Hydroformylation of recalcitrating biorenewable compounds containing trisubstituted double bonds

Article abstract of DOI:10.1016/j.apcata.2019.117406

Hydroformylation is a useful tool in industrial organic syntheses and it is easily applied to substrates containing terminal C[sbnd]C double bonds. Applying this reaction to substrates containing other functionalities or trisubstituted C[sbnd]C double bonds is not straightforward. Herein, the hydroformylation of several biorenewable alkenes with these features, namely: α-terpineol, terpinen-4-ol, limonene (double hydroformylation), and α-ionone, is presented. Only by the judicious choice of the catalytic system and reaction conditions, was it possible to obtain good yields and selectivity for these substrates.

Full text of DOI:10.1016/j.apcata.2019.117406

AppliedCatalysisA,General591(2020)117406
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Hydroformylation of recalcitrating biorenewable compounds containing
trisubstituted double bonds
Amanda de Camargo Faria, Mileny P. de Oliveira, Amanda C. Monteiro, Rayssa L.V. Mota,
Kelley C.B. Oliveira, 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:
Hydroformylation is a useful tool in industrial organic syntheses and it is easily applied to substrates containing
terminal CeC double bonds. Applying this reaction to substrates containing other functionalities or trisubstituted
CeC double bonds is not straightforward. Herein, the hydroformylation of several biorenewable alkenes with
these features, namely: α-terpineol, terpinen-4-ol, limonene (double hydroformylation), and α-ionone, is pre-
sented. Only by the judicious choice of the catalytic system and reaction conditions, was it possible to obtain
good yields and selectivity for these substrates.
Biorenewables
α-Ionone
Hydroformylation
Limonene
Terpineol
Rhodium
1. Introduction
aldehyde functionality usually confers a pleasant scent on the whole
molecule. The hydroformylation of simple terpenes containing terminal
Transition-metal-catalyzed hydroformylation is one of the most
powerful tools for the formation of carbon-carbon bonds in industrial
organic synthesis. The reaction offers an access to aldehydes starting
from olefins through the addition of carbon monoxide and hydrogen
(“syngas“) and represents one of the largest industrial applications of
homogeneous catalysis [1–6]. A high potential of hydroformylation for
the production of flavor&fragrance compounds is widely recognized;
however, commercial applications of this reaction in fine chemical in-
dustry are still limited [5,7,8]. The use of biorenewable olefins as
substrates in hydroformylation is especially valuable within the green
chemistry concept as it allows decreasing the dependence of chemical
industry on fossil resources [9–12]. The solvent nature is another im-
portant issue for liquid phase chemical processes [13–15]. Most of the
hydroformylation reactions in both academia and industry are currently
performed in the presence of solvents, usually rather toxic aromatic
compounds such as benzene and toluene. The replacement of these
solvents for eco-friendly alternatives would also contribute in the sus-
tainability of hydroformylation processes.
olefinic bonds, such as limonene, β-pinene and camphene, is well
documented in the literature [8,22–30]. On the other hand, examples of
using functionalized terpenes, especially those with sterically hindered
CeC double bonds, as substrates in hydroformylation are much more
scarce [8,31–35].
α-Terpineol is the most important monoterpenic alcohol and one of
the top 30 commonly used fragrance ingredients. This alcohol has a
pleasant lilac smell and occurs in various essential oils, such as conifer,
lavander, cajuput and petitgrain oils. For technical applications α-ter-
pineol is produced from α-pinene or limonene by acid catalyzed iso-
merization/hydration or through the partial dehydration of terpin hy-
drate obtained from turpentine oil [17,36,37]. The commercial
mixtures of synthetic α-terpineol usually contain considarable amounts
of γ-terpineol, which is much less common in nature than α-terpineol.
Another isomer, terpinen-4-ol, is also a by-product in the production of
α-terpineol from terpin hydrate. Terpinen-4-ol presents a spicy, woody-
earthy, nutmeg-like and lilac smell and occurs in significant amounts in
many essential oils, such as pine and eucalyptus oils [17]. The hydro-
formylation of terpineols could provide an access to polifunctionalized
compounds with interesting olfactory characteristics; however, it is a
challenging task as the CeC double bonds in their molecules are en-
docyclic and sterically hindered. We could find only one work on the
hydroformylation of α-terpineol, which described the formation of a
complex mixture of aldehydes under drastic reaction conditions
(160 °C, 70 bar) in toluene solutions [38]. The aldehyde mixture
Terpenes are natural compounds easily available from essential oils
of plants and flowers. Due to remarkable olfactory properties and bio-
logical activity, terpenes are traditionally used as ingredients in fra-
grance, cosmetic and pharmaceutical compositions [16–18]. Moreover,
terpenes can be transformed into other commercially valuable products
through a variety of chemical reactions [8,9,19–21]. Hydroformylation
is a particulary atractive method for upgrading of terpenes as an
⁎ Corresponding authors.
E-mail addresses: nicolau@ufmg.br (E.N.d. Santos), elena@ufmg.br (E.V. Gusevskaya).
https://doi.org/10.1016/j.apcata.2019.117406
Received 21 October 2019; Received in revised form 27 December 2019; Accepted 29 December 2019
Availableonline31December2019
0926-860X/©2020ElsevierB.V.Allrightsreserved.

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
Table 1
Hydroformylation of α-terpineol (1) in toluene solutions: ligand effecta.
