A dissociative route to C-H bond activation: Ligand cyclometalation in cis-dimethyl[(sulfinyl-κS)bis[methane]][tris(2-methylphenyl)-phosphine] platinum(2+)

Article abstract of DOI:10.1002/hlca.200590035

The dialkyl compound cis-dimethyl[(sulfinyl-κS)bis[methane]][tris(2- methylphenyl)phosphine]platinum(2+) (cis-[Pt(Me)₂(dmso)(P(o-tol) ₃]; 1) has been isolated from the reaction of cis- dimethylbis[(sulfinyl-κS)bis[methane]]platinum(2+) (cis-[Pt(Me) ₂(dmso)2]) with tris(2-methylphenyl)phosphane (P(o-tol)₃). Restricted rotation around the P-C_ipso bonds of the phosphane ligand generates two different conformers, 1a and 1b, in rapid exchange in non-polar solvents at low temperature. Strong through-space contacts between the ortho-Me substituent groups on the ligand and the cis-Me groups in the coordination plane were determined, which proved useful for identifying the atropisomers formed. At room temperature, ¹H-NMR spectra of 1 maintain a 'static' pattern upon onset of easy and rapid orrto-platination, leading to [[2-[bis(2- methylphenyl)-phosphino-κP]phenyl]methyl-κC]methyl[(sulfinyl- κS)bis[methane]]platinum(2+) (2), a new C,P-cyclometalated compound of platinum(II), with liberation of methane. The process has been studied by ¹H- and ³¹P{¹H}-NMR in CDCl3, and kinetics experiments were performed by conventional spectrophotometric techniques. The first-order rate constants k_c decrease with the addition of dimethyl sulfoxide until the process is blocked by the presence of a sufficient excess of sulfoxide. This behavior reveals a mechanism initiated by ligand dissociation and formation of a three-coordinate species. The value of the rate constant for dimethyl sulfoxide dissociation k₁ has been measured independently over a wide temperature range by both ¹H-NMR ligand exchange (isotopic labeling experiments) and ligand substitution (stopped-flow pyridine for dimethyl sulfoxide substitution). The rates of the two processes are in reasonable agreement at the same temperature, and a single Eyring plot can be constructed with the two sets of kinetics data. However, the value of the derived dissociation constant at 308 K (k₁ = 6.5 ± 0.3 s ^-1) is at least two orders of magnitude higher than that of cyclometalation (k_c = 0.0098 ± 0.0009 s^-1 at 308 K). Clearly, the dissociation step is not rate-determining for cyclometalation. A multistep mechanism consistent with mass-law retardation is derived, which involves a pre-equilibrium that controls the concentration of an unsaturated three-coordinate, 14-electron T-shaped cis-[PtMe₂{P(o-tol) ₃}] intermediate. Cyclometalation is initiated in this latter by an agostic interaction with the o(C-H) orbital of a methyl group. Oxidative addition of the C-H bond follows, yielding a cyclometalated-hydrido 16-electron Pt(IV) five-coordinate intermediate. Finally, reductive elimination and re-entry of dimethyl sulfoxide with liberation of methane should yield the cyclometalated species 2.

Full text of DOI:10.1002/hlca.200590035

Helvetica Chimica Acta ± Vol. 88 (2005)
507
A Dissociative Route to CÀH Bond Activation: Ligand Cyclometalation in
cis-Dimethyl[(sulfinyl-kS)bis[methane]][tris(2-methylphenyl)-
phosphine]platinum(2‡)
by Raffaello Romeo*a), Maria Rosaria Plutinob), and Andrea Romeoa)
a
Á
) Dipartimento di Chimica Inorganica, Chimica Analitica e Chimica Fisica ± Universita di Messina,
Salita Sperone, 31± Vill. S. Agata, I-98166 Messina (fax: ‡ 39-090-393756; e-mail: raf.romeo@chem.unime.it)
b
Á
) Istituto Materiali Nanostrutturati, ISMN ± CNR, Sezione di Palermo, Unita di Messina, Salita Sperone,
31± Vill. S. Agata, I-98166 Messina
Â
Dedicated to Prof. Andre E. Merbach in recognition of his prominent contribution to the advancement of
inorganic and organometallic reaction mechanisms
The dialkyl compound cis-dimethyl[(sulfinyl-kS)bis[methane]][tris(2-methylphenyl)phosphine]platin-
um(2‡) (cis-[Pt(Me)2(dmso)(P(o-tol)3]; 1) has been isolated from the reaction of cis-dimethylbis[(sulfinyl-
kS)bis[methane]]platinum(2‡) (cis-[Pt(Me)2(dmso)2]) with tris(2-methylphenyl)phosphane (P(o-tol)3). Re-
stricted rotation around the PÀCipso bonds of the phosphane ligand generates two different conformers, 1a and
1b, in rapid exchange in non-polar solvents at low temperature. Strong through-space contacts between the
ortho-Me substituent groups on the ligand and the cis-Me groups in the coordination plane were determined,
which proved useful for identifying the atropisomers formed. At room temperature, 1H-NMR spectra of 1
maintain a ꢀstaticꢁ pattern upon onset of easy and rapid ortho-platination, leading to [[2-[bis(2-methylphenyl)-
phosphino-kP]phenyl]methyl-kC]methyl[(sulfinyl-kS)bis[methane]]platinum(2‡) (2), a new C,P-cyclometa-
lated compound of platinum(II), with liberation of methane. The process has been studied by 1H- and 31P{1H}-
NMR in CDCl3, and kinetics experiments were performed by conventional spectrophotometric techniques. The
first-order rate constants kc decrease with the addition of dimethyl sulfoxide until the process is blocked by the
presence of a sufficient excess of sulfoxide. This behavior reveals a mechanism initiated by ligand dissociation
and formation of a three-coordinate species. The value of the rate constant for dimethyl sulfoxide dissociation k1
has been measured independently over a wide temperature range by both 1H-NMR ligand exchange (isotopic
labeling experiments) and ligand substitution (stopped-flow pyridine for dimethyl sulfoxide substitution). The
rates of the two processes are in reasonable agreement at the same temperature, and a single Eyring plot can be
constructed with the two sets of kinetics data. However, the value of the derived dissociation constant at 308 K
(k1 ˆ 6.5 Æ 0.3 sÀ1) is at least two orders of magnitude higher than that of cyclometalation (kc ˆ 0.0098 Æ
0.0009 sÀ1 at 308 K). Clearly, the dissociation step is not rate-determining for cyclometalation. A multistep
mechanism consistent with mass-law retardation is derived, which involves a pre-equilibrium that controls the
concentration of an unsaturated three-coordinate, 14-electron T-shaped cis-[PtMe2{P(o-tol)3}] intermediate.
Cyclometalation is initiated in this latter by an agostic interaction with the s(CÀH) orbital of a methyl
group. Oxidative addition of the CÀH bond follows, yielding a cyclometalated-hydrido 16-electron Pt(IV) five-
coordinate intermediate. Finally, reductive elimination and re-entry of dimethyl sulfoxide with liberation of
methane should yield the cyclometalated species 2.
Introduction. ± Cycloplatinated compounds have been extensively studied [1] and
are of interest for several reasons, including their attractive photochemical and
photophysical properties [2], their potential use as molecular devices [3], and, more
generally, for their potential utility in catalysis and organic synthesis [4]. Several
examples can be found in platinum ± phosphane chemistry [5]. Recently, we focused
ꢂ 2005 Verlag Helvetica Chimica Acta AG, Zürich

