Platinum group thiophenoxyimine complexes: syntheses and crystallographic/computational studies

Article abstract of DOI:10.1021/om0609057

Monomeric thiosalicylaldiminate complexes of rhodium(I) and iridium(I) were prepared by ligand transfer from the homoleptic zinc(II) species. In the presence of strongly donating ligands, the iridium complexes undergo insertion of the metal into the imine carbon-hydrogen bond. Thiophenoxyketimines were prepared by nontemplated reaction of o-mercaptoacetophenone with anilines and were complexed with rhodium(I), iridium(I), nickel(II), and platinum(II). X-ray crystallographic studies showed that while the thiosalicylaldiminate complexes display planar ligand conformations, those of the thiophenoxyketiminates are strongly distorted. Results of a computational study predicted that all synthetically accessible thiophenoxyketiminates will display distorted geometries.

Full text of DOI:10.1021/om0609057

Organometallics 2007, 26, 897-909
897
Platinum Group Thiophenoxyimine Complexes: Syntheses and
Crystallographic/Computational Studies
Jamin L. Krinsky, John Arnold,* and Robert G. Bergman*
Department of Chemistry, UniVersity of California, Berkeley, California, 94720
ReceiVed October 3, 2006
Monomeric thiosalicylaldiminate complexes of rhodium(I) and iridium(I) were prepared by ligand
transfer from the homoleptic zinc(II) species. In the presence of strongly donating ligands, the iridium
complexes undergo insertion of the metal into the imine carbon-hydrogen bond. Thiophenoxyketimines
were prepared by nontemplated reaction of o-mercaptoacetophenone with anilines and were complexed
with rhodium(I), iridium(I), nickel(II), and platinum(II). X-ray crystallographic studies showed that while
the thiosalicylaldiminate complexes display planar ligand conformations, those of the thiophenoxyketimi-
nates are strongly distorted. Results of a computational study predicted that all synthetically accessible
thiophenoxyketiminates will display distorted geometries.
Introduction
Transition metal complexes bearing salicylaldiminate ligands
have been employed successfully in a wide variety of catalytic
systems. Research regarding this ligand class has been greatly
facilitated by their ease of synthesis: A large range of structures
can be accessed, often requiring only one synthetic step from
the many commercially available salicylaldehydes and amines
Figure 1. a. Jacobsen epoxidation catalyst (salen-type ligand) b.
Olefin polymerization catalyst (bidentate salicylaldiminate ligand).
(eq 1). Jacobsen et al. have developed a highly efficient olefin
molecule reactivity has been explored.6 However, ligand lability
of the bidentate salicylaldiminates has limited their utility as
supporting ligands.7
In place of the phenolate functionality of salicylaldiminates,
thiosalicylaldiminates possess a thiophenolate moiety which
might be expected to form a relatively inert bond with soft,
polarizable metals. Homoleptic, first-row transition and main-
group element complexes are known for iron through copper
and the zinc triad, and incorporate bidentate, tridentate, and
tetradentate ligand variants;8 in addition, a few palladium species
have been prepared.9 While these complexes have been known
for some time, their behavior as supporting ligands during metal-
centered transformations is largely unexplored.10 A major
obstacle to their investigation has been the instability11 of free
epoxidation system employing a tetradentate (salen-type) deriva-
tive (Figure 1a).1 Titanium and zirconium complexes of biden-
tate salicylaldiminates have proven to be extremely active
catalysts in the production of polyolefins.2 Grubbs and others
have further extended the utility of these ligands with respect
to olefin polymerization, developing neutral, functional-group-
tolerant nickel catalysts (Figure 1b).3 These late transition metal
catalysts, however, tend to undergo deactivating ligand redis-
tribution reactions unless very sterically demanding salicyl-
aldiminates are used.4 Examples of heavier platinum-group metal
complexes are present in the literature,5 and some small-
(6) (a) Chatterjee, D.; Mitra, A.; Roy, B. C. React. Kinet. Catal. Lett.
2000, 70, 147-151; (b) Anderson, D. J.; Eisenberg, R. Organometallics
1996, 15, 1697-1706; (c) El-Hendawy, A. M.; Alkubaisi, A. H.; El-
Kourashy, A. E.; Shanab, M. M. Polyhedron 1993, 12, 2343-2350; (d)
Pasini, A.; Caldirola, C.; Colombo, A.; Ghilotti, M. J. Organomet. Chem.
1988, 345, 201-208.
(7) (a) Sariego, R.; Carkovic, I.; Martinez, M. Transition Met. Chem.
1984, 9, 106-108; (b) Mague, J. T.; Nutt, M. O. J. Organomet. Chem.
1979, 166, 63-77.
(1) Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.; Deng, L. J.
Am. Chem. Soc. 1991, 113, 7063-7064.
(2) (a) Matsui, S.; Mitani, M.; Saito, J.; Tohi, Y.; Makio, H.; Matsukawa,
N.; Takagi, Y.; Tsuru, K.; Nitabaru, M.; Nakano, T.; Tanaka, H.; Kashiwa,
N.; Fujita, T. J. Am. Chem. Soc. 2001, 123, 6847-6856; (b) Mitani, M.;
Furuyama, R.; Mohri, J.; Saito, J.; Ishii, S.; Terao, H.; Kashiwa, N.; Fujita,
T. J. Am. Chem. Soc. 2002, 124, 7888-7889; (c) Hustad, P. D.; Coates, G.
W. J. Am. Chem. Soc. 2002, 124, 11578-11579.
(3) (a) Wang, C. M.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs,
R. H.; Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149-
3151; (b) Younkin, T. R.; Conner, E. F.; Henderson, J. I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleben, D. A. Science 2000, 287, 460-462.
(4) Connor, E. F.; Younkin, T. R.; Henderson, J. I.; Waltman, A. W.;
Grubbs, R. H. Chem. Commun. 2003, 2272-2273.
(5) (a) Kerber, W. D.; Nelsen, D. L.; White, P. S.; Gagne, M. R. Dalton
2005, 1948-1951; (b) Alteparmakian, V.; Robinson, S. D. Inorg. Chim.
Acta 1986, 116, L37-L38; (c) Leipoldt, J. G.; Basson, S. S.; Grobler, E.
C.; Roodt, A. Inorg. Chim. Acta 1985, 99, 13-17; (d) Bonnaire, R.; Potvin,
C.; Manoli, J. M. Inorg. Chim. Acta 1980, 45, L255-L256.
(8) (a) Marin-Becerra, A.; Stenson, P. A.; McMaster, J.; Blake, A. J.;
Wilson, C.; Schroder, M. Eur. J. Inorg. Chem. 2003, 2389-2392; (b)
Kaasjager, V. E.; Puglisi, L.; Bouwman, E.; Driessen, W. L.; Reedijk, J.
Inorg. Chim. Acta 2000, 310, 183-190; (c) Mugesh, G.; Singh, H. B.;
Butcher, R. J. Eur. J. Inorg. Chem. 1999, 1229-1236; (d) Goswami, N.;
Eichhorn, D. M. Inorg. Chem. 1999, 38, 4329-4333; (e) Marini, P. J.;
Berry, K. J.; Murray, K. S.; West, B. O.; Irving, M.; Clark, P. E. J. Chem.
Soc., Dalton Trans. 1983, 879-884.
(9) (a) Hoskins, B. F.; Robson, R.; Williams, G. A.; Wilson, J. C. Inorg.
Chem. 1991, 30, 4160-4166; (b) Hoskins, B. F.; McKenzie, C. J.; Robson,
R.; Lu, Z. R. J. Chem. Soc., Dalton Trans. 1990, 2637-2641.
10.1021/om0609057 CCC: $37.00 © 2007 American Chemical Society
Publication on Web 01/17/2007

898 Organometallics, Vol. 26, No. 4, 2007
Krinsky et al.
Scheme 1
Scheme 2
Figure 2. ORTEP diagram of 4 at 50% probability (H atoms
omitted for clarity). Selected bond lengths (Å) and angles (deg):
Ir1-S1, 2.275(1); Ir1-N1, 2.081(3); Ir1-C20, 2.141(4); Ir1-C21,
2.149(4); Ir1-C24, 2.148(4); Ir1-C25, 2.138(5); S1-C1, 1.730-
(4); N1-C7, 1.304(5); C6-C7, 1.436(5); S1-Ir1-N1, 92.74(8);
N1-C7-C6-C5, 173.7(4).
dium with sodium 2-formylbenzenethiolate resulted in a product
mixture that appeared by 1H NMR to contain the desired iridium
thiolate. However, this species decomposed over 1 h at ambient
temperature in solution, precluding its use as a synthon for
further elaboration. Since the templated syntheses of the desired
ligands were already proven effective when starting from the
stable homoleptic zinc(II) aldehyde complexes,14 it was postu-
lated that reaction of thiosalicylaldiminate zinc species with
platinum-group chlorides might provide an alternate synthetic
route, much as dialkylzinc reagents alkylate platinum group
chlorides (Scheme 2). Indeed, the bis(thiosalicylaldiminate)zinc
complexes cleanly exchange both ligands with rhodium(I) and
iridium(I) chlorides, generating zinc chloride as an easily
removable byproduct. Two ligands of differing steric demand
were synthesized by the above templated route, one from 2,6-
thiosalicylaldimines: condensation of 2-mercaptobenzaldehyde
with amines frequently results in the formation of dithiocin
derivatives (Scheme 1a).12 This limitation can be overcome
using a templated synthetic approach wherein a preformed
metal-aldehyde complex is treated with the appropriate amine
(Scheme 1b).13 However, the approach is limited to cases where
the metal-aldehyde complex precursor is stable, which we have
found excludes the second- and third-row platinum group metals
in low oxidation states. Reported herein are rational synthetic
routes to heteroleptic thiosalicylaldiminate and thiophenoxy-
ketimine complexes of rhodium, iridium, nickel, and platinum.
The first two sections describe the synthesis and reactivity of
new thiosalicylaldiminate complexes and highlight some issues
of ligand stability. Sections 3 through 5 concern the preparation
of more robust thiophenoxyketimines, their complexation with
transition metal synthons, and some ligand substitution chemistry
of the resulting complexes. Following that is a crystallographic/
computational study that correlates ligand steric factors with
their coordination behavior. Finally, a modified aryl thiophen-
oxyketimine is presented which displays a striking flexibility
in its metal-coordination chemistry.
diisopropylaniline (prepared as L1 Zn (1)) and one from
2
p-anisidine (prepared as L2 Zn (2)).15 Both 1 and 2 undergo
2
reaction with [M(cod)Cl]2 to form L1M(cod) (M ) Rh (3), Ir
(4)) and L2M(cod) (M ) Rh (5), Ir (6)16) in isolated yields
ranging from 27% to 96%. All are stable, slightly air-sensitive
solids which decompose in solution upon extended exposure
to ambient atmosphere. The solids can be stored indefinitely
under inert atmosphere and at ambient temperature. NMR
spectra of each show only a single species present in solution
that displays Cs symmetry. The solid-state structures of 4, 5
and 6 were determined by X-ray diffraction analysis and are
shown in Figures 2-4.17
Results and Discussion
1. Preparation and Structural Properties of Thiosalicyl-
aldiminate Complexes. Early in our investigations it became
evident that established synthetic routes are not amenable to
the preparation of late transition metal thiosalicylaldiminate
complexes. Treatment of dichlorobis(1,5-cyclooctadiene)diiri-
The monomeric nature of complexes 4, 5, and 6 is somewhat
surprising, given the tendency of sterically unencumbered aryl
sulfides to bridge two metal centers through the sulfur atom.18
This structural feature is particularly noteworthy given the
(10) (a) Fallon, G. D.; Nichols, P. J.; West, B. O. J. Chem. Soc. Dalton
Trans. 1986, 2271-2276; (b) Marini, P. J.; Murray, K. S.; West, B. O. J.
Chem. Soc. Dalton Trans. 1983, 143-151; (c) Marini, P. J.; Murray, K.
S.; West, B. O. J. Chem. Soc., Chem. Commun. 1981, 726-728.
(11) Akine, S.; Nabeshima, T. Inorg. Chem. 2005, 44, 1205-1207.
(12) Corrigan, M. F.; Rae, I. D.; West, B. O. Aust. J. Chem. 1978, 31,
587-594.
(13) (a) Frydendahl, H.; Toftlund, H.; Becher, J.; Dutton, J. C.; Murray,
K. S.; Taylor, L. F.; Anderson, O. P.; Tiekink, E. R. T. Inorg. Chem. 1995,
34, 4467-4476; (b) Other methods available, see Garnovskii, A. D.;
Nivorozhkin, A. L.; Minkin, V. I. Coord. Chem. ReV. 1993, 126, 1-69.
(14) Olekhnovich, L. P.; Kurbatov, V. P.; Osipov, O. A.; Minkina, L.
S.; Minkin, V. I. J. Gen. Chem. USSR 1968, 38, 2512.
(15) An iron complex of L2 has previously been prepared; see ref. 10b.
(16) Ir(cod)2Cl was used as the iridium source in our reported synthetic
procedure, giving identical results.
(17) Due to the number of structures present in this report, comments
will be made only on structural features pertinent to this study. CIF-format
files for all structures can be found in the Supporting Information.
(18) Duff, S. E.; Barclay, J. E.; Davies, S. C.; Hitchcock, P. B.; Evans,
D. J. Eur. J. Inorg. Chem. 2005, 4527-4532.

