Hydroformylation and one-pot hydroformylation/epoxy ring cleavage of limonene oxide: A sustainable access to biomass-based multi-functional fragrances

Article abstract of DOI:10.1016/j.apcata.2021.118082

Naturally occurring, renewable limonene oxide is converted into aldehydes with pleasant scent under oxo (hydroformylation) conditions. Good activity and selectivity are only reached by the careful choice of the catalytic system and the reaction conditions. [Rh(COD)(OMe)]₂ was used as the catalyst precursor and PPh₃ or (2,4-di-^tBuC₆H₃O)₃P as auxiliary ligands, the latter showing better results. The process can be carried in eco-friendly solvents such as anisole, ethanol, dimethylcarbonate (DMC) and diethylcarbonate (DEC). Employing Rh/(2,4-di-^tBuC₆H₃O)₃P catalytic system and acetic anhydride or acetic acid as co-reactants, a concomitant one-pot epoxide opening takes place along with hydroformylation, resulting in new acetylated aldehydes potentially useful for the fragrance industry. With the former co-reactant diacetylated products are preferred, while with the latter co-reactant hydroxy acetoxy aldehydes are formed in high yields. The one pot hydroformylation/epoxy ring cleavage process can be also performed in green solvents: DMC, DEC and anisole.

Full text of DOI:10.1016/j.apcata.2021.118082

Applied Catalysis A, General 616 (2021) 118082
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Hydroformylation and one-pot hydroformylation/epoxy ring cleavage of
limonene oxide: A sustainable access to biomass-based
multi-functional fragrances
Mileny P. de Oliveira, F ´a bio G. Delolo, Jesus A.A. Villarreal, Eduardo N. dos Santos*,
Elena V. Gusevskaya*
Departamento de Química, Universidade Federal de Minas Gerais, 31270-901, Belo Horizonte, MG, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords:
Naturally occurring, renewable limonene oxide is converted into aldehydes with pleasant scent under oxo
Epoxide ring opening
Green solvents
Limonene oxide
Oxo-process
(
hydroformylation) conditions. Good activity and selectivity are only reached by the careful choice of the cat-
alytic system and the reaction conditions. [Rh(COD)(OMe)] was used as the catalyst precursor and PPh or (2,4-
P as auxiliary ligands, the latter showing better results. The process can be carried in eco-friendly
solvents such as anisole, ethanol, dimethylcarbonate (DMC) and diethylcarbonate (DEC). Employing Rh/(2,4-
2
3
t
di- BuC
6
H
3
O)
3
Renewable feedstock
t
di- BuC
6
H
3
3
O) P catalytic system and acetic anhydride or acetic acid as co-reactants, a concomitant one-pot
epoxide opening takes place along with hydroformylation, resulting in new acetylated aldehydes potentially
useful for the fragrance industry. With the former co-reactant diacetylated products are preferred, while with the
latter co-reactant hydroxy acetoxy aldehydes are formed in high yields. The one pot hydroformylation/epoxy
ring cleavage process can be also performed in green solvents: DMC, DEC and anisole.
1
. Introduction
Terpenes are a widespread class of natural compounds which are
intermediate products [5,16–21].
For several years, our group has been interested in the hydro-
formylation of natural terpenic compounds, including both simple ter-
penes, e.g. limonene and β-pinene, and functionalized terpenoids, much
scarcely described as substrates for hydroformylation [4]. Currently, we
have focused our interest on the hydroformylation of terpene epoxides.
Limonene oxide is a naturally occurring compound extracted from citric
essential oils and also obtained by the epoxidation of limonene, one of
the most widespread terpenes in nature. Limonene is an important
platform molecule in chemical industry and is produced commercially
from waste orange peel [22]. We have found only one published report
on the limonene oxide hydroformylation: an US Shell patent, which
presents only one reaction involving limonene oxide, without the
specification of either reaction selectivity or product yield [23]. The
isolated aldehyde was described as a compound with a sweet green
fruity floral odor.
easily accessible from essential oils. Because of their unique olfactory
characteristics and biological activity, terpenic compounds traditionally
represent a valuable biorenewable hydrocarbon feedstock for pharma-
ceutical and flavor&fragrance industries [1–3]. Besides direct applica-
tions, chemical transformations of terpenic molecules allow to obtain
many value-added commercially interesting products [3–9]. Hydro-
formylation, the reaction of olefins with molecular hydrogen and carbon
monoxide, is a particularly useful strategy to upgrade terpenic sub-
strates. The reaction results in the incorporation of an aldehyde moiety,
which is often responsible for the pleasant smell of the compound.
Hydroformylation is used industrially to manufacture many commodity
and fine chemicals [4,9–15]. The range of potential products can be
expanded considerably if hydroformylation is integrated in multi-step
combinations, which include several tandem or parallel chemical
transformations in the same reactor. Such one-pot processes allow for
more sustainable synthesis of complex molecules as compared with
stepwise synthetic technics, avoiding separation and purification of
–
–
The limonene oxide molecule has a terminal C C bond and an
endocyclic oxirane functionality. The main challenge in its hydro-
formylation appears to be a selectivity control because of the undesired
–
facile isomerization of the C C bond or epoxide ring-opening
–
*
Corresponding authors.
E-mail addresses: nicolau@ufmg.br (E.N. dos Santos), elena@ufmg.br (E.V. Gusevskaya).
https://doi.org/10.1016/j.apcata.2021.118082
Received 8 January 2021; Received in revised form 1 March 2021; Accepted 4 March 2021
Available online 9 March 2021
0
926-860X/© 2021 Elsevier B.V. All rights reserved.

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
rearrangements which usually give a mixture of isomeric compounds
ethyl acetate in hexane (Rf = 0.31) for aldehyde 2 and 35 % ethyl ac-
etate in hexane (Rf = 0.36) for acetylated aldehydes 3. Isolated yields
usually were 90–95 % from the GC yields.
[
24]. However, the controlled modification of the oxirane group opens
an opportunity to create additional functionalities in the limonene oxide
molecule, parallelly with the hydroformylation of its double bond.
Acetate esters also are an important group of ingredients for flavors
and fragrances because of their sweet fruit aroma [25]. Poly-
functionalized compounds containing the aldehyde and the acetate
moieties are especially valuable because of the simultaneous presence of
two types of olfactory functions which account for both top (fruits) and
heart (floral) notes in fragrance formulations.
