Aerobic oxidation of naturally occurring α-bisabolol catalyzed by palladium(II) salts as sole catalysts

Article abstract of DOI:10.1016/j.apcata.2016.06.027

α-Bisabolol, a sesquiterpene found in various essential oils, has numerous direct commercial uses in cosmetic and therapeutic formulations. In the present work, we report two novel liquid-phase processes for the catalytic oxidation of α-bisabolol with dioxygen, in which palladium salts (PdCl₂ or Pd(OAc)₂) are employed as sole catalysts. The addition of co-catalysts conventionally used for the re-oxidation of reduced palladium species or special ligands for their stabilization is not required. The reactions occur in non-acidic solvents, i.e., aqueous methanol or dimethylacetamide, and give exclusively the products derived from the interaction of the acyclic olefinic bond with palladium, whereas the other olefinic bond remains intact. The method allows to perform the green and simple synthesis of poly-functionalized compounds potentially useful for fine chemical industry as fragrance ingredients and synthetic intermediates, starting from a bio-renewable substrate.

Full text of DOI:10.1016/j.apcata.2016.06.027

Applied Catalysis A: General 524 (2016) 126–133
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Aerobic oxidation of naturally occurring ␣-bisabolol catalyzed by
palladium(II) salts as sole catalysts
Luciana A. Parreiraa,b, Débora C. Gonc¸ alvesc, Luciano Meninic, Elena V. Gusevskayaa,∗
a Departamento de Química, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil
b Departamento de Química e Física, Campus de Alegre—CCA, Universidade Federal do Espírito Santo, 29500-000 Caixa Postal 16 Alegre, ES, Brazil
c Campus de Alegre Instituto Federal de Educac¸ ão, Ciência e Tecnologia do Espírito Santo, 29500-000 Caixa Postal 47 Alegre, ES, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 May 2016
Received in revised form 16 June 2016
Accepted 18 June 2016
Available online 23 June 2016
␣-Bisabolol, a sesquiterpene found in various essential oils, has numerous direct commercial uses in cos-
metic and therapeutic formulations. In the present work, we report two novel liquid-phase processes
for the catalytic oxidation of ␣-bisabolol with dioxygen, in which palladium salts (PdCl2 or Pd(OAc)2)
are employed as sole catalysts. The addition of co-catalysts conventionally used for the re-oxidation of
reduced palladium species or special ligands for their stabilization is not required. The reactions occur
in non-acidic solvents, i.e., aqueous methanol or dimethylacetamide, and give exclusively the products
derived from the interaction of the acyclic olefinic bond with palladium, whereas the other olefinic bond
remains intact. The method allows to perform the green and simple synthesis of poly-functionalized com-
pounds potentially useful for fine chemical industry as fragrance ingredients and synthetic intermediates,
starting from a bio-renewable substrate.
Keywords:
Bio-renewables
␣-Bisabolol
Oxidation
Oxygen
Palladium
Terpenoids
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
The palladium catalyzed oxidation of olefins is a convenient
and widely used in organic synthesis route to various oxygenated
Terpenes are natural compounds present in a variety of essential
oils, often as main components. They are widely used as ingre-
dients and raw materials in flavor&fragrance and pharmaceutical
industries due to specific olfactory characteristics and interesting
therapeutic properties [1,2]. In particular, a sesquiterpenic alcohol
␣-bisabolol occurs in high concentrations in chamomile, candeia
and sage essential oils (up to 50, 85 and 90%, respectively) [3].
As natural sources do not currently satisfy a high commercial
demand for ␣-bisabolol, it is also produced synthetically from other
sesquiterpenic alcohols farnesol and nerolidol [4–7].
␣-Bisabolol is a component in over 1000 cosmetic, fragrance
and therapeutic formulations [3,8]. Despite of such important and
numerous direct uses, the catalytic oxidation of ␣-bisabolol, which
contains two olefinic bonds, can open new perspectives for the
applications of this compound. It has been reported that ␣-bisabolol
derivatives, both natural and synthetic, also show an important
therapeutic potential presenting, for example, anti-inflammatory
and anti-tumoral activities [5,9,10].
compounds, in particular, for the industrial transformation of ethy-
lene into acetaldehyde (the Wacker process) [11–13]. The attractive
feature of these reactions from economic and environmental view-
points is the use of molecular oxygen as the terminal oxidant.
On the other hand, the disadvantage of most of the conven-
tional palladium-based catalytic systems for aerobic oxidations is a
requirement for auxiliary co-catalysts. The role of these co-catalysts
is to act as reversible co-oxidants capable to re-oxidize reduced
palladium species before their irreversible agglomeration in a bulk
metal and to be re-oxidized back by molecular oxygen. CuCl2 is the
most commonly used co-catalyst in these systems (Wacker cat-
alyst); however, corrosion and selectivity issues often become a
problem due to the high concentrations of chloride ions and acidity
of reaction solutions.
Heteropoly compounds, nitrates and p-benzoquinone (BQ) have
been studied among other compounds in palladium-based systems
to replace CuCl2 and create more benign catalytic systems which do
not contain chloride ions [14–21]. Furthermore, in the past decade
several examples of the aerobic oxidation of olefins catalyzed by
palladium(II) complexes in the absence of auxiliary co-oxidants
have been published [22–33]. In these systems, the regeneration
of palladium was performed directly by dioxygen through the sta-
∗
Corresponding author.
