Novel gem-dithiolato-bridged rhodium hydroformylation catalysts: Bridging the gap in dinuclear rhodium thiolate chemistry

Article abstract of DOI:10.1002/ejic.200700756

The direct protonation of the bridging hydroxo ligands in [Rh(μ-OH)(cod)]₂ by 1,1-dimercaptocyclohexane [Chxn(SH) ₂] yields the gem-dithiolato-bridged compound [Rh₂(μ- S₂Chxn)-(cod)₂] (1). The dinucle

Full text of DOI:10.1002/ejic.200700756

FULL PAPER
DOI: 10.1002/ejic.200700756
Novel gem-Dithiolato-Bridged Rhodium Hydroformylation Catalysts:
Bridging the Gap in Dinuclear Rhodium Thiolate Chemistry
Marc A. F. Hernandez-Gruel,[a] Gustavo Gracia-Arruego,[a] Angel B. Rivas,[b]
Isabel T. Dobrinovitch,[a] Fernando J. Lahoz,[a] Alvaro J. Pardey,[b] Luis A. Oro,*[a] and
Jesús J. Pérez-Torrente*[a]
Keywords: Homogeneous catalysis / Hydroformylation / Dinuclear complexes / gem-Dithiolato ligands / Rhodium
The direct protonation of the bridging hydroxo ligands in
[Rh(µ-OH)(cod)]2 by 1,1-dimercaptocyclohexane [Chxn(SH)2]
obtained with the catalytic system 1/P(OMe)3. The dinuclear
compound [Rh2(µ-S2Chxn)(CO)2(PPh3)2] (2) was isolated
yields the gem-dithiolato-bridged compound [Rh2(µ-S2Chxn)- from the catalytic solutions resulting from the system 1/PPh3
(cod)2] (1). The dinuclear framework in 1 is supported by a
1,1-cyclohexanedithiolato ligand exhibiting a 1:2κ2S,1:2κ2SЈ
coordination mode. Compound 1 is an active catalyst precur-
sor in the presence of P-donor ligands for the hydrofor-
mylation of oct-1-ene under mild conditions of pressure and
temperature (100 PSI, 353 K). The best results were obtained
with phosphite ligands as modifying ligands. An aldehyde
selectivity of 97%, a regioselectivity towards the linear alde-
hyde of 81%, and turnover frequencies of up to 198 h–1 were
and was characterized by spectroscopic means and by X-ray
diffraction to be the trans isomer. The mixed-ligand dinu-
clear complexes 2 and [Rh2(µ-S2Chxn)(CO)2(PCy3)2] (3) (Cy
= cyclohexyl) were independently prepared by reaction of
Chxn(SH)2 with the mononuclear complexes [Rh(acac)(CO)-
(PR3)] in the appropriate molar ratio.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
impart the appropriate electronic and steric influence on the
reactions, and more importantly, produce flexible structures
allowing the accommodation of the metal centers in close
proximity but preventing it from fragmenting.[4]
Introduction
Dimetallic complexes can serve as catalysts because the
expected cooperation between the metal atoms should re-
sult in more active and selective catalysts than monometal-
lic systems.[1] However, fragmentation has been a major
problem in polymetallic catalysts and, in spite of the inten-
sive research in this field, the number of active dimetallic
catalysts actually operating through a dimetallic mechanism
are scarce.[2,3] Stanley and co-workers have demonstrated
that the homodimetallic rhodium complex rac-[Rh2(nbd)2-
(et,ph-P4)]2+, containing a binucleating tetraphosphane li-
gand, is a precursor of a highly active and selective catalyst
for the hydroformylation of 1-alkenes by a mechanism in-
volving dimetallic cooperation between the two rhodium
centers.[3] As evidenced by the Stanley’s hydroformylation
system, the design of the binucleating ligands is of major
importance as the catalytic activity largely depends on the
structure of the complex. In particular, the ligands must
fulfil the electronic requirements of the active metal centers,
In this context, it is well known that dinuclear thiolato-
bridged complexes [Rh(µ-SR)(CO)(PRЈ3)]2 (R = tBu, Ph;
RЈ = OMe, OPh, Ph) are effective catalysts in the hydrofor-
mylation of olefins at moderate pressure and temperature
(Figure 1a).[5] However, the dinuclear structure of the active
catalytic species has been questioned, as kinetic studies sug-
gested the involvement of mononuclear species.[6] Similarly
to the Kalck’s systems, fluorothiolato- and aminothiolato-
bridged dinuclear rhodium complexes have been described
as active precursors for the hydroformylation of alkenes un-
der mild conditions.[7] A step forward in rhodium thiolate
chemistry was the preparation by Claver and co-workers
of di- and tetranuclear dithiolato rhodium complexes with
catalytic activity in the hydroformylation of 1-hexene (Fig-
ure 1c).[8,9] Monodentate thiolato-bridging ligands provide
flexible structures that support a wide range of bonding and
nonbonding metal distances by modification of the hinge
angle between the rhodium coordination planes. In con-
trast, the bridging and chelating coordination mode of the
dithiolato ligand results in more rigid dinuclear structures
with a possible influence on the catalytic activity. In ad-
dition, chirality was introduced at the backbone of the di-
thiolato ligand, which gives rise to chiral dinuclear com-
plexes that show very good regioselectivities in the hydro-
[a] Departamento de Química Inorgánica, Instituto Universitario
de Catálisis Homogénea, Instituto de Ciencia de Materiales de
Aragón, Universidad de Zaragoza-C.S.I.C.,
50009-Zaragoza, Spain
Fax: +34-976761143
E-mail: oro@unizar.es
perez@unizar.es
[b] Centro de Equilibrios en Solución, Escuela de Química, Facul-
tad de Ciencias, Universidad Central de Venezuela,
Caracas, Venezuela
Eur. J. Inorg. Chem. 2007, 5677–5683
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5677

