Evaluation of C4 diphosphine ligands in rhodium catalysed methanol carbonylation under a syngas atmosphere: Synthesis, structure, stability and reactivity of rhodium(i) carbonyl and rhodium(iii) acetyl intermediates

Article abstract of DOI:10.1039/b712974b

The carbonylation of methanol to acetic acid is a hugely important catalytic process, and there are considerable cost and environmental advantages if a process could be designed that was tolerant of hydrogen impurities in the CO feed gas, while eliminating by-products such as propionic acid and acetaldehyde altogether. This paper reports on an investigation into the application of rhodium complexes of several C₄ bridged diphosphines, namely BINAP, 1,4-bis(diphenylphosphino)butane (dppb), bis(diphenylphosphino) xylene (dppx) and 1,4-bis(dicyclohexylphosphino)butane (dcpb) as catalysts for hydrogen tolerant methanol carbonylation. An investigation into the structure, reactivity and stability of pre-catalysts and catalyst resting states of these complexes has also been carried out in order to understand the observations in catalysis. Rh(i) carbonyl halide complexes of each of the ligands have been prepared from both [Rh₂(CO)₄Cl₂] and dimeric μ-Cl-[Rh(L)Cl]₂ complexes. These Rh(i) carbonyl complexes are either dimeric with bridging phosphine ligands (dppb, dcpb, dppx) or monomeric chelate complexes. The reaction of the complexes with methyl iodide at 140 °C has been studied, which has revealed clear differences in the stability of the corresponding Rh(iii) complexes. Surprisingly, the dimeric Rh(i) carbonyls react cleanly with MeI with rearrangement of the diphosphine to a chelate co-ordination mode to give stable Rh(iii) acetyl complexes. The Rh acetyls for L = dppb and dppx have been fully characterised by X-ray crystallography. During the catalytic studies, the more rigid dppx and BINAP ligands were found to be nearly 5 times more hydrogen tolerant than [Rh(CO) ₂I₂]^-, as revealed by by-product analysis. The origin of this hydrogen tolerance is explained based on the differing reactivities of the Rh acetyls with hydrogen gas, and by considering the structure of the complexes. The Royal Society of Chemistry.

Full text of DOI:10.1039/b712974b

View Article Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions themed issue on:
Emerging Strategies in Catalysis
Guest Edited by Carmen Claver
Universitat Rovira i Virgili, Tarragona
Published in issue 47, 2007 of Dalton Transactions
Images reproduced by permission of Christian Muller (outside) and Challa Kumar (inside)
Other papers published in this issue include:
Access to well-defined isolated Fe(II) centers on silica and their use in oxidation
Charbel Roukoss, Steven Fiddy, Aimery de Mallmann, Nuria Rendón, Jean-Marie Basset,
Emile Kuntz and Christophe Copéret, Dalton Trans., 2007, DOI: 10.1039/b711015d
Phosphinines as ligands in homogeneous catalysis: recent developments, concepts
and perspectives
Christian Müller and Dieter Vogt, Dalton Trans., 2007, DOI: 10.1039/b712456m
Solventless continuous flow homogeneous hydroformylation of 1-octene
Anja C. Frisch, Paul B. Webb, Guoying Zhao, Mark J. Muldoon, Peter J. Pogorzelec and
David J. Cole-Hamilton, Dalton Trans., 2007, DOI: 10.1039/b712683b
Palladium catalyzed Suzuki C–C couplings in an ionic liquid: nanoparticles responsible
for the catalytic activity
Fernando Fernández, Beatriz Cordero, Jérôme Durand, Guillermo Muller, François Malbosc,
Yolande Kihn, Emmanuelle Teuma and Montserrat Gómez, Dalton Trans., 2007, DOI:
10.1039/b713449e
Visit the Dalton Transactions website for cutting-edge inorganic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Evaluation of C4 diphosphine ligands in rhodium catalysed methanol
carbonylation under a syngas atmosphere: synthesis, structure, stability and
reactivity of rhodium(I) carbonyl and rhodium(III) acetyl intermediates†
Gareth Lamb,a Matthew Clarke,*a Alexandra M. Z. Slawin,a Bruce Williamsb and Lesley Keyb
Received 23rd August 2007, Accepted 25th September 2007
First published as an Advance Article on the web 10th October 2007
DOI: 10.1039/b712974b
The carbonylation of methanol to acetic acid is a hugely important catalytic process, and there are
considerable cost and environmental advantages if a process could be designed that was tolerant of
hydrogen impurities in the CO feed gas, while eliminating by-products such as propionic acid and
acetaldehyde altogether. This paper reports on an investigation into the application of rhodium
complexes of several C4 bridged diphosphines, namely BINAP, 1,4-bis(diphenylphosphino)butane
(dppb), bis(diphenylphosphino)xylene (dppx) and 1,4-bis(dicyclohexylphosphino)butane (dcpb) as
catalysts for hydrogen tolerant methanol carbonylation. An investigation into the structure, reactivity
and stability of pre-catalysts and catalyst resting states of these complexes has also been carried out in
order to understand the observations in catalysis. Rh(I) carbonyl halide complexes of each of the
ligands have been prepared from both [Rh2(CO)4Cl2] and dimeric l-Cl-[Rh(L)Cl]2 complexes. These
Rh(I) carbonyl complexes are either dimeric with bridging phosphine ligands (dppb, dcpb, dppx) or
monomeric chelate complexes. The reaction of the complexes with methyl iodide at 140 ◦C has been
studied, which has revealed clear differences in the stability of the corresponding Rh(III) complexes.
Surprisingly, the dimeric Rh(I) carbonyls react cleanly with MeI with rearrangement of the diphosphine
to a chelate co-ordination mode to give stable Rh(III) acetyl complexes. The Rh acetyls for L = dppb
and dppx have been fully characterised by X-ray crystallography. During the catalytic studies, the more
rigid dppx and BINAP ligands were found to be nearly 5 times more hydrogen tolerant than
[Rh(CO)2I2]−, as revealed by by-product analysis. The origin of this hydrogen tolerance is explained
based on the differing reactivities of the Rh acetyls with hydrogen gas, and by considering the structure
of the complexes.
have now been explored in the ‘unmodified’ catalyst systems, and
phosphine modified catalysts may offer the best opportunity for
fine-tuning the selectivity of the catalysts.18
Introduction
The carbonylation of methanol to acetic acid is one of the
most important applications in homogeneous catalysis. Large
scale industrial processes have typically used either [Rh(CO)2(I)2]−
or [Ir(CO)2(I)2]− as catalysts with methyl iodide1–3 and some
ruthenium halide scavengers as promoters.4–6 The current CativaTM
process uses an iridium catalyst with a high activity and selectivity
towards acetic acid.
