-ocimene.
During preliminary experiments, α-pinene was subjected to
flow reactor conditions in which limonene was obtained in 63%
yield under contact time of 15.0 s with the molar ratio of α-
pinene to water of 1:100 at 400 °C and 30 MPa. No other
monoterpene nor diterpene was detected under this short
contact time.30
1486
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491

Organic Process Research & Development
Article
Conversion of β-pinene was slightly slower than the α-isomer
in water with half-lives of ca. 46 and 8 min at 250 and 300 °C,
respectively. Myrcene and limonene were produced at similar
levels of 14−16% from β-pinene at 370 °C and 21 MPa (Table
S2). Similarly to the case of α-isomer, the yield of limonene
decreased with time at temperatures higher than 300 °C
(Figure 2). The yield of myrcene did not increase substantially
Scheme 2. Interconversion of Chemicals Related to β-Pinene
in Pressurized Hot Water
Since myrcene and β-pinene have the standard molar
enthalpy of formation of 98.8 and 38.7 kJ/mol at 298 K,
respectively, the formation of myrcene from β-pinene is an
endothermic reaction by ca. 60 kJ/mol and therefore should be
favored thermodynamically as the temperature is increased.
The diradical intermediate formed by the homolysis of the
cyclobutane ring is considered to reside near the late transition
state of the reaction path and undergo β-cleavage to give
myrcene (Scheme 2).5,6 The reverse reaction from myrcene to
β-pinene has never been reported and provides novel product
finding and postulated mechanism. There are at least two
representative mechanisms possible for this reverse process.
One is a microscopically reversible diradical mechanism in
which the exothermic reaction product has to be quenched
thermally by its collision with a second body for isolation at
least in the gas phase. It might be possible that near-critical
water molecules could have served as quenchers of the excess
internal energy of the formed β-pinene molecules. Alternatively,
the reverse reaction observed in this study could be ionic as are
many terpenoid ring closure reactions under acid catalysis and
take a route through a cyclobutylmethyl carbonium ion
different from the pyrolytic forward reaction (Scheme 2).
II. Effect of the Amount of Water on the Reaction of
α- and β-Pinenes in Supercritical Water under Argon. A
series of experiments were performed for 1 min at 400 °C by
changing the molar amount of water in the range 0−300
relative to one mole of α-pinene as starting materials. The
results are given in Table S4 and Figure 4. Under these
conditions the amount of the unconverted α-pinene was less
than 1%. Limonene was obtained in 65% yield even in the
Figure 2. Effect of temperature (250−400 °C) and reaction time (1−
30 min) on limonene yield from β-piene under argon.
compared to pyrolysis studies in the literature.5,28,29 As shown
in a typical GC-FID chromatogram given in Figure S3, β-
pinene always gave other products: many small peaks in the C10
and several peaks in the diterpene C20 regions remained to be
identified. In preliminary experiments, β-pinene was subjected
to a flow reactor in which limonene and myrcene were obtained
in 24.3% and 9.4% yield, respectively, under contact time of 1.9
s with the molar ratio of β-pinene to water of 1:300 at 370 °C
and 21 MPa. The GC peaks in the diterpene C20 region
increased considerably in the contact time of 20 s.31,32
Since myrcene seemed far from the end product of β-pinene
in pressurized hot water, it was heated at 370 °C in pressurized
hot water in separate batch reactions. It disappeared rapidly and
limonene and β-pinene were detected in 6% and 2% yields,
respectively, in 1 min, both of which decreased gradually. The
C20 fractions were also detected. Their temporal variation is
given in Table S3 and Figure 3. Interconversion among these
hydrocarbons as depicted in Scheme 2 is concluded.
Figure 3. Effect of reaction time (1−30 min) on myrcene conversion
and selectivity to limonene and β-pinene at 370 °C.
Figure 4. Effect of the amount (0−300 molar) of water relative to one
mole of α-pinene on the yield of limonene for 1 min at 400 °C.
1487
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491

Organic Process Research & Development
Article
absence of water for a short contact time of 1 min. The yield of
limonene increased slightly with increasing relative amount of
water, reaching the highest yield of 70% at the water to
substrate molar ratio of 50−100:1. However, it dropped to ca.
