Cationic rhodium (I) complexes of n-phosphino-tert-butylsulfinamide ligands: Synthesis, Structure, and Coordination Modes

Article abstract of DOI:10.1021/om8006823

Here we describe the rhodium (I) complexes of N-phosphino-iert- butylsulfinamide ligands. This novel class of hemilabile ligands (PNSO) can work as either P,0 or P,S chelating ligands when attached to the square planar rhodium center. Complexes bearing di

Full text of DOI:10.1021/om8006823

480
Organometallics 2009, 28, 480–487
Cationic Rhodium (I) Complexes of
N-Phosphino-tert-butylsulfinamide Ligands: Synthesis, Structure,
and Coordination Modes
Thierry Achard, Jordi Benet-Buchholz, Antoni Riera,* and Xavier Verdaguer*
Unitat de Recerca en S´ıntesi Asime`trica (URSA-PCB), Institute for Research in Biomedicine (IRB
Barcelona) and Departament de Qu´ımica Orga`nica, UniVersitat de Barcelona, c/Baldiri Reixac 10,
E-08028 Barcelona, Spain, and Institute of Chemical Research of Catalonia (ICIQ),
AVda. Pa¨ısos Catalans 16, 43007 Tarragona, Spain
ReceiVed July 18, 2008
Here we describe the rhodium (I) complexes of N-phosphino-tert-butylsulfinamide ligands. This novel
class of hemilabile ligands (PNSO) can work as either P,O or P,S chelating ligands when attached to the
square planar rhodium center. Complexes bearing diene ligands such as [Rh(PNSO)(NBD)][TfO] and
[Rh(PNSO)(COD)][TfO] provided P,O coordination, whereas [Rh(PNSO)2][TfO]-type complexes provided
P,S coordination. Ligand exchange experiments with mono and bidentate phosphines afforded evidence
of the hemilabile behavior of the PNSO ligands.
Introduction
Bidentate ligands have become the most powerful tool in
metal-catalyzed asymmetric processes.1 Chelation provides the
rigidity required to firmly allocate the chiral information around
the metal center. Usually, the chirality of these ligands resides
in the carbon backbone; however, some have a chiral phosphorus
or sulfur center. The synthesis of chiral-phosphorus compounds
is complex and difficult. Conversely, chiral-sulfur compounds
can be prepared efficiently, and some are commercially available
in large scale.2-5 Our group has recently reported N-phosphino-
tert-butylsulfinamides (I) as new class of chiral bidentate ligands
(PNSO) and their application in the intermolecular asymmetric
Pauson-Khand reaction.6 In response to dicobalt carbonyl
complexes, the PNSO ligand acts as a hemilabile bridging P,S
ligand, thereby providing type II complexes (Figure 1).
Figure 1
Scheme 1. Possible Coordination Modes of N-Phosphino
Sulfinamide Ligands with Mono- and Bimetallic Complexes
An essential feature of hemilabile ligands is the presence of
at least one labile donor function while the other donor group
remains firmly bound to the metal center(s). Hemilabile ligands
have the capacity to provide open coordination sites at the metal
center during reactions that are “masked” in the ground-state
structure and to stabilize reactive intermediates.7 PNSO ligands
are an unusual class of hemilabile ligands, since the sulfoxide
moiety can bind the metal through either sulfur or oxygen, thus
enabling the ligand to provide both P,S and P,O chelation
(Scheme 1). We have previously shown that PNSO ligands work
solely as P,S bridging ligands toward bimetallic cobalt-carbonyl
complexes; however, there is no data on the coordination
behavior of these ligands toward monometallic systems. Here,
we report on the cationic rhodium (I) complexes of N-benzyl-
N-phosphino-tert-butylsulfinamide ligands, their structure, co-
ordination modes, and their behavior toward ligand exchange
reactions.
Results and Discussion
* Corresponding author. E-mail: xavier.verdaguer@irbbarcelona.org.
(1) Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H. In ComprehensiVe
Asymmetric Catalysis; Springer: Berlin, 1999; Vol. I-III.
(2) Fernandez, I.; Khiar, N. Chem. ReV. 2003, 103, 3651–3706.
(3) Pellissier, H. Tetrahedron 2006, 62, 5559–5601.
(4) Senanayake, C. H.; Krishnamurthy, D.; Lu, Z.; Han, Z.; Gallou, I.
Aldrichim. Acta 2005, 38, 93–104.
Synthesis of the Ligands. Chiral PNSO ligands were
synthesized from the commercially available tert-butylsulfina-
mide (+)-1 (Scheme 2). The synthesis of 1 in two steps from
the inexpensive starting material tert-butyl disulfide has been
reported by Ellman and co-workers.8 From (+)-1, reductive
amination protocol with Ti(OEt)4 in the presence of benzalde-
hyde produced an intermediate sulfinimine, which was reduced
in situ to the corresponding sulfinamide (2).9 In an optimized
(5) Mellah, M.; Voituriez, A.; Schulz, E. Chem. ReV. 2007, 107, 5133–
5209.
(6) Sola`, J.; Reve´s, M.; Riera, A.; Verdaguer, X. Angew. Chem., Int.
Ed. 2007, 46, 5020–5023. Reve´s, M.; Achard, T.; Sola`, J.; Riera, A.;
Verdaguer, X. J. Org. Chem. 2008, 73, 7080–7087.
(7) Braunstein, P.; Naud, F. Angew. Chem., Int. Ed. 2001, 40, 680–
699.
(8) Cogan, D. A.; Liu, G.; Kim, K.; Backes, B. J.; Ellman, J. A. J. Am.
Chem. Soc. 1998, 120, 8011–8019.
