Tandem hydroformylation/hydrogenation over novel immobilized Rh-containing catalysts based on tertiary amine-functionalized hybrid inorganic-organic materials

Article abstract of DOI:10.1016/j.apcata.2021.118266

Non-phosphorous rhodium-containing catalysts for direct conversion of olefins to alcohols via tandem hydroformylation/hydrogenation have been designed and synthesized. Interaction between Rh(acac)(CO)₂ and tertiary amino groups on the surface of mesoporous hybrid organic-inorganic supports yielded materials which were successfully used in the tandem process. Data obtained for a selected catalyst K_N demonstrate that rhodium is in the Rh⁺¹ state highly dispersed on the surface and is bonded with nitrogen atoms both before and after use. Evaluation of the catalytic performance shows high activity (hydroformylation TOF 312 h^?1), chemoselectivity and stable hydroformylation yield at least in the first 5 cycles with a decrease in alcohol selectivity. The influence of temperature, reaction time, total pressure, and molar CO/H₂ ratio of syngas on oxygenate yields is described. Type of the hydroformylation active sites and possible pathways for the observed decrease in hydrogenation are discussed.

Full text of DOI:10.1016/j.apcata.2021.118266

Applied Catalysis A, General 623 (2021) 118266
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Tandem hydroformylation/hydrogenation over novel immobilized
Rh-containing catalysts based on tertiary amine-functionalized hybrid
inorganic-organic materials
,
Dmitry Gorbunova *, Maria Nenashevaa, Evgeny Naranovb, Anton Maximovb,
,
Edward Rosenbergc, Eduard Karakhanova *
a Department of Petroleum Chemistry and Organic Catalysis, Faculty of Chemistry, Moscow State University, Moscow, 119991, Russia
b Topchiev Institute of Petrochemical Synthesis RAS, Moscow, 119991, Russia
c Department of Chemistry and Biochemistry, University of Montana, Missoula, MT, 59812, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
Non-phosphorous rhodium-containing catalysts for direct conversion of olefins to alcohols via tandem hydro-
formylation/hydrogenation have been designed and synthesized. Interaction between Rh(acac)(CO)2 and tertiary
amino groups on the surface of mesoporous hybrid organic-inorganic supports yielded materials which were
successfully used in the tandem process. Data obtained for a selected catalyst KN demonstrate that rhodium is in
the Rh+1 state highly dispersed on the surface and is bonded with nitrogen atoms both before and after use.
Evaluation of the catalytic performance shows high activity (hydroformylation TOF 312 hꢀ 1), chemoselectivity
and stable hydroformylation yield at least in the first 5 cycles with a decrease in alcohol selectivity. The influence
of temperature, reaction time, total pressure, and molar CO/H2 ratio of syngas on oxygenate yields is described.
Type of the hydroformylation active sites and possible pathways for the observed decrease in hydrogenation are
discussed.
Tandem hydroformylation/hydrogenation
Reductive hydroformylation
Hydroxymethylation
Hybrid inorganic-organic materials
Tertiary amines
Rhodium complexes
1. Introduction
useful and promising reaction. It avoids construction of a separate hy-
drogenation unit, and a decrease in capital investment, making the
Oxo synthesis is one of the most important industrial petrochemical
processes. The key stage of oxo process is hydroformylation – the reac-
tion of unsaturated compounds with syngas leading to aldehydes used
for the production of alcohols, acids, amines, esters, and other com-
pounds. The oxo process’ worldwide production is approximately 11
million tons per year [1]. About 50 valuable chemicals are produced via
this process, which are in high demand in such products as plasticizers,
detergents, solvents, agrochemicals and in the food industry. Hydro-
formylation can be used for a wide range of substrates, from ethylene to
higher olefins and oleochemicals [2–4].
process more economically attractive. This process, which enables the
integration of hydroformylation and aldehyde hydrogenation stages
already exists in industry (Shell process [5]); it is based on the use of a
classical cobalt/phosphine catalytic system and characterized by C7-C14
olefin to alcohol conversion of about 80 %. However, the current version
requires high pressures and temperatures (160–200 ◦C, 5.0–10.0 MPa),
and the yield of side products is comparatively high. So, the search for
new, more effective systems for the process continues, with some
promising results. The different performances of the catalytic systems on
the basis of rhodium, cobalt, palladium, and ruthenium with typical
organic ligands, reaction conditions, and selected applications are
completely described in a review [6]. More recently, homogeneous Ru
[7] and Rh/Ru bimetallic catalytic systems [8,9] were investigated in
tandem reductive hydroformylation. Several new heterogeneous sys-
tems were also reported [10,11]. It should be noted, that alcohols can be
obtained via hydroformylation as the main product when alternative CO
sources, like methyl formate, formaldehyde, formic acid, CO/H2O, and
Among the most frequent industrial processes used for the aldehydes
obtained by hydroformylation is transformation into the corresponding
alcohols. Hydrogenation is often conducted as a separate stage. Many
researchers make an effort to develop catalytic systems for one-step
production of alcohols from unsaturated compounds; the direct and
selective conversion of terminal olefins to n-alcohols (reductive hydro-
formylation, or tandem hydroformylation/hydrogenation) is a very
* Corresponding authors at: Lomonosov Moscow State University, Faculty of Chemistry, 119991, Moscow 1, GSP-1, 1-3 Leninskiye Gory, Russia.
E-mail addresses: gorbunovdn@petrol.chem.msu.ru (D. Gorbunov), kar@petrol.chem.msu.ru (E. Karakhanov).
https://doi.org/10.1016/j.apcata.2021.118266
Received 24 April 2021; Received in revised form 9 June 2021; Accepted 14 June 2021
Available online 17 June 2021
0926-860X/© 2021 Elsevier B.V. All rights reserved.

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
CO2/H2 mixtures are used [12–16].
and further became dark brown (in 3 h). After reaction, the solid product
was separated by centrifugation and washed with dichloromethane (2 ×
10 mL). Next, it was dispersed in dichloromethane and dried in vacuum.
KS became lighter as it dried and turned to dark yellow. Both catalysts
(0.365 g of KR and 0.370 g of KS) were obtained as yellow powders.