508
Helvetica Chimica Acta ± Vol. 88 (2005)
our interest towards understanding whether there is a specific role for cyclometalation
and for the in-plane disposition of single aryl ligands in determining the reactivity
(through metal-to-ligand p-back-bonding) and/or the reaction mechanism in substitu-
tion reactions at PtII complexes. A clear demonstration that this is not the case came
from the comparison of the structural features and the reactivities of cycloplatinated
compounds of the type [Pt(NÀNÀC)Cl] (NÀNÀC ˆ anionic ligands derived from
deprotonated 6-substituted-2,2'-bipyridines) [6] or [Pt(CÀC)(SR2)2] (CÀC ˆ 2,2'-
biphenyl dianion; R ˆ Me and Et) [7] with those of corresponding substrates having the
same set of donor atoms but less-constrained arrangements of ligands. We found also
that complexes of the type cis-[Pt(Me)2(dmso)(PR3)] (PR3 ˆ isosteric tertiary
phosphanes) [8] behave like complexes having a set of donor atoms cis-[Pt(C,C)(S,S)]
(C ˆ strong s-donor carbanions; S ˆ thioethers or sulfoxides) for which a dissociative
mechanism for ligand exchange and substitution of the S-bonded ligand is well
established [9]1). The introduction of a s-donor PR3 in the set of ligand donor atoms
neither produces a change of mechanism nor greatly affects the ease with which the
dissociative activation takes place. During the synthetic work-up, we observed that the
phosphane-containing compound cis-dimethyl[(sulfinyl-kS)bis[methane]][tris(2-meth-
ylphenyl)phosphine]platinum(2‡) (cis-[Pt(Me)2(dmso)(P(o-tol)3)]; 1) in non-polar
solvents undergoes easy and rapid ortho-platination, leading to [{2-[bis(2-methylphe-
nyl)phosphino-kP]phenyl}methyl-kC]methyl[(sulfinyl-kS)bis[methane]]platinum(2‡)
(2) with liberation of methane. Therefore, we decided to perform a detailed kinetics
study of this process to gain some insight into the role of ligand dissociation in the
mechanism of intramolecular CÀH bond activation.
Results. ± 1H-NMR Characterization. The synthesis of compound 1 from cis-
[Pt(Me)2(dmso)2], must be carried out with care at low temperature, avoiding local
excess of the reagent P(o-tol)3. The reaction product was isolated in high yield as a solid
1
and was characterized by elemental analysis and H- and 31P{1H}-NMR spectroscopy.
1
The low-temperature H- and 31P-NMR spectra in CDCl3 showed that the molecule
exists in two differently populated conformations (1a/1b ca. 4 :1), each of which display
the following features: i) a 31P{1H} singlet flanked by platinum satellites (d 28.3,
1J(Pt,P) ˆ 1824 Hz for 1a and d 27.1, 1J(Pt,P) ˆ 1881 for 1b), ii) a set of aromatic proton
signals, iii) a set of three different singlets for the ortho-Me substituent groups of the
triarylphosphane ligand (d 2.84, 1.83, and 1.55 for 1a and d 3.00, 2.05, and 1.53 for 1b),
3
iv) two signals for the Me protons of the DMSO ligand (d 2.94, J(Pt,H) ˆ 13.9 and
3
3
3
2.20, J(Pt,H) ˆ 13.1 for 1a; d 3.06, J(Pt,H) ˆ 15.6 and 2.88, J(Pt,H) ˆ 16.0 for 1b),
each showing that the two Me groups on the S-atom are diastereotopic2). These
observations are entirely compatible with hindered rotation around the PÀCipso aryl
bonds while MÀP rotation remains rapid. This has been suggested before for P(o-tol)3
and other bulky phosphanes in organometallic compounds [10]. When the temperature
was increased, the sulfinyl Me signals underwent only minor broadening, and coupling
1
)
This reference should also be consulted for a discussion of three coordinate intermediates and previous
work in the field.
Because the Me groups in this molecule are diastereotopic, they would not be expected to be equivalent
under conditions where rotation about the PtÀS bond is fast.
2
)