10.1021/om8006823 CCC: $40.75
2009 American Chemical Society
Publication on Web 12/23/2008

Cationic Rhodium (I) Complexes
Organometallics, Vol. 28, No. 2, 2009 481
Scheme 2. Synthesis of Optically Pure N-Phosphino-N-benzyl-
tert-butylsulfinamide Ligands
Scheme 3. Synthesis of Cationic Rhodium (I) Complexes
with Ligand 3
analysis. The crystalline compound (5) was the corresponding
(PNSO)RhI(NBD) triflate complex with the PNSO ligand acting
as a bidentate P,O ligand (Figure 2). 1H NMR of pure 5 shows
resonances at 1.34 ppm for the t-Bu group and two low-field
broad doublets for the benzylic PhCH2N protons at 4.47 and
4.58 ppm. The metal coordinated NBD showed the olefinic
protons as a broad resonance centered at 4-5 ppm. Unfortu-
nately, we could obtain only small amounts of pure 5 with this
methodology, and its preparation was not consistently reproduc-
ible. In addition to this, compound 5 showed limited stability
in solution.
protocol, anion formation with n-BuLi at low temperature
followed by the addition of either Ph2PCl or Cy2PCl and
protection of the phosphine moiety with borane provided the
corresponding borane-protected ligands 3-BH3 and 4-BH3
(Scheme 2). In the case of diphenylphosphine, the borane-free
ligand was obtained by deprotection with DABCO at room
temperature to afford (+)-3 as a crystalline solid in good yield.
Conversely, borane-free ligand 4 could not be isolated because
of the undesired sulfur-to-oxygen migration that leads to the
corresponding phosphine oxide.10,11 Therefore, the cyclohexy-
lphosphine ligand was used in its borane-protected form, 4-BH3,
throughout the present study.
Alternatively, the analogous (PNSO)RhI(COD) complex was
prepared in good yield (85%) using [RhCl(COD)]2 as a metal
source (Scheme 3). Thus, chloride abstraction with AgOTf in
dry THF and addition of ligand 3 under N2 provided the desired
compound (7) after 40 min at room temperature. This compound
was isolated as an air-stable orange solid by layering the reaction
N-Benzyl-N-diphenylphosphino-tert-butylsulfinamide
(3): Synthesis and Characterization of Its Cationic Rho-
dium Complexes. The first attempts to obtain a cationic Rh
complex with ligand 3 followed the procedure described by
Alcock and co-workers, who reported the synthesis of a
phosphine sulfoxide ligand and its RhI-NBD complex.12 In that
study, Alcock used (NBD)Rh(acac) as the rhodium source and
TMSOTf to remove the acac ligand from the metal. Following
this procedure, we treated a solution of (NBD)Rh(acac) in dry
THF at room temperature under N2 with 1 equiv of TMSOTf,
followed by 1 equiv of ligand 3 (Scheme 3). After 20 min at
room temperature, the mixture was added to a vigorously stirred
hexane solution to give a yellow precipitate. 1H NMR (CDCl3)
of this solid showed both noncoordinated and metal-coordinated
NBD and no trace of the corresponding original ligand (t-Bu,
0.93 ppm). Two new major metal species were present with
resonances at 1.47 and 1.34 ppm for the t-Bu group. Consis-
tently, 31P NMR showed two doublets at 69.5 and 115.8 ppm,
respectively, with a phosphorus-rhodium coupling constant of
141 and 193 Hz, respectively. This mixture was crystallized
upon layering a hexane/Et2O mixture over CH2Cl2. This
approach produced a few dark orange crystals suitable for X-ray
1
mixture with diethyl ether. H NMR of 7 showed the t-Bu
resonance at 1.54 ppm in CDCl3 and four signals corresponding
to the olefinic protons of the metal-coordinated COD ligand,
as would be expected for a complex with no symmetry. Suitable
crystals for X-ray analysis were obtained by slow diffusion of
Et2O into a CH2Cl2 solution of 7. The resulting structure revealed
that 3 again acts as a P,O bidentate ligand (Figure 2), in analogy
to the structure of complex 5.
Metal-phosphorus and metal-oxygen, as well as oxygen-
sulfur distances for 5 and 7 are consistent with the structure of
[Ph2PCH2S(O)Ph]RhI(NBD) reported by Alcock and co-workers
(Figure 2).12 The Rh-O-S-N-P atoms formed a five-member
ring with an envelope-like conformation in which the oxygen
bonded to rhodium was slightly out of plane (0.44 Å). This
observation differs from the structure reported by Alcock, in
which it is the central carbon atom of the ligand that is out of
plane. A strong trans effect was observed for both structures
on the Rh-olefin distances. For example, 5 showed Rh-C
(olefinic) distance trans to phosphorus of 2.231(2) and 2.237(2),
whereas the Rh-C (olefinic) distance trans to oxygen was
2.093(1) and 2.093(2). The olefinic group placed trans to the
most electronegative donor atom presented a shorter and stronger
bond to the metal. Alignment of structures 5 and 7 without the
olefinic ligands produced an almost perfect fit, except for the
N-benzyl group (Figure 3). In complex 5, the benzyl group was
perpendicular to the phenyl ring of the phosphine, whereas in
complex 7, it showed a parallel-type disposition. The observed
(9) Tanuwidjaja, J.; Peltier, H. M.; Ellman, J. A. J. Org. Chem. 2007,
72, 626–629.
(10) Hiroi, K.; Suzuki, Y.; Kawagishi, R. Tetrahedron Lett. 1999, 40,
715–718.
(11) Vedejs, E.; Meier, G. P.; Powell, D. W.; Mastalerz, H. J. Org. Chem.
1981, 46, 5253–5254.
(12) Alcock, N. W.; Brown, J. M.; Evans, P. L. J. Organomet. Chem.
1988, 356, 233–247.

482 Organometallics, Vol. 28, No. 2, 2009
Achard et al.
Figure 2. Ortep plot for crystal structures 5 and 7 with thermal ellipsoids shown at 50% probability. Only the metal cations are shown. For
5, a single molecule of the two independent complexes that form the original unit cell is shown. Selected bond distances (Å) for 5 and 7,
respectively: Rh-O, 2.116(1) and 2.100(1); Rh-P, 2.262(1) and 2.250(1); S-O, 1.527(1) and 1.523(1).
to one of the substances obtained in the initial experiments with
(NBD)Rh(acac). Finally, ESI mass spectroscopy confirmed the
new substance (6) as a bis-(PNSO) rhodium (I) triflate.
Traditionally, IR spectroscopy is a good tool to distinguish
whether O-M or S-M bonding occurs in sulfoxide metal
complexes.14 In our case, although noncoordinated 3 displayed
an intense SdO absorption band at 1074 cm-1, no diagnostic
bands were observed for complexes 5, 6, or 7.
To elucidate the structure of the dimeric PNSO complex and
the coordination mode of the sulfinamide moiety, single crystals
for X-ray analysis were obtained by layering Et2O over a CH2Cl2
solution of 6 (Figure 4). The Rh center presented a slightly
distorted square-planar geometry, with the two phosphorus
atoms (and sulfur) in a cis arrangement in close analogy with
other reported P-S, P-N, or P-O ligands.15-18 The Rh, S, N,
Figure 3. Alignment of Rh-P-N centers for complexes 5 and 7.
NBD and COD ligands are omitted for clarity.
and P atoms adopted a planar double diamond disposition in
which the metal center was the connection point of two identical
four-member rings. The complex showed a C2 symmetry axis
O-bonding mode to the metal led to a considerable lengthening
in the coordination plane of the metal. The oxygen atoms and
of the SdO bond (1.527(1) and 1.523(1) Å for 5 and 7,
the t-Bu groups of the sulfinamide were in anti position to each
respectively) in comparison with noncoordinated DMSO 1.493(av)
other to minimize steric and electronic interactions. The PNSO
Å. O-Coordination could, in principle, be expected in response
ligand provided bite angle (P-Rh-S) values of 71.33(2) and
to a hard cationic-RhI center. 31P NMR spectra of complexes 5
71.20(2)°, which fall between dppm (70.80°) and MiniPHOS
and 7 showed resonances at 115.8 and 116.5 ppm, respectively,
(72.96°) ligands in analogous RhI complexes.19,20 The Rh-P
whereas the free ligand 3 displayed the corresponding resonance
(2.251(1) and 2.249(1) Å) and Rh-S (2.289(1) and 2.288(1)
at 60.0 ppm. In addition, Milstein and co-workers reported a
Å) distances were quite similar. In this respect, the Rh-S
[Rh(COD)(DMSO)2][BF4] complex in which both DMSO
distances observed were in the long range for an S-bonding
molecules bind the metal through oxygen.13 These findings
mode.13,14,21 This observation could be attributed to the strong
indicate that the presence of a diolefinic ligand on the metal
trans effect exerted by the phosphine ligands. Moreover, the
makes this center more electron-deficient, thereby favoring the
occurrence of S-bonding in complex 6 can be reasoned on the
O-coordination mode of the PNSO ligand.
We next examined the preparation of rhodium species with
(14) Calligaris, M. Coord. Chem. ReV. 2004, 248, 351–375.
two PNSO ligands. With this purpose, [RhCl(COD)]2 was
treated with AgOTf in dichloromethane, and the resulting
solution was added to a solution of 2 equiv of ligand 3 and
stirred for 24 h at room temperature (Scheme 3). The resulting
yellow mixture was poured over stirred Et2O to provide an air-
stable yellow precipitate. Spectroscopic analysis of this solid
(15) Dick, D. G.; Stephan, D. W. Can. J. Chem. 1986, 64, 1870–1875.
(16) Esquius, G.; Pons, J.; Yanez, R.; Ros, J.; Mathieu, R.; Donnadieu,
B.; Lugan, N. Eur. J. Inorg. Chem. 2002, 299, 9–3006.
(17) Kuznetsov, V. F.; Facey, G. A.; Yap, G. P. A.; Alper, H.
Organometallics 1999, 18, 4706–4711.
(18) Braunstein, P.; Chauvin, Y.; Naehring, J.; DeCian, A.; Fischer, J.;
Tiripicchio, A.; Ugozzoli, F. Organometallics 1996, 15, 5551–5567.
(19) Gridnev, I. D.; Yamanoi, Y.; Higashi, N.; Tsuruta, H.; Yasutake,
M.; Imamoto, T. AdV. Synth. Catal. 2001, 343, 118–136.
(20) Fornika, R.; Six, C.; Gorls, H.; Kessler, M.; Kruger, C.; Leitner,
W. Can. J. Chem. 2001, 79, 642–648.
1
revealed the presence of a single substance with H NMR
resonances at 1.47 ppm (t-Bu). 31P NMR showed a doublet at
69.5 ppm and coupling constant of 141 Hz, which corresponded
(21) For the structure of a related neutral sulfinamide Rh(I) (COD)
complex displaying sulfur coordination, see: (a) Souers, A. J.; Owens, T. D.;
Oliver, A. G.; Hollander, F. J.; Ellman, J. A. Inorg. Chem. 2001, 40, 5299–
5301.
(13) Dorta, R.; Rozenberg, H.; Shimon, L. J. W.; Milstein, D.
Chem.sEur. J. 2003, 9, 5237–5249.

Cationic Rhodium (I) Complexes
Organometallics, Vol. 28, No. 2, 2009 483
Figure 4. Ortep plot for crystal structure 6 with thermal ellipsoids shown at 50% probability. Only the metal cation is shown. Selected bond
distances (Å) and angles (deg): Rh-P, 2.251(1) and 2.249(1); Rh-S, 2.289(1) and 2.288(1); P-Rh-S, 71.33(2) and 71.20(2).
Scheme 4. Synthesis of Cationic Rhodium Complexes with
Ligand 4
The product was purified by flash chromatography of the
reaction mixture on SiO2 using hexane/acetone mixtures as
mobile phase. This approach provided the desired bis PNSO
complex 8 in an excellent 92% yield. 1H NMR revealed a single
t-Bu resonance at 1.46 ppm and for the benzylic PhCH2N an
AB system coupled to phosphorus at 4.29 and 4.46 ppm. The
31P NMR spectrum displayed a single doublet at 86 ppm with
a coupling constant to Rh of 121 Hz. ESI mass spectroscopy
confirmed complex 8 as a bis-PNSO-rhodium complex. Again,
no concluding information on the coordination mode of the
sufoxide fragment could be drawn from the IR spectra of 8.
Finally, crystallization of 8 in a hexane/THF mixture provided
green-yellow plated crystals suitable for X-Ray analysis (Figure
5). In contrast to what we expected, complex 8 showed a trans
configuration of the two phosphines and the two sulfinamide
groups around the square planar rhodium center. The complex
now showed a C2 axis perpendicular to the Rh coordination
plane. This configuration positioned the two t-Bu groups on the
same face of the complex and caused a distortion of the square
planar geometry around the metal. Thus, the P-Rh-P angle
was 165.28(1)°, whereas the analogous S-Rh-S angle was
167.97(1)°. Longer Rh-P distances were observed (2.312(1)
and 2.297(1) Å) for 8 because of the dicyclohexylphosphine
residue’s being a weaker back-bonding ligand than dipheny-
phosphine. Like complex 6, double S-coordination mode was
observed for the sufinamide fragment. The Rh-S distances were
2.251(1) and 2.269(1) Å, again falling in the long-range for
this type of bonding. This distance is in agreement with the
trans S-bonded DMSO molecules in rhodium complexes
reported by Milstein and co-workers and can be explained on
the basis that S-bound sulfoxides have a relatively higher trans
influence than O-bound sulfoxides.13,22
basis that the metal center, which lacks the bis-olefin ligand, is
more electron-rich than complexes 5 and 7 and thus prefers the
softer S-coordination mode. This notion is corroborated by the
31P NMR of complex 6 (69.5 ppm), which is very close to that
of the free ligand 3 (60.0 ppm).
N-Benzyl-N-dicyclohexylphosphino-tert-butylsulfinamide
(4): Synthesis and Characterization of Its Cationic Rh
Complexes. The use of ligand 4 in the preparation of rhodium
complexes presented an additional challenge; that is, it could
not be isolated pure in its borane-free form. Deprotection of
borane-protected PNSO ligands occurs in toluene in the presence
of DABCO. Under these conditions in the presence of a base
and prior to purification, the ligand preserves its integrity, as
determined by 1H and 31P NMR (Scheme 4). Under these
circumstances, we attempted the synthesis of the corresponding
bis-PNSO-Rh1 complex. Thus, chloride removal on
[RhCl(COD)]2 with AgOTf, an addition of the resulting solution
to deprotected 4, was left to stir for 24 h at room temperature.
After some experimentation, we observed that the bis-PNSO-
Rh complex 8 was not configurationally stable. Solving 8 in
1
THF provided a distinct PNSO complex. H NMR analysis of
the new complex 9 showed a t-Bu resonance at 1.63 ppm, and
in the benzylic region, the original AB system for 8 changed to
a multiplet. Additionally, a new doublet at 95 ppm and JRh
)
132 Hz was observed in the 31P NMR spectrum. Compound 9
crystallized as air-stable yellow needles (THF/hexanes). High-
resolution mass spectroscopy (ESI) revealed a composition
(22) Dorta, R.; Rozenberg, H.; Milstein, D. Chem. Commun. 2002, 710–
711.

484 Organometallics, Vol. 28, No. 2, 2009
Achard et al.
Figure 5. Ortep plot for crystal structure 8 with thermal ellipsoids shown at 50% probability (left). Lateral view of the central coordination
core (right). Cyclohexyl and benzyl groups are omitted for clarity. Only the metal cation is shown. Selected bond distances (Å) and angles
(deg): Rh-P, 2.312(1) and 2.297(1); Rh-S, 2.251(1) and 2.269(1); P-Rh-P, 165.28(1); S-Rh-S, 167.97(1).
Scheme 5. Reactivity of Rhodium (I) Complexes with Phosphines
identical to that of 8. Solving 8 in a coordinating solvent such
as THF (24 h) and acetonitrile (24 h) provided full conversion
to the new complex 9 (Scheme 4). However, no isomerization
from 8 to 9 was observed in CDCl3 or toluene for 24 h.
Moreover, the isomerization was not reversible, even when
solving 9 in a nonpolar solvent for a long period of time (i.e.,
toluene). These experimental observations indicated that com-
plex 8 is a kinetic product and 9 is the thermodynamically more
stable complex. From the electronic point of view, the similarity
of the 31P NMR resonances for 8 and 9 suggests that both are
S-bonded complexes. All things considered, the proposed
structure for complex 9 is the corresponding S-coordinated cis
analog of 8 (Scheme 4). The cis configuration should be the
thermodynamically preferred one. In structure 9 the t-Bu groups
are placed at opposite faces of the complex, thus releasing the
steric strain built up in 8 that resulted in a distortion of the
square planar geometry.
Reaction of [Rh(PNSO)(COD)][TfO] (7) with two equivalents
of PPh3 provided the corresponding [Rh(PNSO)(PPh3)2] [TfO]
complex 10, which was synthesized and isolated in an excellent
91% yield in preparative scale. Compound 10 showed the
characteristic pattern for three nonequivalent phosphorus atoms
attached to a square-planar Rh center in the 31P NMR: three
resonances with three distinct couplings each, two couplings to
phosphorus, and one to rhodium. In an analogous fashion, after
75 min, the reaction of 7 with dppe provided the expected
[Rh(PNSO)(dppe)] (11) as a major complex, along with smaller
amounts of [Rh(dppe)2] (12) and [Rh(PNSO)2] (6) in a 16:4:1
ratio (Scheme 5). The reaction spectrum at 8 min (Figure 6)
showed two species that disappeared over time: free ligand 3
(s, 60.0 ppm) and the [Rh(dppe)(COD)] complex 13 (d, 56.5
ppm, JRh ) 149 Hz).23 The reactivity of 7 with PPh3 and dppe
is consistent with 3 being a labile ligand. Thus, the PNSO ligand
provided coordination sites for the incoming phosphine to
Reactivity of Rhodium (I) Complexes with Phosphines.
To examine how competent the PNSO ligands were with respect
to other phosphines, ligand exchange reactions with PPh3 and
dppe were carried out and monitored by 31P NMR (Scheme 5).
(23) de Wolf, E.; Spek, A. L.; Kuipers, B. W. M.; Philipse, A. P.;
Meeldijk, J. D.; Bomans, P. H. H.; Frederik, P. M.; Deelman, B.; van Koten,
G. Tetrahedron 2002, 58, 3911–3922.

Cationic Rhodium (I) Complexes
Organometallics, Vol. 28, No. 2, 2009 485
Figure 6. 31P NMR monitoring for the reaction of complex 7 with 1 equiv of dppe in CDCl3.
coordinate the metal. This behavior allowed the displacement
of the COD ligand with PPh3 for 7. It is known that the COD
ligand cannot be displaced from [Rh(PPh3)2(COD)] complex
with excess PPh3.24 In addition, the emergence of [Rh(dppe)-
(COD)] (13) and free PNSO ligand at the initial stages of the
reaction reveals that dppe preferentially removes the PNSO
rather that the COD ligand and that only later does the PNSO
displace the COD ligand. With an electron-rich RhI center, the
PNSO ligand in complexes 10 and 11 was presumed to display
an S-coordination mode, in concordance with complexes in 6
and 8.25
Finally, interesting and distinct behaviors were observed when
bis-PNSO complexes 6 and 9 were treated with PPh3 and dppe
(Scheme 5). Although 6 and 9 were similarly unreactive toward
PPh3, they showed a different reaction profile with a diphos-
phine. Reaction of dppe with 6 produced a complete ligand
exchange to yield the corresponding bis-dppe rhodium complex
12 in only 8 min. However, no reaction was observed after 5 h
when cyclohexyl PNSO complex 9 was treated with dppe. The
reactivity observed allows us to confirm that PNSO ligands are
more competent ligands than monophosphines and that they
cannot be replaced by PPh3. Furthermore, a chelating diphos-
phine such as dppe can easily replace the diphenylphosphino
PNSO ligand 3 from the Rh coordination sphere, but not the
cyclohexyl analog 4.
Conclusions
Here, we report on the first rhodium (I) complexes with PNSO
ligands. We have shown that these compounds can work as
(24) Albers, J.; Dinjus, E.; Pitter, S.; Walter, O. J. Mol. Catal. A 2004,
219, 41–46.
(25) No characteristic absorption bands were identified for the SdO
stretch in these complexes; see reference 14.

486 Organometallics, Vol. 28, No. 2, 2009
Achard et al.
flask protected from light with aluminum foil. The flask was then
purged with N2, and dry CH2Cl2 (2 mL) was added and stirred for
2 h. The resulting white precipitate (AgCl) was removed by filtration
over Celite under nitrogen and subsequently washed with extra
CH2Cl2 (3 × 2 mL). This orange solution was added dropwise into
a Schlenk tube containing a solution of ligand (+)-3 (0.320 g, 0.808
mmol) in CH2Cl2 (2 mL), and the mixture turned yellow. The
reaction was stirred for 24 h at room temperature, then the yellow
solution was evaporated under vacuum to a small volume (2-3
mL) and poured over vigorously stirred Et2O (50 mL). The resulting
precipitate was filtered, washed with 2 × 10 mL of Et2O and dried
to give complex 6 as a yellow shiny powder (0.36 g, 90%). Suitable
crystals for X-ray analysis were obtained upon layering Et2O over
a CH2Cl2 solution of 6. mp: 177 °C. [R]D ) +34 (c ) 0.1, CHCl3).
either P,O or P,S chelating ligands, depending on the electronic
environment around the metal center. Thus, N-benzyl-N-
diphenylphosphino-tert-butylsulfinamide (3) provides P,O co-
ordination when olefin ligands (NBD or COD) are attached to
the metal center. Alternatively, rhodium complexes with two
PNSO ligands provides P,S coordination. [Rh(PNSO)2] com-
plexes preferentially show the thermodynamically most stable
cis configuration. However, we have identified the kinetic trans
complex with a dicyclohexylphosphino analog of 3. Finally,
ligand exchange experiments with PPh3 and dppe show that
PNSO ligands act as hemilabile ligands by providing coordina-
tion sites for the incoming phosphines.
IR (KBr): νmax ) 1439, 1272, 1149, 1131, 1106, 1062, 1031 cm-1
.
Experimental Section
1H NMR (400 MHz, CDCl3): δ 1.47 (s, 18H), 4.35 (dd, J ) 11
and 17 Hz, 2H), 4.43 (dd, J ) 10 and 17 Hz, 2H), 6.66 (d, J ) 8
Hz, 4H), 6.95-7.02 (m, 8H), 7.06-7.12 (m, 6H), 7.47-7.56 (m,
10H), 7.70 (t, J ) 8 Hz, 2H) ppm. 13C NMR (100 MHz, {31P},
CDCl3): δ 24.7 (3 × CH3), 49.7 (CH2), 68.6 (C), 125.5 (C), 128.0
(C), 128.3 (CH), 128.4 (CH), 128.7 (CH), 129.4 (CH), 130.0 (CH),
132.2 (CH), 132.6 (CH), 133.7 (C), 133.9 (CH), 134.2 (CH) ppm.
31P NMR (121 MHz, CDCl3): 69.5 (d, J ) 141 Hz) ppm. MS (ESI,
H2O/CH3CN (1:1) 1% formic) m/z: 893 [100, (M - OTf-)+].
HRMS (ESI): calcd. for C46H52N2O2P2S2Rh+, 893.1995; found,
893.1988. Anal. calcd. for C47H52F3N2O5P2RhS3: C, 54.12; H, 5.03;
N, 2.69. Found: C, 54.16; H, 4.93; N, 2.60.
(+)-(R)-N-Benzyl-N-dicyclohexylphosphino-tert-butylsulfina-
mide Borane (4-BH3). A 1.3 mL portion of n-BuLi 2.5 M in hexane
(3.2 mmol) was added dropwise to a cooled (-78 °C) solution of
the corresponding (R)-(-)-N-benzyl-tert-butylsulfinamide (600 mg,
2.8 mmol) in THF (20 mL). After 15 min, 0.72 mL (0.76 g, 3.3
mmol) of chlorodicyclohexylphosphine was added via syringe. The
mixture was stirred at -78 °C for 1 h, and then the temperature
was allowed to rise to -30 °C. At this point, 0.35 mL (0.28 g, 3.7
mmol) of BH3-SMe2 was added via syringe. The mixture was
stirred at this temperature for 30 min and then warmed to 0 °C.
The reaction was then quenched with 15 mL of water (caution:
bubbling occurs) and extracted with Et2O (30 mL). The organic
layer was dried over MgSO4, filtered, and concentrated under
reduced pressure. The crude products were purified by flash column
chromatography (SiO2, hexanes/AcOEt, 80/20) to afford 1.00 g
(85%) of the desired ligand as a foamy oil. [R]D ) +49.3 (c 1.0,
(SR)-[Rh(PNSO)(COD)][TfO] (7). [RhCl(COD)]2 (0.2 g, 0.404
mmol) and AgOTf (0.208 g, 0.809 mmol) were placed in a Schlenk
flask protected from light with aluminum foil. The flask was then
purged with N2, and dry THF (2 mL) was added and stirred for
2 h. The resulting white precipitate (AgCl) was removed by filtration
over Celite under nitrogen, and subsequently washed with extra
THF (3 × 2 mL). The resulting orange solution was transferred
into a Schlenk tube, and a solution of ligand 3 (0.320 g, 0.808
mmol) in THF was added dropwise. The reaction was stirred for
40 min at room temperature. Finally, the solvent was removed under
vacuum to a small volume (2-3 mL), and Et2O (10 mL) was
layered over the reaction mixture. After one night, this afforded a
crop of orange crystals (0.518 g, 85%) of complex 7. Suitable
yellow/orange crystals for X-ray analysis were obtained upon
layering Et2O over a CH2Cl2 solution of 7. [R]D ) -115 (c ) 0.1,
CHCl3); IR (KBr): νmax ) 1274, 1223, 1148, 1031 cm-1. 1H NMR
(400 MHz, CDCl3): δ 1.54 (s, 9H), 2.05-2.52 (br m, 8H), 3.55
(br, 1H), 3.74 (br, 1H), 4.49 (dd, J ) 15 Hz and JP ) 4 Hz, 1H),
4.70 (dd, J ) 15 Hz and JP ) 4 Hz, 1H), 5.37 (br, 1H), 5.68 (br,
1H), 6.69 (d, J ) 7 Hz, 2H), 7.12-7.19 (m, 3H), 7.37 (br, t, J )
10 Hz, 2H), 7.57-7.71 (m, 6H), 8.01 (t, J ) 8 Hz, 2H) ppm. 13C
NMR (100 MHz, CDCl3): δ 22.6 (3 × CH3), 27.8 (CH2), 28.4
(CH2), 32.0 (CH2), 33.5 (CH2), 53.9 (CH2), 61.9 (C), 70.4 (d, JRh
) 14 Hz, CH), 73.8 (d, JRh ) 14 Hz, CH), 105.6 (CH), 109.6 (CH),
127.0 (d, JP ) 56 Hz, C), 128.5 (C), 128.9 (C), 129.1 (CH), 130.0
(d, JP ) 10 Hz, CH), 130.2 (d, JP ) 11 Hz, CH), 130.5 (d, JP ) 11
Hz, CH), 131.2 (d, JP ) 56 Hz, C), 132.7 (d, JP ) 3 Hz, CH),
133.8 (CH), 134.4 (C), 135.5 (d, JP ) 16 Hz, CH) ppm. 31P NMR
(121 MHz, CDCl3): 116.5 (d, J ) 167 Hz) ppm. MS (ESI, H2O/
CH3CN (1:1) 1% formic) m/z: 893 [100, (2M - OTf - 2COD)+],
539 [37, (M + CH3CN - COD)+]. HRMS (ESI): calcd. for
C24H29N2OPSRh+, 539.0787; found, 539.0786. Anal. calcd. for
C32H38F3NO4PRhS2: C, 50.86; H, 5.07; N, 1.85. Found C, 50.59;
H, 4.90; N, 1.79.
1
CHCl3). IR (KBr): νmax 2932, 2853, 2390, 1451 cm-1. H NMR
(400 MHz, CDCl3): δ 0.20-0.80 (br s, 3H, BH3), 1.10 (s, 9H),
1.14-1.38 (m, 6H), 1.50-1.98 (m, 11H), 2.00-2.24 (m, 5H), 4.46
(dd, J ) 14 and 17 Hz, 1H), 4.68 (dd, J ) 10 and 17 Hz, 1H),
7.21-7.38 (m, 5H). 13C NMR (100 MHz, CDCl3): δ 24.2 (CH3),
26.0-26.2 (m, 2 × CH2), 26.9-27.3 (m, 4 × CH2), 36.2 (d, JP )
36 Hz, CH), 37.2 (d, JP ) 27 Hz, CH), 44.2 (d, JP ) 3 Hz, CH2),
60.9 (C), 127.3 (CH), 128.2 (CH), 128.3 (CH), 139.0 (C) ppm. 31
P
NMR (121 MHz, CDCl3): δ -63.9 (m) ppm. MS (CI, NH3) m/e )
422 ([M + H]+, 25%), 421 ([M]+, 38%), 420 ([M - H]+, 100%),
364 ([M
-
C4H9]+, 21%). HRMS (ESI-TOF) calcd. for
C23H41BNOPS: H, 420.2661; found, 420.2656.
(SR)-[Rh(PNSO)(NBD)][TfO] (5). Me3SiOSO2CF3 (0.062 mL,
0.339 mmol) was added to a yellow solution of (nbd)Rh(acac) (0.1
g, 0.339 mmol) in dry THF (4 mL) under argon. The yellow-orange
solution was stirred for 5 min, and (+)-3 (0.134 g, 0.339 mmol)
was then added in one portion. The solution briefly changed to
orange, and a yellow precipitate separated to leave an orange liquor.
The suspension was stirred at room temperature for 1 h and then
added to vigorously stirred hexane (50 mL); a yellow-orange
precipitate was formed. Most of the mother liquor was removed
by canula and additional hexane (10 mL) was added. The precipitate
was filtered off and dried to give a yellow solid (206 mg) as a
mixture of complexes. Dark orange crystals of 5, suitable for X-ray
analysis, were obtained upon layering a THF solution of the
resulting solid with Et2O and hexane. IR (KBr): νmax ) 1260, 11173,
1
1035 cm-1. H NMR (400 MHz, CDCl3): δ 1.34 (s, 9H), 1.62 (br
2), 3.99 (br, 2H), 4.42-4.48 (br m,, 1H), 4.55-4.60 (br m,, 1H),
6.83 (br, 2H), 7.16-7.20 (br, 4H), 7.42 (br, 3H), 7.51-7.60 (m,
8H), 7.82-7.86 (br, 3H) ppm. 31P NMR (121 MHz, CDCl3): 115.8
(d, JRh ) 193 Hz) ppm. MS (ESI, H2O/CH3CN (1:1) 1% formic)
m/z: 893 [100, (2M - OTf - 2NBD)+], 590 [92, (M - OTf)+].
HRMS (ESI): calcd. for C30H34NOPSRh+, 590.1153; found,
590.1141.
(SR)-trans-[Rh(PNSO)2][TfO] (8). Ligand 4-BH3 (100 mg, 0.237
mmol, 2 equiv), DABCO (28 mg, 0.249 mmol, 2.1 equiv), and dry
toluene (3 mL) were placed in a Schlenk tube under nitrogen. The
solution was warmed to 70 °C for 4 h and then cooled to room
temperature. In a separate Schlenk flask, [RhCl(COD)]2 (29 mg,
0.059 mmol, 0.5 equiv), AgOTf (30 mg, 0.118 mmol, 1 equiv),
(SR)-cis-[Rh(PNSO)2][TfO] (6). [RhCl(COD)]2 (0.1 g, 0.202
mmol) and AgOTf (0.104 g, 0.404 mmol) were placed in a Schlenk

Cationic Rhodium (I) Complexes
Organometallics, Vol. 28, No. 2, 2009 487
and dry THF (2 mL) were stirred at room temperature under
nitrogen for 2 h. The resulting suspension was filtered over Celite
under nitrogen. The Rh solution was then added to the ligand/
DABCO mixture and allowed to stir for 24 h at room temperature.