L. A. Oro, J. J. Pérez-Torrente et al.
FULL PAPER
formylation of styrene, although the observed enantio-
selectivities were low, indicating that the effect of the pres-
ence of a chiral dithiolato ligand is rather small.[10]
Results and Discussion
The reaction between [Rh(µ-OH)(cod)]2 and 1,1-dimer-
captocyclohexane [Chxn(SH)2] in dichloromethane gave a
red-orange solution of the compound [Rh2(µ-S2Chxn)-
(cod)2] (1), which was isolated as an orange-red microcrys-
talline solid in good yield (Figure 2). Interestingly, com-
pound 1 can be obtained in similar yields from other di-
and mononuclear standard starting materials in rhodium
chemisty such as [Rh(µ-OMe)(cod)]2 and [Rh(acac)(cod)],
although an external base (NEt3) is necessary to drive the
reaction to completion. The dinuclear formulation of the
complex is supported both by the determination of the mo-
lecular weight in chloroform and by the FAB+ spectra that
Figure 1. Different thiolato-bridged dinuclear complexes.
1
shows the dinuclear ion at m/z = 568. The H NMR spec-
The nuclearity of the dithiolato rhodium complexes
[Rh2(µ-S(CH2)nS)(L2)2]x is influenced both by the number
of methylenic units between the two sulfur atoms and by
the auxiliary ligands. Tetranuclear diolefin complexes (L2 =
cod, x = 2) were generally obtained from dithiolato ligands
with large n values (i.e. n = 4). However, the tetranuclear
compounds were converted to dinuclear complexes by car-
bonylation at atmospheric pressure (L = CO, x = 1), which
suggests labile Rh–S bonds.[9] In fact, the ion-pair com-
pounds [Rh(diphos)2][Rh(dithiolato)(CO)2] were observed
in the reaction between a dinuclear carbonyl dithiolato-
bridged complex and diphosphanes.[11] In addition, high-
pressure spectroscopic techniques (HPNMR and HPIR)
have shown that some thiolato- and dithiolato dinuclear
rhodium complexes evolve to mononuclear rhodium hy-
dride complexes under hydroformylation conditions.[12]
In order to reinforce the dinuclear framework, we envis-
aged dinuclear rhodium complexes supported by gem-diti-
olato ligands (Figure 1b). This type of ligand, although
closely related to the standard dithiolato ones, should pro-
vide access to new dinuclear complexes with a number of
features that could be of interest both in stoichiometric and
catalytic reactions. Firstly, the presence of a single bridge-
head carbon atom between both sulfur atoms should lead
to a more compact [Rh(µ-S2CR2)Rh] core that is probably
more resistant to fragmentation. Secondly, the structure
and the coordination mode of the ligand should generate
much more rigid dinuclear systems with a likely smaller an-
gle between the coordination planes of the rhodium centers
trum in CDCl3 at room temperature shows sharp reso-
nances and is in agreement with the expected rigid frame-
work with C2v symmetry. Thus, two resonances for the ole-
finic =CH protons and the carbon atoms of the equivalent
1,5-cyclooctadiene ligands were observed in the 1H and
13C{1H} NMR spectra, respectively. The protons of the 1,1-
cyclohexylene fragment display three resonances, which in-
dicates a rapid equilibrium between the possible chair con-
formations at room temperature.
Figure 2. Synthesis of rhodium gem-dithiolato-bridged dinuclear
complexes. (* cis isomer Ͻ 5%).
Compound 1 was used as a catalyst precursor, in the
and shorter metal–metal distances, which favor the cooper- presence of monodentate P-donor ligands, for the hydrofor-
ative effects between the metal centers. Finally, it is impor- mylation of oct-1-ene under mild conditions of temperature
tant to note that the R groups on the sp3 bridgehead carbon and pressure (353 K and 100 PSI) (Figure 3). It has been
atom are directly oriented toward the rhodium atoms and found that the catalytic activity is strongly dependent on the
not toward the center of the dinuclear unit, which could P/Rh ratio. In the absence of P-donor ligands, no catalytic
have a determining steric influence on the hydroformylation activity was observed at 100 PSI, although extensive isom-
reaction.
erization to internal alkenes was observed at 200 PSI. Al-
Herein we wish to report on the synthesis of gem-dithiol- most certainly, compound 1 is transformed into the inactive
ato-bridged dinuclear rhodium complexes and their cata- tetracarbonyl complex [Rh2(µ-S2Chxn)(CO)4] under hydro-
lytic activity in the hydroformylation of oct-1-ene. Although formylation conditions, and an excess of PR3 ligands is nec-
a few mono- and dinuclear methanedithiolato and gem-di- essary to maintain a sufficient concentration of the possibly
thiolato complexes have been reported,[13] these dinuclear active phosphane-containing species [Rh2(µ-S2Chxn)-
compounds are, to the best of our knowledge, the first ex- (CO)4–x(PR3)x]. The optimum P/Rh ratio was found to be
ample of gem-dithiolato complexes directly synthesized approximately 4, as higher ratios produce a slight decrease
from a gem-dithiol compound.
in the catalytic activity. The results obtained in the hydro-
5678
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5677–5683

Novel gem-Dithiolato-Bridged Rhodium Hydroformylation Catalysts
formylation of oct-1-ene under these optimized conditions the activity and the regioselectivity (54%) are considerably
are shown in Table 1. When P(OMe)3 was used as the mod- smaller (entry 5).
ifying ligand, conversions of 67.9% or 88.4% were obtained
The investigation of the catalytic solutions after the cata-
in 2 or 3 h (entries 1 and 2), respectively. In both catalytic lytic runs when using PPh3 as the P-donor ligand allowed
runs, the aldehyde selectivity was as high as 97%, with re- the isolation of the dinuclear complex [Rh2(µ-S2Chxn)(CO)2-
gioselectivities of up to 81% for the linear aldehyde (only (PPh3)2] (2). Compound 2 can also be prepared in a
1-nonanal and 2-methyl-octanal were obtained in the reac- straightforward manner with an excellent yield from the re-
tions). The by-products of these reactions were octane, the action of [Rh(acac)(CO)(PPh3)] and Chxn(SH)2 in a 2:1
hydrogenation product, detected in trace amounts (Ͻ1%), molar ratio (Figure 2). The molecular structure of com-
and internal n-octenes, resulting from olefin isomerization pound 2 was determined by X-ray diffraction and is shown
(≈ 2%).
in Figure 4. Selected bond lengths and angles are collected
in Table 2. The dinuclear skeleton of 2 is held up by a 1,1-
cyclohexanedithiolato ligand exhibiting a bridging and che-
lating coordination mode (1:2κ2S, 1:2κ2SЈ) that results in
the formation of two fused four-membered metallacycles.
