Palladium catalyzed aerobic oxidation for the incorporation of an olfactory group on naturally occurring Β-caryophyllene

Article abstract of DOI:10.1016/j.apcata.2017.07.042

β-Caryophyllene is one of the most widespread sesquiterpenes, found as a main hydrocarbon component in various essential oils, e.g., copaiba oil, which is used for years in folk medicine and occupies an important position in Brazilian pharmaceutical export. The novel selective oxidation of β-caryophyllene by molecular oxygen using the chloride-free Pd(OAc)₂/p-benzoquinone (BQ) catalytic system has been developed. The reaction gives two main products, both arising from the allylic oxidation of the sterically encumbered endocyclic double bond, whereas the terminal double bond of the substrate remains intact. The catalytic process efficiently operates under 10 atm of oxygen in the absence of auxiliary metal co-catalysts, which are commonly used in related systems. The reaction can also occur under atmospheric pressure in the presence of Cu(OAc)₂ to accelerate the re-oxidation of p-hydroquinone by molecular oxygen. Both allylic acetates, obtained from β-caryophyllene in 75–85% combined yields in both systems, are natural compounds found in essential oils of some plants and have well pronounced perfume properties. To the best of our knowledge, the present work reports the first synthesis of these functionalized sesquiterpenic compounds potentially useful as ingredients of synthetic perfumes and pharmacological compositions.

Full text of DOI:10.1016/j.apcata.2017.07.042

Accepted Manuscript
Title: Palladium catalyzed aerobic oxidation for the
incorporation of an olfactory group on naturally occurring
␤-caryophyllene
Authors: Amanda de Camargo Faria, Taciano A. dos Santos
Wanderley, Eduardo E. Alberto, Elena V. Gusevskaya
PII:
S0926-860X(17)30352-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2017.07.042
APCATA 16347
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
30-4-2017
13-7-2017
25-7-2017
Please cite this article as: Amanda de Camargo Faria, Taciano A.dos Santos Wanderley,
Eduardo E.Alberto, Elena V.Gusevskaya, Palladium catalyzed aerobic oxidation for the
incorporation of an olfactory group on naturally occurring ␤-caryophyllene, Applied
Catalysis A, Generalhttp://dx.doi.org/10.1016/j.apcata.2017.07.042
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Palladium catalyzed aerobic oxidation for the incorporation of an
olfactory group on naturally occurring -caryophyllene
Amanda de Camargo Faria, Taciano A. dos Santos Wanderley, Eduardo E. Alberto,
Elena V. Gusevskaya*
Departamento de Química, Universidade Federal de Minas Gerais, 31270-901,
Belo Horizonte, MG, Brazil.
Dedicated to Prof. Martin Schmal on the occasion of his 80th birthday
_________________
 Corresponding author. Tel.: +55 31 34095741; fax: +55 31 34095700.
E-mail address: elena@ufmg.br (E.V. Gusevskaya)
Graphical abstract

Research Highlights
 -Caryophyllene is one of the most widespread sesquiterpenes, found as a main
hydrocarbon component in various essential oils, e.g., copaiba oil, which is used
for years in folk medicine.
 A novel selective oxidation of -caryophyllene by molecular oxygen using the
chloride-free Pd(OAc)2/p-benzoquinone catalytic system has been developed.

Poly-functionalized sesquiterpenic compounds obtained in these reactions are
potentially useful as ingredients of synthetic perfumes and pharmacological
compositions.
Abstract
-Caryophyllene is one of the most widespread sesquiterpenes, found as a main
hydrocarbon component in various essential oils, e.g., copaiba oil, which is used for
years in folk medicine and occupies an important position in Brazilian pharmaceutical
export. The novel selective oxidation of -caryophyllene by molecular oxygen using the
chloride-free Pd(OAc)2/p-benzoquinone (BQ) catalytic system has been developed. The
reaction gives two main products, both arising from the allylic oxidation of the
sterically encumbered endocyclic double bond, whereas the terminal double bond of the
substrate remains intact. The catalytic process efficiently operates under 10 atm of
oxygen in the absence of auxiliary metal co-catalysts, which are commonly used in
related systems. The reaction can also occur under atmospheric pressure in the presence
of Cu(OAc)2 to accelerate the re-oxidation of p-hydroquinone by molecular oxygen.

Both allylic acetates, obtained from -caryophyllene in 75–85 % combined yields in
both systems, are natural compounds found in essential oils of some plants and have
well pronounced perfume properties. To the best of our knowledge, the present work
reports the first synthesis of these functionalized sesquiterpenic compounds potentially
useful as ingredients of synthetic perfumes and pharmacological compositions.
Keywords: Biorenewables; -Caryophyllene; Oxidation; Palladium; Terpenes
1. Introduction
The involvement of bio-renewable reagents in industrial chemical processes is
among the requirements of green organic synthesis along with the use of catalytic rather
than stoichiometric reactions. In particular, numerous terpene hydrocarbons, the main
components of essential oils of plants and flowers, represent an important feedstock for
the fragrance&flavor and pharmaceutical industries [1–3]. The molecules of terpenes
often contain multiple olefinic bonds and can be upgraded into more valuable
compounds through a variety of chemical reactions, such as hydroformylation,
oxidation, epoxidation, hydrogenation and acid-catalyzed transformations [4–9].
-Caryophyllene is one of the most widespread sesquiterpenes found as a main
hydrocarbon component in various essential oils, such as clove (Eugenia
Caryophyllata) and copaiba (Copaifera) oils [9–12]. A remarkable biological activity of
caryophyllanes is responsible for the well recognized therapeutic properties of copaiba
oil, which is used for years in popular medicine and traditionally occupies an important
position in the Brazilian export of pharmaceutical products [12]. In addition,
caryophyllane derivatives usually have pleasant woody, floral or fruity scents and are
widely used as aromatic ingredients in perfumes, food and tobacco products [11,12].
Therefore, the catalytic functionalization of the -caryophyllene molecule, which
contains two olefinic bonds, could open an access to the compounds important for the
fragrance&flavor and pharmaceutical industries.
For several years our group has been involved in the catalytic upgrading of
naturally occurring products [6,9], in particular, through the palladium catalyzed
oxidation of their C=C bonds [13–21]. The majority of the palladium catalyzed
oxidations of organic molecules by molecular oxygen require an additional redox co-