Phosphine-modified Rh catalysts have also received intense
attention, with the main goal being to further increase activity and
selectivity.7–17 The commercial processes for carbonylation uses
pure CO, which is purified from syngas (1 : 1 CO/H2) in an energy
intensive and expensive process. Propionic acid is the main liquid
by-product requiring distillation to produce high grade acetic acid.
A step change in methanol carbonylation technology would be to
reduce propionic acid levels such that fractional distillation was
not required, and/or to develop catalysts able to utilise lower
grade CO that contains hydrogen. Most of the possible variables
An extremely interesting paper by Moloy and Wegman in 1989
described reductive carbonylation of methanol with (2 : 1 H2/CO)
syngas to give ethanol and ethanal.19,20 Good selectivities towards
reductive carbonylation products were observed with C3 diphos-
phines such as 1,3-bis(diphenylphosphino)propane (dppp). These
authors also studied key organometallic intermediates in the dppp-
based system to shed light on the basic mechanism. This paper
notes that the C4 diphosphine, 1,4-bis(diphenylphosphino)butane
(dppb), was particularly ineffective for reductive carbonylation,
predominantly giving acetic acid derivatives as products. The poor
selectivities observed with the dppb catalysts in this application
suggested that this type of phosphine could offer opportunities
in hydrogen tolerant methanol carbonylation. In this paper,
the synthesis and characterisation of rhodium complexes of C4
diphosphines and their application in methanol carbonylation
using syngas is reported.
aSchool of St. Andrews, University of St. Andrews, St. Andrews, Fife, UK
KY16 9ST. E-mail: mc28@st-andrews.ac.uk; Fax: +44 (0)1334 463808;
Tel: +44 (0)1334 463808
Results and discussion
Coordination chemistry
bBP Chemicals Ltd, Saltend, Hull, UK HU12 8DS
The reaction of dppb with [Rh(CO)2Cl]2 has been reported to give
the dimeric complex [Rh(dppb)CO(Cl)]2, 1.21 1 was prepared by
† CCDC reference numbers 658319–658321. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b712974b
5582 | Dalton Trans., 2007, 5582–5589
This journal is
The Royal Society of Chemistry 2007
©

the literature route, and our more extensive data agrees with the
proposed formulation. In particular, FAB MS shows a M+ 1184.1
peak as the highest mass ion and fragments assigned to [M − Cl]+
(1149.1), and [M − 2CO]+ (1128.1), that would seem unlikely to
arise from monomeric or polymeric complexes.
The original report on the synthesis of [Rh(dppb)CO(Cl)]2 1,
notes that it is not clear if this complex with bridging bidentate
diphosphine ligands is a kinetic or thermodynamic product. To
investigate this further, [Rh(dppb)Cl]2 2,22 in which dppb acts as
a chelate ligand, was treated with CO (Scheme 1). Quantitative
formation of the bridging complex is observed: the fact that the
dppb ligand rearranges from chelate to bridging ligand during
the reaction with CO supports 1 being the thermodynamically
preferred arrangement for this complex.
Fig. 1 Rhodium/C4-phosphine complexes.
Scheme 1 Alternative methods for the formation of [Rh(dppb)CO(Cl)]2.
The carbonylation of [Rh(dppb)Cl]2 2 by decarbonylation
of aldehydes23 was also investigated as an alternative route
to rhodium–diphosphine carbonyl complexes. All of the alkyl
and aryl aldehydes of varying electronic properties (fluoroben-
zaldehyde, 4-methoxybenzaldehyde, etc.) gave quantitative con-
version of the Rh starting material to [Rh(dppb)CO(Cl)]2 after
◦
10 min at 120 C. The bulky aldehyde 2-carboxymethyl-2ꢀ-
methylpropionaldehyde24 on the other hand gave a mixture of
products.
Fig. 2 X-Ray structure analysis of [Rh(BINAP)CO(I)] 6.
Reactions of 1,4-bis(dicyclohexylphosphino)butane (dcpb) and
bis(diphenylphosphino)xylene (dppx) with [Rh(CO)4Cl]2 also gave
sparingly soluble dimeric complexes, 3 and 4 respectively with
bridging bidentate phosphine ligands (Fig. 1).
product. The stability and reactivity of other carbonyl complexes
including [Rh(PPh3)2(CO)Cl] was then examined in greater detail.
Stability was assessed by heating with a large excess of MeI
(300 eq.) for an exact period of time (60 min) at a set temperature,
followed by recording a 31P NMR spectrum of an accurately
In contrast, reaction of [Rh(CO)2Cl]2 with rac-BINAP or
25
reaction of [Rh(rac-BINAP)Cl]2 with CO gave the monomeric
chelate complex, 5.26 This complex reacts with NaI in acetone
to give the iodo analogue, 6 (Fig. 2). The structure of [Rh(rac-
BINAP)(CO)I] was determined by X-ray crystallography.
measured aliquot (0.50 ml) relative to a calibrated external
n
=
standard of Bu3P O (solution in C6D6 in a capillary tube).
This study was complicated by the low solubility of some of the
complexes, which precipitated from solution. The data reported
in Table 1 therefore shows both the relative concentration of Rh
phosphine species in solution and the % of the phosphorus species
observed being the expected Rh–carbonyl or Rh–acetyl species.
The differences between the ligands are quite striking.
Structurally [Rh(rac-BINAP)CO(I)] is quite typical of other
Rh/BINAP complexes with respect to bond lengths and angles.27,28
The expected trans influence of the CO is also seen, with the P(2)–
˚
Rh bond length (2.3769(12) A) being significantly longer than that
˚
of the bond trans to the halide (2.2374(12) A).