Rhodium complexes with tertiary amines [15–22] are also known to
catalyze tandem hydroformylation/hydrogenation. The interaction of
rhodium with tertiary nitrogen in the ligand sphere enhances the ac-
tivity of catalytic centers in aldehyde hydrogenation. Jurewicz and
co-workers [17] found polymeric material with N,N-dimethylbenzyl-
amine fragments (anion exchange resin) to be an effective support for Rh
tandem hydroformylation/hydrogenation catalysts, starting a series of
works aimed at heterogenization of the process, which is of great
importance for easier separation and recycling [23]. The catalysts based
on it were studied in several further works [24–27] and showed
exceptional stability among other polymer-based catalysts [25,28]. The
development of the idea led to the design of silica/tertiary
amine-supported Rh catalysts for the process [29,30]. They demon-
strated higher alcohol selectivity than the catalyst supported by N,
N-dimethylbenzylamine-containing polymer being tested under
comparatively mild conditions (100 ◦C, 7 MPa) [30]. However,
silica-based N-containing catalysts (supported by SIL-CH2-CH2-CH2-NR2
materials) are characterized by considerable Rh leaching [29]. The
combination of the advantages of silica-based heterogeneous catalysts
with a higher stability would be desirable progress in this field.
The approach to this challenge described here is the use of meso-
porous silica with anchored N-containing polymers providing a specific
microenvironment for Rh active centers due to the high concentration of
nitrogen atoms. We report the design of new heterogeneous rhodium
catalysts based on the materials BP-1 and WP-1, where poly(allylamine)
and poly(ethyleneimine) chains, respectively, are anchored on an
amorphous silica gel surface.
2.4. KN and KW synthesis
BP-1-NMe2 or WP-1-NMe2 (0.400 g) and Rh(acac)(CO)2 (0.095 g,
0.37 mmol) were placed into a round-bottom flask (25 mL) equipped
with a magnetic stirrer, then dichloromethane (5 ml) was added. The
mixture was stirred at room temperature for 12 h. In both cases, a color
change was detected: the initially yellow mixture turned orange (in 0.5
h), and further became dark brown (in 3 h). After reaction, the solid
product was separated by centrifugation and washed with dichloro-
methane (2 × 10 mL). Next, it was dispersed in dichloromethane and
dried in vacuum. Both catalysts became lighter as they dried, KN was
obtained as a light brown powder (0.360 g), and KW – as a yellow one
(0.350 g).
2.5. Characterization
IR spectra of the materials were taken with a Nicolet IR2000
(Thermo Scientific) instrument using multiple distortions of the total
internal reflection method with multi-reflection HATR accessories,
containing a 45◦ ZnSe crystal for different wavelengths with a resolution
of 4 cmꢀ 1 in the range of 4000–400 cmꢀ 1. The spectrum was taken by
averaging 100 scans. The structures of catalysts were investigated by X-
ray photoelectron spectroscopy with an LAS-3000 instrument equipped
with a OPX-150 photoelectron retarding-potential analyzer. For photo-
2. Experimental
electron excitation, aluminium anode radiation (Al Kα = 1486.6 eV) was
2.1. BP-1-NMe2 and WP-1-NMe2 synthesis
used at a tube voltage of 12 kV and an emission current 20 mA.
Photoelectron peaks were calibrated with reference to the carbon C1s
corresponding to a binding energy of 285 eV. The TEM study was done
using FEI’s Tecnai Osiris TEM equipped with X-FEG gun at 200 keV.
High angle annular dark field (HAADF) with EDX elemental mapping
was used to identify the composition of the samples. Nitrogen desorp-
tion/desorption isotherms were recorded at 77 K with a Micromeritics
Gemini VII 2390 instrument (Micromeritics, Norcross, GA, United
States). All samples were degassed at 120 ◦C for 12 h before measure-
The initial material (BP-1 or WP-1, 0.600 g) was put into a round-
bottom flask (25 ml) equipped with a magnetic stirrer and a reflux
condenser. The flask was placed into a cold water bath. The mixture of
formic acid (95 %, 3 ml) and water (0.3 mL) was added. In the case of
BP-1, the bubbles could be observed at this stage, for WP-1 they were not
detected. Next, formaldehyde (37 % water solution, 3 mL) was added,
and the mixture was stirred for 15 min on cooling. Further, it was
refluxed for 7 h and then stirred at room temperature overnight. After
reaction, the mixture was centrifuged; the solid product was decanted
and washed with methanol (2 × 10 ml). The resulting white powder was
suspended in dichloromethane and dried in vacuum to yield 0.490 g (for
BP-1) or 0.460 g (for WP-1) of methylated support in the form of white
powder.
ment. The surface area (SBET
nauer–Emmett–Teller (BET) method based on adsorption data in the
relative pressure range P/P0 = 0.05–0.2. The total pore volume (Vtot
) was calculated using the Bru-
)
was determined by the amount of nitrogen adsorbed at a relative pres-
sure of P/P0 = 0.995. Quantitative determination of rhodium in the
samples was performed by ICP AES methods using an IRIS Interpid II
XPL spectrometer (Thermo Electron Corp., USA) at the wavelength
343.49 nm. Percentage of carbon, nitrogen and hydrogen was deter-
mined by elemental analysis on a Vario Microcube Elementar instru-
ment. The samples were combusted at 950 ◦C using helium as a carrier
gas (flow rate 120 mL/min). The separation of the products (nitrogen,
carbon dioxide, and water) was carried out on a thermal desorption
column of the device in a helium flow, using a thermal conductivity
detector. Thermogravimetric analysis (TGA) of the samples was carried
out using a Mettler Toledo instrument TGA/DSC1 in air flow (70 mL/
min) at the heating rate 10 ◦C/min from 30 to 1000 ◦C. The accuracy of
the temperature measurement was ±0.3 ◦C. The results were analyzed
with the service program STARe.
2.2. Rh(acac)(CO)2 synthesis
Rh(acac)(CO)2 was prepared as described in [31]. RhCl3*3H2O (1.07
g, 0.35 mmol) was dissolved in dimethylformamide (20 mL), then
freshly distilled acetylacetone (4 mL, 39 mmol) was added. The solution
was refluxed for 30 min. Further, it was cooled to the room temperature,
and cold water (50 mL) was added. Red precipitate was filtered out and
washed with water and ethanol. Then the complex was recrystallized
from hexane. Red-green dichroic crystals were obtained (0.72 g, yield 75
%), mp = 154–155 ◦C. IR: 2085 cmꢀ 1, 1987 cmꢀ 1
.
