A highly efficient catalyst precursor for ethanoic acid production: [RhCl(CO)(PEt3)2]; X-ray crystal and molecular structure of carbonyldiiodo(methyl)bis(triethylphosphine)rhodium(III)

Article abstract of DOI:10.1039/a703895j

Under mild conditions, [RhI(CO)(PEt₃)₂], is more active for the carbonylation of methanol to ethanoic acid than [Rh(CO)₂I₂]^-, which is widely used industrially; intermediates in the catalytic cycle have been identified and characterised.

Full text of DOI:10.1039/a703895j

View Article Online / Journal Homepage / Table of Contents for this issue
A highly efficient catalyst precursor for ethanoic acid production:
[RhCl(CO)(PEt3)2]; X-ray crystal and molecular structure of
carbonyldiiodo(methyl)bis(triethylphosphine)rhodium(iii)
Joanne Rankin,a Andrew D. Poole,b Attila C. Benyeia† and David J. Cole-Hamilton*a
a School of Chemistry, University of St Andrews, Fife, Scotland, UK KY16 9ST
b B.P. Chemicals Ltd, Hull Research and Technology Centre, Saltend, Hull, UK HU12 8DS
Under mild conditions, [RhI(CO)(PEt3)2], is more active for
the carbonylation of methanol to ethanoic acid than
[Rh(CO)2I2]2, which is widely used industrially; interme-
diates in the catalytic cycle have been identified and
characterised.
The mechanistic cycle has been studied using high pressure
and variable temperature NMR and high pressure IR spectro-
scopies10 as well as stoichiometric reactions. Thus, addition of
methyl iodide to [RhI(CO)(PEt3)2] in CH2Cl2 leads smoothly to
the formation of [RhMeI2(CO)(PEt3)2], which has been fully
characterised spectroscopically‡ and by single crystal X-ray
diffraction§ (Fig. 1). The methyl group is cis to the carbon
monoxide, as required for migratory insertion. There have been
no previously published crystallographic analyses of complexes
of the type [RhX2Me(CO)(PR3)2]. However, there are examples
of crystal structures of other six-coordinate rhodium phosphine
complexes where methyl iodide has been oxidatively
added.11–13 Most have iodide and methyl ligands in mutually
trans positions.
The isolation of the methyl complex for a catalytically active
system is rather unusual since, generally, the insertion of carbon
monoxide into the Rh–C bond is extremely rapid. For
[Rh(CO)2I2]2, the methyl complex has a very short lifetime and
has only been detected as a transient by IR spectroscopy in neat
methyl iodide,14 whilst for the related [RhCl(CO)(PPh3)2],
oxidative addition of methyl iodide gives the six-coordinate
[RhMeClI(CO)(PPh3)2], but this is in equilibrium with the five-
coordinate insertion product, [RhClI(MeCO)(PPh3)2].15 Pre-
sumably the high electron density on the metal in [RhMeI2-
(CO)(PEt3)2] is responsible for the less facile methyl migration
in this case. Despite the stability of the methylrhodium(iii)
complex, preliminary kinetic studies suggest that oxidative
addition of MeI is still rate determining. The catalytic reaction
The [Rh(CO)2I2]2 catalysed carbonylation of methanol is the
most widely used commercial route to ethanoic acid.1 The rate
determining step for the catalytic cycle2 is oxidative addition of
iodomethane to [Rh(CO)2I2]2, so that increasing the electron
density on the metal centre should increase the rate of oxidative
addition and hence the overall rate of production of ethanoic
acid. There have been attempts to modify [Rh(CO)2I2]2 by
using electron donating ligands.3–6 For instance, Baker et al.3
used [RhI(CO)(Ph2PCH2P(S)Ph2)] to enhance the rate of
carbonylation at 185 °C. Cavell and coworkers4 prepared
complexes of the form, [RhCl(CO)(PPh2CH2PPh2NR)], where
R is a substituted phenyl group. Wegman et al.5 compared the
methanol carbonylation rate at 80 °C using [Rh(CO)2I2]2 and
[RhCl(CO)2(Ph2P(CH2)2P(O)Ph2)] and noted an 80-fold in-
crease with the latter. Dilworth et al.6 used rhodium(i) carbonyl
complexes containing phosphinothiolate and thioether ligands
to promote the rate of carbonylation of methanol under the
relatively forcing conditions of 70 bar and 185 °C. To our
knowledge, the use of trialkylphosphines as promoters for
rhodium based carbonylation catalysts has not been investi-
gated, despite the fact that they are strongly electron donating.