Helvetica Chimica Acta ± Vol. 88 (2005)
509
with 195Pt was lost, but the signals did not become isochronous. Likewise, the expected
collapse of the 2-methylphenyl subspectrum was not observed, and three distinct
signals were still seen at room temperature for each rotamer. Thus, the two
conformations maintain apparently stereochemically rigid structures in solution
because of the high activation barrier for rotation around the PÀCipso bonds. Typically,
restricted PÀCPh bond rotation has been observed at low temperature [10][5b], and
only one case in which it was still evident at room temperature has been reported [5a].
A proper kinetics study of the fluxional behavior was not possible because 1 was easily
1
converted to the cyclometalated compound 2. Interestingly, the H- and 31P-NMR
spectra of analogous cis-[Pt(Me)2(dmso)(PR3)] complexes in which the phosphanes
are unhindered were temperature independent, thus indicating that MÀP, M ÀS, and
PÀC bond rotations remain rapid on the NMR time scale [8]. The 1H-NMR spectrum
of 1 is completed by two sets of multiplets for the coordinated Me groups with values at
d 0.47, 2J(Pt,H) ˆ 79.2 and 0.43, 2J(Pt,H) ˆ 69.8 for 1a and d 0.30, 2J(Pt,H) ˆ 69.8 and
2
0.22, 2J(Pt,H) ˆ 78.4 for 1b. The lowest value of J(Pt,H) belongs to the Me group trans
to the strong s-donor PR3.
The proton resonances belonging to the species 1a and 1b were assigned through
NMR spectroscopy from the connectivities in 2D-COSY and NOE cross-peaks in 2D-
NOESY experiments. Exo2 conformations3) for both rotamers 1a and 1b are fully
confirmed by NOE cross-peaks between the ortho-Me substituents of the triarylphos-
phane and in the sulfoxide ligands and those coordinated to the metal, e.g., MeÀPt
trans to the S and P ligands (see Supplemental Material, Table S1 for the relevant
observed NOE dipolar contacts4)). In compound 1a, strong inter-ligand NOE cross-
peaks are evident between PPhbÀMe and PPhcÀMe with PtÀMe (trans to the sulfinyl
group), and between the C(6)ÀH of the PPha fragment with PPhbÀMe, PPhcÀMe and
with the sulfinyl Me groups. This experimental evidence is represented with the sketch
on the left side of Scheme 1, in which the two ortho-Me substituents that lie above and
below the coordination plane are directed toward the PtII center (i.e., PPhbÀMe and
PPhcÀMe), while the third ortho-Me (i.e., PPha) is rotated with the C(6)ÀH closer to
the metal. As far as the rotamer 1b is concerned, an exo2 conformation is confirmed by
selective cross-peaks between the PtÀMe trans to the phosphane ligand with
PPhbÀMe, and the PtÀMe trans to the sulfoxide group with PPhcÀMe , PPhbÀMe,
and the C(6)ÀH of PPha. This conformation is represented in Scheme 1 (right side),
where the positions of all three 2-methylphenyl groups are exchanged relative to the
left side. The presence of an intramolecular ring exchange via correlated ring rotation
for the triarylphosphine conformers 1a and 1b, together with mutual intermolecular
exchange between the two rotamers 1a and 1b is confirmed by the multiplicity of very
diagnostic exchange cross-peaks between all six ortho-Me substituents (1a d 2.84,
PPhcÀMe; 1.83, PPhbÀMe; 1.55, PPhaÀMe; 1b d 3.00, PPhc'ÀMe; 2.05, PPhb'ÀMe;
1.53, PPha'ÀMe) and the four Me signals relative to the DMSO group (1a d 2.94,
SÀMeb; 2.20, SÀMea; 1b d 3.06, SÀMeb'; 2.88, SÀMea'). As shown in Fig. 1, the Me
3
)
For a regular trigonal pyramid constructed with the metal center as apex and the three para ring C-atoms
as the base, a proximal (exo) substituent points away from the base while a distal (endo) substituent points
toward the base. The term exo2 defines the number of proximal ortho Me groups.
4
)
To view Supplemental Material, please contact the corresponding author.

510
Helvetica Chimica Acta ± Vol. 88 (2005)
Scheme 1
Fig. 1. Aliphatic section of the phase-sensitive 1H-2D-NOESY spectrum of a CDCl3 solution of the two rotamers
1a and 1b at 255.6 K. The black cross-peaks arise from exchange, whereas the red cross-peaks arise from NOEs.
resonances are related to the others by five exchange cross-peaks for the ortho-Me
substituents of the triarylphosphane ligands, three exchange cross-peaks for the Me
groups bound to the sulfur, and a selective exchange cross-peak between the Me groups

Helvetica Chimica Acta ± Vol. 88 (2005)
511
directly bound to the PtII center (1a, 0.47 vs. 1b, 0.22 for PtÀMe trans to DMSO; 1a,
0.43 vs. 1b, 0.30 for PtÀMe trans to PR3). This experimental evidence accounts for
intramolecular ring rotation and for intermolecular exchange between the species 1a
and 1b. The mechanisms of ring exchange in triarylphosphanes via correlated ring
rotation have been extensively analyzed by Mislow et al. [11]. The most-likely
mechanisms involve a three-ring flip in the exo2 conformation, which exchanges the
proximal 2-methylphenyl groups but retains the identity of the distal ring, and a two-
ring flip mechanism in which ring exchange occurs via an exo3 intermediate.
The reaction of 1 with pyridine was carried out in situ in an NMR tube. On
exchanging pyridine for DMSO, the molecule loses the rigid structure of its precursor,
1
and the H-NMR spectrum at room temperature shows a single resonance for all Me
substituents of the 2-methylphenyl groups at d 2.04.
Cyclometalation Kinetics. On standing at room temperature in CHCl3, compound 1
yielded the cyclometalated 2 by ortho metalation with loss of CH4 as shown in
Scheme 2. Compound 2 was characterized by elemental analysis and 1H- and 31P-NMR
spectra. The 1H-NMR spectrum at room temperature showed the expected pattern of
signals for the H-atoms of the CH2 group (d 3.18, 2J(Pt,H) ˆ 87.9), of the Me groups of
the coordinated DMSO group (d 2.89, 3J(Pt,H) ˆ 14.4), and of the Me groups of the (2-
methylphenyl)phosphane (at d 2.59). These data indicate fast rotation on the NMR
time-scale of the PÀCipso bond of the 2-methylphenyl rings. The resonance of the Me
group bound to Pt appears as a doublet of pseudotriplets due to coupling with 195Pt
(69 Hz) and with 31P (6.6 Hz). The low value of the coupling in the 31P-NMR spectrum
(1J(Pt,P) ˆ 1920) gives unmistakable evidence that the P-atom is trans to the strong s-
donating Me group.
Scheme 2
1
The cyclometalation reaction shown in Scheme 2 can be followed easily by H- or
31P-NMR in CDCl3. There is no evidence for the accumulation of significant amounts of
any other intermediate species, and only the starting complex 1, the cyclometalated
complex 2, and methane are found in solution. The primary kinetics data were collected
1
by monitoring the decrease of the H-NMR signals of the coordinated Me group and
the concomitant increase of the methane signal. The reaction goes to completion in
30 min at 358. The molar fraction F of free CH4 was obtained by integration of the
signals monitored, and the time dependence of F did not follow a simple exponential
relationship as expected for a first-order process. Estimates of the initial rates showed F
to be also dependent, to some extent, on the concentration of the starting complex; F