The 1,1-cyclohexylene fragment adopts the usual chair con-
formation, and both rhodium atoms exhibit a distorted
square-planar geometry by coordination to two additional
CO and PPh3 ligands. In contrast with dinuclear thiolato
[Rh(µ-SR)(CO)(PR3)]2 complexes, where the PR3 ligands
are usually in a cis arrangement to accommodate the anti
conformation of the thiolate ligands,[14–16] the PPh3 ligands
in 2 adopt a mutually trans disposition.
Figure 3. Hydroformylation of oct-1-ene (R = –C6H13).
Table 1. Hydroformylation of oct-1-ene with the complex [Rh2(µ-
S2Chxn)(cod)2] (1) as catalyst precursor.[a]
Run Ligand P/Rh t [h] % Conv.[b] % Aldehyde[b] % n[b] TOF [h–1 [c]
]
1
2
3
4
5
6[d]
P(OMe)3
P(OMe)3
P(OPh)3
PPh3
PCy3
PPh3
4
4
4
4
4
4
2
3
2
12
12
12
67.9
88.4
96.5
65.2
43.9
68.7
97.2
97.4
76.7
93.2
93.0
92.3
81
80
83
76
54
74
198
172
222
30
20
32
[a] Reaction conditions: 100 PSI (CO/H2, 1/1), 353 K, oct-1-ene
(10.2 mmol, 0.6 ), [Rh2(µ-S2Chxn)(cod)2] (0.017 mmol, 1 m). [b]
Conversion, selectivity for aldehyde, and regioselectivity for the
linear aldehyde (n) determined by GC. [c] TOF = molaldehyde
/
[molcatalyst]·h–1 corresponds to the reaction time. [d] [Rh2(µ-
S2Chxn)(CO)2(PPh3)2] (2) as catalyst precursor.
The catalytic system resulting from P(OPh)3 (entry 3) is
more active, reaching a 96.5% conversion in 2 h, with a
similar regioselectivity. In contrast, this system is much less
selective (aldehyde selectivity 76.7%) as a consequence of
the high isomerization activity that produces internal n-oct-
enes. However, neither 2-ethylheptanal nor 2-propylhexanal
were detected by GC, which indicates that under these ex-
perimental conditions, the internal olefins were not hydro-
formylated.
The catalytic performance with phosphite ligands is su-
perior than that observed with phosphane ligands, as they
provided higher conversions for the same reaction times.
The TOF for aldehyde production in these phosphite cata-
lytic systems was found to be about 200 turnover/h. How-
ever, the catalytic systems obtained with triphenyl- or tricy-
clohexylphosphane as auxiliary ligands provided TOF
numbers for the aldehyde of about 30 turnover/h (see
Table 1). For example, when PPh3 was used as the mod-
ifying ligand, a 65.2% conversion was attained in 12 h with
good aldehyde selectivity (93.2%) and 76% regioselectivitiy
for the linear aldehyde (entry 4). Although the same chemo-
selectivity was observed in the PCy3 catalytic system, both
Figure 4. Molecular structure of the dinuclear complex [Rh2(µ-
S2Chxn)(CO)2(PPh3)2] (2).
Table 2. Selected bond lengths [Å] and angles [°] for dinuclear com-
pound [Rh2(µ-S2Chxn)(CO)2(PPh3)2] (2).
Rh(1)–S(1)
Rh(1)–S(2)
Rh(1)–P(1)
Rh(1)–C(1)
S(1)–C(3)
C(1)–O(1)
C(3)–C(4)
S(1)–Rh(1)–S(2)
S(1)–Rh(1)–P(1)
S(1)–Rh(1)–C(1)
S(2)–Rh(1)–P(1)
S(2)–Rh(1)–C(1)
P(1)–Rh(1)–C(1)
Rh(1)–S(1)–Rh(2)
S(1)–C(3)–S(2)
S(1)–C(3)–C(4)
S(1)–C(3)–C(8)
2.3934(7)
2.3943(8)
2.2700(8)
1.836(3)
1.866(3)
1.150(4)
1.527(4)
71.09(2)
99.58(3)
167.19(10)
170.07(3)
96.44(10)
93.03(10)
74.031(18)
96.82(12)
113.6(2)
111.9(2)
Rh(2)–S(1)
Rh(2)–S(2)
Rh(2)–P(2)
Rh(2)–C(2)
S(2)–C(3)
C(2)–O(2)
C(3)–C(8)
S(1)–Rh(2)–S(2)
S(1)–Rh(2)–P(2)
S(1)–Rh(2)–C(2)
S(2)–Rh(2)–P(2)
S(2)–Rh(2)–C(2)
P(2)–Rh(2)–C(2)
2.4074(6)
2.3821(7)
2.2476(6)
1.834(3)
1.855(3)
1.151(4)
1.523(4)
71.06(2)
163.82(3)
101.20(8)
95.04(2)
172.22(8)
92.73(8)
Rh(1)–S(2)–Rh(2) 74.47(2)
S(2)–C(3)–C(4)
S(2)–C(3)–C(8)
C(4)–C(3)–C(8)
111.2(2)
112.35(19)
110.5(2)
Eur. J. Inorg. Chem. 2007, 5677–5683
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5679