Compound 2 (a mixture of four diastereoisomers not separable by
+
+
GC): m/z (%): 164 (4) [M–H
2
O] , 139 (36) [M–CH
2
CHO] , 138 (32),
125 (20), 124 (25), 123 (20), 121 (21), 111 (100), 109 (60), 108 (21),
107 (31), 97 (33), 95 (50), 94 (31), 93 (68), 91 (30), 85 (21), 84 (39), 83
(39), 81 (53), 79 (50), 77 (23), 71 (57), 69 (59), 68 (29), 67 (58), 55
1
◦
(93); H NMR (400 MHz, CDCl
3
, 25 C, Me
4
Si): δ = 0.81, 0.82, 0.83 and
3
9
7
0.84 (four doublets, J = 3.3 Hz, 3 H; C H
3 3
), 1.23 (s, 3 H; C H ),
5
3
4
5
The present work reports a first systematic study of the limonene
oxide hydroformylation and a novel one-pot process, which includes the
hydroformylation and parallel cleavage of the oxirane ring in the
limonene oxide molecule to give acetoxy derivatives of the corre-
sponding diol. The method allows for the functionalization of limonene
oxide simultaneously at two remote reactive sites to approach novel
products with multiple functionalities, which have new olfactory char-
acteristics. It is noteworthy that most of previously reported multi-step
processes based on hydroformylation include first the aldehyde forma-
tion and then sequential transformations of the aldehyde group in situ to
1.05–1.25 (m, 1 H; C HH), 1.30–1.60 (m, 3 H; C HH, C H and C HH),
3 6 8 10
1.65–2.05 (m, 4 H; C HH, C H and C H), 2.10–2.20 (m, 1 H; C HH),
2
3
10
2.35 and 2.39 (two triplets, J = 5.0 Hz, 1 H; C HH), 2.91 and 2.97 (d,
3
2
3
J = 5.3 Hz and br.s, respectively, 1 H; C H), 9.67 ppm (t, J = 1.2 Hz,
1 H; C11HO); C NMR (100 MHz, CDCl
13
, 25 C, TMS): δ = 16.37, 16.66,
◦
3
9
5
16.76 and 16.78 (C ), 21.71, 22.85, 24.26 and 25.41 (C ), 22.98 and
7
3
24.35 (C ), 26.80, 28.01, 29.15 and 29.45 (C ), 28.49, 29.03, 30.66 and
6
8
30.72 (C ), 31.55, 31.67, 31.91 and 32.05 (C ), 33.90, 37.49 and 37.63
4
10
1
(C ), 48.00, 48.18, 48.41 and 48.50 (C ), 57.54 and 57.78 (C ), 59.13,
2
11
59.20, 60.66 and 60.70 (C ), 202.69 ppm (C HO). Numbering of atoms:
give other products [5,16–20]. To our knowledge, the only report of
Scheme 1. Compound described in [23].
–
–
C
C hydroformylation with controlled simultaneous functionalization
Compound 3a: (a ≈1/1 mixture of two diastereoisomers, (4R8S) and
+
at a remote reactive site was presented in our recent work, which de-
scribes the hydroformylation of propenylbenzenes along with the ace-
toxylation of their phenolic groups [21].
(4R8R), not separable by GC): m/z (%): 164 (1) [M–2H
2
O] , 139 (6),
138 (9), 123 (7), 121 (9), 112 (10), 111 (23), 109 (14), 108 (6), 107
(12), 97 (16), 95 (18), 94 (9), 93 (22), 91 (11), 85 (7), 84 (9), 83 (15), 81
(
22), 80 (6), 79 (14), 77 (6), 71 (100), 70 (7), 69 (23), 68 (10), 67 (20),
1
◦
2
. Experimental
58 (19), 57 (8), 56 (8), 55 (28). H NMR (400 MHz, CDCl , 25 C,
3
3
9
Me
4
Si): δ = 0.89 and 0.91 (two doublets, J = 8.0 Hz, 3 H; C H
3
), 1.21
7
3
5
6
(
+)-Limonene oxide [(4R)-4-isopropenyl-1-methyl-1-cyclohexene
,2-epoxide, a mixture of cis and trans isomers in a ratio of 43/57, 97
], triphenylphosphine (PPh , 99 %), tris(2,4-di-tert-butylphenyl)
P, 98 %], acetic anhydride (99.5 %), p-
toluenesulfonic acid (monohydrate, 98 %) and p-xylene (99 %) were
from Sigma-Aldrich. The synthesis of [Rh(COD)(OMe)]
(COD = 1,5-
(s, 3 H; C H
3
), 1.30–1.60 (m, 6 H; C H
2 2 2
, C H and C H
), 1.60–1.70 (m,
4
8
10
1
1 H; C H), 1.90–2.00 (m, 1 H; C H), 2.10–2.25 (m, 1 H; C HH),
1
0
2
%
3
2.35–2.50 (m, 1 H; C HH), 3.57 (br.s, 1 H; C H), 9.70 and 9,71 ppm
t
(two triplets, 3J = 2.0 Hz, 1 H; C HO); C NMR (100 MHz, CDCl
11 13
phosphite [(2,4-di- BuC
6
H
3
O)
3
3
,
◦
9
25 C, TMS): δ = 17.00 and 17.21 (C ), 21.47 (OCCH
3
), 23.51 and 24.61
5
7
3
8
6
(C ), 26.88 (C ), 31.73 and 32.79 (C ), 31.88 (C ), 33.44 and 33.53 (C ),
2
4
10
1
2
cyclooctadiene) was performed accordingly to a well-established pro-
cedure [26]. Anisole (anhydrous, 99.7 %), toluene (anhydrous, 99.8 %)
and dimethylcarbonate (DMC) (anhydrous, 99.8 %) were acquired in a
Sure/Seal™ bottle (Sigma-Aldrich) and kept in a glove box. Ethanol was
distilled under argon after refluxing for 6 h over magnesium turnings
and solid iodine. Diethylcarbonate (DEC) (Sigma-Aldrich) was distilled
and kept under argon over 4 Å molecular sieves.
35.04 (C ), 48.52 and 48.74 (C ), 71.47 (C ), 73.70 and 73.77 (C ),
203.61 and 203.64 ppm (C11HO). Numbering of atoms: Scheme 1.
Compound 3b (a ≈1/1 mixture of two diastereoisomers, (4R8S) and
+
(4R8R), not separable by GC): m/z (%): 182 (2) [M–HOAc] , 164 (11)
+
[M–HOAc–H
2
O] , 139 (23), 138 (26), 123 (27), 121 (31), 112 (25), 111
(60), 109 (33), 108 (21), 107 (31), 105 (22), 97 (31), 95 (55), 94 (23),
93 (61), 84 (49), 83 (25), 81 (36), 79 (34), 71 (100), 69 (62), 68 (27), 67
(40), 58 (21), 55 (58); 1H NMR (400 MHz, CDCl
, 25 C, Me
◦
Si):
3
), 1.14 (s, 3 H;
The hydroformylation was carried in a 100 mL homemade stainless-
steel high pressure reactor with magnetic stirring and a valved dip tube
for sampling under pressure. The products analysis was performed by
gas chromatography (GC) (Shimadzu GC2010, Rtx®-5MS capillary
column, FID detector, internal standard: p-xylene). Calculations of
conversions and selectivities were based on the limonene oxide con-
verted. Average turnover frequencies (TOFs, mol of limonene oxide
consumed per mol of rhodium per hour) were calculated either up to ca.
3
4
3
9
δ = 0.91 and 0.92 (two doublets, J = 4.0 Hz, 3 H; C H
7
5
3
4
C H
3
), 1.35–1.50 (m, 2 H; C H
2
), 1.50–1.75 (m, 5 H; C H
2
, C H and
6
8
C H
2
), 1.95–2.05 (m, 1 H; C H), 2.04 (s, 3 H; OCOCH
3
), 2.15–2.25 (m,
1
0
10
2
1 H; C HH), 2.35–2.50 (m, 1 H; C HH), 4.76 (br.s, 1 H; C H),
9.72 ppm (br. s, 1 H; C HO); C NMR (100 MHz, CDCl
1
1
13
◦
3
, 25 C, TMS):
9
5
δ = 16.77 and 17.09 (C ), 21.41 (OCCH
3
), 23.17 and 24.43 (C ), 27.22
7
3
8
6
(C ), 28.87 and 29.92 (C ), 32.06 and 32.12 (C ), 34.29 and 34.39 (C ),
4
10
1
4
0–50 % conversion or for the first reaction hour. Typically, the reactor
was charged under argon with a solution (20.0 mL) containing limonene
oxide (4 mmol), [Rh(COD)(OMe)] (5.0
mol), P-ligand (0–0.20 mmol),
acylating agent (if any) and p-xylene (2 mmol, GC internal standard) in
the indicated solvent. After the pressurization with a CO/H mixture (1/
35.57 and 35.61 (C ), 48.43 and 48.60 (C ), 70.06 (C ), 75.42 and
2
3
75.49 (C ), 170.47 and 170.50 (OCCH ), 201.07 and 203.10 ppm
1
1
2
μ
(C HO). Numbering of atoms: Scheme 1.
Compound 3c: (a ≈1/1 mixture of two diastereoisomers, (4R8S) and
+
(4R8R), not separable by GC): m/z (%): 164 (10) [M–HOAc–H
2
O] , 139
2
1
) to 40–80 at m, the reactor was placed in a previously heated
(15), 138 (64), 123 (23), 121 (28), 112 (14), 111 (57), 109 (100), 108
(34), 107 (42), 105 (25), 97 (25), 95 (50), 94 (32), 93 (65), 84 (15), 83
(24), 81 (67), 79 (47), 77 (21), 71 (47), 69 (46), 68 (15), 67 (48), 55
aluminum block with thermostat control and magnetic stirring was
started.
1
◦
The isolation of the main products was performed by column chro-
matography (silica gel 60, eluents: hexane and ethyl acetate). The
product identification was performed by GC–MS (Shimadzu QP2010-
(78). H NMR (400 MHz, CDCl
3
, 25 C, Me
4
Si): δ = 0.85 and 0.86 (two
3
9
7
doublets, J = 8.0 Hz, 3 H; C H
3
), 1.45 (s, 3 H; C H
3
), 1.35–1.75 (m,
3
4
5
6
8
7 H; C H
OCOCH
), 2.10–2.20 (m, 1 H; C HH), 2.35–2.45 (m, 1 H; C HH), 3.98
(br.s, 1 H; C H), 9.67 ppm (br. s, 1 H; C HO); C NMR (100 MHz,
2
, C H, C H
2
and C H
2
), 1.90–2.00 (m, 1 H; C H), 1.94 (s, 3 H;
10 10
PLUS, 70 eV) and NMR spectroscopy (Bruker 400 MHz, CDCl
To determine isolated yields, the reaction mixture was washed with
saturated aqueous solution of NaHCO (4x) and with brine. The organic
phase was dried with anhydrous MgSO , filtered, and concentrated in
3
, TMS).
3
2
11
13
◦
9
7
CDCl
3
, 25 C, TMS): δ = 16.93 and 17.17 (C ), 21.79 (C ), 22.43
3
5
3
(OCCH ), 23.49 and 24.60 (C ), 30.26 and 30.35 (C ), 31.60 and 32.64
3
4
6
8
4
10
vacuum. Then, the desired product was isolated by silica gel column
chromatography using (ethyl acetate/hexane) mixtures as eluents: 15 %
(C ), 32.02 and 32.05 (C ), 34.61 and 34.63 (C ), 48.54 and 48.77 (C ),
2
1
3
70.31 and 70.38 (C ), 82.99 (C ), 170.45 (OCCH ), 203.31 and
2

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
Scheme 1. Hydroformylation and epoxy ring cleavage of limonene oxide (1). The best GC conversion/selectivity values are presented.
2
03.35 ppm (C11HO). Numbering of atoms: Scheme 1.
and 202.69 ppm (C11HO); minor isomer (longer GC retention time): m/z
+
+
Compound 3d: major isomer (a ≈1/1 mixture of two di-
(%): 182 (51) [M–Ac
2
O] , 164 (48) [M–Ac
2
O–H
2
O] , 145 (31), 138
astereoisomers, (4R8S) and (4R8R), not separable by GC): m/z (%): 182
(41), 131 (31), 123 (23), 121 (64), 120 (72), 119 (33), 111 (40), 109
(82), 108 (43), 107 (46), 106 (39), 105 (74), 97 (24), 95 (50), 94 (40),
93 (100), 92 (29), 91 (70), 84 (22), 81 (45), 79 (51), 77 (31), 71 (46), 69
(40), 67 (36), 55 (46). Numbering of atoms: Scheme 1.
+
+
2
O] , 138 (59), 131 (25), 123
(
(
(
(
(
82) [M–Ac
2
O] , 164 (52) [M–Ac
2
O–H
19), 121 (55), 120 (72), 119 (28), 111 (44), 109 (100), 108 (57), 107
33), 106 (25), 105 (77), 97 (19), 95 (37), 94 (35), 93 (86), 92 (22), 91
50), 84 (19), 81 (30), 79 (37), 77 (26), 71 (35), 69 (31), 67 (27), 55
1
◦
38); H NMR (400 MHz, CDCl
3
, 25 C, Me
4
Si): δ = 0.88 and 0.89 (two
3. Results and discussion
3
9
5
doublets, J = 4.0 Hz, 3 H; C H ), 1.10–1.20 (m, 1 H; C HH), 1.38 (s,
3
7
3 4 5 6
3
H; C H
3
), 1.40–1.60 (m, 4 H; C HH, C H, C HH and C HH), 1.65–1.70
3.1. Hydroformylation of limonene oxide
3
8
(
m, 1 H; C HH), 1.95–2.00 (m, 1 H; C H), 1.98 (s, 3 H; OCOCH
3
), 2.04
1
0
(
s, 3 H; OCOCH
3
), 2.15–2.22 (m, 1 H; C HH), 2.30–2.45 (m, 2 H;
Commercial (+)-limonene oxide used as a substrate was a 43:57
mixture of cis/trans isomers with a different position of the methyl and
isopropenyl groups relatively to the carbon ring (Schemes 1 and 2). Both
cis and trans isomers of 1 have R-configuration at asymmetric carbon C-
4, as both are originated from natural (R)-(+)-limonene.