2.3. KR and KS synthesis
2.6. Catalytic experiments
BP-1 or WP-1 (0.400 g) and Rh(acac)(CO)2 (0.095 g, 0.37 mmol)
were placed into a round-bottom flask (25 ml) equipped with a magnetic
stirrer, then dichloromethane (5 mL) was added. The mixture was stirred
at room temperature for 12 h. For BP-1 (KR synthesis) the mixture was
yellow during the reaction. In the case of WP-1 (KS synthesis), the color
change was dedicated: initially yellow mixture turned orange (in 5 min),
A typical procedure for tandem hydroformylation/hydrogenation of
octene-1 was conducted as follows. A stainless steel autoclave (20 mL)
equipped with a manometer and a magnetic stirrer was charged with KN
(30 mg), octene-1 (0.6 mL, 4 mmol), and toluene (2 mL). The autoclave
was sealed, purged with hydrogen, and then consequently pressurized
2

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
Fig. 1. Initial hybrid organic-inorganic materials used for design of the catalysts.
Scheme 1. Synthesis of BP-1-NMe2.
with hydrogen to 3.0 MPa and syngas (CO:H2 = 1:1) to a total pressure
6.0 MPa. The mixture was stirred in an oil bath at 80 ◦C for 5 h. After
reaction, the reactor was cooled down and depressurized. The reaction
mixture was centrifuged, the liquid phase was removed and analyzed by
GLC. The catalyst was washed with toluene (2 mL), separated by
centrifugation, and reused under the same conditions. Liquid products
were analyzed by gas-liquid chromatography on a Khromos chromato-
graph equipped with a flame ionization detector and a 50-m-long
capillary column coated with the DB-5 phase using temperature pro-
gramming of 60–230 ◦C and helium as a carrier gas.
Table 1
Elemental analysis data for initial and modified materials.
Species
C/wt %
H/wt %
N/wt %
H:N (mol.)
C:N (mol.)
BP-1
10.85
13.50
11.88
14.34
2.36
2.80
2.79
3.48
3.05
2.56
4.29
3.04
10.83
15.30
9.12
4.14
6.15
3.23
5.51
BP-1-NMe2
WP-1
WP-1-NMe2
16.04
Table 2
Surface characteristics of initial and modified materials.
3. Results and discussion
Material
SBET [a]/m2
Total pore volume/cm3
Average pore size[b]
nm
/
gꢀ 1
gꢀ 1
3.1. Synthesis and characterization
BP-1
221
194
171
167
0.81
0.69
0.87
0.85
8.7
BP-1-NMe2
WP-1
8.7
For the synthesis of the heterogeneous catalysts, silica composites
BP-1 and WP-1 were used. These materials (Fig. 1) combine a high
density of functional groups with the rigidity of the silica gel matrix
[32–36]; they can be applied in the mining industry for the removal of
contaminating metals from liquid waste streams [37] and in catalysis
[38–41]. In BP-1, poly(allylamine) chains are anchored on an amor-
phous silica gel surface. For WP-1, polymeric constituent is poly(ethyl-
eneimine) (Fig. 1).
11.9
12.1
WP-1-
NMe2
[a]
[b]
Brunauer–Emmett–Teller (BET) method.
Barrett-Joyner-Halenda (BJH) model.
rhodium and impact catalytic activity. In Table 1, the results of
elemental analysis of starting materials and supports obtained are
shown.
As mentioned above, according to the literature, rhodium complexes
with tertiary alkyl amines are exclusively active in aldehyde hydroge-
nation [17–22]. For this reason, we modified terminal amino groups of
initial materials to obtain methylated supports BP-1-NMe2 and
WP-1-NMe2 (Scheme 1).
In both cases, carbon and hydrogen content increased after modifi-
cation. For BP-1, the molar increase in hydrogen content is 2.2 times
more than that of carbon (in theory, it should be 2 times when amino
groups get methylated). However, for WP-1 this ratio is bigger (about 3);
probably, some solvent stayed unremoved after the synthesis.
The surface area analysis showed that both initial and modified
materials can be classified as mesoporous (Table 2). The modification
One more reason for such a modification is a possible reaction be-
tween unmodified primary amino groups and aldehydes obtained during
the catalytic process, which would change the microenvironment of
3

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
dichloromethane by the procedure described above. In the case of BP-1,
the suspension was yellow for all the 12 h of stirring, and the yellow
powder KR was obtained after washing and drying. The mixture of WP-1
and Rh(acac)(CO)2 was undergoing the same color changes as for BP-1-
NMe2 and WP-1-NMe2, so we can suggest that they are related to the
presence of tertiary nitrogen atoms in the material (WP-1 initially con-
tains some (30 %), while BP-1 doesn’t). The final product obtained from
unmodified WP-1 was a yellow powder KS.
Table 3
Rh content in catalysts[a]
.
Sample
Support
Rh content/wt %
KN
KW
KR
KS
BP-1-NMe2
WP-1-NMe2
BP-1
1.93
2.14
1.86
2.08
WP-1
[a]
Determined by ICP-AES.
The ICP-AES data on Rh content in the four catalysts is shown in
Table 3.
doesn’t change the surface area drastically, and pore size remains
unchanged.
In all samples, about 2 % of Rh was found. Methylation of amino
groups doesn’t result in a significant change in Rh content.
To obtain the two rhodium catalysts KN and KW, modified supports
BP-1-NMe2 and WP-1-NMe2, respectively, were suspended in dichloro-
methane. Next, Rh(acac)(CO)2 was added (1 mol Rh per 2 mol N), and
the mixtures were stirred for 12 h at room temperature. In the begin-
ning, the mixtures were light yellow, but in 0.5 h they became orange.
Then, they get darker and turned brown in 3 h. After reaction, the cat-
alysts were isolated by centrifugation, washed twice with dichloro-
methane, and dried in vacuum, becoming lighter as they dried. Two
solid products were obtained: KN as a light brown powder and KW as a
yellow one.
To indicate the interaction between Rh and nitrogen atoms of the
support, we conducted a model experiment. We used PEI (average
Mw~25,000, branched, Sigma-Aldrich) as a soluble analog of hetero-
geneous supports and Rh(acac)(CO)2 as Rh source. These compounds
(taken in 1:1 M N:Rh ratio) were solved in DMSO-d6 and mixed for 1 h at
60 ℃. Right after dissolution, the mixture became green and then
changed its color to yellow. After it was cooled, the 1H NMR spectrum of
the mixture was registered. In the spectrum of PEI, broadened multiplet
in 2.38–2.54 ppm region (CH2 protons) and broadened signal at
2.70–3.70 (NH and NH2 protons) can be observed. The spectrum of Rh
(acac)(CO)2 has one main signal – singlet at 3.37 ppm – which corre-
sponds to protons of CH3 groups of acetylacetonate anion bonded with
In order to confirm that methylation of BP-1 and WP-1 amino groups’
at the first stage is crucial for the performance of the resulting catalysts,
we also treated unmodified BP-1 and WP-1 with Rh(acac)(CO)2 in
Scheme 2. Tandem hydroformylation/hydrogenation and side reactions.