E-mail address: elena@ufmg.br (E.V. Gusevskaya).
http://dx.doi.org/10.1016/j.apcata.2016.06.027
0926-860X/© 2016 Elsevier B.V. All rights reserved.

L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
127
1.90-2.10, 2H, m
1.60-1.70, 1H, m
30.87
1.15-1.30, 1H, m
30.81
23.96
1.64, s
23.23
1.31, s
29.71
14
23.23
6
1.08, s
22.87
22.88
1
5
7
9
134.06
133.73
OH
70.51
70.59
OH
13
4
8
12
11
2
5.36, br.s
120.35
74.05
74.62
3
1.55-1.65, m
43.08
5.66-5.72, m
142.19
120.52
44.15
140.10
15
1.31, s
29.71
1.75-1.85, 1H, m
1.90-2.10, 1H, m
26.82
10
2.20, br.d
3J = 7.2 Hz
42.83
26.40
5.66-5.72, m
121.84
123.50
46.58
Fig. 1. The attribution of the NMR signals for product 3 (two isomers).
1.27 s
14
25.49; 29.75
1.25-1.35, 1H, m
1.24 s; 27.98
1.65-1.85, 1H, m
23.31; 24.03
9 1.12 s; 22.75
13
1.90-2.00, m
31.09; 31.13
15
O
70.03
74.38
6
5
4
1.21; 1.24 s
30.99; 31.07
1.63 s; 1.64 s
23.41; 23.46
75.49
8
12
74.38
7
1
5.59 d, 3J=10.4, 131.52
5.67, br.s; 134.38
134.15
133.48
1.40-1.50, m; 45.58
1.65-1.75, m; 42.96
2
3
10
11 5.80-5.85. m; 122.78
5.67, br.s; 120.46
5.36, br.s
120.85
121.31
1.70-1.90, 1H, m
1.95-2.15, 1H, m
26.12; 26.78
1.80-1.90, 1H, m
1.90-2.00, 1H, m
35.94
1.70-1.80, 1H, m
2.05-2.15, 1H, m
31.70
Fig. 2. The attribution of the NMR signals for product 5 (two isomers).
bilization of palladium(0) species in solutions by special ligands or
by coordinating solvents.
under chloride free conditions to obtain several novel oxygenated
sesquiterpenoid compounds. Further, we have started a search for
the systems which do not require auxiliary co-catalysts at all.
Herein, we present novel processes for the palladium catalyzed
direct-dioxygen-coupled oxidation of ␣-bisabolol. The reactions
occur under mild nonacidic conditions and involve molecular oxy-
gen as the stoichiometric oxidant and palladium salts (PdCl2 or
Pd(OAc)2) as the sole catalysts. Oxygenated poly-functionalized
compounds derived from bio-renewable ␣-bisabolol are poten-
tially useful in fine chemical industry as fragrance ingredients and
synthetic intermediates.
Within our current project aiming to add value to natural
essential oils by catalytic reactions, we have studied the Pd(II) cat-
alyzed oxidation of various naturally occurring alkenes and alkenyl
alcohols, i.e., nerolidol ␣-terpineol and linalool [17–21,27,28].
Our particular interest is currently directed to the oxidation of
␣-bisabolol as a route to interesting poly-functionalized sesquiter-
penic compounds, which would by hardly accessible by other
synthetic methods [34,35].
Before starting the project, we could find very limited informa-
tion on the catalytic oxidation of ␣-bisabolol [36,37]. Both these
publications reported the processes in which hydrogen peroxide
or other peroxo derivatives were used as stoichiometric oxidants.
As far as we know, two our recently published works [34,35] are
the only available reports on the application of catalytic chemistry
of palladium to the oxidation of ␣-bisabolol.
As the attempts to apply the conventional Wacker catalyst
(PdCl2/CuCl2) to ␣-bisabolol were not encouraging in terms of
reaction selectivity, we first tried to find suitable co-catalysts
which could be an alternative to cupper chloride. The catalytic
combinations of palladium(II) acetate with either BQ [34] or cup-
per(II) acetate [35] allowed to perform the aerobic oxidation of
␣-bisabolol in acetic acid or aqueous methanol media, respectively,
2. Experimental
Natural ␣-(−)-bisabolol originated from the candeia (Ereman-
thus erythropappus) extract was received from CITRÓLEO Ind. Com.
Óleos Essenciais as a kind donation. Palladium (II) acetate (>99.9%)
was purchased from Aldrich and palladium (II) chloride (99.9%)
from Strem. The copper content in both palladium salts deter-
mined by atomic absorption spectroscopy was less than 0.005%.
Lithium chloride and copper (II) acetate were heated for dehy-
dration. The catalytic tests were performed in a stainless steel
autoclave (total volume of 100 mL) with magnetic stirring. The
reaction progress was monitored by gas chromatography (GC) be

128
L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
Fig. 3. The attribution of the NMR signals for product 6.