1
0
6
2
C
HH and C HH), 5.11 (br.s, 1 H; C H), 9.70 and 9.71 ppm (two trip-
3
11
13
◦
lets, J = 2 Hz, 1 H; C HO); C NMR (100 MHz, CDCl
3
, 25 C, TMS):
9
7
δ = 16.64 and 16.97 (C ), 21.24 and 21.25 (OCCH
3
), 21.81 (C ), 22.26
3
5
(
3
OCCH ), 23.11 and 24.43 (C ), 28.65 and 29.74 (C ), 30.67 and 30.77
6
8
4
10
(
C ), 32.02 and 32.09 (C ), 35.13 and 35.17 (C ), 48.36 and 48.55 (C ),
The hydroformylation of limonene oxide occurred with high regio-
selectivity to give exclusively the terminal aldehyde 2 (Scheme 1).
2
1
7
3
2.45 and 72.53 (C ), 80.75 (C ), 169.74 and 169.97 (OCCH ), 202.68
Scheme 2. Schematic presentation of the limonene oxide (1) hydroformylation/epoxy ring cleavage. The best GC conversion/selectivity values are presented.
3

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
t
Although the aldehyde products appeared as a single GC peak under the
employed conditions, NMR analysis showed a mixture of four di-
with (2,4-di- BuC
6
H
3
O)
3
P was much faster than in that with triphenyl-
phosphine; furthermore, selectivity for 2 was 88–91 % at 92–94 %
conversions (Table 1, runs 6 and 7 vs. run 5).
1
astereoisomers in a ratio of ca. 4:4:6:6 with subtle differences in H and
1
3
C resonances. Considering that a new asymmetric center (C-8) was
Bulky phosphites are known to impact positively the hydro-
formylation of certain olefins due to special electronic and steric pa-
rameters [10,29,30]. Their large cone angles disfavor the entrance of a
second phosphite ligand into the coordination sphere of rhodium and, in
spite of the high P/Rh atomic ratio, Rh species with only one P(III)
ligand prevails, giving enough room to coordinate the substrate. In
generated at the formation of the aldehyde and that a mixture of (4R) cis
and (4R) trans isomers was employed as the starting material, four iso-
mers of aldehyde 2 having the methyl and isopropenyl groups in cis-trans
positions and different configurations at C-8 were indeed expected. The
proportion of the products derived from cis- and trans-limonene oxide
was close to that in the starting limonene oxide (ca. 4:6, respectively),
with the (8R) and (8S) diastereoisomers in both pairs being formed in
nearly equal amounts.
addition, due to their -acceptor character, phosphites make the Rh
π
center less rich in electron density as compared to their phosphine
counterparts, which also favors the substrate coordination. Thus, raising
t
The presence of the oxirane functional group makes the limonene
oxide molecule highly sensitive to rearrangements [24,27,28]. In fact,
the concomitant modifications involving the oxirane group showed to be
a serious complicating factor in limonene oxide hydroformylation. In the
run without phosphorous ligands (non-promoted system), a plethora of
isomerization products (mainly dihydrocarvone) was observed and the
selectivity for aldehyde 2 was only 65–70 % (Table 1, run 1). Decreasing
the (2,4-di- BuC
6
H
3
O)
3
P concentration (P/Rh from 10 to 20) or the total
gas pressure (from 40 to 80 atm) did not significantly influence the rate
of the limonene oxide conversion or the aldehyde selectivity (ca. 90 %
conversion and ca. 90 % selectivity in 6 h, Table 1, runs 7 and 8 vs. run
6).
Aiming to improve the sustainability of the process, we have tested
several other solvents considered green in modern guides for solvent
selection [31,32]. Representative results on the limonene oxide hydro-
formylation under optimized conditions in various reaction media are
collected in Table 2. The reaction in ethanol was slower than that in
toluene but resulted in an excellent selectivity for aldehyde 2 (95 %,
Table 2, cf. runs 1 and 2). The corresponding acetals were not formed in
appreciable amounts under the relatively low temperature applied.
Biodegradable and low toxicity anisole, dimethylcarbonate (DMC)
and diethylcarbonate (DEC) are promising alternatives as eco-friendly
solvents in organic reactions. These compounds appear in modern sol-
vent guides with high sustainability ranking, close to those of water and
ethanol [31,32]. The excellent potential of DEC, DMC and anisole as
reaction media for hydroformylation was shown in our recent works
[33–36].
◦
temperature from 100 to 80 C resulted in better selectivity; however,
ca. 20 % of limonene oxide was still rearranged into isomerization
products and the corresponding aldehydes originated from these iso-
mers (Table 1, run 2).
A substantial improvement in the yield of aldehyde 2 was attained by
3
adding triphenylphosphine (PPh ) as auxiliary ligand, which signifi-
cantly suppressed undesired isomerization reactions (Table 1, runs 3–5).
Raising the triphenylphosphine concentration led to a gradual increase
in the yield of aldehyde 2; however, with a simultaneous drop in the rate
of the conversion of limonene oxide probably due to the formation of bis
3
and tris PPh -coordinated rhodium centers, which are hindered and
make it difficult the coordination of the considerably bulky limonene
oxide. At P/Rh = 10, the reaction gave aldehyde 2 almost quantita-
tively; however, the conversion was relatively slow, reaching only 80 %
DMC showed excellent results as solvent in the limonene oxide
hydroformylation. The reaction rate and selectivity for 2 were compa-
rable with those obtained in toluene (Table 2, run 3 vs. run 1). In anisole
and DEC, the rate of the limonene oxide conversion was approximately
the same (DEC) or even higher (anisole) than in toluene; however, the
within 24 h (Table 1, run 5).
t
Differently from triphenylphosphine, (2,4-di- BuC
6
3
H O)
3
P did not
decelerate the reaction; however, improved selectivity for aldehyde 2
Table 1 run 6 vs. run 2). The conversion of limonene oxide in the system
(
Table 1
Hydroformylation of limonene oxide (1) in toluene .