L. A. Oro, J. J. Pérez-Torrente et al.
FULL PAPER
2
It is worth noting that the average Rh–S–Rh and S–Rh– (JRh–C = 75 Hz, JP–C = 17 Hz). The IR spectra of both
S bond angles, 74.25(2)° and 71.07(2)°, are significantly compounds in dichloromethane show a broad ν(CO) band
smaller than those found in the related dinuclear dithiolate for the terminal carbonyl groups at 1957 (2) and 1960 cm–1
complexes cis-[Rh(µ-SPh)(CO)(PMe3)]2 [79.3(5) and (3), which is in good agreement with a trans disposition of
81.0(1)°][15] and cis-[Rh(µ-StBu)(CO)(PPh3)]2 [81.6(3) and the ligands.[7b,7f,10b]
80.7(3)°].[16] Both parameters are strongly influenced by the
The mixed-carbonyl compound 2 is also an active pre-
narrow angle of 96.82(12)° centered on the bridgehead car- cursor in the hydroformylation of oct-1-ene. Although re-
bon atom of the 1,1-cyclohexanedithiolato ligand [S(1)– lated dinuclear thiolate systems have shown that diolefin
C(3)–S(2)] that enables the S donor atoms to come closer complexes, in the presence of PR3 ligands, are more active
together [nonbonding S···S distance of 2.7833(10) Å] and, than the mixed carbonyl-phosphane species under the same
in turn, this results in a very small angle of 91.04(2)° be- experimental conditions,[7f] in the present case comparable
tween both rhodium coordination planes (defined only by chemoselectivity, regioselectivity, and activity are obtained
the metal-coordinated atoms) and a short Rh···Rh distance when using the same P/Rh ratio (entry 6).
of 2.8903(3) Å [112.25(3) and 111.61(12)°, 3.061(1) and
The results presented in Table 1 indicate that the activity
3.103(6) Å in the above-mentioned dithiolate complexes, of the catalytic systems decrease with the basicity of the P-
respectively]. The RhS2Rh torsion angle of 95.76(2)°, which donor ligands. Interestingly, the reverse trend was shown
is closely related to the Rh···Rh distance, is slightly larger for dinuclear systems based on functionalized amino-thiol-
than the angle between the coordination planes, as a conse- ate ligands in the hydroformylation of hex-1-ene.[7e,7f] On
quence of the separation of the metal atoms from their co- the other hand, it is evident that the regioselectivity is not
ordination planes by 0.0635(2) and 0.1670(2) Å [Rh(1) and controlled exclusively by steric factors since the more steri-
Rh(2), respectively]. This fact reflects the existence of a fee- cally demanding ligand (PPh3) affords the lower regioselec-
ble repulsion between the metal atoms that is due to the tivities. This fact was already observed in the hydrofor-
ligand-forced, short metal–metal nonbonding distance, as mylation of hex-1-ene with a cationic dinuclear catalyst pre-
suggested before in other similar cases.[9,17]
cursor having an aminothiolato-bridged ligand.[7f]
The geometrical constraints imposed by the gem-dithiol-
As far as the nuclearity of the active species during catal-
ato ligand in 2 relative to the other dithiolato ligands are ysis is concerned, we are aware that, under hydrofor-
largely reflected both in the smaller S–Rh–S angles, 79.02(6) mylation conditions, some dinuclear rhodium complexes
and 84.49(19)° in the complexes [Rh(µ-S(CH2)2S)(cod)2] containing thiolate bridging ligands are precursors of the
and [Rh(µ-S(CH2)3S)(cod)2],[9] and in the smaller angle be- mononuclear rhodium(I) hydrido species that probably ac-
tween the rhodium coordination planes, 96.95 and 103.99 count for the catalytic activity.[12] In spite of the obtained
respectively, although the Rh–S–Rh angles and the Rh···Rh regioselectivities, that are roughly comparable to those ob-
distances are of comparable magnitude. On the other hand, served in the catalytic systems [Rh(acac)(CO)2]/PPh3 and
the structural parameters of the central core in 2 compares [Rh(acac)(CO)2]/P(OPh)3, the recovery of the dinuclear
well with those observed in the structurally related dinu- compound 2 after the catalytic reaction in the system
clear compound [Rh2{µ-S2CN(Me)(Ph)}(cod)2], which has 1/PPh3 and the singular structural features of the compact
a dithiocarbamate bridging ligand exhibiting the same co- [Rh(µ-S2CR2)Rh] core strongly motivate us to look further
ordination mode.[18]
into the chemical behavior of these types of compounds.
The spectroscopic data indicate that compound 2 exists Further studies on the synthesis and reactivity of dinuclear
in solution mainly as the trans isomer, which was observed rhodium complexes containing new gem-dithiolato ligands
as a complex resonance in the 31P{1H} NMR (C6D6) spec- are currently under way, in order to determinate the influ-
trum at δ = 41.90 ppm. This signal correlates well with the ence of the bridging ligand on the catalytic activity and to
calculated spectrum using the parameters reported in the analyze a potential intermetallic cooperative mechanism in
Experimental Section, which resulted from the consider- these dimetallic species.
2
ation of small JRh–P
,
3JP–P, and JRh–Rh coupling con-
stants.[19] However, the cis isomer was also observed (Ͻ5%)
as a doublet at δ = 39.6 (JRh–P = 162 Hz). The dinuclear
compound [Rh2(µ-S2Chxn)(CO)2(PCy3)2] (3) was prepared
Conclusions
We have shown that novel dinuclear gem-dithiolato-
in excellent yield following a similar synthetic protocol bridged rhodium complexes can easily be obtained in high
starting from [Rh(acac)(CO)(PCy3)] (Figure 2). The yields directly by double deprotonation of a gem-dithiol
31P{1H} NMR (C6D6) spectrum only shows a resonance at compound by using mono- or dinuclear rhodium complexes
δ = 53.10 ppm with a similar pattern to that found in the containing basic ligands. The diolefin compound [Rh2(µ-
spectrum of compound 2 and suggests that compound 3 S2Chxn)(cod)2] (1) is an active catalyst precursor in the
exists exclusively as the trans isomer. This fact is probably presence of P-donor ligands for the hydroformylation of
associated to the bulkiness of the PCy3 ligands that com- oct-1-ene under mild conditions. The performance of the
pletely disfavors the formation of the cis isomer. The signals resulting catalytic systems is strongly dependent on the na-
for the equivalent carbonyl groups are observed in the ture of the modifying P-donor ligand, and it has been found
13C{1H} NMR (C6D6) spectra of both compounds as a that ligands of the type P(OR)3 are better than PR3 ligands
doublet of doublets at δ = 192.2 (2) and 191.8 ppm (3) in terms of both activity and selectivity.