catalyst in order to provide the efficient re-oxidation of reduced palladium species
before their irreversible agglomeration into inactive bulk palladium metal. The original
and most commonly used catalytic system for these reactions contains CuCl2 as the co-
catalyst and the presence of chloride ions in high concentrations is required to maintain
a catalytic cycle (Wacker catalyst [22]). Many efforts were directed to search more
environmentally acceptable and less corrosive alternatives for CuCl2, such as p-
benzoquinone, Cu(OAc)2, heteropoly acids or special ligands, which can stabilize
palladium in solutions allowing its direct re-oxidation by molecular oxygen [23–28].
As a part of our general project aiming to the valorisation of essential oils, we
have recently turned our attention to the functionalization of -caryophyllene (1). The
molecule of -caryophyllene has an unusual structure in which a nine-membered ring is
fused in a trans manner with a cyclobutane ring (Scheme 1). The molecule is highly
strained and enters in structural rearrangements with a remarkable facility to release this
strain. This makes a great challenge the development of the selective reactions to obtain
one or two particular products. Probably for this reason, only few reports on the
selective catalytic transformations of -caryophyllene have been described in the
literature, despite the industrial importance of its derivatives [29–34]. In particular, our
group has recently reported two reactions of -caryophyllene: the heteropoly acid
catalyzed acetoxylation to give -caryolanyl compounds, i.e. the products with an
isomerized carbon skeleton [32], and rhodium catalyzed hydroformylation in which the
caryophyllane structure was retained [34].
To the best of our knowledge, the palladium catalyzed oxidation of -
caryophyllene has not been studied before. Moreover, we could find only two
previously published reports on the catalytic oxidation of -caryophyllene in general,
both describing the reactions of epoxidation with hydrogen peroxide over titanium
catalysts [29,30]. In the present work, we report the application of palladium catalytic
systems for the oxidation of -caryophyllene with molecular oxygen. All the reactions
gave exclusively the products derived from the interaction of palladium with the
endocyclic double bond of -caryophyllene. The products obtained from -
caryophyllene in the present work and their mixtures have well pronounced perfume
properties and could be used directly in perfume compositions even without the
separation of individual compounds.

2. Experimental
The chemicals acquired from commercial sources were used as received, except
when indicated. CuCl22H2O and LiCl were dehydrated by heating. p-Benzoquinone
was purified by column chromatography (silica) using dichloromethane as the eluent. -
Caryophyllene [(-)-trans caryophyllene] was purchased from Sigma-Aldrich and glacial
acetic acid used as the solvent from VETEC. The catalytic tests under 1 atm of O2 were
performed in a glass reactor, whereas those under superatmospheric pressure in a
homemade stainless steel 100-mL autoclave equipped with a valve dip tube to take the
samples of the reaction solution. Magnetic stirring was used in both reactors. The glass
reactor was equipped by a gas burette to measure oxygen consumption during the
reactions. The reaction progress was monitored by gas chromatography (GC) taking the
samples without the depressurization of the reactor. In a typical catalytic test, the
solution of the substrate (0.20 M), palladium salt (0.005 – 0.02 M), copper salt (if any,
0.025 – 0.05 M), BQ (if any, 0.05 – 0.20 M), and internal standard (dodecane or p-
xylene, 0.10 M) in the indicated solvent was transferred to the reactor. The total volume
of the solution was 10 mL. Reaction variables (concentrations, temperature, oxygen
pressure, etc.) and reaction time are presented in the Tables. The reactor was pressurized
with oxygen to 1–10 atm, heated in an oil bath and then intensive stirring was turned on
to start the reaction. At the end of the run, the autoclave was cooled to room temperature
and oxygen was allowed to slowly vent out. CAUTION: reactions with oxygen under
elevated pressure can lead to an unforeseen explosion. It must only be performed using
the appropriate equipment and under rigorous safety precautions. The samples of the
reaction solutions periodically taken from the reactor were analyzed by GC on a
Shimadzu GC2010 equipment using a Carbowax 20 m capillary column and a FID
detector. Relative GC response factors were determined using isolated reaction
products. A GC mass balance was based on the -caryophyllene converted using
internal standards (dodecane or p-xylene). The difference was attributed to the
formation of high-boiling products, which were not GC determinable. Initial reaction
rates were calculated at low conversions (up to 20-40%).
The product isolation was performed by a column chromatography (silica gel
60) using mixtures of hexane and CH2Cl2 as eluents. Before passing the reaction
solution through the column, acetic acid was neutralized by sodium bicarbonate and

