Reaction mechanism of transmetalation between tetraorganostannanes and platinum(II) aryltriflate complexes. Mechanistic model for Stille couplings

Article abstract of DOI:10.1021/om051021a

The complexes trans-[PtPh(OTf)(PMe₂Ph)₂] (1) and trans-[PtPh(OTf)(PPh₃)₂] (4) were synthesized from the corresponding chloro complexes in moderate yields. Complex 1 is slowly hydrolyzed in solution, giving the d

Full text of DOI:10.1021/om051021a

Organometallics 2006, 25, 1285-1292
1285
Reaction Mechanism of Transmetalation between
Tetraorganostannanes and Platinum(II) Aryltriflate Complexes.
Mechanistic Model for Stille Couplings
Patrik Nilsson,† Graeme Puxty,‡ and Ola F. Wendt*,†
Organic Chemistry, Department of Chemistry, Lund UniVersity, P.O. Box 124, S-221 00 Lund, Sweden,
and ICB, Safety and EnVironmental Technology Group, ETH Zu¨rich, CH 8093 Zu¨rich, Switzerland
ReceiVed NoVember 28, 2005
The complexes trans-[PtPh(OTf)(PMe2Ph)2] (1) and trans-[PtPh(OTf)(PPh3)2] (4) were synthesized
from the corresponding chloro complexes in moderate yields. Complex 1 is slowly hydrolyzed in solution,
giving the dinuclear hydroxo-bridged complex [Pt(µ-OH)(PMe2Ph)2]2(CF3SO3)2 (6), which was character-
ized by X-ray crystallography. In solution 1 and 4 undergo fast solvolysis to give the corresponding
solvento cations. The reactivities of 1 and 4 with Me3SnPh were investigated in different solvents, and
with 1 two products, trans-[PtPh2(PMe2Ph)2] (8) and cis-[PtPh2(PMe2Ph)2] (9), were always formed
simultaneously. In THF an intermediate, trans-[PtPhMe(PMe2Ph)2] (14), was characterized on the path
to 9. The kinetics for these reactions were evaluated numerically, and on the basis of rate laws, activation
parameters, and reactivity trends a mechanism involving parallel equilibria to 8 and 9 with associative
activation in all steps is proposed. In this mechanism the initial attack always takes place trans to the
phenyl group, giving 8 and 14, respectively, via an open transition state. In a subsequent reaction, 14
reacts with another molecule of stannane, giving 9 as the final product via a cyclic transition state. Complex
4 gives exclusive formation of cis products: cis-[PtPhMe(PPh3)2] (12) and cis-[PtPh2(PPh3)2] (13). These
results are discussed in relation to the reaction mechanism of the transmetalation step in the Stille reaction.
of the reagents in the Stille reaction mainly stems from the low
polarity of the Sn-C bond, and this property is perhaps mostly
pronounced in the transmetalation step. Thus, the mechanistic
insight gained from studies of transmetalation from organotin
reagents are also relevant in other processes involving M-C
bonds of moderate polarity: e.g., Suzuki and Hiyama coupling.6
Also, phosphapalladacycles including PCP pincer complexes
have become known as very promising catalysts for C-C bond
formation reactions.7 Although these types of complexes are
not very commonly used in the Stille coupling, examples have
appeared, from which some mechanistic insight into the
transmetalation reaction has been generated.8
Introduction
In the Stille coupling reaction a palladium catalyst is used to
couple aryl halides with organotin nucleophiles and generate
carbon-carbon bonds. It has become an attractive method in
modern organic synthesis.
The mechanism originally proposed for the Stille reaction
considered a Pd(0)/Pd(II) catalytic cycle with four consecutive
reaction steps (oxidative addition, transmetalation, trans-to-cis
isomerization, and reductive elimination), in which the trans-
metalation step was suggested to be rate-determining.1 Recent
work has provided a greater mechanistic understanding of the
catalytic cycle and the individual steps therein,2 and it has
become apparent that the reaction conditions used strongly
influence the mechanism.3 However, due to its intrinsic
complexity, the transmetalation reaction is still the most poorly
understood step in the catalytic cycle and is therefore a matter
of intense study.4
From a mechanistic standpoint, the use of platinum instead
of palladium should make the complexes less prone to reductive
(4) (a) Farina, V.; Krishna, B. J. Am. Chem. Soc. 1991, 113, 9585. (b)
Louie, J.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 11598. (c) Casado,
A. L.; Espinet, P.; Gallego, A. M. J. Am. Chem. Soc. 2000, 122, 11771. (d)
Casares, J. A.; Espinet, P.; Salas, G. Chem. Eur. J. 2002, 8, 4844. (e)
AÄ lvarez, R.; Faza, O. N.; Lo´pez, C. S.; de Lera, AÄ . R. Org. Lett. 2006, 8,
35.
(5) (a) Ricci, A.; Angeluci, F.; Bassetti, M.; Lo Sterzo, C. J. Am. Chem.
Soc. 2002, 124, 1060. (b) Napolitano, E.; Farina, V.; Persico, M. Organo-
metallics, 2003, 22, 4030. (c) Amatore, C.; Bahsoun, A. A.; Jutand, A.;
Meyer, G.; Ntpede, A. N.; Ricard, L. J. Am. Chem. Soc. 2003, 125, 4212.
(6) (a) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457. (b) Hiyama,
T.; Hatanaka, Y. Pure Appl. Chem. 1994, 66, 1471.
Many different model systems have been used to study the
transmetalation step.5 The high tolerance toward air and moisture
* To whom correspondence should be addressed. E-mail: ola.wendt@
organic.lu.se.
† Lund University.
‡ ETH Zu¨rich.
(1) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508.
(2) (a) Mateo, C.; Cardenas, D. J.; Fernandez-Rivas, C.; Echavarren, A.
M. Chem. Eur. J. 1996, 2, 1596. (b) Amatore, C.; Jutand, A.; Suarez, A. J.
Am. Chem. Soc. 1993, 115, 9531. (c) Casado, A. L.; Espinet, P. Organo-
metallics 1998, 17, 954. (d) Brown, J. M.; Cooley, N. A. Chem. ReV. 1988,
88, 1031.
(7) (a) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100, 3009.
(b) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41, 4176. (c)
Crabtree, R. H. Chem. ReV. 1985, 85, 245. (d) van der Boom, M. E.;
Milstein, D. Chem. ReV. 2003, 103, 1759.
