Ethylene hydroformylation in imidazolium-based ionic liquids catalyzed by rhodium-phosphine complexes

Article abstract of DOI:10.1016/j.cattod.2012.06.031

In this research, the catalytic activity of a rhodium-based (Rh) catalyst with imidazolium-based ionic liquids (IBILs) as solvents for ethylene hydroformylation was studied. The structures of IBILs had an important influence on the activity and stability of the Rh catalyst. The IBILs with longer cation side chains, which were the strong steric hindrances around the Rh catalyst, were more unfavorable for the catalytic activity. The turnover frequency (TOF) of the Rh catalyst was 10627 h^-1 when [Bmim][BF₄] was used as solvent. The activity of the Rh complexes in the ionic liquid is better than they do in toluene. We used electrospray ionization mass spectrometry to characterize the catalyst after the reaction and found that [Bmim]⁺ acts as a ligand of the Rh catalyst to form a new active catalytic site [Rh(CO)(PPh₃)₂(Bmim)(BF₄)]⁺ through the coordination of the Rh atom with the imidazole-2-C group of [Bmim][BF ₄], and it was essential for the stabilization of the Rh catalyst and prevented the formation of low-active Rh clusters. In addition, the catalyst recycling test showed that the Rh catalyst could be reused with [Bmim][BF ₄] as solvent without obvious loss of catalytic activity.

Full text of DOI:10.1016/j.cattod.2012.06.031

Catalysis Today 200 (2013) 54–62
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Ethylene hydroformylation in imidazolium-based ionic liquids catalyzed by
rhodium–phosphine complexes
Yanyan Diaoa, Jing Lia, Ling Wanga, Pu Yanga, Ruiyi Yana, Li Jianga,b, Heng Zhanga,c, Suojiang Zhanga,∗
a Beijing Key Laboratory of Ionic Liquids Clean Process, State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing
100190, PR China
b Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China
c School of Chemistry and Chemical Engineering, Qufu Normal University, Qufu 273165, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
In this research, the catalytic activity of a rhodium-based (Rh) catalyst with imidazolium-based ionic
liquids (IBILs) as solvents for ethylene hydroformylation was studied. The structures of IBILs had an
important influence on the activity and stability of the Rh catalyst. The IBILs with longer cation side chains,
which were the strong steric hindrances around the Rh catalyst, were more unfavorable for the catalytic
activity. The turnover frequency (TOF) of the Rh catalyst was 10627 h−1 when [Bmim][BF4] was used as
solvent. The activity of the Rh complexes in the ionic liquid is better than they do in toluene. We used
electrospray ionization mass spectrometry to characterize the catalyst after the reaction and found that
[Bmim]+ acts as a ligand of the Rh catalyst to form a new active catalytic site [Rh(CO)(PPh3)2(Bmim)(BF4)]+
through the coordination of the Rh atom with the imidazole-2-C group of [Bmim][BF4], and it was essential
for the stabilization of the Rh catalyst and prevented the formation of low-active Rh clusters. In addition,
the catalyst recycling test showed that the Rh catalyst could be reused with [Bmim][BF4] as solvent
without obvious loss of catalytic activity.
Received 30 March 2012
Received in revised form 26 June 2012
Accepted 29 June 2012
Available online 13 August 2012
Keywords:
Hydroformylation
Ionic liquids
Ethylene
Rh catalyst
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
CH3
Rh-Cat.
R CH=CH2
+
CO
H2
R CH2CH2CHO +
R CHCHO
+
The hydroformylation reaction (Scheme 1), which was discov-
ered by Roelen in 1938, is a powerful method for functionalizing
R=H, CH3, CnH2n+1
C
C bonds to provide aldehyde and alcohol compounds. It is also
Scheme 1. Hydroformylation reaction of olefins to produce aldehydes.
an important industrial homogeneous catalytic process [1–3]. Thus,
hydroformylation has attracted continuous and increasing aca-
demic and industrial interest. The general metal catalysts used in
the hydroformylation of common olefins are rhodium (Rh) [4–8],
cobalt [9,10] and platinum [11]. The supported metal catalysts
have also been used in an attempt to combine the efficiency of
a homogeneous catalyst with the practical advantages of a het-
erogeneous system [12,13]. Moreover, different reaction media
have been employed for hydroformylation, such as normal organic
solvents [14,15], aqueous media [16], and supercritical CO2 [17].
However, the unsatisfactory catalytic activity, high cost, low ther-
mal stability of catalysts, and the requirement of toxic organic
solvents (DMF or toluene), as well as the difficulty of separation
are still the disadvantages of the process that need to be overcome.
Ionic liquids have variable physicochemical characteristics, neg-
ligible vapor pressure, and structures that can be tailored to
the reaction requirements, which make them possible alterna-
tives to conventional organic solvents. The hydroformylation of
olefins catalyzed by rhodium complexes in ionic liquids has been
reported [18–22]. The first application of ionic liquids in hydro-
formylation was reported by Chauvin et al. [23]. They reported
the biphasic hydroformylation of 1-pentene with (acetylaceto-
nato)dicarbonylrhodium(I) [Rh(CO)2acac] as catalyst in 1-butyl-3-
methylimidazolium hexafluorophosphate ([Bmim][PF6]). For the
Rh-catalyzed hydroformylation of methyl-3-pentenoate, Wasser-
scheid [24] showed that the addition of an ionic liquid significantly
increased the catalyst’s lifetime and overall productivity. The cat-
alyst was stabilized by the non-volatile ionic liquid during the
distillation process. However, no reports have been published on
the analysis of the active Rh catalyst in ionic liquid solution, partic-
ularly the interaction between the ionic liquid and Rh catalyst.