Platinum Group Thiophenoxyimine Complexes
Organometallics, Vol. 26, No. 4, 2007 899
ligands. Surprisingly, no free cod is observed. Instead, reso-
nances exist that correspond to symmetrically bound cod (eq
2). These cod resonances are quite broad, suggesting fluxional
behavior at ambient temperature. The products arising from 4
and 6 were therefore assigned the formulas L1Ir(cod)(PMe3)2
(7) and L2Ir(cod)(PMe3)2 (8), respectively. Proton-decoupled 31
P
Figure 3. ORTEP diagram of 5 at 50% probability (H atoms
omitted for clarity). Selected bond lengths (Å) and angles (deg):
Rh1-S1, 2.2755(6); Rh-N1, 2.098(2); Rh1-C15, 2.160(2); Rh1-
C16, 2.175; Rh1-C19, 2.147(2); Rh1-C20, 2.165(2); S1-C1,
1.727(2); N1-C7, 1.311(3); C2-C7, 1.436(3); S1-Rh1-N1,
92.68(5); N1-C7-C2-C3, 178.8(2).
NMR spectra show only a single phosphine resonance for each
species. The structural motif that is most consistent with these
data is a square pyramidal geometry with two phosphine ligands
and cod occupying the basal positions and the thiolate sulfur
occupying the apical position. Thus the thiosalicylaldiminate
ligand is probably bound in a monodentate fashion.
Phosphine complex 8 decomposed to a complex mixture of
unidentified compounds over a period of about 1 day in benzene
solution or upon attempted isolation. While 7 was equally
unstable, its decomposition products were 4 (regenerated) and
the new phosphine-containing compound 9. Heating a mixture
of 4 and 9 for 30 min in the presence of added trimethylphos-
phine resulted in the clean conversion of the remaining 4 to 9.
Heating a mixture of 4 and three equivalents of the phosphine
1
also cleanly generated 9. Compound 9 exhibits an H NMR
signal consisting of an apparent doublet of triplets centered at
-10.75 ppm (J ) 156 Hz, 18.2 Hz) that we attributed to a metal
hydride species. The absence of a signal for an imine proton
suggested that site as the source of the hydride. Proton and
proton-decoupled 31P NMR spectra are consistent with the
presence of three nonequivalent phosphine ligands. These data
support the assignment of 9 as the iridium(III) species (L1-H)-
Ir(PMe3)3H where L1-H is the product of oxidative addition of
the imine C-H bond of L1 to the iridium(I) center of
intermediate 7 (eq 3). It is likely that 8 also undergoes this
Figure 4. ORTEP diagram of 6 at 50% probability (H atoms
omitted for clarity). Selected bond lengths (Å) and angles (deg):
Ir1-S1, 2.268(2); Ir1-N1, 2.077(7); Ir1-C15, 2.146(9); Ir1-C16,
2.133(9); Ir1-C19, 2.141(9); Ir1-C20, 2.162(9); S1-C1, 1.734-
(9); N1-C7, 1.32(1); C2-C7, 1.45(1); S1-Ir1-N1, 93.5(2); N1-
C7-C2-C3, 177.8(10).
absence of any sterically demanding substituents on L2. The
ligand backbone in each structure is essentially planar, with the
metal lying close to this approximate plane. In the case of 6,
the greatest deviation from the least-squares plane defined by
the iridium center and the atoms making up the chelate ring is
0.033(9) Å. Compound 4 displays a slight bending of the iridium
center out of the plane defined by the chelate ring, a distortion
common among complexes containing π-conjugated chelating
ligands such as â-diketiminates.19 The thiophenoxy-imine C-C
bond lengths of 4, 5, and 6 are 1.436(5), 1.436(3), and 1.448-
(10) Å, respectively. These are contracted relative to those of a
carbon-carbon single bond and thus are consistent with the
conclusion that L1 and L2 behave as rigid, π-conjugated ligands.
2. Reaction of Iridium Complexes with Trimethylphos-
phine. Attempts to exchange the coordinated cod ligands on 4
or 6 with trimethylphosphine did not yield the expected bis-
phosphine complexes. Treatment of 4 or 6 with excess tri-
methylphosphine at room temperature resulted in the immediate
formation of yellow solutions having 1H NMR spectra consistent
with the coordination of two geometrically equivalent phosphine
transformation, but the presumed iridium(III) product apparently
reacts further to generate the complex mixture observed.
Structural elucidation of 9 by X-ray diffraction analysis
confirmed the assignment made on the basis of spectroscopic
data. The structure depicted in Figure 5 possesses a distorted
octahedral geometry, with the atoms of the first coordination
sphere bent toward the site occupied by the hydride ligand (not
located). The P1-Ir1-P2 and P1-Ir1-P3 bond angles of
96.69(5) and 100.35(5)°, respectively, illustrate this distortion.
The L1-H chelate backbone is approximately coplanar with the
plane defined by Ir1, S1, and C7. The orientation of the N-aryl
ring explains the presence of two sets of isopropyl resonances
1
in the H NMR spectra of 9. One isopropyl group is oriented
near the phosphine ligand containing P1, rendering its two
methyl groups diastereotopic. The other is situated relatively
far from the sterically nondemanding hydride ligand. This
effectively results in a local mirror symmetry and thus its methyl
substituents are spectroscopically equivalent. Rotation around
the N-Ar bond is apparently slow on the NMR time scale.
(19) (a) Nikiforov, G. B.; Roesky, H. W.; Magull, J.; Labahn, T.; Vidovic,
D.; Noltemeyer, M.; Schmidt, H. G.; Hosmane, N. S. Polyhedron 2003,
22, 2669-2681; (b) Basuli, F.; Bailey, B. C.; Tomaszewski, J.; Huffman,
J. C.; Mindiola, D. J. J. Am. Chem. Soc. 2003, 125, 6052-6053.