(8) (a) Louie, J.; Hartwig, J. F. Angew. Chem., Int. Ed. Engl. 1996, 35,
2359. (b) Cotter, W. D.; Barbour, L.; McNamara, K. L.; Hechter, R.;
Lachicotte, R. J. J. Am. Chem. Soc. 1998, 120, 11016. (c) Olsson, D.;
Nilsson, P.; El Masnaouy, M.; Wendt, O. F. Dalton Trans. 2005, 1924.
(3) Espinet, P.; Echavarren, A. M. Angew. Chem., Int. Ed. 2004, 43,
4704.
10.1021/om051021a CCC: $33.50 © 2006 American Chemical Society
Publication on Web 01/27/2006

1286 Organometallics, Vol. 25, No. 5, 2006
Nilsson et al.
acetone, was added and the reaction solution was shielded from
light. After 48 h at room temperature the solution was filtered before
the solvent was removed. The crystals were dried under vacuum
for 2 h. To eliminate residual silver metal and triflate, the product
was dissolved in benzene and placed at 4 °C for 2 h before another
filtration was performed and the solvent removed. The off-white
product was washed with petroleum ether and recrystallized from
dichloromethane-petroleum ether. The yield was 137 mg (0.197
mmol, 55%). Anal. Calcd for C23H27F3O3P2PtS: C, 39.60; H, 3.90.
elimination of the coupling product, and this could possibly aid
in the detection of intermediates from the different reaction steps
in the Stille coupling reaction. A system where the active metal
center is platinum(II) instead of palladium(II) therefore seemed
to be a good choice for a thorough kinetic and mechanistic
investigation of the transmetalation step in the Stille reaction.
A number of investigations on the reaction between tetraor-
ganotin compounds and different platinum complexes have been
reported in the literature.9 As stoichiometric probes of trans-
metalation in d8 complexes, we have studied the reactions of
trans-[PtPhX(L)2] (X ) F, OTf; L ) PPh3, PMe2Ph, AsPh3)
with tetraorganostannanes, such as Me3SnPh and Bu3Sn-
(vinyl).10,11 Here we report a kinetic study of the transmetalation
reaction between trans-[PtPh(OTf)(PMe2Ph)2] (1) and Me3SnPh
in different solvents. The effects on the reactivity, as well as
the product distribution, in the transmetalation reaction in terms
of the nature of the leaving group and the ancillary phosphine
ligands are discussed, and a mechanistic picture, including the
characterization of an intermediate, is reported.
1
Found: C, 39.50; H, 4.02. H NMR (CDCl3): δ 1.56 (t, 12H,
3
2JP-H ) 3.6 Hz, JPt-H ) 27 Hz), 6.54-6.84 (m, 5H), 7.32-7.44
1
(m, 10H). 31P{1H} NMR (CDCl3): δ 1.09 (s, JPt-P ) 2929 Hz).
19F NMR (CDCl3): δ 0.47 (s, 3F).
trans-[PtPh(OTf)(PPh3)2] (4). An 81.6 mg portion (0.098 mmol)
of 3 was dissolved in 5 mL of dichloromethane in a Schlenk flask.
A 30.5 mg portion (0.119 mmol) of AgOTf was dissolved in 2 mL
of acetone and added to the dichloromethane solution. After 48 h
at room temperature with shielding from light, the solution was
filtered and the solvent removed under reduced pressure. A
brownish oil was formed, which was dissolved in benzene and
filtered through a 1 cm layer of Celite, giving a yellow solution.
After evaporation a yellow oil containing white crystals was
achieved. The product was dissolved in a minimum of dichlo-
romethane, which was allowed to evaporate under atmospheric
pressure before the white crystals were washed with pentane. The
yield was 51.6 mg (0.056 mmol, 57%). Anal. Calcd for C43H35F3O3P2-
PtS: C, 54.60; H, 3.73. Found: C, 54.52; H, 3.82. 1H NMR
Experimental Section
General Procedures and Materials. All experiments involving
air-sensitive compounds were carried out using standard high-
vacuum-line or Schlenk techniques or in a glovebox under nitrogen.
If nothing else is stated, all reagents and solvents were of the best
quality available and were used as received from Aldrich. The
complexes trans-[PtPhCl(PMe2Ph)2] (2) and trans-[PtPhCl(PPh3)2]
(3) were prepared according to the literature.10,12 Elemental analyses
was performed by H. Kolbe Mikroanalytisches Laboratorium,
Mu¨lheim an der Ruhr, Germany. Fast atom bombardment (FAB)
mass spectroscopic data were obtained on a JEOL SX-102
spectrometer using 3-nitrobenzyl alcohol as the matrix. Conductivity
measurements were performed using a Metrohm 644 conductome-
3
(C6D6): δ 6.08 (t, 2H, JH-H ) 7.50 Hz, m-H, Pt-Ph), 6.35
3
3
(t, 1H, JH-H ) 7.01 Hz, p-H, Pt-Ph), 6.70 (d, 2H, JH-H ) 7.50
Hz, o-H, Pt-Ph), 7.02-7.05 (m, 12H), 7.41-7.48 (m, 6H), 7.62-
7.67 (m, 12H). 31P{1H} NMR (C6D6): δ 28.2 (s, JPt-P ) 3201
1
Hz). MS (FAB+): m/z 796 [PtPh(PPh3)2+]. MS (FAB-): m/z
149[OSO2CF3-].
Me3Sn(OTf) (5). The following synthesis is a modification of a
procedure already described in the literature.13 A 100 mL round-
bottom flask, connected to a swivel frit, was charged with 0.534 g
(2.08 mmol) of AgOTf and 0.634 g (2.60 mmol) of Me3SnBr.
Approximately 20 mL of THF was vacuum-transferred from Na/
Ph2CO into the flask under reduced pressure before the reaction
solution was stirred at room temperature for 2 h. The reaction
solution was filtered, and the solvent was evaporated under reduced
pressure, giving a white product which was stored under N2(g) at
ter. The cell constant was determined to be 0.87 cm-1
.
NMR Measurements. 1H, 31P, and 19F NMR spectra were
recorded on a Varian Unity 300 or a Varian Unity Inova 500
spectrometer. Chemical shifts are given in ppm downfield from
TMS using residual solvent peaks (1H and 13C NMR) or H3PO4
(31P NMR δ 0) and CFCl3 (19F NMR δ 0) as external references.
The temperature was measured using the temperature-dependent
shift of the CH2 and OH protons of ethylene glycol.
1
-20 °C. The yield was 0.416 mg (1.33 mmol, 64%). H NMR
trans-[PtPh(OTf)(PMe2Ph)2] (1). A Schlenk flask was charged
with 208 mg (0.356 mmol) of complex 2 dissolved in 15 mL of
dichloromethane. While the mixture was stirred, 103 mg (0.402
mmol) of AgOTf (Acros Organics, 99+%), dissolved in 5 mL of
1
(CDCl3): δ 0.78 (s, 2JSn-H ) 64.0)14 (assignment of the H NMR
1
signal was confirmed by a H{13C} HSQC NMR experiment).