The dppb and dppx carbonyl complexes react smoothly with
MeI at 140 ◦C, and after 60 min of reaction time, only very
small traces of anything other than the Rh–acetyl complexes are
detectable in solution (entries 1 and 3). [Rh(PPh3)2(CO)Cl] reacts
with MeI at 60 ◦C, but after only 10 min at a temperature of 100 ◦C
or above, extensive decomposition to Ph2P(O)OH and Ph3PMe+I−
takes place. This is quantitative after 10 min at 140 ◦C. For dppb
and dppx complexes, a higher temperature of 150 ◦C was needed
Investigating stability and reactivity of Rh(I) carbonyl complexes
The reactivity and stability of complexes 1 and 3–5, in the
presence of methyl iodide, was then examined. [Rh(dppb)(CO)Cl]2
was previously reported to be unreactive with methyl iodide.29
However, in our hands, heating 1 with MeI at 140 ◦C for 10 min
in a glass pressure vessel in a microwave smoothly generates
[Rh(dppb)(C(O)CH3)(I)2] 7, as the only phosphorus containing
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5582–5589 | 5583
©

Table 1 Stability of Rh(I) carbonyl complexes in the absence of CO
Table 2 Comparison of X-ray data for [Rh(L)COMe(I)2] complexes for
C1-, C2-, C3- and C4-diphosphines
Entry
Ligand
T/◦C
% in solutiona
% Rh–P speciesb
a
˚
˚
Rh–P /A
C–O/A
I–Rh–I/◦
P–Rh–P/◦
1
2
3
4
5
6
7
dppb
dppb
dppx
dppx
dcpb
BINAP
PPh3
140
150
140
150
140
140
140
44.3
58.3
19.5
48.3
0
0
0
94.4
33.5
> 99
83.3
0
dppm31
dppe32
dppp19
dppb 7
dppx 8
2.265
2.276
2.288
2.312
2.309
1.160(3)
1.178(9)
1.182(7)
1.196(4)
1.210(13)
90.55(10)
91.62(4)
89.15(2)
86.84(16)
85.63(4)
73.30(2)
84.74(7)
90.49(5)
99.13(3)
103.31(11)
0
0
a Average bond length of Rh–P1 and Rh–P2.
a % of complex remaining in solution relative to an external standard of
tributylphosphine oxide calibrated against Rh starting material. b % of the
NMR signals that can be attributed to Rh complexes relative to the singlets
attributed to P(V) species.
to differentiate between their stability. At 150 ◦C dppb complexes
start to decompose more readily resulting in only 33.5% stability
whereas dppx complexes are able to maintain a greater stability
even at this higher temperature.
The high stability of these complexes is somewhat surprising,
since the dppb and dppx analogues have to rearrange from a
bridging bidentate co-ordination mode to a monomeric chelate
during the oxidative addition (or migratory insertion) process, and
this presumably involves dissociation of one end of the diphos-
phine. This rearrangement must be fast relative to quaternisation
of the phosphine. More surprising was the observation that despite
many attempts, [Rh(rac-BINAP)(CO)Cl] reacts with MeI with
severe decomposition: several phosphorus peaks without 31P–Rh
coupling are visible after 60 min at 140 ◦C. Thus in this case, a
Rh(I) chelate complex shows lower stability than a Rh(I) bridging
bidentate complex.
Fig. 3 Molecular structure of [Rh(dppb)(COMe)(I)2] 7.
Rh(III) acetyls for dppb and dppx could be isolated by this
preparative method in good yield. Using the rhodium acetonitrile
30
complex [Rh(NCMe)CO(I)2COMe]2 it was hoped that the
BINAP and dcpb acetyls could be synthesised. However, both
reactions did not go cleanly producing a mixture of products.
In the case of BINAP, the reaction most likely formed an acetyl
1
species (31P{ H} NMR (CD2Cl2): d 32.6 (dd, 1JRh–P 140.2 Hz, 2JP–P
1
2
14.9 Hz), 21.5 (dd, JRh–P 138.1 Hz, JP–P 14.9 Hz); IR (KBr),
v(CO): 1706.9 cm−1) and a set of minor signals (31P{ H} NMR
1
(CD2Cl2): d 43.7 (dd, 1JRh–P 161 Hz, 2JP–P 43.1 Hz), 21.5 (dd, 1JRh–P
125.1 Hz, 2JP–P 43.1 Hz); IR (KBr), v(CO): 2001 cm−1) relating to
the Rh(I) complex [Rh(rac-BINAP)CO(I)]. The formation of the
latter product for BINAP in the absence of CO, presumably, by
a reverse process of migration and reductive elimination of MeI
may imply lower stability for the Rh(III) acetyl. In any case, there
seems to be a stark difference in stability and reactivity of these
C4-diphosphine complexes.
Molecular structures of rhodium acetyl complexes
Fig. 4 Molecular structure of [Rh(dppx)(COMe)(I)2] 8 with a mirror
plane at (x, −y + 0.5, z) bisecting the molecule.
Crystals of [Rh(dppb)(C(O)CH3)(I)2] 7 and [Rh(dppx)(C(O)CH3)-
(I)2] 8 suitable for X-ray diffraction could be grown by slowly
blowing nitrogen over the reaction mixture. The structures of these
complexes were then determined by X-ray crystallography (Fig. 3
and 4).
Going from a C1 to a C4 backbone it can be seen that Rh–P and
C–O bond lengths steadily increase (Table 2). Bond angles are also
affected with a P–Rh–P angle increasing from 73 to 103◦, coupled
with a decreasing I–Rh–I angle. Although there is little difference
between the bond lengths of dppb and dppx complexes, the bond
angles follow the previous trends of an increasing P–Rh–P and
decreasing I–Rh–I angle.
With a greater understanding of Rh co-ordination chemistry
of the C4 ligands in hand, we undertook catalytic studies in
carbonylation of methanol using CO/H2 mixtures.
5584 | Dalton Trans., 2007, 5582–5589
This journal is
The Royal Society of Chemistry 2007
©

Methanol carbonylation
of acetaldehyde produced, as well as the gaseous side-product
methane.
Several carbonylation reactions were run under differing condi-
tions. Our main aim was to compare a range of C4 ligands in a series
of in situ experiments and come up with a rational design for a
hydrogen tolerant system for methanol carbonylation (Scheme 2).
By looking at the amount of methyl acetate and acetic acid
produced we can start to build up a picture of the activity of
these systems and what affect the phosphine ligand has upon this.
The dppb system at 140 ◦C appears to be the most active (965
TOF/h−1) followed by the non-modified system (581 TOF/h−1).
The least reactive under these conditions were the dppx and the
BINAP systems.