30% when the molar amount of water relative to α-pinene was
increased to 150, 200, and 300:1. The decrease in the yield of
limonene, namely, the decomposition of limonene in the
presence of larger amount of water is somewhat puzzling. The
reaction time of 1 min is slightly ambiguous for the batch
reactions of immersing reaction vessels into a preheated salt
bath in that the tube might have thermal gradients from the
reaction vessel wall to the center of the tube and the
temperature of the reactants might not be equilibrated at that
temperature.
Therefore, a series of experiments on β-pinene were
performed by changing the molar amount of water in the
range of 0−400 relative to one mole of β-pinene as a starting
material at longer reaction time, 7 min, at slightly lower
temperature, 370 °C. The results are given in Table S5 and
Figure 5. Just as in the α-isomer, limonene was obtained in the
unique reactivity under these conditions. Second, the observed
increase in the yield of limonene with water could be
interpreted in terms of the contribution of the added acid-
catalyzed reaction pathway (Scheme 1) that is typical for the
production of limonene from α-pinene in the presence of acid
catalysts in the literature.12,13,28,29 Further support may be
found in a recent report by Anikeev showing that the formation
of limonene from α-pinene in supercritical ethanol at 384 °C
and 23.3 MPa proceeded via diradical mechanism and the yield
increased gradually by addition of water. The latter effect was
interpreted in terms of the growth of the hydronium ion
(H3O+) concentration.33 For a third, the observed decrease in
the limonene yield as the relative amount of water was
increased over 150 °C (Figure 4) and 350 °C (Figure 5) is still
a problem that remains to be solved.
III. Reactions in the Presence of Oxygen. When α-
pinene in pure water was not flushed with argon, the ca. 70%
yield of limonene decreased to 41% in 1 min and continued to
decrease with time. In 5 min, the product limonene was
overtaken by p-cymene and its yield reached ca. 23% in 30 min
at 400 °C (Table S6, Figure 6). Throughout the reaction, α-
pinene was consumed.
Figure 5. Relationship of molar ratio of water to β-pinene on the
molar yield of limonene for 7 min at 370 °C and 21 MPa.
Figure 6. Effect of reaction time (1−30 min) on the yields of limonene
and p-cymene from α-pinene at 400 °C and 30 MPa in the presence of
oxygen.
absence of water though in reduced 6.1% yield. The yield of
limonene increased to 11.9% at the molar ratio 300:1 of water
to β-pinene: an increase of almost twice as compared to the
reaction in the absence of water.
p-Cymene is formed from α-pinene when heated with
various catalysts.9,10 Mesoporous silica-supported 12-tungste-
nophosphoric acid is also effective in these reactions.34 Since
the presence of air was suggested to be responsible for the
present dehydrogenative aromatization reaction (Scheme 3), it
was hoped that limonene and p-cymene might be selectively
synthesized from the reaction of α-pinene in hot pressurized
water by controlling the amount of oxygen. However,
saturation of water with oxygen and filling the empty space
of the reactor with oxygen gas did not improve the yield of p-
The trends observed for the effect of the amount of water on
the two isomeric pinenes (Figures 4 and 5) are qualitatively
similar in spite of the lower reaction temperature and longer
reaction time for β-pinene. Contrary to the low reactivity of the
pinenes in the absence of any catalyst expected from the
literature references in the Introduction, they showed
significant conversion to limonene and other products at
water/pinene = 0. We recall that the reactor wall made of SS
316 contains Ni, Cr, and Mn components which may act as
catalyst for such reactions as hydrogenation. However,
seasoned reaction vessels have, to our knowledge, not shown
accelerating effect for pyrolytic reactions. The reaction of α-
pinene to limonene is estimated from their heats of formation
to be exothermic by 38 kJ/mol, and therefore the diradical
intermediate in Scheme 2 is considered to reside near the early
transition state of the reaction path. The pinenes under our
reaction conditions in the absence of water are in or near the
supercritical conditions of their own: critical points are 356.70
°C, 2.93 MPa for α-pinene and 371.35 °C, 2.92 MPa for β-
pinene. Further studies are necessary to see if they exhibit their
cymene (yield 23
2%) and the vessel was soot-blackened
Scheme 3. Dehydrogenative Aromatization of α-Pinene
without Exclusion of Air
1488
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491

Organic Process Research & Development
Article
after 60 min. Addition of dilute aqueous hydrogen peroxide did
not improve the yield of p-cymene.
different carbocations, and two common cyclobutylcarbonium
ion.20 The last possibility is easily excluded.