512
Helvetica Chimica Acta ± Vol. 88 (2005)
decreased with increasing concentration of 1. These deviations could be due to auto-
dissociation of both 1 and 2 and were compensated by adding a sufficient excess of
DMSO to ensure that its concentration was essentially constant throughout the
reaction. Thus, a simple exponential relationship holds (see Fig. S1, Supplemental
Material4)).
1
Although the H- and 31P-NMR spectra provide a clear demonstration of the
identities of the reagents and products as well as the absence of any detectable
intermediate species, the method suffers from the disadvantage that relatively large
concentrations of the complex must be used. The UV/VIS spectrophotometric
technique requires far less of the complex, which can be examined at concentrations
of less than 0.5 Â 10À4 m, allowing a sufficient excess of DMSO to ensure pseudo-first-
order kinetics. Spectrophotometric kinetics experiments were carried out in CHCl3
solutions in a two-chambered silica cell in the thermostatted cell compartment of the
spectrophotometer and were monitored by repetitive scanning of the spectrum at
suitable times. The reactions, carried out in the presence of sufficient excess DMSO,
went to completion, the final spectra being identical to those of solutions of authentic
samples of 2. All reactions obeyed a first-order rate law until well over 90%
conversion. Rate constants kobs (sÀ1) were obtained from non-linear least-squares fits of
the experimental data to At ˆ A1 ‡ (A0 À A1) exp(À kobs t) with A0, A1, and kobs as the
parameters to be optimized (A0 ˆ absorbance after mixing of reagents, A1 ˆ absorb-
ance at completion of reaction) [12]. The pseudo-first-order rate constants kobs for the
reaction shown in Scheme 2 for a range of DMSO concentrations are reported in
Table 1 and illustrated in Fig. 2, which also shows the dependence of kobs (and of the
reciprocal of kobs (in the inset)) on the DMSO concentration. The variable-temperature
³
rate constants were fit to the Eyring equation to give estimates of DH ˆ 82 Æ 4 kJ molÀ1
³
À1
and DS ˆ À50 Æ 1 0 J K molÀ1.
Fig. 2. Plot showing the dependence of the observed pseudo-first-order rate constants (kobs/sÀ1) as a function of
different amounts of free DMSO for the reaction in Scheme 2. Reaction conditions: 0.05 mm 1, CHCl3, 308 K.

Helvetica Chimica Acta ± Vol. 88 (2005)
513
Table 1. Concentration of DMSO and the Temperature Dependence for Rates of Cyclometalation of 1a)
T [K]
DMSO Concentration [mm]
1
4 kobs [sÀ1
]
0
308
308
308
308
308
308
308
308
308
308
308
308
298
303
308
313
318
0.0246
0.0320
0.0490
0.118
0.192
0.246
0.493
0.737
0.985
1.46
2.10
4.93
1.55
1.55
1.55
1.55
1.55
95.7 Æ 0.3
57.4 Æ 0.2
52.9 Æ 0.1
33.7 Æ 0.1
21.9 Æ 0.1
17.3 Æ 0.1
9.06 Æ 0.02
7.24 Æ 0.01
5.20 Æ 0.01
3.80Æ 0.01
2.61 Æ 0.01
1.11Æ 0.01
1.00Æ 0.01
1.50Æ 0.01
2.70Æ 0.01
5.00 Æ 0.01
7.90 Æ 0.01
a) In CHCl3 by spectrophotometry.
Substitution Reactions. Nucleophilic substitution reactions of 1 with pyridine in
CHCl3 to yield dimethyl(pyridine)[tris(2-methylphenyl)phosphine]platinum(2‡) (3)
were carried out over a range of temperatures and pyridine concentrations and
required the use of the stopped-flow technique. Rate constants were evaluated by the
Applied Photophysics software package [13] and are reported in Table 2 as average
values from five to seven independent experiments.
Table 2. Concentration of Pyridine and Temperature Dependence of the Substitution Rates of 1a)
T [K]
Pyridine concentration [mm]
kobs [sÀ1
]
278.0
283.0
288.0
288.2
293.125
296.6
296.6
296.6
303.0
25
25
25
25
0.250 Æ 0.008
0.368 Æ 0.006
0.782 Æ 0.005
0.945 Æ 0.004
Æ 0.05
.62
1
5
10
25
25
1.66Æ 0.05
1.78 Æ 0.08
1.72Æ 0.07
4.34 Æ 0.05
a) In CHCl3 by spectrophotometry.
Isotope-Exchange Reactions. The kinetics of ligand exchange between (D6)DMSO
and 1 were performed by adding known volumes of (D6)DMSO with a microsyringe to
a prethermostatted solution of precisely weighed 1 in CDCl3. The excess of (D6)DMSO
over complex used in each experiment was at least fivefold. The isotope exchange was
followed by monitoring the increase in intensity of the 1H-NMR signal of free

514
Helvetica Chimica Acta ± Vol. 88 (2005)
nondeuterated DMSO at d 2.63 and the concomitant decrease of the signals of the Me
groups on the S-atom coordinated to the metal (at d 2.98 and 2.24), as shown in Fig. 3.
The spectra were recorded at appropriate time intervals. The mole fraction F
(ˆ [DMSO]f/([DMSO]f ‡ [DMSO]b) of free DMSO was obtained by integration of the
signals and the first-order rate constants kexch (sÀ1) for the exchange of the label were
obtained from a non-linear least-squares fit of the experimental data to F ˆ c1 ‡ c2
exp(À kexch t), with c1, c2, and kexch as the parameters to be optimized [12]. A similar
analysis can be performed on the mole fraction of bound DMSO. From the McKay
equation, Rexch ˆ kexch ab (a ‡ b)À1 (where Rexch is the rate of the exchange process; a is
the concentration of complex and b is the concentration of free sulfoxide) the pseudo-
first-order rate constants, kobs ˆ Rexch/a, were calculated and are reported in Table 3.
Fig. 3. 1H-NMR Spectra (300-MHz) showing DMSO exchange for 2.37 mm 1 and 84.3 mm (D6)DMSO in
3
CDCl3 at 257.8 K as a function of time. Coordinated DMSO (open circles, d 2.94 ppm, J(Pt,H) ˆ 13.9 Hz and
2.20 ppm, 3J(Pt,H) ˆ 13.1); free DMSO (filled circles, d 2.63 ppm). Inset: Exponential curves for the molar
fractions of coordinated (open circles) and free DMSO (filled circles) as a function of time.