Table 4
Homogeneous and heterogeneous catalysts in tandem hydroformylation/hydrogenation of octene-1[a]
.
Catalyst
Conv., %
Selectivities
Int. octenes, %
Octane, %
Aldehydes, %
n:iso ald+alc
Alcohols, %
n:iso alc
[b]
Rh(acac)(CO)2
100
100
<1
<1
2
97
99
0.9
2.4
–
<1
–
–
Rh(acac)(CO)2/
<1
[c]
PPh3
Rh(acac)(CO)2/ TMPDA[d]
97
97
98
94
98
98
98
71
96
25
29
23
21
24
25
23
30
29
3
56
27
32
7
2.0
2.1
2.0
2.1
2.0
1.9
1.9
2.1
2.0
16
34
34
52
41
6
3.0
3.1
3.1
2.4
3.0
5.0
–
[e]
KN
10
11
20
10
7
KN*[e]
[e]
KW
[e]
KW*
25
62
70
56
61
[e]
KR
KR*[e]
7
[e]
KS
4
10
5
2.5
4.0
KS*[e]
[a]
5
General conditions: octene-1 0.6 mL (4 mmol), toluene 2 mL, 80 ◦C, syngas (CO:H2 = 1:3) 6.0 MPa, 5 h.
[b]
[c]
[d]
[e]
Rh(acac)(CO)2 2 mg (0.008 mmol).
Rh(acac)(CO)2 2 mg (0.008 mmol), PPh3 6 mg (0.023 mmol).
Rh(acac)(CO)2 2 mg (0.008 mmol), N,N,N′,N′-Tetramethyl-1,3-propanediamine 0.005 mL (0.03 mmol).
Heterogeneous catalyst 30 mg, octene/Rh (mol.) = 670, * – recycle.
4

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
Fig. 2. Tandem hydroformylation/hydrogenation of octene-1 at different temperatures. “Total oxygenates” is a sum of all aldehydes and alcohols produced from
octene-1. Conditions: KN 30 mg, octene/Rh (mol.) = 670, octene-1 0.6 mL (4 mmol), toluene 2 mL, syngas (CO:H2 = 1:3) 6.0 MPa, 5 h.
rhodium. After the model reaction, the spectrum of the mixture contains
both the signals of individual acetylacetone (two tautomers) and ace-
tylacetonate anion. This indicates that the anion was partly displaced
from the Rh coordination sphere. At the same time, the signals of the ion
are slightly shifted compared to original Rh(acac)(CO)2 spectrum (3.33
vs 3.37 ppm). This shift is most probably caused by the interaction be-
tween Rh center and nitrogen atoms. Another evidence of this interac-
tion is the lack of the signals of initial polymer in the spectrum. All NMR
spectra are given in the SI, p.23-25.
unmodified BP-1 are very close to those possessed by KN, and the
aldehyde yield doesn’t change from the first use to the second. At the
same time, both KN and KW are significantly more effective in aldehyde
hydrogenation than KR and KS, which proves that for tandem hydro-
formylation/hydrogenation the modification of starting materials was
favorable.
At the next step, we studied temperature effect on tandem hydro-
formylation/hydrogenation of octene-1 over the KN catalyst. The cata-
lyst stability can be significantly influenced by temperature, so we used
every portion of KN twice at each temperature (first run and recycle)
(Fig. 2). At 60 ◦C, total conversion of octene-1 after 5 h is about 30–35 %
in both reactions. For higher temperatures, it reaches 98–100 %.
Oxygenate selectivity (a sum of all aldehyde and alcohol selectivities)
increases with increasing temperature, but hydrogenation of aldehydes
obtained apparently has an optimum value in the studied temperature
range. Probably, deactivation of hydrogenation centers occurs at high
temperatures (120–140 ◦C), whereas hydroformylation selectivity
doesn’t fall under the examined conditions. An increase in temperature
also promotes a decline in alcohol yields from the first use to the second.
So, we observe no maxima coincidence between the “first” and “reused”
curves: in the first cycle, the best alcohol selectivity is achieved at 100
◦C, while in the second run the selectivity significantly decreases at this
temperature. At 80 ◦C, the yield is sufficient, and no loss in alcohol
selectivity is observed after the first use. For this reason, we have chosen
temperature of 80 ◦C for our further experiments.
3.2. Catalytic study
Tandem hydroformylation/hydrogenation of octene-1 was studied as
a model reaction to optimize reaction conditions. Octene-1 molecular
transformations observed are shown in Scheme 2.
In Table 4, conversions and selectivities to alcohols and aldehydes in
the presence of catalysts KN, KW, KR, and KS are compared with those in
homogeneous rhodium-containing systems. In ligand-free system, no
alcohol formation can be observed. In classical Rh/PPh3 one only traces
of nonanol-1 were found. The addition of tertiary amine leads to hy-
drogenation of aldehydes.
Both KN and KW appear to be more selective to alcohols than ho-
mogeneous analog. In all the cases where alcohols are obtained n:iso
ratio for them exceeds this for aldehydes. It can be explained by the
difference between n- and iso-aldehyde reactivities. Total n-product
selectivities for supported catalysts (n:iso ratio about 2) are also close to
Rh/TMPDA homogeneous system. KN and KW have the same hydro-
formylation activity (characterized by total yield of aldehydes and al-
cohols), but KW is more active in alcohol formation, probably due to the
higher tertiary nitrogen number in the support structure. The hydro-
formylation yield and regioselectivity over KR synthesized from
The effect of syngas pressure and composition on alcohol selectivity
is illustrated in Fig. 3. It can be concluded that the general pressure of
syngas is more critical for alcohol yields than CO:H2 ratio. At the same
time, total oxygenate selectivity decreases with reduction in CO partial
pressure. We observe that an excess of hydrogen promotes octene hy-
drogenation resulting in the growth of octane content in the reaction
5

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
Fig. 3. Tandem hydroformylation/hydrogenation of octene-1 at different pressures. Conditions: KN 30 mg, octene/Rh (mol.) = 670, octene-1 0.6 mL (4 mmol),
toluene 2 mL, 80 ◦C, 5 h.
conditions, which are able to catalyze hydrogenation of aromatic com-
Table 5
–
pounds [42]. The inhibition of C C double bond hydrogenation of
K
N in octene-1 and nonanal hydrogenation[a]
.
cinnamaldehyde in the Rh/NEt3 system with CO added is described in
[43]. On the contrary, hydrogen in the absence of CO is far less effective
in nonanal hydrogenation than syngas, which also corresponds to data
for cinnamaldehyde obtained in [43]. It confirms the conventional point
of view that in systems of the type, hydrocarbonyl rhodium complexes
with tertiary amine ligands catalyze an aldehyde to alcohol trans-
formation [20].