Electrospray ionization mass spectrometry (ESI-MS) is a tech-
nique that allows the ions in solution to be transferred into
the gas phase, where they can be analyzed and eventually
∗
Corresponding author. Tel.: +86 10 8262 7080; fax: +86 10 8262 7080.
E-mail address: sjzhang@home.ipe.ac.cn (S. Zhang).
0920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.06.031

Y. Diao et al. / Catalysis Today 200 (2013) 54–62
55
characterized with minimal fragmentation. Thus, the straight-
PPh
HCHO
3
Rh(PPh3)3Cl
RhCl3·H2O
HRh(CO)(PPh3)3
forward analytical applications of ESI-MS technique to the
characterization of organometallic intermediates have caught the
attention of chemists [25]. ESI-MS is also rapidly becoming the pre-
ferred technique for mechanistic studies and the high-throughput
screening of homogeneous catalysis [26,27].
KOH 80ºC
Scheme 2. Synthesis of the Rh catalyst.
0.053 g (2 mmol) rhodium trichloride 3-hydrate in 4 mL ethanol
was added to a vigorously boiling solution of 0.53 g (20 mmol)
triphenylphosphine in 20 mL of ethanol. After stirring for 15 s,
an aqueous formaldehyde solution (2 mL, 40%, w/v solution) and
a solution of 0.16 g potassium hydroxide in 20 mL hot ethanol
were added rapidly and successively into the above mixture while
stirring. The mixture was heated under reflux for 10 min and then
cooled to room temperature. The bright yellow crystalline product
was filtered, washed with ethanol and water, then were dried and
kept in Schlenk bottle under vacuum conditions before use.
To the best of our knowledge, ionic liquids are mainly used as
solvent in the biphasic hydroformylation of higher olefins to sim-
ple the separation process. To extend the previous work, we chose
the Rh-catalyzed ethylene hydroformylation to demonstrate the
great potential of ionic liquids as solvents for homogeneous cataly-
sis even in cases where the reaction mixture is monophasic. The use
of ionic liquids as solvent significantly enhanced the overall produc-
tivity and the catalyst’s lifetime. Imidazolium-based ionic liquids
(IBILs), particularly the structures of the cations and anions, have
an important influence on the activity and stability of the Rh cata-
lyst. With [Bmim][BF4] as solvent, the effects of parameters, such as
the reaction temperature, pressure, the amount of catalyst used, as
well as the ratio of ligand to Rh catalyst, on ethylene hydroformyla-
tion were also investigated. The catalyst can be reused several times
without additional regeneration and no obvious loss in activity and
selectivity.
Having recognized the pivotal importance of characterizing
organometallic intermediates, we decided to take advantage of the
ability of ESI-MS to transfer ionic species to the gas phase smoothly
and to investigate the structure of the active Rh catalyst in ionic
liquids solution and the interaction between Rh analyst and ionic
liquids. We also aim to probe the effect of the ionic liquids on the
activity and stability of the Rh catalyst.
2.3. Characterization of ionic liquids and catalyst
ESI-MS experiments were performed in positive ion mode on
a Bruker micrOTOF-Q II mass spectrometer, with m/z from 50 to
2500. The infusion flow rate of 180 ␮L h−1 was maintained by a
syringe pump. Basic vacuum ESI condition at 4.0 × 10−7 mbar was
employed. The fourier transform infrared (FT-IR) spectra in the
regions (400–4000 cm−1) were recorded via the KBr pellet tech-
nique using 0.5 mg of sample and 100 mg of KBr on a Thermo Nicolet
380 spectrometer (Thermo Electron Company). Elemental analysis
was performed on a vario EL element analyser.
2.4. Hydroformylation procedure
2. Experimental
The experimental apparatu used for the both batch and
recycling reaction is described in Scheme 3. Batch experiments
were conducted in a flat-bottomed, magnetically stirred 150 mL
steel autoclave equipped with a 120 mL teflon tube. The cat-
alyst HRh(CO)(PPh3)3 (0.0270 g, 0.03 mmol), and the ligand
triphenylphosphine (0.0774 g, 0.3 mmol; Rh:PPh3 = 1:10) were dis-
solved in a solvent of propionaldehyde (5 mL) and IBILs (5 mL), and
then added into the 120 mL teflon tube in the stainless steel auto-
clave. The reactor was vacuumed first, then purged with feed gas
(C2H4:CO:H2 = 1:1:1), and finally, placed under continuous pres-
sure of 2 MPa. After attaining 2 MPa pressure, the reactor was
heated to 100 ◦C using an oil bath and magnetically stirred at
800 rpm. The reaction was performed at a constant pressure of
2 MPa by feeding mixed gas. The total amount of feed gas was mea-
sured using a mass flow meter. The reaction was conducted for 2 h
and before it was allowed to cool to room temperature, and then,
the excess gas was slowly vented.
2.1. Reagents
All reagents and gases used in this study were either of chemical
or analytical grades. The rhodium chloride trihydrate (Sinopharm;
99%) was supplied by the Beijing Cuibolin Non-ferrous Metal Tech-
nology Development Center. The propionaldehyde (Sinopharm;
>98%), triphenylphosphine (Sinopharm; CP; abbreviated as PPh3),
1-methylimidazole (Acros; 99%), 1-propylimidazole (Acros; 99%)
and 1-butylimidazole (Acros; 99%) were produced by the Beijing
Chemical Plant. The hydrogen (99.99%), carbon monoxide (99.9%),
and ethylene (99.99%) were supplied by the Beijing Analytical
Instrument Factory. All reagents were used as received.
2.2. Preparation of ionic liquids and catalyst
2.2.1. Preparation of ionic liquids
The reaction mixtures were analyzed via GC (Agilent 6890,
equipped with a DB-624 capillary column and an FID detector)
and GC–MS. The catalytic activities were expressed in terms of
ethylene conversion (C, %), propionaldehyde selectivity (S, %), and
turnover frequency (TOF, h−1), which were determined via GC using
n-hexane as internal standard. These parameters are defined as:
A series of IBILs were prepared using standard methods [28–36].
Briefly, N-methylimidazole was reacted with a little excess of 1-
bromobutane in a round-bottom flask at room temperature for 5 h
to produce [Bmim]Br. The halide-based ionic liquids were purified
by recrystallization and dried overnight under vacuum conditions
at room temperature. For [Bmim][BF4] or [Bmim][PF6], [Bmim]Br
was ion-exchanged with equal mole of sodium tetrafluoroborate
(NaBF4) or sodium hexafluorophosphate (NaPF6) in aqueous solu-
tion. The obtained IBILs were extracted using dichloromethane, and
the amount of halide impurities in the ionic liquids were qualita-
tively determined by adding silver nitrate. Then IBILs were dried
under vacuum conditions at 343 K for two to three days before use.
22400 · np
Q1 − Q2
np
Q1 − Q2
C =
× 100%
S =
× 100%
TOF =
Q1
ncat · t
where Q1 is the volume of ethylene into the reactor (mL); Q2 is the
volume of ethylene unreacted (mL); np is the mole number of pro-
pionaldehyde produced during reaction (mol); TOF is the Turnover
frequency (h−1); t is the Reaction time (h); ncat is the mole number
of catalyst added in the reaction (mol). The mole number of gas was
calculated using standard conditions.
2.2.2. Preparation of catalyst
The
catalyst
HRh(CO)(PPh)3
[Carbonylhydridotris
Recycling experiments were similar to batch experiments. When
the reaction was finished after 2 h and cooled to room tempera-
ture, the vacuum pump was opened and propionaldehyde product
was collected in a cooling batch. The catalyst HRh(CO)(PPh3)3,
(triphenylphosphine)rhodium(I)] was synthesized as follows
(Scheme 2) [37]. The experiment was performed using standard
Schlenk techniques under nitrogen atmosphere. A solution of

56
Y. Diao et al. / Catalysis Today 200 (2013) 54–62
Intens.
694.7
+MS, 5.0-12.1s #(4-11)
HRh(CO)(PPh3) 3
6
x10
655.0785
0.8
1477.2
1921.0
2036.3
1086.6
742.4
0.6
0.4
0.2
0.0
513.6
1432.1
PPh3
917.1712
279.0943
500
1000
1500
m/z
3500
3000
2500
2000
1500
1000
500
Fig. 2. ESI-MS analysis of catalyst HRh(CO)(PPh3)3.
Wavenumber(cm-1)
Fig. 1. FT-IR spectra of PPh3 and HRh(CO)(PPh3)3.
the Rh-H stretching frequency centered at about 2036.3 cm−1 [37],
and nine typical peaks centered at 3449.5, 3053.6, 1583.7, 1477.2,
1432.1, 1086.6, 742.4, 694.7, and 513.6 cm−1 associated with the
stretching and bending frequencies of triphenylphosphine.
ionic liquids, PPh3 and part of the propionaldehyde remained in
the autoclave, to which fresh feeding gas was added. Then, a new
catalytic reaction was started. The same procedure was repeated
for several cycles. The amount of feed gas was measured using
a mass flow meter in every reaction, and the catalytic activity of
HRh(CO)(PPh3)3 in every reaction was expressed by the amount of
feed gas.
The theoretical contents of different elements of the
HRh(CO)(PPh3)3 catalyst were 71.90% C, 5.05% H, 1.74% O,
meanwhile, the experimental results of the element analysis
were 71.80% C, 5.12% H, 1.83% O. From the results of the FT-IR
and element analysis, the structure of the synthesized catalyst
was HRh(CO)(PPh3)3. In methanol solution, the structure of
HRh(CO)(PPh3)3 was analyzed via ESI-MS in positive ion mode
(Fig. 2). The main peak at 655.0785, is attributed to the active
species [Rh(CO)(PPh3)2]+, and the peak at 917.1712, is attributed to
the active species [Rh(CO)(PPh3)3]+. Fig. 3 shows the comparison
of the experimental isotopic distribution of the mixture and the
theoretical isotopic distribution as simulated by the Bruker Xmass
6. 1. 2 software for the peak at 655.0785 and 917.1712. The appear-
ance of [Rh(CO)(PPh3)2]+ was mainly caused by the dissociation of
[Rh(CO)(PPh3)3]+ in the solution, as shown in Scheme 4.