Helvetica Chimica Acta ± Vol. 88 (2005)
515
Table 3. Concentration of (D6)DMSO and the Temperature Dependence for Rates of DMSO Exchange on 1a)
T [K]
(D6)DMSO Concentration [mm]
1
3 kobs [sÀ1
]
0
230.5
233.5
234.5
241.8
244.0
246.9
248.8
253.185.2
255.6
255.6
255.6
255.6
255.6
257.8
261.7
263.8
267.2
84.6
162.5
85.2
85.0
81.1
84.6
81.7
0.20 Æ 0.05
0.27 Æ 0.01
0.38 Æ 0.04
0.74Æ 0.02
1.57Æ 0.09
1.5Æ 0.2
2.2Æ 0.1
Æ 0.4
5.4
50.3
81.8
120
180
250
84.3
81.7
84.2
81.2
6.6 Æ 0.2
6.4Æ 0.3
6.4 Æ 0.1
6.3 Æ 0.2
6.5 Æ 0.4
8.8 Æ 0.6
15.5Æ 0.8
22 Æ 1
39Æ 2
a
1
) At 2 ± 5 mm in CHCl3 by H-NMR isotope-exchange.
Discussion. ± Substitution Mechanism. The rates of both reactions, i.e., DMSO
exchange and pyridine for DMSO substitution, on 1 are not affected by the
concentration of the added ligand (pyridine or (D6)DMSO; rate data in Tables 2 and
3). The temperature dependency of the substitution reactions (with pyridine) was
determined at a higher temperature range than that of the isotopic labeling
((D6)DMSO). From the overall set of data, it was possible to construct a single
Eyring plot, shown in Fig. 4, which encompasses the wide temperature range and
indicates that DMSO exchange and pyridine substitution have identical values at the
³
same temperature. Values for the activation parameters, DH ˆ 81 Æ 2 kJ molÀ1 and
³
DS ˆ 32 Æ 6 J KÀ1 molÀ1 were calculated by linear-regression analysis of this plot.
The experimental kinetics evidence points unequivocally to a dissociative mode of
activation. The assessment of a dissociative mechanism can be made essentially on the
basis of: i) the dissociation rate being independent of the nature and concentration of
the entering group, ii) the rate of dissociation being identical to the rate of ligand
³
exchange, and iii) the positive sign and magnitude of DS . The solvent used was not
sufficiently coordinating to actively contribute to associative processes. The pattern of
behavior is strictly similar to that found in our previous study on cis-[PtMe2(PR3)-
(dmso)] complexes [8] or on cis-[PtMe2(dmso)2] [14][15] and other dialkyl and diaryl
systems [9]1) [16], where the easy dissociation of thioethers or sulfoxides was shown to
be the rate-determining step of the substitution. The same dissociative mechanism
(Scheme 3) applies, involving i) dissociative loss of DMSO from the substrate (k1 path)
to yield a three-coordinate 14-electron T-shaped intermediate, ii) competition for the
intermediate between re-entry of DMSO (via kÀ1) and attack of another ligand (either
(D6)DMSO or pyridine via k2) to yield the observed products. Application of the
steady-state approximation to the three-coordinate intermediate gives the rate law in
Eqn. 1:

516
Helvetica Chimica Acta ± Vol. 88 (2005)
Fig. 4. Eyring plot constructed with rate data from isotope-labeling experiments (filled squares) and pyridine/
DMSO substitution reactions (open squares) on 1
Scheme 3
kobs ˆ k1 k2[Nu]/{(kÀ1[DMSO] ‡ k2[Nu]}
(1)
The steady-state treatment requires that the two processes in competition for the
intermediate, kÀ1[DMSO] and k2[Nu], have comparable values. For the limiting
situation k2[Nu] ꢀ kÀ1[DMSO], Eqn. 1 reduces to kobs ˆ k1. For 1 at 308 K, k1 ˆ 6.5 Æ
³
³
0.3 sÀ1; DH ˆ 81 Æ 2 kJ molÀ1, and DS ˆ 32 Æ 6 J KÀ1 molÀ1, compared to k1 ˆ 0.062 sÀ1,
DH ˆ 90 Æ 2 kJ molÀ1, and DS ˆ 32 Æ 6 J KÀ1 molÀ1 for cis-[PtMe2(dmso)(P(p-tol)3)]
[8]. As expected, steric congestion (the Tolmanꢁs cone angle for P(o-tol)3 is 1788)
makes dissociation and removal of DMSO from 1 faster and easier than in a
comparatively unhindered parent compound (the cone angle for P(p-tol)3 is 1458).
Strong s-donation from the carbanions and from the coordinated phosphane confers
³
³

Helvetica Chimica Acta ± Vol. 88 (2005)
517
sufficient thermodynamic stability to the T-shaped intermediate formed from
dissociation of 1.
Cyclometalation Mechanism. The conversion of 1 to 2 shows a remarkable rate
effect: the process is slowed by the addition of DMSO and halted in the presence of
high concentrations of added ligand. The rate law (CHCl3, 308 K) is of the type kobs
ˆ
a/(b[DMSO] ‡ 1) and values of a ˆ 0.0098 Æ 0.0009 sÀ1 and b ˆ (1.76Æ 0.01)  104 mÀ1
were derived from a non-linear least-squares analysis of the curve in Fig. 2 or from a
linear fit of the plot in the inset5). The value of a calculated (rate of cyclometalation at
308 K, in the absence of added DMSO) differs markedly and is very much smaller than
that of k1 ˆ 6.51 Æ 0.3 sÀ1 at 308 K. As matter of fact, it must be concluded that
dissociation of DMSO from 1 and formation of the three-coordinate 14-electron
intermediate is a preliminary step but not the rate-determining step for cyclometala-
tion, because the subsequent cyclometalation takes much longer to occur. In
formulating a mechanism, it must be assumed that, even in the absence of added
DMSO, the rates for the dissociation and reassociation steps in Scheme 3 are much
greater than the rate of CÀH bond activation in the three-coordinate intermediate
(k2 ꢁ k1 or kÀ1). In other words, this means that a pre-equilibrium (defined by k1/kÀ1
ˆ
Ke) is established, and the cyclometalation is controlled by two factors, namely i) the
extension of the dissociation pre-equilibrium, which determines the concentration of
the three coordinate intermediate and, ii) the subsequent CÀH bond activation on this
latter, k2 assumed as the rate-determining step. A pre-equilibrium treatment to a
reaction scheme similar to that in Scheme 3, where dissociation enables cyclometala-
tion instead of nucleophilic attack leads to the rate law in Eqn. 2:
kobs ˆ k2/(KÀe 1 [DMSO] ‡ 1
)
(2)
From this rate expression, it is possible to estimate a value for k2, the rate constant
for cyclometalation of the three-coordinate intermediate, which corresponds to a ˆ
0.0098 Æ 0.0009 sÀ1 and for Ke, the equilibrium constant for dissociation, which
corresponds to bÀ1 ˆ Ke ˆ (5.7 Æ 0.1)  10À5 m. A much smaller value of Ke ˆ (4.3 Æ
0.8) Â 10À7 m was reported by Espinet and co-workers [17] for the equilibrium in CDCl3
between the 16-electron species cis-[Pd(2-C6BrF4)2(tht)2] (tht ˆ tetrahydrothiophene)
and the 14-electron species obtained from ligand (tht) dissociation, a key intermediate
for the subsequent overall atropisomerization process. On the basis of the value of Ke
derived by us is reliable, at the low concentrations of complex (ca. 0.05 mm) used for the
spectrophotometric kinetics experiments, we would have expected extensive dissoci-
ation of the complex in solution, which does not seem to be the case.
Once we have provided convincing evidence that the 16-electron metal complex 1 is
kinetically inert and prepare for subsequent intramolecular CÀH activation by
vacating a coordination site, we must address the problem of understanding the
intimate mechanism of CÀH bond cleavage by the central atom. An Eyring treatment
of the kinetics data at various temperatures (obtained in the presence of an excess of
³
added DMSO) affords the following activation parameters: DH ˆ 81.5Æ 4 kJ molÀ1
³
À1
and DS ˆ À48 Æ 1 0 J K molÀ1. The interpretation of these values is not straightfor-
The reciprocal plot is preferred because dispersion of the aÀ1 values is more regular than that of a.
5
)