Run
Ligand
TOFb
Time
Conversion
Selectivity for aldehydes
(P/Rh)
PPh3 (5)
(h−1
2
)
(h)
6
24
6
24
6
24
6
24
6
24
6
24
6
24
(%)
4
8
Total
100
100
62
77
87
1a
49
55
45
58
57
62
66
66
65
65
65
68
65
68
1b
51
45
17
19
30
35
34
34
35
32
35
32
35
32
1
2
3
4
5
6
7
none
22
40
36
34
30
26
26
66
50
92
44
91
47
91
43
90
36
84
(2,4-di-tBuC6H3O)3P (2)
(2,4-di-tBuC6H3O)3P (5)
(2,4-di-tBuC6H3O)3P (10)
(2,4-di-tBuC6H3O)3P (20)
(2,4-di-tBuC6H3O)3P (30)
97
100
100
100
100
100
100
100
100
a
Conditions: [α-terpineol] – 0,20 M (4 mmol), [Rh(COD)(OMe)]2 – 0,25 mM (5 μmol), 120 °C, gas phase – 40 atm (CO/H2 = 1/1), toluene – 20 mL. Conversion
and selectivity were calculated based on the substrate reacted using an internal standard (p-xylene); the rest of the mass balance was due to the formation of isomers
of α-terpineol.
b
TOF – initial turnover frequency (mol of the substrate converted per mol of Rh per hour) measured at low conversions (≤ ca. 30–40 %).
presented a pleasant woody and nutty aroma. As concerns terpinen-4-
ol, we could not find any reports on the hydroformylation of this sub-
strate starting the present project.
using a GC- Shimadzu GC2010 chromatograph equipped with a Rtx®-
5MS capillary column and FID detector. Conversion and selectivity
were calculated based on the reacted substrate using p-xylene as an
internal standard. Initial turnover frequencies (TOFs) were calculated at
low conversions, usually less than 40 %. In a typical run, a solution
(20.0 mL) of the substrate (4 mmol), [Rh(COD)(OMe)]2 (5.0 μmol),
phosphorus ligand (0–0.30 mmol) and p-xylene (2 mmol, GC internal
standard) in a specified solvent was placed into the reactor under argon;
the reactor was pressurized with “syngas” to 40–80 at m (CO/H2 = 1/2
to 2/1) and heated to the specified temperature. Then, the reaction
mixture was stirred with a magnetic stirrer for the specified time.
After the reaction, the solutions were analyzed by gas chromato-
graphy/mass spectrometry (GC–MS) on a Shimadzu QP2010-PLUS
equipment operating at 70 eV. Main reaction products were isolated
from the reaction solutions by column chromatography (silica gel 60,
hexane/ethyl acetate mixtures) and identified by NMR spectroscopy
(DEPT, COSY, HMQC, HMBC and NOESY experiments) on a Bruker
400 MHz spectrometer (CDCl3, TMS) (see Supplementary Material).
α−Ionone is a natural product found in various essential oils, such
as rose oil, and is also produced synthetically in a large industrial scale.
α−Ionone is also a terpenoid compound belonging to the specific class
of norisoprenoids. Ionones are highly required as comercial fragrance
ingradients and also known to present important pharmacological
properties [17,39]. The catalytic functionalization of the molecule of α-
ionone, which contains two olefinic bonds and a ketone moiety, could
offer the access to new compounds with pharmaceutical and fragrance
potential [40]. No studies on the hydroformylation of α-ionone were
previously reported in the literature as far as we know.
Herein, we present a first systematic study on the hydroformylation
of α-terpineol and terpinen-4-ol. Catalytic systems and conditions were
found to allow the synthesis of corresponding hydroxyaldehydes in
good yields. The expertise obtained in these studies has been extended
to related compounds, α-ionone and limonene, which have a similar
trisubstituted endocyclic CeC double bond in their structures. We also
demonstrated that the reactions can be performed in environmentally
more benign solvents, as compared to conventionally used toluene,
such as anisole, p-cymene, and diethylcarbonate.
3. Results and discussion
3.1. Hydroformylation of α-terpineol
As the endocyclic olefinic bond in the molecule of α-terpineol (1) is
sterically hindered, the hydroformylation was expected to be a chal-
lenging task for this substrate. The attempts to run the reaction in the
presence of triphenylphosphine were unsuccessful. Only 8 % of α-ter-
pineol was consumed for 24 h at 120 °C and 40 atm of syngas (Table 1,
run 1). However, in a non-promoted system, i.e, in the absence of
auxiliary phosphorous ligands, α-terpineol showed much higher re-
activity. Under the same conditions as in run 1 for 24 h, nearly a 70 %
substrate conversion was achieved with the formation of hydro-
formylated products in 77 % combined selectivity (Table 1, run 2). The
rest of the mass balance was due to the appearance of several isomers of
α-terpineol, such as other para-menthenic alcohols and isoborneol
(identified based on their characteristic mass-spectra and GC retention
times).
The hydroformylation products were isolated from the reaction
solutions by column chromatography and characterized by MS and
NMR spectroscopy. The major product (ca. 60 % selectivity) was
identified as endocyclic aldehyde 1a derived from the carbonylation of
the less substituted olefinic carbon in the α-terpineol molecule. The
second main product formed in smaller amounts (ca. 20 % selectivity)
was terminal aldehyde 1b with the formyl group at exocyclic carbon
2. Experimental
All chemical products were from commercial sources and used
without special treatment, unless otherwise indicated. (R)-(+)-limo-
nene (97 %), α-terpineol ((≥95 %), (−)-terpinen-4-ol (≥95 %), α-io-
none (90 %), triphenylphosphine (PPh3) and tris(2,4-di-tert-butyl-
phenyl)phosphite, (2,4-di-tBuC6H3O)3P, were received from Sigma-
Aldrich. The catalyst precursor, [Rh(COD)(OMe)]2 (COD = 1,5-cy-
clooctadiene), was synthesized by a previously reported method [41].
Toluene (anhydrous, 99.8 %), diethyl carbonate (DEC) (anhydrous,
≥99 %), anisole (anhydrous, 99.7 %) and p-cymene (99 %) were pur-
chased from Sigma-Aldrich. Toluene was passed through a silica
column and stored in an argon-filled glove box. Anisole, purchased in a
Sure/Seal™ bottle, was opened, stored in the glove box and used
without special treatment. DEC was distilled under argon and stored
over 4 Å molecular sieves. p-Cymene was distilled in a Kugelrohr dis-
tillation apparatus, collected under argon and stored in the glove box.
Catalytic reactions were run in a 100 mL homemade stainless steel
autoclave with magnetic stirring. Aliquots were periodically taken from
the reaction solutions using a valved dip tube without depressurization
of the reactor. The samples were analyzed by gas chromatography (GC)
2

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
Scheme 1. Hydroformylation of α-terpineol (1).
C7. Quaternary aldehyde 1c with the formyl group at carbon C1 was
detected only in trace amounts. The determined structures of products
1a, 1b and 1c are presented in Schemes 1 and 2.
intermediate B with rhodium at the quaternary carbon originates al-
dehydes 1b and 1c. The Markovnikov type addition of Rh-H to the
olefinic bond gives intermediate B, but the attack of the tertiary alkyl
ligand by the carbonyl, the next step for the hydroformylation cycle, is
disfavored by steric reasons and the β-elimination to yield intermediate
C is the preferred path. For this reason, quaternary aldehyde 1c is ex-
pected to be formed only in small amounts. On the other hand, terminal
aldehyde 1b, which is also derived from intermediate B, is formed in
significant amounts (20–30 % selectivity based on the reacted α-terpi-
neol). Aldehyde 1b is the product of the hydroformylation of terminal
olefin C, the isomer of the original α-terpineol (Scheme 2). Due to the
much higher reactivity of the terminal double bond in hydroformyla-
tion, terpineol C was not observed in the reaction solutions, it was
readily converted into aldehyde 1b.
The appearance of terminal aldehyde 1b in significant amounts
among the products suggests two alternative reactions pathways. The
first alternative (not shown in Scheme 2) would consist in the formation
of rhodium η3-allylic intermediates involving C1, C2 and C7 carbons,
followed by their carbonylation. However, this pathway looks unlikely
because i) the carbonylation of η3-allylic complexes would give term-
inal aldehyde 1b in greater amounts as compared to endocyclic alde-
hyde 1a for steric reasons; ii) rhodium η3-allylic complexes are known
to be very resistant to the insertion of carbon monoxide [28,29,42]; iii)
quaternary aldehyde 1c could not be formed at the carbonylation of η3-
allylic complexes. The alternative which involves the formation of
rhodium η1-alkyl intermediates shown in Scheme 2 seems more rea-
sonable for the hydroformylation of α-terpineol.
The NMR analysis showed the presence of ca. 5 % of γ-terpineol (2)
in the commercial α-terpineol used as the substrate. It was observed
that under the hydroformylation conditions, γ-terpineol reacted faster
than α-terpineol yielding hydroformylation product 2a (Scheme 3), in
The major aldehyde 1a arises from intermediate A with rhodium
attached to the less substituted carbon of the olefinic bond; whereas
Scheme 2. Mechanistic scheme for the hydroformylation of α-terpineol (1) and terpinen-4-ol (3).
3

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
both for (2,4-di-tBuC6H3O)3P promoted (runs 1 and 2) and non-pro-
moted (runs 3 and 4) systems. Under doubled pressure of the equimolar
carbon monoxide/hydrogen mixture, the reaction occurred at much
higher rates, under both conditions: with and without the auxiliary li-
gand. Considering that our first attempts to apply the non-promoted
rhodium system, the selectivity for the hydroformylation products was
significantly lower than in the Rh/phosphite system: 77% at 66%
conversion (Table 1, run 2), working with higher pressure (80 atm)
showed to be an excellent alternative for non-promoted systems.
Moreover, the isomers of α-terpineol seem to be consumed to yield
exclusively aldehydes 1a and 1b at longer reaction times under higher
pressures (Table 2, run 4).
In a further work we directed our efforts to substitute toluene, a
fossil-based and rather toxic solvent conventionally used in hydro-
formylation, by the solvents considered greener in modern solvent se-
lection guides [45,46]. As the solvent usually responds for the major
part of the environmental impact caused by the industrial chemical
process, the replacement of solvent for more benign alternative would
significantly contribute for the process sustainability. The hydro-
formylation of α-terpineol in different solvents was compared with that
in toluene for the systems with (Table 4) and without (Table 3) the
auxiliary (2,4-di-tBuC6H3O)3P ligand. p-Cymene, considered as a re-
newable and low-toxic compound, showed nearly the same perfor-
mance as a reaction medium for the hydroformylation of α-terpinol as
compared to that of toluene, in terms of both catalytic activity and
selectivity. p-Cymene, a natural compound occurring in various essen-
tial oils, is currently produced from petrochemical materials; however,
it can be also obtained in a large scale from limonene, one of the most
abundant in nature terpenic compound.
