Synthesis of Rh(I) and Ir(I) metal complexes with the first two chiral dithiolate ligands derived from carbohydrates

Article abstract of DOI:10.1016/S0022-328X(99)00251-X

New chiral dithiol compounds 1,2-O-isopropylidene-3,5-dithiol-α-D-ribofuranose RiSSH₂ and 1,2-O-isopropylidene-3,5-dithiol-α-D-xylofuranose XySSH₂ and their chiral dithiolate olefinic Rh(I) and Ir(I) complexes [M₂(μ-RiSS)(cod)₂] (M=Rh 1, Ir 3) and [M₂(μ-XySS)(cod)₂] (M=Rh 2, Ir 4) were synthesized and characterized. The Rh(I) complexes reacted with CO to give the corresponding binuclear tetracarbonyls [Rh₂(μ^_RiSS)(CO)₄] (5) and [Rh₂(μ-XySS)(CO)₄] (6), which reacted with two equivalents of PPh₃ to form mixed carbonyl-phosphine complexes [Rh₂(μ-RiSS)(CO)₂(PPh₃)₂] (7) and [Rh₂(μ-XySS)(CO)₂(PPh₃)₂] (8). The structures of [Rh₂(μ-RiSS)(CO)₄] (5) and the two possible conformers of [Rh₂(μ-XySS)(CO)₄] (6a) and (6b) were optimized and their relative stability determined by theoretical methods based on density functional theory (DFT). An interaction between one Rh atom and the oxygen atom of the ring was found in the most stable conformer of [Rh₂(μ-XySS)(CO)₄]. The complexes [Rh₂(μ-RiSS)(cod)₂] (1) and [Rh₂(μ-XySS)(cod)₂] (2) were used as catalytic precursors in the hydroformylation of styrene. The results suggest that mononuclear rhodium hydride carbonyl species are responsible for the catalytic activity.

Full text of DOI:10.1016/S0022-328X(99)00251-X

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 586 (1999) 125–137
Synthesis of Rh(I) and Ir(I) metal complexes with the first two
chiral dithiolate ligands derived from carbohydrates
Oscar Pa`mies, Gemma Net, Aurora Ruiz *, Carles Bo, Josep Maria Poblet,
Carmen Claver
Departament de Qu´ımica F´ısica i Inorga`nica, Uni6ersitat Ro6ira i Virgili, Pl. Imperial Ta`rraco 1, E-43005 Tarragona, Spain
Received 23 February 1999; received in revised form 24 April 1999
Abstract
New chiral dithiol compounds 1,2-O-isopropylidene-3,5-dithiol-a-
D
-ribofuranose RiSSH2 and 1,2-O-isopropylidene-3,5-dithiol-
a- -xylofuranose XySSH2 and their chiral dithiolate olefinic Rh(I) and Ir(I) complexes [M2(m-RiSS)(cod)2] (M=Rh 1, Ir 3) and
D
[M2(m-XySS)(cod)2] (M=Rh 2, Ir 4) were synthesized and characterized. The Rh(I) complexes reacted with CO to give the
–
corresponding binuclear tetracarbonyls [Rh2(m RiSS)(CO)4] (5) and [Rh2(m-XySS)(CO)4] (6), which reacted with two equivalents
of PPh3 to form mixed carbonyl-phosphine complexes [Rh2(m-RiSS)(CO)2(PPh3)2] (7) and [Rh2(m-XySS)(CO)2(PPh3)2] (8). The
structures of [Rh2(m-RiSS)(CO)4] (5) and the two possible conformers of [Rh2(m-XySS)(CO)4] (6a) and (6b) were optimized and
their relative stability determined by theoretical methods based on density functional theory (DFT). An interaction between one
Rh atom and the oxygen atom of the ring was found in the most stable conformer of [Rh2(m-XySS)(CO)4]. The complexes
[Rh2(m-RiSS)(cod)2] (1) and [Rh2(m-XySS)(cod)2] (2) were used as catalytic precursors in the hydroformylation of styrene. The
results suggest that mononuclear rhodium hydride carbonyl species are responsible for the catalytic activity. © 1999 Elsevier
Science S.A. All rights reserved.
Keywords: Rhodium(I); Iridium(I); Chiral dithiolates; Carbohydrate derivatives; DFT molecular orbital calculations
1. Introduction
ands with carbohydrates as building blocks are P-lig-
ands derived from glucose but there are also examples
derived from other monosaccharides or other carbohy-
drates [4,5]. Some of them have given very high enan-
tioselectivities in asymmetric catalysis [5–9] and have
In recent decades the efficiency of asymmetric homo-
geneous catalysis with chiral organometallic complexes
in the synthesis of optical pure drugs, agrochemicals,
flavors or fragrancies has meant that great efforts have
been made to design new chiral ligands [1]. One of the
simplest methods of obtaining chiral substances is the
transformation or derivatization of readily available
natural chiral compounds (the so-called chiral pool).
Some carbohydrates are particularly advantageous be-
cause they are inexpensive compounds [2]. Thus, in the
last three decades, carbohydrates have been used as
starting material in the synthesis of chiral auxiliaries for
asymmetric synthesis [3]. Nevertheless, despite the ac-
cessibility of carbohydrate synthons, the full potential
of the carbohydrate chiral pool for providing chiral
ligands has hardly been exploited [4]. Most chiral lig-
been applied in industry to synthesize
L-DOPA [10].
The S-ligands derived from carbohydrates have been
studied much less than their P-counterparts. To the best
of our knowledge, the only contributions to this area
study compounds derived from glucose, one of which is
a ligand with only sulfur as donor atom [11], while the
rest are P–S ligands [12–14].
In homogeneous catalysis, sulfur ligands are efficient
for reactions such as the hydroformylation or hydro-
genation of olefines [15]. This feature, together with
their advantages over phosphorous ligands (less expen-
sive, less toxic and less oxidisable) [16] is an incentive
for further investigation in this field.
Here we report the synthesis of the first dithiolate
ligands with a backbone derived from carbohydrates.
Two epimer ligands derived from ribofuranose and
* Corresponding author. Fax: +34-977-559563.
E-mail address: aruiz@quimica.urv.es (A. Ruiz)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00251-X

126
O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
xylofuranose, respectively (Scheme 1) and their corre-
sponding Rh(I) and Ir(I) complexes have been synthe-
sized. Following our previous work in hydro-
formylation catalyzed by dithiolate rhodium complexes
[17], we also report the results obtained in the hydro-
formylation of styrene by using rhodium complexes
with the new dithiolate ligands as precatalysts.
2. Results and discussion
2.1. Synthesis of chiral dithiols RiSSH2 and XySSH2
RiSSH2 and XySSH2 were prepared according to
Scheme 1. The bisthioacetate ribose derivative V has
already been described in the literature starting from
the corresponding ditosylate [18]. All attempts to repro-
Scheme 2. Synthesis of the complexes with XySS and RiSS (1–8).
duce this reaction led mainly to decomposition prod-
ucts or, in milder conditions, to the substitution of only
one of the tosyl groups. The bisthioacetate ribose and
xylose derivatives V and VI were finally obtained from
the corresponding ditriflates III and IV. The yield ob-
tained in the case of V was much lower because the
nucleofilic attack took place through the most sterically
hindered face. This unfavorable kinetics led to sec-
ondary reactions, principally the elimination of the
triflyl group. The dithiol derivatives RiSSH2 and
XySSH2 were easily obtained by reacting V and VI with
the exchange resin IRA-420 in methanol. NaBH4 in
ethanol (75%) was also able to reduce the thioacetate
groups to thiol groups though with lower yields.
2.2. Synthesis of the dithiolate complexes
2.2.1. Olefinic Rh(I) and Ir(I) complexes
The [Rh2(m-RiSS)(cod)2] (1) and [Rh2(m-XySS)(cod)2]
(2) red complexes were synthesized easily from the
bisthioacetate derivatives, V and VI, and [Rh(m-
OMe)(cod)]2 with sodium methoxide acting as catalyst
(Scheme 2). Synthesis from the corresponding dithiols
RiSSH2 and XySSH2 and [Rh(m-OMe)(cod)]2 gave sim-
ilar yields. Though both compounds are air-stable
solids, they decompose slowly in solution if not kept
under nitrogen. Complex 1 is by far the least stable.
Dithiolate Rh(I) complexes are known to be dinuclear
or tetranuclear [19]. The FAB mass spectra for 1 and 2
show the heaviest peaks at m/z 642. This suggests that
both compounds are dinuclear. The anomeric proton of
1 (5.55 ppm) appears at higher field than the anomeric
proton of RiSSH2 (5.82 ppm), as expected from the
coordination to the metal [20], whereas the anomeric
Scheme 1. Synthesis of H2RiSS and H2XySS.