518
Helvetica Chimica Acta ± Vol. 88 (2005)
ward as for the kinetics of ligand dissociation, because they cannot be assigned to a
single step, and cannot discriminate between the two most-conceivable reaction
mechanisms, electrophilic attack and oxidative addition. The electrophilic-attack
mechanism is a very common pathway for cyclopalladation and has been proposed
occasionally also for PtII chemistry [18]. It does not require formation of a metal
hydride, the central atom does not change oxidation number, and the H-atom
dissociates as a free or a bound proton (Scheme 4). However, there is a continuously
growing wealth of evidence for the operation of an oxidative-addition mechanism not
only in intramolecular but also in intermolecular processes of CÀH bond activation by
PtII compounds. Hydridoplatinum(IV) intermediates have been proposed, although
not detected directly, in the synthesis of cyclometalated complexes containing
bidentate (C,N) or terdentate (C,N,N') ligands [19]. Similar intramolecular CÀH
bond activation was proposed in the thermolysis of different platinum dialkyl
complexes that lack b-H-atoms, leading to the formation of metallacycles along with
the corresponding hydrocarbons [20]. Intermolecular CÀH activation of alkanes and
arenes is accomplished under mild conditions by complexes of the form
‡
[Pt(NÀN)(Me)(solv)] , where NÀN is a bidentate N-centered ligand and ꢀsolvꢁ is a
weakly coordinating solvent, and involve coordinatively unsaturated PtII species as
intermediates and a stepwise process of oxidative addition/reductive elimination [21].
A useful synthetic route to PtIV complexes [PtX(H)R2(NÀN)] (X ˆ halide, R ˆ alkyl
or aryl) is the addition of acid to [PtR2(NÀN)] and the decomposition of these PtIV
species by reductive elimination of RH to yield [PtXR(NÀN)] [22]. Hy-
drido(methyl)platinum(4‡) complexes with triethylphosphane ligands have also been
identified by low-temperature NMR techniques [21a][23].
Scheme 4
On the basis of the above, we are inclined to think that the most likely mechanism
for the cyclometalation of 1 involves preliminary PtÀS bond dissociation, as depicted
in Scheme 5. The coordinative and electronic unsaturation at the metal atom in the
three-coordinate, 14-electron T-shaped 1 might be stabilized by an agostic interaction
involving the empty s orbital on Pt and the electron pair of the CÀH bond at one of the
ortho sites. This hypothesis is strongly supported by the observation that bulky ligands
favor the formation of agostic interactions between the metal and CÀH bonds [24] and
that cyclometalation is initiated by an agostic interaction with the s(CÀH) orbital
followed by back-donation to the s*(CH) orbital [25]. Oxidative addition of the CÀH
bond follows this preliminary PtÀCH interaction, yielding a cyclometalated-hydrido
16-electron PtIV five-coordinate intermediate. Several theoretical studies have indi-
cated that these reaction intermediates should have square-pyramidal geometry [26],
but, only recently, two papers appeared describing the isolation and structural
characterization of such compounds [27]. Finally, reductive elimination and re-entry of
DMSO should yield the cyclometalated species 2 with liberation of methane.

Helvetica Chimica Acta ± Vol. 88 (2005)
519
Scheme 5
Cyclometalation on 1 develops along the same key steps recognized for the dinuclear
PtII compound of formula [Pt2(m-SEt2)2(Hbph)4] (HbphÀ ˆ biphenyl-4-yl monoanion)
in forming [Pt2(m-SEt2)2(bph)2], containing the planar biphenyl-2,2'-diyl ligand (bph2À)
[9]1).
Considering the equilibrium dissociation constant Ke ˆ k1/kÀ1, and applying the
steady-state approximation to the cyclometalated-hydrido PtIV five-coordinate inter-
mediate, we obtain Eqn. 3:
kobs ˆ k2 k3 Ke/((kÀ2 ‡ k3)[DMSO] ‡ (kÀ2 ‡ k3) Ke)
(3)
that is formally similar to Eqn. 2 with a ˆ k2k3/(kÀ2 ‡ k3) and still b ˆ KÀe 1 and also
explains the retarding effect of the addition of DMSO. The two expressions become
identical when one assumes that k3 ꢀ kÀ2. An appropriate measure of the extent to
which dissociation proceeds deserves further investigations.
Summing up, although the cycloplatination process discussed follows indeed a
dissociative pathway, the dissociation step is not rate determining. It influences the rate
because a pre-equilibrium is established that controls the concentration of the
unsaturated three-coordinate T-shaped intermediate. A multistep mechanism of
oxidative addition/reductive elimination follows, and, at this stage, it is difficult to
assess with confidence which of these two steps is rate limiting.

