Butadiene hydroformylation to adipaldehyde with Rh-based catalysts: Insights into ligand effects

Article abstract of DOI:10.1016/j.mcat.2019.110721

Rh-catalyzed hydroformylation of butadiene to adipaldehyde is a promising alternative route for producing valuable C₆ compounds such as adipic acid and hexamethylenediamine. Fundamental insights into reaction pathways, aimed at enhancing adipaldehyde yield, were obtained from temporal concentration profiles and in situ ReactIR studies of butadiene hydroformylation on Rh complexes at 80 °C and 14 bar syngas (molar CO/H₂ = 1) pressure in a batch reactor. Specifically, the effects of operating conditions and eight commercially available ligands on activity and selectivity were systematically investigated. It was found that the adipaldehyde selectivity is independent of the ligand/Rh ratio, rhodium concentration, butadiene concentration and syngas pressure, but significantly dependent on the type of ligand used. For example, while the DIOP ligand provided an adipaldehyde yield of ～40% with butadiene as a substrate, the 6-DPPon ligand gave a maximum adipaldehyde yield of ～93% with 4-pentenal as substrate. Furthermore, the adipaldehyde selectivity correlates well with the natural bite angle of the various ligands. ReactIR studies suggest that the preferential formation of the stable rhodium η³-crotyl complex with the various Rh complexes may be the main reason for the low adipaldehyde selectivity.

Full text of DOI:10.1016/j.mcat.2019.110721

Molecular Catalysis xxx (xxxx) xxxx
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Butadiene hydroformylation to adipaldehyde with Rh-based catalysts:
Insights into ligand effects
Si-min Yu , William K. Snavely , Raghunath V. Chaudharia,b, Bala Subramaniama,b,
a
a
*
a
Center for Environmentally Beneficial Catalysis, University of Kansas, Lawrence, KS, 66047, United States
Department of Chemical and Petroleum Engineering, University of Kansas, Lawrence, KS 66045, United States
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
Butadiene
Hydroformylation
Rh complexes
Temporal concentration profiles
ReactIR
Rh-catalyzed hydroformylation of butadiene to adipaldehyde is a promising alternative route for producing
valuable C6 compounds such as adipic acid and hexamethylenediamine. Fundamental insights into reaction
pathways, aimed at enhancing adipaldehyde yield, were obtained from temporal concentration profiles and in
situ ReactIR studies of butadiene hydroformylation on Rh complexes at 80 °C and 14 bar syngas (molar CO/
2
H = 1) pressure in a batch reactor. Specifically, the effects of operating conditions and eight commercially
available ligands on activity and selectivity were systematically investigated. It was found that the adipaldehyde
selectivity is independent of the ligand/Rh ratio, rhodium concentration, butadiene concentration and syngas
pressure, but significantly dependent on the type of ligand used. For example, while the DIOP ligand provided an
adipaldehyde yield of ∼40% with butadiene as a substrate, the 6-DPPon ligand gave a maximum adipaldehyde
yield of ∼93% with 4-pentenal as substrate. Furthermore, the adipaldehyde selectivity correlates well with the
natural bite angle of the various ligands. ReactIR studies suggest that the preferential formation of the stable
3
rhodium η -crotyl complex with the various Rh complexes may be the main reason for the low adipaldehyde
selectivity.
1. Introduction
hydroformylation to adipaldehyde. The use of either Co/monopho-
sphine [10] or Rh/monophosphine catalysts [6,9] requires severe re-
Adipaldehyde is an attractive precursor for synthesizing C
6
com-
action conditions (120−175 °C, 200−300 bar), resulting in complex
aldehyde products with low C -dialdehyde selectivity (< 20%). Rho-
pounds such as adipic acid and hexamethylenediamine, important in-
dustrial monomers for the production of polyamides (e.g., Nylon-6,6),
polyesters and polyurethanes [1,2]. The current commercial process for
adipic acid manufacture is based on multistep oxidation of cyclohexane
using nitric acid as the oxidant. The over oxidation of cyclohexane and
6
dium/diphosphine catalysts provide better selectivities and activities
under mild conditions (p < 20 bar, T < 100 °C) [4,7,8,11]. Ohgomori
et al. [4] reported 37% adipaldehyde selectivity with a catalyst com-
posed of Rh and the commercially available DIOP ligand. Smith et al.
[5] reported a higher adipaldehyde selectivity of 50% with congeners of
the bisphosphine triptyphos as ligands. By using the concept of iso-
merizing hydroformylation, Mormul et al. achieved 73% selectivity
toward adipaldehyde bis-acetal derivative [20]. However, the adi-
paldehyde yield (50 %) is still low for practical viability.