O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
127
proton of 2 (6.34 ppm) appears further downfield than
XySSH2 (5.93 ppm). Below we shall discuss this shift
downfield with coordination.
DFT level were carried out for 5 and the two conform-
ers of 6. The optimized structures and their relative
energies are shown in Fig. 1. Their most representative
intramolecular bond distances and angles are summa-
rized in Table 1. From these results it can be concluded
that the most stable conformer for complex 6 is 6a.
The three conformers can be described as bent dou-
ble square–planar structures. The difference between
the torsion angle (RhS2Rh) and the dihedral angle
between the coordination planes (the metals are not
included) suggests a slightly distorted square planar
coordination around each rhodium atom. The dihedral
angles between the coordination planes for the three
conformers are in the range from 127 to 129° and are
slightly larger than the values reported for other dinu-
clear dithiolate-bridged Rh complexes [17e,19a,21,23].
As expected from the large dihedral angles, the Rh–Rh
distances in the three conformers are longer than the
ones reported for the above mentioned compounds.
These distances are outside the accepted range for
metal–metal interactions in these kinds of compound,
The [Ir2(m-RiSS)(cod)2] (3) and [Ir2(m-XySS)(cod)2]
(4) brown complexes were synthesized analogously to
the parent rhodium compounds (Scheme 2). Their FAB
mass spectra suggest that the iridium complexes 3 and
4 are dinuclear. They are relatively air stable in the
solid state but they are very unstable in solution even
under nitrogen atmosphere, which meant that their
reactivity could not be studied. They could not be
1
characterized by H-NMR, even at low temperatures,
because of the presence of peaks that were too wide.
2.2.2. Rh (I) carbonylic complexes
Bubbling CO through solutions of 1 and 2 in
methylene chloride afforded the corresponding tetracar-
bonylic compounds [Rh2(m-RiSS)(CO)4] (5) and [Rh2(m-
XySS)(CO)4] (6) (Scheme 2). All attempts to isolate
these complexes failed because of their high solubility in
organic solvents. In contrast to their olefinic precursors
1 and 2, solutions of the carbonylic compounds 5 and 6
are air stable over long periods (weeks).
,
2.617–2.796 A [24]. The Rh–Rh distances in 5 and 6b
are practically identical, probably due to the similar
conformation of the dithiolate ligand in these two
conformers, and shorter than the Rh–Rh distance in
6a. No striking features can be observed in the bond
distances between the metal and the donor atoms. The
Rh–S bond distances (average 2.405, 2.394 and 2.393
All the dithiolate-bridged tetracarbonyl Rh(I) or Ir(I)
complexes that have been structurally characterized by
X-ray [21] or IR [17e,19] spectrometry are dinuclear.
The IR spectra of methylene chloride solutions of 5 and
6 show four stretching carbonyl frequencies near 2000
cm−1 and suggest that these complexes are also dinu-
clear (see the corresponding section).
,
A) are similar to those reported for other dithiolate-
bridged Rh complexes [17e,19,21,23]. The Rh–C bond
,
distances (average 1.865, 1.864 and 1.867 A) are close
1
Like the olefinic precursors, the H-NMR spectra of
to those of other dicarbonyl rhodium complexes [21].
The intramolecular distance between the sugar ring
oxygen and one of the rhodium atoms in 6a is only
5 and 6 show the same pattern. Again, the anomeric
proton of 5 (5.66 ppm) appears further upfield than the
anomeric proton of RiSSH2 (5.82 ppm), as expected
from the coordination to the metal, whereas the
anomeric proton of 6 (6.16 ppm) appears further
downfield than XySSH2 (5.93 ppm).
,
2.840 A at the local level and 2.982 at BP86 level, which
suggests a weak interaction between the two atoms.
According to the theory of atoms in molecules (AIM),
the presence of a bond critical point in the charge
density distribution linking two nuclei at the equi-
librium geometry is a necessary and sufficient condition
for the existence of a bond [25]. Such a bond critical
point could be characterized in both computed density
distributions of 6a between the ring O and the Rh
nuclei. The same analysis was made between the two
metal atoms. The absence of a bond critical point
indicates the absence of a metal–metal interaction as
suggests the intermetal distance.
2.2.3. Theoretical structures for [Rh2(RiSS)(CO)4] and
[Rh2(XySS)(CO)4]
Much interest has recently been shown in methods
based on density functional theory (DFT) as an alterna-
tive to ab initio schemes for organometallic complexes
[22]. Initial geometries were generated by means of
molecular mechanics calculations for complexes 5 and 6
as well as for olefinic complexes 1 and 2. During this
study, it was realized that whilst the coordinated
RySS2− ligand (complexes 1 and 5) can only have one
conformation, XySS2− complexes can be present as
two conformers, 2a and 2b for olefinic complexes and
6a and 6b for carbonylic complexes. The bridging dithi-
olate ligands of conformers 1, 5, 2b and 6b have a
similar conformation, but it is noteworthy that in con-
formers 2a and 6a the XySS2− ligand bends in such a
way that the oxygen from the sugar ring is above one of
the rhodium atoms. Full geometry optimizations at the
On the basis of the structures obtained for 6a and 6b,
1
the H-NMR shift downfield observed for the anomeric
proton of 2 and 6 upon coordination can be interpreted
by the anisotropic effect of the cod double bond in 2a
and the triple bond of one of the carbonyls in con-
former 6a, because both ligands are very near the
anomeric proton in this ligand conformation. Between
1
90 and −80°C, the H-NMR spectra of 2 and 6 did
not change, thus indicating that the only conformers
detected were 2a and 6a.

128
O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
It should be pointed out that for this kind of com-
C26 symmetric bisthiolate-bridged dicarbonyl Rh or Ir
complexes, Poilblanc and Kalck studied the relative
intensity of the three expected bands (A1, B1 and B2) as
a function of q [26]. The q angle is the dihedral angle
between the planes of the CO groups of the two
Rh(CO)2 moieties. This study cannot be applied in our
case because the asymmetry of the dithiolate bridging
ligand lowers the symmetry of the complex to C1. The
shoulder observed in the spectra of 5 and 6 at lower
energies is a fourth band (A2 in a C26 symmetry) which
is not active for a C26 symmetry but which becomes
active for a C1 symmetry. The intensity of this band is
much higher for complex [Rh2(XySS)(CO)4] than for
[Rh2(RiSS)(CO)4]. This agrees with the ‘higher asymme-
try’ of the XySS compound due to the Rh–O interac-
tion in only one of the Rh(CO)2 groups.
plexes the inexpensive molecular mechanics calculations
(MM) give quite good results. Calculated MM struc-
tures for 5, 6a and 6b are not very different from those
computed from DFT calculations (the differences in
bond angles and distances are often less than 0.5% and
never more than 2%). MM calculations were also able
to show that conformer 6b has greater energy than 6a.
2.2.4. Theoretical IR spectrum in the carbonylic region
for [Rh2(XySS)(CO)4]
The structures of [Rh2(RiSS)(CO)4] (5) and
[Rh2(XySS)(CO)4] (6) were optimized assuming that
they were dinuclear. The presence of four bands (Fig. 2)
in the carbonyl region of their IR spectra (eight bands
would be expected for a tetranuclear compound) and
the dinuclearity found in all solved crystal structures of
similar compounds [21] suggested this structure. In
order to confirm this, the IR spectra in the carbonylic
region of these compounds were further studied. For
The IR spectrum in the carbonylic region for one of
these complexes, [Rh2(XySS)(CO)4], was theoretically
calculated using the DFT methodology at the LDA
Fig. 1. Calculated structures of [Rh2(m-RiSS)(CO)4] (5) and [Rh2(m-XySS)(CO)4] (6a and 6b) with their relative energies at the non-local level
(values at the local level in parentheses).

O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
129
Table 1
1006 in the FAB mass spectra suggested that these
complexes were dinuclear. The stoichiometry [Rh2(m-
SS)(CO)2(PPh3)2], where m-SS is an asymmetric dithio-
late ligand, allows the four molecular isomeric
structures depicted in Fig. 3. In the case of 8, since the
coordinated dithiolate ligand XySS2− can be present in
two different conformations, the number of possible
isomers increases to eight. The 31P-NMR spectrum for
both complexes had eight doublets which suggested
that four possible structural isomers for this stoi-
chiometry were present in solution for 7 and 8 but in
different proportions (all the possible isomers for 7 and
four of the eight possible isomers for 8). IR spectra in
solution of 7 and 8 show three broad bands in the metal
carbonyl region, between 1950 and 2025 cm−1, which
confirm the presence of different isomers in solution.
In contrast, solid state IR spectra for these complexes
show a single broad band in this region at 1967 and
Selected geometrical parameters for [Rh2(m-RiSS)(CO)4] (5) and the
two conformers of [Rh2(m-XySS)(CO)4] (6a and 6b) calculated at the
non local level. Values obtained at the LDA level are shown in
parentheses a
5
6a
6b
,
Bond lengths (A)
Rh(1)–Rh(2)
Rh(1)–S(3)
Rh(1)–S(4)
Rh(2)–S(3)
Rh(2)–S(4)
Rh(1)–C(5)
Rh(1)–C(6)
Rh(2)–C(7)
Rh(2)–C(8)
q(RhS2Rh) b
3.252 (3.169) 3.295 (3.202) 3.267 (3.168)
2.455 (2.405) 2.441 (2.386) 2.447 (2.391)
2.474 (2.415) 2.439 (2.387) 2.450 (2.392)
2.446 (2.398) 2.453 (2.398) 2.442 (2.390)
2.458 (2.400) 2.459 (2.405) 2.457 (2.398)
1.892 (1.864) 1.897 (1.867) 1.897 (1.867)
1.891 (1.863) 1.895 (1.866) 1.896 (1.866)
1.895 (1.867) 1.887 (1.859) 1.897 (1.868)
1.895 (1.866) 1.891 (1.864) 1.899 (1.869)
126.8 (127.4) 128.5 (129.0) 126.9 (129.0)
Bond angles (°)
c
ƒ
128.7(129.4)
93.2 (93.7)
93.8 (93.7)
84.3 (84.7)
84.8 (84.1)
83.1 (84.0)
82.5 (83.8)
127.6(128.1)
93.7 (94.0)
92.9 (93.2)
83.7 (84.8)
83.0 (84.7)
84.6 (83.0)
84.5 (82.8)
128.1(129.4)
93.3 (93.3)
93.55 (93.6)
83.6 (85.6)
83.6 (85.6)
83.8 (82.6)
83.5 (82.3)
C(5)–Rh(1)–C(6)
C(7)–Rh(2)–C(8)
S(3)–Rh(1)–S(4)
S(3)–Rh(2)–S(4)
Rh(1)–S(3)–Rh(2)
Rh(1)–S(4)–Rh(2)
1965 cm−1
,
respectively. Trans-[Rh2(m-SS)(CO)2-
(PPh3)2] complexes with symmetrical dithiolate bridges
have two very close bands in the carbonylic area with
relative intensities 4.7:10, which sometimes can be seen
as a single broad band [26,28]. Thus, the IR spectra of
7 and 8 suggest that these compounds are present in the
solid state only as the trans isomers. In [Rh2(m-
SS)(CO)2(PPh3)2] complexes there is often only one
isomer in the solid state (cis or trans) and two isomers
in solution [26,28,29].
a
,
Distances in A and angles in °.
b q: torsion angle.
c
ƒ: dihedral angle between the two mean planes containing the
four coordinated atoms (C,C,S,S).
1
In the H-NMR spectra of 7 and 8 there are two
groups of signals for the dithiolate protons. In 7 the
chemical shifts are very similar whereas in 8 they are
significantly different. Thus, the anomeric proton ap-
pears at 6.29 ppm for the major isomers and at 4.89
ppm for the minor isomers of 8. Molecular mechanics
calculations were carried out for complex 8. When
triphenylphosphine enters the L4 position (Fig. 3), due
to steric hindrance, the puckered conformation of the
coordinated ligand XySS2− (in which the ring oxygen
is above one of the rhodium atoms) is not possible.
This reduces the number of possible structures for 8
from eight to six. The relative strain energies for the six
possible isomers of 8 present in solution are listed in
Table 3. These results show that the strain energy of the
isomers depends on the conformation of the bridging
ligand as observed for the other XySS2− compounds 2
and 6. Again, the most stable isomers are those with the
ring oxygen above one of the rhodium atoms (confor-
mation a). Only when PPh3 enters the L4 position is this
conformation no longer possible and the conformation
of the ligand becomes more extended (b). Thus, the
four isomers in the 31P-NMR spectrum are the first four
in Table 3 and the two groups of isomers seen in
1H-NMR correspond to the two possible geometries of
the ligand in the four possible isomers.
level. The frequencies of the experimental and calcu-
lated stretching CO bands are listed in Table 2. The
calculated values are different from the experimental
values by 0.8–2.4% and these differences are similar to
those found for other carbonylic compounds [27].
[Rh2(S-(CH2)3S)(CO)4] is a similar compound with a
symmetric dithiolate for which the presence and intensi-
ties of three carbonyl bands in the IR suggests a
dinuclear structure [19a]. The same calculation was
carried out for this compound (Table 2). The differ-
ences between experimental and calculated frequen-
cies were very similar to those found with
[Rh2(XySS)(CO)4]. In summary, the similarity of the
experimental and theoretical IR frequencies is further
proof of the dinuclearity of these complexes.
2.2.5. Rh(I) mixed carbonyl–phosphine complexes
The complexes [Rh2(m-RiSS)(CO)4] (5) and [Rh2(m-
XySS)(CO)4] (6) react at room temperature with two
equivalents of PPh3 to form the orange species [Rh2(m-
RiSS)(CO)2(PPh3)2] (7) and [Rh2(m-XySS)(CO)2(PPh3)2]
(8), which are air stable in the solid state (Scheme 2).
These mixed dithiolate bridged complexes are known to
be di- or tetranuclear depending on the dithiolate
bridge [17c,19b]. The fact that the heaviest ion was at

130
O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
The two conformations account for the considerable
for preparing of enantiomerically pure aldehydes,
which are very important not only as precursors for
biologically active compounds but also for new mate-
rials such as biodegradable polymers and liquid crys-
tals.
Table 4 shows the results of the hydroformylation
of styrene with the catalyst precursors 1 and 2 in the
absence of phosphorous ligands.
The catalyst precursor [Rh2(m-XySS)(cod)2] (2) is
not active at 20 bar and 75°C and is only slightly
active at 30 bar (entries 1,2). Although the catalytic
activity of 2 increases with time, its regioselectivity
decreases, until it disappears (entries 1,2). The depen-
dence of activity and regioselectivity on time indicates
a change in the active species.
differences in chemical shifts between the two groups
1
of signals for the dithiolate protons in H-NMR. The
major isomers (with the lowest strain energy) are the
two isomers with the bent conformation of the ligand
(conformation a), that is to say, in which the
anomeric proton is affected by the anisotropic cur-
rents of the carbonyl ligand (H1: 6.29 ppm), and the
minor isomers are the two in which the ligand has
conformation b and PPh3 in L4. (H1: 4.89 ppm).
2.3. Hydroformylation of styrene
Asymmetric hydroformylation has attracted much
attention because of its potential as a synthetic tool
Fig. 2. Experimental IR spectrum in the carbonylic region of [Rh2(RiSS)(CO)4] (5) (right) and [Rh2(XySS)(CO)4] (6) (left) with its Lorentzian
adjustment.

O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
131
Table 2
at 10 and 20 bar (entries 5–6). The regioselectivities
are again the same as the ones obtained with
[Rh(acac)(CO)2] (1) in the same conditions (entries 8
and 9). These results and the practical absence of
chiral induction supports the hypothesis formulated
for the catalyst precursor 2.
If the results obtained for catalyst precursors 1 and
2 are compared, it can be observed that 1 is more
active, even at lower pressures, than 2. According to
our hypothesis, this means that, under hydroformyla-
tion conditions, 2 evolves more slowly than 1 to
[RhH(CO)3], the major active species for this reaction.
The same reaction was carried out in the presence
of triphenylphosphine and the conclusions were not
very different (Table 5). The progressive formation of
Wavenumber (cm−1
) of the stretching carbonyl vibrations for
[Rh2(m–XySS)(CO)4] (6) and [Rh2(m-(S(CH2)3S)(CO)4]
v1
v2
v3
v4
[Rh2(m-XySS)(CO)4]
Experimental
Calculated
2084
2102
18
2062
2089
27
2012
2041
29
1989
2037
48
Difference
[Rh2(m-(S(CH2)3S)(CO)4]
Experimental
Calculated
Difference
2086
2104
21
2065
2085
24
2017
2042
30
It is generally accepted that the catalyst precursor
[Rh(acac)(CO)2] evolves rapidily under hydroformyla-
tion conditions to the active species [RhH(CO)3] (not
regioselective in these conditions). The results in en-
tries 1 and 2 suggest a similar, though slower, evolu-
tion for the catalyst precursor 2. In order to confirm
this hypothesis, styrene was hydroformylated in the
same conditions but with [Rh(acac)(CO)2] (9) as the
precatalyst (entry 7). It is worth noting that the re-
gioselectivity obtained in this experiment coincides
with the value to which the results of entries 1 and 2
evolve. This indicates that complex 2 has little cata-
lytic activity and that it transforms slowly and gradu-
ally to [RhH(CO)3]. The fact that there are practically
no enantiomeric excesses (less than 4%) further sup-
ports this hypothesis.
the mononuclear active species [RhH(CO)x(PPh3)3−x
]
under these conditions, as also described for
[Rh(acac)(CO)2]/PPh3 system [30], is responsible for
the catalytic activity observed. In this case the rup-
ture of the dinuclear compound was faster than in
the absence of triphenylphosphine.
These results agree with very recent high-pressure
spectroscopic studies, carried out on rhodium com-
plexes containing bridge dithiolate ligands and
triphenylphosphine under hydroformylation conditions
[31].
3. Conclusions
The epimer dithiolate ligands RiSS2− and XySS2−
react with [M(m-OMe)(cod)]2 (M=Rh, Ir) to form
the dinuclear dithiolate bridged complexes [M2(m-
RiSS)(cod)2] and [M2(m-XySS)(cod)2]. The rhodium
complexes react with carbon monoxide to afford the
corresponding [Rh2(m-RiSS)(CO)4] and [Rh2(m-XySS)-
(CO)4] complexes which in the presence of tryphenyl-
phosphine evolve to [Rh2(m-RiSS)(CO)2(PPh3)2] and
[Rh2(m-XySS)(CO)2(PPh3)2]. All the complexes with
XySS2− as bridging dithiolate can be present as two
conformers. In the absence of steric hindrance, the
most stable conformer is the one in which the oxygen
of the furanose ring is above one of the rhodium
atoms. DFT theoretical studies show the existence of
a bond interaction between these two atoms which
stabilizes the structure. Computed CO stretching fre-
quencies for [Rh2(m-XySS)(CO)4] agree with the exper-
imental values and confirm the dinuclearity of these
complexes.
The catalyst precursor [Rh2(m-RiSS)(cod)2] (1) is
more active at 30 bar and 75°C in the hydroformyla-
tion of styrene (entries 3 and 4). The regioselectivity
again coincides with the result obtained with
[Rh(acac)(CO)2] (entry 7). This system is also active
Between 5 and 30 bar syn gas, [Rh2(m-RiSS)(cod)2]
and [Rh2(m-XySS)(cod)2] undergo a progressive rup-
ture to mononuclear complexes that do not contain
the dithiolate ligand and are the major active species
in the hydroformylation of styrene.
Fig. 3. Possible geometric isomers for the stoichiometry [Rh2(m-
SS)(CO)2(PPh3)2] where SS is a bridging asymmetric dithiolate ligand.