5680
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5677–5683

Novel gem-Dithiolato-Bridged Rhodium Hydroformylation Catalysts
C 56.68, H 5.15, S 6.85. 1H NMR (300.08 MHz, C6D6, 293 K):
δ = 7.93 (m, 12 H), 7.05 (m, 18 H, PPh3), 2.63 (m, 4 H, ϾCH2),
1.45 (m, 4 H, ϾCH2), 1.07 (m, 2 H, ϾCH2, Chxn) ppm. 13C{1H}
Experimental Section
General: All manipulations were performed under a dry argon
atmosphere by using Schlenk-tube techniques. Solvents were dried
by standard methods and distilled under argon immediately prior
to use. Standard literature procedures were used to prepare the
complexes [Rh(µ-OH)(cod)]2,[20] [Rh(acac)(CO)(PPh3)],[21] and
[Rh(acac)(CO)(PCy3)].[22] 1,1-Dimercaptocyclohexane was pre-
pared according to the reported method.[23] Oct-1-ene was pur-
chased from Aldrich and was distilled prior to use.
2
NMR (75.46 MHz, C6D6, 293 K): δ = 192.2 (dd, JRh–C = 75, JP–
= 17 Hz, CO), 135.5 (d, JP–C = 45 Hz), 134.4 (d, JP–C = 12 Hz),
C
130.1, 128.5 (d, JP–C = 12 Hz, PPh3), 86.7 (C1), 57.6 (C2 and C6),
24.6 (C4), 21.7 (C3 and C5, Chxn) ppm. 31P{1H} NMR
(121.47 MHz, C6D6, 293 K): δ = 41.90 (AAЈXXЈ spin system, A =
2
31P and X = 103Rh, calcd. spectrum: JRh–P = 163.72 Hz, JRh–P
=
3
–1.47 Hz, JP–P = 6.60 Hz, JRh–Rh = 3.59 Hz, trans isomer), 39.6
(d, JRh–P = 162 Hz, cis isomer). MS (FAB+, CH2Cl2): m/z (%) =
932 (25) [M+] 904 (20) [M+ – CO], 876 (15) [M+ – 2CO], 532 (100)
[M+ – Chxn – 2CO – PPh3]. Mol. Weight (CHCl3): calcd 932; found
940. IR (pentane ): ν(CO) = 1957 (s) cm–1.
1
Physical Measurements: H, 31P{1H}, and 13C{1H} NMR spectra
were recorded on a Varian Gemini 300 spectrometer operating at
1
300.08 MHz for H. Chemical shifts are reported in parts per mil-
lion and referenced to SiMe4 by using the residual resonances of
the deuterated solvents (1H and 13C) and 85% H3PO4 (31P) as ex-
ternal reference. Assignments in complex NMR spectra were made
by simulation with the program gNMR© v 3.6 (Cherwell Scientific
Publishing Limited) for Macintosh. The initial choice of chemical
shifts and coupling constants were optimized by successive itera-
tions following a standard least-squares procedure; a numerical as-
signment of the experimental frequencies was used. IR spectra were
recorded on a Nicolet-IR 550 spectrometer. Elemental C, H and N
analysis were performed with a Perkin–Elmer 2400 microanalyzer.
Molecular weights were determined with a Knauer osmometer by
using chloroform solutions of the complexes. Mass spectra were
recorded in a VG Autospec double-focusing mass spectrometer op-
erating in the FAB+ mode. Ions were produced with the standard
Cs+ gun at ca. 30 Kv; 3-nitrobenzylic alcohol (NBA) was used as
matrix. Hydroformylation experiments were carried out in a stain-
less steel magnetically stirred autoclave (100 mL) equipped with a
thermocouple and an external heating mantle. The syngas (CO/H2
= 1) was supplied at constant pressure from a ballast. The drop in
pressure in the ballast was monitored by using a pressure trans-
ducter.
Preparation of [Rh2(µ-S2Chxn)(CO)2(PCy3)2] (3): [Rh(acac)-
(CO)(PCy3)] (0.367 g, 0.719 mmol) and Chxn(SH)2 (50 µL,
0.365 mmol, ρ = 1.083 gmL–1) were reacted in diethyl ether
(15 mL) for 15 min to give an orange suspension. The suspension
was concentrated under vacuum to about one half the volume and
cooled to –85 °C. The orange microcrystalline solid was filtered,
washed with cold pentane (2ϫ5 mL), and dried under vacuum.
Yield: 0.319 g (92%). C44H76O2P2Rh2S2 (968.96): calcd. C 54.54, H
7.90, S 6.62; found C 54.22, H 7.98, S 6.50. 1H NMR (300.08 MHz,
CDCl3, 293 K): δ = 2.85 (m, 6 H, PCy3), 2.19–2.06 (m, 28 H), 1.80–
1.65 (m, 28 H), 1.25–1.10 (m, 14 H), (PCy3, Chxn) ppm. 13C{1H}
NMR (75.46 MHz, CDCl3, 293 K): δ = 191.8 (dd, JRh–C = 75,
2JP–C = 17 Hz, CO), 84.5 (C1), 57.1 (C2 and C6, Chxn), 35.7 (d,
JP–C = 21Hz), 26.8 (d, JP–C = 11 Hz), 26.7 (d, JP–C = 10 Hz), 25.7
(PCy3), 23.9 (C4), 21.1 (C3 and C5, Chxn) ppm. 31P{1H} NMR
(121.47 MHz, C6D6, 293 K): δ = 53.10 (AAЈXXЈ spin system, A =
2
31P and X = 103Rh, calcd. spectrum: JRh–P = 158.30 Hz, JRh–P
=
3
–0.54 Hz, JP–P = 3.79 Hz, JRh–Rh = 3.74 Hz, trans isomer). MS
(FAB+, CH2Cl2): m/z (%) = 968 (100) [M+], 938 (96) [M+ – CO –
2 H], 908 (65) [M+ – 2CO – 4 H]. Mol. weight (CHCl3): calcd 968;
found 970. IR (pentane ): = ν(CO) = 1960 (s) cm–1.