Reaction of 1 and 4 with Stannane. In a typical experiment, a
J. Young NMR tube was loaded with the aryltriflate platinum(II)
complex and solvent. The reaction mixture was thermostated prior
to addition of an excess of the tin compound, and the reaction was
monitored using 31P NMR spectroscopy. In most cases the products
were not separated and isolated but characterized in situ. In addition,
some qualitative measurements were performed in the presence of
Bu4N(PF6), Bu4N(OTf), or free PMe2Ph. In these cases the reagents
were administered as stock solutions in dichloromethane-d2.
(9) (a) Cetinkaya, B.; Lippert, M. F.; McMeeking, J.; Palmer, D. E. J.
Chem. Soc., Dalton Trans. 1973, 1202. (b) Eaborn, C.; Odell, K.; Pidcock,
A. J. Organomet. Chem. 1975, 96, C38. (c) Cardin, C. J.; Cardin, D. J.;
Lappert, M. F. J. Chem. Soc., Dalton Trans. 1977, 767. (d) Eaborn, C.;
Odell, K. J.; Pidcock, A. J. Organomet. Chem. 1978, 146, 17. (e) Eaborn,
C.; Odell, K. J.; Pidcock, A. J. Chem. Soc., Dalton Trans. 1978, 357. (f)
Eaborn, C.; Odell, K. J.; Pidcock, A. J. Chem. Soc., Dalton Trans. 1979,
758. (g) Eaborn, C.; Kandu, K.; Pidcock, A. J. Chem. Soc., Dalton Trans.
1981, 933. (h) Eaborn, C.; Kandu, K.; Pidcock, A. J. Chem. Soc., Dalton
Trans. 1981, 1223. (i) Stang, P. J.; Kowalski, M. H.; Schiavelli, M. D.;
Langford, D. J. Am. Chem. Soc. 1989, 111, 3347. (j) Stang, P. J.; Kowalski,
M. H. J. Am. Chem. Soc. 1989, 111, 3356. (k) Mateo, C.; Fernandez-Rivas,
C.; Cardenas, D. J.; Echavarren, A. M. Organometallics 1998, 17, 3661.
(10) Nilsson, P.; Plamper, F.; Wendt, O. F. Organometallics 2003, 22,
5235.
Data Analysis. A kinetic model was fitted to the measured
concentration vs time data by nonlinear least-squares (NLLS)
regression using the program Pro-KII.15 The observed data were
arranged into the matrix C, where each column of C contained the
time-dependent concentration profile of a particular species. A
proposed kinetic model was entered into the software program using
an intelligent model parser, which extracts the number of species,
(11) OTf ) trifluoromethane sulfonate, [OSO2CF3]-.
(12) (a) Cox, E. G.; Saenger, H.; Wardlaw, W. J. Chem. Soc. 1934, 182.
(b) Otto, S. Structural and Reactivity Relationships in Platinum(II) and
Rhodium(I) Complexes. Thesis, University of The Orange Free State,
Bloemfontein, South Africa, 1999; Chapter 3. (c) Clark, H. C.; Dixon, K.
R. J. Am. Chem. Soc. 1969, 91, 596. (d) Kukushkin, V. Y.; Lo¨vqvist, K.;
Nore´n, B.; Oskarsson, A° .; Elding, L. I. Phosphorus, Sulfur Silicon Relat.
Elem. 1992, 73, 253. (e) Kapoor, P.; Kukushkin, V. Y.; Lo¨vqvist, K.;
Oskarsson, A° . J. Organomet. Chem 1996, 517, 71.
(13) (a) Schmeisser, M.; Sartori, P.; Lippsmeier, B. Chem. Ber. 1970,
103, 868. (b) Kapoor, R.; Sood, V.; Kapoor, P. Polyhedron 1995 14, 489
(14) No resolution was obtained between 119Sn and 117Sn satellites.
(15) Puxty, G.; Maeder, M.; Neuhold, Y.-M.; King, P. Pro-Kineticist II;
Applied Photophysics Ltd, Leatherhead, England, 2001.

Mechanistic Model for Stille Couplings
Organometallics, Vol. 25, No. 5, 2006 1287
species names, their corresponding stoichiometric coefficients, and
the number of reactions. The program produces a list of parameters
(rate constants) with their reaction coefficients and constructs a
system of simultaneous ordinary differential equations (ODEs) that
describe the change in concentration of each species with time.16
With knowledge of the initial concentrations of the species in the
chemical model, the differential equations were integrated to yield
calculated concentrations of each species at the same times as the
measured data. The calculated concentration data were arranged
into a matrix of concentration profiles, Cˆ .
Scheme 1
A matrix of residuals, Rˆ , representing the difference between
the measured and calculated concentration data, was calculated
according to eq 1. The rate constants of the kinetic model were
The 31P{1H} NMR spectrum of 1 in chloroform solvent is a
singlet with platinum satellites (1JPt-P ) 2929 Hz), which
confirms the trans configuration. The virtual triplet at δ 1.56
(2JP-H ) 3.6 Hz) for the methyl protons on the phosphorus
ligands, as seen in the 1H NMR spectrum, also indicates a trans
Rˆ ) C - Cˆ
(1)
1
configuration. Similarly, 4 displays a JPt-P coupling constant
then refined by NLLS regression until a best fit was found between
C and Cˆ (as measured by the sum of the squared residuals). The
NLLS algorithm used was the Levenburg-Marquardt algorithm.17
As a comparison the commercially available FACSIMILE
chemical modeling package, implemented on a PC, was also
used.
of 3201 Hz, which is consistent with a trans geometry.
From a benzene-d6 solution of 1 single crystals of the
dinuclear complex [Pt(µ-OH)(PMe2Ph)2]2(CF3SO3)2 (6) were
formed after several weeks. The molecular structure is given in
Figure 1 together with selected bond distances and angles.
Crystal data and details of the data collection are given in Table
1. The coordination geometry around the Pt(II) atoms is distorted
square planar, with two bridging hydroxo ligands. This gives
rise to a cis arrangement between the phosphine ligands and a
four-membered ring where the O-Pt-O and Pt-O-Pt angles
are 79.2(3) and 100.7(2)°, respectively. Hydroxo-bridged di-
nuclear complexes of palladium and platinum are far from
unusual, and several crystal structures of tetraphosphine com-
pounds are known, including the cation of complex 6; it was
earlier solved in the space group P1h with nitrate as the
counteranion.23 Although the angles in the reported nitrate
complex are similar to those in 6, packing effects are far from
negligible, since all the distances in the coordination sphere of
the present triflate crystals are substantially longer.
Most reasonably, the formation of 6 takes place via proto-
nolysis (by adventitious water) of the phenyl group of 1 in
solution. It is known that coordination of a water molecule, by
displacement of the triflate, renders it acidic enough to explain
the protonolysis, although Pt-C bonds are normally stable
toward water.24 In the presence of hydroxide, formed in the
protonolysis, dicationic bis(phosphine) complexes are known
to form hydroxo-bridged dinuclear complexes.23 In addition, the
lability of coordinated triflate anions has been used in developing
facile routes to ligand-bridged, dinuclear transition-metal com-
plexes.25
Structure Determination of 6. Single crystals of the dinuclear
complex [Pt(µ-OH)(PMe2Ph)2]2(CF3SO3)2 (6) suitable for X-ray
diffraction were obtained after thawing a benzene-d6 stock solution
of 1 that had been stored at -20 °C. The intensity data set was
collected at 293 K with a Bruker SMART CCD system using ω
scans and a rotating anode with Mo KR radiation (λ ) 0.710 73
Å).18 The intensities were corrected for Lorentz, polarization, and
absorption effects using SADABS.19 The first 50 frames were
collected again at the end to check for decay; no decay was
observed. All reflections were integrated using SAINT.20 The
structure was solved by Patterson methods and refined by full-
matrix least-squares calculations on F2 using SHELXTL 5.1.21
Non-H atoms were refined with anisotropic displacement param-
eters. Hydrogen atoms were constrained to parent sites, using a
riding model. The crystals were fairly weak scatterers, giving rise
to a high Rint value.22 The data are more than 99% complete out to
θ ) 29°. The benzene solvent molecule displayed a large disorder,
and to resolve this, the ring was constrained to be a planar hexagon
with identical C-C distances.
Results
Synthesis and Characterization. Complexes 1 and 4 were
synthesized by chloride for triflate substitution as described
earlier for similar Pd complexes4c (cf. Scheme 1). This reaction
operates in moderate yields and gives a reaction mixture with
residual silver metal, which made the purification difficult.
However, repeated filtration and recrystallization gave pure
products that were characterized by NMR spectroscopy and
elemental analysis.
All crystallographic data in the CIF format are given in the
Supporting Information.
A modified procedure13 to synthesize compound 5 was
developed, using dry conditions and THF as solvent instead of
benzene. The 1H and 13C NMR spectroscopic data of the crude
product were in good agreement with the literature data, and
no further purification or characterization was done.
(16) Dyson, R.; Maeder, M.; Puxty, G.; Neuhold, Y.-M. Inorg. React.
Mech. 2003, 5, 39-46.
(17) Press: W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T.
Numerical Recipes in C: The Art of Scientific Computing; Cambridge
University Press: Cambridge, England, 1992.
(18) BrukerAXS SMART Area Detector Control Software; Bruker
Analytical X-ray Systems, Madison, WI, 1995.
Solvolysis. Even though 1 obviously undergoes a slow
protonolysis reaction in solution, the general stability to air and
moisture is good and in the solid state the compound can be
kept at ambient temperature and under air for a very long time.
Since the triflate ligand is known to be a good leaving group
(19) Sheldrick, G. M. SADABS, Program for Absorption Correction;
University of Go¨ttingen, Go¨ttingen, Germany, 1996.
(20) BrukerAXS SAINT Integration Software; Bruker Analytical X-ray
Systems, Madison, WI, 1995.
(21) Sheldrick, G. M. SHELXTL5.1, Program for Structure Solution and
Least-Squares Refinement; University of Go¨ttingen, Go¨ttingen, Germany,
1998.
(23) (a) Trovo´, G.; Bandoli, G.; Casellato, U.; Corain, B.; Nicolini, M.;
Longato, B. Inorg. Chem. 1990, 29, 4616. (b) Longato, B.; Bandoli, G.;
Dolmella, A. Eur. J. Inorg. Chem. 2004, 1092.
(24) Chatt, J.; Shaw, B. L. J. Chem. Soc. 1959, 4020. Cotton, F. A.;
Wilkinson, G. AdVanced Inorganic Chemistry, 4th ed.; Wiley: New York,
1980.
(22) Wiess, M. S.; Hilgenfeld, R. J. Appl. Crystallogr. 1997, 30, 203.
(25) Lawrance, G. A. Chem. ReV. 1986, 86, 17.

1288 Organometallics, Vol. 25, No. 5, 2006
Nilsson et al.
of Et4NCl) to a CD2Cl2 solution of 1 gave instant formation of
[PtPh(PMe2Ph)3]+ (7) and 2, respectively (cf. Table 2). The
reactions take place with equal amounts, as well as an excess,
of the entering ligands.
Transmetalation Reactions. The reaction of complex 1 with
Me3SnPh was studied in CD2Cl2, acetone-d6, and THF at room
temperature by 31P NMR spectroscopy (cf. Scheme 2).
Scheme 2
Figure 1. Diamond drawing with atomic numbering of the
molecular structure of complex 6. The ellipsoids denote 30%
probability. Bond distances (Å) and angles (deg) with estimated
standard deviations: Pt1-O4 ) 2.122(5), Pt1-P2 ) 2.260(2), Pt2-
O4 ) 2.117(5), Pt2-P1 ) 2.258(2); O4-Pt1-O4 ) 79.2(3), O4-
Pt1-P2 ) 170.26(17), 93,64(16), P2-Pt1-P2 ) 94.18(12), O4-
Pt2-O4 ) 79.5(3), P1-Pt2-P1 ) 94.38(12), O4-Pt2-P1 )
170.83(16), 93.32(16), Pt1-O4-Pt2 ) 100.7(2).
Complex 1 reacts with Me3SnPh, and there is a conversion
into two different products, trans-[PtPh2(PMe2Ph)2], (8) and cis-
[PtPh2(PMe2Ph)2], (9). The products were characterized by NMR
1
spectroscopy and display JPt-P values of 2899 and 1755 Hz,
typical of trans and cis configurations, respectively. The
chemical shifts for the cis product were confirmed by compari-
son with a complex synthesized elsewhere.27 An independent
synthesis of 8 by the addition of the Grignard reagent BrMgPh
to a THF solution of 2 was attempted. Instantly, only compound
9 was formed, as seen by 31P NMR spectroscopy. However, if
the THF solvent was removed and CH2Cl2 was added, 8 was
also formed within 2 days at room temperature. In this Grignard
synthesis two additional species were also observed, identified
as trans-[PtPhBr(PMe2Ph)2] (10) and trans-[PtBr2(PMe2Ph)2]
Table 1. Crystal Data and Details of Data Collection and
Refinement for Compound 6
chem formula
mol wt
cryst syst
space group
a/Å
C46H58F6O8P4Pt2S2
1431.10
monoclinic
C2/c
25.078(5)
13.604(3)
18.511(4)
110.12(3)
5930(2)
b/Å
c/Å
â/deg
1
V/Å3
(11), respectively.28 10 and 11 were characterized by H and
31P NMR spectroscopy (cf. Table 2).
Z
D
4
calcd/g cm-3
1.603
4.954
1.73-32.01
28 702
8972 (Rint ) 0.128)
298
0.0589, 0.1155
0.1402
The initial formation of 8 is faster than the formation of 9,
regardless of reaction conditions. The reactivity and the product
distribution ratio vary depending on the amount of added tin
compound, reaction temperature, and solvent. Moreover, in CD2-
Cl2 and acetone there is always 1 left when the reaction is over.
This indicates that the reaction in Scheme 2 is reversible, at
least under certain conditions.
µ/mm-1
θ range/deg
no. of collected rflns
no. of unique rflns
no. of params
R(F),a Rw(F2)b (I > 2σ(I))
Rw(F2) (all data)
Sc
0.995
To verify this reversibility, the equilibrium was accessed also
from the product side in separate experiments. To a solution of
1 and Me3SnPh in dichloromethane-d2, considered to be at
equilibrium, 5 was added in a 1:1 molar ratio with respect to
the platinum complex. This caused a small change of the
compound distribution in solution in favor of compound 1.
Leaving the reaction solution for 20 h at room temperature gave
no further changes. However, when an additional amount of 5
was added, the reaction was shifted further to the left, making
1 the most abundant platinum compound. The uncatalyzed cis-
trans isomerization is very slow, and addition of free PMe2Ph
to a solution of only 9 resulted in no changes even after several
days, although free phosphines are known to catalyze such
isomerizations.29 The reversible nature of the transmetalation
2
a R ) ∑(|Fo| - |Fc|)/∑|Fo|. b Rw ) [∑w(|Fo| - |Fc|)2/∑|Fo| ]1/2
.
c S )
[∑w(|Fo| - |Fc|)2/(m - n)]1/2
.
and is often used to activate metal complexes, one of our first
goals was to investigate whether solvolysis of the triflate ligand
in 1 occurs in solution. Our results point to a fast solvolysis of
the triflate ligand. Dissolving 1 in CD2Cl2 gives only one (apart
from coupling) sharp resonance in both 31P and 19F NMR
spectroscopy. Upon addition of LiOTf or Bu4N(OTf) to the
solution, the 19F NMR data gave no additional signal, indicating
that there is only free triflate present in solution. Addition of
Et4NBF4, on the other hand, results in an additional signal at
-
higher field, corresponding to the BF4 anion. Similar results
were obtained in other solvents. In addition, conductivity
measurements were performed on dichloromethane solutions of
1, 2, Bu4N(OTf), and Bu4NCl. The molar conductivities are
given in the Supporting Information, and they support the
occurrence of solvolysis of the triflate ligand, where the molar
conductivity of the solution of 1 is much higher than that of
the corresponding solution of 2. We cannot fully exclude the
possibility of a very fast exchange between free and coordinated
triflate, but there is no sign of any broadening of the signals in
the NMR spectra, and substitution reactions of similar com-
pounds are known to be slow on the NMR time scale.26
The lability of the triflate ion was further demonstrated by
the fact that addition of free PMe2Ph or chloride (in the form
(26) (a) Alibrandi, G.; Romeo, R.; Monsu` Scolaro, L.; Tobe, M. L. Inorg.
Chem. 1992, 31, 5061. (b) Romeo, R.; Arena, G.; Monsu` Scolaro, L.;
Plutonia, M. R. Inorg. Chim. Acta 1995, 240, 81.
(27) Wendt, O. F. Platinum(II) and Palladium(II) Complexes with Group
14 and 15 Donor Ligands. Synthesis, Structure and Substitution Mechanisms.
Thesis, Lund University, Lund, Sweden, 1997.
(28) No references for the NMR data were found in the literature.
However, the assignment was based on the fact that PR3 is known to have
a higher trans influence than Br. For references concerning analogous
platinum(II) complexes, see for example ref 9g and: Pham, E. K.; West,
R. J. Am. Chem. Soc. 1989, 111, 7667. Rahn, J. A.; O’Donnell, D. J.; Palmer,
A. R.; Nelson, J. H. Inorg. Chem. 1989, 28, 2631.
(29) Tobe, M. L.; Burgess, J. Inorganic Reaction Mechanisms; Long-
man: New York, 1999.

Mechanistic Model for Stille Couplings
Organometallics, Vol. 25, No. 5, 2006 1289
Table 2. NMR Data for Compounds 7-11
chem shift/δ (multiplicity, coupling const/Hz, assignta)
compd
nucleus
1H
7
1.15 (d, 12H, 2JP-H ) 8.79, 3JPt-H ) 26.0), 1.30 (t, 6H, 2JP-H ) 4.03, 3JPt-H ) 24.9),
6.93-7.00 (m, 3H, Pt-Ph), 7.07-7.14 (m, 2H, Pt-Ph), 7.27-7.49 (m, 15H, P-Ph)
-5.69 (d, 1JPt-P ) 2658, 2JP-P ) 21.9), -15.16 (t, 1JPt-P ) 1748, 2JP-P ) 22)
1.50 (vt, 12H, 2JP-H ) 3.66, 3JPt-H ) 17.2)
31P{1H}
1H
8b
9
31P{1H}
1H
-5.76 (s, 1JPt-P ) 2899)
1.10 (d, 12H, 2JP-H ) 7.49, 3JPt-H ) 28.8), 6.63-6.71 (m, 4H), 6.84-6.92 (m, 6H),
7.28-7.50 (m, 10H)
31P{1H}
-14.8 (s, 1JPt-P ) 1755)
10
11
1H
1.50 (t, 12H, 2JP-H ) 3.66, 3JPt-H ) 32.6), 6.76-6.82 (m, 3H, Pt-Ph), 7.08-7.11 (m, 2H, Pt-Ph),
7.34-7.40 (m, 6H, P-Ph), 7.63-7.68 (m, 4H, P-Ph)
-5.10 (s, 1JPt-P ) 2840)
31P{1H}
1H
0.96 (t, 12H, 2JP-H ) 5.49, 3JPt-H ) 68.12), 7.33-7.42 (m, 6H), 7.62-7.69 (m, 4H)
-15.9 (s, 1JPt-P ) 1929)
31P{1H}
a The assignment is only displayed for clarification when necessary. In all cases the solvent has been CD2Cl2. b Not all protons could be assigned due to
interference with protons from the Me3SnPh compound.
thus implies that the organotin reagents are catalysts for cis-
trans isomerization. On the basis of the Grignard experiment
this seems to be true also for organomagnesium reagents.
The reaction of complex 4 with Me3SnPh was investigated
1
in CD2Cl2 and benzene-d6 at room temperature using H and
31P NMR spectroscopy (cf. Scheme 3). Complex 4 reacts with
Me3SnPh, and within 6 h there is a complete conversion into
one product, identified as cis-[PtPhMe(PPh3)2] (12) by NMR
spectroscopy. After 18 h also cis-[PtPh2(PPh3)2] (13) was
identified in the reaction mixture, and the molar ratio between
12 and 13 was about 7:3. Over the course of another 48 h no
further changes were seen, making compound 12 the dominant
transmetalation product of the reaction in Scheme 3. Similar
observations were done in both CD2Cl2 and benzene-d6.
Complexes 12 and 13 have been identified elsewhere, giving
the same NMR signatures as observed here.30
Figure 2. Observed concentrations of reactant (0), intermediate
(]), trans (O), and cis (4) product as a function of time: (a) [1] )
9.6 mM, [Me3SnPh] ) 414 mM in CD2Cl2 at 25 °C (solid lines
denote the best fit using FACSIMILE); (b) [1] ) 11.0 mM, [Me3-
SnPh] ) 388 mM in THF/THF-d8 at 50 °C (solid lines denote the
best fit using Pro-KII).
Scheme 3
Table 3. Activation Parameters and Rate Constants for the
Different Pathways of the Proposed Equilibrium in Scheme
2 in CD2Cl2 at 25 °C
pathway
103k/M-1 s-1
∆Hq/kJ mol-1
∆Sq/J K-1 mol-1
k1
k-1
k2
3.4 ( 0.3
42 ( 11
1.3 ( 0.3
20 ( 5
36 ( 4
31 ( 4
55 ( 4
19 ( 4
-178 ( 13
-176 ( 13
-123 ( 14
-221 ( 13
k-2
As noted, triflate ions are easily substituted in solution by,
for example, chloride or free phosphine, which are typically
present in catalytic mixtures. This prompted us to investigate
the reactivity toward transmetalation also for complexes 2 and
7, respectively. Addition of Me3SnPh or Bu3Sn(CHdCH2) to a
solution of 2 in CD2Cl2 clearly showed that no transmetalation
reaction occurs between 2 and either stannane. Complex 7
displayed the same inertness against transmetalation reactions
with Me3SnPh. The transmetalation reaction in both cis and trans
positions is obviously facilitated by the lability of the triflate
ligand. Espinet and co-workers4c have shown that the formation
of trans-[Pd(C6Cl2F3)Cl(PPh3)2] (upon addition of LiCl to the
reaction solution) gives a retardation of the transmetalation rate
in comparison to the case when no free Cl- is used. Since the
stability of a Pt-Cl bond is even higher than that of the
corresponding Pd-Cl bond, the total inertness toward trans-
metalation in our system is in agreement with this. Also the
inertness of complex 7 can be explained by thermodynamic
considerations. Early, qualitative observations on Pt(II) systems
in transmetalation reactions concluded that strong σ-donors
dramatically slow the reaction rate.4c,9d-h Hence, the additional
coordination of a phosphine ligand, which is a relatively strong
σ-donor and a π-acceptor, hampers the transmetalation reaction
for complex 7. The question remains, though, why the presence
of strongly bonded ligands with a good bridging ability affects
the cis transmetalation so dramatically.
Kinetic Analysis. To understand the mechanism of these
transmetalation reactions, we performed a detailed kinetic study
of the equilibrium in Scheme 2. In this way it was possible to
study the transmetalation in both the cis and trans positions. It
was soon clear to us that any closed-form solution of a kinetic
scheme involving two equilibria would be excessively compli-
cated, and we turned to numerical fittings of the concentration-
time data. The reactions were always followed with the tin
compound in large excess compared to the concentration of 1.
This pushed the equilibrium to the right to a convenient extent.
Typical kinetic traces for the reaction are shown in Figure 2.
Two different kinetic modeling programs were used to fit the
experimental data to different mechanisms; very good correlation
between the programs was seen, and reasonably good fits were
obtained (cf. Figure 2). Experimental results are summarized
in Tables 3 and 4, and complete data are reported as Supporting
Information.
(30) For NMR characterization data of 12, see ref 10, and for complex
13, see ref 9g.

1290 Organometallics, Vol. 25, No. 5, 2006
Nilsson et al.
Table 5. Qualitative Influences on the Initial Rate of
Reactant Decay and Product Formation in the Reaction
Depicted in Scheme 2 upon Addition of Different Salts to the
Reaction Solutiona
Table 4. Rate Constants (103 M-1 s-1) for the Different
Pathways of the Proposed Mechanism in Scheme 4 in THF
at 50 °Ca
pathway
influence on the reacn rate
k3
k-3
k4
k-4
k5
k-5
entryb
additive
[additive]/mM
1
8
9
1.8 ( 0.4 4.8 ( 2.2 5.4 ( 0.9 330 ( 130 1.0 ( 0.3 18 ( 9
1
2
3
4
Bu4N(PF6)
Bu4N(PF6)
Bu4N(OTf)
Bu4N(OTf)
8.92
89.2
8.43
84.3
+
+
-
-
-
-
a The platinum(II) complex was trans-[PtPh(OTf)(PMe2Ph)2] (1). The
errors are given as 1 standard deviation.