organic components were extracted from aqueous solutions with dichloromethane.
Isolated products were identified by 1H and 13C-NMR with TMS as an internal standard
and/or GC-MS (Bruker 400 MHz spectrometer; Shimadzu QP2010-PLUS instrument,
70 eV).
Compound 2 (two isomers): MS (70 eV, EI): m/z (%) first isomer (shorter GC
retention time): 202 (10) [M+-HOAc], 187 (22) [M+-HOAc-CH3], 173 (17), 159 (34),
146 (18), 145 (28), 133 (27), 132 (18), 131 (54), 120 (16), 119 (42), 118 (67), 117 (58),
107 (23), 106 (19), 105 (61), 93 (37), 92 (22), 91 (100), 81 (17), 79 (51), 77 (36), 69
(40), 67 (29), 65 (16), 60 (20), 55 (26); second isomer (longer GC retention time): 202
(9) [M+-HOAc], 187 (20) [M+-HOAc-CH3], 173 (18), 159 (32), 146 (17), 145 (28), 133
(25), 132 (15), 131 (53), 120 (15), 119 (40), 118 (60), 117 (52), 107 (24), 106 (19), 105
(66), 93 (37), 92 (21), 91 (100), 81 (20), 79 (51), 77 (34), 69 (35), 67 (28), 65 (16), 60
(20), 55 (27). 1H NMR (400 MHz, CDCl3, 25°C, Me4Si): δ=0.92 (s, 6H; C14H3 and
C15H3), 1.40–2.00 (m, 4H; C2H2 and C6H2), 1.50–1.70 (m, 3H; C10H2 and C1H), 1.80–
2.00 (m, 2H; C3HH, and C7HH), 1.95 and 1.96 (s, 3H; OCOCH3), 2.20–2.45 (m, 3H;
C3HH, C7HH and C9H), 4.60 and 4.66 (s, 1H; C13HH), 4.74 (s, 1H; C13HH), 4.92 and
4.99 (s, 1H; C12HH), 4.97 and 5.04 (s, 1H; C12HH), 5.05–5.10 ppm (m, 1H; C5H); 13C
NMR (100 MHz, CDCl3, 25°C, TMS): δ=21.20 and 22.25 (OCOCH3), 21.82 and 22.15
(C14), 29.85 and 29.91 (C15), 29.96 and 30.16 (C2 or C6), 30.35 and 30.43 (C2 or C6),
30.94 and 32.48 (C7), 33.44 and 33.48 (C11), 33.79 (C3), 36.66 and 37.73 (C10), 43.41
and 43.85 (C9), 53.03 and 54.49 (C1), 76.65 and 76.92 (C5), 109.35 and 109.50 (C13),
115.36 and 116.35 (C12), 146.26 and 147.16 (C4), 151.54 and 152.34 (C8), 169.85 and
170.01 ppm (OCOCH3). For atom numbering see Scheme 1.
Compound 3: MS (70 eV, EI): m/z (%) 202 (13) [M+-HOAc], 187 (26) [M+-
HOAc-CH3], 173 (24), 159 (36), 146 (21), 145 (34), 133 (30), 132 (20), 131 (63), 120
(16), 119 (39), 118 (31), 117 (48), 107 (23), 106 (20), 105 (63), 93 (39), 92 (22), 91
1
(100), 81 (15), 79 (51), 77 (34), 69 (39), 67 (27), 65 (16), 60 (15), 55 (24). H NMR
(400 MHz, CDCl3, 25°C, Me4Si): δ=0.91 (s, 3H; C14H3), 0.92 (s, 3H; C15H3), 1.40–1.60
(m, 2H; C2H2), 1.50–1.70 (m, 2H; C10H2), 1.70–1.80 (m, 1H; C1H), 1.99 (s, 3H;
OCOCH3), 2.05–2.15 (m, 1H; C3HH), 2.10–2.30 (m, 5H; C3HH, C6H2 and C7H2), 2.35–
4
2.45 (m, 1H; C9H), 4.40 (d, J=4.0 Hz, 2H; C12H2), 4.68 (s, 1H; C13HH), 4.76 (s, 1H;
3
C13HH), 5.51 ppm (t, 1H; J=8.0 Hz, C5H); 13C NMR (100 MHz, CDCl3, 25°C, TMS):
δ=20.93 (OCOCH3), 22.80 (C14), 25.15 (C3), 26.13 (C2), 27.38 (C6), 29.85 (C15), 33.06
(C11), 34.65 (C7), 40.00 (C10), 40.17 (C9), 51.62 (C1), 68.42 (C12), 110.69 (C13), 130.29

(C5), 134.55 (C4), 155.09 (C8), 170.78 ppm (OCOCH3). For atom numbering see
Scheme 1.
3. Results and Discussion
The molecule of -caryophyllene has two olefinic bonds which could be
involved in the interaction with palladium: terminal disubstituted and endocyclic
trisubstituted bonds. The conventional Wacker system, PdCl2/CuCl2, was tested first for
the oxidation of -caryophyllene (1) (Table 1, run 1). The reaction was performed under
typical conditions in the solution of glacial acetic acid and resulted in nearly complete
-caryophyllene conversion within 3 hours. More than twenty new peaks appeared in
the chromatogram of the reaction solutions after this run. In a blank reaction with no
palladium salt added, -caryophyllene was also consumed almost completely giving
nearly the same products (Table 1, run 2). The products were clearly divided in two
groups on the chromatogram. The nature of these compounds was suggested based on
their characteristic retention times and mass spectra. The products from the first group
(ca. ten GC peaks) with retention times close to that of -caryophyllene were attributed
to -caryophyllene isomers. The second set of GC peaks with longer retention times
was suggested to belong to the products of the solvent (HOAc) addition to -
caryophyllene and its isomers. Thus, the Wacker catalytic system promoted almost
exclusively the acid-catalyzed transformations of -caryophyllene, i.e., the
isomerization and solvent addition, also accompanied by the isomerization of carbenium
ion intermediates. CuCl2 alone promoted the same reactions of -caryophyllene in
acetic acid solutions. In the presence of palladium, no products which could arise from
the oxidation of -caryophyllene (accordingly to their mass spectra) were detected in
the reaction solutions in appreciable amounts. The results of these preliminary tests
clearly indicated that the attempts to oxidize -caryophyllene, such a sensitive to the
reaction conditions molecule, should be continued in alternative systems: preferentially
with non-acidic solvents or at least free of CuCl2.
The most attractive candidate would be a system recently discovered by Kaneda
et al. using dimethylacetamide (DMA) solutions with PdCl2 which acts as a sole catalyst
[28,35]. The use of the coordinating DMA solvent in this system allows to dispense the
presence of auxiliary co-oxidants because Pd(0) species stabilized by the solvent react