Substrate
Gas composition
Results
Octene-1
Nonanal
Nonanal
H2, 4.5 MPa
Yield of octane 100 %
Yield of nonanol-1 2 %
Yield of nonanol-1 51 %
H2, 4.5 MPa
Syngas (CO:H2 = 1:3),
6.0 MPa
[a]
Conditions: KN 30 mg, substrate 4 mmol, toluene 2 mL, 80 ◦C, 5 h.
Next, different solvents were tested as a reaction media for octene-1
hydroformylation/hydrogenation (Fig. 4).
products at CO:H2 ratios 1:2 and 1:3 compared with 1:1 at the same total
pressure values, which can explain the drop of oxygenate selectivity by
about 10 % in both cases. However, the percentage of aldehydes un-
dergoing hydrogenation rises as hydrogen content in syngas increases,
so no loss in the final yield of alcohols can be noted. Regioselectivity of
hydroformylation declines with an increase in pressure.
The highest alcohol selectivity is achieved in THF. Nonanol-1 is the
main product in both cycles, however, its yield decreased after the first
use. The analogous drop occurs in ethanol and heptane, while in toluene
the yields of alcohols are almost the same in the first and second uses. In
a solvent-free system alcohol molar yields are comparable to yields
obtained in the presence of solvents. Total n:iso ratios are close to those
in non-polar solvents, and undoubtedly the ratios for alcohols will also
decrease if the reaction continues.
To investigate the role of CO and H2 in the two stages of the tandem
reaction, we conducted a series of hydrogenation experiments: in the
presence of KN, octene-1 and nonanal were treated with hydrogen. Also,
the action of syngas on nonanal was studied (Table 5).
The time dependence profile of the reaction is presented in Fig. 5.
Firstly, octene-1 content declines as it transforms into aldehydes,
undergoes isomerization, or hydrogenation. At 1,5 h the amount of al-
dehydes reaches a peak and then falls with their conversion to alcohols.
Nonanal reacts with hydrogen more readily than its isomer, but after 12
h the contents of two C9 alcohols are almost equal, and the yield reaches
92 %. Most of the rest is octane formed in the first three hours, also some
sterically-hindered iso-aldehydes stay untransformed (about 3 % of the
product mixture).
Hydrogen readily reacts with octene-1 over KN in the absence of CO.
It is worth mentioning that toluene used as a solvent was partly hy-
drogenated in the system (about 7 % of toluene transformed to meth-
ylcyclohexane). We have never observed this process in other entries,
neither with syngas nor with nonanal and hydrogen. Probably, the
–
catalyst is specifically active toward C C double bonds and even aro-
matic compound hydrogenation when no carbon monoxide or other
carbonyl group (e.g. aldehyde) can interact with rhodium. Also, this
may be connected to the formation of Rh nanoparticles under CO-free
As it shown in Table 6, KN was successfully used for tandem
6

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
Fig. 4. Tandem hydroformylation/hydrogenation of octene-1 in various solvents. Conditions: KN 30 mg, octene-1 0.6 mL (4 mmol), octene/Rh (mol.) = 670, solvent
2 mL, 80 ◦C, syngas (CO:H2 = 1:3) 6.0 MPa, 5 h. For no-solvent experiments: octene-1 2.6 mL (17 mmol).
Fig. 5. Time course plot for tandem hydroformylation/hydrogenation of octene-1. Conditions: KN* 50 mg (reused after the first reaction cycle under the same
conditions, 5 h), octene-1 0.9 mL (6 mmol), octene/Rh (mol.) = 600, toluene 3.3 mL, 80 ◦C, syngas (CO:H2 = 1:3) 6.0 MPa.
hydroformylation/hydrogenation of various unsaturated substrates.
Hexene-1 converts to C7 alcohols with selectivity close to that
observed for hydroformylation/hydrogenation of octene-1. For styrene
and cyclohexene, the corresponding alcohols are the main products, as
well as for the model substrate. Due to electronic effects,
hydroformylation of styrene leads to branched aldehyde (2-phenyl-
propanal) predominantly [44], so 2-phenylpropanol prevails among
oxygenates obtained. Surprisingly, although the conversion of ethylene
to propanal is as high as 70 %, propanol-1 cannot be produced under the
same conditions in sufficient yield. It may be connected to the lower
7

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
Table 6
Tandem hydroformylation/hydrogenation of various substrates[a]
.
Substrate
Conv./ %
Selectivities
Int. olefins/ %
Hydrogenated substrate/ %
Aldehydes/ %
n:iso ald+alc
Alcohols/ %
n:iso alc
Octene-1
97
29
29
–
10
27
37
3
2.1
2.3
–
0.3
–
34
32
97
64
1
3.1
3.3
–
0.3
–
Hexene-1
Cyclohexene
Styrene
94
2
55
<1[d]
13[e]
n/a
77
–
23
99
Ethylene [b]
70[c]
–
[a]
Conditions: KN 30 mg, substrate 4 mmol, substrate/Rh (mol.) = 670, toluene 2 mL, 80 ◦C, syngas (CO:H2 = 1:3) 6.0 MPa, 5 h.
p (CH2CH2) = 0.5 MPa.
[b]
[c]
Conversion to oxygenates, estimated by the amount of liquid product using heptane as an internal standard. Gaseous products were not analyzed after reaction, so
we cannot determine ethane yield.
[d]
Cyclohexane.
[e]
Ethylbenzene.
Fig. 6. Activity of KN (a) and KW (b) in tandem hydroformylation/hydrogenation of octene-1 in five consequent reactions. Conditions: KN or KW 30 mg, octene-1 0.6
mL (4 mmol), substrate/Rh (mol.) = 670, toluene 2 mL, 80 ◦C, syngas (CO:H2 = 1:3) 6.0 MPa, 5 h.
solubility of propanal in toluene compared to aldehydes produced from
the other unsaturated substrates in Table 6, which leads to complica-
tions in propanal transfer to active sites of the catalyst.
results show that after nine reactions Rh content decreased from 1.93 to
1.13 % wt.
The results of HR-TEM images (Fig. 7) of KN reveal the absence of
rhodium clusters and nanoparticles, which suggests that Rh/amine
complexes are the most probable state of rhodium in KN. This is also
indicated by the fact that Rh and N are evenly distributed over the
sample (Fig. 8). TEM images of KN* are very similar to those of KN, and
no significant difference could be observed. Rhodium does not aggregate
during the catalytic reactions; the elemental distribution (Rh, N, Si)
remains the same. The selected area electron diffraction (SAED) patterns
do not display rings formed by discrete diffraction spots, which may be
indexed as diffraction planes of Rh nanoparticles.
KN and KW were tested in 5 consequent reactions (octene-1 hydro-
formylation/hydrogenation) at 80 ◦C (Fig. 6). In both cases, total
oxygenate yield was consistently high, but alcohol selectivity gradually
decreased. However, for KN this process was slower than for KW.
Hydroformylation regioselectivity remained constant with an n:iso ratio
of about 2.0.
3.3. Study of the changes in KN catalyst structure
K
N and KN* samples were investigated by XPS (Fig. 9). Two rhodium
As KN demonstrated the most stable activity in tandem hydro-
formylation/hydrogenation, we selected it to investigate its structure by
XPS, IR, EDS mapping, and high-resolution TEM. To indicate a possible
change in the structure of KN during the reaction, it was also studied
after nine sequential reactions. We took a bigger amount of catalyst for
this series (120 mg), and the loading of octene-1 and toluene was
enlarged to 2 and 4 mL, respectively (T = 80 ◦C, p = 6.0 MPa (CO:H2 =
1:3), t = 5 h for each reaction). The syngas pressure declined rapidly
with such a big loading of substrate and catalyst, so we introduced
additional portions of syngas (CO:H2 = 1:3) as it was consumed to bring
conditions closer to the standard. After nine reactions, the catalyst’s
weight decreased from 120 to 75 mg because of mechanical losses, but in
the 9th cycle it was still active in both hydroformylation and hydroge-
nation (conv. 98 %, aldehyde select. 56 %, alcohol select. 20 %).
Further, the catalyst after nine cycles will be referred to as KN*. ICP-AES
peaks are observed at binding energies 307.7 eV and 312.4 eV. Ac-
cording to the spectra, rhodium atoms are found in one oxidation state.
Most probably, these bands belong to Rh+1 (literature data: 307.8 eV
and 312.5 eV for Rh3d5/2 and Rh3d3/2, respectively [45]). Binding en-
ergies are also close to those reported in reference [46], where they were
attributed to Rh+1/amine complexes. The X-ray photoelectron spectrum
of KN* demonstrates slight changes in rhodium coordination sphere. The
signals are shifted to higher binding energies compared to the initial KN
spectrum. The binding energies (308.3 eV and 313.0) can still be
attributed to Rh+1 bonded to amino groups of the support [47].
We observe that the yield of alcohols decreases much more readily
than total oxygenate yield, which may indicate a change in the rhodium
microenvironment. In fact, after four reaction cycles (Fig. 6) the per-
formances of KN become closer to those of KR, where amino groups had
8

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
Fig. 7. TEM images of KN (1a-3a), KN* (1b-3b) and SAED pattern of KN (4)*.
Fig. 8. HAADF-STEM images of KN (1a-3a) and KN* (1b-3b) with corresponding element mapping analysis.
9

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
reaches 98 % with 73 % aldehyde selectivity and n:iso ratio of 1.4, which
obviously indicates the presence of rhodium in the solution. Neverthe-
less, no aldehyde hydrogenation is observed. The use of Rh(acac)(CO)2
as a precursor in the homogeneous reaction (Table 4) also leads to al-
dehydes being the only oxygenates produced from octene-1. We also
conducted the same experiment with KR to find out if methylation is
responsible for Rh leaching. The results were close to those for KN:
octene-1 conversion was 98 %, aldehyde selectivity 63 %, n:iso = 1.9, no
alcohols were found. The difference in aldehyde yield led us to the
conclusion that Rh content in the solution after reaction is slightly lower
in case of unmethylated support, so, probably, methylation, though not
causing the leaching, can affect it. It should be mentioned here that KN
initially had contained a little more Rh than KR (Table 3).
We can suggest the different nature of the two stages of the tandem
reaction in the presence of KN. Both rhodium hydrocarbonyls leached to
a liquid phase and rhodium atoms coordinated by amino groups on the
material’s surface can contribute to hydroformylation of olefins [17].
The reversibility of the transfer between these two states cannot be
excluded. However, hydrogenation of aldehydes turns out to be het-
erogeneous process, as it doesn’t take place in the absence of Rh-N
bonding under the investigated conditions (Fig. 10).
4. Conclusions
The materials obtained by the interaction between rhodium pre-
cursor Rh(acac)(CO)2 and surface tertiary amino groups of mesoporous
hybrid organic-inorganic materials (BP-1 and WP-1 modified by NH2-
groups methylation) could be considered as possible non-phosphorous
catalysts for the tandem liquid-phase hydroformylation/hydrogenation
reaction. For the selected catalyst KN high-resolution TEM microscopy
and XPS analysis showed high dispersion of rhodium atoms in oxidation
state +1 bonded with nitrogen both in the fresh and used catalyst. No
huge Rh◦ clusters or nanoparticles are observed on the surface of the
catalyst after 9 cycles (total run time 40 h) in octene-1 hydro-
formylation/hydrogenation. The catalyst remains active in hydro-
formylation with some activity losses in the intermediate aldehyde
hydrogenation step, giving a stable total yield of oxygenates with a drop
in alcohol selectivity. The reason for the drop is leaching, possibly
accompanied by the changes of the microenvironment of Rh atoms, their
rearrangement, and, as a result, less intensive interaction between Rh
and NMe2-fragments compared to the fresh catalyst. We showed that the
range of different unsaturated substrates such as hexene-1, styrene,
cyclohexene could be converted into corresponding alcohols with suf-
ficient yields. In the case of ethylene, only propanal can be achieved
using this protocol. The influence of the main reaction parameters such
as temperature, total pressure, composition of syngas, solvent type, and
Fig. 9. X-ray photoelectron spectrum of KN and KN*.
not been methylated before rhodium was introduced, so rhodium is most
likely attached to secondary and primary nitrogen atoms not effecting its
hydrogenation activity as much as tertiary ones. Possibly, in KN either
rhodium leaches from its complexes with tertiary nitrogen atoms and
after several reactions is only bonded with secondary/primary nitrogen
atoms remain unmodified, or the tertiary nitrogen atoms themselves
undergo some transformations. Several versions of what can happen to
N-containing rhodium hydroformylation catalyst during the reaction are
presented in [25].