We now report the results of such a study.
Complexes of the form [RhX(CO)(PEt3)2] (X = Cl, Br or I)
have nCO at ca. 1960 cm21 compared with 1988 and 2059 cm21
for [Rh(CO)2I2]2, suggesting that the rhodium centre is more
electron rich in the triethylphosphine complexes. We have
shown that these complexes are active for the double carbonyla-
tion of CH2I27 and of allylic halides8 and so have attempted to
increase the rate of carbonylation of methanol by using these
complexes and have found that [RhCl(CO)(PEt3)2] is an
extremely active catalyst precursor for ethanoic acid produc-
tion. We report the first direct comparison of [Rh(CO)2I2]2 with
a rhodium phosphine complex under the conditions of
120–150 °C and 27 bar carbon monoxide pressure. The initial
carbonylation rates using [RhCl(CO)(PEt3)2] and [Rh(CO)2Cl]2
as sources of rhodium at 120 and 150 °C in the presence of
differing water concentrations are shown in Table 1.
The results of Table 1 show that, in the presence of 17.1%
m/m H2O under the mild conditions of 120 or 150 °C and 27 bar
pressure (runs 1–4), [RhI(CO)(PEt3)2] catalyses the carbonyla-
tion of methanol at a rate nearly double that of [Rh(CO)2I2]2.
Water acts to maintain the catalyst in its active form9 i.e. as a
rhodium(i) complex and decreases the formation of inactive
rhodium(iii) complexes such as [Rh(CO)2I4]2 or [RhI3(CO)-
(PEt3)2]. Thus, as shown in Table 1, the rate of carbonylation is
dramatically enhanced in the presence of a significant concen-
tration of water. Under lower water conditions (runs 5 and 6), no
appreciable benefit is obtained by using [RhCl(CO)-
(PEt3)2] over [Rh(CO)2Cl]2 as catalyst precursors.
Table 1 Results of carbonylation experimentsa
Initial water
(% m/m)
Initial rate/
mol dm23
h21
Run
T/°C
Rhodium source
1
2
3
4
5
6
120
120
150
150
120
120
[RhCl(CO)(PEt3)2]b
17.1
17.1
17.1
17.1
3.2
1.6
0.9
9.2
5.0
0.6
0.6
c
[Rh(CO)2Cl]2
[RhCl(CO)(PEt3)2]b
c
[Rh(CO)2Cl]2
[RhCl(CO)(PEt3)2]b
c
[Rh(CO)2Cl]2
3.2
a
Experiments were carried out in a 300 cm3 Hastelloy B2 autoclave
equipped with a ballast vessel, Magnedrive stirrer, liquid injection facility
and cooling coils. For reactions requiring 17.1% H2O m/m, a solution of
MeCOOH (987 mmol), MeOAc (474 mmol), H2O (1.82 mol) and MeI (396
mmol) was charged to the vessel. Similarly for reactions requiring 3.2%
H2O m/m, a solution of MeCOOH (1.75 mol), MeOAc (286 mmol), H2O
(354 mmol) and MeI (396 mmol) was charged to the vessel. The reaction
vessel was heated to the required temperature whilst stirring under ca. 5 bar
CO pressure. Once the desired temperature had been reached, the rhodium
source, dissolved in MeOAc (13.5 mmol) and MeCO2H (117 mmol), was
injected into the autoclave to initiate the reaction. The autoclave was
pressurised to 27 bar and stirred at 600 rpm. The rate of carbonylation was
monitored by recording the rate of carbon monoxide uptake from a ballast
b
vessel every 2 s. Assumed to form [RhI(CO)(PEt3)2] rapidly under the
reaction conditions. c Forms [Rh(CO)2I2]2 under the reaction conditions.