Table 1
Oxidation of ␣-bisabolol with dioxygen in methanol.a
Run
Extra water (vol%)
T (◦C)
P (atm)
Time (h)
Conversion (%)
Selectivity (%)
TOFb (h−1
)
2
3
4
c
8
8
8
15
4
8
8
8
8
80
80
60
60
60
60
60
60
60
10
10
5
10
10
10
10
10
10
5
87
63
8
10
4.8
4.4
1.8
2.2
2.6
2.3
2.0
18.0
2.0
2
3
4
5
4 7
3 7
7 10
7 10
9
10
2
9
55 57
27 32
47 51
48 50
92
83
100
50
45 48
43 45
32 33
46 45
48
81
–
tr.
6 7
38 37
52 50
27 26
37
7
–
tr.
9 9
9 10
10 12
9 12
5
10
–
tr.
6
7c
8d
9e
a
Conditions: ␣-bisabolol (0.20 M), [Pd(OAc)2] (0.01 M); gas phase – O2, total volume 12 mL. Conversion and selectivity were obtained from GC data and referred to the
amounts of the reacted substrate; tr. – trace amounts.
b
Average rate (turnover frequency) of the substrate conversion per mol of palladium for the first reaction hour.
Cu(OAc)2 added (0.05 M).
Pd(OAc)2 was replaced by PdCl2.
Pd(OAc)2 was replaced by Pd(OCOCF3)2.
c
d
e
OH
OR
2 (R=CH3)
3 (R=H)
Pd(OAc)2 (cat)
O2, CH3OH
(up to 85%)
OH
O
1
(~10%)
4
Scheme 1. Oxidation of ␣-bisabolol in methanol.
periodical sampling. Typically, ␣-(−)-bisabolol, palladium catalyst,
water and bornyl acetate (used as a GC standard, 0.10 M) were dis-
solved in indicated concentrations in a specified solvent (12 mL
total volume). The solution was put in the autoclave and the indi-
cated pressure was attained with dioxygen as the gas phase. An
oil bath was used to heat the reactor and to maintain the reac-
tion temperature. The samples of the reaction solutions for the GC
analysis were taken using a special device maintaining the pres-
sure inside the reactor. GC analyses were performed on a Shimadzu
GC-2010 Plus chromatograph using a Rtx-Wax 30 m or Rtx-5MS
30 m capillary column and a flame ionization detector. Conversions
and selectivities were calculated using as a reference the amount
of the substrate reacted. Average turnover frequencies, TOF, were
measured for the first reaction hour (in most cases up to 20–40%
conversions).
The identification of the products was performed by 1H and
13C NMR and GC–MS spectroscopy after their isolation from the
reaction solutions by a column chromatography (silica gel 60); hex-
ane and dichloromethane and their solutions were used as eluents.
NMR experiments were conducted on a Bruker 400 MHz instru-
ment (CDCl3 as the solvent, TMS as the internal standard). Mass
spectra were acquired on a Shimadzu QP2010-PLUS equipment
(70 eV).

L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
129
Table 2
Oxidation of ␣-bisabolol with dioxygen in dimethylacetamide.a
Run
Added water (vol%)
[PdCl2] (mM)
T (◦C)
Time (h)
Conversion (%)
Selectivity (%)
TOFb (h−1
)
3
4
5
6
1
15
15
15
15
15
15
15
15
15
8
10
60
60
60
60
60
60
80
40
30
60
60
60
2
4
4
1
6
4
1
6
95
0
0
13
–
–
2
–
–
7
8
2
tr.
11
6
35
–
–
36
–
–
11.0
–
–
2
none
none
20
5
10
10
10
10
10
3c
4
90
57
93
94
86
70
95
45
70
16
13
10
10
10
10
11
tr.
–
30
38
35
27
25
19
25
14
tr.
37
22
34
21
40
51
30
24
tr.
9.0
9.6
6.0
18.8
8.0
4.8
9.0
4.0
1.0
5
6d
7d
8
9
11
10
4
10
11
12e
3
3
20
0
15
10
10
13
a
Conditions: ␣-bisabolol (0.20 M), [Pd(OAc)2] (0.01 M); 10 atm of O2, total volume 12 mL. Conversion and selectivity were obtained from GC data and referred to the
amounts of the reacted substrate; tr. – trace amounts.
b
Average rate (turnover frequency) of the substrate conversion per mol of palladium for the first reaction hour.
LiCl was added (0.02 M).
5 atm.
c
d
e
PdCl2was replaced by Pd(OAc)2.
Product 2 (isomers 2a and 2b) (first description was published
in [35]): 2a: MS (70 eV, EI): m/z (%): 202 (1) [M+−H2O−CH3OH)],
139 (32), 125 (9), 121 (18), 95 (46), 82 (100), 71 (18%), 67 (32);
2b (longer GC retention time): MS (70 eV, EI): m/z (%): 202 (1)
[M+−H2O−CH3OH)], 125 (4), 95 (4), 73 (100), 67 (3). For NMR data
see [35].
Product 3 (novel compound as far as we know, isomers 3a
and 3b): 3a: MS (70 eV, EI): m/z (%): 220 (1) [M+−H2O], 202 (1)
[M+−H2O−H2O)], 147 (15), 139 (30), 125 (31), 121 (30), 99 (38), 95
(69), 85 (61), 82 (100), 67 (46), 59 (61). 3b: MS (70 eV, EI): m/z (%):
220 (1) [M+−H2O], 202 (1) [M+−H2O−H2O)], 147 (22), 125 (31), 95
(36), 94 (51), 85 (100), 82 (25), 79 (37), 67 (26), 59 (90), 43 (94). The
attribution of NMR signals is shown in Fig. 1.