3. Results and discussion
3.1. Characterization of catalyst
To confirm the structure of the Rh catalyst, FT-IR spectroscopic
studies were conducted. Fig. 1 shows the IR spectra of the catalyst
and triphenylphosphine. The spectrum of the catalyst displayed a
very strong peak corresponding to the CO stretching frequency cen-
tered at about 1921.0 cm−1, a typical strong peak corresponding to
Scheme 3. Experimental apparatus used for both batch and recycling hydroformylation.

Y. Diao et al. / Catalysis Today 200 (2013) 54–62
57
Intens.
x10
Intens.
x10
+MS, 5.0-12.1s #(4-11)
+MS, 7.0-11.0s #(6-10)
6
5
917.1712
655.0785
Experomental
Experomental
0.8
0.6
0.4
0.2
0.0
918.1751
2
656.0835
1
919.1768
657.0870
920.1787
0
6
5
x10
x10
C 55 H 45 O P 3 Rh ,917.17
C 37 H 30 O P 2 Rh ,655.08
655.0821
917.1733
3
2
1
0
Theoretical
Theoretical
0.8
0.6
0.4
0.2
0.0
918.1766
656.0855
919.1800
657.0889
657
920.1833
920
655
656
658
m/z
917
918
919
921
m/z
Fig. 3. Comparison of experimental isotopic distribution and theoretical isotopic distribution simulated by Bruker Xmass 6.1.2 software at the peaks of 655.0785 and 917.1712.
carbon chain length in the N1 position have adverse effects on
catalytic activity. Similarly, when [Bmim][BF4], [Bpim][BF4] and
[Bbim][BF4] were used as solvents, the length of carbon chains in
the N3 position of the imidazolium cation increased from 1 to 4,
the TOF of the catalyst decreased from 10627 h−1 to 8237 h−1. This
result shows that the increased lengths of the chains in the N1 or
N3 position could more or less reduce the catalytic activity of the
Rh catalyst.
PPh
3
[Rh(CO)(PPh3)2]+
[Rh(CO)(PPh3)3]+
Scheme 4. The equilibrium of [Rh(CO)(PPh3)3]+ and [Rh(CO)(PPh3)2]+ in methanol
solvent.
3.2. Activities of Rh catalyst with various ionic liquids as solvents
on ethylene hydroformylation
The anions of the ionic liquids also had an important influ-
ence on the activity of the Rh catalyst. The TOF of 7912 h−1 in
the [Bmim][PF6] solution was lower than that in the [Bmim][BF4]
solution. Some reports [38,39] on the hydroformylation of olefins
The activities of the HRh(CO)(PPh3)3 catalyst in different sol-
vents during ethylene hydroformylation were examined. The
corresponding results are presented in Table 1. The catalytic activ-
ity of the Rh catalyst in the [Bmim][BF4] solution (Table 1, entry
3) was better than that in the traditional toluene and propionalde-
hyde solutions (Table 1, entries 1and 2). Table 1 showed that the
structures of the ionic liquids had a significant influence on the
activity of the Rh catalyst during ethylene hydroformylation. Based
on the activity of the Rh catalyst with [Bmim][BF4], [Hmim][BF4]
and [Omim][BF4] as solvents, the carbon numbers of the chain in
the N1 position (the mark of C and N in the imidazole ring of ionic
liquids as shown in Scheme 5) of the imidazolium-based cation
increased from 4 to 8, the TOF of the catalyst decreased from 10627
h−1 to 1123 h−1. This result indicates that the lengthening of the
−
in [Bmim][PF6] have shown that the PF6 group could be disso-
ciated weakly, which is possibly why the TOF of the Rh catalyst
in the [Bmim][PF6] solution was slightly lower than that in the
[Bmim][BF4] solution. The Rh catalyst had no activity in the
[Bmim][OTs], [Bmim][OAc], [Bmim][SCN] and [Bmim][HSO4] solu-
tions.
3.3. Analysis of Rh catalyst after ethylene hydroformylation by
ESI-MS
After ethylene hydroformylation, the existential state of the
Rh catalyst in toluene and ionic liquid solution were all analyzed
by ESI-MS. As shown in Fig. 4, there are four typical peaks in
toluene centered at 627.1042, 655.0925, 889.1886, and 917.1809,
which could be attributed to the active catalytic sites [Rh(PPh3)2]+,
[Rh(CO)(PPh3)2]+, [Rh(PPh3)3]+, and [Rh(CO)(PPh3)3]+ respectively.