Chem. Commun., 1997
1835

View Article Online
Et3P
OC
I
Rh
PEt3
C(6)
C(5)
1
C(7)
C(8)
H2O
MeCOI
MeI
C(3)
P(1)
Me
CO
C(4)
MeCOOH
MeOH
Me
Et3P
I
Et3P
OC
I
HI
Rh
Rh
OC
PEt3
PEt3
CO
C(2)
I
I
I(2)
3
2
Rh
Scheme 1 Proposed cycle for the carbonylation of methanol catalysed by
[RhI(CO)(PEt3)2]
I(1)
C(1)
O(1)
‡ Selected spectroscopic data: all NMR spectra were recorded in CD2Cl2
unless otherwise stated. For 1: 31P{1H}: d 20.2 (d, 1JRhP 111 Hz); 13C{1H}:
d (CO) 186.0 (dt, 1JRhC 79 Hz, 2JPC 16 Hz); nCO/cm21 (CH2Cl2) 1961. For
2: 31P{1H}: d 2.4 (d, 1JRhP 85 Hz); 13C {1H}: d (CO) 183.3 (dt, 1JRhC 64 Hz,
2JPC 10 Hz), d (CH3) 23.1 (dt, 1JRhC 19 Hz, 2JPC 4 Hz); nCO/cm21 (CH2Cl2)
2043. For 3: 31P{1H}: d 2.1 (d, 1JRhP 88 Hz); 13C{1H} (C6D6): d (CO) 179.3
(dt, 1JRhC 69 Hz, 2JPC 8 Hz), d (COCH3) 226 (dt, 1JRhC 25 Hz, 2JPC 5 Hz),
d (COCH3) 47.6 (d, 2JRhC 3 Hz); nCO/cm21 (CH2Cl2) 2065, 1665.
C(12)
P(2)
C(9)
C(11)
C(13)
C(10)
C(14)
§ Crystal data for 2, [RhMeI2(CO)(PEt3)2], C14H33I2OP2Rh, Mr = 636.08,
monoclinic, space group Pn (no. 7), a
= 7.402(5), b = 11.785(5),
Fig. 1 X-Ray crystal structure and numbering scheme for [RhMeI2-
c = 12.520(4) Å, b = 96.01(4)°, U = 1086.2(9) Å3, Z = 2, Dc = 1.95
g cm23, F(000) = 612, m(Mo-Ka) = 37.7 cm21, l = 0.71069 Å, T = 220
K. Crystal size, 0.5 3 0.4 3 0.4 mm. Of the 2787 reflections which where
collected (3 < 2q < 50°) on a Rigaku AFC7S diffractometer, 2598 were
unique and 2453 were observed. Structure solved by direct methods and
expanded using Fourier techniques. The non-hydrogen atoms were refined
anisotropically. The maximum and minimum residual electron densities are
1.29 and 21.19 e Å23. R = 0.026, Rw = 0.027. CCDC 182/566.
Important dimensions: bond lengths (Å): Rh–P(1) 2.387(2), Rh–P(2)
2.390(2), Rh–C(1) 1.836(9), Rh–C(2) 2.109(9), Rh–I(1) 2.7823(7), Rh–I(2)
2.721(1), C(1)–O(1) 1.113(10); bond angles (°) I(1)–Rh–I(2) 94.18(4),
I(1)–Rh–P(2) 91.36(5), I(1)–Rh–C(2) 173.8(2), I(2)–Rh–P(2) 88.22(5),
I(2)–Rh–C(2) 91.8(2), P(1)–Rh–C(1) 92.2(3), P(2)–Rh–C(1) 90.0(3),
C(1)–Rh–C(2), 85.0(3), I(1)–Rh–P(1) 90.34(5), I(1)–Rh–C(1) 89.0(2),
I(2)–Rh–P(1) 89.48(5), I(2)–Rh–C(1) 176.3(2), P(1)–Rh–P(2) 177.24(7),
P(1)–Rh–C(2) 88.1(2), P(2)–Rh–C(2) 90.4(2).
(CO)(PEt3)2]
is zero order in pCO and first order in [MeI]. However, we have
found that at 298 K the rate of oxidative addition of MeI for
[RhI(CO)(PEt3)2] (1.37 3 1023 dm3 mol21 s21) is 57 times
faster than that for [Rh(CO)2I2]2 (2.39 3 1025 dm3 mol21 s21);
this compares with 120 times faster16 for [Ir(CO)2I2]2 relative
to [Rh(CO)2I2]2.
Migratory insertion of carbon monoxide occurs under CO
pressure (27 bar) to form [RhI2(MeCO)(CO)(PEt3)2] which has
been fully spectroscopically characterised.‡ The rate of the
insertion reaction at room temperature and 27 bar is 38 times
slower than for [RhMe(CO)2I3]2,¶ but this slowing is not
sufficient to make the migratory insertion step rate determining
at 150 °C (shifts to 1740 cm-1, MeOAc, on addition of MeOH).
In contrast, for [Ir(CO)2I2]2, the insertion reaction is slowed by
105–106 times16 compared to [Rh(CO)2I2]2 and becomes rate
determining.