900 Organometallics, Vol. 26, No. 4, 2007
Krinsky et al.
Figure 6. ORTEP diagram of 10 at 50% probability level.
Calculated H atoms omitted for clarity. Selected bond lengths (Å)
and angles (deg): S1-C1, 1.739(2); N1-H1, 1.04(3); N1-C7,
1.308(3); C1-C2, 1.418(3); C2-C3, 1.360(3); C3-C4, 1.398(3);
C4-C5, 1.371(3); C5-C6, 1.407(3); C1-C6, 1.444(3); C6-C7,
1.456(3); S1-C1-C2, 117.0(2); S1-C1-C6, 126.5(2); C5-C6-
C7-C8, 0.2(3); C7-N1-C9-C14, 49.5(3).
Figure 5. ORTEP diagram of 9 at 50% probability (H atoms
omitted for clarity). Hydride not located. Selected bond lengths (Å)
and angles (deg): Ir1-S1, 2.380(1); Ir1-P1, 2.351(2); Ir1-P2,
2.288(1); Ir1-P3, 2.337(2); Ir1-C7, 2.097(5); S1-C1, 1.736(5);
S1-Ir1-P1, 90.07(5); S1-Ir1-P2, 171.68(5); S1-Ir1-P3, 87.84-
(5); S1-Ir1-C7, 84.2(1); P1-Ir1-P2, 96.69(5); P1-Ir1-P3,
100.35(5); P1-Ir1-C7, 94.1(1).
C6, and C1-C6 bond lengths are 1.418(3), 1.407(3), and 1.444-
(3) Å, respectively. These bond lengths are considerably greater
than those of a completely delocalized phenyl ring, which in
this structure average 1.388 Å for the N1-bound aryl ring. The
C2-C3 and C4-C5 bond lengths are significantly contracted
(1.360(3) and 1.371(3) Å, respectively) and the C1-S1 bond
length of 1.739(2) Å is somewhat short, lending support for
contribution of the thioketone resonance form. The iminium-
thiolate backbone is essentially planar, with a greatest deviation
of 0.012(2) Å (C7) from the least-squares plane defined by S1,
C1, C6, C7, and N1. The N1-bound aryl group is rotated by
49.5(3)° with respect to this plane and thus is out of conjugation
with the rest of the π-electron system.
4. Synthesis of Thiophenoxyketimine Complexes. The alkali
metal salts of L3 (generated in situ from 10) reacted with a
variety of metal halides to form the corresponding thiophenox-
yketimine complexes. Complexes of formulas L3M(cod) (M )
Rh (12), Ir (13)) and L3M(PPh3)nX (M ) Ni, X ) Br, n ) 1
(14); M ) Pt, X ) Cl, n ) 1 (15), 2 (16)) were prepared from
[M(cod)Cl]2 and (Ph3P)2MX2, respectively. The iridium system
yielded multiple products when the reaction was scaled past 50
mg of 10. However, when Ir(cod)2Cl was used in place of [Ir-
(cod)Cl]2 the reaction proceeded to high yield (by NMR) on a
200 mg scale although separation of L3Ir(cod) (13) from the
small percentage of byproducts drastically reduced the isolated
yield. The known compound Ir(cod)2Cl exists as a mixture of
[Ir(cod)Cl]2 and free cod in THF solution, hence the increase
in selectivity must be due in some way to the presence of the
additional free cod. Stabilization of a reactive intermediate
during the ligand substitution process is a possible explanation
for this observation.
Scheme 3
3. Preparation of Thiophenoxyketimines. The observed
proclivity toward imine C-H bond insertion of the thiosalicyl-
aldiminates prompted us to explore derivatives of this ligand
architecture which might be less prone to unwanted reactivity
with the metal center. The most obvious derivatization is
replacement of the aldimine hydrogen with a methyl group,
affording the simplest ketimine (Scheme 3). To our knowledge
only a few previous examples of this ligand class exist, all being
of the tetradentate type prepared by templated routes discussed
previously.20 We found that the ketimines derived from sterically
unhindered anilines can be prepared directly and in good yield
if the reaction mixtures are kept at ambient temperature.
Compounds L3H (10, Ar ) p-C6H4Me) and L4H (11, Ar )
p-C6H4OMe) are bright red, crystalline solids that are stable in
the solid state and in solution provided they are stored under
inert atmosphere. Like the free thiosalicylaldimines that have
been successfully isolated,12 they display extremely downfield-
shifted thiol resonances in their 1H NMR spectra (17.3 and 18.2
ppm, respectively) which are substantially outside of the typical
diamagnetic region.
X-ray diffraction studies on a sample of 10 revealed that the
thiol proton is associated with the imine nitrogen, and the
structure is better classified as intermediate between a thiolate
zwitterion and a thioketone (Figure 6). This configuration had
previously been proposed for the free thiosalicylaldimines based
on spectroscopic data.21 Iminium hydrogen H1 was located and
its positional coordinates refined, while the other hydrogen
atoms were placed in calculated positions. The C1-C2, C5-
In the case where M ) Ni, one phosphine ligand was cleanly
displaced by L3 to produce L3Ni(PPh3)Br (14) quantitatively.
However, incomplete phosphine displacement was observed
with the platinum analogue, resulting in an equilibrium mixture
of L3Pt(PPh3)Cl (15) and L3Pt(PPh3)2Cl (16). In the case of 16,
L3 is presumably bound only through the thiophenylate sulfur.
The proposed structure is analogous to that bearing a similar
ligand which was crystallographically determined (see below).
Reasonably pure 16 was isolated by concentrating a THF
solution of the crude mixture in the presence of excess
triphenylphosphine (analytically pure samples could not be
obtained; NMR spectra are available as Supporting Information)
(20) (a) Karsten, P.; Strahle, J. Acta Crystallogr. Sect. C: Cryst. Struct.
Commun. 1999, C55, 488-489; (b) Coombes, R. C.; Costes, J. P.; Fenton,
D. E. Inorg. Chim. Acta 1983, 77, L173-L174.
(21) Corrigan, M. F.; West, B. O. Aust. J. Chem. 1976, 29, 1413-1427.

Platinum Group Thiophenoxyimine Complexes
Organometallics, Vol. 26, No. 4, 2007 901
Scheme 4
(Scheme 4). While heating resolvated 16 generated only a small
amount of 15, slow crystallization from the crude mixture (by
pentane vapor diffusion) yielded crystalline 15 only. This
observation suggests that crystal packing forces are sufficient
to offset the energy increase in forming 15. Surprisingly, pure
15 was also prepared by repeatedly washing solid 16 with Et2O.
The use of platinum starting materials containing more easily
displaced ligands resulted only in intractable mixtures: no
desired product was observed when (PhCN)2PtCl2 or (Me2S)2-
PtCl2 were treated with L3Na.
Figure 7. ORTEP diagram of 19 at 50% probability level. H atoms
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Pt1-S1, 2.391(1); Pt1-N1, 2.198(4); Pt1-C16, 2.060(5); Pt1-
C17, 2.071(5); Pt1-C18, 2.052; S1-C1, 1.754(5); C2-C7, 1.490-
(7); S1-Pt1-N1, 89.6(1); S1-Pt1-C16, 95.1(2); S1-Pt1-C17,
174.5(2); S1-Pt1-C18, 89.4(2); N1-Pt1-C18, 178.4(2); C3-
C2-C7-C8, 32.5(7).
In order to verify that classical organometallic complexes
bearing thiophenoxyketimine ligands can exist, alkyl substituents
were then investigated. Attempted alkylations of 14 with MeLi,
MeMgBr, or Me2Zn all resulted in decomposition to intractable
material. This route seems to be generally avoided with respect
to the salicylaldiminate complexes, suggesting a lack of compat-
ibility with the ligand’s imine fragment. Due to the compara-
tively facile synthesis of alkylplatinum starting materials, these
were chosen for subsequent study. The reaction of (Ph3P)2-
PtMeCl with L3Na yielded multiple uncharacterized species.
While heating (Ph3P)2PtMe2 in the presence of 10 or 11 resulted
only in the decomposition of the ligand, (nbd)PtMe2 (nbd )
norbornadiene) reacted with 11 at ambient temperature in
acetonitrile (eq 4), spontaneously precipitating a modest yield
When L4 was employed in place of L3, a mixture of mono-
and bis-phosphine complexes was again isolated. However,
when (Ph3P)2PtCl2 was treated with L4Na at -78 °C under
extremely dilute conditions (2.5 mM), L4Pt(PPh3)Cl (17) and
triphenylphosphine were the principal observed products. This
observation implies that 17 is the kinetically favored product
of this reaction, while L4Pt(PPh3)2Cl (18) and 17 are of
comparable thermodynamic stability. This is puzzling if one
assumes the reaction proceeds via salt metathesis followed by
displacement of the phosphine by the imine functionality of L4.
The heterogeneous nature of the reaction mixture has so far
precluded kinetic study, and thus a satisfactory explanation for
this reactivity has been elusive. Interestingly, the most effective
method for isolation of pure 17 was the simple addition of
diethyl ether over an amorphous sample of the crude product.
Over the course of several hours all amorphous material was
converted to crystalline product, leaving in solution the tri-
phenylphosphine liberated during the reaction. The lattice energy
gained apparently drives the transformation from amorphism
to crystallinity through the small equilibrium concentration of
dissolved product.
Compounds 12-15 and 17 are stable, crystalline materials
that were unchanged after exposure to air for weeks in the solid
state. In solution, no decomposition was observed over several
months in the absence of oxygen. No ligand disproportionation
was observed when nickel complex 14 was heated at 75 °C in
benzene for 24 h; thus it appears that thiophenoxyketimines do
afford substitutionally inert ligands. With this in mind we wished
to prepare hydrocarbyl nickel complexes that might be reactive
toward olefins. However, treatment of (Ph3P)2NiPhCl with alkali
metal salts of L3 resulted only in intractable tars. There is
evidence in the literature that aryl substituents on group 10
metals can reductively couple with aryl sulfides.22 The coupling
product was not observed in the product mixture, although this
decomposition pathway cannot be ruled out, as this particular
diaryl sulfide might react further with the highly reactive
zerovalent nickel species generated by the coupling reaction.
of an impure material formulated as L4PtMe3 (19) on the basis
of 1H NMR spectra (available as Supporting Information).
Purification of this material has proven problematic, and thus
the complex is not fully characterized.
A few high-quality crystals of 19 formed when the reaction
mixture was allowed to stand, and this material was subjected
to X-ray crystallographic analysis. Complex 19 is a centrosym-
metric dimer wherein each sulfur atom bridges both platinum
centers and each imine nitrogen binds to one of the two platinum
centers (Figure 7). Interestingly, 19 is a structural analogue of
the known salicylaldiminate complex formulated as [(salnr)-
PtMe3]2.23 The geometry about the platinum(IV) center is only
slightly distorted from the idealized octahedron. The Pt-Pt
separation is 3.68 Å. Unlike the thiosalicylaldiminate complexes
described above, the chelating ligand backbone is nonplanar
with a twist about the C2-C7 bond of 32.5(7)°. The C2-C7
bond length is 1.490(7) Å which is significantly longer than
the analogous bond length of 1.456(3) Å in the free ligand 10.
(22) (a) Mann, G.; Baranano, D.; Hartwig, J. F.; Rheingold, A. L.; Guzei,
I. A. J. Am. Chem. Soc. 1998, 120, 9205-9219; (b) Osakada, K.; Maeda,
M.; Nakamura, Y.; Yamamoto, T.; Yamamoto, A. J. Chem. Soc. Chem.
Commun. 1986, 442-443.
(23) Hall, J. R.; Swile, G. A. Aust. J. Chem. 1975, 28, 1507-1511.