Scheme 6 shows that these species are in equilibriums between
Table 1
Effect of various ionic liquids as solvent for ethylene hydroformylation.a
Entry
Solvents
Conversion (%)
Selectivity (%)
TOF (h−1
)
1
2
3
4
5
6
7
8
Toluene
98.90
97.64
98.19
95.22
71.41
97.18
97.12
96.92
99.10
77.80
9856
7558
Propionaldehyde
[Bmim][BF4]
[Hmim][BF4]
[Omim][BF4]
[Bpim][BF4]
[Bbim][BF4]
[Bmim][PF6]
[Bmim][OTs]
[Bmim][OAc]
[Bmim][SCN]
[Bmim][HSO4]
98.14
94.79
92.64
92.32
92.26
92.85
–
–
–
–
10627
5411
1123
8450
8237
7912
–
H
(2)C
H
C
m
b
+
N(1)
m+2
9
–
C H
n
n+2
10
11
12
–
–
–
–
–
–
(3)N
(4)C
C(5)
a
Conditions: 3 mmol L−1, HRh(CO)(PPh3)3, Rh: PPh3 = 1:10, 100 ◦C, 2.0 MPa, C2H4:
H2: CO = 1: 1: 1, 800 rpm, 2 h.
b
H
H
No
reaction,
[Bmim] = 1-butyl-3-methylimidazolium;
[Hmim] = 1-
[Omim] = 1-octyl-3-methylimidazolium;
hexyl-3-methylimidazolium;
[Bpim] = 1-butyl-3-propylimidazolium; [Bbim] = 1-butyl-3- butylimidazolium;
[OTs] = Tosylate; [OAc] = Acetate.
Scheme 5. The structure of imidazolium-based cation and the mark of C and N in
the imidazole ring.

58
Y. Diao et al. / Catalysis Today 200 (2013) 54–62
Fig. 4. ESI-MS catalysis of active Rh catalyst structure in toluene solution after ethylene hydroformylation.
CO
HRh(CO)(PPh3)2
HRh(PPh3)2
PPh3
PPh3
CO
HRh(PPh3)3
HRh(CO)(PPh3)3
Scheme 6. The equilibrium of Rh catalyst on ethylene hydroformylation.
them, and they are all active catalysts. As shown in Fig. 5 and Fig. 6,
the Rh catalyst also existed as several states in the [Bmim][BF4],
notably at the peak of 793.1966, which was attributed to the new
active catalytic site [Rh(Bmim)(CO)(PPh3)2]+, and at the peak of
365.2457, 591.3706, 817.5091, and 1043.6361, which could be
attributed to the clusters of [(Bmim)2(BF4)]+, [(Bmim)3(BF4)2]+,
[(Bmim)4(BF4)3]+, and [(Bmim)5(BF4)4]+ respectively.
Fig. 5. ESI-MS catalysis of active Rh catalyst structure in [Bmim][BF4] after ethylene
hydroformylation.
with a pair of electrons that can be used as ligands of the Rh cat-
alyst. In the Figs. 7 and 8, in the [Omim][BF4], [Omim]+ can also
act as a ligand of the Rh catalyst to form a new active catalytic
site [Rh(Omim)(CO)(PPh3)2]+ at 849.2590. Another active catalytic
site [Rh(CO)(PPh3)2]+ was also observed at 655.0823. The peak at
477.3761 is attributed to the cluster of [(Omim)2(BF4)]+.
Based on the analysis, [Bmim]+ acts as a ligand of the Rh cat-
alyst to form a new active catalytic site for hydroformylation. In
addition, previous studies [20] have reported on the analysis of
imidazole-based heterocyclic fragment ionic liquids used in the
rhodium catalyzed hydroformylation process via 1H NMR. The H
atoms of imidazole-2 are unstable, and can easily release free H+.
The remaining imidazole ring constitutes a heterocyclic carbine
In the catalytic reaction, the acyl-rhodium complex reacts with
di-hydrogen is a rate-limiting step. The cations of the IBILs can be
Intens.
x10
Intens.
+MS, 4.2-9.2s #(4-9)
+MS, 4.2-9.2s #(4-9)
4
793.1969
1.5
1.0
0.5
365.2507
2000
Experomental
Experomental
1500
794.2002
1000
364.2536
500
366.2527
795.2032
0.0
4
0
x10
C 45 H 44 N 2 O P 2 Rh ,793.20
C 16 H 30 B F 4 N 4 ,365.25
2500
1.50
1.25
1.00
0.75
0.50
0.25
0.00
793.1978
Theoretical
2000
Theoretical
365.2497
1500
794.2012
1000
364.2530
364
500
366.2527
795.2045
0
363
365
366
367
m/z
792
793
794
795
796
797
m/z
Fig. 6. Comparison of experimental isotopic distribution and theoretical isotopic distribution simulated by Bruker Xmass 6.1.2 software at the peaks of 365.2507 and 793.1969.

Y. Diao et al. / Catalysis Today 200 (2013) 54–62
59
h−1 to about 1.0 × 104 h−1. As Rh concentration further increased
from 3.0 mmol L−1 to 3.5 mmol L−1, the TOF and conversion rate
remained almost constant. Within the range of Rh concentrations,
the selectivity remained almost constant. Therefore, the suitable
Intens.
x10
+MS, 4.2-10.2s #(4-10)
5
477.3761
1.25
1.00
0.75
0.50
0.25
0.00
Rh concentration was set to 3.0 mmol L−1
.
3.4.2. Effect of PPh3/Rh ratio
Fig. 10 shows the conversion, selectivity and TOF of a series of
PPh3/Rh ratios on ethylene hydroformylation. The PPh3/Rh ratio
was found to have an important effect on catalyst activity. The low-
est catalytic activity was observed with the lowest PPh3/Rh ratio in
the test range. As the PPh3/Rh ratio increased from 1 to 70, the TOF
849.2590
655.0823
increased from 2000 h−1 to about 1.0×104
h
−1. In addition, the cat-
849.2590
800
alytic activity also increased; whereas the selectivity was remained
above 96% in all test PPh3/Rh ratios. As the PPh3/Rh ratio continued
to increase to 90, the change in TOF was not obvious. Therefore, the
PPh3/Rh ratio of 70 was chosen as the most appropriate ratio in this
test range.