520
Helvetica Chimica Acta ± Vol. 88 (2005)
Experimental Part
General. All syntheses were performed under a dry, O2-free N2 atmosphere by standard Schlenk-tube
techniques, or in a glove box; however, the products could be worked up and handled in air. All solvents were
analytical reagent grade (Lab-Scan Ltd.) and were used in the synthetic procedures after distillation under a N2
atmosphere from appropriate drying agents: Et2O from sodium benzophenone ketyl; CH2Cl2 from BaO;
DMSO from CaH2, at low pressure after preliminary filtration through alumina. All solvents were stored in N2-
filled flasks over activated 4-Š molecular sieves. CHCl3 for use in kinetics experiments by spectrophotometry
was degassed by means of a series of ꢀfreezing-pumpingꢁ cycles and eventually stored in Schlenk tubes. CDCl3
(99.96 ‡ %, C. I. L. Inc.) was dried by allowing it to stand many days over CaH2, distilled under N2 over
activated MgSO4 and Na2CO3, and then stored over activated 4-Š molecular sieves. K2PtCl4 (Strem) was
purified by dissolving in H2O and filtering. cis-[Pt(Me)2(dmso)2] was prepared according to a published method
and was crystallized several times from Et2O [28]. All phosphane ligands used in these studies were purchased
from Strem or Aldrich, and crystallized from EtOH. All the other reagents were of the highest commercial grade
available and were used as received or were purified by distillation or recrystallization when necessary. UV/VIS:
Rapid-Scanning Hewlett-Packard 8452 A diode-array spectrophotometer or Applied Photophysics Bio
Sequential SX-MX Stopped-Flow ASVD spectrofluorometer. (temp. accuracy Æ 0.058). Rate constants were
evaluated with the SCIENTIST software package. IR Spectra: Perkin-Elmer 883 spectrophotometer; n in cmÀ1
(nujol mulls). 1H- and 31P{1H}-NMR Spectra: Bruker AMX-R-300; d in ppm referenced to TMS for 1H and to
H3PO4 for 31P; J in Hz. The probe temp. was checked by means of the MeOH or ethylene glycol method [29].
2D-1H-NMR Spectra: Bruker AMX-R-300; 2D-COSY experiments were performed with gradient selection in
QF mode; phase-sensitive 2D-1H-NOESY experiments were performed with a standard pulse sequence and a
mixing time of 0.4 s (1a,b). Elemental analyses were performed at the Analytical Services Unit, Department of
Chemistry, University College Dublin, Ireland.
Preparation of Complexes. cis-Dimethyl[(sulfinyl-kS)bis[methane]][tris(2-methylphenyl)phos-
phine]platinum(2‡) (cis-[PtMe2(dmso){P(o-tol)3}]; 1a,b). A cold soln. of tris(2-methylphenyl)phosphane
(P(o-tol)3; 91.3 mg, 0.3 mmol) in CH2Cl2 (10 ml) was added very slowly dropwise and under continuous stirring
to a CH2Cl2 soln. (30 ml) of cis-dimethylbis[(sulfinyl-kS)bis[methane]]platinum(2‡) (cis-[PtMe2(dmso)2];
114.4 mg, 0.3 mmol) immersed in a ice bath. The mixture was kept for 1 h at 08 and then concentrated under
vacuum with the temp. maintained at ca. 58 until the first white crystals of the product began to form. Addition
of 1vol. petroleum ether with cooling at À 358 promoted complete precipitation of the solid. IR: 1108 cmÀ1
(SˆO). Anal. calc. for C25H33OPPtS: C 49.20, H 5.90, S 5.25; found: C 49.28, H 5.73, S 5.13.
1
3
3
4
Data of 1a: H-NMR (CDCl3, 255.6 K): 8.85 (m, J(P,H) ˆ 17.2, J(H,H) ˆ 6.7, J(H,H) ˆ 2.5, C(6)ÀH of
Pha); 7.34 (m, C(3)ÀH of Phc); 7.32 (m, C(5)ÀH of Pha); 7.24 (m, C(6)ÀH of Phc); 7.18 (m, C(6)ÀH of Phb);
7.14 ( m, C(4)ÀH of Pha); 7.10 (m, C(5)ÀH of Phb); 7.09 (m, C(5)ÀH of Phc); 7.07 (m, C(3)ÀH, C(4)ÀH of
3
3
Phb); 7.04 (m, C(3)ÀH, C(4)ÀH of Pha); 2.94 (s, J(Pt,H) ˆ 13.9, SMeb); 2.84 (s, PhcMe); 2.20 (s, J(Pt,H) ˆ
13.1, SMea); 1.83 (s, PhbMe); 1.55 (s, PhaMe); 0.47 (m, 3J(P,H) ˆ 8.8, 2J(Pt,H) ˆ 79.2, PtMe trans to dmso); 0.43
(m, J(P,H) ˆ 6.7, J(Pt,H) ˆ 69.8, 3 H, PtMe trans to PR3). 31P{1H}-NMR (CDCl3, 255.6 K): 28.3 (1J(Pt,P) ˆ
3
2
1824).
Data of 1b: 1H-NMR (CDCl3, 255.6 K): 8.70 (m, 3J(P,H) ˆ 16.0, 3J(H,H) ˆ 7.6, 4J(H,H) ˆ 2.5, C(6)ÀH of
Pha); 7.39 ± 7.03 (m, 11 arom. H); 3.06 (s, 3J(Pt,H) ˆ 15.6, SMeb'); 3.00 (s, Phc'Me); 2.88 (s, 3J(Pt,H) ˆ 16.0,
3
2
SMea'); 2.05 (s, Phb'Me); 1.53 (s, Pha'Me); 0.30 (m, J(P,H) ˆ 6.7, J(Pt,H) ˆ 69.8, PtMe trans to PR3); 0.22 (m,
3J(P,H) ˆ 8.4, 2J(Pt,H) ˆ 78.4, PtMe trans to dmso). 31P{1H}-NMR (CDCl3, 255.6 K): 27.1( 1J(Pt,P) ˆ 1881).
[[2-[Bis(2-methylphenyl)phosphino-kP]phenyl]methyl-kC]methyl[(sulfinyl-kS)bis[methane]]platinum(2‡)]
(2). The procedure described above for the synthesis of the precursor 1 was repeated at r.t. Afterwards, the soln.
was refluxed for 2 h until most of the solvent evaporated, and precipitation of a white solid was incipient,
petroleum ether (1vol.) was added to the soln., and the mixture was placed in the freezer ( À 308) overnight. IR:
1
2
3
1102 (SˆO). H-NMR (CDCl3, 298.3 K): 7.4 ± 6.8 (m, 12 H); 3.18 (m, J(Pt,H) ˆ 87.9, J(P,H) ˆ 3.8, PtCH2);
2.89 (s, 3J(Pt,H) ˆ 14.4, 2 SMe); 2.59 (s, 2 PhMe); 0.45 (m, 2J(Pt,H) ˆ 69.3, 3J(P,H) ˆ 6.6, PtMe). 31P{1H}-NMR
(CDCl3, 298.3 K): 36.3 (1J(Pt,P) ˆ 1920). Anal. calc. for C24H29OPPtS: C 48.72, H 4.94; found: C 48.63, H 4.92.
cis-Dimethyl(pyridine)[tris(2-methylphenyl)phosphine]platinum(2‡) (cis-[PtMe2(py){P(o-tol)3)]; 3). A
1
slight excess of pyridine was added in situ to 0.5 ml of a CDCl3 soln. of 1a in a NMR tube. H-NMR (CDCl3,
298.3 K): 8.20 (m, HÀC(2), HÀC(6) of py); 7.46 (m, HÀC(4) of py); 7.40 ± 6.90 (m, 12 arom. H); 6.86 (m,
3
2
HÀC(3), HÀC(5) of py); 2.04 (m, 3 PhMe); 0.47 (m, J(P,H) ˆ 7.4, J(Pt,H) ˆ 70.6, PtMe trans to PR3); 0.27
3
2
(m, J(P,H) ˆ 8.8, J(Pt,H) ˆ 84.6, PtMe trans to py).