902 Organometallics, Vol. 26, No. 4, 2007
Krinsky et al.
Scheme 5
Thus the imine moiety and the thiophenylate ring are extensively
deconjugated. The platinum-bound methyl group containing C17
is situated over the nitrogen-bound aryl ring. The close proximity
of these two substituents apparently prevents rotation about the
N1-C9 bond, which explains the observation that all of the
1
Figure 8. ORTEP diagram of 12 at 50% probability. One of two
molecules in the asymmetric unit displayed. H atoms omitted for
clarity. Selected bond lengths (Å) and angles (deg): Rh1-S1,
2.314(2); Rh1-N1, 2.145(5); Rh1-C16, 2.183(7); Rh1-C17,
2.163(6); Rh1-C20, 2.125(7); Rh1-C21, 2.128(6); S1-C1, 1.744-
(6); N1-C7, 1.307(8); C2-C7, 1.477(9); S1-Rh1-N1, 88.2(1);
C3-C2-C7-C8, 35.0(7).
aryl proton resonances (for each asymmetric unit) in the H
NMR spectrum of 19 are nonequivalent.
5. Reactivity of a Triarylphosphine with Thiophenox-
yketimine Complexes 12, 13. Rhodium complex 12 reacted
cleanly with two equivalents P(p-tol)3 to form L3Rh(P(p-tol)3)2
(20) over 8 h (Scheme 5). Monitoring the reaction by 1H NMR
(C6D6) revealed that no 12 was left in solution after 1 h. In
addition to 20 and free cod, a third species was present in
solution that gave rise to olefinic cod resonances at 5.09 and
3.45 ppm, one of which was significantly shifted relative to
that of 12 (4.64 and 3.45 ppm). Proton-decoupled 31P NMR
spectra of the mixture show resonances corresponding to 20,
free P(p-tol)3, and a doublet (JP-Rh ) 153.9 Hz) consistent with
a monophosphine rhodium complex. These data are consistent
with an intermediate wherein one phosphine is bound to rhodium
with retention of the cod ligand. The downfield cod resonance
is significantly upfield-shifted relative to free cod (5.58 ppm),
indicating that both olefinic moieties are still bound to rhodium.
Thus the intermediate is assigned as L3Rh(cod)P(p-tol)3 (21)
where L3 is bound only through the thiolate sulfur.
Complex 22, the iridium analogue of 21, was observed after
1 h in C6D6 and was the only species present in solution (other
1
than 1 equiv of free P(p-tol)3). Olefinic resonances in the H
NMR spectra occur at 5.07 and 3.89 ppm (compared with 4.54
and 3.05 ppm for 13). Phosphorus NMR spectra display only
two resonances in a one-to-one ratio, one corresponding to free
P(p-tol)3 and the other occurring at 22.10 ppm. No final product
analogous to 20 was observed after allowing the reaction to
proceed for 48 h. While this species appeared stable in solution
at ambient temperature, heating or attempted isolation yielded
intractable material.
The hemilabile nature of L3 indicates that this ligand behaves
more like a thiophenylate ligand with a tethered imine fragment
than a rigid, conjugated ligand such as a â-diketiminate. The
instability of the ligand toward alkylating agents supports this
conclusion, as the alkylation of â-diketiminate-bearing com-
plexes is a well-known transformation.24 Therefore we wished
to investigate structural aspects of these complexes that may
impart the observed reactivity.
Figure 9. ORTEP diagram of 14 at 50% probability. H atoms
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Br1-Ni1, 2.362(1); Ni1-S1, 2.149(2); Ni1-P1, 2.184(2); Ni1-
N1, 1.911(5); S1-C1, 1.759(7); N1-C7, 1.284(8); C6-C7, 1.473-
(9); Br1-Ni1-P1, 89.50(6); Br1-Ni1-N1, 93.8(2); S1-Ni1-P1,
90.77(7); S1-Ni1-N1, 89.9(2); C5-C6-C7-C8, 27.3(9).
subjected to X-ray crystallographic analysis, confirming the
structural assignments made by spectroscopic means. All are
square planar complexes with unremarkable bond lengths and
angles around the metal centers. The nickel-containing species
14 is slightly distorted toward a tetrahedral conformation.
Representative structures are depicted in Figures 8-10 (see
Schemes 4 and 5 for line drawings).
The most striking feature revealed by X-ray crystallographic
investigation is the lack of ligand planarity observed for the
ketimine complexes. Whereas the solid-state structures of the
6. Crystallographic/Computational Analysis of Ligand
Conformation. Several species bearing chelating L3 or L4 were
(24) Budzelaar, P. H. M.; van Oort, A. B.; Orpen, A. G. Eur. J. Inorg.
Chem. 1998, 1485-1494.

Platinum Group Thiophenoxyimine Complexes
Organometallics, Vol. 26, No. 4, 2007 903
Figure 12. ORTEP diagram of LRIr(cod) where R ) F. Optimized
structure at B3LYP/LACVP** level of theory.
Figure 10. ORTEP diagram of 15 at 50% probability. H atoms
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Pt1-Cl1, 2.333(2); Pt1-S1, 2.283(2); Pt1-P1, 2.220(2); Pt1-N1,
2.101(5); S1-C1, 1.779(7); N1-C7, 1.293(8); C6-C7, 1.473(9);
Cl1-Pt1-P1, 95.42(6); Cl1-Pt1-N1, 89.4(1); S1-Pt1-P1, 88.25-
(6); S1-Pt1-N1, 87.0(1); C5-C6-C7-C8, 37.3(9).
Figure 13. ORTEP diagram of LRIr(cod) where R ) NO2.
Figure 11. a. Side view of LRM(cod) defining φ, the cant between
aryl ring and metal coordination plane. b. Top view defining γ,
the torsion (twisting) angle between aryl ring and imine moiety.
Optimized structure at B3LYP/LACVP** level of theory.
agree very well with experiment.26 Because the most obvious
difference between the aldimines and ketimines is the increased
steric demand of the alkyl group, our investigation focused
primarily on geometry changes with varying imine substituent
size. All calculations were performed at the hybrid B3LYP/
LACVP** level of theory in the gas phase. As a starting point,
the structures of the crystallographically determined aldimine
and ketimine complexes were optimized and were found to agree
well with the experimentally observed structures.
Next, the structures of the rhodium and iridium aldimine
complexes were reoptimized after replacing the imine hydrogen
atom with several different substituents of intermediate steric
demand. The substituents chosen were fluoro, nitro, and phenyl,
under the assumption that the flat π-electron system of the latter
two would be oriented normal to the plane of the ligand
backbone. The optimized structures of LRIr(cod) where R ) F
and NO2 are shown in Figures 12 and 13, respectively. The
geometries are discussed in more detail below, but one can see
that in the case where R ) NO2, the plane defined by the atoms
of R is approximately perpendicular to the ligand backbone.
Because the ligands in question contain an unsaturated (imine)
fragment that can exist in conjugation with the thiophenylate
π-electron system, one would expect that there should be
resistance to breaking this conjugation. Therefore, little deviation
from planarity should be observed until the steric situation
becomes sufficiently problematic that ligand twisting is then
favorable. After this point, π-electron conjugation is lost and
the ligand is free to adopt the most favorable conformation based
solely on sterics. That is, while the ligand twist angle γ (Figure
aldimine complexes 4, 5, and 6 display approximate Cs
symmetry with the metal and chelating ligand backbone defining
the mirror plane, the ketimine analogues possess an extremely
nonplanar ligand conformation. In these structures the thiolate
phenyl ring is sharply canted with respect to the plane containing
the metal and atoms of the first coordination sphere (φ , 180°,
Figure 11a). The carbon-carbon bond connecting the thiophe-
nylate moiety with the imine fragment is twisted, bringing the
imine π-bond out of conjugation with the thiophenylate (γ .
0°, Figure 11b). The apparent lack of π-electron conjugation
may account for the reactivity trends discussed above. This
geometry is observed in the solid-state structures of some nickel
thiosalicylaldiminate complexes; however, there appears to be
no trend in substitution pattern that would explain why some
are planar and some twisted.8b,25 Thus it appears that in those
reported systems the lack of planarity is due primarily to
crystallographic packing forces.
We wished to investigate, by computational methods, the
source of the large differences in ligand conformation when
seemingly minor structural modifications are introduced. This
knowledge could then be used to ascertain whether a ketimine
could be synthesized that would have the planar, conjugated
conformation of the aldimines without the unwanted reactivity
of the aldimine C-H bond. While energies predicted by DFT
are often inaccurate compared to more sophisticated (and costly)
ab initio methods, calculated geometries have been shown to
(25) (a) Bouwman, E.; Henderson, R. K.; Powell, A. K.; Reedijk, J.;
Smeets, W. J. J.; Spek, A. L.; Veldman, N.; Wocadlo, S. J. Chem. Soc.
Dalton Trans. 1998, 3495-3499; (b) Christensen, A.; Jensen, H. S.; McKee,
V.; McKenzie, C. J.; Munch, M. Inorg. Chem. 1997, 36, 6080-6085.
(26) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger, M.
D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445-11459.

904 Organometallics, Vol. 26, No. 4, 2007
Krinsky et al.
Table 1. Computeda and Experimental Ligand Torsion
Anglesb in LRM(cod). Entries from Crystallographically
Determined Structures Bear Compound Numbers in
Boldface
entry
Ar
R
M
γ (deg)
1
2(5)
3
4(6)
5
6
7(4)
8
9
10
11
p-C6H4Me
p-C6H4OMe
p-C6H4Me
p-C6H4OMe
p-C6H4Me
p-C6H4Me
p-2,6-iPr2Ph
p-C6H4Me
p-C6H4Me
p-C6H4Me
p-C6H4Me
p-C6H4Me
p-C6H4Me
p-C6H4Me
p-C6H4Me
H
H
H
H
F
F
H
NO2
NO2
Ph
Me
Me
Me
Ph
Me
Ir
0.6
1.2
1.3
2.1
3.7
5.2
6.2
7.1
19.4
26.3
30.7
33(1)
33.8
34.6
35.0(7)
Rh
Rh
Ir
Ir
Rh
Ir
Ir
Rh
Ir
Ir
Ir
Figure 14. Selected region of the 1H NMR spectrum (ppm) of 12
in CD2Cl2, showing broadened cod resonances at 50 °C. Ligand
methyl resonances truncated.
12(13)
13
14
Rh
Rh
Rh
Scheme 6
15(12)
a B3LYP/LACVP**. b See Figure 11b for definition of γ.
11) is a function of the steric extent of R, it is not a linear
relationship but a step function.
Table 1 summarizes the results of the computational study,
which agree reasonably well with the predictions stated above.
Only a slight perturbation in geometry was observed where R
) F (entries 5, 6). An increase in γ accompanies replacement
of R with NO2, especially in the case of M ) Rh (entry 9)
where γ is approximately two-thirds of the value for the methyl
ketimine complexes. In the cases where R ) Ph (entries 10,
14), the γ parameters reach values comparable to those of the
methyl ketimine complexes, and thus the proposed barrier to
ligand twisting has been surpassed. Therefore, this study predicts
that the π-electron extent (normal to the plane containing the
atoms) of a nitro group is the size limit for any substituent R,
and anything larger will cause the ligand to adopt a nonplanar,
twisted conformation. It should be noted that this size limit
depends somewhat on the metal size, as a contraction of the
N-S distance could make room for a larger R. However, further
studies indicated that only in the extreme case where M ) H
(the free ligand) does this effect result in a planar methyl
ketimine. It follows that there is unlikely to be a synthetically
accessible thiophenoxyketimine that would exhibit a planar
conformation upon complexation with a transition metal.
Given that the methyl ketimine complexes prefer a twisted
ligand conformation, they should either have C1 symmetry in
solution or undergo a ring flip with some activation barrier.
Computations on 12 indicate that this barrier should be small,
with a planar configuration approximately 5 kcal/mol (not ZPE-
corrected) higher in energy than that of the twisted conformation.
tures investigated computationally, no synthetically feasible
ketimine was predicted to bind in a planar conformation. We
wondered if a highly electron-deficient arene might possess a
sufficiently contracted π-electron cloud that it could be used in
place of a nitro group, yielding a planar-binding ligand. Thus,
an aryl ketimine containing a 3,5-bis(trifluoromethyl)phenyl
moiety was synthesized. The aryl ketone precursor was prepared
in a manner analogous to that used for 2-mercaptobenzaldehyde
(Scheme 6). Thiocresol was deprotonated with n-butyllithium
and treated with Weinreb amide 23. The resulting ketone (24)
was converted to ketimine ligand L5H (25) in good yield by
treatment with p-toluidine in benzene at ambient temperature.
This ligand exhibited reactivity toward metal-complex precursors
similar to that of the methyl ketimine analogs discussed above.
Thus, L5Ni(PPh3)2Br (26), L5Pt(PPh3)2Cl (27), and [L5Rh(cod)]2
(28) were prepared by treatment of the appropriate metal com-
plex with L5K which was generated in situ from 25 (Scheme
7).
1
At ambient temperature or above, H NMR spectra of 12 and
13 exhibit somewhat broad, featureless cod resonances (Figure
14). These resonances sharpened upon cooling, resolving into
complex multiplets that may result from overlapping of non-
equivalent peaks (which would be the case for a C1-symmetric
complex). The peak broadening is decidedly small, but is not
surprising given the minor change in electronic environment
that the cod protons would experience during the ring inversion
process. This broadening was not observed in spectra of the
analogous (planar) aldimine complexes, a result which supports
the conclusion that it is due to a facile ring inversion. Despite
efforts to model this process by DFT, no planar transition state
was located using several methods. It is unclear whether this is
due to an extremely shallow potential energy surface or the
presence of a multistep pathway for ring inversion.
The solid-state structure of 26 depicted in Figure 15 shows
a twisted ligand conformation similar to those of the methyl
ketimine complexes. This datum is inconclusive with respect
to the conformational study (see above), as even the nickel
aldimine complexes display a variability in ligand conformation.
The more informative structures would be the platinum and
rhodium complexes. However, unlike the platinum systems
containing L3 or L4, crystallization of the crude product did not
yield L5Pt(PPh3)Cl. Instead, both phosphine ligands were
retained as shown in Figure 16.
X-ray structural investigation of 28 revealed that it is a C2-
symmetric dimer in the solid state, with the thiolate sulfur atoms
bridging the rhodium centers (Figure 17). The imine function-
alities are both uncoordinated. In solution, 28 is apparently in
fast exchange with its monomer and displays highly temperature-
dependent NMR resonances (spectra available in Supporting
7. Preparation and Structures of 3,5-Bis(trifluoromethyl)-
phenyl Ketimine Complexes. Of the various ligand architec-

Platinum Group Thiophenoxyimine Complexes
Organometallics, Vol. 26, No. 4, 2007 905
Table 2. Crystallographic Data for Compounds 4-6, 9, 10, 12-15, 19, and 26-28
compd no.
4
5
6
9
10
P212121
7.5851(5)
11.2611(7)
14.9642(9)
90
90
90
1278.2(1)
4
1.254
0.229
7021
1444
0.014
0.0323, 0.0433
0.0310, 0.0431
space group
a, Å
b, Å
P21/n
P1h
P1h
P1h
14.4457(9)
11.3894(7)
15.7491(9)
90
112.433(1)
90
8.9880(6)
10.3914(7)
11.1130(7)
94.331(1)
103.799(1)
105.673(1)
959.6(1)
2
8.9758(7)
10.4233(8)
11.0658(9)
93.684(1)
104.019(1)
106.112(1)
955.3(1)
2
9.4131(7)
11.4413(9)
15.924(1)
94.483(1)
89.982(1)
113.337(1)
1568.9(2)
2
c, Å
R, deg
â, deg
γ, deg
V, Å3
Z
2395.1(3)
4
F
calcd, g cm-3
1.655
5.691
12110
3446
1.569
1.007
5260
3160
1.887
7.127
4892
2643
1.517
4.503
8900
4336
µ(Mo KR), mm-1
no. rflns measd
no. indep rflns
Rint
0.024
0.016
0.038
0.025
R1, wR2a (all)
R1, wR2 (obsdb)
0.0285, 0.0280
0.0215, 0.0263
0.0261, 0.0311
0.0233, 0.0305
0.0378, 0.0337
0.0291, 0.0324
0.0383, 0.0485
compd no.
12
13
14
15
19
space group
a, Å
b, Å
P21/c
P21/c
Pbcn
P21/n
P21/c
18.801(1)
19.655(1)
10.9890(6)
90
106.274(1)
90
3898.1(4)
8
1.538
0.988
17237
5274
0.036
18.850(2)
19.622(2)
11.0043(9)
90
106.214(1)
90
3908.4(5)
8
1.838
6.964
16864
4300
0.053
22.475(2)
14.050(1)
17.948(2)
90
90
90
5667(1)
8
1.503
2.253
25445
2868
12.0768(8)
14.864(1)
17.228(1)
90
110.004(1)
90
2906.0(3)
4
1.676
5.051
13161
3517
0.047
8.2563(6)
21.618(2)
9.9255(7)
90
96.223(1)
90
1761.2(2)
4
1.873
8.055
8876
2538
0.023
c, Å
R, deg
â, deg
γ, deg
V, Å3
Z
F
calcd, g cm-3
µ(Mo KR), mm-1
no. rflns measd
no. indep rflns
Rint
0.077
R1, wR2 (all)
R1, wR2 (obsd)
0.0655, 0.0980
0.0524, 0.0853
0.0689, 0.0470
0.0353, 0.0384
0.1213, 0.0955
0.0510, 0.0500
0.0585, 0.0533
0.0334, 0.0375
0.0324, 0.0330
0.0229, 0.0298
compd no.
26
27
28
space group
a, Å
b, Å
P21/c
P(-1)
C2/c
13.5569(9)
14.7725(9)
22.151(2)
90
91.618(1)
90
4434.4(5)
4
1.454
1.486
22642
5358
0.039
13.492(1)
14.743(1)
15.712(1)
109.336(2)
102.825(1)
109.274(2)
2583.9(4)
2
26.474(1)
9.4643(5)
26.832(1)
90
118.924(1)
90
5884.2(5)
4
1.579
0.713
15000
4220
0.020
c, Å
R, deg
â, deg
γ, deg
V, Å3
Z
F
calcd, g cm-3
1.552
2.920
15022
6612
µ(Mo KR), mm-1
no. rflns measd
no. indep rflns
Rint
0.036
R1, wR2 (all)
R1, wR2 (obsd)
0.0652, 0.0537
0.0396, 0.0476
0.0574, 0.0465
0.0376, 0.0417
0.0375, 0.0427
0.0309, 0.0416
2
2
2
aR1 ) Σ||Fo| - |Fc||/Σ|Fo|; wR2 ) [Σ{w(Fo - Fc )2}/Σw(Fo )2]1/2
.
b I > 3.00σ(I).
1
Information). All H NMR resonances are broad, and only a
single broad peak is observed in the olefinic region. At -80 °C,
this region displays many peaks. It is not clear whether these
belong to the dimer and monomer or whether more than one
isomer of the dimer are present. The broad olefinic peak
observed at ambient temperature began to decoalesce into two
peaks at 70 °C, presumably belonging to the monomer.
From these data it appears that the electron-withdrawing
capabilities of the 3,5-bis(trifluoromethyl)phenyl fragment render
the imine nitrogen a poor donor relative to its methyl ketimine
analogues. This decrease in donor ability forces the ligand to
adopt alternative coordination modes when bound to the heavier
platinum group metals. Whether or not the arene substituent is
sufficiently compact to allow for planar ligand binding, the
observed conformational flexibility of L5 probably precludes
its use as a â-diketiminate analogue.
Conclusion
This paper describes a rational synthesis of thiosalicylaldi-
minate complexes containing heavy platinum group metals. The
ligands were prepared using zinc as a template and metal
chlorides were treated with the resulting complexes to yield the
target species in moderate to good yield. The iridium complexes
were found to undergo insertion of the metal center into the
imine C-H bond in the presence of strong donor ligands.
The thiophenoxyketimines synthesized in this study display
unexpected coordination behavior and in the solid state adopt
twisted conformations when bound to a metal. This twisting
apparently reduces π-electron conjugation of the thiophenylate
fragment with the imine functionality. A computational study
predicted that all synthetically accessible thiophenoxyketimines
would be too sterically demanding to allow a planar conforma-

906 Organometallics, Vol. 26, No. 4, 2007
Krinsky et al.
Figure 16. ORTEP diagram of 27 at 50% probability. H atoms
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Pt1-Cl1, 2.338(1); Pt1-S1, 2.362(2); Pt1-P1, 2.253(2); Pt1-P2,
2.281(2); S1-C1, 1.757(6); N1-C8, 1.278(7); C2-C8, 1.503(8);
Cl1-Pt1-S1, 88.07(5); Cl1-Pt1-P2, 90.51(5); P1-Pt1-P2,
98.58(6); P1-Pt1-S1, 82.77(5).
Figure 15. ORTEP diagram of 26 at 50% probability. H atoms
omitted for clarity. Selected bond lengths (Å) and angles (deg):
Ni1-Br1, 2.3554(6); Ni1-S1, 2.169(1); Ni1-P1, 2.157(1); Ni1-
N1, 1.940(3); S1-C1, 1.777(4); N1-C7, 1.283(5); C6-C7, 1.464-
(5); Br1-Ni1-P1, 90.21(3); Br1-Ni1-N1, 96.98(9); S1-Ni1-
P1, 90.67(4); S1-Ni1-N1, 89.42(10); C5-C6-C7-C9, 41.6(5).
Scheme 7
Figure 17. ORTEP diagram of 28 at 50% probability. H atoms
omitted for clarity. Selected bond lengths (Å) and angles (deg ):
Rh1-S1, 2.3887(8); Rh1-S1′, 2.3801(8); Rh1-C24, 2.147(3);
Rh1-C25, 2.148(3); Rh1-C28, 2.140(3); Rh1-C29, 2.137(3); S1-
C1, 1.791(3); N1-C8, 1.273(4); C6-C8, 1.510(4); S1-Rh1-S1′,
80.64(3); Rh1-S1-Rh1′, 94.19(3).
tion when metal-bound. In agreement with the computational
results, the reaction chemistry of these species indicates that
their behavior is consistent with the lack of a rigid, conjugated
ligand backbone. As imines are generally comparatively poor
ligands for heavy, soft metal centers it is therefore of little
surprise that the imine functionality of this ligand class should
be easily displaced by strongly binding ligands such as
phosphines. A corollary to this argument is that the imine
substituent would be expected to be much more susceptible to
nucleophilic attack than one that is part of a conjugated system
such as a â-diketiminate. The synthetic strategies now in place
will facilitate investigations into the catalytic utility of this ligand
class.
NMR tubes which were then flame-sealed under reduced pressure.
Solvents were dried by passage through a column of activated
alumina,27 degassed with nitrogen, and stored over 4 Å molecular
sieves. Nonchlorinated solvents were tested with Na/benzophenone
prior to use. Deuterated solvents were vacuum-transferred from Na/
benzophenone, except CD2Cl2 and CDCl3 which were vacuum-
transferred from CaH2. Molecular sieves were activated by heating
at 250 °C under high vacuum for 2 d. NMR spectra were obtained
with Bruker AMX-300, AMX-400, or DRX-500 spectrometers.
Chemical shifts were calibrated relative to residual solvent peaks
or external standards and are reported relative to TMS (1H, 13C),
85% H2PO4 (31P), and CFCl3 (19F). Infrared spectra were recorded
neat on a Thermo Nicolet Avatar 370 spectrometer equipped with
a Smart Performer ATR crystal. Elemental analyses were performed
Experimental Section
General. All manipulations were performed using standard
Schlenk-line techniques or an N2-atmosphere glove box. Reactions
followed by NMR spectroscopy were performed in 5 mm Pyrex
(27) Alaimo, P. J.; Peters, D. W.; Arnold, J.; Bergman, R. G. J. Chem.
Ed. 2001, 78, 64.

Platinum Group Thiophenoxyimine Complexes
Organometallics, Vol. 26, No. 4, 2007 907
at the UC, Berkeley Microanalytical Facility with a Perkin-Elmer
2400 Series II CHNO/S Analyzer. All starting materials obtained
from commercial sources were used without further purification.
2-Mercaptobenzaldehyde28 and its sodium salt,29 copper tert-butyl
sulfide,30 [Ir(cod)Cl]2,31 [Rh(cod)Cl]2,32 (nbd)PtMe2,33 (Ph3P)2-
PtMe2,34 (Me2S)2PtCl2,35 (Ph3P)2PtMeCl,36 and (Ph3P)2NiPhCl37
were prepared according to literature procedures. Ir(cod)2Cl was
prepared by cooling a concentrated solution of [Ir(cod)Cl]2 in neat
cod.38
remaining atoms are treated with Pople’s 6-31G** basis set,51 which
includes a set of d polarization functions for non-hydrogen atoms
and a set of p polarization functions for hydrogens. This level of
theory was sufficient to provide optimized structures that agreed
quite well with those determined experimentally. All optimized
structures converged well employing the DIIS convergence scheme
with default convergence criteria. Increasing the size of the SCF
grids and the accuracy of the numerical integrations from the default
values did not have a significant effect on the optimized geometries.
Vibrational frequencies were calculated for all converged structures,
and their character as energy minima were confirmed by the absence
of negative (imaginary) frequencies. Graphical representations of
all computed structures were generated using ORTEP-3. In order
to check for the presence of a planar structure as a local energy
minimum in the case of the methyl ketimine complexes, the
geometry of 12 was first optimized with the ligand backbone and
metal-chelate ring constrained in a planar conformation. The
constraints were then removed, and the optimization was performed
once again. A twisted structure identical to that previously calculated
was obtained, suggesting that there is no energy minimum at the
planar geometry. When the optimized structure of 12 was modified
such that the imine methyl group was replaced by a hydrogen atom,
reoptimization of that structure yielded a planar geometry. Thus,
the aldimine and ketimine complexes should each possess only one
minimum-energy ligand conformation.
Representative Procedure for X-ray Crystallography. A
crystal of appropriate size was mounted on a glass fiber or Kaptan
loop using Paratone-N hydrocarbon oil. The crystal was transferred
to a Siemens SMART or APEX diffractometer/CCD area detector,39
centered in the beam, and cooled by a nitrogen flow low-
temperature apparatus that had been previously calibrated by a
thermocouple placed at the same position as the crystal. Preliminary
orientation matrices and cell constants were determined by collec-
tion of 60 10-s frames, followed by spot integration and least-
squares refinement. An arbitrary hemisphere of data was collected
and the raw data were integrated using SAINT.40 Cell dimensions
reported were calculated from all reflections with I > 10 σ. The
data were corrected for Lorentz and polarization effects, but no
correction for crystal decay was applied. Data were analyzed for
agreement and possible absorption using XPREP.41 An empirical
absorption correction based on comparison of redundant and
equivalent reflections was applied using SADABS.42 The structure
was solved and refined with teXsan software package.43 ORTEP
diagrams were created using ORTEP-3 software package.44
Computational Methods. All calculations were performed using
the Jaguar 5.5 package45 with Maestro as the graphical user
interface.46 The hybrid DFT functional B3LYP47 was used through-
out, which consists of the Becke three-parameter functional48 and
the correlation functional of Lee, Yang, and Par.49 The LACVP**
basis set was employed,50 which uses an effective core potential
and valence double-ú contraction basis functions for metals. The
Selected Specific Experimental Procedures. For reasons of
space, procedures in close analogy with those listed below (or are
not of central importance to the discussion) are present only in the
Supporting Information.
2′-tert-Butylsulfanylacetophenone. A Schlenk flask was charged
with 2′-iodoacetophenone (5.00 g, 20.3 mmol, 1.00 equiv) and
copper tert-butyl sulfide (3.90 g, 25.5 mmol, 1.26 equiv) and flushed
with nitrogen gas. Anhydrous DMF (100 mL) was then added via
cannula, and the reaction mixture was heated at 100 °C for 12 h
with stirring in the absence of light. The resulting black solid was
allowed to settle and the pale yellow solution decanted into a
separatory funnel containing 100 mL of H2O under ambient
atmosphere. The gray, opaque suspension was extracted with Et2O
(6 × 50 mL). The Et2O extracts were then combined, washed with
H2O (4 × 100 mL), and dried over MgSO4. Filtration and removal
of the solvent in vacuo yielded a light yellow oil of high purity
(28) Still, I. W. J.; Natividad-Preyra, R.; Toste, F. D. Can. J. Chem.
1999, 77, 113-121.
(29) Kagano, H.; Itsuda, H.; Yoshida, K.; Nakano, M. (Sumitomo Seika
KK, Japan). Jpn. Kokai Tokkyo Koho 05246981, 1993.
(30) Adams, R.; Reifschneider, W.; Ferretti, A. Org. Synth. 1962, 42,
22-25.
(31) Herde, J. L.; Lambert, J. C.; Senoff, C. V. Inorg. Synth. 1974, 15,
18-20.
(32) Giordano, G.; Crabtree, R. H. Inorg. Synth. 1979, 19, 218-220.
(33) Appleton, T. G.; Hall, J. R.; Williams, M. A. J. Organomet. Chem.
1986, 303, 139-149.
1
(4.08 g, 19.6 mmol, 97%). H NMR (300 MHz, CDCl3): δ 7.57
(m, 1H), 7.37 (m, 3H), 2.63 (s, 3H), 1.23 (s, 9H). 13C{1H} NMR
(100.6 MHz, CDCl3): δ 204.90, 148.66, 138.86, 129.74, 129.12,
128.99, 127.05, 47.65, 32.26, 31.03. IR (cm-1): 1693 (s), 1584
(w). Anal. Calcd. for C12H16OS: C, 69.19; H, 7.74. Found: C,
68.96; H, 7.69.
(34) Chatt, J.; Shaw, B. L. J. Chem. Soc. 1959, 705-716.
(35) Roulet, R.; Barbey, C. HelV. Chim. Acta 1973, 56, 2179-2186.
(36) (a) Adapted from Puddephatt, R. J.; Thompson, P. J. J. Chem. Soc.
Dalton Trans. 1977, 1219-1223; (b) Spectra match Davies, J. A.; Eagle,
C. T.; Otis, D. E.; Venkataraman, U. Organometallics 1989, 8, 1080-1088.
(37) Hidai, M.; Kashiwag.T.; Ikeuchi, T.; Uchida, Y. J. Organomet.
Chem. 1971, 30, 279-282.
o-Mercaptoacetophenone. A Schlenk flask was charged with
4.40 g (33.0 mmol, 1.00 equiv) of AlCl3 under nitrogen atmosphere
and suspended in 80 mL of toluene. The flask was cooled in an
IPA/dry ice bath, and a solution of 2′-tert-butylsulfanylacetophenone
(4.00 g, 19.2 mmol, 0.582 equiv) in 20 mL of toluene was added
via cannula. The reaction mixture was allowed to warm to ambient
temperature and was stirred for 20 h. Water (100 mL) was added
to the orange suspension, and the mixture was transferred to a
separatory funnel under ambient atmosphere. Extraction caused the
mixture to become colorless. The organic layer was extracted with
5% aqueous NaOH (4 × 50 mL). All aqueous extracts were then
combined and acidified with concd HCl. The resulting white
suspension was extracted with Et2O (4 × 50 mL), and the Et2O
extracts were combined, washed with H2O (2 × 50 mL), and dried
over MgSO4. Filtration and removal of the solvent in vacuo resulted
(38) Onderdelinden, A. L.; Van der Ent, A. Inorg. Chim. Acta 1972, 6,
420-426.
(39) SMART: Area-Detector Software Package; Bruker AXS: Madison,
WI, 2001-2003.
(40) SAINT: SAX Area Detector Integration Program, version 6.40;
Bruker AXS: Madison, WI, 2003.
(41) XPREP, version 6.12; part of the SHELXTL Crystal Structure
Determination Package; Bruker AXS: Madison, WI, 2001.
(42) SADABS: Bruker-Nonius Area Detector scaling and Absoption
Correction Program, version 2.05; Bruker AXS: Madison, WI, 2003.
(43) teXsan: Crystal Structure Analysis Package; Molecular Structure
Corporation, 1985 & 1992.
(44) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(45) Jaguar, version 5.5; Schrodinger, LLC: Portland, OR, 2003.
(46) Maestro, version 6.5; Schrodinger, LLC: Portland, OR, 2004.
(47) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623-11627.
(51) (a) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon,
M. S.; DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654-3665;
(b) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213-222;
(c) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257-
2261; (d) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971,
54, 724-728.
(48) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(49) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989,
157, 200-206.
(50) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310.

908 Organometallics, Vol. 26, No. 4, 2007
Krinsky et al.
in a brown oil which was pure by 1H NMR spectroscopy.
Dissolution in CH2Cl2 and filtration through a plug of silica removed
the brown color, yielding a light yellow oil of similar purity (2.61
g, 17.1 mmol, 92%). 1H NMR (300 MHz, CDCl3): δ 7.89 (d, J )
7.5 Hz, 1H), 7.31 (m, 2H), 7.22 (m, 1H), 4.46 (s, 1H), 2.64 (s,
3H). Proton NMR spectra match literature values.52
IR (cm-1): 1605, 1590, 1574, 1523. Anal. Calcd. for C22H24-
NORhS: C, 58.28; H, 5.34; N, 3.09. Found: C, 58.14; H, 5.21; N,
3.15.
(L1-H)Ir(PMe3)3H (9). Compound 4 (73.5 mg, 0.123 mmol, 1.00
equiv) was dissolved in 1 mL of benzene, and 0.70 mmol (5.7
equiv) of trimethylphosphine was added via vacuum transfer. The
reaction mixture was then heated at 65 °C for 1 h. The solvent and
excess phosphine were removed under reduced pressure, and the
residue was dissolved in 1 mL of CH2Cl2. Crystalline material was
obtained by slow vapor diffusion of pentane (2 mL) into the CH2-
L1 Zn (1). Sodium 2-formylbenzenethiolate (400 mg, 2.50 mmol,
2
1.00 equiv) was suspended in 15 mL of THF. Solid anhydrous
ZnCl2 (180 mg, 1.32 mmol, 0.528 equiv) was added, and the
mixture was stirred for 1 h. 2,6-Diisopropylaniline (448 mg, 2.53
mmol, 1.01 equiv) was then added, followed by 5 g of 4 Å
molecular sieves. This mixture was allowed to stand for 2 d, after
which time the cloudy solution was filtered and the solvent was
removed under vacuum. The residue was dissolved in 5 mL of CH2-
Cl2 and filtered once more. The volume of the resultant yellow
solution was reduced to 2 mL and cooled to -30 °C. The
supernatant was decanted from the yellow crystalline product that
had formed, and the solid was washed with Et2O (4 × 2 mL)
yielding pure 1 (581 mg, 0.882 mmol, 71%). Solvent removal from
the supernatant and washes resulted in a residue which, after
washing with pentane (5 × 1 mL), yielded an additional 125 mg
1
Cl2 solution (64.5 mg, 0.0900 mmol, 73%). H NMR (500 MHz,
CD2Cl2) δ 7.26 (d, J ) 7.0 Hz, 1H), 7.03 (m, 1H), 6.91 (m, 1H),
6.78 (m, 1H), 6.65 (m, 2H), 6.15 (t, J ) 7.0 Hz, 1H), 3.27 (sept,
J ) 7.0 Hz, 1H), 2.91 (sept, J ) 7.0 Hz, 1H), 1.76 (d, JH-P ) 10
Hz, 9H), 1.65 (d, JH-P ) 8.0 Hz, 9H), 1.36 (d, JH-P ) 8.5 Hz,
9H), 1.22 (d, J ) 7.0 Hz, 3H), 1.13 (d, J ) 7.0 Hz, 3H), 1.01 (d,
J ) 7.0 Hz, 3H), 0.47 (d, J ) 7.0 Hz, 3H), -11.22 (d of t, JH-P
)
191 Hz, 22.5 Hz, 1H). 13C{1H} NMR (100.6 MHz, CD2Cl2): δ
190-185 (m), 156.66, 154.15, 151.01, 136.21, 133.86, 127.99,
127.66, 127.53, 124.38, 123.17, 120.29, 118.40, 28.44 (d, JC-P
)
49 Hz), 25.32 (d, JC-P ) 25 Hz), 23.71 (d, JC-P ) 36 Hz), 22.22,
21.94, 21.79, 21.42, 17.28, 17.01. 31P{1H} NMR (162.0 MHz, CD2-
Cl2): δ -44.65 (m), -58.58 (m), -63.80 (m). IR (cm-1): 2026,
1579, 1545. Anal. Calcd. for C28H49IrNP3S: C, 46.91; H, 6.89; N,
1.95. Found: C, 47.16; H, 6.96; N, 1.93.
1
(0.190 mmol) of 1 of sufficient purity for further use. H NMR
(500 MHz, CD2Cl2): δ 8.17 (s, 2H), 7.28 (br s, 2H), 7.17 (m, 6H),
6.96 (m, 4H), 6.87 (br s, 2H), 3.23 (br s, 2H), 2.67 (br s, 2H), 1.41
(br s, 6H), 1.04 (d, J ) 6.5 Hz, 12H), 0.60 (br s, 6H). 13C{1H}
NMR (125.8 MHz, CD2Cl2): δ 173.83, 149.39, 146.97, 142.16 (br),
139.73 (br), 137.32, 136.81, 131.78, 130.62, 127.26, 124.37 (br),
123.09 (br), 122.63, 29.52, 29.28, 25.83, 23.69, 21.64. IR (cm-1):
1620, 1604, 1585, 1541. Anal. Calcd. for C38H44N2S2Zn: C, 69.33;
H, 6.74; N, 4.26. Found: C, 68.99; H, 6.67; N, 4.11.
L3H (10). o-Mercaptoacetophenone (591 mg, 3.88 mmol, 1.00
equiv) was diluted with 15 mL of benzene and p-toluidine (416
mg, 3.88 mmol, 1.00 equiv) was added, followed by 5 g of 3 Å
molecular sieves. The reaction mixture was allowed to stand for 2
days, during which time the solution became deep red. After solvent
removal, the resulting amorphous powder was dissolved in 3 mL
of THF and gently heated. During heating, pentane (15 mL) was
slowly added, resulting in the formation of bright red crystals upon
cooling. The reaction vessel was then placed in a -30 °C freezer
overnight. The supernatant was removed, and the crystalline material
(791 mg, 3.28 mmol, 84%) was washed with pentane. Note: the
L1Rh(cod) (3). A solution of 1 (200 mg, 0.304 mmol, 1.00 equiv)
in 5 mL of THF was slowly added to [Rh(cod)Cl]2 (150 mg, 0.304
mmol, 1.00 equiv) dissolved in 12 mL of THF. After the reaction
mixture was stirred for 4 h, the solvent was removed under vacuum.
The residue was extracted with Et2O (4 × 5 mL), and the filtered
extracts were combined and reduced to 10 mL. The solution was
held at -30 °C for 24 h, resulting in the precipitation of 252 mg
1
crude product was pure by H NMR and could be used without
purification. Crystals suitable for X-ray crystallographic analysis
1
(0.496 mmol, 82%) of pure 3. H NMR (500 MHz, CD2Cl2): δ
1
were grown by cooling a concentrated solution of 10 in Et2O. H
8.19 (d, J ) 2.5 Hz, 1H), 7.83 (d, J ) 7.5 Hz, 1H), 7.34 (t, J ) 7.5
Hz, 1H), 7.25 (m, 3H), 7.09 (t, J ) 7.5 Hz, 1H), 4.48 (m, 2H),
3.49 (m, 2H), 3.35 (sept, J ) 7.0 Hz, 2H), 2.32 (m, 4H), 2.02 (m,
2H), 1.90 (m, 2H), 1.40 (d, J ) 7.0 Hz, 6H), 1.02 (d, J ) 7.0 Hz,
6H). 13C{1H} NMR (125.8 MHz, CD2Cl2): δ 167.96, 149.82,
140.87, 138.11, 132.68, 131.93, 130.11, 127.56, 124.18, 123.90,
122.65, 86.50, 78.23, 30.96, 30.53, 28.74, 25.72, 22.63. IR (cm-1):
1590, 1572, 1524. Anal. Calcd. for C27H34NRhS: C, 63.90; H,
6.75; N, 2.76. Found: C, 63.92; H, 6.55; N, 2.72.
L2Rh(cod) (5). A solution of 2 (129 mg, 0.234 mmol, 1.00 equiv)
in 5 mL of THF was slowly added to [Rh(cod)Cl]2 (116 mg, 0.235
mmol, 1.00 equiv) dissolved in 10 mL of THF. After the reaction
mixture was stirred for 6 h, the solvent was removed under vacuum.
The residue was extracted with 3 mL of CH2Cl2, filtered, and heated
to boiling while 2 mL of pentane was added. The vessel was then
cooled to -30 °C overnight. The supernatant was decanted, and
the orange solid was washed with Et2O (3 × 1 mL) (204 mg, 0.450
mmol, 96%). Crystals suitable for X-ray diffraction analysis were
grown by vapor diffusion of pentane into a solution of 5 in THF.
1H NMR (500 MHz, THF-d8): δ 8.39 (s, 1H), 7.72 (d, J ) 8.0
Hz, 1H), 7.46 (d, J ) 6.5 Hz, 1H), 7.22 (m, 1H), 7.07 (d, J ) 8.5
Hz, 2H), 6.96 (m, 1H), 6.89 (d, J ) 8.5 Hz, 2H), 4.37 (br s, 2H),
3.78 (s, methyl and cod olefinic overlapping, 5H), 2.27 (m, 4H),
1.95 (br s, 2H), 1.85 (br s, 2H). 13C{1H} NMR (125.8 MHz, THF-
d8): δ 168.61, 158.84, 151.46, 149.23, 138.51, 132.45, 131.43,
130.47, 125.28, 121.70, 114.25, 84.35, 76.85, 55.79, 31.36, 30.90.
NMR (500 MHz, CD2Cl2): δ 17.33 (s, 1H), 7.74 (d, J ) 1.5 Hz,
1H), 7.72 (d, J ) 1.0 Hz, 1H), 7.31 (d, J ) 8.0 Hz, 2H), 7.13 (m,
3H), 6.98 (m, 1H), 2.55 (s, 3H), 2.41 (s, 3H). 13C{1H} NMR (125.8
MHz, CD2Cl2): δ 172.08, 162.19, 138.35, 138.32, 136.57, 131.97,
130.57, 130.53, 128.11, 124.37, 121.36, 21.37, 18.40. IR (cm-1):
1615, 1585, 1531. Anal. Calcd. for C15H15NS: C, 74.65; H, 6.26;
N, 5.80. Found: C, 74.47; H, 6.41; N, 5.76.
L3Rh(cod) (12). Yellow [Rh(cod)Cl]2 (203 mg, 0.412 mmol,
1.00 equiv) was dissolved in 10 mL of THF and cooled to -30 °C.
Ligand 10 (200 mg, 0.829 mmol, 2.01 equiv) was dissolved in 4
mL of THF, and NaH (21.3 mg, 0.843 mmol) was added, causing
vigorous gas evolution. After 30 min, the solution had become light
yellow, at which time it was slowly added to the cooled [Rh(cod)-
Cl]2 solution. The mixture was allowed to stir for 4 h, and the
solvent was removed under vacuum. 1H NMR spectra showed
quantitative conversion to product. The residue was dissolved in 3
mL of CH2Cl2 and filtered, and the solution was gently heated while
10 mL of pentane was slowly added. The reaction vessel was placed
in a -30 °C freezer overnight. The supernatant was removed from
the resulting orange crystalline material, and the solid was washed
with pentane (342 mg, 0.758 mmol, 92%). Slow evaporation of a
CH2Cl2 solution of 12 yielded crystals of sufficient quality for X-ray
diffraction analysis. 1H NMR (500 MHz, CD2Cl2): δ 7.62 (d, J )
7.5 Hz, 1H), 7.45 (d, J ) 8.0 Hz, 1H), 7.21 (d, J ) 8.0 Hz, 2H),
7.08 (m, 1H), 6.99 (m, 1H), 6.91 (d, J ) 8.0 Hz, 2H), 4.14 (br s,
2H), 3.35 (br s, 2H), 2.38 (s, 3H), 2.27 (br s, 2H), 2.14 (br s, methyl
and cod methylene overlapping, 5H), 1.81 (br s, 2H), 1.67 (br s,
2H). 13C{1H} NMR (125.8 MHz, CD2Cl2): δ 171.42, 148.67,
(52) Topolski, M. J. Org. Chem. 1995, 60, 5588-5594.