Water under near-critical and supercritical conditions proved
to be a novel media for ring-opening reactions of α- and β-
pinenes. Lowering of its dielectric constant to those of organic
solvents at ambient temperature and increase of the ion
product almost by 3 orders of magnitude are considered to be
responsible for the reactions. It goes without saying that
thermal activation of the reaction by a factor lnT and provision
for nonflammable media for combustible reactants are added
merits of near-critical and supercritical water in addition to the
above two factors.
Without exclusion of oxygen, the yield of limonene in ca.
11% from β-pinene also decreased rapidly with time and was
overtaken by that of p-cymene in 10 min. Its yield reached ca.
24% in 30 min at 400 °C (Table S7) with neither β-pinene nor
limonene remaining.
Since limonene was considered to be an intermediate from
both isomeric pinenes to p-cymene, we treated limonene in hot
pressurized water in the presence of oxygen. The results are
given in Table S8 and Figure 7. Limonene was stable in the
absence of water.
Whereas the size of the reaction was limited by the reaction
vessel used in these exploratory studies (Figure 8), it could
Figure 8. Batch reactor handmade from a SUS316 1/2-in. tube.
readily be expanded by employing larger size reactors. One of
the highest yields (70%) of limonene from α-pinene was
obtained at 9 MPa and 300 °C in 20 min. Under these
conditions, it must not be difficult to find a commercially
available pressure bottle (an autoclave) of the inner volume of
500−l000 mL (e.g., a Parr 4605 high pressure 600−1200 mL
vessel for maximum pressure: 34.5 MPa and temperature: 350
°C). Furthermore, it has been a definite improvement in the
design of flow reactors of recent date using water under sub-
and supercritical conditions in various laboratories since its
prototype debuted in the early 1990s.37,38 Our preliminary
study using a homemade desktop flow reactor achieved the
average limonene yield of 67% with each contact time of 20 s in
6 h for the feed speed 0.03 g/min of α-pinene. When
extrapolated to the continuous operation for 24 h, it would
allow us to obtain ca. 25 g of limonene.31
Our future efforts should be directed toward three aspects.
Whereas linear dimers and Diels−Alder dimers due to myrcene
are reported in the literature,39 the reaction products including
C20 fractions have to be identified. It is desirable to find
effective radical scavengers usable for studying radical reactions
in pressurized hot water. Lastly, optimization of the formation
of p-cymene from both pinenes in pressurized hot water should
be made using hydrogen acceptors other than silicates, copper
chromite, iodine, or PCl3.21
Figure 7. Effect of reaction time (1−120 min) on the yields of
limonene conversion and p-cymene yield in the presence of oxygen at
400 °C and 30 MPa.
Recently, the formation of p-cymene was reported in the
oxidation of α-pinene in moist air by the atmospheric oxidants
OH, O3, and NO3. Sorbrerol,35 a cyclic monoterpene diol was
proposed as an intermediate for the formation of p-cymene.36
Our results are not contradictory with this observation but
would require some more mechanistic studies for confirmation.
CONCLUSION AND FUTURE DIRECTIONS
■
Our findings described in the previous sections are similar to
but not the same as any of the reported cases in the references
summarized in the Introduction. There are two most novel
aspects. First, limonene had never been obtained in yields as
high as 70% from α-pinene without using a catalyst. A series of
reactions have been studied in supercritical lower alcohols
where the ring-opening is mostly derived from pyrolytic
diradical reactions33 and the best result in terms of the
formation of limonene is 40.0% yield for 300 s, at 300 °C, 10.1
MPa in 0.1 mol L−1 in propanol.5 The lack of formation of
significant amounts of dimers and polymeric materials from α-
pinene seems to be unique to the pressurized hot water
reaction presented here. Second, the formation of myrcene
from β-pinene was found to be reversible. It is generally
understood that terpenes undergo ionic cyclization under acid
catalysis and biradical ring-opening.7,19 It is not yet clear from
the limited amount of our data if this rule applies to the partial
ring-opening of the isomeric pinenes in pressurized hot water.