+
+
none
-
none
-
In THF and acetone an intermediate is observed, and this of
course provided us with additional information when defining
the reaction mechanism. Starting with the data for THF
solutions, we tested a number of mechanisms (see the Support-
ing Information) and the best agreement was obtained with a
scheme consisting of two parallel and reversible reaction paths
leading to 8 and 9 with no uncatalyzed interconversion of 8
and 9 on the time scale of the reaction. The results from our
fitting procedure strongly suggest that the long-lived intermedi-
ate is on the reaction path to 9 only. Cases where the
intermediate is on the reaction path to 8, or common to the
paths to both products, were considered. In these cases, and
especially in the former one, the results displayed a low
significance in some of the determined rate constants as well
as large errors, which indicates that these models can be
discarded. In THF, we were able to identify the intermediate
by 31P and 1H NMR spectroscopy. In the 31P spectrum it has a
chemical shift of -4.1 ppm and a 1JPt-P value of 2960 Hz. This
strongly suggests a Pt(II) trans species. In the 1H NMR spectrum
a transient peak at -0.19 ppm was observed (2JP-H ) 6.5 Hz,
2JPt-H ) 49 Hz). Taken together, these data are compatible with
the intermediate being trans-[PtMePh(PMe2Ph)2] (14), and
chemical shifts and coupling constants agree with those
reported.10
a All comparisons refer to when no additive is used. All reactions were
done in dichloromethane-d2 solvent at room temperature with [1] ) 10.5-
14.3 mM and [Me3SnPh] ) 204-208 mM.
effects and reagent concentration effects. Furthermore, the salt
effects can be divided into two categories: one thermodynamic,
where the final distribution between the reactant and the products
is changed, and one kinetic, where the initial rates of formation
of the different products are affected. All these results are
qualitative in nature.
Addition of both the hexafluorophosphate and triflate salts
to the reaction solution favors the formation of products in
Scheme 2. This is most reasonably a consequence of an
increased stabilization of the products: e.g., the tin triflates.
One can also note that 9 is favored compared to 8, when neutral
salt is added, and this most probably has its origin in the higher
charge separation in solution, favoring the product with higher
dielectricity constant. With increased free triflate concentration,
8 is instead the favored product.
The qualitative influences on the reaction rate, corresponding
to the reactant decay and the product formation, are depicted
in Table 5. For the hexafluorophosphate the overall effect is
that the rate of reactant decay and the formation of 8 are
increased, whereas the rate of formation of 9 is decreased. When
the triflate salt is added, the trend is instead a retardation of the
reaction rate for all species involved (at lower concentrations
this concentration effect seems to be counterbalanced by the
“salt” effect that triflate also gives rise to). This observation
goes well in line with our suggestion of the necessity of a
solvolysis prior to transmetalation, and the presence of triflate
ions in solution probably disfavors the formation of the solvated
complex.
In CD2Cl2 no intermediate was observed, and the best fit was
obtained with two parallel reversible pathways leading to cis
and trans products. The fitting suggests that there is no common
intermediate in this case either, but of course one cannot exclude
the existence of intermediates in steady-state concentrations
along one or both pathways. In all cases, the adopted models
were fitted at various concentrations of [Me3SnPh] and [Pt],
giving good agreement for the calculated parameters in most
cases.
Activation parameters were determined for the reaction in
CD2Cl2 by fitting the Eyring equation to the rate constants at
different temperatures, and they are given in Table 3.
Discussion
Much effort has been devoted to the elucidation of the
mechanism of transmetalation reactions in the Stille coupling.
It has been concluded that both the outcome of the reaction
and the reaction mechanism are strongly dependent on the
reaction conditions used. Our observations are no exception,
but we will try to present a unified picture that explains most
of our observations. The present investigation was not compli-
cated by a consecutive reductive elimination reaction, as is
usually the case when palladium complexes are used. On the
other hand, one can always question the relevance of using
platinum when trying to understand palladium catalysis.
First of all, it seems clear that it is the solvento complex that
is the reactive species in all cases in the current system. This is
supported by a number of observations. First, we see a
significant change of the reactivity and product distribution in
different solvents and the reactivity follows the well-known
coordinating ability of the different solvents, indicating that the
solvent molecule is actually coordinated to the metal center and
that this coordination plays a crucial role in the overall reaction.
Without solvent coordination it is expected that the solvent with
Equilibrium constants were calculated from the determined
rate constants, and these agree with the qualitative estimations:
i.e., the reaction in Scheme 2 is an equilibrium in CD2Cl2,
whereas it goes more or less to completion in THF. In CD2Cl2
the calculated equilibrium constants for formation of 8 and 9
were 0.081 ( 0.023 and 0.065 ( 0.023, respectively. The final
thermodynamic product distribution was determined at room
temperature in CD2Cl2. The molar ratios for 1, 8, and 9 changed
from being 49:48:3 after 100 min to 6:35:59 after 6 days. No
further changes were observed even after several more days.
This indicates that 9 is slightly more stable than 8, which is
somewhat off the calculated ratio based on the equilibrium
constants from kinetics, which indicate a slightly higher stability
for the trans compound 8. It is clear that when the kinetic runs
were terminated, equilibrium was not yet established.
Salt Effects. The ionic strength effects during transmetalation
in CD2Cl2 were also investigated. They were adjusted in two
different ways, with the addition of Bu4N(PF6) or Bu4N(OTf).
In this way we were able to distinguish between proper salt

Mechanistic Model for Stille Couplings
Organometallics, Vol. 25, No. 5, 2006 1291
Scheme 4
the highest dielectricity constant, acetone, would support the
charge separation the best and give rise to the highest rate.
Second, we see a retarding effect upon adding a triflate salt, in
line with the formation of a less reactive neutral triflate complex.
Third, the fact that no transmetalation reaction takes place on
complex 2 or 7, where a chloro or phosphine ligand is
coordinated to the metal center, suggests that solvolysis and
the lability of the leaving group are important for the reactivity.
In addition, the conductivity measurements indicate the presence
of a cationic complex in solution. It should be noted that all
these effects apply to the reaction to both cis and trans products
(vide infra).
All the elementary steps in the mechanistic models are
bimolecular, first order in both platinum and stannane, and this
together with the activation parameters strongly indicates an
associatively activated reaction mode for the transmetalation
reaction in Scheme 2. Also, the fact that we have a strong steric
effect on the outcome of the transmetalation reaction indicates
associatively activated reaction paths. The entropies of activation
are all largely negative, and our values are in the same range
as those derived earlier for different palladium(II) systems.4c It
should be noted that the errors in the activation parameters are
large and that one should not draw any far-reaching conclusions
on the basis of their numerical values.
presence of a chloride or a THF molecule, instead of a
dichloromethane, would therefore hamper the reactivity in the
trans position, whereas the cis effect would be minor. Still, we
see an almost equal effect in both the cis and trans positions,
and this is easily rationalized if 14 is an intermediate on the
path to 9, also in dichloromethane. This is also in line with the
notion put forward by Espinet and co-workers that transmeta-
lations are similar to nucleophilic substitutions.4c Reacting 1
with chloride exclusively gives 2, and reacting 2 with iodide
exclusively results in substitution of chloride.10 From our work
on fluoro complexes where trans complexes are exclusively
formed, we know that these can be transformed into cis products
also with exchange of alkyl/aryl ligands.10 The fact that we see
more isomerization in the present system can probably be
explained by the large excess of organotin reagent used here.
