Rhodium catalysts bearing mixed thioether-phosphine ligands for carbonylation of methanol

Article abstract of DOI:10.1246/bcsj.82.1009

Two kinds of thioether-trialkylphosphine ligands and their Rh complexes were synthesized and applied to the carbonylation of methanol to aceticacid to reveal that one of the two complexes showed higher activity compared to Rh catalyst without any ligands.

Full text of DOI:10.1246/bcsj.82.1009

Short Articles
Bull. Chem. Soc. Jpn. Vol. 82, No. 8, 1009–1011 (2009) 1009
Rh cat.
CH3OH
+
CO
CH3CO2H
CH3I
Rhodium Catalysts Bearing Mixed
Thioether-Phosphine Ligands for
Carbonylation of Methanol
CO
CO
I
I
–CH3COI
Rh
Ken Sakakibara and Kyoko Nozaki*
1
CH3
CO
CH3
I
CO
CO
I
I
Department of Chemistry and Biotechnology,
Graduate School of Engineering, The University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656
Rh
Rh
I
CO
OC
I
I
Received March 23, 2009; E-mail: nozaki@chembio.t.u-
tokyo.ac.jp
4
2
CH3
CO
I
I
CO
Rh
Two kinds of thioether-trialkylphosphine ligands and
their Rh complexes were synthesized and applied to the
carbonylation of methanol to acetic acid to reveal that one of
the two complexes showed higher activity compared to Rh
catalyst without any ligands.
CO
I
3
CH3CO2H
CH3COI
CH3OH
CH3I
HI
The metal-catalyzed carbonylation of methanol is a process
of great importance because of overwhelming global acetic
acid production.1,2 However, since the first report of the
catalyst, there has been little improvement on the intrinsic
activity of the rhodium catalyst [RhI2(CO)2] . Thus, drastic
improvement on the catalytic activity is desirable.
The overall reaction mechanism is described in Scheme 1.
The rate-determining step of the catalytic cycle is the oxidative
addition of CH3I (From 1 to 2 in Scheme 1) and thus the
catalyst design focused on the improvement of this oxidative
addition step. The basic idea was that ligands which increase
the electron density at the metal should promote oxidative
addition, and consequently increase the overall rate of the
reaction. For this purpose, a variety of rhodium complexes have
been synthesized in the past, and they have been shown to be
active catalysts of comparable or better performance compared
to rhodium carbonyl species without donor ligands.1
H2O
Scheme 1. Rh-catalyzed methanol carbonylation.
¹
(a)
S
Cy
Cy
S R
P
Cl
R
LiPCy2
+
5 : R = Ph 6 : R = Et
R = Ph, Et
(b)
Cy
5 or 6 + 0.5 [RhCl(CO)2]2
P
S
R
Cy
Rh
7
OC
Cl
8 : R = Ph 9 : R = Et
Scheme 2. Phosphine-thioether ligands and Rh complexes.
Mixed P-S ligands, such as phosphine-phosphine
sulfide Ph2PCH2P(=S)Ph2 (dppms) and phosphine-thiolate
Ph2PC2H4S (dppes), are particularly attractive for incorporat-
much more electron-rich than triarylphosphines.10 Thus, the
introduction of a trialkylphosphine moiety was expected to
make up for the modest electron-donating ability of thioether
ligands so that the overall electron-donating ability would be
comparable to dppms and dppes. The new ligands 5 and 6 were
easily prepared by the reaction of 2-chloroethyl sulfides with
LiPCy2 (Scheme 2a). The complexes 8 and 9 were prepared
from ligands 5 and 6 and [RhCl(CO)2]2 7 (Scheme 2b).
Single-crystal X-ray analyses of 8 and 9 revealed their
structures and their ORTEP views are shown in Figures 1 and
2. Judging from their bond angles and bond lengths (See CIF
files), all the local geometries at the palladium atoms of 8 and 9
are square planar with Rh, P, S, C, and Cl. The CO groups
and phosphorus atoms of 8 and 9 are coordinated cis on the
¹
ing an electron-rich sulfur anion together with a neutral
phosphorus atom to accelerate oxidative addition and enhance
activity of the catalytic carbonylation of methanol.3-5
On the other hand, thioether ligands have higher stability,
allowing easy storage and handling. In addition, the synthetic
pathway to thioether-bearing phosphine ligands is a simple one
and thus both their electronic states and steric states can be
easily varied,6 compared to dppms7 and dppes derivatives.8,9
Herein we report syntheses of thioether-bearing phosphine
ligands, their Rh complexes, and their application to the
carbonylation of methanol.
Since the electron-donating ability of thioether (R2S) is
¹
clearly lower than that of thiolate (RS ) or sulfide (R3P=S), we
rhodium centers. Complexes 8 and 9 have ¯(CO) absorption
¹1
focused on thioethers-bearing trialkylphosphines which are
centered around 1975 (8) and 1979 cm
(9). The small
Published on the web August 7, 2009; doi:10.1246/bcsj.82.1009