132
O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
Table 3
Strain energies for the possible isomers of the complex [Rh2(m-XySS)(CO)2(PPh3)2] (8)
Isomer
L4
Relative conformation of the furanose ring
Strain energy (kcal mol−1
)
cis
trans
cis
trans
cis
trans
CO
CO
PPh3
PPh3
CO
a
a
b
b
b
b
130.6
131.6
140.0
141.2
138.4
139.6
CO
4. Experimental
4.2. Computational details
4.1. General comments
Density functional calculations were made with the
ADF package [37]. For the correlation, the spin local
density approximation for the electron gas exchange
(Xh, h= 2/3) was used with the Vosko–Wilk–Nusair
parametrization [38]. Non-local corrections were applied
self consistently, taking the gradient-corrected Becke’s
functional [39] for the exchange and Perdew’s functional
[40] for correlation, i.e. the so-called BP86 functional. A
double-n+polarization Slater basis set was used to
describe the valence electrons of C, O and S. For the Rh
atoms, a frozen core composed of the 1s to 4sp shells was
described with simple Slater functions. 5s and 4d elec-
trons were described by triple-n Slater functions and 5p
by a single orbital. The H atoms were described by a
double-n Slater basis set. Quasi relativistic corrections
were made using the Pauli formalism with corrected core
potentials. The quasi relativistic frozen core shells were
generated with the auxiliary program DIRAC [36].
The molecular mechanics calculations were carried
out with the program CERIUS2 [41] developed by
Molecular Simulations (MSI) and the force field UFF
developed by Rappe et al. [42]. The electrostatic inter-
actions were taken into account from atomic changes
generated by the Qeq method [43].
All syntheses were performed by standard Schlenk
techniques under a nitrogen atmosphere. The complexes
–
–
[Rh(m OMe)(cod)]2 [32], [Ir(m OMe)(cod)]2 [32b,33]
[Rh(acac)(CO)2] [34], and the compounds 1,2-O-iso-
propylidene-a-
D-xylofuranose [35] and 1,2-O-isopropyli-
dene-a- -ribofuranose [35,36] were prepared by
D
previously described methods. Solvents were purified by
standard procedures. All other reagents were used as
commercially available.
Elemental analyses were performed on a Carlo Erba
EA-1108 instrument. Infrared spectra (KBr pellet or
solution) were obtained on a FT-IR MIDAC Prospect
PRS instrument. IR Lorentz line shape analyses were
1
performed with the Microcal Origin 4.1 Program. H-,
13C{1H}- and 31P{1H}-NMR spectra were recorded on a
Variant Gemini 300 MHz Spectrometer. Chemical shifts
are relative to SiMe4 (1H and 13C) as internal standard
or H3PO4 (31P) as external standard. All assignments in
NMR spectra were determined by means of COSY and
HETCOR spectra. EI Mass spectra were obtained on an
HP 5989 A spectrometer. A VG-Autospect was used for
FAB mass spectral analyses. The matrix was m-nitroben-
zylalcohol. Gas chromatographic analyses were run on
a Hewlett–Packard 5890A instrument (split/splitless in-
jector, J&W Scientific, Ultra-2 25 m column, internal
diameter 0.2 mm, film thickness 0.33 mm, carrier gas: 150
kPa He, F.I.D. detector) equipped with a Hewlett–Pack-
ard 3396 series II integrator. Enantiomeric excesses were
measured after oxidation of the aldehydes to the corre-
sponding carboxylic acids on a Hewlett–Packard 5890A
gas chromatograph (split/splitless injector, J&W Scien-
tific, FS-Cyclodex b-I/P 50 m column, internal diameter
0.2 mm, film thickness 0.33 mm, carrier gas: 100 kPa He,
F.I.D. detector). Absolute configuration was determined
by comparing the retention times with optically pure
(S)-(+)-2-phenylpropionic and R-(−)-2-phenylpropi-
onic acids. Optical rotations were measured at 25°C on
a Perkin–Elmer 241 MC polarimeter. The specific rota-
tions are given in deg cm3 g−1 dm−1 units. Hydroformy-
lation reactions were carried out in a home-made 100 cm3
stainless steel autoclave.
Table 4
Hydroformylation of styrene with [Rh2(m-SS)(cod)2], SS=RiSS (1) or
XySS (2) as precatalyst a
Entry
Precursor
P (bar)
t (h)
%Conv.
%2-PP
1
2
3
4
5
6
7
8
9
2
2
1
1
1
1
9
9
9
30
30
30
30
20
10
30
20
10
24
48
3
6
9
20
2
2.5
6
19
82
47
99
85
57
80
58
45
66
55
51
51
45
41
55
46
43
a Conditions: Substrate: Rh=300. Solvent: tetrahydrofuran (7.5
cm3). Precursor (0.05 mmol). T=75°C. CO: H2=1.% Conv.=con-
version in aldehydes (chemoselectivity\99%). 2-PP=2-phenyl-
propanal.%2-PP=100×(2-PP)/(2-PP+3-PP). 3-PP=3-phenylpro-
panal. 9: [Rh(acac)(CO)2].

O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
133
Table 5
4.73 (1H, dd, H-2, J1–2=3.5 Hz, J2–3=4.1 Hz), 4.78
(1H, dd, H-5, J5–5%=11.9 Hz, J5–4=2.2 Hz), 4.86 (1H,
dd, H-3, J3–2=4.1 Hz, J3–4=8.7 Hz) and 5.80 (1H, d,
H-1, J1–2=3.5 Hz); lC(CDCl3) 26.5 (CH3), 71.2 (C-5),
74.3 (C-4), 77.0 (C-2), 80.0 (C-3), 104.0 (C-1), 114.9
(CMe2), 118.4 (q, CF3, JC–F=320 Hz) and 118.5 (q,
CF3, JC–F=318 Hz).
Hydroformylation of styrene with [Rh2(m-SS)(cod)2], SS=RiSS (1) or
XySS (2) as precatalyst in the presence of PPh3
a
Entry
Precursor
P (bar)
t (h)
%Conv.
%2-PP
1
2
3
4
5
6
2
2
1
1
9
9
5
30
5
30
5
30
20
2
7
1.5
1
0.6
13
100
76
100
98
96
63
89
61
89
61
89
4.4. Synthesis of 3,5-di-S-acetyl-1,2-O-isopropylidene-
3,5-dithio-h-
1,2-O-isopropylidene-3,5-dithio-h-
D
-ribofuranose (V) and 3,5-di-S-acetyl-
-xylofuranose (VI)
a Conditions: Substrate: Rh=200. PPh3:Rh=1. Solvent: tetrahy-
drofuran (7.5 cm3). Precursor (0.05 mmol). T=75°C. CO:H2=1. %
Conv.=conversion in aldehydes (chemoselectivity\99%). 2-PP=2-
D
Potassium thioacetate (6.5 g, 59.9 mmol) was added
to a acetonitrile solution (35 cm3) of compound III or
IV (2.0 g, 4.4 mmol) and the mixture was refluxed for
90 min. The solution was evaporated to dryness and the
residue was extracted with CH2Cl2 (3×20 cm3). The
product, obtained after evaporating the dried solution
(MgSO4), was purified by column chromatography (5:1
hexane–ethyl acetate) to give a yellow solid.
phenylpropanal.
phenylpropanal. 9: [Rh(acac)(CO)2].
%2-PP=100×(2-PP)/(2-PP+3-PP).
3-PP=3-
Finally, the topological properties of the charge den-
sity were computed with a modified version of the
AIMPAC package [44] and the Xaim software [45].
4.3. Synthesis of 1,2-O-isopropylidene-3,5-di-O-
trifluoromethansulfonyl-h-
1,2-O-isopropylidene-3,5-di-O-trifluoromethan-
sulfonyl-h- -ribofuranose (IV)
D-xylofuranose (III) and
D
4.4.1. Compound V
Yellow–orange product (310 mg, 23%), [h]2D5 (c 1.25
in CHCl3)= +55.2 (Found: C, 47.32; H, 6.05; S, 21.36.
Calc. for C12H18O5S2: C, 47.52; H, 5.94; S, 21.12%);
lH(CDCl3) 1.32 (3H, s, CH3), 1.50 (3H, s, CH3), 2.35
(3H, s, CH3–CO), 2.39 (3H, s, CH3–CO), 3.10 (1H, dd,
H–5%, J5%–5=14.3 Hz, J5%–4=6.3 Hz), 3.32 (1H, dd,
H-5, J5–5%=14.3 Hz, J5–4=4.1 Hz), 3.74 (1H, dd, H-3,
Anhydrous pyridine (0.8 cm3, 10 mmol) was added to
a solution of 1,2-O-isopropylidene-a- -xylofuranose or
1,2-O-isopropylidene-a- -ribofuranose (0.7 g, 3.7
D
D
mmol) in 20 cm3 methylene chloride. After 10 min,
trifluoromethanesulfonic anhydride (1.5 cm3, 8.9 mmol)
was added dropwise at −20°C and the mixture was
allowed to react at room temperature for 45 min after
which the solvent was evaporated. The residue was
purified by column chromatography (10:1 hexane–ethyl
acetate) to give the ditriflate as a white solid.
J
J
3–4=10.3 Hz, J3–2=4.1 Hz), 4.12 (1H, ddd, H-4,
4–3=10.3 Hz, J4–5%=6.3 Hz, J4–5=4.1 Hz), 4.70 (1H,
dd, H-2, J2–1=3.9 Hz, J2–3=4.1 Hz) and 5.84 (1H, d,
H-1, J1–2=3.9 Hz); lC(CDCl3) 26.2 (CH3), 26.5 (CH3),
29.9 (C-5), 33.0 (C6 H3–CO), 47.5 (C-3), 78.7 (C-4), 80.9
(C-2), 104.4 (C-1), 112.3 (CMe2), 194.0 (CO) and 194.7
4.3.1. Compound III
(0.9 g, 54%), [h]D25 (c 1.25 in CHCl3)= −19.2
(Found: C, 26.77; H, 2.68; S, 13.98. Calc. for
C10H12O9S2F6: C, 26.44; H, 2.66; S, 14.12%);
lH(CDCl3) 1.35 (3H, s, CH3), 1.56 (3H, s, CH3), 4.68
(3H, m, H-5, H-5%, H-4), 4.80 (1H, d, H-2, J2–1=3.9
Hz), 5.26 (1H, d, H-3, J3−4=2.3 Hz) and 6.06 (1H, d,
H-1, J1–2=3.9 Hz); lC(CDCl3) 26.1 (CH3), 26.4 (CH3),
70.5 (C-5), 75.6 (C-4), 83.0 (C-2), 86.8 (C-3), 104.8
(CO); m/z 291 [M–CH3], 263 [M–COCH3].
4.4.2. Compound VI
Pale yellow product (935 mg, 70%), [h]2D5 (c 1.25 in
CHCl3)= −34 (Found: C, 26.97; H, 2.96; S, 13.82.
Calc. for C10H12O9S2F6: C, 26.44; H, 2.66; S, 14.12%);
lH(CDCl3) 1.30 (3H, s,CH3), 1.52 (3H, s, CH3),
2.38(3H, s, CH3–CO), 2.41 (3H, s, CH3–CO), 3.04
(1H, dd, H-5%, J5%–5=13.7 Hz, J5%–4=6.4 Hz), 3.21 (1H,
dd, H-5, J5–5% =13.7 Hz, J5–4=7.7 Hz), 4.13 (1H, d,
H-3, J3–4=3.7 Hz), 4.48 (1H, ddd, H-4, J4–5%=6.4 Hz,
(C-1), 113.8 (CMe2), 118.3 (q, CF3,
JC−F=305
Hz) and 118.4 (q, CF3, JC–F=304 Hz); m/z: 439
[M−CH3].
4.3.2. Compound IV
J
4–5=7.7 Hz, J4–3=3.7 Hz), 4.58 (1H, d, H-2, J2–1
=
(0.9 g 54%), [h]2D5 (c 1.25 in CHCl3)= +54 (Found:
C, 26.97; H, 2.96; S, 13.82. Calculated for
C10H12O9S2F6: C, 26.44; H, 2.66; S, 14.12%);
lH(CDCl3) 1.40 (3H, s, CH3), 1.59 (3H, s, CH3), 4.37
(1H, dt, H-4, J4–5=2.2 Hz, J4–5%=2.2 Hz, J4–3=8.7
Hz), 4.51 (1H, dd, H-5%, J5%–5=11.9 Hz, J5%–4=2.2 Hz),
3.7 Hz) and 5.80 (1H, d, H-1, 1–2=3.7 Hz);
J
lC(CDCl3) 26.2 (CH3), 26.5 (CH3), 29.1 (C-5), 30.5
(CH3–CO), 31.0 (CH3–CO), 51.0 (C-3), 77.0 (C-4),
86.3 (C-2), 104.5 (C-1), 112.1 (CMe2), 192.9 (CO) and
194.6 (CO); m/z 263 [M−COCH3], 220 [M−
2COCH3].

134
O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
4.5. Synthesis of 1,2-O-isopropylidene-3,5-dithiol-h-
ribofuranose (RiSSH2) and 1,2-O-isopropylidene-
3,5-dithiol-h-D-xylofuranose (XySSH2)
D
-
dd, H-2, J2–1=4.1 Hz, J2–3=3.9 Hz), 4.52 (3H, m, CH
COD, H-3), 4.76 (2H, m, CH COD) and 5.55 (1H, d,
H-1, J1–2=4.1 Hz); lC(CD2Cl2) 26.5 (CH3), 26.6
(CH3), 31.6 (CH2 COD), 31.7 (CH2 COD), 31.9 (CH2
COD), 32.0 (CH2 COD), 32.1 (CH2 COD), 32.2 (CH2
COD), 34.6 (C-5), 51.8 (C-3), 80.5 (d, CH COD,
Compound V or VI (0.05 g, 0.16 mmol) was added to
a suspension of 15.0 mg of the ion-exchange resin IRA
420 (Fluka) in 5 cm3 methanol and the mixture was
allowed to react for 24 h. After filtration and evapora-
tion of the solvent the product was obtained as a white
solid.
J
C–Rh=10.7 Hz), 81.8 (d, CH COD, JC–Rh=10.3 Hz),
81.9 (d, CH COD, JC–Rh=11.8 Hz), 82.1 (d, CH COD,
C–Rh=10.6 Hz), 82.2 (d, CH COD, JC–Rh=12.5 Hz),
J
82.4 (C-2), 82.6 (d, CH COD, JC–Rh=10.6 Hz), 83.7
(d, CH COD, JC–Rh=10.3 Hz), 83.9 (d, CH COD,
4.5.1. RiSSH2
JC–Rh=11.2 Hz), 87.0 (C-4), 104.6 (C-1) and 111.9
(29.5 mg, 80%), [h]2D5 (c 1.25, CHCl3)= +100.8
(Found: C, 42.92; H, 6.35; S, 28.15. Calc. for
C8H14O3S2: C, 43.23; H, 6.35; S, 28.85%); lH(CDCl3)
1.36 (3H, s, CH3), 1.52 (3H, s, CH3), 1.60 (1H, dd,
HS-C-5, JSH–5%=11.7 Hz, JSH–5=4.7 Hz), 1.79 (1H, d,
HS–C-3, JSH–3=13.1 Hz), 2.74 (1H, ddd, H-+5%,
(CMe2); m/z 642 [M+].
4.6.2. [Rh2(v-XySS)(cod)2] (2)
(36 mg, 60%) (Found: C, 44.74; H, 5.72; S, 9.87.
Calc. for Rh2C24H36O3S2: C, 44.87; H, 5.65; S, 9.98%);
lH(CD2Cl2) 1.27 (3H, s, CH3), 1.39 (3H, s, CH3), 1.88
(4H, m, CH2 COD), 2.11 (4H, m, CH2 COD), 2.29 (4H,
m, CH2 COD), 2.50 (4H, m, CH2 COD), 2.62 (1H, dd,
H-5%, J5%–5=12.7 Hz, J5%–4=3.4 Hz), 2.86 (2H, m, H-5,
H-3), 3.91 (1H, m, CH COD), 3.97 (1H, m, CH COD),
J5%–5=14.0 Hz, J5%–SH=11.7 Hz, J5%–4=5.1 Hz), 3.07
(2H, m, H–5, H-3), 3.94 (1H, m, H-4), 4.63 (1H, dd,
H-2, J2–3=4.0 Hz, J2–1=3.9 Hz) and 5.82 (1H, d, H-1,
J1–2=3.9 Hz);.lC(CDCl3) 25.1 (C-5), 26.4 (CH3), 26.5
(CH3), 42.3 (C-3), 81.3 (C-2), 82.8 (C-4), 104.0 (C-1)
4.15(3H, m, CH COD, H-4), 4.38 (1H, d, H-2, J2–1=
and 112.0 (CMe2); m/z 220 [M+ –2H], 189 [M –SH].
+
4.2 Hz), 4.42 (2H, m, CH COD), 4.58 (2H, m, CH
COD) and 6.34 (1H, d, H-1, J1–2=4.2 Hz); lC(CD2Cl2)
26.3 (CH3), 26.8 (CH3), 30.2 (C-5), 31.3 (CH2 COD),
’
’
4.5.2. XySSH2
(35.0 mg, 95%), [h]2D5 (c 1.25 in CHCl3)= +289
(Found: C, 43.92; H, 5.71; S, 27.23. Calc. for
C8H14O3S2: C, 43.23; H, 6.35; S, 28.85%); lH(CDCl3)
1.35 (3H, s, CH3), 1.54 (3H, s, CH3), 2.99 (1H, dd,
H-5%, J5’–5=12.7 Hz, J5%–4=3.0 Hz), 3.41 (1H, d, H-5,
J5–5%=12.7 Hz), 4.11 (1H, d, H-3, J3–4=5.4 Hz), 4.76
(1H, d, H-2, J2–1=3.9 Hz), 5.48 (1H, m, H-4) and 5.93
(1H, d, H-1, J1–2=3.9 Hz); lC(CDCl3) 26.9 (CH3), 27.5
(CH3), 49.3 (C-5), 61.1 (C-3), 88.0 (C-2), 88.2 (C-4),
106.5 (C-1) and 112.7 (CMe2); m/z 220 [M−2H], 189
[M–SH].
31.5 (CH2 COD), 31.9 (CH2 COD), 32.0 (CH2 COD),
6
32.6 (CH2 COD), 34.0 (CH2 COD), 49.7 (C-3), 77.8
(C-2), 79.7 (d, CH COD, JC–Rh=9.4 Hz), 80.2 (d, CH
COD, JC–Rh=9.4 Hz), 80.9 (d, CH COD, JC–Rh=9.5
Hz), 81.6 (d, CH COD, JC–Rh=10.0 Hz), 81.9 (d, CH
COD, JC–Rh=9.5 Hz), 82.4 (d, CH COD, JC–Rh=9.5
Hz), 83.1 (d, CH COD, JC–Rh=9.5 Hz), 90.0 (C-4),
105.1 (C-1) and 111.7 (CMe2); m/z 642 [M+].
4.7. Synthesis of [Ir2(v-RiSS)(cod)2] (3) and
[Ir2(v-XySS)(cod)2] (4)
4.6. Synthesis of [Rh2(v-RiSS)(cod)2] (1) and
[Rh2(v-XySS)(cod)2] (2)
Bisthioacetate V or VI (0.07 mmol, 21.0 mg) and
sodium methoxide (0.07 mmol, 4.0 mg) were added to a
solution of [Ir(m-OMe)(cod)]2 (0.06 mmol, 40.0 mg) in 2
cm3 of methylene chloride and the mixture was allowed
to react for 2 h. The red solution was filtered over celite
in order to remove the sodium methoxide, and the
product was obtained after precipitation with cold
petroleum ether.
Bisthioacetate V or VI (0.1 mmol, 30.8 mg) and
sodium methoxide (0.1 mmol, 5.5 mg) were added to a
solution of [Rh(m-OMe)(cod)]2 (0.1 mmol, 44.4 mg) in 2
cm3 of methylene chloride and the mixture was allowed
to react for 6 h. After precipitation with methanol, the
red product was filtered, washed with cold methanol
and vacuum dried.
4.7.1. [Ir2(v-RiSS)(cod)2] (3)
(26 mg, 53% yield) (Found: C, 33.34; H, 4.22; S, 7.13.
Calc. for Ir2C24H36O3S2+CH2Cl2: C, 33.14; H, 4.23; S,
7.07%); m/z 819 [M+ −2H].
4.6.1. [Rh2(v-RiSS)(cod)2] (1)
(39 mg, 65%) (Found: C, 45.15; H, 5.65; S, 9.98.
Calc. for Rh2C24H36O3S2: C, 44.87; H, 5.65; S, 9.98%);
lH(CD2Cl2) 1.30 (3H, s, CH3), 1.66 (3H, s, CH3), 1.95
(4H, m, CH2 COD), 2.12 (4H, m, CH2 COD), 2.33 (4H,
m, CH2 COD), 2.49 (5H, m, CH2 COD, H-5’), 2.64
(1H, m, H-5), 4.16 (5H, m, CH COD, H-4), 4.39 (1H,
4.7.2. [Ir2(v-XySS)(cod)2] (4)
(10 mg, 21% yield) (Found: C, 34.66; H, 4.45; S, 6.95.
Calc. for Ir2C24H36O3S2: C, 35.11; H, 4.42; S, 7.81%);
m/z 820 [M+ −H].

O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
135
4.8. Synthesis of [Rh2(v-RiSS)(CO)4] (5) and
[Rh2(v-XySS)(CO)4] (6)
i H-4), 5.81 (1H, d, H-1, J1–2=3.8 Hz) and 7.2–8.0 (CH,
Ph); lP(CD2Cl2) 39.4 (1P, d, JP–Rh=11.9 Hz), 40.1 (1P,
d, JP–Rh=10.7 Hz), 40.7 (1P, d, JP–Rh=12.3 Hz) and
41.2 (1P, m).
Carbon monoxide was bubbled for 15 min through a
solution of [Rh2(m-RiSS)(cod)2] or [Rh2(m-XySS)(cod)2]
(40 mg, 0.06 mmol) in 5 cm3 deuterated dichloro-
methane. The red solution turned green.
4.9.3. Major isomers
lH(CD2Cl2) 1.49 (3H, s, CH3), 1.78 (3H, s, CH3), 2.50
(2H, m, H-5 i H-5%), 2.72 (1H, m, H-3), 4.39 (2H, m, H-2
i H-4), 5.79 (1H, d, H-1, J1–2=3.8 Hz) and 7.2–8.0 and
(CH, Ph) lP(CD2Cl2) 39.9 (1P, d, JP–Rh=14.9 Hz), 41.2
(2P, m) and 42.6 (1P, d, JP–Rh=14.8 Hz).
4.8.1. [Rh2(v-RiSS)(CO)4] (5)
wmax/cm−1 (CO) 2086, 2065, 2017 and 1990 (CH2Cl2);
lH(CD2Cl2) 1.32 (3H, s, CH3), 1.62 (3H, s, CH3), 2.36
(8H, m, CH2 free COD), 2.44 (1H, dd, H-5%, J5%–4=10.7
Hz, J5%–5=10.1 Hz), 2.89 (1H, dd, H-5, J5–5%=10.1 Hz,
–
4.9.4. [Rh2(v XySS)(CO)2(PPh3)2] (8)
J
5–4=4.4 Hz), 3.00 (1H, dd, H-3, J3–4=11.8, J3–2=3.3
Hz), 4.09 (1H, m, H-4), 4.58 (1H, dd, H-2, J2–1=3.6 Hz,
2–3=3.3 Hz), 5.54 (4H, m, CH free COD) and 5.66 (1H,
(35 mg, 57%) (Found: C, 55.38; H, 4.32; S, 5.98. Calc.
for Rh2C46H42O5S2P2: C, 54.88; H, 4.20; S, 6.37%);
wmax/cm−1 (CO) 2020, 1991 and 1966 (CH2Cl2); 1965
(KBr); m/z 1006 [M+].
J
d, H-1, J1–2=3.6 Hz); lC(CD2Cl2) 25.3 (CH3), 25.6
(CH3), 27.3 (CH2 free COD), 33.5 (C-5), 50.4 (C-3), 80.2
(C-2), 85.0 (C-4), 104.1 (C-1), 111.6 (CMe2), 127.9 (CH
free COD) and 183.6 (d, CO, JRh–CO=72 Hz).
4.9.5. Minor isomers
lH(CD2Cl2)1.05 (3H, s, CH3), 1.28 (3H, s, CH3), 2.61
(2H, m, H-5 i H-5%), 3.18 (1H, m, H-3), 3.55 (1H, d, H-2,
4.8.2. [Rh2(v-XySS)(CO)4] (6)
wmax/cm−1 (CO) 2084, 2062, 2013 and 1989 (CH2Cl2);
lH(CD2Cl2) 1.29 (3H, s, CH3), 1.43 (3H, s, CH3), 2.34
(8H, m, CH2 free COD), 3.00 (1H, dd, H-5%, J5%–4=2.9
Hz, J5%–5=14.6 Hz), 3.15 (1H, dd, H-5, J5–5%=14.6 Hz,
J
J
2–1=3.3 Hz), 4.30 (1H, m, H-4), 4.89 (1H, d, H-1,
1–2=3.3 Hz) and 7.2–8.0 (CH, Ph); lP(CD2Cl2) 39.8 (d,
1P, JP–Rh=15.6 Hz), 40.3 (d, 1P, JP–Rh=13.5 Hz), 41.1
(d, 1P, JP–Rh=11.9 Hz) and 41.6 (d, 1P, JP–Rh=13.9
Hz).
J5–4=2.2 Hz), 3.30 (1H, d, H-3, J3–4=2.9 Hz), 4.43
(1H, td, H-4, J4–3=J4–5%=2.9 Hz, J4–5=2.2 Hz), 4.60
(1H, d, H-2, J2–1=3.7 Hz), 5.54 (4H, m, CH free COD)
and 6.16 (1H, d, H-1, J1–2=3.7 Hz); lC(CD2Cl2) 26.2
(CH3), 26.3 (CH3), 28.3 (CH2 free COD), 31.0 (C-5), 49.7
(C-3), 76.1 (C-2), 88.8 (C-4), 104.5 (C-1), 111.9 (CMe2),
128.8 (CH free COD) and 182.6 (d, CO, JRh–CO=70
Hz).
4.9.6. Major isomers
lH(CD2Cl2) 1.21 (3H, s, CH3), 1.43 (3H, s, CH3), 2.89
(2H, m, H-5 i H-5%), 3.18 (1H, m, H-3), 4.45 (1H, d, H-2,
J
J
2–1=3.6 Hz), 4.53 (1H, m, H-4), 6.29 (1H, d, H-1,
1–2=3.6 Hz) and 7.2–8.0 (CH, Ph); lP(CD2Cl2) 39.6 (d,
1P, JP–Rh=14.2 Hz), 40.8 (d, 1P, JP–Rh=11.9 Hz), 42.0
(d, 1P, JP–Rh=12.3 Hz) and 42.3 (d, 1P, JP–Rh=13.9
Hz).
4.9. Synthesis of [Rh2(v-RiSS)(CO)2(PPh3)2] (7) and
–
[Rh2(v XySS)(CO)2(PPh3)2] (8)
Carbon monoxide was bubbled through a solution of
[Rh2(m-RiSS)(cod)2] or [Rh2(m-XySS)(cod)2] (40 mg, 0.06
mmol) in the minimum amount of dichloromethane. The
red solution turned green. After 15 min, triphenylphos-
phine (38.9 mg, 0.12 mmol) was added and the green
solution turned orange. The mixture was allowed to react
for 20 min at room temperature. An orange product was
obtained by precipitation with hexane.
4.10. Hydroformylation of styrene
The autoclave was purged three times with CO. In
the case of hydroformylations in the absence of
triphenylphosphine, a THF solution (7.5 cm3) of [Rh2(m-
RiSS)(cod)2] or [Rh2(m-XySS)(cod)2] (0.025 mmol) and
styrene (15 mmol) was introduced. When the hydro-
formylations were run in the presence of triphenylphos-
phine, the solution was formed from [Rh2(m-RiSS)(cod)2]
or [Rh2(m-XySS)(cod)2] (0.025 mmol), triphenylphos-
phine (0.125 mmol) and styrene (10 mmol) in THF (7.5
cm3). After pressurizing to the desired pressure with syn
gas and heating the autoclave to the reaction tempera-
ture, the reaction mixture was stirred. During the reac-
tion several samples were taken from the autoclave. After
the desired reaction time, the autoclave was cooled to
room temperature and depressurized. The reaction mix-
ture was analysed by gas chromatography.
4.9.1. [Rh2(v-RiSS)(CO)2(PPh3)2] (7)
(32 mg, 52%) (Found: C, 55.38; H,4.32; S, 5.98. Calc.
for Rh2C46H42O5S2P2: C, 54.88; H, 4.20; S, 6.37%);
wmax/cm−1 (CO) 2021, 1992 and 1969 cm−1 (CH2Cl2);
1967 (KBr); m/z 1006 [M+].
4.9.2. Minor isomers
lH(CD2Cl2) 1.54 (3H, s, CH3), 1.61 (3H, s, CH3), 3.01
(2H, m, H-5 i H-5%), 3.27 (1H, m, H-3), 4.78 (2H, m, H-2

136
O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
(1993) 1829. (b) J.E. Babiarz, S.D. Pastor, Phosphorus, Sulfur and
In order to determine the enantioselective excess, the
Silicon 91 (1994) 263. (c) G.J.H. Buisman, M.E. Martin, E.J. Voss,
A. Klootwijk, P.C.J. Kamer, P.W.N.M. van Leeuwen, Tetrahe-
dron: Asymmetry 6 (1995) 719. (d) J.-C. Shi, X.-Y. Huang, D.-X.
Wu, Q.-T. Liu, B.-S. Kang, Inorg. Chem. 35 (1996) 2742. (e) M.
Beller, J.G.E. Krauter, A. Zapf, Angew. Chem. Int. Ed. Engl. 36
(1997) 772. (f) M.T. Reetz, S.R. Waldvogel, Angew. Chem. Int.
Ed. Engl. 36 (1997) 865.
aldehydes obtained from the hydroformylation were
oxidized to carboxylic acids. A small portion of the
hydroformylation reaction mixture (2 cm3) was added
to the solution formed with 10 cm3 of a potassium
permanganate 1 M solution and 10 cm3 of a 1.25 M
potassium dihydrogenphosphate solution. After 1 h
with agitation, 5 cm3 of a saturated solution of sodium
sulfite was added and then diluted hydrochloric acid
until the brown precipitate of manganese (IV) oxide
disappeared. The acids formed were extracted with
diethyl ether (3×10 cm3) and the organic phase was
concentrated to dryness. 10 cm3 of a 2 M solution of
sodium hydroxide were added. After washing the solu-
tion with diethyl ether, 10 cm3 of concentrated hy-
drochloric acid were added and the product extracted
with diethyl ether (3×10 cm3). The carboxylic acids
were obtained after washing the etheric phase with
water, drying it with magnesium sulfate and evaporat-
ing it to dryness.
[6] (a) DuPont, A.L. Casalnuovo, T.V. RajanBabu, US Patent, 5 484
902 (16 Jan. 1966). (b) T.V. Rajanbabu, A.L. Casalnuovo, Pure
and Appl. Chem. 66 (1994) 1535. (c) A.L. Casalnuovo, T.V.
RajanBabu, T.A. Ayers, T.H. Warren, J. Am. Chem. Soc. 116
(1994) 9869. (d) T.V. RajanBabu, A.L. Casalnuovo, J. Am. Chem.
Soc. 118 (1996) 6325.
[7] (a) D. Lafont, D. Sinou, G. Descotes, J. Organomet. Chem. 169
(1979) 87. (b) S. Saito, Y. Nakamura, Y. Morita, Chem. Pharm.
Bull. 33 (1985) 5284. (c) R. Selke, H. Pracejus J. Mol. Catal. 37
(1986) 213. (d) R. Selke, J. Organomet. Chem. 370 (1989) 241. (e)
R. Selke, J. Organomet. Chem. 370 (1989) 249. (f) R. Selke, C.
Facklam, H. Foken, D. Heller, Tetrahedron: Asymmetry 4 (1993)
369. (g) T.V. RajanBabu, T.A. Ayers, A.L. Casalnuovo, J. Am.
Chem. Soc. 116 (1994) 4101. (h) A. Kumar, G. Oehme, J.P. Roque,
M. Schwarze, R. Selke, Angew. Chem 106 (1994) 2272; Angew.
Chem. Int. Ed. Engl. 33 (1994) 2197. (i) T.V. RajanBabu, T.A.
Ayers, G.A. Halliday, K.K. You, J.C. Calabrese, J. Org. Chem.
62 (1997) 6012.
[8] A. Iida, M. Yamashita, Bull. Chem. Soc, Jpn. 61 (1988) 2365.
[9] T.V. Rajanbabu, T.A. Ayers, Tetrahedron Lett. 35 (1994) 4295.
[10] W. Vocke, R. Ha¨nel, F. -U. Flo¨ther, Chem. Tech. (Leipzig) 39
(1987) 123.
[11] M. Spescha, Helv. Chim. Acta 76 (1993) 1833.
[12] C. Li, B. Bernet, A. Vasella, E.A. Broger, A. Meili, Carbohydr.
Res. 216 (1991) 149.
[13] (a) P. Barbaro, A. Currao, J. Herrmann, R. Nesper, P.S. Pregosin,
R. Salzmann, Organometallics 15 (1996) 1879. (b) A. Albinati, P.S.
Pregosin, K. Wick, Organometallics 15 (1996) 2419. (c) A.
Albinati, J. Eckert, P.S. Pregosin, H. Ru¨egger, R. Salzmann, C.
Sto¨ssen, Organometallics 16 (1997) 579.
[14] A. Borwitzky, T. Schareina, E. Paetzold, G. Oehme, Phosphorus
Sulfur and Silicon and the related elements 114 (1996) 115.
[15] (a) P. Kalck, J.M. Frances, P.M. Pfister, T.G. Southern, A.
Thorez, J. Chem. Soc. Chem. Commun. (1983) 510. (b) J.C.
Bayo´n, J. Real, C. Claver, A. Polo, A. Ruiz, J. Chem. Soc. Chem.
Commun. (1989) 1056. (c) S. Gladiali, J.C. Bayo´n, C. Claver,
Tetrahedron: Asymmetry 8 (1995) 1453. (d) M. Die´guez, A.
Orejo´n, A.M. Masdeu-Bulto´, R. Echarri, S. Castillo´n, C. Claver,
A. Ruiz, J. Chem. Soc. Dalton Trans. (1997) 4611. (e) M. Die´guez,
A. Ruiz, C. Claver, M.M. Pereira, A.M. d’A. Rocha Gonsalves,
J. Chem. Soc., Dalton Trans. (1988) 3517.
Acknowledgements
We thank the Spanish Ministerio de Educacio´n y
Cultura and the Generalitat de Catalunya (CIRIT) for
financial support (QFN-95-4725-C03-2; PB95-0639-
C02-02; CICYT-CIRIT) and the Generalitat de
Catalunya (CIRIT) for awarding a research grant (to
O.P.). We are also very much indebted to Prof. Sergio
Castillo´n for his very useful discussions and suggestions
on this work and to Ms Cinta Pujol for the IR Lorentz
line shape analyses.
References
[1] (a) J.D. Morrison (Ed.), Asymmetric Synthesis, vol. 5, Academic
Press: NY, 1985. (b) P.A. Chaloner, Handbook of Coordination
Catalysis in Organic Chemistry, Butterworths, London, 1986. (c)
H. Brunner, Topics in Stereochem. 18 (1988) 129. (d) H. Brunner,
Synthesis (1988) 645. (e) M. Noyori, M. Kitamura, Modern
Synthetic Methods 5 (1989) 155. (f) I. Ojima, M. Clos, C. Bastos,
Tetrahedron 45 (1989) 6901. (g) R. Noyori, H. Takaya, Acc.
Chem. Res. 23 (1990) 345. (h) M. Sasaya, Y. Ito, Chem. Rev. 92
(1992) 857. (i) H.U. Blaser, Chem. Rev. 92 (1992) 935. (j) I. Ojima
(Ed.), Catalytic Asymmetric Synthesis, VCH, Weinheim, 1993. (k)
R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley,
New York, 1994.
[2] (a) M. Bols, Carbohydrate Building Blocks, Wiley, New York,
1996. (b) W.A. Nugent, T.V. RajanBabu, M.J. Burk, Science 259
(1993) 479.
[3] (a) J.D. Morrison, J.W. Scott, Asymmetric Synthesis, vol. 14,
Academic Press, San Diego, 1984. (b) J.S. Penne, Chiral Auxil-
iaries and Ligands in Asymmetric Synthesis, Wiley, New York,
1995.
[16] F. Fache, P. Gamez, F. Nour, M. Lemaire, J. Mol. Catal. 85 (1993)
131.
[17] (a) C. Claver, S. Castillo´n, N. Ruiz, G. Delogu, D. Fabbri, S.
Gladiali, J. Chem. Soc. Chem. Commun. (1993) 1833. (b) A.M.
Masde´u, A. Orejo´n, A. Ruiz, S.Castillo´n, C. Claver, J. Mol. Catal.
94 (1994) 149. (c) A. Aaliti, A.M. Masde´u, A. Ruiz, C. Claver,
J. Organomet. Chem. 489 (1995) 101. (d) A. Castellanos-Pa´ez, S.
Castillo´n, C. Claver, J. Organomet. Chem. 539 (1997) 1. (e) N.
Ruiz, A. Aaliti, J. Fornie´s-Ca´mer, A. Ruiz, C. Claver, C.J. Cardin,
D. Fabbri, S. Gladiali, J. Organomet. Chem. 545–546 (1997) 79.
(f) J. Fornie´s-Ca´mer, A. Aaliti, N. Ruiz, A.M. Masde´u-Bulto´, C.
Claver, C.J. Cardin, J. Organomet. Chem. 539 (1997) 199.
[18] D.M.C. Hull, P.F. Ochard, L.N. Owen, J. Chem. Soc. Perkin
Trans. I 10 (1977) 1234.
[19] (a) A.M. Masdeu, A. Ruiz, S. Castillo´n, C. Claver, P.B. Hitchcock,
P. Chaloner, C. Bo, J.M. Poblet, P. Sarasa, J. Chem. Soc. Dalton
Trans. (1993) 2689. (b) N. Ruiz, S. Castillo´n, A. Ruiz, C. Claver,
A. Aaliti, A. Alvarez-Larena, F. Piniella, G. Germain, J. Chem.
Soc. Dalton Trans. (1996) 969.
[4] H. Brunner, W. Zettlmeier, Handbook of Enantioselective Catal-
ysis, VCH, Weinheim, 1993.
[5] (a) M. Sawamura, K. Kitayama, Y. Ito, Tetrahedron Asymm. 4