400
600
1000
1200 m/z
Fig. 7. ESI-MS analysis of active Rh catalyst structure in [Omim][BF4] after catalytic
reaction.
associated with the [Rh-H] species [40,41]. These associations made
the [Rh-H] species more active and enhanced their reactions with
dihydrogen, thus increasing the catalytic activity. However, the
experimental results showed that an increase of the carbon chain
length, regardless of whether it is in the N1 or N3 position, would
lead to a decline in catalytic activity. This finding indicates that the
imidazolium cation around the rhodium center speeds up the reac-
tion of Rh-H. Meanwhile, the steric effect of the imidazolium cation
also has an effect on the association and dissociation of the Rh–PPh3
complex.
3.4.3. Effect of temperature
The Effect of temperature on ethylene hydroformylation is
shown in Fig. 11. The reaction was very sensitive to temperature. As
the temperature increased from 80 ◦C to 110 ◦C, the TOF increased
remarkably, from below 4000 h−1 to about 1.3 × 104 h−1. As the
temperature further increased from 110 ◦C up to 130 ◦C, the TOF
decreased to 7000 h−1, As the temperature increased from 80 ◦C
to 110 ◦C, the increasing of thermal energy makes the molecules
of materials move faster, and increase the collision probability
of molecules and catalysts, thus enhance catalytic activity. How-
ever, at temperatures over 110 ◦C, the rhodium complex began to
become unstable. Thus the catalytic activity continued to decline
as the temperature kept rising. The decline in selectivity began
at 100 ◦C rather than at 110 ◦C. This phenomenon showed that a
great number of side reactions were produced when temperature
increased to above 100 ◦C. Hence, the most appropriate reaction
temperature in the test range should be 100 ◦C.
3.4. Influence of reaction conditions on ethylene
hydroformylation in [Bmim][BF4]
The reaction conditions generally have important influence on
the activity of catalyst in the homogeneous reaction. Thus, we
investigated the TOF of Rh catalyst on the different reaction con-
ditions (catalyst concentration, PPh3/Rh ratio, temperature and
pressure) using [Bmim][BF4] as solvent.
3.4.4. Effect of pressure
3.4.1. Effect of Rh concentration
Fig. 12 presents the conversion, selectivity and TOF of a series
of pressure conditions on ethylene hydroformylation. From 0.5 MPa
to 1.5 MPa, the TOF rapidly increased from 1500 h−1 to about 11500
h−1 and then moderately decreased at pressures ranging from
1.5 MPa to 2.5 MPa until the TOF became constant. The conversion
Fig. 9 shows the effect of a series of Rh concentration on ethy-
lene hydroformylation. As the Rh concentration increased from
1.0 mmol L−1 to 3.0 mmol L−1, the number of active rhodium
centers also increased and the TOF steadily increased from 2000
Intens.
+MS, 4.2-10.2s #(4-10)
Intens.
x10
+MS, 4.2-10.2s #(4-10)
5
849.2590
477.3761
Experomental
1.25
5000
Experomental
1.00
0.75
0.50
0.25
4000
850.2627
3000
2000
478.3783
476.3786
1000
851.2633
479.3798
0
0.00
x10
5
C 24 H 46 B F 4 N 4 ,477.38
C 49 H 52 N 2 O P 2 Rh ,849.26
6000
849.2604
477.3751
Theoretical
Theoretical
1.25
1.00
0.75
0.50
0.25
0.00
5000
4000
3000
2000
1000
0
850.2638
478.3780
476.3782
851.2671
852.2705
479.3813
479 480
476
477
478
m/z
849
850
851
852
853 m/z
Fig. 8. Comparison of experimental isotopic distribution of and theoretical isotopic distribution simulated by Bruker Xmass 6.1.2 software at the peaks of 477.3761and
849.2590.

60
Y. Diao et al. / Catalysis Today 200 (2013) 54–62
1.2x104
1.0x104
100
90
80
70
60
50
8.0x103
Conversion
6.0x103
Selectivity
4.0x103
2.0x103
1.0
1.5
2.0
2.5
3.0
3.5
1.0
1.5
2.0
2.5
3.0
3.5
-1
-1
Rh Concentration /( mmol·L )
Rh Concentration /(mmol·L )
Conditions: PPh3/Rh = 10, 100 °C, 2.0 MPa, C2H4: H2: CO= 1: 1: 1, 2 h.
Fig. 9. Effect of Rh concentration on ethylene hydroformylation.
4
100
95
90
85
80
75
2.0x10
4
1.5x10
4
1.0x10
Conversion
Selectivity
3
5.0x10
0.0
10
1
5
30
50
70
90
1
10
5
30
PPh3/Rh
50
70
90
PPh3/Rh
Conditions: 3 mmol·L-1 HRh(CO)(PPh3)3, 100 °C, 2.0 MPa, C2H4: H2: CO= 1: 1: 1, 2 h.