a
b
Run
Ligand
P/Rh)
TOF
Time
(h)
C
Selectivity (%)
Isomers
ꢀ
1
(
(h
)
(%)
Hydroformylation
c
of 1
2
other aldehydes
d
1
6
1
6
1
6
2
64
87
43
91
38
93
23
83
13
80
77
94
70
92
73
90
35
29
15
13
16
10
1
65
70
85
82
84
80
99
91
99
99
93
88
95
91
90
89
0
1
0
5
0
10
0
7
0
0
0
8
0
5
4
7
1
none
none
256
172
2
3
4
PPh
3
3
(2)
(5)
152
40
PPh
2
2
2
2
6
2
6
2
6
4
2
1
5
PPh
3
(10)
28
4
1
e
t
6
(2,4-di- BuC
6
H
3
O)
3
P
176
7
(
10)
4
t
5
7
(2,4-di- BuC
6
H
H
3
O)
O)
3
P (20)
P
160
156
4
f
t
8
(2,4-di- BuC
6
3
3
6
(
10)
4
a
◦
Reaction conditions: [limonene oxide] – 0.20 M (4 mmol), [Rh(COD)(OMe)]
2
– 0.25 mM (5
μ
mol), 80 C, 40 atm (CO/H
2
= 1/1), solvent – toluene, total volume –
2
0 mL. The values for selectivity and conversion (C) were based on the limonene oxide reacted.
b
Average turnover frequency (mol of limonene oxide consumed per mol of Rh per hour) up to ca. 40–50 % conversion; only for rough comparison because the
reactions in most cases showed high conversions already at first sampling.
c
Mainly dihydrocarvone.
d
e
f
◦
1
00 C.
Isolated yield of 2 – 79 %.
0 atm (CO/H
= 1/1).
8
2
4

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
Table 2
3.2. Hydroformylation/epoxy ring cleavage of limonene oxide
a
Hydroformylation of limonene oxide (1) in different solvents .
b
Aiming to perform the modification of the limonene oxide molecule
in a single procedure at two different sites in a parallel way, we have
combined the hydroformylation reaction with a simultaneous cleavage
of the epoxy ring (Table 3). First, we compared the reactions in pure
Run
Solvent
TOF
Time
(h)
C
Selectivity (%)
ꢀ
1
(
h
)
(%)
Isomers
Hydroformylation
c
of 1
2
other aldehydes
2
6
2
6
2
6
2
6
2
4
77
94
58
85
80
92
84
99
88
95
7
4
4
3
6
4
8
5
8
6
93
88
96
95
88
90
70
72
81
82
0
toluene and in toluene containing acetic anhydride (Ac
acylating agent (5 equivalents relatively to limonene oxide, which cor-
responds to 10 vol.%, Table 3, runs 1 and 2). Fortunately, Ac O did not
2
O) used as the
1
2
Toluene
Ethanol
176
120
8
0
2
2
affect the hydroformylation rate significantly. During the first hour,
nearly 45 % of limonene oxide was converted in both runs giving mainly
aldehyde 2, while the selectivity remained only slightly lower in the
presence of acetic anhydride. After 24 h, the amounts of side products
6
3
4
DMC
DEC
200
180
6
22
23
10
12
5
a
Anisole
240
were in the range of 10–15 % in both systems. In the presence of Ac
2
O,
aldehyde 2 was slowly converted to several products with higher boiling
points, which were identified as diol derivatives 3a, 3b, 3c and 3d
Reaction conditions: [limonene oxide]
–
0.20 M (4 mmol), [Rh(COD)
t
P (P/Rh = 10), 80 ◦C,
(
OMe)]
2
– 0.25 mM (5
μ
mol), ligand – (2,4-di- BuC
6 3 3
H O)
(
Schemes 1 and 2), with structures confirmed by MS and NMR
spectroscopy.
In run 2, the diol derivatives 3 were formed in 37 % combined yield,
4
0 atm (CO/H
2
= 1/1), total volume – 20 mL. The values for selectivity and
conversion (C) were based on the limonene oxide reacted. DEC: diethylcar-
bonate, DMC: dimethylcarbonate.
b
with nearly 50 % of aldehyde 2 remaining unconverted as the rate of the
ring opening was relatively low. Under the conditions used in this run,
the average rate of the hydroformylation of limonene oxide was much
higher (47 % conversion in 1 h) than the average rate of the ring
cleavage step (37 % conversion for diol derivatives 3 in 24 h).
Average turnover frequency (mol of limonene oxide consumed per mol of Rh
per hour) up to ca. 40–50 % conversion; only for rough comparison because the
reactions in most cases showed high conversions already at first sampling.
c
Mainly dihydrocarvone.
◦
Then, the reaction was run at higher temperature (100 C) in order to
isomerization reactions were more pronounced (Table 2, runs 4 and 5).
Nevertheless, aldehyde 2 could be obtained in 70–80 % individual yield
in DEC and anisole solutions, with the total yield for the aldehydes being
nearly 90 % at nearly complete substrate conversions. It is worthwhile
mentioning that the mixture of the aldehydes obtained from limonene
oxide hydroformylation has a pleasant scent and, in this regard, might
be useful for fragrance formulations even without separation.
promote faster cleavage of the oxirane ring (Table 3, run 3). In 48 h,
aldehyde 2 was essentially converted into diol derivatives 3, with their
combined yield reaching 87 %. The main products 3b and 3d were
formed with ca. 40 % selectivity each. Compound 3b is a 1,2-hydroxya-
cetate with tertiary hydroxyl (C-1) and secondary acetoxy (C-2) groups
and compound 3d is a 1,2-diacetate (Schemes 1 and 2). All these
products 3a, 3b, 3c and 3d were formed as a nearly 1:1 mixture of their
diastereoisomers, which were not resolved by GC under the employed
analysis conditions. It is possible to infer that the isomers in each pair
Table 3
a
Hydroformylation/epoxy ring cleavage of limonene oxide (1) .
Run
Acylating
agent
Solvent
Time
(h)
C
Selectivity (%)
(%)
Isomers
Aldehydes
Diol derivatives
b
(
mmol)
of 1
2
others
3a
3b
3c
3d
3
(total)
c
1
44
4
2
5
2
4
2
0
7
0
9
0
9
0
8
0
9
0
7
0
6
0
96
85
91
52
96
22
6
0
0
0
0
3
0
4
4
0
8
0
9
0
5
3
2
2
9
2
4
2
5
0
0
0
0
0
0
1
none
Toluene
Toluene
2
1
2
1
4
94
13
4
0
0
0
c
47
0
0
0
2
Ac
2
2
O (20)
O (20)
4
97
9
15
0
11
0
8
37
0
92
0
0
3
4
Ac
Toluene
24
>99
>99
89
5
31
35
0
6
28
41
0
69
82
0
48
1
48
1
24
1
24
1
48
1
48
1
48
1
48
7
2
88
3
5
0
Ac
2
2
O (40)
O (20)
Toluene
Toluene
Toluene
Toluene
DMC
>99
93
8
24
0
0
45
0
77
0
d
66
0
12
20
15
4
0
5
Ac
>99
94
10
2
0
41
0
60
4
d
60
0
2
6
HOAc (20)
HOAc (20)
HOAc (20)
HOAc (20)
HOAc (20)
>99
94
27
2
5
0
37
7
e
70
4
15
2
2
0
7
>99
96
44
1
34
1
0
80
4
62
13
55
9
20
14
28
15
17
7
0
8
9
>99
97
30
1
20
1
2
61
4
0
DEC
>99
97
27
2
24
4
1
56
8
68
5
0
1
a
0
Anisole
>99
40
28
2
75
t
P (P/Rh = 10), 100 ◦C, 40 atm
Reaction conditions: [limonene oxide] – 0.20 M (4 mmol), [Rh(COD)(OMe)]
2
– 0.25 mM (5
μ
mol), ligand – (2,4-di- BuC
6 3 3
H O)
(
CO/H
2
= 1/1), total volume – 20 mL. The values for selectivity and conversion (C) were based on the limonene oxide reacted. DEC: diethylcarbonate, DMC:
dimethylcarbonate.
b
Mainly dihydrocarvone.
c
d
e
◦
8
0
C.
2
mol% of p-toluenesulfonic acid was added.
Isolated yield of 3 – 71 %.
5

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
have different configurations at C-8 as they were derived from the 1:1
mixture of (8R) and (8S) isomers of aldehyde 2. Product 3c is a
regioisomer of monoacetate 3b with the reverse positions of hydroxyl
2
converted into diacetate 3d in the presence of larger amounts of Ac O
(3b/3d ≈ 1/2, Table 3, run 4). In this run, 45 % selectivity for diacetate
3d was observed at nearly complete substrate conversion, which is the
best result obtained for this product in the present work. As the rate of
the ring-opening step was much lower than that of the hydroformylation
step, we tried to use an additional acidic catalyst. Really, in the presence
of the catalytic amounts of p-toluenesulfonic acid (PTSA, 2 mol%), the
cleavage of the epoxy ring occurred faster; however, the selectivity
decreased significantly (Table 3, run 5). The drop in selectivity was due
to the increased isomerization of limonene oxide and also due to the
formation of high-boiling products not detectable by GC, both processes
being promoted by PTSA.
(
secondary) (C-2) and acetoxy (tertiary) (C-1) groups (Schemes 1 and 2).
Diol 3a was formed only in low concentrations, probably, by the inter-
action of the epoxide with residual water or by the reaction of mono-
acetates 3b and 3c with acetic acid (which appeared in the reaction
solutions as described whereafter) to give 3a and acetic anhydride.
At early reaction stages, monoacetates 3b and 3c were detected in
comparable concentrations (Table 3, run 3; Fig. 1). Then, the amounts of
3
b (tertiary alcohol) continued raising, whereas the amounts of 3c
(
secondary alcohol) gradually decreased with simultaneous increase in
the concentration of diacetate 3d. These results are fully consistent with
higher resistance of tertiary alcohols to acetylation with acetic anhy-
dride as compared to secondary alcohols. As can be seen in Fig. 1, the
average rate of the hydroformylation step in this run was also nearly 30
times higher (92 % conversion in 1 h) than the average rate of the ring
cleavage step (69 % conversion for diol derivatives 3 in 24 h).
When acetic anhydride was replaced by acetic acid (with the same
amounts of PTSA), monoacetate 3b became a predominant ring-opening
product with no diacetate 3d being formed at all (Table 3, cf. runs 6 and
5). Several unidentified and high-boiling products (not detectable by
GC) were formed in this reaction so that the overall selectivity for
aldehyde 2 and corresponding diol derivatives was low. These results
provide an additional argument for the suggestion that monoacetate 3b
with a tertiary OH group is formed from trans-limonene oxide, whereas
monoacetate 3c with a secondary OH group from cis-limonene oxide, as
explained whereafter. cis-Limonene oxide is known to be much more
susceptible to isomerization reactions than trans-limonene oxide
because of steric reasons and, therefore, it reacts much faster under
acidic conditions; this feature has been even used to develop a strategy
for a kinetic resolution of trans- and cis-isomers [28,40]. Thus, in the
presence of acetic acid and PTSA in run 6 (Table 3), cis-limonene oxide
and cis-aldehyde 2 suffered more hardly from undesirable isomerization
rearrangements than trans-limonene oxide and trans-aldehyde 2. For this
reason, monoacetate 3b originated by trans-limonene oxide became the
main ring-opening product, whereas monoacetate 3c originated from
cis-limonene oxide, appeared only in small or even trace amounts. As
Related studies on the interaction of cis- and trans-limonene oxide
with acetic acid revealed that the formation of main products followed
the Fürst-Plattner rule of the diaxial opening of the epoxy ring [37,38].
In these studies, the cis-isomer gave exclusively a 1,2-hydroxyacetate
with tertiary OAc (at C-1) and secondary OH (at C-2) groups, whereas
the trans-isomer gave the reversed product with secondary OAc (at C-2)
and tertiary OH (at C-1) groups. All the products contain the OH and
OAc groups in a relative trans configuration, with the methyl and iso-
propenyl fragments being also in mutually trans positions. In a more
recent work, it has been also reported that trans-limonene oxide reacted
with secondary amines resulting exclusively in amino alcohols with
tertiary OH groups in trans position to the amine moiety [39]. Consid-
ering these data, we suppose that monoacetate 3b with a tertiary OH
group is formed from trans-limonene oxide, whereas monoacetate 3c
with a secondary OH group from cis-limonene oxide as depicted in
Scheme 2. As expected, secondary alcohol 3c reacted more promptly
with acetic anhydride to give diacetate 3d, whereas tertiary alcohol 3b
remains essentially intact. For this reason, in all the samples of the re-
2
expected, acetic acid as the acylating agent was not as efficient as Ac O
because it was not able to promote the acetylation of monoacetates to
give diacetate 3d (Table 3, cf. runs 6 and 5).
As displayed in Scheme 2, the formation of the diol monoacetates 3b
and 3c from the epoxide requires the availability of acetic acid in stoi-
chiometric amounts. Therefore, diol monoacetates should not be ex-
2
actions performed with Ac O, monoacetate 3c was found only in low
concentrations as a minor product.
The reaction performed with higher concentration of Ac
2
O (10
pected in significant concentrations when Ac
acylating agent. Instead, a stoichiometric reaction of the epoxide group
with Ac O should result mainly in diacetate. In the runs when Ac O was
2
O alone is employed as the
equiv.) also gave mainly monoacetate 3b and diacetate 3d (Table 3, run
). The fraction of diacetate 3d in the reaction products was higher
4
2
2
because not only monoacetate 3c but also monoacetate 3b was partially
used as the acylating agent, the GC analysis of the reaction mixture
before the pressurization of the autoclave with “syngas” expectedly
revealed the peak of Ac O and no peak which could be assigned to acetic
2
acid. Nevertheless, starting from the first sampling (usually after 1 h),
GC analyses indicated that acetic acid was present in the reaction media.
2
As anhydrous toluene and 99.5 % purity Ac O were employed in the
reactions, a significant amount of acetic acid that appeared in the re-
action solutions could not come from the reaction of residual water with
Ac
2
O.
A plausible explanation for the formation of acetic acid could be the
hydrogenolysis of Ac O (Scheme 3). In other words, we suppose that the
source of hydrogen atoms to supply the formation of monoacetate 3b is
O with H is a well-
2
H
2
from the “syngas”. The hydrogenolysis of Ac
2
2
known process, which is usually catalyzed by transition metals,
including rhodium [41–43]. For example, it was reported the production
of acetic acid and ethylidene diacetate from Ac
2
O and H
2
using a
◦
RhCl
3
/phosphine catalytic system at 100 C and 70 atm of the CO/H
2
gas phase, i.e, under hydroformylation conditions [44]. Ethylidene
diacetate formally results from the reaction between acetic anhydride
and acetaldehyde.
2
To verify the hypothesis of the partial hydrogenolysis of Ac O and
Fig. 1. Product distribution in the course of the hydroformylation/epoxy ring
cleavage of limonene oxide (1) under the conditions of run 3 in Table 3:
the role of rhodium in the process, two experiments without limonene
oxide were carried out using the conditions of run 3 in Table 3: with and
[
limonene oxide] – 0.20 M (4 mmol), [Rh(COD)(OMe)]
2
– 0.25 mM (5
P (P/Rh = 10), 100 C, 40 atm (CO/H = 1/1),
O – 2 mL (20 mmol), solvent – toluene, total volume – 20 mL.
μmol),
t
◦
2
without the rhodium complex. Ac O was stable in toluene solution
ligand – (2,4-di- BuC
Ac
6
H
3
O)
3
2
(0.10 M) for at least 6 h without the rhodium complex [(2,4-
2
6

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
Scheme 3. Rhodium catalyzed hydrogenolysis of acetic anhydride.
t
◦
di- BuC
6
H
3
O)
3
P – 5 mM, 100 C, gas phase – 40 atm (CO/H
2
= 1/1)].
CRediT authorship contribution statement
However, in the run under the same conditions but with [Rh(COD)
OMe)] O were converted
(
2
(0.25 mM, P/Rh = 10), ca. 30 % of Ac
2
Mileny P. de Oliveira: Investigation, Writing - original draft. F a´ bio
G. Delolo: Conceptualization, Methodology, Investigation, Validation.
Jesus A.A. Villarreal: Methodology, Investigation. Eduardo N. dos
Santos: Conceptualization, Project administration, Writing - review &
editing. Elena V. Gusevskaya: Conceptualization, Supervision, Writing
- review & editing.
within 2 h into acetic acid. Thus, we suggest that the formation of
monoacetates 3b and 3c was supplied by acetic acid originated from
2
Ac O due to its hydrogenolysis under hydroformylation conditions
(
Schemes 2 and 3).
In a further study, we checked the possibility of using acetic acid as
the starting acylating agent instead of acetic anhydride. The reaction
performed with 5 equivalents of acetic acid gave the diol derivatives 3 in
Declaration of Competing Interest
2
nearly the same yield as with Ac O (ca. 80 %); albeit it required more
time to complete the acylation of aldehyde 2 (Table 3, run 7 vs run 3).
However, the composition of the product mixture was different: mon-
oacetates 3b and 3c were main reaction products, with only trace
amounts of diacetate 3d. In this run, 44 and 34 % selectivity for mon-
oacetates 3b and 3c were observed at nearly complete substrate con-
version, which is the best result obtained for these products in the
present work. Thus, acetic acid as the acylating agent is not as efficient
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgments
The authors acknowledge Fundaç a˜ o de Amparo a` Pesquisa do Estado
de Minas Gerais (FAPEMIG, Brazil), Conselho Nacional de Desenvolvi-
mento Científico e Tecnol o´ gico (CNPq, Brazil), Coordenaç a˜ o de Aper-
feicoamento de Pessoal de Nível Superior (CAPES, Brazil), and Instituto
Nacional de Ci ˆe ncia e Tecnologia de Cat a´ lise em Sistemas Moleculares e
Nanoestruturados (INCT-Cat a´ lise, Brazil) for the financial support.
as Ac
2
O to convert even secondary alcohol 3c in diacetate 3d. It can be
O, the hydroformylation step was less
noticed that, as compared with Ac
2
selective in the presence of acetic acid, which catalyzed undesirable
isomerization of starting limonene oxide (Table 3, runs 3 and 7, the data
for 1 h). However, the overall selectivity for products 3 at the end of the
reactions was nearly the same. The rate of the hydroformylation step in
the run with acetic acid (Table 3, run 7) was also much higher (94 %
conversion in 1 h) than the rate of the ring cleavage step (80 % con-
version for diol derivatives 3 in 48 h).
References
[
[
[
1] N. Armanino, J. Charpentier, F. Flachsmann, A. Goeke, M. Liniger, P. Kraf, Angew.
Chem. Int. Ed. 59 (2020) 2–37.
The green solvents DMC, DEC and anisole, can also be used as re-
action media to perform the one-pot hydroformylation/O-acylation of
limonene oxide. DMC and DEC underperformed toluene in terms of the
overall yield for diol derivatives 3; whereas anisole showed a similar
performance as toluene (Table 3, runs 8, 9 and 10 vs. run 7), being
preferred in terms of sustainability of the process.
2] C. Sell, in: 2nd ed., in: C. Sell (Ed.), The Chemistry of Fragrances: from Perfumer to
Consumer, vol. 2, RSC Publishing, Dorset, UK, 2006, pp. 52–88.
3] E. Breitmaier, Terpenes. Flavors, Fragrances, Pharmaca, Pheromones, Wiley-VCH,
Weinheim, 2006.
[
[
4] E.V. Gusevskaya, J. Jim `e nez-Pinto, A. B o¨ rner, ChemCatChem 6 (2014) 382–411.
5] A. Behr, A.J. Vorholt, K.A. Ostrowski, T. Seidensticker, Green Chem. 16 (2014)
9
82–1006.
[
[
6] A.E. Harman-Ware, in: M. Crocker, E. Santillan-Jimenez (Eds.), Chemical Catalysts
for Biomass Upgrading, Wiley, Weinheim, 2019, pp. 529–568.
7] A. Behr, L. Johnen, in: P.T. Anastas (Ed.), Handbook of Green Chemistry, vol. 7,
Wiley-VCH, Weinheim, 2012, pp. 69–92.
4
. Conclusions
In conclusion, in the present work we report a systematic study of the
[8] E.V. Gusevskaya, ChemCatChem 6 (2014) 1506–1515.