Preparation of [Rh2(µ-S2Chxn)(cod)2] (1): To a solution of [Rh(µ-
OH)(cod)]2 (0.502 g, 1.100 mmol) in CH2Cl2 (5 mL) was added 1,1-
dimercaptocyclohexane (Chxn(SH)2, 170 µL, 1.241 mmol, ρ =
1.083 gmL–1) to give a red-orange solution, which was stirred for
15 min. The addition of EtOH (10 mL) gave a red suspension,
which was concentrated under vacuum to ca. 5 mL and then fil-
tered to give a red-orange microcrystalline solid, which was washed
with EtOH (2ϫ3 mL) and dried under vacuum. Yield: 0.511 g
(82%). C22H34Rh2S2 (568.44): calcd. C 46.48, H 6.03, S 11.28;
found C 46.53, H 6.05, S 11.53. 1H NMR (300.08 MHz, CDCl3,
293 K): δ = 4.54 (m, 4 H, =CH), 4.23 (m, 4 H, =CH, cod), 2.43
(m, 12 H, ϾCH2, cod and Chxn), 1.98 (m, 4 H, ϾCH2), 1.83 (m,
4 H, ϾCH2, cod), 1.44 (m, 4 H, ϾCH2), 1.26 (m, 2 H, ϾCH2,
Chxn) ppm. 13C{1H} NMR (75.46 MHz, CDCl3, 293 K): δ = 84.0
(C1, Chxn), 79.8 (d, JRh–C = 12 Hz, =CH), 79.1 (d, JRh–C = 12 Hz,
=CH, cod), 57.2 (C2 and C6, Chxn), 31.4 and 31.1 (ϾCH2, cod),
24.3 (C4), 22.0 (C3 and C5, Chxn) ppm. MS (FAB+, CH2Cl2): m/z
(%) = 568 (100) [M+] 460 (60) [M+ – cod]. Mol. weight (CHCl3):
calcd 568; found 562.
Standard Hydroformylation Experiment: In a typical run, a solution
of the catalyst precursor [Rh2(µ-S2Chxn)(cod)2] (1) (0.017 mmol),
which contains the phosphane or phosphite ligand (0.20–
0.60 mmol), oct-1-ene (10.2 mmol), and toluene (15.4 mL), was
transferred from a Schlenk tube under argon to the autoclave by
using a stainless steel cannula. The autoclave was purged with
syngas three times at 120 PSI and then pressurized at 50 PSI and
heated at 80 °C. When thermal equilibrium was reached, the pres-
sure was adjusted to 100 PSI, and the mixture stirred for 8 h with
the continuous supply of syngas at constant pressure. After the
reaction time, the autoclave was cooled to room temperature and
depressurized. The reaction mixture was analyzed by gas
chromatography with
equipped with
a
Hewlett–Packard 5890 instrument
a
capillary column (HP, ULTRA 1.
25 mϫ0.32 mmϫ0.17 µm) and a flame-ionization detector. The
products were quantified by the internal standard method by using
anisole.
Preparation of [Rh2(µ-S2Chxn)(CO)2(PPh3)2] (2): To a suspension
of [Rh(acac)(CO)(PPh3)] (0.501 g, 1.018 mmol) in diethyl ether
Crystal Structure Determination of [Rh2(µ-S2Chxn)(CO)2(PPh3)2]
(2): Suitable crystals for X-ray diffraction of compound 2 were ob-
tained from a saturated solution of the complex in dichlorometh-
ane/(diethyl ether) at 258 K. A summary of the crystal data, data
collection, and refinement parameters are given in Table 3. Inten-
sity data were collected at low temperature [150(2) K] on a Bruker
SMART diffractometer (equipped with a CCD area detector) by
using graphite-monochromated Mo-Kα radiation (λ = 0.71073 Å).
Data were integrated with the Bruker SAINT package,[24] and ab-
sorption corrections were applied by the SADABS program.[25] The
(15 mL) was added Chxn(SH)2 (73 µL, 0.533 mmol,
ρ =
1.083 g mL–1) to immediately give an orange solution, which was
stirred for 15 min. The addition of MeOH (15 mL) gave an orange
suspension, which was stirred for 5 min and concentrated under
vacuum to about one half of the original volume and then filtered
to give an orange solid, which was washed with cold MeOH
(2ϫ5 mL) and dried under vacuum. Yield: 0.433 g (92%).
C44H40O2P2Rh2S2 (932.68): calcd. C 56.66, H 4.32, S 6.87; found
Eur. J. Inorg. Chem. 2007, 5677–5683
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5681

L. A. Oro, J. J. Pérez-Torrente et al.
FULL PAPER
[4]
[5]
E. K. Van der Beuken, B. L. Feringa, Tetrahedron 1998, 54,
structure was solved by direct methods and completed by subse-
quent difference Fourier techniques. Refinement on F2 was carried
out by full-matrix least-squares methods (SHELXL97).[26] All non-
hydrogen atoms were refined with anisotropic displacement param-
eters; all hydrogen atoms were observed in the difference Fourier
maps and refined as free isotropic atoms. CCDC-654127 contains
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
12985.
a) P. Kalck, J. M. Frances, P. M. Pfister, T. G. Southern, A.
Thorez, J. Chem. Soc. Chem. Commun. 1983, 510; b) P. Kalck
in Organometallics in Organic Syntheses (Eds.: A. de Meijere,
H. Tom Dick), Springer Verlag, Weinheim, 1987, p. 297; c) P.
Kalck, Pure Appl. Chem. 1989, 61, 967; d) P. Kalck, Polyhedron
1988, 7, 2441; e) F. Monteil, R. Queau, P. Kalck, J. Organomet.