with molecular oxygen faster than form an inactive bulk metal. We have previously
applied this so called “palladium solo” system for the oxidation of several terminal and
internal alkenes [20,36,37]; however, for -caryophyllene the results were not
promising (Table 1, runs 3–5). Under typical conditions (PdCl2 - 5–10 mol %, solvent –
DMA with 10–20 vol % of H2O, 80oC, 10 atm of oxygen), only the products of the -
caryophyllene isomerization were detected in the reaction solutions. Varying the
reaction conditions did not result in the formation of any oxidation products. The
isomerization of -caryophyllene was observed even in the absence of the palladium
salt (Table 1, run 6).
Thus, on the one hand, both olefinic bonds of -caryophyllene behave as quite
resistant to the interaction with palladium and, on the other hand, the molecule is
extremely sensitive to the presence of CuCl2 which promotes various acid catalyzed
transformations of -caryophyllene. Our further attention was directed to the catalytic
systems based on the use of p-benzoquinone (BQ), which is a valuable candidate to
replace CuCl2 in Wacker oxidations [23,38]. Representative results obtained are shown
in Tables 2 and 3.
The strategy was to study the reactions of -caryophyllene with stoichiometric
amounts of BQ using Pd(OAc)2 as the catalyst and then, if succeed, aim to make the
reaction catalytic also in BQ. We were very pleased to observe that the reactions of -
caryophyllene with BQ in acetic acid solutions containing catalytic amounts of
Pd(OAc)2 gave only two main products (2 and 3) with apparently high combined
selectivity (Table 2, runs 1–3). The formation of neither of these products was observed
in a blank reaction performed under the same conditions without Pd(OAc)2, with only
BQ added.
Compounds 2 and 3 presented close GC retention times and very similar mass
spectra. They were isolated from the reaction solutions by column chromatography as a
mixture (with a pleasant woody-floral smell) and identified by GC-MS and NMR
spectroscopy (Scheme 1). As can be seen, compounds 2 and 3 are closely related regio-
isomeric allylic acetates and are both derived from the interaction of the endocyclic
double bond of -caryophyllene with palladium. Acetate 2 is the product of the
molecule oxidation at the endocyclic carbon (C-5), whereas acetate 3 – at the terminal
carbon (C-12). The oxidation of -caryophyllene occurred with high site-selectivity: no
products derived from the oxidation of the other double bond were detected in

appreciable amounts. It is a remarkable result as the terminal double bond is expected
to show higher reactivity towards the interaction with palladium than the sterically
encumbered trisubstituted endocyclic bond.
Allylic acetates 2 and 3 were detected in the reactions of -caryophyllene with
o
BQ at 60–70 C with a combined selectivity of ca. 80% (Table 2, runs 1 and 2). The
major product, acetate 2, was responsible for ca. 60% of the mass balance at high
substrate conversions, whereas the selectivity for minor acetate 3 in most of the runs
was 15–20%. In addition, the products of the -caryophyllene isomerization were
detected in small amounts (ca. 10%) along with several unidentified products with
longer GC retention times (probably, over-oxidation products).
Acetate 2 was formed in all the runs as two GC-separable isomers (with virtually
identical mass spectra and very similar NMR spectra) in approximately equal amounts.
Judging from the structure of 2, we suggest that the difference between the two isomers
is the relative orientation of the acetate group and the hydrogens H-1 and H-9 at the
carbons shared by both rings (Scheme 1). Compound 2 was previously described as a
natural ingredient of the essential oil produced from the European alpine plant
Leontopodium alpinum, commonly known as Edelweiss [39]. Acetate 3 is also a natural
compound isolated from the essential oil of Betula species, well-known as birch tree,
widely distributed in the northern hemisphere and used for years in folk medicine and
cosmetic products [40]. To the best of our knowledge, the present work reports the first
synthesis of these functionalized sesquiterpenic compounds, which are potentially
useful as components of synthetic perfumes and pharmacological compositions.
Although the runs with stoichiometric amounts of BQ were performed under
oxygen atmosphere, no oxygen uptake was observed as expected. The re-oxidation of
the reduced form of BQ, p-hydroquinone (BQH2), with molecular oxygen under
atmospheric pressure is known to require an auxiliary metal catalyst, such as cobalt,
copper or manganese salts [24,38]. Aiming to make the reaction catalytic in BQ,
Cu(OAc)2 was added to the reaction solutions to accelerate the regeneration of BQ
(Table 2, runs 4–13). It was verified that the presence of Cu(OAc)2 did not affect
negatively the reaction selectivity as well as the reaction rate (Table 2, run 4 vs. run 2).
In further runs, BQ was used in sub-stoichiometric amounts and was successfully
regenerated by molecular oxygen so that the reactions were catalytic not only with
respect to palladium and copper but also with respect to BQ. The consumption of