Compared to terminal mono-alkene, conjugated dienes as substrates
are very resistant to hydroformylation and show atypical reaction be-
havior resulting in much slower reaction rates and lower regioselec-
tivity [21]. Even though the hydroformylation of several conjugated
dienes (such as butadiene [22], isoprene [23], 1,3-pentadiene [24] and
myrcene [25,26]) has been reported, systematic investigations of the
correlation between phosphine structure and catalytic performance are
rare. Previous literature on the subject of butadiene hydroformylation
2
release of large amounts of N O emissions are major environmental
problems associated with the conventional technology [3]. Hexam-
ethylenediamine is mainly produced by Ni-catalyzed hydrocyanation
and subsequent hydrogenation, but it requires handling of relatively
expensive and extremely toxic HCN gas. However, adipaldehyde as a
starting material provides a promising alternative route for the pro-
duction of adipic acid by oxidation and hexamethylenediamine by re-
ductive amination (Scheme 1).
The selective hydroformylation of butadiene is an atom economical
process for producing adipaldehyde [4–15]. With increased availability
of inexpensive natural gas liquids, the use of butadiene as the feedstock
is receiving renewed interest. Among homogeneous catalysts [16–19],
transition-metal-complexes are the most reported for butadiene
⁎
Corresponding author at: Center for Environmentally Beneficial Catalysis, The University of Kansas, Lawrence, KS, 66047, United States.
E-mail address: bsubramaniam@ku.edu (B. Subramaniam).
https://doi.org/10.1016/j.mcat.2019.110721
Received 15 October 2019; Received in revised form 13 November 2019; Accepted 15 November 2019
2468-8231/ © 2019 Elsevier B.V. All rights reserved.
Please cite this article as: Si-min Yu, et al., Molecular Catalysis, https://doi.org/10.1016/j.mcat.2019.110721

S.-m. Yu, et al.
Molecular Catalysis xxx (xxxx) xxxx
Scheme 1. Butadiene as building block for commodity chemicals.
2
. Experimental
.1. Materials
CO) Rh(acac) and phosphine ligands L1-L8 were purchased from
2
(
2
commercial suppliers and stored in an argon atmosphere in a glovebox
before usage. Butadiene (chemically pure) and syngas (99.99% purity,
molar H /CO ratio of 1:1) were supplied by Matheson in cylinders. The
2
toluene solvent (anhydrous, 99.8%) was purchased from Alfa Aesar and
stored under an atmosphere of argon.
2.2. General procedure for hydroformylation experiments
The butadiene hydroformylation experiments were carried out in a
0 mL Parr reactor equipped with a six-port valve for periodic sampling
5
of reactor contents (see Fig. S1 for a schematic of the unit). (CO)
2
Rh
(
acac) (10 mg, 0.039 mmol) and ligand (0.116 mmol) were dissolved in
13 mL toluene, and 0.1 mL decane was added as an internal standard.
After purging the mixture with Ar (high purity) thrice, the reactor was
heated to 80 °C, and then 0.5 mL pure butadiene (metered with an ISCO
pump at −6 °C/∼12 bar) was charged into the reactor followed by 2
mL of toluene from a HPLC pump. Once the reactor temperature sta-
2
bilized at a predetermined value, syngas (molar CO/H = 1) was
charged into the reactor from an external reservoir to the desired re-
actor pressure. Following this step, the stirrer was promptly started and
set at 1000 rpm to initiate the hydroformylation reaction. Samples from
the liquid phase were collected by a six-port valve equipped with a
Scheme 2. Overall reaction stoichiometry for formation of adipaldehyde and
other byproducts from butadiene hydroformylation.
2
50 μL sample loop which was flushed thrice with N . The first sample
that typically showed the maximum 4-pentenal selectivity was collected
immediately (< 1 min) after syngas was charged into the reactor and
the stirring was commenced. The concentrations of the various re-
actants and products in the reaction mixture were analyzed by GC/FID
analysis. The continuous decrease in the external reservoir pressure
following reaction initiation is due to syngas consumption and was
monitored on a LabVIEW® data acquisition and control system. The
overall carbon balance was estimated based on end-of-run analysis of
the hydroformylation products in the liquid phase. To get an accurate
estimate of butadiene remaining in the reactor following a batch run,
the reactor was sufficiently cooled (to −20 °C) to totally condense the
butadiene prior to sampling the liquid phase.