Helvetica Chimica Acta ± Vol. 88 (2005)
521
REFERENCES
[1] A. D. Ryabov, Chem. Rev. 1990, 90, 403; A. J. Canty, in ꢀComprehensive Organometallic Chemistry II: A
Review of the Literature 1982 ± 1994ꢁ, Eds. E. F. Abel, F. G. A. Stone, G. Wilkinson, Pergamon, Oxford,
New York, 1995, Vol. 9, p. 225; G. R. Newkome, W. E. Puckett, V. K. Gupta, G. Kiefer, Chem. Rev. 1986,
86, 451; C. Anderson, M. Crespo, J. Organomet. Chem. 2004, 689, 1496.
[2] V. W. W. Yam, K. M. C. Wong, N. Zhu, J. Am. Chem. Soc. 2002, 124, 6506; S. W. Lai, M. C. W. Chan, T. C.
Cheung, S. M. Peng, C. M. Che, Inorg. Chem. 1999, 38, 913; A. Caubet, C. Lopez, X. Solans, M. Font-
Bardia, J. Organomet. Chem. 2003, 669, 164; G. Y. Zheng, D. P. Rillema, Inorg. Chem. 1998, 37, 1 392; J.
Brooks, Y. Babayan, S. Lamansky, P. I. Djurovich, I. Tsyba, R. Bau, M. E. Thompson, Inorg. Chem. 2002,
41, 3055.
[3] M. Hissler, J. E. Mc Garrah, W. B. Connick, D. K. Geiger, S. D. Cummings, R. Eisenberg, Coord. Chem.
Rev. 2000, 208, 115, and refs. cit. therein.
[4] B. S. Williams, P. Dani, M. Lutz, A. L. Speck, G. van Koten, Helv. Chim. Acta 2001, 84, 3519; A. D. Ryabov,
I. M. Panyashkina, V. A. Polyakov, J. A. K. Howard, L. G. Kuzꢁmina, M. S. Datt, C. Sacht Organometallics
1998, 17, 3615; J. Campora, P. Palma, E. Carmona, Coord. Chem. Rev. 1999, 193 ± 195, 207; I. Omae, Chem.
Rev. 1979, 79, 287; I. Omae, Coord. Chem. Rev. 1988, 83, 137.
[5] a) J. Fornies, A. Martin, R. Navarro, V. Sicilia, P. Villarroya, Organometallics 1996, 15, 1826; b) S. D.
Chappell, D. J. Cole-Hamilton, J. Chem. Soc., Dalton Trans. 1983, 1051; c) A. J. Cheney, B. L. Shaw, J.
Chem. Soc., Dalton Trans. 1972, 754; d) E. C. Alyea, G. Ferguson, J. Malito, B. L. Ruhl, Organometallics
1989, 8, 1188; e) E. C. Alyea, J. Malito, B. L. Ruhl, J. Organomet. Chem. 1988, 340, 119; f) C. Scheffknecht,
A. Rhomberg, E. P. Müller, P. Peringer, J. Organomet. Chem. 1993, 463, 245; g) H. C. Clark, A. B. Goel,
R. G. Goel, S. Goel, Inorg. Chem. 1980, 19, 3220.
Á
[6] R. Romeo, M. R. Plutino, L. Monsu Scolaro, S. Stoccoro, G. Minghetti, Inorg. Chem. 2000, 39, 2712.
Á
[7] M. R. Plutino, L. Monsu Scolaro, R. Romeo, A. Grassi, Inorg. Chem. 2000, 39, 4749.
Á
[8] R. Romeo, L. Monsu Scolaro, M. R. Plutino, F. Fabrizi de Biani, G. Bottari, A. Romeo, Inorg. Chim. Acta
2003, 350, 143.
Á
[9] M. R. Plutino, L. Monsu Scolaro, A. Albinati, R. Romeo, J. Am. Chem. Soc. 2004, 126, 6470.
[10] J. A. S. Howell, M. G. Palin, P. McArdle, D. Cunningham, Z. Goldschmidt, H. E. Gottlieb, D. Hezroni-
Langerman, Inorg. Chem. 1991, 30, 4685; J. A. S. Howell, M. G. Palin, P. McArdle, D. Cunningham, Z.
Goldschmidt, H. E. Gottlieb, D. Hezroni-Langerman, Organometallics 1993, 12, 1694.
[11] K. Mislow, Chemtracts: Org. Chem. 1989, 2, 151; K. Mislow, Acc. Chem. Res. 1976, 9, 26; K. Mislow, Pure
Appl. Chem. 1971, 225, 549; K. Mislow, D. Gust, P. Finocchiaro, R. J. Boettcher, Topics Curr. Chem. 1974,
47, 1 .
[12] SCIENTIST, Micro Math Scientific Software, Salt Lake City, UT.
[13] Applied Photophysics Bio Sequential SX-17 MV, sequential stopped flow ASVD spectrofluorimeter,
Software manual: Applied Photophysics Ltd., 203/205 Kingstone Road, Leatherhead KT22 7PB, U.K.
[14] D. Minniti, G. Alibrandi, M. L. Tobe, R. Romeo, Inorg. Chem. 1987, 26, 3596.
[15] U. Frey, L. Helm, A. E. Merbach, R. Romeo, J. Am. Chem. Soc. 1989, 111, 8161.
[16] R. Romeo, Comments Inorg. Chem. 1990, 11, 21 .
[17] A. C. Albeniz, A. L. Casado, P. Espinet, Inorg. Chem. 1999, 38, 2510.
[18] D. L. Thorn, Organometallics 1998, 17, 348.
[19] C. M. Anderson, M. Crespo, M. C. Jennings, A. J. Lough, G. Fergurson, R. J. Puddephatt, Organometallics
1991, 10, 2672; C. M. Anderson, M. Crespo, M. Font-Bardia, A. Klein, X. Solans, J. Organomet. Chem.
2000, 601, 22; M. E. van der Boom, J. Ott, D. Milstein, Organometallics 1999, 18, 3873; M. Crespo, X.
Solans, M. Font-Bardia, Organometallics 1995, 14, 355.
[20] P. Foley, G. M. Whitesides, J. Am. Chem. Soc. 1979, 101, 2732; P. Foley, R. Di Cosimo, G. M. Whitesides, J.
Am. Chem. Soc. 1980, 102, 6713; S. S. Moore, R. Di Cosimo, A. F. Sowinski, G. M. Whitesides, J. Am. Chem.
Soc 1981, 103, 948; R. Di Cosimo, S. S. Moore, A. F. Sowinski, G. M. Whitesides, J. Am. Chem. Soc. 1982,
104, 124; S. D. Chappell, D. J. