Influence of phosphorus and oxygen donor diphosphine ligands on the reactivity of rhodium(I) carbonyl complexes

Article abstract of DOI:10.1016/j.molcata.2010.05.008

The dimeric rhodium precursor [Rh(CO)₂Cl]₂ reacts with two molar equivalent of 9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene [xantphos] (a), bis(2-diphenylphosphinophenyl)ether [DPEphos] (b) and their corresponding dioxide analogues xantphos dioxide (c), DPEphos dioxide (d) to afford the mono- and dicarbonyl complexes of the type [Rh(CO)Cl(L)] (1a,1b) and [Rh(CO)₂Cl(L)] (1c,1d) respectively, where L = a-d. The complexes 1a-1d have been characterized by elemental analyses, IR and NMR (¹H, ³¹P and ¹³C) spectroscopy, and the structure of the ligand d was determined by single crystal X-ray diffraction. 1a-1d undergo oxidative addition (OA) reactions with different electrophiles such as CH₃I, C₂H₅I and I₂ to give Rh(III) complexes of the types [Rh(CO)_y(COR)ClXL] {R = -CH₃ (2a-2d), -C₂H₅ (3a-3d); X = I and y = 0, L = a, b; y = 1, L = c, d} and [Rh(CO)ClI₂L] (4a-4d) respectively. Kinetic data for the reactions of 1a-1d with CH₃I indicate a pseudo-first-order reaction. The catalytic activity of 1a-1d for the carbonylation of methanol to acetic acid and its ester was evaluated at different CO pressure 15, 20 and 33 bar at 130 °C and a higher Turn Over Number (TON) (679-1768) were obtained compared to that of the well-known commercial species [Rh(CO)₂I₂]^- (TON = 463-1000) in each case under the similar experimental conditions.

Full text of DOI:10.1016/j.molcata.2010.05.008

Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Editor’s choice paper
Influence of phosphorus and oxygen donor diphosphine ligands on the reactivity
of rhodium(I) carbonyl complexes
Biswajit Deb, Dipak Kumar Dutta∗
Materials Science Division, North East Institute of Science and Technology (Council of Scientific and Industrial Research), Jorhat 785 006, Assam, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The dimeric rhodium precursor [Rh(CO)2Cl]2 reacts with two molar equivalent of 9,9-dimethyl-4,5-
bis(diphenylphosphino)xanthene [xantphos] (a), bis(2-diphenylphosphinophenyl)ether [DPEphos] (b)
and their corresponding dioxide analogues xantphos dioxide (c), DPEphos dioxide (d) to afford the mono-
and dicarbonyl complexes of the type [Rh(CO)Cl(L)] (1a,1b) and [Rh(CO)2Cl(L)] (1c,1d) respectively, where
L = a–d. The complexes 1a–1d have been characterized by elemental analyses, IR and NMR (1H, 31P and
13C) spectroscopy, and the structure of the ligand d was determined by single crystal X-ray diffraction.
1a–1d undergo oxidative addition (OA) reactions with different electrophiles such as CH3I, C2H5I and I2
to give Rh(III) complexes of the types [Rh(CO)y(COR)ClXL] {R = –CH3 (2a–2d), –C2H5 (3a–3d); X = I and
y = 0, L = a, b; y = 1, L = c, d} and [Rh(CO)ClI2L] (4a–4d) respectively. Kinetic data for the reactions of 1a–1d
with CH3I indicate a pseudo-first-order reaction. The catalytic activity of 1a–1d for the carbonylation of
methanol to acetic acid and its ester was evaluated at different CO pressure 15, 20 and 33 bar at 130 ◦C and
a higher Turn Over Number (TON) (679–1768) were obtained compared to that of the well-known com-
mercial species [Rh(CO)2I2]− (TON = 463–1000) in each case under the similar experimental conditions.
Received 23 February 2010
Received in revised form 12 May 2010
Accepted 14 May 2010
Available online 11 June 2010
Keywords:
Rhodium
Carbonyl complexes
Oxygen donor
Diphosphine
Carbonylation
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Monsanto’s catalyst [1–17,22–24]. As a part of our continuing
research activity [13–17,25–27], we have chosen two diphosphine
Oxidative addition to organometallic compounds is one of the
key elementary steps in many catalytic processes. For this rea-
son, the reaction has been extensively studied, and in the last few
years many publications have appeared concerning its implication
to ligand-promoted methanol carbonylation to acetic acid produc-
tion [1–17]. The original [Rh(CO)2I2]− catalyst, developed at the
Monsanto’s laboratories [18] and studied in detail by Forster and
co-workers [19,20], is largely used for the industrial production
of acetic acid. It is commonly accepted that the rate determining
step of this reaction is precisely the oxidative addition of MeI to
a square planar d8 complex [21] and therefore the major focus
has been made on the design of catalysts for the improvement
of this reaction. Ligands that increase the electron density at the
metal center should facilitate the oxidative addition step and, con-
sequently, increase the overall rate of acetic acid formation.
ligands viz. 9,9-dimethyl-4,5-bis-(diphenylphosphino)xanthene
(xantphos) and bis(2-diphenylphosphinophenyl)ether (DPEphos)
with different ligand backbone and made them oxygen functional-
ized by oxidation of the phosphorus atoms. In this paper, we report
the synthesis of four rhodium(I) carbonyl complexes containing the
‘Soft’ (phosphorus) and ‘Hard’ (oxygen) donor diphosphine ligands
and their reactivity with different electrophiles like CH3I, C2H5I and
I2. The effects of backbone and donor capacity of the ligands on cat-
alytic carbonylation of methanol under different CO pressure have
also been studied.