typically reports product selectivity at the end of fixed-time batch runs
lasting several hours. Given that butadiene conversion to adipaldehyde
requires two sequential hydroformylation steps with various competing
reactions (Scheme 2), temporal product evolution profiles are needed to
better discern the relative contributions of the two steps toward adi-
paldehyde yield. In this work, we measured product evolution profiles
by periodic sampling to monitor the reaction progress. This allowed us
to systematically investigate the effects of reaction parameters and li-
gands on the two hydroformylation steps. We found that for butadiene
hydroformylation with a given ligand, the adipaldehyde selectivity was
independent of the ligand/Rh ratio, Rh concentration, substrate con-
centration and syngas pressure. Rh/DIOP catalyst provided the best 4-
pentenal selectivity (48%) at short reaction times (i.e., low butadiene
conversion), and 40% adipaldehyde selectivity at the end of a run. In
contrast, for 4-pentenal hydroformylation, the 6-DPPon ligand per-
formed much better than the DIOP ligand, providing nearly 93% adi-
paldehyde selectivity. These insights provide new leads for catalyst
selection and process optimization aimed at maximizing adipaldehyde
selectivity.
The chemoselectivity is defined as moles of aldehydes formed re-
lative to moles of total products formed.
n(aldehydes formed)
n(total products)
Chemoselectivity (%)=
× 100
(1)
The regioselectivity (n/i) is defined as the molar ratio of linear
dialdehyde (adipaldehyde) to branched dialdehydes (2-methylpenta-
nedial) in the products.
n(adipaldehyde)
n(2-methylpentanedial)
Regioselectivity(n/i) =
(2)
Turnover frequency is estimated from the slope of the linear portion
of the products’ concentration vs time plots (Fig. S3) as follows.
2

S.-m. Yu, et al.
Molecular Catalysis xxx (xxxx) xxxx
systematically investigate the effects of reaction operating conditions
n(total products)
−
1
TOF (h ) =
n(Rh) × time
(3)
and ligand on each of the sequential hydroformylation steps (i) and (ii).
The carbon balance for the hydroformylation of either butadiene or
pent-4-enal is estimated as follows
3.2. Reaction parameter effects
nC(aldehydes + butadiene or 4-pentenal)
The effects of the ligand/Rh ratio, rhodium concentration, buta-
product
Carbon balance =
diene concentration and syngas pressure are summarized in Table 1.
Details of the syngas consumption profiles and temporal selectivity
profiles corresponding to the various operating conditions are pre-
sented in the Supplementary Materials (Fig. S4 and Fig. S5). DIOP/Rh
ratios were varied from 1–3 (entries 1–3, Table 1). At a molar [DIOP/
Rh] = 1, the butadiene hydroformylation rate was lower compared to
higher ratios (1.5–3). Temporal concentration profiles reveal that 3-
pentenal (2 + 3) was apparently transformed to pentanal (4) and 2-
methylpentanedial (6) beyond 80 min (Fig. S5). This could be due to
degradation of the phosphine ligand [21,27], resulting in the formation
of different phosphines with lower selectivity toward adipaldehyde. In
the presence of excess DIOP ligand (entries 2 and 3, Table 1), both the
hydrogenation and hydroformylation reactions involving 3-pentenal
were suppressed, resulting in an increased adipaldehyde selectivity of
40%, which is consistent with literature values [4]. Increasing the
nC(CO) + [nC(butadiene or 4-pentenal)]
(4)
feed
nC(CO)=[n(syngas consumption from
reservoir)-n(hydrogenation products)] × 0.5
(5)
Clearly, the C balance will be 100% for total chemoselectivity to
aldehydes. Any deviation is attributed to the formation of hydrogena-
tion products. Details of carbon balance estimation for three temporal
runs are shown in Tables S1-S3.
2.3. GC method
Gas chromatographic analysis was performed using an Agilent
Technologies 7890A GC system equipped with
a 30 m × 320
μm × 0.25 μm HP-5 column. The He flow rate was kept at 0.8 std mL/
min. The column temperature was initially held at 35 °C for 8 min, then
ramped at 20 °C /min to 90 °C and held for 3 min. This was followed by
another ramp of 30 °C /min to 200 °C, where the temperature was held
for 5 min. Details of calibration for quantifying the product species may
be found in the Supplementary Materials (Table S4).
(CO) Rh(acac) concentration from 0.024 to 0.08 mmol (entries 5 and 4,
2
Table 1) enhances the rate (See also Fig. S4). At higher (CO) Rh(acac)
2
concentration (entry 4, Table 1), the maximum 4-pentenal selectivity
was slightly increased to 48%. The higher (CO) Rh(acac) concentration
2
also increases pentanal (4) and 2-methylpentanedial (6) formation, but
the adipaldehyde selectivity (37 %) remains relatively constant. Simi-
larly, the adipaldehyde selectivity was unaffected with increase in ei-
ther the butadiene concentration from 1.1 to 11.4 mmol (entries 6 and
7, Table 1) or the syngas pressure from 7 to 24 bar (entries 8 and 9,
Table 1). It is noteworthy however that, at lower butadiene con-
centration and syngas pressure (entries 7 and 9), the initial 4-pentenal
selectivity was more than 60 %. However, end-of-run analysis showed
that 3-pentenal was the main product in all cases and the overall adi-
paldehyde selectivity was ∼40 %. The higher 4-pentenal selectivities
2
.4. In situ ReactIR experiments
Reactions for in situ infrared spectroscopic analysis were performed
in Mettler Toledo ReactIR 15. The reactor schematic is essentially si-
milar to the one shown in Fig. S1 with the stirred reactor unit being
replaced by another unit fitted with a ReactIR probe. Approximately, 77
mg (0.29 mmol) (CO)
autoclave reactor. The solution was heated to 80 °C at a stirrer speed of
000 rpm. Once the temperature reached 80 °C, the reactor was pres-
2
Rh(acac) was dissolved in 13 mL toluene in an
3
1
(62–65 %) are caused by the formation of stable rhodium η -crotyl
surized with 7 bar syngas and the measurement was started with 2
scans at 1 min intervals. After allowing 1 h for formation of the catalyst
precursor, 0.5 mL butadiene and 2 mL toluene were introduced into the
reactor by a HPLC pump. Syngas consumption from an external re-
servoir was monitored on a LabVIEW® data acquisition and control
system.
complex which does not favor 3-pentenal formation at low butadiene
conversion. Importantly, the adipaldehyde selectivity is more or less
independent of the ligand/Rh ratio, Rh concentration, butadiene con-
centration and syngas pressure.
3.3. Ligand effects
3
. Results and discussion
A series of commercially available bidentate phosphine and phos-
phite ligands L1-L8 (Fig. 2) were tested for butadiene hydroformyla-
tion. The reaction was carried out at 80 °C and 14 bar syngas with molar
3.1. Temporal product evolution profiles
[
2 2
CO/H ] ratio of 1 using 0.04 mmol (CO) Rh(acac) with molar [P/Rh]
Fig. 1 shows the temporal syngas consumption and product se-
ratio of 6. The product selectivities and TOF are summarized in Table 2.
The syngas consumption profiles and temporal selectivity profiles ob-
tained with the eight ligands are provided in the Supplementary Ma-
terials (Fig. S6 and Fig. S7). In general, monoaldehydes (4-pentenal and
3-pentenal) were the primary intermediate products. At the end of 260
min batch run, the 4-pentenal was totally consumed to form adi-
paldehyde.
The observed selectivities vary dramatically depending on the li-
gand used. The bis(diphenylphosphino) alkane ligands L1-L3, show
progressively increasing 4-pentenal selectivity (from 14–39%) and
adipaldehyde selectivity (from 1–30%). Ligands L3-L5 (Fig. 2), which
are reported to be effective for butadiene hydroformylation [4,7], show
high initial 4-pentenal selectivity (39–46 %) as well as high adipalde-
hyde selectivity (30–39%). For ligands L7 and L8, the maximum 4-
pentenal selectivity decreased. At the end of the 260 min run, 3-pen-
tenal was the major product with selectivity ranging from 73–82%. The
adipaldehyde selectivities were correspondingly much lower (7 % and
16 %). Ligand L6, which can display typical bidentate ligand behavior
by self-assembly through hydrogen bonding [28], showed moderate
lectivity profiles during butadiene hydroformylation using Rh/DIOP
complex as a catalyst. As shown in Scheme 2, the first hydroformylation
step rapidly produces 4-pentenal (1) and trans/cis-3-pentenal (2 + 3).
The maximum 4-pentenal selectivity being approximately 48%. During
the first 100 min, the 4-pentenal selectivity decays steadily and is
completely consumed. The adipaldehyde (7) selectivity rises during this
period reaching approximately 40% and remains fairly constant for the
remaining duration of the batch run. Beyond 100 min, the 3-pentenal
was further hydroformylated to 2-ethylbutanedial (5) and 2-methyl-
pentanedial (6), resulting in a lower n/i ratio. Simultaneously, the 3-
pentenal was isomerized to conjugated 2-pentenal, which is rapidly
transformed into pentanal (4) by hydrogenation. At longer reaction
times, pentanal and 2-methylpentanedial were the main products. Bu-
tadiene conversion to adipaldehyde requires the following two steps to
occur preferentially: (i) butadiene (conjugated alkene) hydroformyla-
tion to monoaldehyde 4-pentenal via 1,2-addition; and (ii) 4-pentenal
(
terminal alkene) hydroformylation to adipaldehyde. In order to im-
prove the overall selectivity of the desired adipaldehdye, we sought to
3

S.-m. Yu, et al.