O. Pa`mies et al. / Journal of Organometallic Chemistry 586 (1999) 125–137
137
[20] R.V. Parish, NMR, NQR, EPR, Mo¨ssbauer Spectroscopy in
Inorganic Chemistry, Ellis Horwood, Chichester, UK, 1990.
[21] (a) C. Claver, A.M. Masde´u, N. Ruiz, C. Foces-Foces, F.H.
Cano, M.C. Apreda, L.A. Oro, J. Garc´ıa-Alejandre, H. Tor-
rens, J. Organomet. Chem. 398 (1990) 177. (b) J. Pursiainen,
T. Teppana, S. Rossi, A. Pakkanen, Acta Chem. Scan. 47
(1993) 416. (c) J.J. Bonnet, D. de Montauzon, R. Poilblanc, J.
Galy, Acta Crystallogr. B35 (1979) 832.
[32] (a) G. Giordano, R.H. Crabtree, Inorg. Synth. 19 (1979) 218.
(b) R. Uso´n, L.A. Oro, J.A. Cabeza, Inorg. Synth. 23 (1985)
126.
[33] (a) G. Winkhaus, H. Singer, Chem. Ber. 99 (1966) 3610. (b) J.
Chatt, J.M. Davidson, J. Chem. Soc. (1965) 843.
[34] (a) J. Gallay, D. de Montouzon, R. Poilblanc, J. Organomet.
Chem. 38 (1972) 179. (b) F. Bonati, G. Wilkinson, J. Chem.
Soc. (1964) 3156.
[22] (a) E. Folga, T. Ziegler, J. Am. Chem. Soc. 115 (1995) 5169.
(b) J. Li, G. Schreckenbach, T. Ziegler, Inorg. Chem. 34
(1995) 3245. (c) R.G. Parr, W. Yang, Density-functional The-
ory of Atoms, Molecules, Oxford University Press, New York,
1989. (d) E.S. Kryachko, E.V. Ludena, Density Functional
Theory of Many Electron Systems, Kluwer Press, Dordrecht,
1990.
[23] A. Elduque, L.A. Oro, M.T. Pinillos, A. Tiripicchio, F. Ugoz-
zoli, J. Chem. Soc. Dalton Trans. 3 (1994) 385.
[24] M. Cowie, J.T. Magne, A.R. Sanger, J. Am. Chem. Soc. 100
(1978) 3628.
[25] (a) R.F.W. Bader, Atoms in Molecules. A Quantum Theory;
Clarendon Press, Oxford, 1990. (b) R.F.W. Bader, Chem. Rev.
91 (1992) 893.
[26] P. Kalck, R. Poilblanc, Inorg. Chem. 14 (1975) 2779.
[27] (a) A.W. Ehler, Y. Ruiz-Morales, E.I. Baerends, T. Ziegler,
Inorg. Chem. 36 (1997) 5031. (b) R.K. Szilagyi and G. Frenk-
ing, Organometallics 16 (1997) 4807. (c) V. James, W. Thiel,
Organometallics 17 (1998) 353.
[28] (a) J.J. Bonnet, J.P. Kalck, A. Maisonnat, R. Poilblanc, Inorg.
Chem. 14 (1975) 743. (b) J.J. Bonnet, J.P. Kalck, R. Poilblanc,
Inorg. Chem. 16 (1977) 1514.
[29] A. Polo, C. Claver, S. Castillo´n, A. Ruiz, J.C. Bayo´n, J. Real,
C. Mealli, D. Masi, Organometallics 11 (1992) 3525.
[30] A.M. Trzeciak, J.J. Ziolkowski, S. Aygen, R. van Eldik, J.
Mol. Catal. 34 (1986) 337.
[35] (a) K.P.R. Kartha, Tetrahedron Let. 27 (1986) 3415. (b) A.I.
Vogel, Vogel’s Textbook of Practical Organic Chemistry, 5th
edition, Longman, NY, 1994.
[36] (a) P.A. Levene, A.L. Raymond, J. Biol. Chem. 102 (1933)
317. (b) D.H. Hollenberg, R.S. Klein, J.J. Fox, Carbohydr.
Res. 67 (1978) 491. (c) G. Ritzmann, R.S. Klein, D.H. Hollen-
berg, J.J. Fox, Carbohydr. Res. 39 (1975) 227. (d) H. Tsut-
sumi, Y. Kawai, Y. Ishido, Carbohydr. Res. 73 (1979) 293.
[37] (a) ADF 2.01–2.2 Theoretical Chemistry, Vrije Universiteit
Amsterdam. (b) E.J. Baerends, D.E. Ellis, P. Ros, Chem.
Phys. 2 (1973) 41. (c) G. te Velde, E.J. Baerends, J. Comp.
Phys. 99 (1992) 84. (d) C. Fonseca Guerra, METECC-95
(1995) 305.
[38] S.H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 58 (1980)
1200.
[39] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[40] J.P. Perdew, Phys. Rev. B 33 (1986) 8822.
[41] CERIUS2, version 3.5, Molecular Simulations INC., 1997.
[42] (a) L.A. Castonguay, A.K. Rappe, J. Am. Chem. Soc. 114
(1992) 583. (b) A.K. Rappe, C.J. Casewit, K.S. Colwell, W.A.
Goddard-III, W.M. Skiff, J. Am. Chem. Soc.114 (1992) 10024.
(c) A.K. Rappe, K.S. Colwell, Inorg. Chem. 32 (1993) 3438.
[43] A.K. Rappe, C.J. Casewit, W.A. Goddard-III, J. Phys. Chem.
95 (1991) 3358.
[44] F.W. Biegler-Ko¨nig, R.F.W. Bader, T.H. Tang, J. Comp.
Chem. 3 (1982) 317.
[45] Xaim software developed by J.C. Ortiz, C. Bo, Universitat
Rovira i Virgili, Tarragona, 1998.
[31] M. Die´guez, C. Claver, A.M. Masde´u-Bulto´, A. Ruiz,
P.W.N.M. van Leeuwen, G.C. Schoemaker, Organometallics
18 (1999) 2107.
.
.