2. Experimental
2.1. General definition
For this purpose, a large variety of rhodium carbonyl complexes
have been synthesized by incorporating different ligands into its
coordination sphere and evaluated for methanol carbonylation,
giving comparable or better activities compared to the original
All solvents were distilled under N2 prior to use. RhCl3·xH2O was
purchased from M/S Arrora Matthey Ltd., Kolkota, India. The ligands
xantphos and DPEphos were purchased from Across Organics, Bel-
gium and used without further purification. H2O2 was obtained
from Ranbaxy, New Delhi, India and estimated before use.
Elemental analyses of C and H were performed on a Perkin-
Elmer 2400 elemental analyzer. The elements P, Cl and Rh were
analyzed quantitatively by standard analytical techniques [28,29],
∗
Corresponding author. Tel.: +91 376 2370081; fax: +91 376 2370011.
E-mail address: dipakkrdutta@yahoo.com (D.K. Dutta).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.05.008

22
B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
1
180.3 (s, CO). 31P{ H} NMR (CDCl3, ppm): ı 45.5 (d, JP–P = 69.2 Hz),
ı 48.04 (d, JP–P = 69.2 Hz). Elemental analyses: Found (Cald. for
C41H32O5P2ClRh), C 60.73 (61.17); H 3.78 (3.98); P 8.01 (7.70); Cl
4.59 (4.41); Rh 12.60 (12.79); O 10.29 (9.94).
and the quantity of O was determined by difference. IR spectra
(4000–400 cm−1) were recorded in KBr discs and CHCl3 on a Perkin-
Elmer system 2000 FT-IR spectrophotometer. The 1H, 13C and 31P
NMR spectra were recorded at room temperature in CDCl3 solu-
tion on a Bruker DPX-300 Spectrometer and chemical shifts were
reported relative to SiMe4 and 85% H3PO3 as internal and exter-
nal standards respectively. Mass spectra of the complexes were
recorded on ESQUIRE 3000 Mass Spectrometer. The carbonylation
reactions of methanol were carried out in a high pressure reac-
tor (Parr-4592, USA) fitted with a pressure gauge and the reaction
products were analyzed by GC (Chemito 8510, FID).
[Rh(CO)2Cl(DPEphos dioxide)] (1d)
IR (KBr, cm−1): 1996, 2071 [ꢀ(CO)], 1189, 1186 [ꢀ(P O)]. 1H
NMR (DMSO-D6, ppm): ı 6.15–6.38, 7.11–7.82 (m, Ph). 13C NMR
(DMSO-D6, ppm): ı 119.2–138.5 (m, Ar), ı 153.8 (s, O–Cphenyl), ı
182.5, 185.9 (s, CO). 31P{ H} NMR (DMSO-D6, ppm): ı 35.8, 24.7
1
(s, P O). Elemental analyses: Found (Cald. for C38H28O5P2ClRh), C
58.98 (59.65); H 3.51 (3.66); P 8.32 (8.10); Cl 4.41 (4.64); Rh 13.82
(13.46); O 10.96 (10.47).
2.2. Synthesis of xantphos dioxide (c) and DPEphos dioxide (d)
2.5. Reactivity of [Rh(CO)ClL] (1a,1b) and [Rh(CO)2ClL] (1c,1d)
The ligands xantphos and DPEphos dioxide were synthesized by
oxidation of xantphos and DPEphos respectively by H2O2 following
the literature protocol [25,27]
with CH3I, C2H5I and I2
2.5.1. Synthesis of [Rh(CO)y(COR)ClXL] R = CH3, X = I, y = 0
Analytical data:
Xantphos dioxide (c)
(2a,2b), y = 1 (2c,2d); R = C2H5, X = I, y = 0 (3a,3b), y = 1(3c,3d)
[Rh(CO)2ClL] (50 mg) was dissolved in dichloromethane (5 cm3)
and each of RX (3 cm3) (RX = CH3I, C2H5I) was added to it. The reac-
tion mixture was then stirred at r.t. for about 2–10 h for CH3I and
C2H5I respectively. The colour of the solution changed from yellow-
ish red to dark reddish brown and the solvent was evaporated under
vacuum. The compounds so obtained were washed with diethyl
ether and stored over silica gel in a desiccator.