Molecular Catalysis xxx (xxxx) xxxx
Fig. 1. Typical temporal profiles of syngas consumption and selectivity during butadiene hydroformylation with Rh/DIOP complex at 80 °C. Reaction conditions: 5.7
mmol butadiene, 0.04 mmol (CO) Rh(acac), molar [DIOP/Rh] ratio = 3, syngas pressure =14 bar; molar [H /CO] ratio = 1, 15 mL toluene, 1000 rpm.
2 2
Table 1
Hydroformylation of butadiene catalyzed by (CO)
C, 260 min, 1000 rpm.
2 2 2
Rh(acac) with DIOP. Reaction conditions: 0.04 mmol (CO) Rh(acac), molar [H /CO] ratio = 1, 15 mL toluene, 80
°
Entry
DIOP/Rh
Butadiene, mmol
P, bar
Maximum 4-pentenal selectivity, %
End-of-run selectivities, %
TOFc h−1
2
+ 3
4
6
7
1
2
3
1.0
1.5
3.0
1.5
5.0
3.0
3.0
3.0
3.0
5.7
5.7
5.7
5.7
5.7
11.4
1.1
5.7
5.7
14
14
14
14
14
14
14
24
7
47
47
46
48
44
48
65
45
62
19
43
45
33
50
45
42
47
43
24
8
6
13
4
6
7
4
8
14
8
7
10
6
7
7
7
7
31
39
39
37
38
40
37
39
38
87
122
121
115
121
176
29
a
4
5
b
6
7
8
9
123
52
a
0
0
.08 mmol (CO)
.024 mmol (CO)
2
Rh(acac).
Rh(acac).
b
c
2
Estimated using Eq. (3)].
selectivity towards 4-pentenal (24 %) and adipaldehyde (22 %). The
maximum 4-pentenal selectivity followed an identical order: DPPE <
DPPP < DPPB < DIOP > 6-DPPon > Xantphos > Biphephos. DIOP
and syngas (molar [H
2
/CO] = 1) pressure of 14 bar using 0.02 mmol
(CO) Rh(acac) with molar [P/Rh] ratio of 6. The results are shown in
2
Table 3. The major products were adipaldehyde (7) and branched 2-
methylpentanedial (6) followed by 3-pentenal (2 + 3) and pentanal
(4).
(
L5) provided the best selectivity toward 4-pentenal (46 %) and adi-
paldehyde (39 %) as well as the highest activity (TOF).
For ligands L1 and L2, the adipaldehyde selectivity ranged from
3
7–46 %, and branched 2-methylpentanedial (6) was the favored pro-
3
.4. 4-pentenal hydroformylation
duct. Ligands L3-L5 showed progressively higher selectivity (84–91 %)
towards the linear dialdehyde with moderate increases in the reaction
rate. In sharp contrast to butadiene hydroformylation, the L6-L8 ligands
The foregoing results prompted us to investigate ligand effects on 4-
pentenal hydroformylation as well. Ligands L1-L8 were tested at 80 °C
Fig. 2. Ligands tested in the hydroformylation of butadiene or 4-pentenal.
4

S.-m. Yu, et al.
Molecular Catalysis xxx (xxxx) xxxx
Table 2
Ligand effects for butadiene hydroformylation. Reaction conditions: 5.7 mmol butadiene, 0.04 mmol (CO)
2
Rh(acac), molar [P/Rh] = 6, syngas pressure =14 bar;
molar [H
2
/CO] ratio = 1, 15 mL toluene, 80 °C, 260 min, 1000 rpm.
Entry
Ligand
Natural bite angle, °
Maximum 4-pentenal selectivity, %
End-of-run selectivities, %
TOFa h−1
1
2 + 3
4
6
7
n/i
L1
L2
L3
L4
L5
L6
L7
L8
DPPE
DPPP
DPPB
DPPF
85
91
98
96
98
–
14
19
39
42
46
24
20
11
0
3
0
0
0
0
0
6
10
70
49
48
42
43
73
82
81
10
6
9
6
25
7
5
4
7
9
9
7
6
2
0
1
4
0.3
0.6
3.3
3.4
5.6
3.7
8.0
–
59
44
93
96
116
34
39
10
30
31
39
22
16
7
DIOP
6-DPPon
Xantphos
Biphephos
111
120
a
Estimated using Eq. (3).
Table 3
Ligand effects on 4-pentenal hydroformylation. Reaction conditions: 2 mmol 4-
pentenal, 0.02 mmol (CO) Rh(acac), molar [P/Rh] ratio = 6, syngas pressure
14 bar; molar [H /CO] ratio = 1, 15 mL toluene, 80 °C, 80 min, 1000 rpm.
Ohgomori et al. also reported that diphosphine ligands with natural
bite angle of 102-113° provided 37% adipaldehyde selectivity from
butadiene hydroformylation [4]. Smith et al. reported increased adi-
paldehyde selectivity (50%) during butadiene hydroformylation with
Rh complexes using chelating bisphosphate ligands [5]. To better
characterize the structures of transient intermediate species, Ir analogs
of the Rh complexes were investigated with bite angles of the ee and ea-
coordinated isomers being 105.5° and 99.0°, respectively [38]. Based on
our results and those reported in the literature, it follows that when
employing Rh catalysts with a single ligand for butadiene hydro-
formylation, the natural bite angle should be close to 100° in order to
maximize adipaldehyde selectivity. The fact that the 4-pentenal hy-
droformylation requires a different ligand to maximize adipaldehyde
selectivity provides the motivation to develop a clear understanding of
how the ligand structure affects the first hydroformylation step in se-
lectively forming 4-pentenal.
2
=
2
TOFa h−1
Entry Ligand
Natural bite,
End-of-run selectivity %
°
2
+ 3
4
6
7
n/i
L1
L2
L3
L4
L5
L6
L7
L8
DPPE
DPPP
DPPB
DPPF
85
91
98
96
98
–
3
0
0
0
0
0
1
3
0
0
0
0
0
0
2
8
45
53
16
12
9
7
3
1
37
46
84
88
91
93
92
87
0.8
0.9
5.3
7.3
10.1
13.3
30.7
87.0
68
20
105
78
198
435
130
4546
DIOP
6-DPPon
Xantphos
Biphephos
111
120
a
Estimated using Eq. (3).
displayed much higher adipaldehyde selectivities (87–93 %) during 4-
pentenal hydroformylation. The 6-DPPon ligand (L6) showed the best
adipaldehyde selectivity (93 %) while the Biphephos ligand (L8) pro-
vided remarkably high reaction rate with a n/i ratio of 87. In sharp
contrast, ligand L8 showed a rather low activity for butadiene hydro-
formylation (Table 2). The n/i ratio observed with various ligands for 4-
pentenal hydroformylation ranked as follows: DPPE < DPPP <
DPPB < DIOP < 6-DPPon < Xantphos < Biphephos.
The foregoing results reveal markedly different dependence on the
various ligands tested, and suggest that the activity and selectivity to-
wards the desired products result from the two sequential hydro-
formylation steps, viz., butadiene to 4-pentenal and 4-pentenal to adi-
paldehyde. As inferred from Table 3, the adipaldehyde selectivities
correlate well with natural bite angle of the ligands. The DPPE (L1) and
DPPP (L2) ligands with bite angles near 85° [29] showed the lowest
selectivities towards the desired products for both the hydroformylation
steps (Tables 2 and 3), thereby resulting in the lowest adipaldehyde
selectivity. In contrast, for the first hydroformylation step, the DPPE
ligand favors 3-pentenal [4,7] which is readily transformed to pentanal
and the branched dialdehyde via hydrogenation and hydroformylation,
respectively, in the second step (Tables 2 and 3). van Leeuwen et al.
also reported high pentanal selectivity (90%) when using rhodium with
DPPE ligand [7]. While the L6 and L8 ligands [28,30,31] with large bite
angles ranging from 111-120° [32,33] are highly selective (87–93 %)
towards the linear dialdehyde during 4-pentenal hydroformylation
3
.5. ReactIR experiments
We employed in-situ ReactIR to investigate intermediate products
formed during butadiene hydroformylation with DIOP/Rh catalysts
Fig. 3). The full IR spectra and the carbonyl region are shown in the
Supplementary Materials (Fig. S8). The catalyst was prepared under 7
bar syngas from the precursors, 20 mmol/L (CO) Rh(acac) and excess
(
2
of DIOP (molar [DIOP/Rh] = 1.5) at 80 °C. The typical spectrum for the
hydrido dicabonyl complex (I) was observed as shown in Fig. 3. The
four carbonyl bands (at
assigned to the two trigonal bipyramidal isomers with DIOP ligand in
an axial-equatorial position (ae) or in equatorial-equatorial position
∼
−1
v
= 2041, 1995, 1976 and 1950 cm ) were
(
ee) (Fig. 3a) [34,35]. After allowing 1 h for the catalyst precursors and
syngas to react, butadiene was added, resulting in the immediate for-
mation of new bands (1947 cm ) in the carbonyl region of the IR
spectrum. This band is assigned to the η crotyl complex (II) (Fig. 3b)
-1
3
[
36]. Upon consumption of butadiene, the complex (I) was regenerated
3
(
Fig. 3c and 3d). The η -crotyl complex is a precursor to 3-pentenal
formation, which is converted to either pentanal by hydrogenation or
the branched dialdehyde by hydroformylation. The preferential for-
mation of the stable rhodium η -crotyl complex thus appears to be the
main reason for the relatively low adipaldehyde selectivity observed
3
3
during butadiene hydroformylation. Stable rhodium η -crotyl com-
plexes are also observed when using other ligands as shown in Figure
S9. This explains why the adipaldehyde selectivity is more or less in-
dependent of the ligand/Rh ratio, Rh concentration, butadiene con-
centration and syngas pressure [37].
(
Table 3), the maximum 4-pentenal selectivity when using butadiene as
a feed is quite low (11–24 %). In comparison, the L3-L5 ligands with
similar bite angle near 100° [29] provided higher 4-pentenal selectivity
for the first step compared with other ligands (Table 2) but moderate
adipaldehyde selectivity for the second step (Table 3). Thus, L3-L5
provide the best combination of selectivities towards the desired pro-
ducts from the two steps resulting in the highest overall selectivity to-
wards adipaldehyde.
4
. Conclusions
Temporal concentration profiles provide fresh insights into the ef-
fects of ligands and reaction parameters on adipaldehyde selectivity
during butadiene hydroformylation on using Rh catalysts. The major
5

S.-m. Yu, et al.
Molecular Catalysis xxx (xxxx) xxxx
Fig. 3. In-situ IR spectra from butadiene hydroformylation with DIOP/Rh catalysts. (CO)
2
Rh(acac) and DIOP dissolved in toluene at 80 °C as background. (a) after
2
adding 7 bar of CO/H at 80 °C for 1 h; (b) immediately after adding butadiene; (c) after 26 min of hydroformylation; (d) after 50 min of hydroformylation.
determinant of adipaldehyde selectivity during the two-step butadiene
hydroformylation appears to be the ligand structure. Of the eight bi-
phosphine ligands tested, the DIOP ligand was preferred for the first
hydroformylation step providing a maximum 4-pentenal selectivity of
approximately 48 %, while the 6-DPPon ligand showed the best per-
formance for the 4-pentenal hydroformylation step with adipaldehyde
selectivity exceeding 93 %. The observed selectivity trends are con-
sistent with previously reported correlations between ligand bite angles
and product selectivity during Rh-catalyzed hydroformylation of ole-
fins. Our finding that the two sequential butadiene hydroformylation
steps require different ligands to maximize adipaldehyde selectivity
suggests an opportunity to rationally design an optimum ligand that
simultaneously maximizes the linear aldehyde selectivity during both
steps.
Germany, 2010.
[
[
[
2] R. Franke, D. Selent, A. Börner, Chem. Rev. 112 (2012) 5675–5732.
3] S. van de Vyver, Y. Roman-Leshkov, Catal. Sci. Technol. 3 (2013) 1465–1479.
4] Y. Ohgomori, N. Suzuki, N. Sumitani, J. Mol. Catal. A Chem. 133 (1998) 289–291.
[5] S.E. Smith, T. Rosendahl, P. Hofmann, Organometallics 30 (2011) 3643–3651.
[
[
[
6] H. Bahrmann, B. Fell, J. Mol. Catal. 8 (1980) 329–337.
7] P.W.N.M. van Leeuwen, C.F. Roobeek, J. Mol. Catal. 31 (1985) 345–353.
8] T. Horiuchi, T. Ohta, E. Shirakawa, K. Nozaki, H. Takaya, Tetrahedron 53 (1997)
7795–7804.
[9] B. Fell, W. Rupilius, Tetrahedron Lett. 10 (1969) 2721–2723.
10] H. Adkins, J.L.R. Williams, J. Org. Chem. 17 (1952) 980–987.
11] S. Bertozzi, N. Campigli, G. Vitulli, R. Lazzaroni, P. Salvadori, J. Organomet. Chem.
487 (1995) 41–45.
[
[
[12] B. Fell, H. Bahrmann, J. Mol. Catal. 2 (1977) 211–218.
[
[
[
13] T. Maji, C.H. Mendis, W.H. Thompson, J.A. Tunge, J. Mol. Catal. A Chem. 424
2016) 145–152.
(
14] N. Herrmann, D. Vogelsang, A. Behr, T. Seidensticker, ChemCatChem 10 (2018)
5342–5365.
15] M.J. Tenorio, R.V. Chaudhari, B. Subramaniam, Ind. Eng. Chem. Res. (2019),
https://doi.org/10.1021/acs.iecr.9b05184.