Product 4 (first description was published in [29]): MS (70 eV,
EI): m/z (%): 220 (1) [M+], 187 (1) [M+−H2O−CH3], 132 (18), 125
(100), 107 (61), 95 (14), 93 (13), 67 (18). For NMR data see [34].
Product 5 (novel compound as far as we know): MS (70 eV, EI):
m/z (%): 220 (2) [M+], 205 (3) [M+−CH3], 187 (2) [M+−CH3−H2O],
125 (100), 107 (34), 83 (12), 43 (46). Product 5 showed only one
peak on chromatograms; however, the NMA analysis revealed the
existence of two isomers with very similar spectra. The attribution
of NMR signals is shown in Fig. 2.
Product 6 (known as norbisabolide): MS (70 eV, EI): m/z (%): 194
(28) [M+], 179 (4) [M+−CH3], 161 (8) [M+−CH3−H2O], 134 (22), 121
(64), 119 (22), 99 (100), 93 (78), 81 (19), 79 (19), 71 (28), 67 (22),
42 (49). The attribution of NMR signals is shown in Fig. 3.
tration of copper. Encouraged by this observation, we performed
the process in the absence of copper acetate. Despite the decrease in
selectivity for product 2 and reaction stagnation at ca. 50% conver-
sion, the reaction was still catalytic with respect to palladium and
showed the dioxygen-coupled turnover number (TON) of nearly 10
(Table 1, run 2). No formation of the palladium mirror on the walls
of the reactor was detected at the end of the reaction.
Varying reaction parameters (Table 1, runs 3–6), we found
that the reaction could be nearly completed in 9 h at 60 ◦C and
10 atm to give methyl ether 2 and corresponding diol 3 in 85%
combined selectivity and comparable amounts (Table 1, run 6). Het-
erocyclic compound 4 was detected in these runs in small amounts
(5–10%). All three products were formed due to the oxidation of the
acyclic C C bond with the participation as nucleophiles of external
methanol or water molecules to give products 2 and 3, respectively,
or the internal tethered hydroxyl group to give cyclization product
4. In addition to the products specified in Table 1, bisabolol oxide B
(identified by MS) was formed in most of the runs in small amounts.
Diol 3, a novel compound to the best of our knowledge, was sep-
arated after the reaction as a mixture of two isomers 3a and 3b. The
NMR spectra of these isomers are very similar (Fig. 1, Experimental
Section). Both 3a and 3b seem to have a trans configuration at the
acyclic double bond as the hydrogens at C-10 and C-12 correlated in
the NOE spectrum and the chemical shift values for allylic carbons
C-13 and C-10 in both isomers were close. We suppose that the
structural difference between the two isomers of diol 3 is related
to the configuration at carbon C-8, similarly to what we suggested
for the isomers of ether 2 [35].
3. Results and discussion
Compound 4 was reported for the first time in our previous work
as a major oxidation product formed from ␣-bisabolol in the acetic
acid solutions containing palladium acetate and BQ in catalytic
amounts [34].
3.1. Oxidation of ˛-bisabolol in methanol solutions
The data on the oxidation of ␣-bisabolol (1) by dioxygen in aque-
ous methanol solutions containing Pd(OAc)2 in catalytic amounts
are collected in Table 1. In our previous work [35], Cu(OAc)2 was
used as the co-catalyst together with Pd(OAc)2 for the oxidation of
␣-bisabolol. The main reaction products were compounds 2 (up to
85% selectivity) and 4 (ca. 10% selectivity) (Scheme 1). Both prod-
ucts were novel bisabolane derivatives formed due to the oxidation
of the acyclic olefinic bond of the substrate. The example of the reac-
tion in the system containing Cu(OAc)2 is presented in Table 1 (run
1).
It is surprising that the reactions with and without Cu(OAc)2
occurred at similar rates and with similar selectivities for diol
products 2 and 3 and for cyclization product 4 (Table 1, run 7 vs.
run 6). However, differently from “palladium solo” reactions, in
the bimetallic system methyl ether 2 was the main diol product,
whereas diol 3 itself was detected only in small amounts (Table 1,
runs 1 and 7). It can be suggested that Lewis acidity of copper ions
favored the etherification of diol 3 with methanol to give ether 2.
However, the second hydroxyl group in diol 3 did not undergo
etherification in these systems. The more plausible explanation
looks a suggestion that the presence of copper ions increases the
relative contribution of methanol as the nucleohile at the oxidation
The oxidation of ␣-bisabolol in the bimetallic system was accel-
erated by the increase in the palladium concentration; however, it
was (quite unexpectedly) only slightly dependent on the concen-

130
L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
O
5 (up to 40%)
PdCl2 (cat)
O2, DMA
OH
O
O
1
6 (up to 50%)
OH
OH
3
(10-15%)
Scheme 2. Oxidation of ␣-bisabolol in dimethylformamide.
step (to give ether 2) as compared to water (to give diol 3), possi-
bly, due to the formation of copper methoxy complexes. In blank
reactions without Pd(OAc)2, none of oxidation products 2, 3 and 4
was expectedly detected in reaction solutions.