Fig. 10. Effect of PPh3/Rh ratio on ethylene hydroformylation.
and selectivity both remained above 95% in all conditions of pres-
sure, and the highest conversion rate and selectivity were observed
at 1.5 MPa. From 0.5 MPa to 1.5 MPa, the increased feed gas pressure
caused the increased solubility of the gas in the liquid phase, which
strengthened the mass transfer between the gas-liquid states and
thus enhanced the TOF of the catalyst. From 1.5 MPa to 2.5 MPa,
the decrease in TOF was caused by the competition of excess CO
and PPh3 in coordination with the Rh catalyst. When the pres-
sure was higher than 2.5 MPa, a delicate balance was established
between the feed gas concentrations in the liquid phase and the
competitive coordination, leading to the observed stable catalytic
activity. Therefore, the pressure of 1.5 MPa would be the most suit-
able choice for the system.
3.5. Recycling experiments of catalyst
Using ethylene, CO and H2 as substrates, we conducted recy-
cling experiments to examine the recyclability of HRh(CO)(PPh3)3
in the [Bmim][BF4] solution using the similar industrial production
conditions of toluene as solvent (100 ◦C and 2.0 MPa). The catalytic
activity of HRh(CO)(PPh3)3 was expressed by the amount of feed
gas. As shown in Fig. 13, no obvious decrease in the gas flow for six
repeated runs indicated the high stability of the Rh catalyst. In addi-
tion, the ESI-MS analysis results (Fig. 14) proved that the Rh catalyst
in the [Bmim][BF4] solution did not change after six runs and the
active catalytic sites of catalyst were [Rh(CO)(PPh3)2]+ (655.0818)
and [Rh(Bmim)(CO)(PPh3)2]+ (793.1871).
1.4x104
100
98
1.2x104
1.0x104
8.0x103
6.0x103
4.0x103
2.0x103
96
94
Conversion
92
Selectivity
90
80
90
100
110
120
130
80
90
100
110
120
130
Temperature /ºC
Temperature /ºC
Conditions: 3 mmol·L-1 HRh(CO)(PPh3)3, PPh3/Rh = 10, 2.0 MPa, C2H4: H2: CO= 1: 1: 1, 2 h.
Fig. 11. Effect of temperature on ethylene hydroformylation.

Y. Diao et al. / Catalysis Today 200 (2013) 54–62
61
100
95
1.2x104
1.0x104
8.0x103
6.0x103
4.0x103
Conversion
Selectivity
90
2.0x103
0.0
85
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Pressure / MPa
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Pressure /MPa
-1
Conditions: 3 mmol·L HRh(CO)(PPh3)3, PPh3/Rh = 10, 100 °C, C2H4: H2: CO= 1: 1: 1, 2 h.
Fig. 12. Effect of pressure on ethylene hydroformylation.
interacts as a ligand with Rh catalyst to form a new active catalytic
site for hydroformylation, which is essential for the stabilization
of the Rh catalyst and prevention of the formation of low active Rh
clusters. Thus, a high TOF (10627 h−1) could be achieved under mild
conditions. In addition, the catalyst recycling test showed that the
Rh catalyst could be reused with [Bmim][BF4] as solvent without
obvious loss of catalytic activity. The organic green process pre-
sented here could have potential applications in industry due to its
easy product separation and high efficiency.
20000
15000
10000
5000
0
Acknowledgements
This work is financially supported by General Program Youth of
National Natural Science Foundation of China (21006106), National
Basic Research Program of China (973 Program) (2009CB219901),
and Key Program of National Natural Science Foundation of China
(21036007).
1
2
3
4
5
6
Recycle times
Conditions: PPh3/Rh = 10, 100 °C, 2.0 MPa, C2H4: H2: CO= 1: 1: 1, 2 h.
Fig. 13. Recycling experiments of ethylene hydroformylation in [Bmim]BF4.
References
Intens.
x105
+MS, 5.2-8.2s #(5-8)
[1] C. Claver, P.W.N.M. van, Leeuwen (Eds.), Rhodium Catalyzed Hydroformyla-
tion;, vol. 1, Kluwer Academic, Dordrecht, The Netherlands, 2000.
[2] E. A. V. Brewester, R. L. Pruett, Cyclic Hydroformylation Process, US patent
4247486, 1981.
[3] J. Falbe, New syntheses with carbon monoxide, Springe-Verlag, Berlin Heidel-
berg, 1985, 24-216.
[4] J.M. Praetorius, M.W. Kotyk, J.D. Webb, R.Y. Wang, C.M. Crudden,
Organometallics 26 (2007) 1057–1061.
[5] Y.J. Yan, X.W. Zhang, X.M. Zhang, Journal of the American Chemical Society 128
(2006) 16058–16061.
[6] A.F. Peixoto, M.M. Pereira, A.M.S. Silva, C.M. Foca, J. Carles Bayón, M.S.M.
Moreno, A.M. Beja, J.A. Paixão, M.R. Silva, Journal of Molecular Catalysis A:
Chemical 275 (2007) 121–129.
[7] R.M. Deshpande, B.M. Bhanage, S.S. Divekar, S. Kanagasabapathy, R.V. Chaud-
hari, Industrial and Engineering Chemistry Research 37 (1998) 2391–2396.
[8] M. Kuil, T. Soltner, P.W.N.M. van Leeuwen, J.N.H. Reek, Journal of the American
Chemical Society 128 (2006) 11344–11345.
[9] P.N. Bungu, S. Otto, Dalton Transaction (2007) 2876–2884.
[10] R.J. Klingler, M.J. Chen, J.W. Rathke, K.W. Kramarz, Organometallics 26 (2007)
352–357.
[11] R. van Duren, J.I. van der Vlugt, H. Kooijman, A.L. Spek, D. Vogt, Dalton Trans-
action (2007) 1053–1059.
365.2538
5
4
3
2
1
0
655.0818
793.1871
591.3795
1043.6382
1269.7581
1494.8915
400
600
800
1000
1200
1400
m/z
[12] Q.R. Peng, D.H. He, Catalysis Letters 115 (2007) 19–22.
[13] S. Christine Bourque, H. Alper, L.E. Manzer Prabhat Arya, Journal of the American
Chemical Society 122 (2000) 956–957.
Fig. 14. The ESI-MS analysis of active Rh catalyst structure in [Bmim][BF4] after
recycling reaction.
[14] M.L. Clarke, G.J. Roff, Green Chemistry 9 (2007) 792–796.
[15] H.J.V. Barros, C.C. Guimaraes, E.N. dos Santos, E.V. Gusevskaya, Catalysis Com-
munications 8 (2007) 747–750.
4. Conclusions
[16] S. Paganelli, M. Marchetti, M. Bianchin, C. Bertucci, Journal of Molecular Catal-
ysis A: Chemical 269 (2007) 234–239.
[17] J. Ke, B.X. Han, M.W. George, H.K. Yan, M. Poliakoff, Journal of the American
Chemical Society 123 (2001) 3661–3670.
[18] M. Haumann, A. Riisager, Chemical Reviews 108 (2008) 1474–1497.
[19] D. Bradley, G. Williams, M. Ajam, Organometallics 26 (2007) 4692–4695.
[20] J.D. Scholten, J. Dupont. Organometallics 37 (2008) 4439–4442.
[21] P. Wasserscheid, H. Waffenschmidt, P. Machnitzki, K.W. Kottsieper, O. Stelzer,
Chemical Communications (2001) 451–452.
IBILs, which are used as solvents on ethylene hydroformylation,
have an important influence on the activity and stability of the Rh
catalyst. The increased length of the chains in the N1 or N3 position
of the imidazole ring reduced the catalytic activity of the catalyst,
and the anions of ionic liquids also had significant influence on
the activity of the catalyst. In the [Bmim][BF4] solution, [Bmim]+

62
Y. Diao et al. / Catalysis Today 200 (2013) 54–62
[22] F. Faure, H. OlivierBourbigou, D. Commereuc, L. Saussine, Chemical Communi-
cations (2001) 1360–1361.
[31] S. Forsyth, J. Golding, D.R. MacFarlane, M. Forsyth, Electrochimica Acta 46
(2001) 1753–1757.
[23] Y. Chauvin, L. Mussmann, H. Olivier, Angew Chemie 34 (1996) 2698–
2700.
[32] D.R. MacFarlane, J. Golding, S. Forsyth, M. Forsyth, G.B. Deacon, Chemical Com-
munications (2001) 1430–1431.
[24] H. Waffenschmidt, P. Wasserscheid, D. Vogt, W. Keim, Journal of Catalysis 186
(1999) 481–484.
[25] Z. Takats, S.C. Nanita, R.G. Cooks, Angew Chemie 115 (2003) 3645–3647.
[26] A.A. Sabino, A.H.L. Machado, C.R.D. Correia, M.N. Eberlin, Angew Chemie 43
(2004) 2514–2518.
[33] A. Paul, P.K. Mandal, A. Samanta, Chemical Physics Letters 402 (2005) 375–379.
[34] S.V. Dzyuba, R.A. Bartsch, Chemical Communications (2001) 1466–1467.
[35] P. Wasserscheid, M. Sesing, W. Korth, Green Chemistry 4 (2002) 134–138.
[36] S.G. Zhang, Q.L. Zhang, Z.C. Zhang, Industrial and Engineering Chemistry
Research 43 (2004) 614–622.
[27] R. Qian, H. Guo, Y.X. Liao, Y.L. Guo, S.M. Ma, Angew Chemie 44 (2005)
4771–4774.
[28] S. Chun, S.V. Dzyuba, R.A. Bartsch, Analytical Chemistry 73 (2001) 3737–3741.
[29] J.D. Holbrey, K.R. Seddon, Journal of the Chemical Society, Dalton Transactions
(1999) 2133–2140.
[37] N. Ahmad, J.J. Levison, S.D. Robinson, Inorganic Syntheses 15 (1974) 59–60.
[38] B.A. Omotowa, J.M. Shreeve, Organometallics 23 (2004) 783–791.
[39] Q. Lin, PhD thesis, Sichuan University, 2006.
[40] J. Blum, A. Rosenfeld, F. Gelman, H. Schumann, D. Avnir, Journal of Molecular
Catalysis A: Chemical 146 (1999) 117–122.
[30] Q.G. Zhang, J.Z. Yang, X.M. Lu, J.S. Gui, Z. Huang, Fluid Phase Equilibria 226
(2004) 207–211.
[41] C. Zuccaccia, G. Bellachioma, G. Cardaci, A. Macchioni, Journal of the American
Chemical Society 123 (2001) 11020–11028.