Several possibilities remain to be explored. The difference in
the chemical structure of α- and β-isomers lies in the position of
the double bond. Mechanistically there are three possibilities
for the initial intermediates: two different diradicals, two
EXPERIMENTAL SECTION
■
Chemicals. α-Pinene ((1R)-(+), ≥80% e.e.) of chemical
purity greater than 98% was purchased from Kanto Chemical
Co., Inc. (ACROS Organics). β-Pinene ((1S)-(-), ≥95.6%),
myrcene (≥77.6%), and alloocimene (mixture of isomers,
≥90.0%) were obtained from Tokyo Chemical Industries Co.,
Ltd., and used without further purification. Authentic samples
of limonene (≥92%) and p-cymene (≥99%) were obtained
from Kanto Chemical Co., Inc. (ACROS Organics).
Ultrapure water that had a specific resistivity greater than
18.2 MΩ and total organic carbon (TOC) smaller than 20 ppb
1489
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491

Organic Process Research & Development
Article
was obtained by treating tap water through a Millipore Milli-
RX75 ultrapure water production system. Other chemicals and
solvents used were commercially available.
required to heat the reactor to the required temperature.
Following a given heating period, the reactors were removed
from the salt bath and submerged in room temperature water.
A pair of heat-resistant gloves and security goggles were worn
during these operations. The reaction mixture usually in the
range 3−7 mL was flushed out of the reactor, collected,
extracted with 3−4 portions of 30 mL of hexane, and analyzed
as follows.
When the amount of water was changed relative to those of
pinenes as described in subsection II of the Results and
Discussion section, the amount of water was kept constant at
3.57 g and that of α-pinene was changed so that the inside
pressure of a 10 mL reactor should be kept constant at 30 MPa
at 400 °C.
Analytical Methods. Qualitative analyses were made on a
Shimadzu Corporation GCMS-QP2010 gas chromatography−
mass spectrometer. The mass spectral fragmentation of each
GC peak was identified by comparison with that of the
authentic samples or by means of similarity analyses using the
provided MS data library in the case of 2,6-dimethyl-1,3,5-
octatriene.41
Quantitative analyses were carried out on a Shimadzu
Corporation GC-2010 gas chromatograph (GC-FID) with the
inlet and detector temperatures of 250 and 280 °C,
respectively. Calibration curves for determining the molar
amounts of α-, β-pinenes, limonene, allo-ocimene, myrcene,
and p-cymene from the corresponding GC peak areas were
drawn in advance with internal standards 1-propanol and 1-
pentanol for the analytes from the reactions of α- and β-pinene,
respectively. A given amount of the internal standard was added
to the hexane extract solutions of α- or β-pinene reactions,
respectively. An aliquot of 1 μL was injected with a split ratio of
100. Chromatographic peaks were identified by comparing their
retention times with those of authentic samples described
under Chemicals.
For the α-pinene analyses, a highly polar capillary column
HP-INNOWax (Agilent Technologies, Co., length: 30 m, inner
diameter (i.d): 0.25 mm, and a film thickness of 0.25 μm) was
used, in conjunction with helium carrier gas. The temperature
program used for the oven of GC and GC-MS analyses started
at 40 °C for 15 min, followed by a temperature ramp of 10 °C/
min up to 70 °C, maintained for 5 min, another temperature
ramp of 20 °C/min up to 250 °C, and then was held at the final
temperature for 10 min.
For the β-pinene analyses, a medium polar capillary column
DB-1701 (Agilent Technologies, Co., length: 30 m, inner
diameter (i.d): 0.25 mm, and a film thickness of 0.25 μm) was
used in conjunction with helium carrier gas. The temperature
program used for the oven of GC and GC-MS analysis started
at 40 °C for 5 min, followed by a couple of temperature ramps
of 3 °C/min to 110 and 0 °C/min to 270 °C, and then was
held there for 5 min.