The decrease in the oxygen pressure did not affect signifi-
cantly the product distribution and initial turnover frequency (TOF)
(Table 1, run 3 vs run 6). However, the reaction at 5 atm turned
stagnated at ca. 30% conversion with the formation of inactive pal-
ladium metal. Therefore, to ensure the effective re-oxidation of
palladium the reaction should be performed under higher oxygen
pressure.
The nature of anionic ligands on palladium is also remarkably
important for the successful operation of palladium as the sole cata-
lyst in the oxidation of ␣-bisabolol. The reactions with the catalysts
generated from PdCl2 and Pd(OCOCF3)2 showed very poor selec-
tivities for products 2, 3 and 4, with numerous unidentified peaks
being detected by GC (Table 1, runs 8 and 9).
3.2. Oxidation of ˛-bisabolol in dimethylacetamide solutions
In 2006, Kaneda and co-workers disclosed that PdCl2 catalyzed
the oxidation of some terminal [39] and internal [26] alkenes with
molecular oxygen in dimethylacetamide (DMA) solutions with-
out the addition of co-oxidants or special ligands. Latter, we have
extended the method to the oxidation of styrene [27] and naturally
occurring allyl benzenes [28], the substrates containing terminal
olefinic bonds. It has been suggested that DMA as a coordinating
solvent stabilizes palladium(0) species thus allowing to perform
direct-dioxygen-coupled catalytic cycles. Based on these results,
we decided to apply the approach to the oxidation of ␣-bisabolol
using dioxygen as the sole oxidant (Table 2).
The reactions were performed in aqueous DMA solutions con-
taining catalytic amounts of PdCl2 and, in most of the runs, 10 atm
of oxygen to ensure the efficient capture of the reduced palladium
species. Compounds 5 and 6 were detected as major products in
ca. 70% combined selectivity along with smaller amounts of diol
3 (10–15%) and tetrahydrofuran product 4 (5–10%) (Scheme 2).
Compounds 5 and 6 were isolated from the reaction solutions and
identified by GC–MS and NMR spectroscopy.
Product 5, a novel compound to the best of our knowledge, was
observed by NMR as two not GC separable isomers with very similar
spectra (Fig. 2, Experimental Section). As natural ␣-(−)-bisabolol
has R configurations at carbons 4 and 8, we suppose that the partial
inversion of the configuration occurs at C-8, which is linked to the
involved in the catalytic transformation OH group, so that the two
isomers of 5 differ from each other by geometry at C-8.
Product 6 has been identified as a ␥-butyrolactone compound
shown in Scheme 2, which is known as norbisabolide, a natural
compound extracted from Atalantia Monophylla citric plant [35]
(NMR attributions are shown in Fig. 3), Experimental Section. Com-
pound 6 was described by Solabannavar et al. [40].
It is remarkable that the reaction with ␣-bisabolol proceeds
in aqueous methanol solutions with dioxygen-coupled turnovers
without the use of auxiliary co-oxidants or special auxiliary lig-
ands for the stabilization of reduced palladium species. In this
context, it is worthwhile to cite the example of the related “palla-
dium solo” reaction which we have found in the literature. As early
as in 1978, Hosokawa et al. described the oxidative cyclization of
allylphenol substrates with molecular oxygen catalyzed by palla-
dium(II) acetate alone in the solutions of aqueous methanol [38].
We suppose that palladium(0) species in these systems are able
to form relatively stable complexes with the solvent and/or with
the substrate which decelerates their aggregation. In other words,
the coordination enables the reoxidation of palladium(0) species
with dioxygen to occur faster than palladium deactivation due to
clustering into inactive bulk metal.
The nature of the olefinic substrate is very important for the
possibility to run the process without Cu(OAc)2, i.e., to perform
a “palladium solo” reaction. Although the related substrate 3,7-
dimethyl-1,6-octadien-3-ol (linalool) did react with dioxygen in
the presence of the bimetallic Pd(OAc)2 + Cu(OAc)2 catalytic com-
bination [19], the attempts to perform the reaction with Pd(OAc)2
alone resulted in the appearance of the palladium mirror on the
walls of the reactor (not shown in the table). Palladium black was
also formed at the attempt to oxidize another monoterpenic acyclic
alcohol, 3,7-dimethyl-2,6-octadien-1-ol (nerol), in the absence of
Cu(OAc)2. All three substrates: linalool, nerol, and ␣-bisabolol, con-
tain in their molecules the same fragment with a trisubstituted
acyclic double bond and namely this bond is oxidized in our reac-
tions. It seems that the presence of a six-membered unsaturated
ring in the ␣-bisabolol molecule plays a key role in its capacity to
stabilize the reduced palladium species in the course of the cat-
alytic process thus allowing to maintain the catalytic cycle without
the need for auxiliary co-catalysts.
All identified products detected at the oxidation of ␣-bisabolol
in DMA solutions were originated by the interaction of the acyclic
olefinic bond with palladium. Products 3, 4 and 5 keep the car-
bon skeleton of the original molecule, whereas the formation of

L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
131
Graphical abstract.doc
revised manuscript.doc
Scheme 3. Suggested mechanism for the formation of products 2, 3, 4 and 5.
ketone 6 involves its oxidative cleavage. The endocyclic double
bond remained intact in all these products.
product 6 became higher at the expense of product 5. However, the
combined selectivity for products 5 and 6 remained nearly constant
(ca. 70%). The process can be performed at nearly ambient temper-
ature. Although the reaction rate became lower, the selectivity of
51% obtained for norbisabolide (product 6) was the best value we
could achieve so far (Table 2, run 9).
At lower water concentrations, the reaction was slower and less
selective (Table 2, cf. runs 1, 10 and 11). The accelerating effect
of water was also reported for the Pd/(−)-sparteine [41] and PdCl2
[27,28] catalyzed oxidation of terminal olefins with molecular oxy-
gen in DMA solutions. However, water concentrations higher than
15 vol% can not be recommended due to miscibility problems.
␣-Bisabolol readily reacted with PdCl2 in aqueous DMA solu-
tions (15 vol% of water). A nearly complete conversion occurred in
a two-hour reaction at 60 ◦C to give compounds 5 and 6 in ca. 35%
selectivity each (Table 2, run 1). The reaction was catalytic in pal-
ladium showing TON of ca. 20. Thus, DMA efficiently stabilizes the
palladium(0) species so that their reaction with dioxygen occurs
faster than the precipitation of palladium black.
In blank reactions, with no PdCl2 or with LiCl instead of PdCl2,
the conversion of ␣-bisabolol was not observed (Table 2, runs 2 and
3). It is important, that no formation of even trace amounts of com-
pound 6, which might be the product of the non-catalytic oxidative
cleavage of the ␣-bisabolol molecule, was detected in these runs.
The reaction rate was found to be roughly proportional to the
PdCl2 amounts as initial TOF values in the experiments with dif-
ferent PdCl2 concentrations were similar (Table 2, runs 1, 4 and 5).
With 10 mol% of PdCl2, the reaction was nearly completed within
1 h to give products 3, 4, 5 and 6 with 90% combined selectivity.
Bisabolol oxide B was mainly responsible for the rest of the mass
balance in this system also. The reaction could be accelerated by the
increase in the oxygen pressure (Table 2, cf. runs 1 and 6). Under
5 atm, a nearly complete conversion occurred in 4 h with no forma-
tion of the palladium mirror. The changes in the oxygen pressure
and catalyst amounts did not affect significantly a product distri-
bution. The positive effects of palladium and oxygen on reaction
kinetics suggest that the rate determining step of the whole pro-
cess can be the regeneration of palladium by dioxygen. It should
be mentioned that in the reaction under atmospheric pressure, the
precipitation of palladium metal was observed.
3.3. Reaction mechanism
The acyclic olefinic bond of ␣-bisabolol was the only one oxi-
dized in “palladium solo” reactions in both methanol and DMA
solutions. No products derived from the oxidation of the endocyclic
olefinic bond were registered in detectable amounts. In this context,
it seems remarkable the difference between these systems and the
Pd(OAc)2/BQ system described in our previous work [34], which
promoted the oxidation of both olefinic bonds in ␣-bisabolol.
The mechanism proposed for the formation of products 2, 3,
4 and 5 is shown in Scheme 3. As it is generally accepted that
the palladium promoted oxidations of alkenyl alcohols involve an
oxypalladation step [4–45], we suggest such interaction as a key
reaction feature in our process. The Markovnikov oxypalladation of
the acyclic double bond leads to products 2, 3 and 5. In this case, the
nucleophile attacks a trisubstituted vinylic carbon atom. The anti-
Markovnikov oxypalladation leads to product 4. The Markovnikov
type interaction is a more favorable alternative because palladium
in corresponding ␴-alkyl palladium intermediates is linked to a
sterically more accessible carbon atom. Probably for this reason,
compound 4 was detected only as the minor product in all the
runs. ␴-Alkyl palladium intermediates A (the precursors of diol
The reaction was much faster at 80 ◦C being nearly completed
for 1 h vs. 4 h at 60 ◦C; however, selectivity was lower, especially for
product 6 (Table 2, run 7 vs. run 6). The tendency was confirmed in
another set of the run (Table 2, runs 1, 8 and 9): at lower temper-
ature the contribution of the carbon–carbon bond cleavage to give

132
L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
Bisabolol, a renewable sesquiterpene found in essential oils of
OH
various plants, contains two olefinic bonds, both trisubstituted.
All reaction products arise exclusively from the oxidation of the
acyclic olefinic bond. Although the endocyclic double bond is usu-
ally involved in palladium catalyzed reactions with more facility,
it remains intact. In general, internal acyclic olefinic bonds in
non-functionalized alkenes are very reluctant to Wacker-type oxi-
dations, with only few examples being reported so far [26]. The
success in the oxidation of ␣-bisabolol was probably enabled by the
existence of the hydroxyl group in a ␥-position to the acyclic dou-
ble bond, which could favour the interaction of th double bond with
palladium. These simple catalytic reactions represent environmen-
tally and commercially interesting routes to poly-functionalized
products potentially useful in fine chemical industry as fragrance
ingredients and synthetic intermediates. The use of the renew-
able biomass-based substrate, low-cost nonacidic non-corrosive
solvents, palladium salts as the sole catalysts and dioxygen as the
stoichiometric oxidant is an advantage particularly important for
the green chemistry concept.
1
[Pd]
OH
[Pd]
OH
O
OH
O
hydroxyaldehyde
hydroxycarboxylic acid
- H2O
[Pd]
O
O
O
OH
- 2[H]
6
hemiacetal
Scheme 4. Suggested mechanism for the formation of ␥-butyrolactone 6.
Acknowledgments
3 and ether 2) are formed through the nucleophilic attack by the
external molecule of water or methanol. The participation of the
internal hydroxyl group at the oxypalladation step leads to the
molecule cyclization. The hydroxyl group can attack both vinylic
carbons coordinated on palladium to form either a tetrahydrofu-
ran or tetrahydropyran ring (intermediates B and C, respectively).
Finally, the abstraction of ␤-hydrogens in ␴-alkyl intermediates A,
B and C results in the formation of products 2 (or 3), 4 and 5.
It is interesting that the direction of the ␣-bisabolol cycliza-
tion (furan vs. pyran products) is affected by the nature of anionic
groups on palladium. The reactions with Pd(OAc)2 (see Table 1 and
our previous works [34,35]) give preferably five-membered prod-
uct 4, which is the result of the anti-Markovnikov oxypalladation.
On the other hand, in the system with PdCl2 the major cycliza-
tion product is six-membered compound 5, which arises from the
Markovnikov oxypalladation. Moreover, when PdCl2 was substi-
tuted by Pd(OAc)2, not only the reaction became slower and less
selective, but also the cyclization selectivity was switched almost
completely from six-membered compound 5 to five-membered
product 4 (Table 2, run 12 vs. run 1).
The authors thank the Citróleo Ind. Com. Óleos Essenciais for
the donating of ␣-(−)-bisabolol, Dr. Patrícia Fontes Pinheiro (Uni-
versidade Federal do Espírito Santo) for the help in contacting the
Citróleo industry and the CNPq, FAPEMIG, FAPES and INCT Catálise
(Brazil) for funding this work.
References
[1] C. Sell (Ed.), The Chemistry of Fragrances: from Perfumer to Consumer, vol. 2,
second ed., RSC Publishing, Dorset, UK, 2006, pp. 52–88.
[2] H. Mimoun, Chimia 50 (1996) 620–625.
[3] G.P.P. Kamatou, A.M. Viljoen, J. Am. Oil Chem. Soc. 87 (2010) 1–7.
[4] O. Taglialatela-Scafati, F. Pollastro, L. Cicione, G. Chianese, M.L. Bellido, E.
¨
Munoz, H.C. Ozen, Z. Toker, G. Appendino, J. Nat. Prod. 75 (2012) 453–458.
[5] K. Massonne, K.P. Pfaff, J. Schubert, G. Gottwald (BASF), DE 102005053329
(2007).
[6] K. Massonne, K.P. Pfaff, J. Schubert, G. Gottwald (BASF), DE 102005053338
(2007).
[7] A.L.P. de Meireles, M. dos Santos Costa, K.A. da Silva Rocha, E.V. Gusevskaya,
Appl. Catal. A 502 (2015) 271–275.
[8] K. Russell, S.E. Jacob, Dermatitis 21 (2010) 57–58.
[9] M. Piochon, J. Legault, C. Gauthier, A. Pichette, Phytochemistry 70 (2009)
228–236.
␥-Butyrolactone 6 is the product of the oxidative cleavage of
the acyclic double bond, the reaction much less common in the
catalytic chemistry of palladium than transformations through the
oxypalladation step shown in Scheme 3. H. Jiang et al. have recently
reported that palladium acetate catalyzes the oxidative cleavage of
multiple C C bonds in alkenes [46] and alkynes [47] by molec-
ular oxygen to give carbonyl and carboxyl products (aldehydes,
ketones or carboxylic acid derivatives). Based on these literature
data we are suggesting two possible routes to product 6 from ␣-
bisabolol (Scheme 4). The oxidative cleavage of the olefinic bond
could give a hydroxyaldehyde, which then undergoes a spon-
taneous intramolecular cyclization resulting in a five-membered
hemiacetal. The subsequent palladium catalyzed dehydrogenation
of the hemiacetal would give ␥-lactone 6. Alternatively, the product
of the oxidative cleavage of ␣-bisabolol could be a hydroxycar-
boxylic acid, whose spontaneous cyclization would give directly
product 6.
[10] A.P. da Silva, M.V. Martini, C.M.A. de Oliveira, S. Cunha, J.E. de Carvalho,
A.L.T.G. Ruiz, C.C. da Silva, Eur. J. Med. Chem. 45 (2010) 2987–2993.
[11] W. Wu, H. Jiang, Acc. Chem. Res. 45 (2012) 1736–1748.
[12] B.V. Popp, S.S. Stahl, Top. Organomet. Chem. 22 (2007) 149–189.
[13] J.A. Keith, P.M. Henry, Angew. Chem. Int. Ed. 48 (2009) 9038–9049.
[14] A. Heumann, K.J. Jens, M. Réglier, in: K.D. Karlin (Ed.), Progress in Inorganic
Chemistry, vol. 42, Wiley, New York, 1994, pp. 542–576.
[15] T. Yokota, A. Sakakura, M. Tani, S. Sakaguchi, Y. Ishii, Tetrahedron Lett. 43
(2002) 8887–8891.
[16] J. Piera, J.-E. Bäckvall, Angew. Chem. Int. Ed. 47 (2008) 3506–3523.
[17] J.A. Gonc¸ alves, E.V. Gusevskaya, Appl. Catal. A 258 (2004) 93–98.
[18] J.A. Gonc¸ alves, O.W. Howarth, E.V. Gusevskaya, J. Mol. Catal. A 185 (2002)
95–104.
[19] M.G. Speziali, P.A. Robles-Dutenhefner, E.V. Gusevskaya, Organometallics 26
(2007) 4003–4009.
[20] M.G. Speziali, V.V. Costa, P.A. Robles-Dutenhefner, E.V. Gusevskaya,
Organometallics 28 (2009) 3186–3192.
[21] A.C. Bueno, Á.O. de Souza, E.V. Gusevskaya, ChemCatChem 4 (2012)
1382–1388.
[22] C.N. Cornell, M.S. Sigman, Inorg. Chem. 46 (2007) 1903–1909.
[23] S.S. Stahl, Science 309 (2005) 1824–1826.
[24] J. Muzart, Chem. Asian J. 1 (2006) 508–515.
[25] K.M. Gligorich, M.S. Sigman, Angew. Chem. Int. Ed. 45 (2006) 6612–6615.
[26] T. Mitsudome, K. Mizumoto, T. Mizugaki, K. Jitsukawa, K. Kaneda, Angew.
Chem. Int. Ed. 49 (2010) 1238–1240.
4. Conclusion
[27] A.C. Bueno, Á.O. de Souza, E.V. Gusevskaya, Adv. Synth. Catal. 351 (2009)
2491–2495.
[28] L.A. Parreira, L. Menini, J.C. da Cruz Santos, E.V. Gusevskaya, Adv. Synth. Catal.
352 (2010) 1533–1538.
[29] P.B. White, J.N. Jaworski, G.H. Zhu, S.S. Stahl, ACS Catal. 6 (2016) 3340–3348.
[30] M.S. Sigman, E.W. Werner, Acc. Chem. Res. 45 (2012) 874–884.
[31] A.B. Weinstein, S.S. Stahl, Angew. Chem. Int. Ed. 51 (2012) 11505–11509.
In summary, we have developed two novel liquid-phase pro-
cesses for the oxidation of ␣-bisabolol in aqueous methanol or
dimethylacetamide solutions using molecular oxygen as the sto-
ichiometric oxidant. The reactions are catalyzed solely by the
commercially available palladium salts, Pd(OAc)2 or PdCl2. ␣-

L.A. Parreira et al. / Applied Catalysis A: General 524 (2016) 126–133
133
[32] A.B. Weinstein, D.P. Schuman, Z.X. Tan, S.S. Stahl, Angew. Chem. Int. Ed. 52
[39] T. Mitsudome, T. Umetani, N. Nosaka, K. Mori, T. Mizugaki, K. Ebitani, K.
Kaneda, Ang. Chem. Int. Ed. 45 (2006) 481–485.
(2013) 11867–11870.
[33] J.E. Redford, R.I. McDonald, M.L. Rigsby, J.D. Wiensch, S.S. Stahl, Org. Lett. 14
(2012) 1242–1245.
[40] B. Solabannavar, V.B. Helavi, U.V. Desai, R.B. Mane, Tetrahedron Lett. 43
(2002) 4535–4536.
[34] L.A. Parreira, L. Menini, E.V. Gusevskaya, Catal. Sci. Technol. 4 (2014)
2016–2022.
[35] L.A. Parreira, A.M. da Cunha, L. Menini, E.V. Gusevskaya, RSC Adv. 5 (2015)
56987–56992.
[36] L.H.B. Baptistella, I.M.O. Sousa, Y. Gushikem, A.M. Aleixo, Tetrahedron Lett. 40
(1999) 2695–2698.
[41] C.N. Cornell, M.S. Sigman, Org. Lett. 8 (2006) 4117–4120.
[42] G. Zeni, R.C. Larock, Chem. Rev. 104 (2004) 2285–2309.
[43] J. Muzart, Tetrahedron 63 (2007) 7505–7521.
[44] R.I. McDonald, G. Liu, S.S. Stahl, Chem. Rev. 111 (2011) 2981–3019.
[45] R.M. Trend, Y.K. Ramtohul, B.M. Stoltz, J. Am. Chem. Soc. 127 (2005)
17778–17788.
[37] A.G.M. Silva, T.S. Rodrigues, A. Dias, H.V. Fajardo, R.F. Gonc¸ alves, M. Godinho,
P.A. Robles-Dutenhefner, Catal. Sci. Technol. 4 (2014) 2016–2022.
[38] T. Hosokawa, S. Miyagi, S.-I. Murahashi, A. Sonoda, J. Org. Chem. 43 (1978)
2752–2757.
[46] A. Wang, H. Jiang, J. Org. Chem. 75 (2010) 2321–2326.
[47] A. Wang, H. Jiang, J. Am. Chem. Soc. 130 (2008) 5030–5031.