Chem. 1994, 480, 177.
[6]
[7]
R. Davis, J. W. Epton, T. G. Southern, J. Mol. Catal. J. Mol.
Catal. A 1992, 77, 159.
Table 3. Crystal data, data collection, and refinement parameters
for the X-ray analysis of [Rh2(µ-S2Chxn)(CO)2(PPh3)2] (2).
a) J. L. G. Fierro, M. Martínez-Ripoll, M. Merchán, A. Rodri-
guez, P. Terreros, H. Torrens, M. A. Vivar-Cerrato, J. Or-
ganomet. Chem. 1997, 544, 243; b) A. Polo, C. Claver, S. Cas-
tillón, A. Ruiz, J. C. Bayón, J. C. Real, J. Mealli, D. Masi, Orga-
nometallics 1992, 11, 3525; c) A. Polo, E. Fernandez, C. Claver,
S. Castillón, J. Chem. Soc. Chem. Commun. 1992, 642; d) C.
Claver, A. M. Masdeu, N. Ruiz, C. Foces-Foces, F. Cano, C.
Apreda, L. A. Oro, J. García-Alejandre, H. Torrens, J. Or-
ganomet. Chem. 1990, 398, 177; e) J. C. Bayón, J. Real, C. Cla-
ver, A. Polo, A. Ruiz, J. Chem. Soc. Chem. Commun. 1989,
1056; f) J. C. Bayón, P. Esteban, J. Real, C. Claver, A. Ruiz, J.
Chem. Soc. Dalton Trans. 1989, 1579.
a) J. C. Bayón, C. Claver, A. M. Masdeu-Bultó, Coord. Chem.
Rev. 1999, 193–195, 73; b) N. Ruiz, S. Castillón, A. Ruiz, C.
Claver, A. Aaliti, A. Alvarez-Larena, J. F. Piniella, G. Germain,
J. Chem. Soc. Dalton Trans. 1996, 969; c) A. Aaliti, A. M. Mas-
deu, A. Ruiz, C. Claver, J. Organomet. Chem. 1995, 489, 101.
Formula
Mr
C44H40O2P2Rh2S2
932.64
Crystal size [mm]
Temperature
Crystal System
Space group
a [Å]
b [Å]
c [Å]
0.28ϫ0.24ϫ0.20
150(2)
monoclinic
P21/n
13.2115(6)
19.2395(9)
16.0070(7)
102.2275(11)
4
3976.4(3)
1.558
β [°]
[8]
Z
V [Å3]
Dcalcd. [gcm–3]
µ [mm–1]
1.052
Θ range [°]
No. measured reflections
No. unique reflections
min/max transmission
2.79–32.06
19622
10256 (Rint = 0.0419)
0.668/0.812
[9]
A. M. Masdeu, A. Ruiz, S. Castillón, C. Claver, P. B. Hitch-
cock, P. A. Chaloner, C. Bó, J. M. Poblet, P. Sarasa, J. Chem.
Soc. Dalton Trans. 1993, 2689.
No. reflections/restrainsts/parameters 10256/0/629
[10]
a) Z. Freixa, E. Martin, S. Gladiali, J. C. Bayon, Appl. Or-
ganomet. Chem. 2000, 14, 57; b) O. Pamies, O. G. Net, A. Ruiz,
C. Bó, J. M. Poblet, C. Claver, J. Organomet. Chem. 1999, 586,
125; c) A. Castellanos-Páez, S. Castillón, C. Claver, J. Or-
ganomet. Chem. 1997, 539, 1; d) N. Ruiz, A. Aaliti, J. Forniés-
Cámer, A. Ruiz, C. Claver, C. J. Cardin, D. Fabbri, S. Gladiali,
J. Organomet. Chem. 1997, 546, 79; e) S. Gladiali, J. C. Bayón,
C. Claver, Tetrahedron: Asymmetry 1995, 6, 1453; f) A. M.
Masdeu-Bultó, A. Orejón, S. Castillón, C. Claver, Tetrahedron:
Asymmetry 1995, 6, 1885; g) A. M. Masdeu, A. Orejón, A.
Ruiz, S. Castillón, C. Claver, J. Mol. Catal. J. Mol. Catal. A
1994, 94, 149; h) C. Claver, S. Castillón, A. Ruiz, G. Delogu, D.
Fabri, S. Gladiali, J. Chem. Soc. Chem. Commun. 1993, 1833.
R1(F) [F2 Ն2σ(F2)]
wR2(F2) (all data)
S (all data)
0.0342
0.0720
0.914
Acknowledgments
The financial support from Ministerio de Educación y Ciencia
(MEC/FEDER) is gratefully acknowledged (Project CTQ2006-
03973/BQU and Factoría de Cristalización, CONSOLIDER IN-
GENIO-2010). A. B. R. thanks the Programa Iberoamericano de
Ciencia y Tecnología para el Desarrollo (CYTED, Project V.9) for
a fellowship. A. J. P. thanks to Fonacit-Venezuela (S1-2002000260)
for financial support.
[11]
[12]
A. Castellanos-Páez, J. Thayaparan, S. Castillón, C. Claver, J.
Organomet. Chem. 1998, 551, 375.
a) A. Castellanos-Páez, S. Castillón, C. Claver, P. W. N. M.
van Leeuwen, W. G. J. Lange, Organometallics 1998, 17, 2543;
b) H. Gao, R. J. Angelici, Organometallics 1998, 17, 3063; c)
M. Diéguez, C. Claver, A. M. Masdeu-Bultó, A. Ruiz,
P. W. N. M. van Leeuwen, G. Schoemaker, Organometallics
1999, 18, 2107; d) J. Forniés-Cámer, A. M. Masdeu-Bultó, C.
Claver, C. Tejel, M. A. Ciriano, C. J. Cardin, Organometallics
2002, 21, 2609.
a) F. Novio, R. Mas-Balleste, I. Gallardo, P. Gonzalez-Duarte,
A. Lledos, N. Vila, Dalton Trans. 2005, 2742; b) T. Gandhi,
B. R. Jagirdar, Inorg. Chem. 2005, 44, 1118; c) Z. Li, W. Zheng,
H. Liu, K. F. Mok, T. S. A. Hor, Inorg. Chem. 2003, 42, 8481;
d) R. P. Hughes, J. M. Smith, C. D. Incarvito, K.-C. Lam, B.
Rhatigan, A. L. Rheingold, Organometallics 2002, 21, 2136; e)
C. Megret, P. Gonzalez-Duarte, Inorg. Chem. 2002, 41, 3218;
f) M. R. DuBois, B. R. Jagirdar, S. Dietz, B. C. Noll, Organo-
metallics 1997, 16, 294; g) H. Hashimoto, M. Y. Shang, T. P.
Fehlner, Organometallics 1996, 15, 1963; h) R. Mas-Balleste,
M. Capdevila, P. A. Champkin, W. Clegg, R. A. Coxall, A.
Lledos, C. Megret, P. Gonzalez-Duarte, Inorg. Chem. 2002, 41,
3218; i) K.-Y. Shih, P. E. Fanwick, R. A. Walton, Inorg. Chem.
1992, 31, 3663; j) D. P. Klein, G. M. Kloster, R. G. Bergman,
J. Am. Chem. Soc. 1990, 112, 2022; and references cited therein.
[1] a) P. Braunstein, L. A. Oro, P. R. Raithby (Eds.), Metal Clus-
ters in Chemistry, Wiley-VCH, Weinheim, 1999; b) R. D. Ad-
ams, F. A. Cotton (Eds.), Catalysis by Di- and Polynuclear
Metal Cluster Complexes, Wiley-VCH, New York, 1998.
[2] a) Y. Ishii, M. Hidai in Multimetallic Catalysts in Organic Syn-
thesis (Eds.: M. Shibasaki, Y. Yamamoto), Wiley-VCH,
Weinheim, 2004, ch. 9, pp. 201–223; b) N. Tsukada, T. Mitsu-
boshi, H. Setoguchi, Y. Inoue, J. Am. Chem. Soc. 2003, 125,
12102; c) N. Wheatly, P. Kalck, Chem. Rev. 1999, 99, 3379; d)
P. Braunstein, J. Rosé in Comprehensive Organometallic Chem-
istry II (Eds.: E. W. Abel, F. G. A. Stone, G. Willkinson), Elsev-
ier Science Ltd., Oxford, 1995, vol. 10, ch. 7, pp. 351–385.
[3] a) G. G. Stanley in Multimetallic Catalysts in Organic Synthesis
(Eds.: M. Shibasaki, Y. Yamamoto), Wiley-VCH, Weinheim,
2004, ch. 10, pp. 225–248; b) M. E. Brousand, B. Juma, S. G.
Train, W.-J. Peng, S. A. Laneman, G. G. Stanley, Science 1993,
260, 1784; c) R. C. Matthews, D. K. Howell, W.-J. Peng, S. G.
Train, W. Dale-Treleaven, G. G. Stanley, Angew. Chem. Int. Ed.
Engl. 1996, 35, 2253.
[13]
5682
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5677–5683

Novel gem-Dithiolato-Bridged Rhodium Hydroformylation Catalysts
[14] a) V. Miranda-Soto, J. J. Pérez-Torrente, L. A. Oro, F. J. Lahoz,
M. L. Martín, M. Parra-Hake, D. B. Grotjahn, Organometal-
lics 2006, 25, 4374; b) J. R. Dilworth, D. Morales, Y. Zheng,
J. Chem. Soc. Dalton Trans. 2000, 3017; c) H. Schumann, N.
[20] R. Uson, L. A. Oro, J. A. Cabeza, Inorg. Synth. 1985, 23, 126.
[21] F. Bonati, G. Wilkinson, J. Chem. Soc. 1964, 3156.
[22] A. Jegorov, J. Podlaha, J. Podlahová, F. Turecek, J. Chem. Soc.
Dalton Trans. 1990, 3295.
Hemling, N. Goren, J. Blue, J. Organomet. Chem. 1995, 485, [23] a) M. Demuynck, J. Vialle, Bull. Soc. Chim. Fr. 1962, 2126; b)
209.
C. Djerassi, B. Tursch, J. Org. Chem. 1962, 22, 1041.
[24] SAINT+: Software for CCD Difractometers, v. 6.01, 2001,
Bruker AXS, Madison, WI, and SAINT, version 6.02.
[25] a) R. H. Blessing, Acta Crystallogr., Sect. A 1995, 51, 33; b)
SADABS: Area-Detector Absorption Correction, 1996, Bruker
AXS, Madison, WI.
[15] J. J. Bonet, P. Kalck, R. Poilblanc, Inorg. Chem. 1977, 16, 1514.
[16] R. A. Jones, S. T. Schwab, J. Crystallogr. Spectrosc. 1986, 16,
577.
[17] J. Forniés-Cámer, A. M. Masdeu-Bultó, C. Claver, C. J. Cardin,
Inorg. Chem. 1998, 37, 2626.
[18] A. Elduque, C. Finestra, J. A. López, F. J. Lahoz, F. Merchán,
L. A. Oro, M. T. Pinillos, Inorg. Chem. 1998, 37, 824.
[19] M. A. F. Hernandez-Gruel, J. J. Pérez-Torrente, M. A. Ciriano,
A. B. Rivas, F. J. Lahoz, I. T. Dobrinovitch, L. A. Oro, Organo-
metallics 2003, 22, 1237.
[26] G. M. Sheldrick, SHELXL-97 Program for Crystal Structure
Refinement, University of Göttingen, Göttingen, Germany,
1997.
Received: July 17, 2007
Published Online: November 15, 2007
Eur. J. Inorg. Chem. 2007, 5677–5683
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5683