Platinum Group Thiophenoxyimine Complexes
Organometallics, Vol. 26, No. 4, 2007 909
147.40, 138.44, 135.71, 132.66, 131.26, 129.90, 129.82, 122.28,
121.84, 86.23 (d, JC-Rh ) 10.1 Hz), 76.68 (d, JC-Rh ) 12.6 Hz),
31.46, 30.59, 24.31, 21.21. IR (cm-1): 1958, 1933, 1903, 1582,
1565, 1534, 1504. Anal. Calcd. for C23H26NRhS: C, 61.19; H, 5.80;
N, 3.10. Found: C, 61.11; H, 5.95; N, 3.18.
1598, 1578, 1539. Anal. Calcd. for C57H56NP2RhS: C, 71.92; H,
5.93; N, 1.47. Found: C, 70.91; H, 6.02; N, 1.41. Analysis of two
independently prepared and purified samples showed similarly low
abundance of carbon. NMR spectra of pure 20 are available as
Supporting Information.
L3Pt(PPh3)Cl (15). NaH (9.98 mg, 0.416 mmol, 1.00 equiv) was
added to a solution of 10 (100 mg, 0.414 mmol, 0.995 equiv) in 5
mL of THF. The mixture was stirred for 30 min and was then added
dropwise to a suspension of (Ph3P)2PtCl2 (326 mg, 0.413 mmol,
0.993 equiv) in 10 mL of THF and stirred overnight. The solvent
was removed under reduced pressure, and the residue was extracted
with 10 mL of benzene and lyophilized. The crude product was
allowed to stand under 10 mL of Et2O for 2 weeks, and the Et2O
was then decanted. This process was then repeated, after which
time the remaining solid was extracted with 2 mL of THF. Vapor
diffusion of pentane (1 mL) into this solution resulted in crystalline
L5H (25). Molecular sieves (3 Å, 4.5 g), 15 mL of benzene,
compound 24 (1.00 g, 2.74 mmol, 1.00 equiv), and p-toluidine (297
mg, 2.77 mmol, 1.01 equiv) were combined and left to stand for
10 d. The reaction mixture was then filtered away from the
molecular sieves and lyophilized. The bright yellow residue was
dissolved in 3 mL of Et2O, and 5 mL of pentane was slowly added
while the solution was gently heated. The vessel was then cooled
to -30 °C overnight. The crystalline solid was collected via
filtration and washed with pentane (6 × 1 mL), yielding 878 mg
of product. The pentane washes were added to the supernatant,
which upon cooling afforded an additional 126 mg of pure 25 (1.00
1
1
15 (64.5 mg, 0.0880 mmol, 21%). H NMR (500 MHz, CD2Cl2):
g total, 2.21 mmol, 81%). H NMR (400 MHz, C6D6): δ 8.43 (s,
δ 7.62 (m, 6H), 7.52 (d, J ) 7.0 Hz, 1H), 7.44 (m, 3H), 7.37 (m,
6H), 7.29 (m, 5H), 7.14 (m, 2H), 2.44 (s, 3H), 2.28 (s, 3H). 13C-
{1H} NMR (125.8 MHz, CD2Cl2): δ 171.32, 146.93, 141.46 (d,
JC-P ) 6.3 Hz), 136.73, 135.10 (d, JC-P ) 11.3 Hz), 132.17, 131.23,
131.15, 130.55, 130.30, 129.79, 129.42, 128.40 (d, JC-P ) 11.3
Hz), 125.13, 124.80, 23.53, 21.38. 31P{1H} NMR (162 MHz, CD2-
Cl2): δ 12.02 (JP-Pt ) 3914 Hz). IR (cm-1): 1594, 1552, 1505.
Anal. Calcd. for C33H29ClNPPtS: C, 54.06; H, 3.99; N, 1.91.
Found: C, 53.70; H, 3.81; N, 1.82. Compound 16, present in the
crude reaction mixture, was not obtained in analytically pure form.
NMR spectra of 16 are available as Supporting Information.
[L4PtMe3]2 (19). A solution of 11 (40.0 mg, 0.155 mmol, 1.00
equiv) in 3 mL of acetonitrile was slowly added to 49.5 mg (0.156
mmol, 1.01 equiv) of (nbd)PtMe2 in 5 mL of acetonitrile. The
mixture quickly developed a dark orange color and then turned
light orange after ca. 5 min. After 6 h, a brick red precipitate had
formed which was collected by filtration and washed with aceto-
nitrile (3 × 1 mL). This sparingly soluble material was not
2H), 7.79 (s, 1H), 6.90 (d, J ) 8.0 Hz, 2H), 6.80 (d, J ) 8.0 Hz,
2H), 6.75 (d, J ) 8.0 Hz, 1H), 6.51 (m, 2H), 2.80 (s, 1H), 1.91 (s,
3H), 1.71 (s, 3H). 13C{1H} NMR (100.6 MHz, C6D6): δ 163.24,
147.56, 141.27, 136.35, 135.62, 134.70, 132.26 (q, JC-F ) 33.2
Hz), 131.32, 130.74, 130.08, 129.54, 128.72 (d, JC-F ) 3.0 Hz),
126.67, 124.04 (quint, JC-F ) 4.0 Hz), 123.77 (q, JC-F ) 273.6
Hz), 121.22, 20.75, 20.50. 19F NMR (376.5 MHz, C6D6): δ -61.93.
IR (cm-1): 2563, 1627, 1596, 1504. Anal. Calcd. for C23H17F6NS:
C, 60.92; H, 3.78; N, 3.09; S, 7.07. Found: C, 61.04; H, 3.92; N,
2.90; S, 7.40.
[L5Rh(cod)]2 (28). Potassium tert-butoxide (13.8 mg, 0.123
mmol, 1.00 equiv) was added to a solution of 25 (55.6 mg, 0.123
mmol, 1.00 equiv) in 3 mL of THF. After stirring for 30 min, the
mixture was added dropwise to a solution of [Rh(cod)Cl]2 (30.2
mg, 0.0612 mmol, 0.498 equiv) in 10 mL of THF. The reaction
mixture was stirred for 3 h, after which time the solvent was
removed under reduced pressure. The residue was extracted with
2 mL of benzene, and the extract was filtered into a small vessel,
into which 1 mL of pentane vapor was allowed to slowly diffuse.
Large crystals of [28‚0.5 pentane] were collected and washed with
Et2O (3 × 1 mL) (56.8 mg, 0.0813 mmol, 66%). 1H NMR spectra
of 28 at three temperatures are included in the Supporting
Information. These spectra, as well as the elemental analysis, were
obtained using the same sample that was used for the X-ray
diffraction studies. IR (cm-1): 1630, 1604, 1595, 1504. Anal. Calcd.
for C31H28F6NRhS‚0.5C5H12: C, 57.51; H, 4.90; N, 2.00. Found:
C, 57.39; H, 5.03; N, 2.00.
1
successfully purified although its H NMR spectrum is consistent
with the crystallographically determined structure (see Supporting
Information). The supernatant from the reaction mixture was
allowed to stand for 2 d, during which time a few high-quality
crystals formed which were used for the X-ray structural determi-
nation.
L3Rh(P(p-tol)3)2 (20). Tri(p-tolyl)phosphine (135 mg, 0.444
mmol, 1.00 equiv) was added to a solution of 12 (100 mg, 0.222
mmol, 0.500 equiv) dissolved in 10 mL of benzene, causing a rapid
color change from orange to dark purple/red. The reaction mixture
was stirred for 24 h, after which time the solvent was removed
under reduced pressure. The residue was dissolved in 2 mL of THF
and heated to boiling while 12 mL of hexane was slowly added.
Upon cooling to ambient temperature, the vessel was cooled at
-30 °C for 24 h. The supernatant was decanted, and the dark purple
solid was washed with Et2O (3 × 2 mL) (167 mg, 0.175 mmol,
79%). 1H NMR (500 MHz, CD2Cl2): δ 7.64 (d, J ) 7.5 Hz, 1H),
7.41 (m, 6H), 7.28 (d, J ) 7.5 Hz, 1H), 7.23 (m, 1H), 7.10 (t, J )
7.5 Hz, 1H), 7.07 (d, J ) 8.0 Hz, 2H), 6.90 (m, 12H), 6.80 (m,
6H), 6.35 (br s, 2H), 2.46 (s, 3H), 2.32 (s, 9H), 2.29 (s, 9H), 1.92
(s, 3H). 13C{1H} NMR (125.8 MHz, CD2Cl2): δ 169.75, 149.06,
146.33, 138.82, 138.66, 135.16, 135.02, 134.93, 134.69, 134.32,
134.22, 133.88, 133.58, 132.89, 130.20, 128.79, 128.06, 127.98,
127.45, 123.92, 121.82, 21.78, 21.56, 21.50, 21.29. 31P{1H} NMR
(162.0 MHz, CD2Cl2): δ 46.51 (dd, JP-P ) 42.1 Hz, JP-Rh ) 179.8
Hz), 40.02 (dd, JP-P ) 42.1 Hz, JP-Rh ) 183.1 Hz). IR (cm-1):
Acknowledgment. We are indebted to Dr. Fred Hollander
and Dr. Allen Oliver at the UC Berkeley CHEXRAY X-ray
crystallographic facility and Dr. Kathleen Durkin at the Mo-
lecular Graphics Facility (NSF grant CHE-0233882) for crystal-
lographic and computational consultation, respectively. We also
thank Prof. F. Dean Toste for advice concerning ligand
syntheses. This work was supported by NSF grant CHE-
0345488 to R.G.B. and DOE funding to J.A.
Supporting Information Available: CIF-format files for all
crystallographically determined structures, experimental procedures
for compounds not listed above, NMR spectra of 16, 19, 20 and
28, positional coordinates of all computed structures. This material
is available free of charge via the Internet at http:/pubs.acs.org.
OM0609057