Most probably, rhodium is leached into the solution in the form of
carbonyl complexes, which are removed with the liquid phase when the
catalyst is separated from the reaction mixture. The liquid phase ob-
tained by centrifugation after hexene-1 hydroformylation/hydrogena-
tion on KN was found to be active in octene-1 isomerization and
hydroformylation. After 3 h, under general conditions (T = 80 ◦C, p(CO:
H2 = 1:3) = 6.0 MPa, octene-1 0.6 mL) octene-1 conversion in this case
Fig. 10. Proposed pathway of active site formation and catalyst deactivation.
10

D. Gorbunov et al.
Applied Catalysis A, General 623 (2021) 118266
[12] L. Wu, Q. Liu, R. Jackstell, M. Beller, Angew. Chem. Int. Ed. (Eng. Transl.) (53)
(2014) 6310–6320, https://doi.org/10.1002/anie.201400793. Angew. Chem., 126
(2014) 6426–6436.
reaction time in octene-1 hydroformylation/hydrogenation is evaluated.
In 12 h’ reaction at 80 ◦C in toluene at 6.0 MPa syngas (CO:H2 = 1:3)
pressure octene-1 transforms into a mixture of n- and iso-nonanols (n/
iso = 1.1) with 92 % yield. The catalytic system demonstrates high TOF
(hydroformylation TOF = 312 hꢀ 1 at 21 % conv.) and selectivity
compared to analogs described in literature.
[13] D.N. Gorbunov, M.V. Nenasheva, Yu.S. Kardasheva, E.A. Karakhanov, Russ. Chem.
Bull. Int. Ed. (Eng. Transl.) 69 (2020) 625–634, https://doi.org/10.1007/s11172-
020-2810-y. Izv. Akad. Nauk. Ser. Khim., 2020, No. 4, 625–634.
[14] D. Gorbunov, M. Nenasheva, M. Terenina, Yu. Kardasheva, A. Maksimov,
E. Karakhanov, ChemistrySelect 5 (2020) 6407–6414, https://doi.org/10.1002/
slct.202001327.
[15] K. Kaneda, M. Yasumura, M. Hiraki, T. Imanaka, S. Teranishi, Chem. Lett. 10
(1981) 1763–1766, https://doi.org/10.1246/cl.1981.1763.
Authors statement
[16] K. Kaneda, T. Imanaka, S. Teranishi, Chem. Lett. 12 (1983) 1465–1466, https://
doi.org/10.1246/cl.1983.1465.
All authors agree to the content and confirm that the manuscript is
original and not being considered for publication by another journal.
[17] A.T. Jurewicz, L.D. Rollmann, D.D. Whitehurst, in: D. Forster, J.F. Roth (Eds.),
Advances in Chemistry, Vol. 132, American Chemical Society, Washington, DC,
1974, pp. 240–251, https://doi.org/10.1021/ba-1974-0132.ch016.
[18] B. Fell, A. Geurts, Chem. Ing. Technol. 44 (1972) 708–712, https://doi.org/
10.1002/cite.330441104.
CRediT authorship contribution statement
[19] L.L.W. Cheung, G. Vasapollo, H. Alper, Adv. Synth. Catal. 354 (2012) 2019–2022,
https://doi.org/10.1002/adsc.201200053.
Dmitry Gorbunov: Conceptualization, Methodology, Investigation,
Writing - review & editing. Maria Nenasheva: Investigation, Formal
analysis, Writing - original draft, Visualization. Evgeny Naranov: Data
curation, Formal analysis. Anton Maximov: Conceptualization, Super-
vision, Validation. Edward Rosenberg: Conceptualization, Writing -
review & editing. Eduard Karakhanov: Supervision.
[20] S. Fuchs, D. Lichte, M. Dittmar, G. Meier, H. Strutz, A. Behr, A.J. Vorholt,
ChemCatChem 9 (2017) 1436–1441, https://doi.org/10.1002/cctc.201601518.
´
[21] T. Vanbesien, E. Monflier, F. Hapiot, Green Chem. 18 (2016) 6687–6694, https://
doi.org/10.1039/c6gc02706g.
[22] K. Cousin, B. Quienne, J. Pinaud, S. Caillol, E. Monflier, F. Hapiot, Eur. J. Lipid Sci.
Technol. (2020), https://doi.org/10.1002/ejlt.202000158. No. 2000158.
[23] D.N. Gorbunov, A.V. Volkov, Yu.S. Kardasheva, A.L. Maksimov, E.A. Karakhanov,
Petrol. Chem. (Eng. Transl.) 55 (2015) 587–603, https://doi.org/10.1134/
S0965544115080046. Neftekhim. 55 (2015) 443–460.
Declaration of Competing Interest
[24] W.H. Lang, A.T. Jurewicz, W.O. Haag, D.D. Whitehurst, L.D. Rollmann,
J. Organometal. Chem. 134 (1977) 85–94, https://doi.org/10.1016/S0022-328X
(00)93615-5.
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.
[25] D.L. Hunter, S.E. Moore, R.A. Dubois, P.E. Garrou, Appl. Catal. 19 (1985) 275–285,
https://doi.org/10.1016/S0166-9834(00)81750-9.
[26] K. Kaneda, H. Kuwahara, T. Imanaka, J. Mol. Catal. 72 (1992) L27–L30, https://
doi.org/10.1016/0304-5102(92)85004-Y.
Acknowledgments
[27] L. Alvila, T.A. Pakkanen, T.T. Pakkanen, O. Krause, J. Mol. Catal. 71 (1992)
281–290, https://doi.org/10.1016/0304-5102(92)85019-C.
[28] D.P. Zhuchkov, M.V. Nenasheva, M.V. Terenina, Yu.S. Kardasheva, D.N. Gorbunov,
E.A. Karakhanov, Pet. Chem. (Engl. Transl.) 61 (2021) 1–14, https://doi.org/
10.1134/S0965544121010011.
The reported study was financially supported by the Russian Science
◦
Foundation (RSF) grant (project N 20-79-00143).
[29] K.G. Allum, R.D. Hancock, I. V.Howell, R.C. Pitkethly, P.J. Robinson, J. Catal. 43
(1976) 322–330, https://doi.org/10.1016/0021-9517(76)90317-1.