Characterization of the Main Products: Limonene,
Allo-Ocimene, Myrcene, and p-Cymene. A typical GC-FID
chromatogram is shown in Figure 2S. Mass spectral analysis of
the most abundant product peak in the gas chromatograph
from α-pinene showed a molecular ion peak at m/z = 136, a
base peak at m/z = 68, and other fragment peaks that were
superimposed to those of an authentic sample of limonene.
One of the peaks in the 25-min region was similarly found to be
2,6-dimethyl-1,3,5-octatriene having a molecular ion peak at m/
z = 136, a base peak at m/z = 93, and other characteristic
fragment peaks.
Experimental Apparatus and Procedures. Batch re-
actions were carried out in a handmade reactor of a SS316 1/2-
in. tube (length: 170 mm, outside diameter: 12.7 mm, wall
thickness: 2.1 mm (Mecc Technica Co.)). It was fitted with a
SS316 1/2-in. cap (Swagelok Co., No. SS-810-C), a 1/16-in. to
1/2-in. reducing connector (Swagelok Co., No. SS-810-6-1),
and a 1/16-in. plug (Swagelok Co., No. SS-100-P) to make an
effective inner volume of ca. 10 cm3 (Figure 8). Prior to its use
in experiments, the reactor was rinsed with acetone and loaded
with water and conditioned for 1 h at 300 °C to remove any
residual lubricants/oils that may have been present due to the
Swagelok manufacturing process. The reactor was cleaned with
acetone and dried overnight at 60 °C prior to repeated use.
Preliminary experiments showed that it took nearly one minute
before the thermocouple placed at the center of the tube
showed thermal equilibration with 3.57 g of water and at 400
°C.
Estimation of the Internal Pressure Due to Water at
250−400 °C and the Sufficient Volume Allowing for
Expansion. In most cases, water is employed in large excess
relative to organic substrates for the reactions in pressurized hot
water. Therefore, the phase diagram of water itself may be used
as an approximation of the reaction mixtures. The isobar curves
in the temperature vs density diagram for water (Figure S1
written by K. Sue on the basis of the equations and data in the
literature40) is instructive in determining the recommended
amount of water for the batch reaction in steel reactors. Under
supercritical conditions (TC ≥ 374.15 °C) in which there is
only one phase of water, the amount of loaded water is directly
reflected in the density, and therefore the inside pressure was
estimated to become ca. 30 and 50 MPa at 400 °C when started
with 3.5 and 5.8 g/10 mL of loaded water, respectively. The
latter is about the limit of the allowed pressure of typical
apparatus made of steel. Under subcritical conditions (TC
<
374.15 °C), the isobaric curves in the temperature vs density
diagram for water become bell-shaped corresponding to the
separation of two phases due to saturated vapor and liquid. The
saturated vapor pressure of 8.6 MPa at 300 °C is in principle
independent of the amount of water. Since most acid-catalyzed
reactions take place in the liquid phase, it would be preferable
to load as much water as possible. However, once the amount
of charged water reached 7.5 g/10 mL at 300 °C, for example,
the pressure approaches that of one-phase water with density:
0.75 g/cm3 and pressure of 30 MPa (see Figure S1).
Furthermore, since the isobaric curves are densely populated
in these density regions, the pressure inside the reactor could
change drastically depending on the slight changes in the
amount of loaded water. When the perturbation due to any
added reactant is considered, it would be advisable to load 5−7
g water for the reactor of inner volume of 10 mL at 300 °C (see
Figure S1 in Supporting Information for further detail).
Typical Runs. High-purity water was deaerated prior to use
by a vacuum deaeration method to remove dissolved oxygen
(O2). Pure water (3.574 g, 199 mmol) and either α- or β-
pinene (0.2705 g, 1.99 mmol) were charged in the reactor to
make the starting materials of a 100:1 molar ratio. The mixture
was further flushed with argon unless otherwise stated and
heated to 250−400 °C in a salt bath (Thomas Kagaku Co. Ltd.,
Celsius 600H) which was preheated to the required temper-
ature. Reaction times were 1−30 min, including ca. 1 min
1490
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491

Organic Process Research & Development
Article
(9) Kirkpatrik, W. J. U. S. Patent 2,393,915, 1946
(10) Parikh, P. A. Chem. Eng. J. 1994, 54, 79.
(11) Gurvich, L. G. J. Russ. Phys. Chem. Soc. 1915, 47, 827.
(12) Volzone, C.; Masini, O.; Comelli, N. A.; Grzona, L. M.; Ponzi,
E. N.; Ponzi, M. I. Appl. Catal. A: General 2001, 214, 213.
(13) Wystrach, V. P.; Barnum, L. H.; Garber, M. J. Am. Chem. Soc.
1957, 79, 5786.
β-Pinene yielded other products as shown in a typical GC-
FID chromatogram shown in Figure S3. In the C10 region, the
formation of myrcene was confirmed by a GC-mass spectral
analysis that showed a molecular ion peak at m/z = 136, a base
peak at m/z = 93, and other fragment peaks that were
superimposable on those of an authentic sample of myrcene. In
addition there were several peaks in the diterpene C20 region of
the GC/FID that remained to be identified.39
(14) Roberge, D. M.; Buhl, D.; Niederer, J. P. M.; Holderich, W. F.
Appl. Catal., A: General 2001, 215, 111.
̈
The main product from α- and β-pinenes in pressurized hot
water without exclusion of air was similarly identified as p-
cymene with a strong parent peak at m/z = 134 and other
characteristic fragment peaks that matched with those of the
authentic sample.
(15) Terpin-Werke, German Patent 223,795, 1907.
(16) Uematsu, M.; Franck, E. U. Ber. Bunsenges. Phys. Chem 1980, 95,
1586; J. Phys. Chem. Ref. Data 1980, 9, 1291.
(17) Fernandez, D. P.; Goodwin, A. R. H.; Lemmon, E. W.; Sengers,
J. M. H. L.; Williams, R. C. J. Phys. Chem. Ref. Data 1997, 26, 1125.
(18) Schneider, G. M. Ber. Bunsenges, Phys. Chem. 1972, 76, 325.
(19) Marshall, W. L.; Franck, E. U. J. Phys. Chem. Ref. Data 1981, 10,
295.
(20) Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.; Takeda, N.;
Adschiri, T.; Arai, K. J. Supercrit. Fluids 1998, 13, 261.
(21) Sasaki, M.; Fang, Z.; Fukushima, Y.; Adschiri, T.; Arai, K. Ind.
Eng. Chem. Res. 2005, 44, 1226.
(22) Aida, T. M.; Tajima, K.; Watanabe, M.; Saito, Y.; Kuroda, K.;
Nonaka, T.; Hattori, H.; Smith, R. L., Jr.; Arai, K. J. Supercrit. Fluids
2007, 42, 110.
ASSOCIATED CONTENT
* Supporting Information
■
S
i. Eight tables of numerical data on the reactions of α- and β-
pinenes, limonene, and myrcene in hot pressurized water
constituting Figures 1−7. ii. A temperature−pressure−density
diagram for water with explanation. iii. Two GC chromato-
grams for the reaction mixture of α- and β-pinenes in hot
pressurized water. This material is available free of charge via
the Internet at http://pubs.acs.org.
(23) Savage, P. E.; Gopalan, S.; Mizan, T. I.; Martino, C. J.; Brock, E.
E. AIChE J. 1995, 41, 1723.
(24) Katritzky, A. R.; Allin, S. M.; Siskin, M. Acc. Chem. Res. 1996, 29,
399.
(25) Savage, P. E. Chem. Rev. 1999, 99, 603.
(26) Kruse, A.; Dinjus, E. J. Supercrit. Fluids 2007, 39, 362.
(27) Liotta, C. L.; Hallett, J. P.; Pollet, P.; Eckert, C. A. Reactions in
AUTHOR INFORMATION
Corresponding Authors
*E-mail: iwamura.hiizu@nihon-u.ac.jp.
*E-mail: hiaki.toshihiko@nihon-u.ac.jp.
Notes
■
Nearcritical Water, In Organic Reactions in Water; Lindstrom, U. M.,
̈
Ed.; Blackwell Publishing, 2007; Chapter 9.
The authors declare no competing financial interest.
(28) Sully, B. D. Chem. Ind. (London) 1964, 263.
(29) Bauer, K.; Garbe, D. Common Flavor and Fragrance Materials;
VCH: Weinheim, 1985.
(30) Szuppa, T.; Stolle, A.; Ondruschka, B. Org. Biomol. Chem. 2010,
8, 1560.
(31) Miura, H. Development of Organic Reaction Processes by the Flow
System Using Pressurized Hot Water, Masters Thesis; College of
Industrial Technology, Nihon University, 2011.
(32) Saida, K. to be published in Thesis for the master’s degree,
College of Industrial Technology, Nihon University, 2014.
(33) Anikeev, V. I. Flavour Fragrance J. 2010, 25, 443.
(34) Stanislaus, A.; Yeddanapalli, L. M. Can. J. Chem. 1972, 50, 113.
(35) Xu, Z.-B.; Qu, J. Chem.Eur. J. 2013, 19, 314.
(36) Gratien, A.; Johnson, S. N.; Ezell, M. J.; Dawson, M. L.; Bennett,
R.; Finlayson-Pitts, B. J. Environ. Sci. Technol. 2011, 45, 2755.
(37) Thornton, T. D.; Savage, P. E. J. Supercrit. Fluids 1990, 3, 240.
(38) Adschiri, T.; Kanazawa, K.; Arai, K. J. Am. Ceram. Soc. 1992, 75,
1019.
ACKNOWLEDGMENTS
■
This paper was presented in part at the First International
Symposium on Process Chemistry-ISPC 08 in July 28−30,
2008 in Kyoto. We dedicate this paper to the memory of
Professor Wesley Cocker (1908−2007), Trinity College
Dublin, in recognition of his personal suggestion and
encouragement as early as in 1969 of an importance of new
approaches to pinene chemistry. We thank Mr. Kohei Saida for
his help in obtaining the additional data on the effect of the
amount of water on the yield of limonene from α-pinene and
his generosity in disclosing his thesis work prior to publication.
This research was supported in part by a Grant in Aid from the
Ministry of Education, Culture, Sports, Science, and Technol-
ogy for promoting multidisciplinary research projects.
(39) Cataldo, F.; Keheyan, Y. Radiat. Phys. Chem. 2006, 75, 572.
(40) Wagner, W.; Pruss, A. J. Phys. Chem. Ref. Data 2002, 31, 387.
(41) Mitzner, B. M.; Theimer, E. T.; Freeman, S. K. Appl. Spectrosc.
1965, 19, 169.
REFERENCES
■
(1) Coppen, J. J. W.; Hone, G. A. Non−wood Forest Products 2, Gum
Naval Stores: Turpentine and Rosin from Pine Resin; Natural Resources
Inst., FAO of UN, FAO Corporate Document Repository, Rome,
1995. Another estimate is 263,000 tons in 1999 by: Morris, M. The
Global Supply Position for Turpentine, Proceeding of Papers Presented
at the IFEAT 2000 International Conference − Aroma Chemicals 2000 &
Beyond; Florida, USA; 29 Oct − 2 Nov 2000; pp 17−31.
(2) Berthelot, M. Ann. Chim. [iii] 1853, 39, 9.
(3) Wallach, O. Ann. Chem. 1885, 227, 282.
(4) Costa, M. C. C.; Johnstone, R. A. W.; Whittaker, D. J. Mol. Catal.
A: Chem. 1996, 104, 251.
(5) Stolle, A.; Bonrath, W.; Ondruschka, B. J. Anal. Appl. Pyrolysis
2008, 83, 26.
(6) Kolicheski, M. B.; Cocco, L. C.; Mitchell, D. A.; Kamiski, M. J.
Anal. Appl. Pyrolysis 2005, 80, 92.
(7) Fuguitt, R. E.; Hawkins, J. E. J. Am. Chem. Soc. 1945, 67, 242.
(8) Savich, T. R.; Goldblatt, L. A. J. Am. Chem. Soc. 1945, 67, 2027.
1491
dx.doi.org/10.1021/op4000863 | Org. Process Res. Dev. 2013, 17, 1485−1491