The salt effects are also in agreement with the proposed
mechanism; neutral salt facilitates the initial formation of trans
product, whereas the isomerization to the cis product is less
influenced, as expected for a nonpolar transition state. The
organotin-catalyzed isomerization from 8 to 9 (k6 path in Scheme
4) was considered as an alternative to the intermediacy of 14.
In the end, this was not the best model, but it seems likely that
this pathway is operating, at least to some extent.
We suggest open transition states for the formation of trans
products, as shown in Scheme 4. We cannot distinguish between
the existence of a so-called σ-complex where the triflate (or
solvent) attacks the coordinated stannane and releases the R3-
SnX and an oxidative path where there is oxidation over the
Sn-C bond to form a Pt(IV) intermediate from which the R3-
Sn group is reductively eliminated. For the formation of the cis
product it seems most likely that we have a cyclic transition
state of the type pictured in Scheme 4. This would directly lead
to the cis configuration.4c,e Such a reaction would be favored
for bulkier phosphine ligands and could explain why the PPh3
complexes give exclusive formation of cis complexes.
It has been claimed that increased bridging ability of the
leaving ligand in transmetalation reactions should increase the
formation of the cis product, due to the enhanced possibility
for formation of a cyclic transition state.31 Recently we
investigated the reactivity of a platinum(II)-fluoro complex.10
Our results displayed an exclusive formation of trans products,
which was somewhat surprising in view of the good bridging
ability of the fluoro ligand. In addition, we were also able to
identify transmetalation over an Sn-C(sp3) bond, which was
rationalized in terms of steric demands of the ancillary phosphine
ligands. Thus, in the current system, where the leaving ligand
has a lower bridging ability than fluoride, an exclusive trans
product formation might also be expected. This is not the case,
however, and we instead observe a mixed product distribution
of cis and trans isomers when PMe2Ph is used as ancillary
ligand. In addition, and even more surprising, only cis products
are formed in the case of the PPh3 complex (Scheme 3). In this
case the steric effect again directs the transmetalation to occur
over the Sn-C(sp3) bond rather than over the Sn-C(sp2) bond.
In THF solution we observe that 14 is an intermediate on
the path to the cis product, and it therefore seems reasonable
that the initial attack on the triflate/solvento platinum compound
always takes place trans to the phenyl ligand. We propose that
this is true also in dichloromethane, although we cannot observe
any intermediate in this case and the kinetic modeling cannot
distinguish between a mechanism where 14 is present in steady-
state concentrations and where there is no intermediate. If the
initial attack takes place trans to the phenyl and the solvent is
always the leaving group, most of our observations are
explained. A comprehensive mechanism is given in Scheme 4.
The rate constants defined in Scheme 4 refer to the measure-
ments in THF. It is a well-known fact that it is the nature of
the trans ligand that, apart from steric reasons, dictates the
reactivity in square-planar complexes, and thus it is very difficult
to comprehend the lower reactivity of the THF complex and 2
and 7 also in the cis position if phosphine is the leaving group.
Especially chloride, with a low steric demand and good bridging
properties, would be expected to be a good cis ligand. The
Overall, it is difficult to know how relevant these results are
for transmetalation reactions of palladium complexes. A retarda-
tion of the reaction rate is often observed when free neutral
ligand is added to a reaction solution in the Stille coupling
reaction,4a,b and one explanation that has been offered is the
formation of a tris(phosphine) complex with lower reactivity.4c
This explanation is clearly substantiated by our finding that the
(31) (a) Labadie, J. W.; Stille, J. K. J. Am. Chem. Soc. 1983, 105, 6129.
(b) Reference 4c.

1292 Organometallics, Vol. 25, No. 5, 2006
Nilsson et al.
tris(phosphine) complex is completely unreactive. Also, it is
clear that if reductive elimination operates on a cis compound
the reversible nature of these reactions could easily funnel all
product out via the cis complex, even if substantial amounts of
trans compound are formed initially. The transformation of
trans-bis(phosphine) to cis-bis(phosphine) complexes mediated
by organotin compounds is evidently a possibility also in
palladium chemistry. In fact, isomerization reactions via bridging
hydrocarbyl complexes have been considered before.32
8 and 14, respectively, via an open transition state. In a
subsequent reaction, 14 (and possibly 8) reacts with another
molecule of stannane and gives 9 as the final product via a cyclic
transition state. Initial reaction taking place in the trans position
only explains the reactivity trends, where the reactivities to both
cis and trans products are found to be extremely sensitive to
the leaving-group ability of the ligand trans to the phenyl group.
Implications for the Stille reaction are further substantiation for
the retardation by added free ligand based on the formation of
unreactive tris(phosphine) complexes. Also, the reversibility of
the transmetalation reactions means that all organic product
could be funneled out via a cis complex, although there are
substantial amounts of trans complex initially. In essence, the
stannanes catalyze the cis-trans isomerization.
Conclusion
By choosing platinum(II) instead of palladium(II), we could
successfully perform a thorough kinetic and mechanistic inves-
tigation of the transmetalation reaction with organotin reagents,
having reasonable reaction rates and no subsequent reductive
elimination. The transmetalation reaction between trans-[PtPh-
(OTf)(PMe2Ph)2] (1) and Me3SnPh was investigated, and we
always observe two products, trans-[PtPh2(PMe2Ph)2] (8) and
cis-[PtPh2(PMe2Ph)2] (9), formed simultaneously. In THF we
were able to characterize, by NMR spectroscopy, an intermedi-
ate, trans-[PtPhMe(PMe2Ph)2] (14), on the path to 9. We
evaluated the kinetics for these reactions numerically and
propose a mechanism involving parallel equilibria to 8 and 9
with associative activation in all steps. In this mechanism the
initial attack always takes place trans to the phenyl group, giving
Acknowledgment. We thank Mr. Felix Plamper for great
assistance with the synthesis of the triphenylphosphine complex.
Financial support from the Swedish Research Council, the
Crafoord Foundation, the Knut and Alice Wallenberg Founda-
tion, and the Royal Physiographic Society in Lund is also
gratefully acknowledged.
Supporting Information Available: Tables giving detailed
crystallographic data, atomic positional parameters, and bond
lengths and angles in the CIF format, tables of all observed rate
constants as a function of temperature and concentration, and
primary data from the mechanistic fitting. This material is available
free of charge via the Internet at http://pubs.acs.org.
(32) Casado, A. L.; Casares, J. A.; Espinet, P. Organometallics 1997,
16, 5730.
OM051021A