[
9] A. B o¨ rner, R. Franke, Hydroformylation. Fundamentals, Processes, and
limonene oxide hydroformylation and for the first time a novel one-pot
process for the hydroformylation of biorenewable limonene oxide with a
parallel cleavage of the oxirane ring to give acetoxy derivatives of the
corresponding diol. The process allows approaching novel products with
multiple functionalities (the aldehyde and acetate moieties), which have
new olfactory features. Either acetic acid or acetic anhydride can be used
as acylating agents. The presence of the oxirane functional group makes
the limonene oxide molecule extremely sensitive to isomeric rear-
rangements, especially in the presence of acidic species. This feature
seriously complicates the hydroformylation of this substrate. In addi-
tion, acids can catalyze undesirable condensation/polymerization re-
actions involving the substrate and the aldehyde products. Nevertheless,
appropriate conditions were found to perform with high to excellent
selectivity both the hydroformylation step separately and hydro-
formylation/oxirane ring cleavage simultaneously in a one-pot proced-
ure. It was also shown that the solvents with excellent sustainability
rankings, such as dimethylcarbonate, diethylcarbonate or anisole, can
be used advantageously as reaction media in both processes as alter-
natives for the conventional more problematic toluene.
Applications in Organic Synthesis, Wiley-VCH, Weinheim, 2016.
10] P.W.N.M. van Leeuwen, C. Claver (Eds.), Rhodium Catalyzed Hydroformylation,
Kluwer Academic Publishers, Dordrecht, 2000.
[
[11] P.W.N.M. van Leeuwen, P.C.J. Kamer, Catal. Sci. Technol. 8 (2018) 26–113.
[
[
[
12] H.-W. Bohnen, B. Cornils, Adv. Catal. 47 (2002) 1–64.
13] R. Franke, D. Selent, A. B o¨ rner, Chem. Rev. 112 (2012) 5675–5732.
14] J. Hibbel, E. Wiebus, B. Cornils, Chem. Ing. Tech. 85 (2013) 1853–1871.
[15] G.T. Whiteker, C.J. Cobley, Top. Organomet. Chem. 42 (2012) 35–46.
[
[
[
16] D.E. Fogg, E.N. dos Santos, Coord. Chem. Rev. 248 (2004) 2365–2379.
17] B.P. Bond ˇz i ´c , J. Mol. Catal. A 408 (2015) 310–334.
18] M. Vilches-Herrera, L. Domke, A. B o¨ rner, ACS Catal. 4 (2014) 1706–1724.
[19] G.M. Torres, R. Frauenlob, R. Franke, A. B o¨ rner, Catal. Sci. Technol. 5 (2015)
3
4–54.
[
[
20] P. Kalck, M. Urrutigoïty, Chem. Rev. 118 (2018) 3833–3861.
21] F.G. Delolo, G.M. Vieira, J.A.A. Villarreal, E.N. dos Santos, E.V. Gusevskaya, Catal.
Today (2021), https://doi.org/10.1016/j.cattod.2020.04.053.
[
[
22] R. Ciriminna, M. Lomeli-Rodriguez, P.D. Car `a , J.A. Lopez-Sanchez, M. Pagliaro,
Chem. Commun. 50 (2014) 15288–15296.
23] R.T. Gray, A.J. De Jong (Shell Oil Company), US Patent 4,352,937, 1982.
[24] R.L. Settine, G.L. Parks, G.L.K. Hunter, J. Org. Chem. 29 (1964) 616–618.
[
25] K. Bauer, D. Garbe, H. Surburg, Common Fragrance and Flavor Materials, Wiley-
VCH, Weinheim, 2001.
[
26] R. Uson, L.A. Oro, J.A. Cabeza, Inorg. Synth. 23 (1985) 126–127.
[27] V.V. Costa, K.A. da Silva Rocha, I.V. Kozhevnikov, E.F. Kozhevnikova, E.
V. Gusevskaya, Catal. Sci. Technol. 3 (2013) 244–250.
[
28] R.F. Cotta, R.A. Martins, M.M. Pereira, K.A. da Silva Rocha, E.F. Kozhevnikova, I.
V. Kozhevnikov, E.V. Gusevskaya, Appl. Catal. A 584 (2019), 117173.
[
[
29] P.W.N.M. van Leeuwen, C.F. Roobeek, J. Organomet. Chem. 258 (1983) 343–350.
30] H. Tricas, O. Diebolt, P.W.N.M. van Leeuwen, J. Catal. 298 (2013) 198–205.
7

M.P. de Oliveira et al.
Applied Catalysis A, General 616 (2021) 118082
[37] J.C. Leffingwell, E.E. Royals, Tetrahedron Lett. 43 (1965) 3829–3837.
[
[
[
[
[
[
31] C.M. Alder, J.D. Hayler, R.K. Henderson, A.M. Redman, L. Shukla, L.E. Shuster, H.
F. Sneddon, Green Chem. 18 (2016) 3879–3890.
[38] E.E. Royals, J.C. Leffingwell, J. Org. Chem. 31 (1966) 1937–1944.
[39] W. Chrisman, J.N. Camara, K. Marcellini, B. Singaram, C.T. Goralski, D.L. Hasha, P.
R. Rudolf, L.W. Nicholson, K.K. Borodychuk, Tetrahedron Lett. 42 (2001)
5805–5807.
32] D. Prat, A. Wells, J. Hayler, H. Sneddon, C.R. McElroy, S. Abou-Shehada, P.J. Dunn,
Green Chem. 18 (2016) 288–296.
33] F.G. Delolo, K.C.B. Oliveira, E.N. dos Santos, E. Gusevskaya, Mol. Catal. 462 (2019)
1
–9.
[40] N. Ravasio, F. Zaccheria, M. Guidotti, R. Psaro, Top. Catal. 27 (2004) 157–168.
[41] P.A. Dub, T. Ikariya, ACS Catal. 2 (2012) 1718–1741.
[42] S. Nakamura, M. Tamura (Kuraray Co.) EP Patent 40414, 1981.
[43] Y. Shima, T. Motoi, K. Nakamura (Mitsubishi Gas Chemical Co.) JP Patent
10120605, 1998.
34] A. de Camargo Faria, K.C.B. Oliveira, A.C. Monteiro, E.N. dos Santos, E.
V. Gusevskaya, Catal. Today 344 (2020) 24–31.
35] A. de Camargo Faria, M.P. de Oliveira, A.C. Monteiro, R.L.V. Mota, K.C.B. Oliveira,
E.N. dos Santos, E.V. Gusevskaya, Appl. Catal. A 591 (2020), 117406.
36] F.G. Delolo, E.N. dos Santos, E.V. Gusevskaya, Green Chem. 21 (2019) 1091–1098.
[44] D. M. Fenton (Oil Company of California) US Patent 3,579,566,1971.
8