16] S. Rej, N. Chatani, Angew. Chem. Int. Ed. 58 (2019) 8304–8329.
17] N. Chatani, Bull. Chem. Soc. Jpn. 91 (2018) 211–222.
18] T. Ohkuma, N. Kurono, N. Arai, Bull. Chem. Soc. Jpn. 92 (2019) 475–504.
19] X. Zhao, Y. Yang, R. Zhu, C. Liu, D. Zhang, Mol. Catal. 471 (2019) 77–84.
20] J. Mormul, J. Breitenfeld, O. Trapp, R. Paciello, T. Schaub, P. Hofmann, ACS Catal.
6 (2016) 2802–2810.
CRediT authorship contribution statement
[
[
[
[
[
Si-min Yu: Methodology, Data curation, Writing - original draft,
Visualization, Investigation. William K. Snavely: Methodology, Formal
analysis, Writing - review & editing. Raghunath V. Chaudhari:
[
[
[
21] C. Claver, P.W.N.M. van Leeuwen, Rhodium Catalyzed Hydroformylation, Kluwer
Academic Publishers, Dordrecht, 2000.
22] S. Schmidt, E. Baráth, C. Larcher, T. Rosendahl, P. Hofmann, Organometallics 34
(2015) 841–847.
Methodology, Supervision, Writing
Subramaniam: Conceptualization, Methodology, Supervision, Writing
review & editing.
-
review
&
editing. Bala
-
23] H.J.V. Barros, C.C. Guimarães, E.N. dos Santos, E.V. Gusevskaya, Organometallics
2
6 (2007) 2211–2218.
Declaration of Competing Interest
[
[
24] P. Neubert, S. Fuchs, A. Behr, Green Chem. 17 (2015) 4045–4052.
25] C.M. Foca, H.J.V. Barros, E.N. dos Santos, E.V. Gusevskaya, J. Carles Bayon, New J.
Chem. 27 (2003) 533–539.
26] H.J.V. Barros, J.G. da Silva, C.C. Guimarães, E.N. dos Santos, E.V. Gusevskaya,
Organometallics 27 (2008) 4523–4531.
27] P.W.N.M. van Leeuwen, Appl. Catal. A Gen. 212 (2001) 61–81.
28] B. Breit, W. Seiche, J. Am. Chem. Soc. 125 (2003) 6608–6609.
29] P. Dierkes, P.W.N.M. van Leeuwen, J. Chem. Soc., Dalton Trans. (1999) 1519–1530.
30] W. Seiche, A. Schuschkowski, B. Breit, Adv. Synth. Catal. 347 (2005) 1488–1494.
31] J. Mormul, M. Mulzer, T. Rosendahl, F. Rominger, M. Limbach, P. Hofmann,
Organometallics 34 (2015) 4102–4108.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
[
[
[
[
[
[
Acknowledgment
The authors gratefully acknowledge support for this work by a grant
from the joint National Science Foundation and Environmental
Protection Agency program Networks for Sustainable Material
Synthesis and Design program (NSF-EPA 1339661).
[
[
[
32] M. Kranenburg, Y.E.M. van der Burgt, P.C.J. Kamer, P.W.N.M. van Leeuwen,
K. Goubitz, J. Fraanje, Organometallics 14 (1995) 3081–3089.
33] C.P. Casey, G.T. Whiteker, M.G. Melville, L.M. Petrovich, J.A. Gavney, D.R. Powell,
J. Am. Chem. Soc. 114 (1992) 5535–5543.
34] L.A. van der Veen, M.D.K. Boele, F.R. Bregman, P.C.J. Kamer, P.W.N.M. van
Leeuwen, K. Goubitz, J. Fraanje, H. Schenk, C. Bo, J. Am. Chem. Soc. 120 (1998)
Appendix A. Supplementary data
1
1616–11626.
[
[
[
35] S. Schmidt, G. Abkai, T. Rosendahl, F. Rominger, P. Hofmann, Organometallics 32
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.110721.
(2013) 1044–1052.
36] S. Schmidt, E. Barath, T. Prommnitz, T. Rosendahl, F. Rominger, P. Hofmann,
Organometallics 33 (2014) 6018–6022.
37] S. Schmidt, P. Deglmann, P. Hofmann, ACS Catal. 4 (2014) 3593–3604.
References
[38] G. Abkai, S. Schmidt, T. Rosendahl, F. Rominger, P. Hofmann, Organometallics 33
2014) 3212–3214.
(
[1] H.-J. Arpe, K. Weissermel, Industrial-Organic Chemistry, Wiley-VCH, Weinheim,
6