IR (KBr, cm−1): 1193 [ꢀ(P–O)]. 1H NMR (CDCl3, ppm): ı
6.74–7.74 (m, 30H, Ph), 2.17 (s, 6H, CH3). 13C NMR (CDCl3, ppm): ı
153.10–124.1 (m, Ar), ı 32.3, 34.9 (s, CH3). 31P{ H} NMR (CDCl3,
1
ppm): ı 30.95 [s, P O]. Elemental analyses; Found (Cald. for
C
39H32O3P2): C 73.98 (74.40); H 5.02 (5.08).
DPEphos dioxide (d)
IR (KBr, cm−1): 1198, 1186 [ꢀ(P–O)]. 1H NMR (DMSO-D6, ppm):
ı 6.11–6.18, 7.2–7.69 (m, 28H, Ar), 13C NMR (DMSO-D6, ppm): ı
2.5.2. Synthesis of [Rh(CO)ClI2L] (4a–4d)
122.5–134.21 (m, Ar), ı 153.4 (s, O–Cphenyl). 31P{ H}NMR (DMSO-
1
[Rh(CO)2ClL] (50 mg) was dissolved in CH2Cl2 (15 cm3) and to
this solution I2 (25 mg) was added. The reaction mixture was then
stirred at r.t. for about 4 h. The solvent was evaporated under vac-
uum and the brown coloured compound so obtained were washed
with hexane and stored over silica gel in a desiccator.
D6, ppm): ı 26.21, 24.21 [s, P O]. Elemental analyses; Found (Cald.
for C36H28O3P2): C 75.18 (75.71); H 4.73 (4.90).
2.3. Synthesis of starting material
[Rh(CO)2Cl]2 was prepared by passing CO gas over RhCl3·3H2O
2.6. X-ray structural analysis
at 100 ◦C in the presence of moisture [30].
Single crystal of d was grown by slow evaporation of a satu-
rated solution of d in acetone. The intensity data of the compounds
were collected on Bruker Smart–CCD with Mo K␣ radiation
(ꢁ = 0.71073 Å) at 293 K. The structure was solved with SHELXS-
97 and refined by full-matrix least squares on F2 using SHELXL-97
computer program [31]. Hydrogen atoms were idealized by using
the riding models.
2.4. Synthesis of the complexes [Rh(CO)xClL] (1a–1d), where
L = xantphos (a), DPEphos (b) and x = 1; L = xantphos dioxide (c),
DPEphos dioxide (d) and x = 2
[Rh(CO)2Cl]2 (100 mg) was dissolved in dichloromethane
(10 cm3) and to that solution, a stoichiometric quantity (Rh:L = 1:1)
of the respective ligands were added. The reaction mixture was
stirred at room temperature (r.t.) for about 10–60 min and the sol-
vent was evaporated under vacuum. The yellowish red coloured
compounds so obtained were washed with diethyl ether and stored
over silica gel in a desiccator.
2.7. Kinetic experiment
The kinetic experiments of OA reaction of complexes 1a–1d
with CH3I were monitored using FT-IR spectroscopy in a solu-
tion cell (CaF2 windows, 1.0 mm path length). In order to obtain
pseudo-first-order condition, excess of CH3I relative to metal com-
plex was used. FT-IR spectra (4.0 cm−1 resolution) were scanned
in the ꢀ(CO) region (2200–1600 cm−1) and saved at regular time
interval using spectrum software. After completion of experiment,
absorbance versus time data for the appropriate ꢀ(CO) frequencies
were extracted by subtracting the solvent spectrum and analyzed
off line using OriginPro 7.5 software. Kinetic measurements were
made by following the decay of lower frequency ꢀ(CO) band of
the complexes in the region 1974–1996 cm−1. The pseudo-first-
order rate constants were found from the gradient of the plot of
ln(Ao/At) versus time, where Ao is the initial absorbance and At is
the absorbance at time t.
Analytical data for the complexes 1a–1d are given as follows:
[Rh(CO)Cl(Xantphos)] (1a)
IR (KBr, cm−1): 1974 [ꢀ(CO)]. 1H NMR (CDCl3, ppm): ı 6.82–7.09,
7.31–7.88 (m, Ph), ı 1.69 (s, CH3). 13C NMR (CDCl3, ppm): ı
152.2–126.5 (m, Ar), ı 63.8 (CMe2), ı 32.5 (s, CH3), ı 183.3 (s, CO).
1
31P{ H} NMR (CDCl3, ppm): ı 22.55 [d, JP–Rh = 123.5 Hz]. Elemental
analyses: Found (Cald. for C40H32O2P2ClRh), C 63.85 (64.48); H 4.21
(4.30); P 8.45 (8.32); Cl 4.89 (4.77); Rh 13.72 (13.82); O 4.88 (4.30).
[Rh(CO)Cl(DPEphos)] (1b)
IR (KBr, cm−1): 1985 [ꢀ(CO)]. 1H NMR (CDCl3, ppm):
ı
6.80–7.20, 7.30–7.37, 7.54–7.79 (m, Ph). 13C NMR (CDCl3, ppm): ı
1
154.8–128.2 (m, Ar), ı 180.9 (s, CO). 31P{ H} NMR (CDCl3, ppm):
ı 27.3 [d, JP–Rh = 119.2 Hz]. Elemental analyses: Found (Cald. for
C37H28O2P2ClRh), C 62.91 (63.06); H 3.73 (3.97); P 8.85 (8.79); Cl
5.15 (5.04); Rh 14.23 (14.61); O 5.13 (4.54).