1010 Bull. Chem. Soc. Jpn. Vol. 82, No. 8 (2009)
© 2009 The Chemical Society of Japan
Table 1. Carbonylation Experimentsa)
CH3CO2CH3 CH3CO2H
Run
Rh cat. T/°C t/h
/mmol
/mmol
1
2
3
4
5
6
7b)
8c)
7
8
9
7
8
9
7
7
120
120
120
180
180
180
120
120
1.0
1.0
1.0
0.5
0.5
0.5
1.0
1.0
2.48
6.24
2.81
5.15
5.00
6.24
1.88
2.48
0.300
0.891
0.491
4.77
5.10
3.80
0.276
0.362
a) Experiments were carried out in a 50-cm3 autoclave
equipped with a vessel. Reaction conditions after 15 min at
room temperature: Rh catalyst (0.0280 mmol), MeOH
(10.5 mL), H2O (2.28 mL), MeI (1.74 mL, 28.0 mmol), and
CO 2.5 MPa at rt (3.2 MPa at 120 °C, 5.2 MPa at 180 °C). Yield
was determined by NMR spectra with internal standard
(CHCl2CHCl2). b) [PBu3Me]+I¹ (0.0280 mmol) was added.
c) O=POct3 (0.0280 mmol) was added.
Figure 1. ORTEP drawing of 8 (all H atoms are omitted for
clarity).
spectra of Rh-P bonds with ligands 5 and 6,14 ligands 5 and 6
did not possess enough stability under the industrial conditions
(180 °C, CO 5.2 MPa), while other thioether-bearing phosphine
ligands such as Ph2P(2-MeSC6H4) and Ph2PC2H4SMe can
increase catalytic activity under the harsh conditions (185 °C,
CO 70 bar).3 It should be noted that the ratio of CH3CO2H
to CH3CO2CH3 was greatly enhanced at 180 °C, compared
to 120 °C. In order to confirm that the enhanced activity
was achieved by not phosphine oxide and/or phosphonium
iodide derived from 5 but ligand 5, phosphine oxide and
phosphonium iodide were added to 7 (Runs 7 and 8). Addition
of [PBu3Me]+I¹ to [RhCl(CO)2]2 (7) inhibited catalytic activity
(Run 7), compared with the result of [RhCl(CO)2]2 (7) (Run 1).
The dissociation and degradation of phosphine ligand into
phosphonium iodide would retard the catalytic cycle.4 Finally,
addition of phosphine oxide O=POct3 to [RhCl(CO)2]2 (7)
showed no influence on catalytic activity (Run 8). Results of
Runs 7 and 8 indicate that enhanced catalytic activity (Run 2)
results from coordination of ligand 5 on the Rh center.
In conclusion, new mixed phosphine-thioether ligands 5 and
6 and their rhodium complexes 8 and 9 were synthesized and
applied to the carbonylation of methanol. Rhodium complex 8
demonstrated higher yield of CH3CO2CH3 and CH3CO2H
at 120 °C than [RhCl(CO)2]2 (7), while rhodium complex 9
gave slightly higher catalytic activity. On the other hand,
at 180 °C, all the complexes 7-9 provided nearly the same
amount of CH3CO2CH3 and CH3CO2H. This may be due to the
dissociation of ligands which were converted into phosphine
oxide and phosphonium iodide.
Figure 2. ORTEP drawing of 9 (all H atoms are omitted for
clarity).
difference of ¯(CO) absorption between 8 and 9 indicates little
difference of electron-donating ability between phenyl sulfide
and alkyl sulfide. Naturally, both rhodium centers of 8 and
9 are more electron-rich than the rhodium complex with
triarylphosphine-thioether, Ph2P(o-MeSC6H4): ¯(CO) 1998
cm¹1.3 Furthermore, no difference in stability between 8 and
9 was seen, judging from decomposition points: 8 (177.2 to
179.3 °C) and 9 (178.2 to 180.2 °C).
The application of the rhodium complexes 8 and 9 to the
carbonylation of methanol was investigated as shown in
Table 1. At 120 °C for 1 h,11 catalyst precursor [RhCl(CO)2]2
(7) provided 2.78 mmol of CH3CO2CH3 and CH3CO2H
(Run 1). Under the same conditions, catalyst precursor 8
catalyzed the carbonylation of MeOH to yield 7.13 mmol of
CH3CO2CH3 and CH3CO2H which is more than double the rate
(Run 2), of [RhCl(CO)2]2 (7). However, the 31P NMR spectrum
of the inorganic residue indicated that some amount of ligand 5
remained bound to the Rh center after 1 h at 120 °C in the
presence of large excess of MeI, while the dissociation and
degradation of ligand 5 occurred.12 The phosphine ligand
would result in phosphine oxide and/or phosphonium iodide.13
On the other hand, another precursor 9 showed slightly higher
catalytic activity than [RhCl(CO)2]2 (7) (Run 3). Thus, sig-
nificant difference in catalytic activity between 8 and 9 was
seen although complexes 8 and 9 have nearly the same v(CO)
absorption values indicating negligible difference of the
electron center on the Rh centers in 8 and 9. At 180 °C,
all the complexes 7-9 provided nearly the same amount
of CH3CO2CH3 and CH3CO2H (Runs 4-6). At 180 °C, the
dissociation of ligands would be promoted to give phosphine
oxide and phosphonium iodide. Judging from no 31P NMR
Experimental
Carbonylation of MeOH.
A solution of Rh catalyst
(28.0 ¯mol) and MeI (28.0 mmol) in MeOH (10.5 mL) and H2O
(2.28 mL) was added to a 50-cm3 autoclave equipped with a vessel.
All reactor valves were closed. The solution was stirred at room
temperature for 15 min under CO pressure (2.5 MPa at room
temperature). The solution in the autoclave was stirred at an
assigned temperature. After the reactor was rapidly cooled to room

K. Sakakibara et al.
Bull. Chem. Soc. Jpn. Vol. 82, No. 8 (2009) 1011
temperature, the reactor valve was opened to release CO gas. As an
internal standard, CHCl2CHCl2 was added to the resultant solution
to determine yield of AcOH and AcOMe.
Crystallographic data have been deposited with Cambridge
Crystallographic Data Centre: Deposition numbers CCDC-724017
and -724018 for compounds 8 and 9, respectively. Copies of the
data can be obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.html (or from the Cambridge Crystallographic
Data Centre, 12, Union Road, Cambridge, CB2 1EZ, UK; Fax:
+44 1223 336033; e-mail: deposit@ccdc.cam.ac.uk).
4
M. J. Baker, M. F. Giles, A. G. Orpen, M. J. Taylor, R. J.
Watt, J. Chem. Soc., Chem. Commun. 1995, 197.
L. Gonsalvi, H. Adams, G. J. Sunley, E. Ditzel, A. Haynes,
J. Am. Chem. Soc. 2002, 124, 13597.
A. M. Masdeu-Bultó, M. Diéguez, E. Martin, M. Gómez,
Coord. Chem. Rev. 2003, 242, 159.
5
6
7
8
S. O. Grim, J. D. Mitchell, Inorg. Chem. 1977, 16, 1762.
E. Hauptman, P. J. Fagan, W. Marshall, Organometallics
1999, 18, 2061.
9
J. Chatt, J. R. Dilworth, J. A. Schmutz, J. A. Zubieta,
J. Chem. Soc., Dalton Trans. 1979, 1595.
10 J. Rankin, A. C. Benyei, D. J. Cole-Hamilton, A. D. Poole,
Chem. Commun. 1997, 1835.
K.S. is grateful for financial support from the Research
Fellowships of the Japan Society for the Promotion of Science
(JSPS) for Young Scientists.
11 Before the carbonylation of MeOH carried out at high
temperature, resultant solution of all material was stirred at room
temperature for 15 min (See experimental section). No acetic acid
or methyl acetate were detected for 15 min at room temperature.
12 The amount of phosphine ligand bound to the Rh center
can not be measured precisely, because an unidentified broad
peak was detected. Moreover, it should be noted that the difference
of the activity between 7 and 8 decreased for 2 h, compared with
1 h (8: CH3CO2CH3 8.77 mmol and CH3CO2H 2.15 mmol, 7:
CH3CO2CH3 5.72 mmol and CH3CO2H 1.10 mmol.).
13 J. Rankin, A. C. Benyei, A. D. Poole, D. J. Cole-Hamilton,
J. Chem. Soc., Dalton Trans. 1999, 3771.
Supporting Information
Experimental procedures and identification data of newly
synthesized compounds. This material is available free of charge
on the web at http://www.csj.jp/journals/bcsj/.
References
1
C. M. Thomas, G. Süss-Fink, Coord. Chem. Rev. 2003,
243, 125.
2
3
G. J. Sunley, D. J. Watson, Catal. Today 2000, 58, 293.
J. R. Dilworth, J. R. Miller, N. Wheatley, M. J. Baker, J. G.
14 31P NMR spectrum of the residue in Run 5 showed only
one singlet peak at 56.5 ppm.
Sunley, J. Chem. Soc., Chem. Commun. 1995, 1579.