The primary liquid by-product was acetaldehyde, under an
atmosphere of CO we would expect to see very little produced
and this is what is observed. By running the reactions using 5 :
1 CO/H2 gas mixtures at 150 ◦C with 15% H2O the hydrogen
tolerance of these systems was evaluated.
The order of reactivity for the catalysts was dppb>
Rh2CO4Cl2>dcpb>BINAP>dppx. The dppb system proved to
be the most active with Rh2CO4Cl2 and the dcpb systems also
showing high activity relative to the dppx and BINAP systems,
that were approximately 50% less active.
The greatest change in the products compared with the previous
experiment was the increase in the by-product acetaldehyde which
increased by as much as 35 times in the case of Rh2CO4Cl2
and dppb. Both Tables 3 and 4 show the mol% of acetalde-
hyde/carbonylation products, from these calculations it can be
seen that the dppx system is 4.5 times as hydrogen tolerant as
[Rh(CO)2I2]− with as little as 0.74% of the carbonylated products
being the undesired by-product acetaldehyde.
As well as acetaldehyde both the dppb and dcpb systems formed
large amounts of methane, in the case of the dppb system >25% of
the autoclave headspace was methane, whereas dppx and BINAP
systems gave very little methane as a gaseous by-product with
BINAP giving <0.1 %vol vol−1.
Scheme 2 Postulated mechanism for carbonylation versus reductive car-
bonylation in a rhodium–diphosphine–methyl iodide catalysed reaction.
The catalytic cycle shown in Scheme 2 demonstrates the com-
peting reactions. When hydrogen is present acetaldehyde may be
formed via the pathway on the left hand side of the cycle. However,
this must compete with the carbonylation reaction forming the
reactive intermediate 9 from which several carbonylated products
are possible depending on the water concentration of the system.
Therefore several carbonylation reactions were run using differ-
ent gas mixtures and water concentrations. At 100 ◦C 22.6 g of
MeI was charged into 100 g of methanol containing the precursor
Rh2CO4Cl2 and the phosphine ligand. The solution was then
heated at 140 ◦C for 40 min under a pressure of 26 bar of CO
and after the reaction was complete both liquid and gas samples
were taken for analysis.
In conclusion, dppx and BINAP systems although suffering
from comparably low activity are very hydrogen tolerant produc-
◦
ing between 135–185 ppm of acetaldehyde at 150 C and almost
neglible amounts of methane in all catalytic experiments. To the
best of our knowledge the dppx system is the most hydrogen
tolerant catalyst for the rhodium catalysed carbonylation of
methanol reported to date.18
The results highlighted in bold, in Tables 3 and 4, comprise
the most significant data. These being the amount of desired
products formed i.e. acetic acid and methyl acetate, the amount
Table 3 Methanol carbonylation using pure COa
Ligand
None
581
dppb
965
dppx
208
dcpb
465
BINAP
TOFb/h−1
174
<0.079
Aldehyde/acetyl ratioc
0.085
0.046
0.080
0.031
Liquid analysisd/%mol mol−1
Water
14.4
78.5
3.0
3.9
0.2
35
17.6
73.1
2.8
6.2
0.4
30
12.3
82.9
3.3
1.4
0.1
12
14.7
78.9
3.2
3.1
0.1
10
11.9
83.7
3.2
1.2
0.1
Methanole
Methyl iodide
Methyl acetate
Acetic acid
MeCHO (ppm)
Gas analysisf/%vol vol−1
H2
<10
0.6
82.5
0.2
<0.1
86.9
0.09
0.08
<0.1
85.8
<0.1
<0.1
4.9
84.2
<0.02
0.4
<0.1
89.0
0.04
CO
CO2
CH4
0.4
<0.05
a Reaction conditions: Rh2CO4Cl2 (0.386 mmol), Rh : ligand = 1 : 1, CO p = 26 bar, for 40 min at 140 ◦C. b TOF was calculated from the total conversion
of methanol and methyl iodide to methyl acetate and acetic acid. c Aldehyde/acetyl ratio was calculated from the ratio of acetaldehyde to carbonylated
products. d Calculated by calibrated GC analysis using an internal standard at BP, Hull. e %mol mol−1 of methanol also comprises ∼5–7 %mol mol−1
DME. f Sampled at room temperature from headspace after cool down.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5582–5589 | 5585
©

Table 4 Methanol carbonylation at 150 ◦C, 15% H2O and 5 : 1 CO : H2
a
Ligand
None
dppb
857
dppx
387
dcpb
690
BINAP
TOFb/h−1
740
3.20
426
0.92
Aldehyde/acetyl ratioc
Liquids analysisd/%mol mol−1
Water
2.59
0.74
2.30
45.4
49.8
1.5
2.7
0.6
44.5
50.2
1.4
3.2
0.7
42.9
53.8
1.5
1.5
0.3
43.9
51.3
1.6
2.6
0.6
43.5
53.0
1.6
1.7
0.3
Methanole
Methyl iodide
Methyl acetate
Acetic acid
MeCHO (ppm)
Gas analysisf/%vol vol−1
H2
1100
1040
135
750
185
37.2
44.4
2.3
27
43.6
1.8
27.0
69.6
0.5
31.4
42.1
1.1
20.2
78.2
0.3
CO
CO2
CH4
g
14.4
25.8
1.7
23.9
<0.1
a Reaction conditions: Rh2CO4Cl2 (0.386 mmol), Rh : ligand = 1 : 1, CO/H2 = 5 : 1, p = 26 bar, for 40 min at 150 ◦C with an initial water concentration
of 15%. b TOF was calculated from the total conversion of methanol and methyl iodide to methyl acetate and acetic acid. c Aldehyde/acetyl ratio was
calculated from the ratio of acetaldehyde to carbonylated products. d Calculated by calibrated GC analysis using an internal standard at BP, Hull. e %mol
mol−1 of methanol also comprises 5–7 %mol mol−1 DME. f Sampled at room temperature from headspace after cool down. g %vol vol−1 of methane
typically represent between 0.0001–0.6 %mol mol−1 converted from the initial %mol mol−1 of methanol and MeI.
Table 5 % of catalyst recovered under different syngas conditions
shielding these must give to the metal centre. Alternatively, given
that we have been unable to isolate Rh species for dcpb in
stoichiometric reactions and much lower recovery at the end of the
catalytic runs had occurred, it is also possible that dcpb catalysts
decompose under the reaction conditions.
For dppx and dppb on the other hand, it was possible to collect
a substantial amount of precipitate formed at the end of each
catalytic run. These precipitates were fully analysed and the resting
state shown to be the Rh(III) species, [Rh(L)(C(O)Me)(I)2]. The
most stable catalyst system is dppx which manages 85% recovery
as a precipitate under catalytic conditions at 140 ◦C, and was also
seen to be the most stable in the previous study in hot MeI. In the
runs carried out in aqueous methanol at 150 ◦C, a similar crop of
[Rh(dppx)(C(O)Me)(I)2] was obtained (79%) in comparison to the
dppb catalyst that provided very little of the Rh acetyl precipitate.
This is most likely due to decomposition of dppb complexes at
higher temperatures.
2 : 1 H2 : CO 140 ◦C, 40 min
5 : 1 H2 : CO 150 ◦C, 40 min
dppb
dppx
dcpb
60.2
85
27.8
63.9
2.7
78.7
5.9
BINAP
53.8
Catalyst resting states and catalyst stability
On cooling, the catalysts precipitated out of solution. It should be
noted that the % recovery of precipitated catalyst does not com-
pletely decouple stability from solubility differences so the results
contained in Table 5 must be interpreted cautiously. However, since
all the Rh complexes involved show very limited solubility in the
product mix it is very likely that these reflect catalyst stability. The
precipitate collected from the BINAP systems was shown to be
the Rh(I) carbonyl iodide species [Rh(BINAP)CO(I)]. Although
HPIR studies during the reaction were inconclusive, given that we
also observed a lower stability of BINAP Rh(III) acetyls in the
coordination studies, [Rh(BINAP)(CO)I] is the proposed resting
state. The higher stability of the BINAP catalyst under catalytic
conditions can be ascribed to the presence of CO which enables
the system to turnover and produce a Rh(I) resting state.
The Rh/dcpb catalyst that precipitated was only sparingly
soluble with the 31P NMR showing the presence of a Rh(III) species
with a typical P–Rh coupling of 138 Hz. However, FT-IR of the
bulk sample showed that the majority to be the Rh(I) species
[Rh(dcpb)CO(I)] (1949 cm−1) with a minor peak at 1710 cm−1
relating to the acetyl. Therefore, it is likely that the nature of the
catalyst resting state is finely balanced. However, it is noted that
the recovery of these sparingly soluble species was relatively low.
One may possibly expect the dcpb system to be much more
active than dppb due to the strongly electron donating properties
of the phosphines which may facilitate oxidative addition, with
the bulky groups favouring reductive elimination. However, a
reasonable explanation for similar performance is the increase
in steric hindrance of the cyclohexyl groups and the considerable
Reaction of dppp, dppb and dppx complexes with hydrogen
The greater hydrogen tolerance of the dppx catalyst could reflect
a more efficient migratory insertion/reductive elimination for this
ligand, or a reduced propensity for the complexes to undergo
hydrogenolysis. According to Moloy and Wegman it is possible to
generate the hydride dimer and acetaldehyde (as dimethyl acetal)
by reacting [Rh(dppp)(C(O)Me)(I)2] with hydrogen (Scheme 3).20
We have investigated ligand effects on this by reacting the isolated
acetyls [Rh(dppp)(C(O)Me)(I)2], [Rh(dppb)(C(O)Me)(I)2] and
[Rh(dppx)(C(O)Me)(I)2] with hydrogen under similar conditions
set out above.
Scheme 3 Reaction of [Rh(dppp)COMe(I)2] with H2 and the formation
of the hydride dimer.
5586 | Dalton Trans., 2007, 5582–5589
This journal is
The Royal Society of Chemistry 2007
©

[Rh(dppp)(C(O)Me)(I)2] was heated at 120 ◦C for 5 h under
7 bar of H2, this yielded a complex with the same spectroscopic
data as previously described for the hydride dimer (31P NMR
stability and hydrogen tolerance of phosphine modified rhodium
carbonylation catalysts are ongoing.
d 27.3, d, J
= 125 Hz). Under these conditions we also
Rh–P
Experimental
observed the hydrogenolysis of [Rh(dppb)(C(O)Me)(I)2] and the
formation of several new complexes that show hydride signals
General
1
in the H NMR. Unfortunately we were unable to recover this
new complex in an analytically pure form. Interestingly, when
[Rh(dppx)(C(O)Me)(I)2] was reacted under the same forcing
conditions no reaction occurred and the starting material was
recovered in quantitative yield.
All manipulations were performed using standard Schlenk line
techniques under nitrogen supplied by BOC. All chemicals and
solvents were obtained through commercial sources. All complexes
used were prepared by standard literature procedures. Microanal-
yses were by the University of St. Andrews microanalytical service.
NMR spectra were recorded on Bruker Advance 300 instruments.
Chemical shifts are reported in ppm with 1H and 13C NMR spectra
referenced to tetramethylsilane (external). 31P NMR spectra were
referenced externally to 85% H3PO4. Proton and carbon signal
multiplicities are given as s (singlet), d (doublet), t (triplet), q
(quartet), m (multiplet), b (broad) or a combination of these.
All spectra were recorded at room temperature and the solvent
for a particular spectrum is given in parentheses. 13C and 31P
NMR spectra were recorded with broad band proton decoupling.
IR spectra were recorded using a Perkin Elmer Spectrum GX
FT-IR system. All solids were analysed as KBr disks. Gas
chromatography was carried out at BP PLC, Saltend. Mass
spectroscopy data were obtained from the EPSRC National Mass
Spectroscopy Service Centre, Swansea. In addition to detecting the
ions shown, all Rh complexes showed excellent agreement between
calculated and expected isotope patterns.
When this experiment was repeated at the lower temperature
of 80 ◦C the complexes containing dppp and dppb react to form
Rh–hydrides. However, [Rh(dppx)(C(O)Me)(I)2] once again does
not react with hydrogen and in fact at the lower temperature it
starts to equilibrate back to the Rh(I) species [Rh(dppx)CO(I)].
This is backed up by both 31P NMR data as well as FT-IR of
the mixture, that clearly shows peaks for both the acetyl and the
Rh(I) carbonyl. Interestingly, the [Rh(dppx)CO(I)] being formed
by this reverse process is a monomeric chelate species as opposed
to a dimeric bridging species formed from Rh2CO4Cl2. There is
therefore a clear difference in the reactivity of the Rh(III) acetyls
with hydrogen that co-incides with the observations under catalytic
conditions.
One possible explanation for hydrogen tolerance can be seen
when the crystal structures of the dppb and dppx acetyls are
compared. The more rigid dppx backbone forces the phenyl groups
inwards increasing the steric hindrance around the Rh–acetyl
bond. It seems reasonable that this more congested environment
around the metal centre coupled with the fact that dppx cannot
easily change conformation to accommodate incoming reactants,
accounts for the differences in catalytic products.
Catalysis experiments
These experiments were carried out at BP chemicals Ltd in
Hull in a 300 cm3 Hastelloy high pressure infrared (HPIR)
autoclave fitted with a bursting disc, catalyst injector, overhead
MagnedriveTM stirrer, impeller with gas sparging facility, gas
inlet, high pressure (195 bar) carbon monoxide reservoir and
thermocouple. Additionally the apparatus featured equipment
for pressure measurements and sampling. Carbon monoxide
consumption was measured by monitoring the pressure drop in
the reservoir. A computer collected the data automatically by use
of a Dynamic Data Exchange (DDE) ink between an Orsi process
control package and a data logging package on Excel.
Conclusion
In summary, Rh(I) carbonyl complexes of the C4-diphosphines
dppb, BINAP, dcypb and dppx have been prepared and their
reactivity studied. These studies reveal that only the dppb
and dppx complexes smoothly give the expected Rh(III) acetyl
complexes on reaction with a large excess of methyl iodide at
140 ◦C. The complexes have been studied as catalysts for methanol
carbonylation using CO that contains hydrogen. Examination of
the insoluble catalyst residues from these reactions shows that
dppx and dppb have Rh(III) acetyls as resting states and that the
dppx complex is more stable with respect to elimination of P(V)
by-products. In contrast, the dcpb system consisted of both Rh(I)
carbonyl and Rh(III) acetyl species. By-product analysis of the
organic carbonylation products reveals that the more rigid BINAP
and dppx systems give much lower proportions of acetaldehyde
and methane side products. An explanation for this comes from
the relative reactivity of the Rh(III) acetyls with hydrogen: whereas
dppb and dppp react with hydrogen to give various hydride species,
there is no reaction for the dppx complex. Examination of the
X-ray crystal structures of these two complexes suggest that the
rigidity of the dppx backbone makes the diphenylphosphine group
shield the acetyl from hydrogenolysis, which may be the ultimate
origin of the greater hydrogen tolerance of the dppx carbonylation
catalysts. Further studies to gain a greater understanding of the
CAUTION: These experiments were carried out using custom
made facilities dedicated to methanol carbonylation. MeI and CO
are extremely toxic and should be used with extreme care.
For the reactions carried out with an initial water level of
15% w/w, methanol (100 g, 3.125 mol), water (19 g, 1.06 mol),
Rh2CO4Cl2 (150 mg, 0.386 mmol) and diphosphine ligand (2 eq,
0.772 mmol) were added to the autoclave under CO/syngas.
The quantity of water was altered for the remaining reactions
incorporating no initial water content. The autoclave was then
sealed and pressurised with carbon monoxide (5 bar) and heated to
100 ◦C, controlled by the use of a thermocouple in a thermowell in
the reaction solution, whilst stirring at 900 rpm. Using the catalyst
injector iodomethane (22.6 g, 0.159 mol) was then added to the
autoclave with an over pressure of CO/syngas. Simultaneously,
the autoclave was pressurised with CO/syngas (27 bar) from the
high pressure ballast vessel. The autoclave was then heated to
the desired temperature and the data collection activated. The
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5582–5589 | 5587
©

pressure was maintained at 27 bar by a pressure regulating value
which transferred CO/syngas from the high pressure ballast to the
autoclave when the pressure dropped below the set-point.
After the reaction, the autoclave was allowed to cool before a
gas sample was collected from the headspace in the autoclave. This
gas sample was analysed by gas chromatography using in-house
facilities.
The autoclave was then disconnected from the kinetics rig and
a liquid sample of the reaction solution was analysed by gas
chromatography. Liquid analysis was carried out for the main
constituents, that fall into the bulk liquids category, as well as a
range of liquid by-products such as acetaldehyde.
This was stirred for 10 min before adding the aldehyde (100 eq.) in
THF (0.5 cm3). The solution was then heated at 110 ◦C for 10 min
in a BiotageTM microwave and the resulting yellow solution was
then analysed.
Note: Several of these complexes were too insoluble to record
meaningful 13C NMR data.
29
Synthesis of [Rh(dppb)CO(Cl)]2
.
Yield: 90% (136.8 mg,
0.116 mmol); IR (KBr), v(CO)/cm−1: 1959.6. LSIMS Found
1184.2 M+; C58H56Cl2O2P4Rh2 requires 1184.1 M+. 31P{ H} NMR
1
(CD2Cl2): d 24.0 (d, 1JRh–P 124 Hz).
Synthesis of [Rh(BINAP)CO(Cl)]25. Yield: 79% (95.6 mg,
0.121 mmol). IR (KBr), v(CO)/cm−1: 2007.1. LSIMS (FAB)
Found 760.1 [M − CO]+; C44H32Cl1P2Rh1 requires 760.1 [M −
X-Ray crystallography
1
CO]+. 31P{ H} NMR (CD2Cl2): d 46.8 (dd, 1JRh–P 161.1 Hz, d 25.0
X-Ray crystallography data were collected at 93 K by using
a Rigaku MM007 High Brilliance RA generator and Mer-
cury/Saturn CCD systems using Mo-Ka radiation. Intensities
were corrected for Lorentz-polarisation and for absorption. The
structures were solved by direct methods. All hydrogen atoms were
refined as idealised riding geometries and structural refinements
were obtained with full-matrix least-squares based on F2 by using
the program SHELXTL.
(dd, 1JRh–P 127.7 Hz).
Synthesis of [Rh(dppx)CO(Cl)]2. Yield: 79.3% (48.4 mg,
0.041 mmol). IR (KBr), v(CO)/cm−1: 1970.3. LSIMS (FAB)
Found 1224.1 [M − 2CO]+, 1245.2 [M − Cl]+; C64H56Cl2P4Rh2
1
requires 1224.1 [M − 2CO] +. 31P{ H} NMR (CD2Cl2): d 30.7 (d,
1JRh–P 127.7 Hz). 1H NMR (CD2Cl2): dH 8.00–6.00 (28H, m), 3.85
(1H, s), 3.60 (1H, s), 2.95 (2H, s), 1.42 (2H, s), 1.25 (1H, s), 1.20
(1H, s).
Crystal structure determination.
Synthesis of [Rh(dcpb)CO(Cl)]2. Yield: 74% (117 mg,
0.095 mmol). IR (KBr), v(CO)/cm−1: 1949.4. LSIMS (FAB)
Found 1232.3 M+; C58H104Cl2O2P4Rh2 requires 1232.4 M+. Anal.
calc. for C58H104Cl2O2P4Rh2: C, 56.45; H, 8.49%. Found: 56.22; H,
Crystal data for [Rh(BINAP)(CO)I]. C45H32IOP2Rh, M =
880.46, triclinic, a = 11.1576(12), b = 11.5084(16), c =
◦
◦
◦
˚
14.6153(15) A, a = 84.203(7) , b = 71.704(7) , c = 89.593(10) ,
3
¯
˚
V = 1772.1(4) A , T = 93(2) K, space group P1, Z = 2, 11 550
reflections measured, 6105 unique of which 4758 were observed
[I < 2rI] (R1 = 0.0474, wR2 = 0.0865, Rint = 0.0409).
1
1
1
8.20%. 31P{ H} NMR (CD2Cl2): d 31.5 (d, JRh–P 118.8 Hz). H
NMR (CD2Cl2): dH 1.99 (16H, bm), 1.74 (48H, bm), 1.46 (16H,
bm), 1.188 (16H, bm).
Crystal data for [Rh(dppb)(COMe)I2]. C30H31I2OP2Rh·
[C3H6O2]0.5, M = 863.24, monoclinic, a = 21.540(4), b =
Synthesis of [Rh(BINAP)CO(I)]
◦
3
˚
˚
8.1514(14), c = 18.908(4) A, b = 96.662(3) , V = 3297.5(11) A ,
T = 93(2) K, space group P2(1)/c, Z = 4, 22 888 reflections
measured, 5963 unique of which 5162 were observed [I < 2rI]
(R1 = 0.0258, wR2 = 0.0519, Rint = 0.0310).
Acetone (5 ml) was added to a Schlenk tube containing
[Rh(BINAP)CO(Cl)] (50 mg, 0.063 mmol) and NaI (1.9 eq.). This
resulted in a colour change from a yellow to an orange solution
which was filtered and dried under vacuum to give the iodo-
complex [Rh(BINAP)CO(I)]. Yield: 96% (53.6 mg, 0.061 mmol).
(IR (KBr), v(CO)/cm−1: 2007.2. LSIMS (FAB) Found 851.9 [M −
Crystal data for [Rh(dppx)(COMe)I2]. C34H31I2OP2Rh·
CHCl3, M = 993.61, orthorhombic, a = 24.241(4), b = 18.209(3),
3
˚
˚
c = 8.1431(14) A, V = 3594.5(11) A , T = 93(2) K, space group
Pnma, Z = 4, 21819 reflections measured, 3338 unique of which
1
CO]+; C44H32IP2Rh requires 852.0 [M − CO]+. 31P{ H} NMR
(CD2Cl2): d 42.9 (dd, 1JRh–P 166 Hz, 2JP–P 42.3 Hz), 21.9 (dd, 1JRh–P
126 Hz, 2JP–P 42.3 Hz). This compound was further characterised
by X-ray crystallography.
2388 were observed [I < 2rI] (R1 = 0.0641, wR2 = 0.1305, Rint
=
0.1380).
General synthesis of [Rh(L)CO(Cl)]2
Synthesis of [Rh(dppb)(C(O)Me)(I)2]30
Method A. THF (2 cm3) was charged into a Schlenk tube
containing the diphosphine (2 eq.) and the precursor Rh2CO4Cl2.
After stirring for a short amount of time a pale yellow solid
precipitated which was filtered off and washed with ice cold
methanol before drying under vacuum. Yields quoted below are
all using method A.
1.5 ml methyl acetate and 0.5 ml methyl iodide was added to a
2.0 ml BiotageTM microwave vial containing [Rh(dppb)CO(Cl)]2
(15 mg, 0.013 mmol). The reaction mixture was then heated
to 140 ◦C for 10 min. High quality X-ray crystals were then
formed from the slow evaporation of this solution. Yield: 86%
(17.8 mg, 0.022 mmol). IR (KBr), v(CO)/cm−1: 1701.8. LSIMS
(FAB) Found 825.8 M+; C30H31I2OP2Rh requires 825.9 M+. Anal.
calc. for C30H31I2OP2Rh: C, 43.61; H, 3.78%. Found: C, 44.06; H,
Method B. THF (2 cm3) was charged into a Schlenk tube
containing the diphosphine (2 eq.) and the precursor [Rh(cyclo-
octene)Cl]2. CO was then bubbled through the solution resulting
in a pale yellow solid which was washed with ice cold methanol
before drying under vacuum.
1
1
1
3.76%. 31P{ H} NMR (CD2Cl2): d 32.9 (d, JRh–P 138.1 Hz. H
NMR (CD2Cl2): dH 7.58 (4H, CH-Ar, dt), 7.23–7.41 (12H, CH-
Ar, m), 3.19 (2H, CH2, m), 2.58 (3H, COCH3, s), 2.35 (2H, CH2,
m), 1.49 (2H, CH2, m), 0.98 (2H, CH2, m). 13C NMR (CD2Cl2): d
212.85 (COMe, dt, 1JRh–C 25.77 Hz), 137.27, 136.54, 130.09, 130.27
Method C. THF (1.5 cm3) was charged into a microwave vial
containing the diphosphine (2 eq.) and the precursor [Rh(coe)Cl]2.
5588 | Dalton Trans., 2007, 5582–5589
This journal is
The Royal Society of Chemistry 2007
©

(C-Ar), 136.44–127.30 (CH-Ar), 44.85 (CH2, s), 23.53 (CH2, d,
1JRh–C 28.64 Hz), 21.46 (CH3, s).
6 J. G. Sunley, in Manufacture of acetic acid by the iridium-catalyzed
carbonylation using one or more metal iodides or iodide-generating
salts or complexes, Br. Pat., GB 2003–16681, 2003.