Removal of solvent under vacuum and purification by flash
chromatography (acetone/hexanes, 60:40) afforded 116 mg (92%)
of complex 8 as an air-stable yellowish solid. Suitable yellow/green
crystals for X-ray analysis were obtained upon layering hexanes
over a THF solution of 8. [R]D ) -34 (c ) 0.0016, CHCl3). IR
(KBr): νmax ) 2931, 2854, 1450, 1264, 1141, 1057, 1028 cm-1. 1H
NMR (400 MHz, CDCl3): δ 0.93-1.38 (m, 18H), 1.46 (s, 18H),
1.73-1.86 (m, 18H), 2.06-2.06 (m,2H), 2.11-2.20 (m, 4H),
2.36-2.48 (m, 2H), 4.29 (dt, J ) 6 and 16 Hz, 2H), 4.46 (dt, J )
4 and 16 Hz, 2H), 7.35-7.46 (m, 10H) ppm. 13C NMR (100 MHz,
{31P}, CDCl3): δ 25.2 (3 × CH3), 25.59 (CH2), 26.64 (CH2), 26.3
(CH2), 26.7 (CH2), 27.4 (CH2), 27.5 (CH2), 27.9 (CH2), 29.8 (CH2),
30.6 (CH2), 34.7 (CH), 39.5 (CH), 51.5 (CH2), 69.5 (C), 128.8 (CH),
129.1 (CH), 129.2 (CH), 135.3 (C) ppm. 31P NMR (121 MHz,
CDCl3): 86.2 (d, J ) 121 Hz) ppm. MS (ESI, H2O/CH3CN (1:1)
1% formic) m/z: 917.38 [59, (M - OTf)+], 121 [100%]. HRMS
(ESI): calcd. for C46H76N2O2P2S2Rh+, 917.3878; found, 917.3858.
(SR)-cis-[Rh(PNSO)2][TfO] (9). Solid complex 8 was suspended
in dry THF under nitrogen and stirred overnight at room temper-
ature. Solvent removal under vacuum afforded pure cis-Rh complex
9 as a yellow solid. Suitable crystals for X-ray analysis were
obtained by layering hexanes over a solution of 9 in THF. [R]D )
-43 (c ) 0.0021, CHCl3). IR (KBr): νmax ) 2931, 2854, 1450,
[51, (MH - OTf)+], 917 [100, (M - OTf)+]. HRMS (ESI): calcd.
for C46H76N2O2P2S2Rh+, 917.3878; found, 917.3879. Anal. calcd.
for C47H76F3N2O5P2RhS3 + 1/2CH2Cl2: C, 51.41; H, 6.99; N, 2.52.
Found: C, 51.34; H, 6.97; N, 2.53.
(SR)-[Rh(PNSO)(PPh3)2][TfO] (10). A Schlenk tube containing
complex 7 (0.05 g, 0.066 mmol), PPh3 (0.034 g, 0132 mmol) and
dry dichloromethane was stirred under nitrogen for 1 h at room
temperature. The resulting yellow-orange solution was poured over
vigorously stirred hexane (50 mL), and the precipitate was filtered
and washed with hexane (2 × 10 mL). The yellow solid was dried
over high vacuum to provide 0.071 g (91%) of complex 10. [R]D
) -16 (c ) 0.1, CHCl3). IR (KBr): νmax ) 1436, 1272, 1148, 1093,
1031 cm-1. 1H NMR (400 MHz, CDCl3): δ 1.12 (s, 9H), 4.05 (dd,
J ) 17 Hz and JP ) 12 Hz, 1H), 4.40 (dd, J ) 17 Hz and JP ) 12
Hz, 1H), 6.40 (d, J ) 7 Hz, 2H), 6.89-6.95 (m, 7H), 7.02-7.09
(m, 7H), 7.17-7.27 (m, overlap with CDCl3 peak, 9H), 7.31-7.48
(m, 17H), 7.50-7.61(m, 2H), 7.64-7.69 (m, 1H) ppm. 31P NMR
(121 MHz, CDCl3): 23.7 (ddd, JPPcis ) 37.2, JPRh ) 138.8, JPPtrans
) 279 MHz), 32.2 (dt, JPPcis ) 37.2, JPRh ) 155.8), 63.6 (ddd,
JPPcis ) 37.3, JPRh ) 127.7, JPPtrans ) 278.9 MHz) ppm. MS (ESI,
H2O/CH3CN (1:1) 1% formic) m/z: 760 [26, (M - PPh3)+], 801
[100], 1022 [8, (M)+]. HRMS (ESI): calcd. for C59H56NOP3SRh,
1022.2345; found, 1022.2341.
Acknowledgment. We thank MEC (CTQ2005-623) and
Enantia S. L. for financial support. T.A. thanks Enantia for
financial support. We thank Prof. Anton Vidal-Ferran and
Dr. Elisenda Reixach (ICIQ) for helpful assistance in 31P
decoupled NMR experiments.
1
1267, 1151, 1123, 1031 cm-1. H NMR (400 MHz, CDCl3): δ
0.82-1.26 (m, 16H), 1.63 (s, 18H), 1.70-1.81 (m, 20H), 1.96-2.03
(m, 2H), 2.15 (br,2H), 2.33 (br, 2H), 4.41 (m, 4H), 7.35-7.42 (m,
6H), 7.445-7.48 (m, 4H) ppm. 13C NMR (100 MHz, {31P}, CDCl3):
δ 25.5 (CH2), 25.72 (CH2), 25.77 (3 × CH3), 26.0 (CH2), 26.2
(CH2), 27.0 (CH2), 28.4 (CH2), 29.9 (CH2), 30.2 (CH2), 35.3 (CH),
39.6 (CH), 51.6 (CH2), 71.2 (C), 128.9 (CH), 129.2 (CH), 129.4
(CH), 135.1 (C) ppm. 31P NMR (121 MHz, CDCl3): 95.5 (d, J )
135 Hz) ppm. MS (ESI, H2O/CH3CN (1:1) 1% formic) m/z: 918
Supporting Information Available: General methods and NMR
spectra for ligand 4-BH3 and complexes 6, 7, 8, and 9. X-ray crystal
data with a complete numbering scheme, atomic distances and
angles for 5, 7, 6, and 8. This material is available free of charge
via the Internet at http://pubs.acs.org.
OM8006823