molecular oxygen in these runs corresponded to the amounts of -caryophyllene
converted into oxidation products.
The reactions with catalytic amounts of BQ occurred at lower rates compared to
their stoichiometric counterparts (Table 2, run 5 vs. run 1; run 6 vs. runs 2 and 4; run 7
vs. run 3), suggesting that the regeneration of BQ is the slowest step of the whole
process. Initial reaction rates were clearly depended on the BQ concentration (Table 2,
cf. runs 6 and 8; runs 9 and 7) and also on the concentration of Cu(OAc)2 (Table 2, cf.
runs 10 and 8) supporting the suggestion that the rate-determining step is the oxidation
of BQH2 with molecular oxygen. Furthermore, a two-fold increase in the palladium
concentration did not accelerate the reaction, suggesting nearly zero order in palladium
(Table 2, run 11 vs. run 8). However, at lower palladium concentrations, the interaction
of palladium with such a difficult substrate as -caryophyllene became slower
(probably, turned kinetically comparable with the BQH2 re-oxidation) affecting the rate
of the whole process (Table 2, run 12 vs. run 8; run 13 vs. run 9).
Thus, the palladium catalyzed oxidation of -caryophyllene successfully occurs
in the presence of Cu(OAc)2 with an oxygen coupled catalytic turnover under
atmospheric pressure. Apparently, Cu(OAc)2 acts as an auxiliary catalyst accelerating
the reaction of BQH2 with molecular oxygen. The combined selectivity for allylic
oxidation products 2 and 3 was 80–90% at nearly complete substrate conversions, with
the major acetate 2 being responsible for 70–75% of the mass balance in most of the
o
runs. The reactions performed at 90 C showed slightly lower selectivity (Table 2, runs
7, 9 and 13).
We have found in our recent studies that under 5–10 atm BQ can be regenerated
from BQH2 directly by molecular oxygen in the absence of metal co-catalysts, which
allowed us to create environmentally more friendly oxidation processes [17,21]. The
applications of this approach to the oxidation of -caryophyllene are presented in Table
3. The reactions were less selective as compared to their counterparts performed under
atmospheric pressure due to the appearance of several unidentified minor products with
longer GC retention times (probably, over-oxidation products). However, the
optimization of the reaction conditions allowed to obtain allylic acetates 2 and 3 with
ca. 80% selectivity at high substrate conversions (Table 3, runs 3, 4 and 7). No
palladium mirror was observed on the reactor walls after the reactions with only 0.25
equiv. of BQ reflecting the efficient regeneration of Pd and BQ (Table 3, runs 5–8). The

ratio between acetates 2 and 3 was slightly dependent on the reaction temperature
varying from 3:1 at 80 oC to 6:1 at 50 oC; with all the reactions resulting predominantly
in the endocyclic acetate 2. The same tendency can also be seen in all the runs under
atmospheric pressure (Table 2). Neither product 2 nor 3 were formed from -
caryophyllene under 1–10 atm of oxygen in acetic acid solutions in the absence of
Pd(OAc)2. At lower BQ concentrations, the reaction occurred slower, albeit with nearly
the same selectivity and product distribution (Table 3, cf. runs 1 and 5; runs 2 and 6;
runs 3 and 7). Therefore, the oxidation of BQH2 by molecular oxygen seems still to be a
rate controlling step at the oxidation of -caryophyllene under superatmospheric oxygen
pressure. On the other hand, with the increase in the Pd(OAc)2 concentration the
reaction occurred faster (Table 3, run 8 vs. run 6). These results could indicate that the
reactions between the olefin and palladium and between BQH2 and molecular oxygen
occur at comparable rates, each one being able to control the rate of the whole process
depending on the reaction conditions.
The mechanistic proposal for the oxidation of -caryophyllene into allylic
acetates 2 and 3 is presented in Scheme 2. The accepted mechanism for the palladium
catalyzed allylic oxidation involves the formation of -allyl palladium complexes as key
intermediates [14,41,42]. The structures of products 2 and 3 imply that they result from
the exocyclic -allyl palladium intermediate (A) formed by the abstraction of the allylic
hydrogen from acyclic terminal carbon atom C-12. It is noteworthy that no products
derived from endocyclic -allyl palladium intermediates (not shown in Scheme 2),
which could be formed by the hydrogen abstraction from C-3 or C-6, were observed in
appreciable amounts at the oxidation of -caryophyllene. The tendency in
regioselectivity at the interaction of the endocyclic double bond with palladium and also
such a great difference in the reactivity of allylic hydrogens in the molecule of -
caryophyllene are surprising. On the one hand, the formation of intermediate A follows
a Markovnikov-type orientation; that is, involving the hydrogen allylic to the more
substituted end of the olefinic bond. On the other hand, a strong preference for the
hydrogen abstraction from acyclic carbon C-12 rather than from endocyclic carbon C-3
is remarkable taking into account a usually observed so-called “cyclic activation”,
which is the enhanced reactivity of cyclic allylic hydrogens compared to acyclic ones.
Such a “cyclic activation” has been previously observed in the palladium catalyzed
allylic functionalizations of limonene [14, 43], -bisabolol [18] and -ionone [44] in