Diethylcarbonate (DEC) and anisole are highly recommended green
solvents by solvent selection guides with even better score than p-
cymene [45,46]. Being low-toxic and biodegradable, DEC and anisole
occupy in these guides prestigious positions with impressive overall
rankings comparable to those of ethanol and water [45,46]. Both DEC
and anisole showed an excellent performance as the solvents to perform
the hydroformylation of α-terpineol. Aldehydes 1a and 1b were ob-
tained with 95–97 % combined selectivity.
Scheme 3. Hydroformylation of γ-terpineol (2).
amounts corresponding to the γ-terpineol content in the original sub-
strate mixture. Aldehyde 2a was isolated from the reaction solutions as
an individual compound and fully characterized by MS and NMR
spectroscopy.
The addition of (2,4-di-tBuC6H3O)3P as an auxiliary ligand allowed
to nearly double the hydroformylation rate as compared to the non-
promoted system and also to increase the combined selectivity for main
aldehydes 1a and 1b up to 97 % (Table 1 run 3 vs. run 2). The ad-
vantageous effect of bulky phosphites in hydroformylation of sterically
hindered olefins can be explained by their steric and electronic prop-
erties [1,43,44]. The large cone angles prevent the coordination of
more than one phosphite ligand on the same rhodium atom to give bis-
ligand species, which are expected to be catalytically less active due to
the lack of space for the olefin coordination. In addition, the weaker
electron-donating and stronger electron-withdrawing ability of bulky
phosphite ligands, as compared to triphenylphosphine, benefit the
dissociation of carbon monoxide and coordination of the olefin on the
metal centre. Low concentrations of (2,4-di-tBuC6H3O)3P (P/Rh = 2
and 5) induced a beneficial effect on the reaction rate; however, with
larger ligand amounts (P/Rh = 10, 20 and 30) the reaction became
slower, probably, due to the difficulty for the bulky substrate to com-
pete with the also bulky ligand for the coordination sites on rhodium
(Table 1, runs 3–7).
As it can be noticed in Tables 3 and 4, p-cymene and DEC are better
alternatives than toluene for this reaction as they give similar outputs as
toluene and are environmentally preferable. Anisole, a low-cost, non-
toxic and biodegradable compound, deserves a special attention as the
greenest option with the highest sustainability rank within the solvent
series tested. Although the manufacture of anisole is currently based on
petrochemicals, it can be also produced from renewable materials such
as lignin and guaiacol. In anisole solutions, the reaction in the
In the system promoted by (2,4-di-tBuC6H3O)3P with P/Rh≥5, al-
dehydes 1a and 1b were virtually exclusive reaction products, with no
substrate isomerization occurring in an appreciable extent (Table 1,
runs 4–7).
We also studied the effects of the total pressure of syngas (Table 2)
Table 2
Table 3
Hydroformylation of α-terpineol (1): effect of pressurea.
Hydroformylation of α-terpineol (1) with non-promoted Rh system in various
solventsa.
Run P(H2) P(CO) TOFb Time Conversion Selectivity for aldehydes
Run
Solvent
TOFb
Time
Conversion
Selectivity for aldehydes
(h−1
36
)
(h)
6
(%)
44
91
60
95
26
66
56
94
Total
100
100
90
1a
66
66
65
68
45
58
60
70
1b
34
34
25
28
17
19
29
30
(h−1
54
)
(h)
6
24
6
24
6
24
6
(%)
56
94
55
94
60
95
58
95
Total
89
100
89
97
91
96
88
95
1a
60
70
60
65
63
64
61
65
1b
29
30
29
32
28
32
27
30
1
20
40
20
40
20
40
20
40
24
6
24
6
24
6
1
2
3
4
Toluene
p-Cymene
DEC
2
56
22
54
96
62
77
89
46
52
50
3c
4c
24
100
Anisole
24
a
Conditions: [α-terpineol] – 0,20 M (4 mmol), [Rh(COD)(OMe)]2 – 0,25 mM
(5 μmol), ligand – (2,4-di-tBuC6H3O)3P (P/Rh = 5), 120 °C, toluene – 20 mL.
Conversion and selectivity were calculated based on the substrate reacted using
an internal standard (p-xylene); the rest of the mass balance was due to the
formation of isomers of α-terpineol.
Conditions: [α-terpineol] – 0,20 M (4 mmol), [Rh(COD)(OMe)]2 – 0,25 mM
80 atm (CO/H2 = 1/1), solvent 20 mL.
a
(5 μmol), 120 °C, gas phase
–
–
Conversion and selectivity were calculated based on the substrate reacted using
an internal standard (p-xylene); the rest of the mass balance was due to the
formation of isomers of α-terpineol. DEC: diethylcarbonate.
b
TOF – initial turnover frequency (mol of the substrate converted per mol of
b
Rh per hour) measured at low conversions (≤ ca. 30–40 %).
TOF – initial turnover frequency (mol of the substrate converted per mol of
c
In the absence of (2,4-di-tBuC6H3O)3P.
Rh per hour) measured at low conversions (≤ ca. 30–40 %).
4

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
Table 4
suggested to occur due to the migration of the original double bond to
the terminal position to give isomeric terpineol C followed by its fast
hydroformylation. The hydroformylation of terpinen-4-ol was success-
fully performed both in non-promoted and (2,4-di-tBuC6H3O)3P pro-
moted systems at similar rates and with similar product distribution
(Table 5, runs 1 and 2). The increase in the concentrations of phosphite
decelerated the reaction due to the more severe competition for the
coordination sites on rhodium between the bulky ligand and the also
bulky substrate. It should be mentioned that although non-promoted
systems have an advantage to be less costly; the presence of phos-
phorous ligands usually allows stabilizing better rhodium species in
solutions and makes easier the catalyst recovery.
Hydroformylation of α-terpineol (1) with Rh/(2,4-di-tBuC6H3O)3P system in
various solventsa.
Run
Solvent
TOFb
Time
Conversion
Selectivity for aldehydes
(h−1
56
)
(h)
6
24
6
24
6
24
6
(%)
60
95
62
97
63
97
73
97
Total
90
1a
65
68
69
70
68
67
68
65
1b
25
28
31
30
31
29
29
33
1
2
3
4
Toluene
p-Cymene
DEC
96
48
50
66
100
100
99
100
97
98
Anisole
24
It is noteworthy that the hydroformylation of terpinen-4-ol occurred
slower than that of α-terpineol. Two hypothesis may explain this fact: i)
stronger deactivation of the double bond toward the interaction with
rhodium (I) species by the electron-withdrawing hydroxyl group in a
homoallylic position; ii) the coordination of 3 on rhodium through both
C]C and OH moieties to give a chelate complex, which is less prone to
react via the hydroformylation mechanism. In addition, the hydro-
formylation of terpinen-4-ol can be also performed in eco-friendly an-
isole even at a higher rate as compared to toluene (Table 5, run 4 vs.
run 2).
a
Conditions: [α-terpineol] – 0,20 M (4 mmol), [Rh(COD)(OMe)]2 – 0,25 mM
(5 μmol), ligand – (2,4-di-tBuC6H3O)3P (P/Rh = 5), gas phase – 80 atm (CO/
H2 = 1/1), 120 °C, solvent – 20 mL. Conversion and selectivity were calculated
based on the substrate reacted using an internal standard (p-xylene); the rest of
the mass balance was due to the formation of isomers of α-terpineol. DEC:
diethylcarbonate.
b
TOF – initial turnover frequency (mol of the substrate converted per mol of
Rh per hour) measured at low conversions (≤ ca. 30–40 %).
phosphite promoted system occurred even faster than in toluene
(Table 4, run 4 vs. run 1), reinforcing the recommendation of this
solvent for hydroformylation, as disclosed in our recent work [15].
3.3. Hydroformylation of endocyclic double bonds in related substrates
3.3.1. Double hydroformylation of limonene
Encouraged by the results obtained at the hydroformylation of ter-
pineols, we directed our attention to limonene, one of the most abun-
dant and low cost terpenic compounds available (Table 6). The mole-
cule of limonene (4) has two CeC double bonds: the terminal and the
endocyclic ones, the latter being similar to the double bonds in the
molecules of terpineols. The relatively facile hydroformylation of li-
monene in the terminal position to give monoaldehyde 4a (Scheme 5)
has been reported in several previous studies [8,22,23]. However, the
reports dealing with the double-hydroformylation of limonene, which
involves in the reaction its trisubstituted endocyclic double bond, are
really scarce [47].
The first hydroformylation of limonene, i.e. the hydroformylation of
the terminal double bond to give monoaldehyde 4a, readily occurred at
80 °C resulting in a complete limonene conversion in ca. 2 h (Table 6,
run 1). However, under these conditions, aldehyde 4a was extremely
resistant toward hydroformylation, with only small amounts of dia-
ldehydes 4b and 4c appearing even after 48 h of the reaction. We ex-
pected that the hydroformylation of the endocyclic double bond in al-
dehyde 4a could be performed under more severe temperature
conditions, similarly to the hydroformylation of terpineols. The in-
crease in the reaction temperature to 100 and then to 120 °C allowed to
accelerate the second hydroformylation step and to achieve a nearly
complete conversion of the monoaldehyde in 48 h (Table 6, runs 2 and
3). The products of double-hydroformylation, dialdehydes 4b and 4c,
were isolated from the reaction solution and identified by NMR spec-
troscopy (Scheme 5). The product 4b contains two formyl groups in
endocyclic and terminal positions, whereas both formyl groups in
minor product 4c are in terminal positions (4b/4c ≈ 2:1 in most of the
runs).
3.2. Hydroformylation of terpinen-4-ol
To the best of our knowledge, this is the first time that the hydro-
formylation of terpine-4-ol is described. Representative results are
presented in Table 5. The products were isolated from the reaction
solutions and identified by MS and NMR spectroscopy. All isolated
compounds and their mixture have a pleasant smell. Two main products
were endocyclic aldehyde 3a derived from the carbonylation of the less
substituted olefinic carbon and terminal aldehyde 3b with the formyl
group at exocyclic carbon C7 (Scheme 4). In the molecule of terpinen-4-
ol, the hydroxyl group is closer to the CeC double bond as compared to
α-terpineol. For this reason, aldehyde 3a undergoes a partial (20–30 %)
intramolecular cyclization to give the bicyclic compound 3c. Such cy-
clization does not take place in the aldehydes 1a-c. All three hydro-
formylation products were obtained with virtually 100 % combined
selectivity. The regioselectivity of hydroformylation was similar for
both substrates: endocyclic/terminal carbonylated products ≈ 70/30.
The reaction scheme for terpinen-4-ol under hydroformylation condi-
tions is analogous to that for α-terpineol presented in Scheme 2, added
the cyclization step. The formation of terminal aldehyde 3b was
Table 5
Hydroformylation of terpinen-4-ol (3)a.
Run
P/Rh
TOFb
Time
Conversion
Selectivity (%)
(h−1
31
)
(h)
6
24
6
24
6
24
6
(%)
41
87
48
90
25
71
55
95
3a
66
64
62
60
61
62
61
62
3b
21
24
21
24
22
24
21
24
3c
13
12
17
16
17
14
18
14
1
0
2
5
35
16
42
Nevertheless, the combined selectivity for dialdehydes 4b and 4c
did not exceed 65 %, even after the attempts to optimize the reaction
conditions (Table 6, runs 3–5). The rest of the mass balance was due to
the isomerization of limonene and the hydroformylation of these iso-
mers to give corresponding aldehydes (identified based on their char-
acteristic mass-spectra and GC retention times). A significant im-
provement in selectivity was achieved by running the reaction in two
consecutive steps: first, transforming completely limonene into mono-
aldehyde 4a at lower temperature and then increase the temperature of
the mixture. With this procedure we expected suppressing isomeriza-
tion because, differently from the original limonene, aldehyde 4c does
3
20
5
4c
24
a
Conditions: [terpinen-4-ol]
– 0,20 M (4 mmol), [Rh(COD)(OMe)]2 –
0,25 mM (5 μmol), ligand – (2,4-di-tBuC6H3O)3P, 120 °C, gas phase – 80 atm
(CO/H2 = 1/1), toluene – 20 mL. Conversion and selectivity were calculated
based on the substrate reacted using an internal standard (p-xylene).
b
TOF – initial turnover frequency (mol of the substrate converted per mol of
Rh per hour) measured at low conversions (≤ ca. 30–40 %).
c
Solvent – anisole.
5

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
Scheme 4. Hydroformylation of terpinen-4-ol (3).
Table 6
Table 7
Hydroformylation of limonene (4)a.
Hydroformylation of α-ionone (5)a.
Run P/Rh
P
T
Time Conversion Selectivity for aldehydes
Run
P/Rh
Time
Conversion
Selectivity (%)
(atm) (ºC) (h)
(%)
94
100
99
100
99
100
99
100
98
100
96
100
96
100
98
4a
97
83
91
31
87
1
80
6
70
6
98
5
96
7
98
3
96
4
4b
0
12
0
44
8
47
10
46
11
46
0
65
0
50
0
4c
0
4
0
17
4
20
5
20
1
15
0
23
0
(h)
5
24
170
5
24
170
5
24
170
(%)
100
100
100
98
100
100
96
6
6a
4
21
43
3
28
46
5
26
49
6b
0
2
32
0
2
30
0
2
1
10
10
10
5
40
40
40
40
80
80
80
80
80
80
2
48
2
48
2
48
2
24
2
24
9
1
0
5
5
96
77
20
97
70
13
95
68
10
2
100
120
120
120
70
120 24
60
120 48
70
120 24
60
120 48
2
3
4
3b
100
100
5
5
30
6b
7b
8bc
9bc
10
0
a
Conditions: [α-ionone] – 0,20 M (4 mmol), [Rh(COD)(OMe)]2 – 0,25 mM (5
μmol), ligand – (2,4-di-tBuC6H3O)3P, 120 °C, gas phase – 80 atm (CO/H2 = 1/
1), toluene – 20 mL. Conversion and selectivity were calculated based on the
substrate reacted using an internal standard (p-xylene); the rest of the mass
balance was due to the formation of the isomers of aldehyde 6a.
4
20
0
24
0
10
0
9
100
95
100
67
0
53
b
Solvent – anisole.
4
28
selectivity for dialdehydes 4b and 4c of ca. 90 % was obtained in both
solvents in the presence of (2,4-di-tBuC6H3O)3P ligand, which is the
best result ever reported for this reaction as far as we know (Table 6,
runs 6 and 8). The process can also be performed in the absence of
phosphite; however, with slightly lower selectivity in the second step
(Table 6, runs 7 and 9). Both dialdehydes derived from limonene and
their mixture presented a pleasant smell. It is important that the di-
hydroformylation of limonene can be successfully performed in the
solutions of environmentally benign anisole without any loss in the
catalyst performance. In the non-promoted system, the reaction in an-
isole showed higher selectivity for dialdehydes (81 %) as compared to
that in toluene (70 %) (Table 6, run 9 vs. run 7).
aConditions: [limonene] - 0,20 M (4 mmol), [Rh(COD)(OMe)]2 - 0,25 mM (5
μmol), ligand – (2,4-di-tBuC6H3O)3P, gas phase −CO/H2 = 1/1, toluene -
20 mL. Conversion and selectivity were calculated based on the substrate re-
acted using an internal standard (p-xylene); the rest of the mass balance was
b
due to the formation of isomers of limonene and their aldehydes. After the
indicated time, the temperature was increased to 120 °C and the reaction was
c
continued for the indicated time. Solvent – anisole.
not contain a terminal double bond and is therefore more stable toward
isomerization. Indeed, it was the case.
Representative data on the two-step hydroformylation of limonene
in toluene and anisole solutions are presented in Table 6. The combined
Scheme 5. Hydroformylation of limonene (4).
6

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
Scheme 6. Transformations of α-ionone (5) under hydroformylation conditions.
3.3.2. Hydroformylation of a-ionone
Finally, it was demonstrated that these reactions can be carried out in
anisole, a highly recommended green solvent, or alternative solvents
such as p-cymene and diethylcarbonate, all having better sustainability
ranking than the now standard toluene.
As hydroformylation is an interesting way of adding a new func-
tionality in the molecule, herein we report for the first time the hy-
droformylation of α-ionone, which is also a terpenoid compound be-
longing to the specific class of norisoprenoids. Representative results
are shown in Table 7. Under hydroformylation conditions, it was ob-
served a relatively fast hydrogenation of the exocyclic double bond in
α-ionone to give compound 6 shown in Scheme 6. This compound
showed to be rather resistant to hydroformylation, even under quite
harsh reaction conditions. With non-promoted system (Table 7, run 1),
after 24 h only 23 % of 6 was transformed into hydroformylation pro-
ducts shown in Scheme 6. After 170 h the yield increase to 80 %, but a
fair amount of aldehyde 6a was reduced to corresponding alcohol 6b.
In the presence of (2,4-di-tBuC6H3O)3P the results were only slightly
better. The reaction can be also performed in the solutions of anisole, a
green eco-friendly solvent (Table 7, run 3).
CRediT authorship contribution statement
Amanda de Camargo Faria: Conceptualization, Investigation,
Writing
- original draft. Mileny P. de Oliveira: Investigation,
Validation. Amanda C. Monteiro: Investigation, Validation. Rayssa
L.V. Mota: Investigation, Validation. Kelley C.B. Oliveira:
Methodology, Project administration. Eduardo N. dos Santos:
Conceptualization, Writing - review & editing. Elena V. Gusevskaya:
Conceptualization, Supervision, Writing - review & editing.
Acknowledgments
Thus, our studies have demonstrated that α-ionone is a particularly
difficult substrate to be transformed through hydroformylation, but the
combined yield of the hydroformylation products may reach ca. 80 %
under harsh conditions and at long reaction times.
CNPq, CAPES, FAPEMIG, and INCT-Catálise (Brazil) are gratefully
thanked for the financial support and fellowships.
Appendix A. Supplementary data
4. Conclusions
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2019.117406.
In this work the reactivity of several biorenewable terpenic sub-
strates recalcitrant to undergo hydroformylation was exploited. To
achieve the goal, it was necessary to employ comparatively harsh re-
action conditions and rhodium systems with electron-withdrawing li-
gands which allow room to facilitate substrate coordination, namely CO
(non-promoted system) or (2,4-di-tBuC6H3O)3P, a bulky phosphite that
produces mostly rhodium species containing only one phosphite ligand.
With this strategy it was possible to carry out the hydroformylation of
α-terpineol, terpinen-4-ol, and α-ionone. A two-step protocol to pro-
mote the double hydroformylation of limonene was also developed with
unprecedented yields and selectivity for corresponding dialdehydes.
References
[1] P.W.N.M. van Leeuwen, C. Claver, Rhodium Catalyzed Hydroformylation, Kluwer
Academic Publishers, Dordrecht, 2000.
[2] A. Börner, R. Franke, Hydroformylation. Fundamentals, Processes, and Applications
in Organic Synthesis, Wiley-VCH, Weinheim, 2016.
[3] C.W. Kohlpaintner, R.W. Fischer, B. Cornils, Appl. Catal. A Gen. 221 (2001)
219–225.
[4] H.-W. Bohnen, B. Cornils, Adv. Catal. 47 (2002) 1–64.
[5] R. Franke, D. Selent, A. Börner, Chem. Rev. 112 (2012) 5675–5732.
[6] J. Hibbel, E. Wiebus, B. Cornils, Chem. Ing. Tech. 85 (2013) 1853–1871.
7

A.d.C. Faria, et al.
AppliedCatalysisA,General591(2020)117406
[7] G.T. Whiteker, C.J. Cobley, Top. Organomet. Chem. 42 (2012) 35–46.
[8] E.V. Gusevskaya, J. Jimènez-Pinto, A. Börner, ChemCatChem 6 (2014) 382–411.
[9] A. Behr, A.J. Vorholt, K.A. Ostrowski, T. Seidensticker, Green Chem. 16 (2014)
982–1006.
[10] A. Behr, L. Johnen, Ch-J. Li (Ed.), Handbook of Green Chemistry, Volume 7: Green
Synthesis, first ed., Wiley-VCH, Weinheim, 2012, pp. 69–92.
[11] T. Vanbésien, E. Monflier, F. Hapiot, Eur. J. Lipid Sci. Technol. 118 (2016) 26–35.
[12] S.A. Jagtapa, E. Monflier, A. Ponchel, B.M. Bhanage, Mol. Catal. 436 (2017)
157–163.
Chem. 27 (2003) 533–539.
[29] H.J.V. Barros, J.G. da Silva, C.C. Guimarães, E.N. dos Santos, E.V. Gusevskaya,
Organometallics 27 (2008) 4523–4531.
[30] C.G. Vieira, J.G. da Silva, C.A.A. Penna, E.N. dos Santos, E.V. Gusevskaya, Appl.
Catal. A Gen. 380 (2010) 125–132.
[31] J.G. da Silva, H.J.V. Barros, E.N. dos Santos, E.V. Gusevskaya, Appl. Catal. A Gen.
309 (2006) 169–176.
[32] M.C. de Freitas, K.C.B. de Oliveira, A. de Camargo Faria, E.N. dos Santos,
E.V. Gusevskaya, Catal. Sci.Technol. 4 (2014) 1954–1957.
[33] C.G. Vieira, E.N. dos Santos, E.V. Gusevskaya, Appl. Catal. A Gen. 466 (2013)
208–215.
[13] C.J. Clarke, W.C. Tu, O. Levers, A. Bröhl, J.P. Hallett, Chem. Rev. 118 (2018)
747–800.
[14] Y.L. Gu, F. Jerome, Chem. Soc. Rev. 42 (2013) 9550–9570.
[15] F.G. Delolo, E.N. dos Santos, E.V. Gusevskaya, Green Chem. 21 (2019) 1091–1098.
[16] C. Sell, The Chemistry of Fragrances: From Perfumer to Consumer, Vol. 2, second
ed., RSC Publishing, Dorset, UK, 2006, pp. 52–88.
[34] C.G. Vieira, M.C. de Freitas, K.C.B. Oliveira, A. de Camargo Faria, E.N. dos Santos,
E.V. Gusevskaya, Catal. Sci.Technol. 5 (2015) 960–966.
[35] F.G. Delolo, K.C.B. Oliveira, E.N. dos Santos, E.V. Gusevskaya, Mol. Catal. 462
(2019) 1–9.
[17] H. Surburg, J. Panten, Common Fragrance and Flavor Materials. Preparation,
[36] P.A. Robles-Dutenhefner, K.A. da Silva, M.R.H. Siddiqui, I.V. Kozhevnikov,
E.V. Gusevskaya, J. Mol. Catal. A Chem. 175 (2001) 33–43.
[37] M. Román-Aguirre, L. De la Torre-Sáenz, W. Antúnez Flores, A. Robau-Sánchez,
A. Aguilar Elguézabal, Catal. Today 107-108 (2005) 310–314.
[38] F. Fujioka, R.M. Boden, J. Schreiber, (to International Flavors & Fragrances), US
4,491,537, (1985).
[39] M. Ansari, S. Emami, Eur. J. Med. Chem. 123 (2016) 141–154.
[40] M. dos Santos Costa, A.L.P. de Meireles, E.V. Gusevskaya, Asian J. Org. Chem. 6
(2017) 1628–1634.
Properties and Uses, Wiley-VCH, Weinheim, 2006.
[18] E. Breitmaier, Terpenes. Flavors, Fragrances, Pharmaca, Pheromones, Wiley-VCH,
Weinheim, 2006.
[19] K.A.D. Swift, Top. Catal. 27 (2004) 143–155.
[20] J.L.F. Monteiro, C.O. Veloso, Top. Catal. 27 (2004) 169–180.
[21] E.V. Gusevskaya, ChemCatChem 6 (2014) 1506–1515.
[22] L. Kollár, J. Bakos, B. Heil, P. Sándor, G. Szalontai, J. Organomet. Chem. 385 (1990)
147–152.
[23] L. Kollár, G. Bódi, Chirality 7 (1995) 121–127.
[24] F. Azzaroni, P. Biscarini, S. Bordoni, G. Longoni, E. Venturini, J. Organomet. Chem.
[41] R. Uson, L.A. Oro, J.A. Cabeza, Inorg. Synth. 23 (1985) 126–127.
[42] P.W.N.M. van Leeuwen, C.F. Roobeek, J. Mol. Catal 31 (1985) 345–353.
[43] P.W.N.M. van Leeuwen, C.F. Roobeek, J. Organomet. Chem. 258 (1983) 343–350.
[44] H. Tricas, O. Diebolt, P.W.N.M. van Leeuwen, J. Catal. 298 (2013) 198–205.
[45] 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.
508 (1996) 59–67.
[25] K. Soulantica, S. Sirol, S. Koinis, G. Pneumatikakis, Ph. Kalck, J. Organomet. Chem.
498 (1995) C10–C13.
[26] M.C. de Freitas, C.G. Vieira, E.N. dos Santos, E.V. Gusevskaya, ChemCatChem 5
(2013) 1884–1890.
[46] D. Prat, A. Wells, J. Hayler, H. Sneddon, C.R. McElroy, S. Abou-Shehada, P.J. Dunn,
[27] C.M. Foca, E.N. dos Santos, E.V. Gusevskaya, J. Mol. Catal. A Chem. 185 (2002)
17–23.
[28] C.M. Foca, H.J.V. Barros, E.N. dos Santos, E.V. Gusevskaya, J.C. Bayon, New J.
Green Chem. 18 (2016) 288–296.
[47] R. Franke, D. Fridag, D. Hess, F. Klasovsky, (to Evonik Industries AG), WO 2014/
135413 A1, (2014).
8