Heating a solution of [RhI2(MeCO)(CO)(PEt3)2] to 75 °C in
the presence of excess methyl iodide and 40 bar carbon
monoxide produces a product with a large absorption at 1799
cm21 arising from ethanoyl iodide.17 This reductive elimination
step completes the catalytic cycle. It has been shown that the
usual mechanism for the formation of ethanoic acid or methyl
ethanoate involves reductive elimination of ethanoyl iodide
followed by reaction with water or methanol;2,18 however, it is
unusual to observe ethanoyl iodide. The high electron density
on the rhodium centre caused by the electron donating system
would be expected to favour the RhIII complex, [RhI2(Me-
CO)(CO)PEt3)2], over [RhI(CO)(PEt3)2] and MeCOI. How-
ever, the way in which the reaction is performed allows the
catalytic production of MeCOI because the RhI complex can be
trapped by addition of methyl iodide when this is in excess. The
overall catalytic cycle for the carbonylation of methanol in the
presence of [RhI(CO)(PEt3)2], for which each complex has been
characterised, is shown in Scheme 1.
¶ For [RhMe(CO)2I3]2 this insertion reaction is independent of carbon
monoxide pressure whilst for [RhMeI2(CO)(PEt3)2] a first order depend-
ence on carbon monoxide is observed.
1 M. J. Howard, M. D. Jones, M. S. Roberts and S. A. Taylor, Catal.
Today, 1993, 18, 325.
2 D. Forster, Adv. Organomet. Chem., 1979, 17, 255.
3 M. J. Baker, M. F. Giles, A. G. Orpen, M. J. Taylor and R. J. Watt,
J. Chem. Soc., Chem. Commun., 1995, 197.
4 K. V. Katti, B. D. Santarsiero, A. A. Pinkerton and R. G. Cavell, Inorg.
Chem., 1993, 32, 5919.
5 R. W. Wegman, Eur. Pat., 171804, 1986; Chem. Abstr., 1986, 105,
78526g.
6 J. R. Dilworth, J. R. Miller, N. Wheatley, M. J. Baker and J. G. Sunley,
J. Chem. Soc., Chem. Commun., 1995, 1579.
7 W. S. Weston, R. C. Gash and D. J. Cole-Hamilton, J. Chem. Soc.,
Chem. Commun., 1994, 745.
8 M. C. Simpson, M. J. Payne and D. J. Cole-Hamilton, J. Chem. Soc.,
Dalton Trans., 1994, 2899.
9 E. C. Baker, D. E. Hendriksen and R. Eisenberg, J. Am. Chem. Soc.,
1980, 102, 1020.
10 W. R. Moser, J. E. Cnossen, A. W. Wang and S. A. Krouse, J. Catal.,
1985, 95, 21.
11 T. G. Schenck, C. R. C. Milne, J. F. Sawyer and B. Bosnich, Inorg.
Chem., 1985, 24, 2338.
At 150 °C the catalyst loses activity over ca. 10 min with the
formation of Et3PO and [Rh(CO)2I2]2. Current studies are
aimed at stabilising the catalyst towards phosphine loss.
We thank Dr Anthony Haynes for useful discussions. We
thank B.P. Chemicals Ltd and the EPSRC for a studentship
(J. R.) and the European Commission for support under a
TEMPUS programme (A. C. B.).
12 G. J. Lamprecht, G. J. Van Zyl and J. G. Leipoldt, Inorg. Chim. Acta,
1989, 164, 69.
13 M. J. Menu, P. Desrosiers, M. Darguenave and Y. Dartiguenave,
Organometallics, 1987, 6, 1822.
14 A. Haynes, B. E. Mann, G. E. Morris and P. M. Maitlis, J. Am. Chem.
Soc., 1993, 115, 4093.
15 I. C. Douek and G. Wilkinson, J. Chem. Soc. A, 1969, 2604.
16 M. Bassetti, D. Monti, A. Haynes, J. M. Pearson, I. A. Stanbridge and
P. M. Maitlis, Gazz. Chim. Ital., 1992, 122, 391.
17 J. Ramsey and J. Ladd, J. Chem. Soc. B, 1968, 188.
18 P. M. Maitlis, A. Haynes, G. J. Sunley and M. J. Howard, J. Chem. Soc.,
Dalton Trans., 1996, 2187.
Footnotes and References
* E-mail: djc@st-and.ac.uk
† Current address: Chemistry Department, Kossuth Lajos University,
Debrecen, Hungary.
Received in Cambridge, UK, 4th June 1997; 7/03895J
1836
Chem. Commun., 1997