2.8. Carbonylation of methanol using complexes 1a–1d as
[Rh(CO)2Cl(Xantphos dioxide)] (1c)
catalyst precursors
IR (KBr, cm−1): 1983, 2058 [ꢀ(CO)], 1190, 1185 [ꢀ(P O)]. 1H
NMR (CDCl3, ppm): ı 6.68–7.61 (m, Ph), ı 1.69 (s, CH3). 13C NMR
(CDCl3, ppm): ı 153.8–125.2 (m, Ar), ı 34.23 (s, CH3), ı 178.5,
CH3OH (0.099 mol, 4 cm3), CH3I (0.016 mol, 1 cm3), H2O
(0.055 mol, 1 cm3) and catalyst (0.0514 mmol) were placed in a

B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
23
Table 1
Crystallographic data and selected bond lengths for the ligand d.
Crystallographic information
Empirical formula
Formula weight
Temperature (K)
ꢁ (Å)
Cryst. Syst.
Space group
Z
C36H28O3P2
570.5
293
0.71073
Triclinic
Pbca
8
a (Å)
b (Å)
18.2628 (4)
15.0368 (3)
21.3338 (4)
90
90
90
c (Å)
˛ (◦)
ˇ (◦)
ꢂ (◦)
ꢃ(Mo K␣) mm−1
Reflections collected
R1 (observed data)
wR2 (all data)
0.184
5647
0.0466 (3982)
0.1132 (5647)
Fig. 1. X-ray crystal structure of the ligand d.
50 mL reaction vessel. The reaction mixture was then purged with
CO for about 5 min and then pressurized with CO gas at around 5, 10
and 20 bar respectively at 25 ◦C. The carbonylation reactions were
carried out at 130 2 ◦C for 1 h with corresponding gas pressure at
around 15 2, 20 2 and 33 2 bar. The products were collected
and analyzed by G.C.
Selected bond lengths (Å)
P(1)–O(2)
P(2)–O(3)
O(3)· · ·O(2) (Spatial)
1.4822 (16)
1.4803 (15)
6.414
3. Results and discussion
type [Rh(CO)Cl(L)] (1a,1b) where L = a, b (Scheme 1). Elemental
analyses of the complexes support the observed molecular com-
position of 1a and 1b. The IR spectra of 1a and 1b exhibit single
intense ꢀ(CO) bands at around 1974 and 1985 cm−1 respectively
indicate the formation of monocarbonyl rhodium(I) complexes
of diphosphine ligands. The complexes 1a and 1b show char-
3.1. Single crystal X-ray structural determination of d
The single crystal X-ray structure of xantphos dioxide (c) was
previously determined by us [25], only the crystal structure of
DPEphos dioxide (d) ligand is reported in this paper (Fig. 1). The
crystal information of d and the selected bond lengths are summa-
rized in Table 1. The compound d crystallizes in a triclinic system
with space group Pbca. The average P–O bond length (1.48 Å) of d is
similar to the P O bond lengths reported in literature [13,25]. The
ligand d exhibits interesting intermolecular interactions involv-
ing shorter distances than the sum of van der Waals radii (Fig. 2).
The interactions between these molecules are presented by dot-
ted lines and summarized in Fig. 2. The molecules are connected
by weak hydrogen bonding between H atoms of the phenyl ring
and O atom of P O group (C–H· · ·O P) of the adjacent molecules
[C(26)–H(26)· · ·O(2) 2.512 Å] to develop an extended one dimen-
sional network (Fig. 2).
1
acteristic 31P{ H} NMR spectra with doublet at around ı 22.55
(d, JP–Rh = 123.5 Hz) and 27.3 (d, JP–Rh = 119.2 Hz) ppm respectively
and 1H NMR spectra with phenylic multiplet in the range ı
1
6.80–7.88 ppm. Thus, 31P{ H} NMR spectra show downfield shift
compared to the free ligands (ı −16.3 (a) and −12.1 (b) ppm) indi-
cating chelate formation through phosphorus donor. The 13C NMR
spectra of 1a and 1b show characteristic resonances at around ı
183 and 181 ppm respectively for the carbonyl carbons. The phenyl
and other carbons in the complexes are found in their respec-
tive ranges. On the other hand, the dimeric precursor [Rh(CO)2Cl]2
reacts with dioxide analogues of a and b i.e. xantphos dioxide (c)
and DPEphos dioxide (d) in CH2Cl2 to yield the dicarbonyl com-
plexes of the type [Rh(CO)2Cl(L)] (1c,1d) where L = c, d (Scheme 1).
The IR spectra of 1c and 1d exhibit two intense ꢀ(CO) bands in
the range 1983–2071 cm−1 indicating the presence of two termi-
nal CO groups. The ꢀ(P–O) bands of the complexes at around 1190,
1185 (1c) and 1189, 1186 (1d) cm−1 respectively are lower than
those of the corresponding free ligands [␯(P–O) = 1193 (c) and 1198,
1186 (d) cm−1] confirming the formation of Rh–O bonds. The com-
3.2. Synthesis and characterization of 1a–1d
The reaction of the chloro-bridge dimer [Rh(CO)2Cl]2 with
two molar equivalent of 9,9-dimethyl-4,5-bis(diphenylphosphino)
xanthene [xantphos] (a) and bis(2-diphenylphosphinophenyl)
ether [DPEphos] (b) in dichloromethane proceeds rapidly with the
evolution of CO gas to yield the monocarbonyl complexes of the
1
plexes 1c and 1d show characteristic 31P{ H} NMR spectra with
two doublets at around ı 45.5, 48.04 (d, JP–P = 69.2 Hz) and singlets
Fig. 2. Extended one dimensional network of d down the crystallographic c-axis stabilized through C–H· · ·O [C(36)–H(36)· · ·O(2) 2.512 Å] hydrogen bonding.

24
B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
Scheme 1. Synthesis and reactivity of 1a–1d.
at around ı 24.7, 35.8 ppm respectively and also 1H NMR spectra
with phenylic multiplet in the range ı 6.15–7.82 ppm. The occur-
rence of lower ꢀ(P–O) bands in 1c and the downfield shift of the
and 1d show characteristic resonances for CO groups in the region
ı 178–186 ppm along with other characteristic bands of the coor-
dinated ligands. It is interesting to mention here that the rhodium
dicarbonyl precursor reacts with a, b to form monocarbonyl com-
plexes, while, the dioxide analogues c, d under similar reaction
condition afford dicarbonyl complexes. This fact may be due to the
moderate ␲ accepting tendency of PPh2 moiety of a, b which com-
pete with the coordinated CO groups facilitating decarbonylation
to stabilize the monocarbonyl complexes, while, the oxygen donor
ligands (c, d) act as only ␴ donor and favour to generate dicarbonyl
complexes.
1
31P{ H} NMR spectrum suggesting the formation of chelate com-
plex. However, in the complex 1d, the appearance of IR band at
around 1186 cm−1 due to uncoordinated P O donor together with
1
the 31P{ H} NMR band at around ı 24.7 ppm indicate the presence
of dangling P–O group in the complex. The observed diversity of
coordination mode of the two dioxide ligands during complexa-
tion under the similar experimental condition depends exclusively
on their inherent structural behaviour which may be partially sub-
stantiated from the single crystal X-ray structures of c [25] and d
as shown in Figs. 3 and 1 respectively. The 13C NMR spectra of 1c
3.3. Reactivity of 1a–1d towards various electrophiles
One of the most important industrial processes utilizing homo-
geneous transition-metal catalysis is the rhodium and iodide
promoted carbonylation of methanol to acetic acid. In this respect,
OA reaction of alkyl halides with metal complexes is a very impor-
tant reaction as it is the key step in the carbonylation catalysis
[32]. Therefore, oxidative reactivities of 1a–1d towards various
electrophiles were evaluated.
The complexes 1a–1d undergo OA reactions with CH3I and
C2H5I to generate Rh(III) complexes of the type [RhCl(CO)y(COR)IL],
where R = –CH3, y = 0 (2a,2b) and y = 1 (2c,2d); R = –C2H5 (3a,3b),
y = 0 (3a,3b) and y = 1 (3c,3d) (Scheme 1). On the other hand, 1a–1d
reacts with I2 in dichloromethane to produce Rh(III) complexes of
the type [RhCl(CO)I2L](4a–4d) (Scheme 1). The IRspectraof theoxi-
dized products show single characteristic terminal ꢀ(CO) bands in
the range 2038–2075 cm−1 and a broad ꢀ(CO) band (except 4a–4d
in which the electrophile is iodine) in the range 1678–1733 cm−1
characteristic of acyl carbonyl groups (Table 2). The 1H NMR spec-
tra of the complexes 2a–2d display singlet resonances in the
range ı 2.58–2.78 ppm suggesting the formation of –OCCH3 group
Fig. 3. X-ray crystal structure of the ligand c. Selected bond lengths (Å): P(2)–O(3)
1.479(4), P(1) O(1) 1.476(4), O(1)· · ·O(3) (spatial) 3.374

B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
25
Table 2
Significant IR, 1H and 13C NMR data for the oxidative addition (OA) adducts.
OA adducts
IR (cm−1
)
1H NMR (ppm)
CH3
13C NMR (ppm)
ꢀ(CO)t
ꢀ(CO)acyl
CH2
–
(CO)t
–
(CO)acyl
2a
–
1678
1.63(s)
2.78(s)
206(s)
2b
2c
–
1702
2.62(s)
–
–
–
205.6(s)
203.4(s)
2044
1709
1733
1.68(s)
2.58(s)
181.5(s)
2d
3a
2057
–
1719
1727
2.72(s)
–
182.5(s)
–
204.1(s)
201.4(s)
1.73(s)
1.21(t)
3.56(q)
3b
3c
–
1691
1725
1.13(t)
3.57(q)
3.42(q)
–
204.8(s)
206.2(s)
2038
1.78(s)
1.22(t)
184.5(s)
3d
4a
4b
4c
4d
2057
2071
2074
2075
2068
1713
1.19(t)
1.68(s)
–
1.76(s)
–
3.24(q)
180.3(s)
185.6(s)
184.3(s)
183.4(s)
188.6(s)
207.6(s)
–
–
–
–
–
–
–
–
–
–
–
–
including other characteristic bands of the ligands. Similarly, the
complexes 3a–3d shows the bands for methylene and methyl pro-
tons of –OCCH2CH3 group as quartet and triplet respectively in the
range ı 1.13–3.57 ppm (Table 2). The 13C NMR spectra of 2a–2d
and 3a–3d exhibit bands in the range 201–208 ppm characteristic
of the acyl carbonyl group.
Attempts to substantiate the structures of different rhodium(I)
and rhodium(III) carbonyl complexes by X-ray crystal structure
determination were not possible because no suitable crystals could
be obtained in spite of numerous attempts.
Kinetic measurements for the OA reaction of the complexes
1a–1d with methyl iodide were carried out using IR spectroscopy
by monitoring the changes in the ꢀ(CO) band(s). Fig. 4 shows a typ-
ical series of spectra of 1c when reacts with CH3I at 25 ◦C, in which
the bands at around 1983 and 2058 cm−1 decay and new bands
grow in the region 2040–2070 and 1700–1750 cm−1. Finally, the
two terminal ꢀ(CO) bands at around 2058 and 1983 cm−1 of 1c
are replaced by two new terminal ꢀ(CO) bands 2044 (intense) and
2067 (weak) cm−1 and two broad acyl ꢀ(CO) bands at around 1709
and 1733 cm−1. The presence of two acyl ꢀ(CO) bands of almost
equal intensity may be due to the formation of equimolar quan-
tity of two acyl Rh(III) complexes in the reaction mixture as shown
in Scheme 2. Absorbance versus time plots for the decay of lower
intensity ꢀ(CO) bands at 1974, 1985, 1983 and 1996 cm−1 of 1a–1d
respectively are shown in Fig. 5. A linear fit of pseudo-first order
was observed for the entire course of the reaction of CH3I with the
complexes 1a–1d as is evidenced from the plot of ln(Ao/At) ver-
sus time, where Ao and At are the absorbance at time t = 0 and t,
respectively (Fig. 6). From the slopes of the plots, the rate constants
were calculated and found as 3.19 × 10−4, 3.44 × 10−4, 6.14 × 10−4
and 1.51 × 10−3 S−1 for the complexes 1a, 1b, 1c and 1d respec-
tively. The rate of oxidative addition of the four complexes varies
as 1a < 1b < 1c < 1d. The observed values of the rate constants indi-
cate that the rate of OA is higher in case of dicarbonyl complexes
of oxygen donor ligands than those of monocarbonyl phosphine
donor complexes.
3.4. Carbonylation of methanol to acetic acid and ester using the
complexes 1a–1d as the catalyst precursors
The results of carbonylation of methanol to acetic acid and
methyl acetate in the presence of 1a–1d as catalyst precursors
are shown in Table 3. The precursor complexes 1a–1d show total
conversion of CH3OH in the range 35.3–91.9% at 130 2 ◦C under
initial CO pressure 5–20 bar (at ∼25 ◦C) for 1 h reaction time with
corresponding TON 679–1768. It has been observed from Table 3,
that as the applied CO pressure increases from 5 to 20 bar for 1a,
total conversion increases from 35.3 to 82.4% with corresponding
increase in TON from 679 to 1585. Like 1a, the complexes 1b–1d
also show a similar increasing trend with the increase in CO pres-
sure and maximum TON of about 1637 (1b), 1718 (1c), and 1768
(1d) with corresponding conversions of 85.1, 89.3, and 91.9% have
been obtained at 20 bar initial CO pressure (∼25 ◦C). It is interest-
ing to note that with increase of initial CO pressure, the selectivity
of the catalyst changes i.e. at 5 bar CO pressure, the major prod-
uct was methyl acetate; while at around 20 bar initial CO pressure,
the major product was acetic acid. Under the similar experimental
conditions, the well-known precursor [Rh(CO)2I2]− generated in
situ from [Rh(CO)2Cl]2 [33] shows lower TON compared to 1a–1d.
Thus, the efficiency of the complexes depends on the nature of the
ligands and follows the order 1d > 1c > 1b > 1a > [Rh(CO)2Cl]2. The
observed trend is also well supported by their kinetic experiments.
The monocarbonyl complexes 1a and 1b undergo OA of methyl
Fig. 4. Series of IR spectra {␯(CO) region} illustrating the reaction of 1c with CH3I
at 25 ◦C. The arrows indicate the behaviour of each band as the reaction progresses.

26
B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
Scheme 2. Plausible Rh(III) acyl complexes (2cꢀand 2cꢀꢀ) generated after OA of CH3I to 1c.
Fig. 5. Kinetic plot of (a)–(d) showing the decay of ␯(CO) bands of 1a–1d respectively during the reaction with neat CH3I at room temperature (∼25 ◦C).

B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
27
Fig. 6. ln(Ao/At) versus time plot in (a)–(d) are presented for the OA reaction of the complexes 1a–1d respectively with neat CH3I at ∼25 ◦C .
Table 3
Results of methanol carbonylation.
Catalyst precursor
[Rh(CO)2I2]−c
CO pressure at 25 ◦C ( 1 bar)
Total conv. (%)
Acetic acida (%)
Methyl acetatea
TONb
5
10
20
24.1
41.4
52.1
4.0
12.3
10.3
20.1
29.1
41.8
463.6
796.4
1000
1a
1b
1c
5
10
20
35.3
56.2
82.4
8.9
24.0
42.1
26.4
32.2
40.3
679.0
1081.0
1585.0
5
10
20
36.5
66.5
85.1
7.5
25.3
48.5
29.0
41.2
36.6
702.1
1279.2
1637.0
5
10
20
42.5
66.9
89.3
12.1
27.8
49.5
30.4
39.1
39.8
839.1
1286.9
1717.8
1d
5
10
20
55.2
72.4
91.9
12.6
30.4
54.4
42.6
42.0
37.5
1061.8
1392.7
1768.4
a
Yield of methyl acetate and acetic acid were obtained from GC analyses after 1 h reaction time.
TON = [amount of product (mol)]/[amount of catalyst (Rh mol)].
Formed from added [Rh(CO)2Cl]2 under catalytic condition.
b
c
iodide with slower rate than those of dicarbonyl complexes 1c and
metal center throughout the entire course of the catalytic reac-
tions.
1d and hence the former show lower TON compared to latter. How-
ever, the catalytic efficiency of 1b and 1d is found to be higher than
the similar type of complexes 1a and 1c respectively under identical
experimental conditions, which may be due to the flexible ligand
backbone of 1b and 1d. Such advantage of flexible ligand back-
bone in catalytic reaction is also reported in literature [26,34–36].
On examining the catalytic reaction mixture by IR spectroscopy at
different time intervals and at the end of the catalytic reaction, mul-
tiple ꢀ(CO) bands are obtained that matched well with the ꢀ(CO)
values of solution containing a mixture of the parent rhodium(I)
carbonyl complexes 1a–1d and rhodium(III) acyl complexes 2a–2d.
Thus, it may be inferred that the ligands remained bound to the
4. Conclusions
The new complexes of the type [Rh(CO)Cl(L)] (1a,1b) and
[Rh(CO)2Cl(L)] (1c,1d) where L = a–d have been synthesized and
characterized, and the structure of the ligand d was determined
by single crystal X-ray diffraction. The complexes 1a–1d undergo
oxidative addition with different electrophiles like CH3I, C2H5I and
I2 to afford acyl Rh(III) complexes. The kinetic data for the OA reac-
tions with CH3I indicate a first order reaction and also exhibit that
the rate of OA for the monocarbonyl complexes 1a, 1b is slower

28
B. Deb, D.K. Dutta / Journal of Molecular Catalysis A: Chemical 326 (2010) 21–28
than those of dicarbonyl complexes 1c, 1d. The catalytic activities
of 1a–1d for the carbonylation of methanol to acetic acid and its
ester exhibit higher TON (679–1768) than that of the well-known
commercial species [Rh(CO)2I2]− (TON = ∼464–1000) under similar
experimental conditions.
[9] C.M. Thomas, G. Suss-Fink, Coord. Chem. Rev. 243 (2003) 125–142.
[10] J.F. Roth, J.H. Craddock, A. Hershman, F.E. Paulik, Chem. Technol. (1971)
600–605.
[11] K.K. Robinson, A. Hershman, J.H. Craddock, J.F. Roth, J. Catal. 27 (1972) 389–396.
[12] H.C. Martin, N.H. James, J. Aitken, J.A. Gaunt, H. Adams, A. Haynes,
Organometallics 22 (2003) 4451–4458.
[13] D.K. Dutta, J.D. Woollins, A.M.Z. Slawin, A.L. Fuller, B. Deb, P.P. Sarmah, M.G.
Pathak, D. Konwar, J. Mol. Catal. A: Chem. 313 (2009) 100–106.
[14] D.K. Dutta, P. Chutia, B.J. Sarmah, B.J. Borah, B. Deb, J.D. Woollins, J. Mol. Catal.
A: Chem. 300 (2009) 29–35.
Supplementary data
[15] B.J. Sarmah, B.J. Borah, B. Deb, D.K. Dutta, J. Mol. Catal. A: Chem. 289 (2008)
95–99.
CCDC-699237(d) contains the supplementary crystallographic
data for this paper. These 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 336 033; e-mail:
deposit@ccdc.cam.ac.uk.