[30] T.O. Mitchell, D.D. Whitehurst, US Pat., 4213921, 1980.
[31] Yu.S. Varshavsky, T.G. Cherkasova, Russ. J. Inorg. Chem. (Engl. Tnansl.) 12 (1967)
1709–1712. Zh. Neorg. Khim., 1 (1967) 1709–1711.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2021.118266.
[32] E. Rosenberg, in: A.S. Abd-El-Aziz, C.E. Carraher Jr., C.U. Pittman Jr., M. Zeldin
(Eds.), Macromolecules Containing Metal and Metal-Like Elements, Vol. 4, John
Wiley & Sons, Inc., Hoboken, New Jersey, 2005, pp. 51–78.
References
[33] J.J. Allen, E. Rosenberg, E. Johnston, C. Hart, ACS Appl. Mater. Interfaces 4 (2012)
1573–1584, https://doi.org/10.1021/am201761m.
¨
[1] R. Franke, D. Selent, A. Borner, Chem. Rev. 112 (2012) 5675–5732, https://doi.
[34] D. Pang, E. Rosenberg, US Pat., 5695882, 1997.
org/10.1021/cr3001803.
[35] R.J. Fischer, E. Rosenberg, US Pat., 6576590, 2003.
[2] D.N. Gorbunov, M.V. Nenasheva, R.P. Matsukevich, M.V. Terenina, F.N. Putilin,
Yu.S. Kardasheva, A.L. Maksimov, E.A. Karakhanov, Petrol. Chem. (Engl. Transl.)
59 (2019) 1009–1016, https://doi.org/10.1134/S0965544119090056. Neftekhim.
59 (2019) 560–568.
[36] R.J. Fischer, E. Rosenberg, US Pat., 7008601, 2006.
[37] E. Rosenberg, D.C. Pang, US Pat., 5997748, 1999.
[38] J. Allen, E. Rosenberg, E. Karakhanov, S.V. Kardashev, A. Maximov,
A. Zolotukhina, Appl. Organometal. Chem. 25 (2011) 245–254, https://doi.org/
10.1002/aoc.1749.
[3] K.U. Künnemann, L. Schurm, D. Lange, T. Seidensticker, S. Tilloy, E. Monflier,
D. Vogt, J.M. Dreimann, Green Chem. 22 (2020) 3809–3819, https://doi.org/
10.1039/D0GC00820F.
[39] E.A. Karakhanov, A.L. Maksimov, A.V. Zolotukhina, S.V. Kardashev, T.
Yu. Filippova, Petrol. Chem. (Engl. Transl.) 52 (2012) 289–298, https://doi.org/
10.1134/S0965544112050052. Neftekhim., 52 (2012) 323–332.
[40] E. Karakhanov, A. Maximov, Yu. Kardasheva, V. Semernina, A. Zolotukhina,
A. Ivanov, G. Abbott, E. Rosenberg, V. Vinokurov, ACS Appl. Mater. Interfaces 6
(2014) 8807–8816, https://doi.org/10.1021/am501528a.
´
[4] K. Cousin, T. Vanbesien, E. Monflier, F. Hapiot, Catal. Commun. 215 (2019) 37–42,
https://doi.org/10.1016/j.catcom.2019.03.018.
[5] Oxo Synthesis, Ullmann’s Encyclopedia of Industrial Chemistry [Online], Wiley-
VCH Verlag GmbH & Co, 2013, https://doi.org/10.1002/14356007.a18_321.pub2.
Posted Jan 15, (Accessed Oct 25, 2020).
[41] D. Gorbunov, D. Safronova, Yu. Kardasheva, A. Maximov, E. Rosenberg,
E. Karakhanov, ACS Appl. Mater. Interfaces 10 (2018) 26566–26575, https://doi.
org/10.1021/acsami.8b02797.
¨
[6] G.M. Torres, R. Frauenlob, R. Franke, A. Borner, Catal. Sci. Technol. 5 (2015)
34–54, https://doi.org/10.1039/c4cy01131g.
¨
[7] C. Rodrigues, F.G. Delolo, J. Norinder, A. Borner, A.L. Bogado, A.A. Batista, J. Mol.
[42] T. Harada, S. Ikeda, Y.H. Ng, T. Sakata, H. Mori, T. Torimoto, M. Matsumura, Adv.
Funct. Mater. 18 (2008) 2190–2196, https://doi.org/10.1002/adfm.200701442.
[43] T. Mizoroki, K. Seki, S. Meguro, A. Ozaki, Bull. Chem. Soc. Jpn. 50 (1977)
2148–2152, https://doi.org/10.1246/bcsj.50.2148.
Catal. A: Chem. 426 (part B) (2017) 586–592, https://doi.org/10.1016/j.
molcata.2016.09.020.
[8] M.R.L. Furst, V. Korkmaz, T. Gaide, T. Seidensticker, A. Behr, A.J. Vorholt,
ChemCatChem 9 (2017) 4319–4323, https://doi.org/10.1002/cctc.201700965.
[9] F. Rodrigues, P. Kucmierczyk, M. Pineiro, R. Jackstell, R. Franke, M. Pereira,
M. Beller, ChemSusChem 11 (2018) 2310–2314, https://doi.org/10.1002/
cssc.201800488.
[44] C.A. Tolman, J.W. Faller, in: L.H. Pignolet (Ed.), Homogeneous Catalysis with
Metal Phosphine Complexes, Plenum Press, New York, 1983, pp. 13–109.
[45] Y. Okamoto, N. Ishida, T. Imanaka, S. Teranishi, J. Catal. 58 (1979) 82–94, https://
doi.org/10.1016/0021-9517(79)90247-1.
[10] C.O. Oseghale, B.M. Mogudi, C.A. Akinnawo, R. Meijboom, Appl. Catal. A 602
(2020), https://doi.org/10.1016/j.apcata.2020.117735. No. 117735.
[11] S.S. Bhagade, S.R. Chaurasia, B.M. Bhanage, Catal. Today 309 (2018) 147–152,
https://doi.org/10.1016/j.cattod.2017.08.022.
[46] P. Hernan, C. Del Pino, E. Ruiz-Hitzky, Chem. Mater. 4 (1992) 49–55, https://doi.
org/10.1021/cm00019a013.
[47] H. Noda, K. Motokura, W.-J. Chun, A. Miyaji, S. Yamaguchi, T. Baba, Catal. Sci.
Technol. 5 (2015) 2714–2727, https://doi.org/10.1039/c5cy00133a.
11