related systems, resulting in the preferential formation of the products derived from
corresponding endocyclic rather than exocyclic -allyl palladium intermediates.
The subsequent nucleophilic attack of acetic acid on -allyl palladium
intermediate A results in the formation of two allylic acetates. Linear allylic acetate 3 is
formed by nucleophilic addition expectedly directed to less hindered carbon C-12.
However, acetate 3 is the minor product; whereas the major product is branched allylic
acetate 2, in which the acetate group bonds to more substituted endocyclic carbon C-5.
In principle, the attack of acetic acid on intermediate A could occur on both faces of the
nonane ring to give in the case of product 2 two steric isomers with different geometry
at carbon C-5 with respect to the hydrogens at shared carbons C-1 and C-9. A strong
effect of reaction variables on the regioselectivity at palladium catalyzed allylic
oxidations is well known [42,45–47]. In particular, it was reported that terminal alkenes
could be selectively oxidized to either linear or branched allylic acetates in different
solvents and with different palladium catalysts [45].
Among the C-C double bonds of -caryophyllene, the endocyclic one is much
more reactive towards palladium species, determining the structure of the oxidation
products. The products that could be derived from the oxidation of the terminal double
bond were not detected in appreciable amounts. The analysis of a 3-D ball-and-stick
model of -caryophyllene reveals that its structure is highly conformationally strained
and both double bonds are sterically encumbered; nevertheless, the reasons for such low
reactivity of the terminal double bond compared to the endocyclic one are not
straightforward. On the other hand, it is noteworthy that in most previously reported
catalytic [29,30,34] and even non-catalytic [10] transformations of -caryophyllene the
endocyclic double bond reacted first; for example, in catalytic epoxidation [29,30] and
hydroformylation [34] reactions. It seems reasonable to suggest that this preference is
due to the weakness of the ring strain during the reactions in which the endocyclic
double bond is involved [10]. The analysis of the models of caryophyllane compounds
supports this suggestion showing that the structures with no double bond in the nonane
ring are much less rigid than -caryophyllene itself or its ring unsaturated derivatives.
Thus, the high site-selectivity observed at the palladium catalyzed oxidation of -
caryophyllene seems to be determined by steric factors related to the attenuation of ring
strain in the original molecule. We also suggest that these factors could be responsible
for the regio-selectivity observed at the oxidation of the endocyclic double bond with

the preferential formation of acetate 2, which contains a saturated ring system,
differently from minor acetate 3. Moreover, all the products which could be derived
from endocyclic -allyl palladium intermediates (mentioned above) would also contain
endocyclic double bonds and therefore would have much more rigid structures than
acetate 2. Probably for this reason, such products are not formed at the palladium
catalyzed oxidation of -caryophyllene in our systems. An interesting example of the
site-selectivity control at the palladium catalyzed oxidation of limonene in related
systems has been recently reported by Czapiewski et.al [47]. In acetic acid solutions, the
products derived from the oxidation of the endocyclic olefinic bond were mainly
obtained; whereas, the addition of dimethyl sulfoxide allowed for the selective
oxidation of the exocyclic disubstituted terminal double bond.
4. Conclusions
The oxidation of the abundant, naturally occurring -caryophyllene represents
an attractive entry to fragrant and biologically active compounds. Herein, we developed
a novel selective oxidation of -caryophyllene by molecular oxygen using a chloride-
free Pd(OAc)2/p-benzoquinone catalytic system. The endocyclic double bond in the -
caryophyllene molecule unexpectedly showed much higher reactivity than the terminal
one. The reaction gave two main products, both arising from the allylic oxidation of the
sterically encumbered endocyclic double bond, whereas the terminal double bond of the
substrate remained intact. We suppose that steric factors related to the attenuation of
ring strain in the original molecule are mainly responsible not only for the high site-
selectivity at the oxidation of -caryophyllene, but also for the regio-selectivity
observed at the oxidation of its endocyclic double bond.
The catalytic process efficiently operates under 10 atm of oxygen in the absence
of auxiliary metal co-catalysts, which are conventionally used in related systems. The
reaction can also occur under atmospheric pressure in the presence of Cu(OAc)2 to
accelerate the re-oxidation of p-hydroquinone by molecular oxygen. Both allylic
acetates obtained from -caryophyllene are natural compounds found in essential oils of
some plants and have well pronounced perfume properties. To the best of our
knowledge, the present work reports the first synthesis of these functionalized

sesquiterpenic compounds potentially useful as components of synthetic perfumes and
pharmacological compositions.
Acknowledgments
We acknowledge CNPq, CAPES, FAPEMIG, and INCT-Catálise (Brazil) for the
financial support and fellowships. Bruno B. N. S. Brandão is thanked for the
experimental assistance.
References
[1]
[2]
[3]
E. Breitmaier, Terpenes. Flavors, Fragrances, Pharmaca, Pheromones, Wiley-
VCH, Weinheim, 2006.
C. Sell, in: C. Sell (Ed.), The Chemistry of Fragrances: from Perfumer to
Consumer, second ed., RSC Publishing, Dorset, UK, vol. 2, 2006, pp. 52–88.
K. Bauer, D. Garbe, H. Surburg, Common Fragrance and Flavour Materials.
Preparation, Properties and Uses, VCH Verlagsgesellschaft, 1990.
K.A.D. Swift, Top. Catal. 27 (2004) 143–155.
[4]
[5]
[6]
J.L.F. Monteiro, C.O. Veloso, Top. Catal. 27 (2004) 169–180.
E.V. Gusevskaya, J. Jimènez-Pinto, A. Börner, ChemCatChem 6 (2014) 382–
411.
[7]
A. Behr, A.J. Vorholt, K.A. Ostrowski,T. Seidensticker Green Chem. 16 (2014)
982–1006.
[8]
[9]
A. Behr, L. Johnen, ChemSusChem 2 (2009) 1072–1095.
E.V. Gusevskaya, ChemCatChem 6 (2014) 1506–1515.
[10] I.G. Collado, J.R. Hanson, A.J. Macías-Sáncheza, Nat. Prod. Rep. (1998) 187–
204.
[11] A.V. Tkachev, Chem. Nat. Comp. (Engl. Transl.) (1987) 393–412.
[12] M.R.R. Tappin, J.F.G. Pereira, L.A. Lima, A.C. Siani, J.L. Mazzei, M.F.S.
Ramos, Quim. Nova 27 (2004) 236–240.
[13] M.J.da Silva, J.A. Gonçalves, O.W. Howarth,R.B. Alves, E.V. Gusevskaya, J.
Organomet. Chem. 689 (2004) 302–308.
[14] J.A Goncalves, M.J. da Silva, D.Pilo-Veloso,O.W. Howarth, E.V. Gusevskaya,
J. Organomet. Chem. 690 (2005) 2996–3003.

[15] M.G. Speziali, P.A. Robles-Dutenhefner, E.V. Gusevskaya, Organometallics 26
(2007) 4003–4009.
[16] M.G. Speziali, V.V. Costa, P.A. Robles-Dutenhefner, E. V. Gusevskaya,
Organometallics 28 (2009) 3186–3192.
[17] A.C. Bueno, Á.O. de Souza, E.V. Gusevskaya, ChemCatChem 4 (2012) 1382–
1388.
[18] L.A. Parreira, L. Menini, E.V. Gusevskaya, Catal. Sci. Technol. 4 (2014) 2016–
2022.
[19] L.A. Parreira, A.M. da Cunha, L. Menini, E.V. Gusevskaya, RSC Adv. 5 (2015)
56987–56992.
[20] L.A. Parreira,D.C. Gonçalves, L. Menini, E.V. Gusevskaya, Appl. Catal. A 524
(2016) 126–133.
[21] L.A. Parreira, A.F. Azevedo, L. Menini, E. V. Gusevskaya, J. Mol. Catal. A 426
(2017) 429–434.
[22] J. Smidt, W. Hafner, R. Jira, J. Sedlmeier, R. Sieber, R. Ruttinger, H. Kojer,
Angew. Chem. 71 (1959) 176–182.
[23] A. Heumann, K.J. Jens, M. Réglier, in: K.D. Karlin (Ed.), Progress in Inorganic
Chemistry, Wiley, New York, 42, 1994, pp. 483–576.
[24] J.-E. Bäckvall, A. Gogoll, Tetrahedron Lett. 29 (1988) 2243–2246.
[25] S.S. Stahl, Angew. Chem. Int. Ed. 43 (2004) 3400–3420.
[26] J. Muzart, Chem. Asian J. 1 (2006) 508–515.
[27] K.M. Gligorich, M.S. Sigman, Angew. Chem. Int. Ed. 45 (2006) 6612–6615.
[28] T. Mitsudome, K. Mizumoto, T. Mizugaki, K. Jitsukawa, K. Kaneda, Angew.
Chem. Int. Ed. 49 (2010) 1238–1240.
[29] N.V. Maksimchuk, M.N. Timofeeva, M.S. Mel'gunov, A.N. Shmakov, Y.A.
Chesalov, D.N. Dybtsev, V.P. Fedin, O.A. Kholdeeva, J. Catal. 257 (2008) 315–
323.
[30] M.V. Barmatova, I.D. Ivanchikova, O.A. Kholdeeva, A.N. Shmakov, V.I.
Zaikovskii, M.S. Mel'gunov, Catal. Lett. 127 (2009) 75–82.
[31] M. Nomura, K. Hotta, Y. Fujihara, Nippon Kagaku Kaishi (1989) 475–480.
[32] K.A. da Silva Rocha, N.V.S. Rodrigues, I.V. Kozhevnikov, E.V. Gusevskaya,
Appl. Catal. A 374 (2010) 87–94.
[33]
J.C. Chalchat, R. Ph. Garry, E. Lecomte, A. Michet, Flavour Fragr. J. 6 (1991)
179–182.

[ 34 ] K.C.B. Oliveira, A. de Camargo Faria, A.C. Monteiro, E.N. dos Santos , E.V.
Gusevskaya, Appl. Catal. A523 (2016) 139–145.
[35] T. Mitsudome, T. Umetani, N. Nosaka, K. Mori. T. Mizugaki, K. Ebitani,
K. Kaneda, Angew. Chem. Int. Ed. 45 (2006) 481–485.
[36] A.C. Bueno,Á.O. de Souza, E. V. Gusevskaya, Adv. Synth. Catal. 351 (2009)
2491–2495.
[37] L.A. Parreira,L. Menini,J.C. da Cruz Santos, E.V. Gusevskaya, Adv. Synth.
Catal.352 (2010) 1533–1538.
[38] J. Piera, J.-E. Bäckvall, Angew. Chem. Int. Ed. 47 (2008) 3506–3523.
[39] N. Comey, A.I. Grey, I.L. Hook, P. James, H. Sheridan, Phytochemistry 50
(1999) 1057–1060.
[40] B. Demirci, K.H.C. Başer, T. Özek, F. Demirci, Planta Med. 66 (2000) 490–493.
[41] H. Grennberg, K. Bergstad, J.E. Bäckvall, J. Mol. Catal. A 113 (1996) 355–358.
[42] S.E. Mann, L. Benhamou, T.D. Sheppard, Synthesis 47 92015) 3079–3117.
[43] J.A Goncalves, E.V. Gusevskaya, J. Mol. Catal. A. 121 (1997) 131–137.
[44] M. dos Santos Costa, A.L.P. de Meireles, E.V. Gusevskaya, Asian J. Org. Chem.
(2017) DOI: 10.1002/ajoc.201700337R1
[45] M.S. Chen, M.C. White, J. Am. Che. Soc. 126 (2004) 1346–1347.
[46] A.F. Barrero, J. F. Quílez del Moral, M. del Mar Herrador, E.M. Sánchez, J.F.
Arteaga, J. Mex. Chem. Soc. 50 (2006) 149–156.
[47] M.von Czapiewski, M.A.R. Meier, Catal. Sci. Technol. 4 (2014) 2318–2325.

Scheme and Table captions
Scheme 1. Oxidation of -caryophyllene.
Scheme 2. Proposed mechanistic scheme for the palladium catalyzed oxidation of -
caryophyllene.
Scheme 1. Oxidation of -caryophyllene.

Scheme 2. Proposed mechanistic scheme for the palladium catalyzed oxidation of -
caryophyllene.

Table 1
Oxidation of -caryophyllene (1) with various catalytic systemsa
Run Pd catalyst
(M)
Co-catalyst
(M)
Solvent
T
P
Time C
(oC) (atm) (h)
(%) oxidation
96 traces
Selectivity for the products of (%)
isomerization addition
1b
2b
3c
4c
5d
PdCl2 (0.02) CuCl2 (0.05) HOAc
none CuCl2 (0.05) HOAc
70
70
80
80
80
80
1
3
3
6
6
6
6
38
40
62
60
0
1
92
15
28
20
5
0
0
0
0
0
PdCl2 (0.01) none
PdCl2 (0.02) none
PdCl2 (0.02) none
DMA
DMA
DMA
DMA
10
10
10
10
100
100
100
100
0
0
6c
none
none
0
a
b
Conditions: [-caryophyllene] = 0.20 M, gas phase – O2. Conversion (C) and selectivity were determined by GC and calculated based on the consumed -caryophyllene;
[LiCl] = 0.17 M. c DMA – dimethylacetamide; 20 vol% of water was added. d 10 vol% of water was added.

Table 2
a
Oxidation of -caryophyllene (1) catalyzed by Pd(OAc)2
Run [BQ]
(M)
Cu(OAc)2
(M)
T
t
Conversion Selectivity (%) Initial rate
(oC) (h) (%)
(M h-1)
2
3
1
0.20
0.20
0.20
0.20
0.10
0.10
0.10
0.05
0.05
0.05
0.05
0.05
0.05
none
70
80
90
80
70
80
90
80
90
80
80
80
90
3
7
75
95
94
98
95
98
92
99
55
78
60
80
94
100
60
95
88
96
37
82
62
96
25
64
60
92
70 15
65 16
62 18
60 20
60 16
48 14
69 21
59 23
80 10
71 13
72 10
68 14
60 20
50 25
72 10
65 15
61 18
54 20
75 10
66 14
83 12
75 15
0.08
2
none
none
0.05
0.05
0.05
0.05
0.05
0.05
0.025
0.05
0.05
0.05
3
0.12
0.14
0.12
0.04
0.06
0.10
0.04
0.07
0.02
0.04
0.02
0.04
4
3
3
5
4
3
4
5
3
10
3
6
8
7
4
6
8
3
10
4
9
6
10
11b
12c
13c
3
10
3
10
3
75
7
10
4
65 11
65 15
50 20
8
a Conditions: solvent – HOAc, [-caryophyllene] = 0.20 M, [Pd(OAc)2] = 0.01 M, gas phase – O2, 1 atm.
Conversion and selectivity were determined by GC and calculated based on the consumed substrate.
Initial reaction rates were calculated at low conversions (up to 20– 40%). b [Pd(OAc)2] = 0.02 M.
c Pd(OAc)2] = 0.005 M.

Table 3
Oxidation of -caryophyllene (1) with molecular oxygen catalyzed by Pd(OAc)2/p-
benzoquinone (BQ)a
Run [BQ]
(M)
T
t
Conversion Selectivity (%) Initial rate
(oC) (h) (%)
(M h-1)
2
3
1
2
3
4
5
6
7
8b
0.10
0.10
0.10
0.10
0.05
0.05
0.05
0.05
80
70
60
50
80
70
60
70
3
77
96
53
100
45
100
27
85
62
96
45
90
35
85
60
95
50 17
45 15
65 11
56 16
70 11
66 11
72 10
70 11
50 17
47 15
0.06
4
3
0.04
0.03
0.02
0.04
0.03
0.02
0.05
8
3
24
3
24
3
6
3
63
9
8
59 11
72 13
68 12
65 10
55 15
3
10
3
6
a Conditions: solvent – HOAc, [-caryophyllene] = 0.20 M, [Pd(OAc)2] = 0.01 M, gas phase – O2, 10 atm.
Conversion and selectivity were determined by GC and calculated based on the consumed substrate.
Initial reaction rates were calculated at low conversions (up to 20– 40%). b [Pd(OAc)2] = 0.02 M.