7 M. J. Baker, M. F. Giles, A. G. Orpen, M. J. Taylor and R. J. Watt,
J. Chem. Soc., Chem. Commun., 1995, 197.
Synthesis of [Rh(dppx)(C(O)Me)(I)2]
8 C.-A. Carraz, A. G. Orpen, D. D. Ellis, P. G. Pringle, E. J. Ditzel and
G. J. Sunley, Chem. Commun., 2000, 1277.
9 J. R. Dilworth, J. R. Miller, N. Wheatley, M. J. Baker and J. G. Sunley,
J. Chem. Soc., Chem. Commun., 1995, 1579.
10 D. K. Dutta, J. D. Woollins, A. M. Z. Slawin, D. Konwar, P. Das, M.
Sharma, P. Bhattacharyya and S. M. Aucott, Dalton Trans., 2003, 2674.
11 P. M. Maitlis, A. Haynes, G. J. Sunley and M. J. Howard, J. Chem. Soc.,
Dalton Trans., 1996, 2187.
12 J. Rankin, A. C. Benyei, A. D. Poole and D. J. Cole-Hamilton, J. Chem.
Soc., Dalton Trans., 1999, 3771.
13 J. Rankin, A. D. Poole, A. C. Benyei and D. J. Cole-Hamilton, Chem.
Commun., 1997, 1835.
14 Z. Freixa, P. C. J. Kamer, M. Lutz, A. L. Spek and P. W. N. M. van
Leeuwen, Angew. Chem., Int. Ed., 2005, 44, 4385.
15 C. M. Thomas, R. Mafua, B. Therrien, E. Rusanov, H. Stoeckli-Evans
and G. Suss-Fink, Chem.–Eur. J., 2002, 8, 3343.
1.5 ml methyl acetate and 0.5 ml methyl iodide was added to a
2.0 ml BiotageTM microwave vial containing [Rh(dppx)CO(Cl)]2
(15 mg, 0.012 mmol). The reaction mixture was then heated
to 140 ◦C for 10 min. High quality X-ray crystals were then
formed from the slow evaporation of this solution. Yield: 78.5%
(56.0 mg, 0.064 mmol). IR (KBr), v(CO)/cm−1: 1710.6. LSIMS
(FAB) Found 873.8 M+, 746.9 [M − I]+; C34H31I2OP2Rh requires
873.9 M+. Anal. calc. for C34H31I2OP2Rh: C, 46.71; H, 3.57%.
1
Found: C, 46.56; H, 3.31%. 31P{ H} NMR (CD2Cl2): d 11.0 (d,
1JRh–P 138.7 Hz). 1H NMR (CD2Cl2): dH 7.86 (4H, CH, dt), 7.23–
7.45 (12H, CH, m), 6.43 (4H, CH2, m), 6.09 (2H, CH2, m), 5.02
(2H, CH2, m), 3.34 (2H, CH2, m), 2.38 (3H, CH3, s).
16 C. M. Thomas and G. Suss-Fink, Coord. Chem. Rev., 2003, 243, 125.
17 S. Burger, B. Therrien and G. Suess-Fink, Helv. Chim. Acta, 2005, 88,
478.
Acknowledgements
18 S. Gaemers and J. G. Sunley, in Carbonylation process using metal-
polydentate ligand catalysts, Pat. Appl., WO 2004101487, 2004.
19 K. G. Moloy and J. L. Petersen, Organometallics, 1995, 14, 2931.
20 K. G. Moloy and R. W. Wegman, Organometallics, 1989, 8, 2883.
21 A. R. Sanger, J. Chem. Soc., Chem. Commun., 1975, 893.
22 J. A. Osborn and G. Wilkinson, Inorg. Synth., 1967, 10, 67.
23 D. Evans, J. A. Osborn and G. Wilkinson, Inorg. Synth., 1968, 11, 99.
24 M. L. Clarke and G. J. Roff, Chem.–Eur. J., 2006, 12, 7978.
25 K. A. Bunten, D. H. Farrar, A. J. Poe and A. Lough, Organometallics,
2002, 21, 3344.
We would like to thank Glenn Sunley for many fruitful discussions.
EPSRC and BP for an industrial CASE studentship, EPSRC
National Mass Spectroscopy Service, Caroline Horsborough from
St. Andrews MS, Melanya Smith and Thomas Lebel from the
NMR services department at the University of St. Andrews.
References
26 S. Jeulin, S. Duprat de Paule, V. Ratovelomanana-Vidal, J.-P. Genet, N.
Champion and P. Dellis, Angew. Chem., Int. Ed., 2004, 43, 320.
27 K. Toriumi, T. Ito, H. Takaya, T. Souchi and R. Noyori, Acta
Crystallogr., Sect. B, 1982, 38, 807.
1 F. E. Paulik and J. F. Roth, Chem. Commun. (London), 1968, 1578.
2 A. Haynes, P. M. Maitlis, G. E. Morris, G. J. Sunley, H. Adams, P. W.
Badger, C. M. Bowers, D. B. Cook, P. I. P. Elliott, T. Ghaffar, H. Green,
T. R. Griffin, M. Payne, J. M. Pearson, M. J. Taylor, P. W. Vickers and
R. J. Watt, J. Am. Chem. Soc., 2004, 126, 2847.
28 T. K. Hollis, L.-S. Wang and F. Tham, J. Am. Chem. Soc., 2000, 122,
11737.
29 A. R. Sanger, J. Chem. Soc., Dalton Trans., 1977, 120.
30 A. Haynes, P. M. Maitlis, I. A. Stanbridge, S. Haak, J. M. Pearson, H.
Adams and N. A. Bailey, Inorg. Chim. Acta, 2004, 357, 3027.
31 H. Adams, N. A. Bailey, B. E. Mann and C. P. Manuel, Inorg. Chim.
Acta, 1992, 198–200, 111.
32 L. Gonsalvi, H. Adams, G. J. Sunley, E. Ditzel and A. Haynes, J. Am.
Chem. Soc., 2002, 124, 13.
3 G. J. Sunley and D. J. Watson, Catal. Today, 2000, 58, 293.
4 A. Haynes, D. J. Law, A. Miller, G. E. Morris, M. J. Payne and J. G.
Sunley, in Carbonylation catalyst and process for the production of
acetic acid from methanol, Br. Pat., WO 2005009939, 2005.
5 L. A. Key, M. J. Payne and A. D. Poole, in Catalytic carbonylation
process for the production of acetic acid from methanol and carbon
monoxide, Br. Pat., 2004026805, 2004.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5582–5589 | 5589
©