[16] D.K. Dutta, J.D. Woollins, A.M.Z. Slawin, D. Konwar, P. Das, M. Sharma, P. Bhat-
tacharyya, S.M. Aucott, Dalton Trans. (2003) 2674–2679.
[17] D.K. Dutta, J.D. Woollins, A.M.Z. Slawin, D. Konwar, M. Sharma, P. Bhatacharyya,
S.M. Aucott, J. Organomet. Chem. 691 (2006) 1229–1234.
[18] D. Forster, J. Am. Chem. Soc. 98 (1976) 846–848.
[19] D. Forster, Adv. Organomet. Chem. 17 (1979) 255–267.
[20] D. Forster, T.C. Singleton, J. Mol. Catal. 17 (1982) 299–303.
[21] A. Haynes, J. McNish, J.M. Pearson, J. Organomet. Chem. 551 (1998) 339–347.
[22] T. Ghaffar, H. Adams, P.M. Maitlis, A. Haynes, G.J. Sunley, M.J. Baker, Chem.
Commun. (1998) 1359–1360.
Acknowledgements
[23] J. Rankin, A.D. Poole, A.C. Benyei, D.J. Cole-Hamilton, Chem. Commun. (1997)
1835–1836.
[24] L. Gonsalvi, H. Adams, G.J. Sunley, E. Ditzel, A. Haynes, J. Am. Chem. Soc. 121
(1999) 11233–11234.
[25] B. Deb, D.K. Dutta, Polyhedron 28 (2009) 2258–2262.
[26] B. Deb, B.J. Borah, B.J. Sarmah, B. Das, D.K. Dutta, Inorg. Chem. Commun. 12
(2009) 868–871.
[27] B. Deb, B.J. Sarmah, B.J. Borah, D.K. Dutta, Spectrochim. Acta A 72 (2009)
339–342.
The authors are grateful to Dr. P.G. Rao, Director, North East Insti-
tute of Science and Technology (CSIR) Jorhat-785 006, Assam, India
for his kind permission to publish the work. The Department of Sci-
ence and Technology (DST), New Delhi (Grant: SR/S1/IC-05/2006)
is acknowledged for the partial financial grant. The author B. Deb
thanks to CSIR, New Delhi for providing the Senior Research Fel-
lowship.
[28] I. Vogel, A Text Book of Quantitative Inorganic Analysis: Theory and Practice,
third ed., Longmans Green and Co., 1951.
References
[29] The analysis of rhodium was carried out by gravimetric method in which the
rhodium metal complex was converted to metallic rhodium by following steps:
The complex was first decomposed by boiling in conc. H2SO4/HNO3. It was
then allowed to react with thiourea when black rhodium sulfide was formed
which on ignititon gives the oxide. Lastly, the oxide was quantitively reduced
by H2 gas (at elevated temperature) to metallic rhodium and weighted as
such.
[30] J.A. McCleverty, G. Wilkinson, Inorg. Synth. 8 (1966) 211–214.
[31] G.M. Sheldrick, Acta Cryst. A 64 (2008) 112–122.
[32] A. Haynes, B.E. Mann, D.J. Gulliver, G.E. Morrison, P.M. Maitlis, J. Am. Chem. Soc.
113 (1991) 8567–8569.
[33] H. Adams, N.A. Bailey, B.E. Mann, C.P. Manuel, C.M. Spencer, A.G. Kent, J. Chem.
Soc., Dalton Trans. (1988) 489–496.
[34] B.P. Morgan, R.C. Smith, J. Organomet. Chem. 693 (2008) 11–16.
[35] M. Sakai, H. Hayashi, N. Miyaura, Organometallics 16 (1997) 4229–4231.
[36] B. Deb, P.P. Sarmah, D.K. Dutta, Eur. J. Inorg. Chem. (2010) 1710–1716.
[1] J.R. Dilworth, J.R. Miller, N. Wheatly, M.J. Baker, J.G. Sunley, J. Chem. Soc., Chem.
Commun. (1995) 1579–1581.
[2] P.M. Maitlis, A. Haynes, G.J. Sunley, M.J. Howard, J. Chem. Soc., Dalton Trans.
(1996) 2187–2196.
[3] M.J. Baker, M.F. Giles, A.G. Orpen, M.J. Taylor, R.J. Watt, J. Chem. Soc., Chem.
Commun. (1995) 197–198.
[4] M.J. Howard, M.D. Jones, M.S. Roberts, S.A. Taylor, Catal. Today 18 (1993)
325–354.
[5] A.J. Pardey, C. Longo, Coord. Chem. Rev. 254 (2010) 254–272.
[6] Z. Freixa, P.C.J. Kamer, M.L. Anthony, P.W.N.M. van Leeuween, Angew. Chem.
Int. Ed. 44 (2005) 4385–4388.
[7] C.M. Thomas, R. Mafua, B. Therrien, E. Rusanov, H.S. Evans, G. Suss-Fink, Chem.
Eur. J. 8 (2002) 3343–3352.
[8] S. Burger, B. Therrien, G. Suss-Fink, Helvet. Chim. Acta 88 (2005) 478–486.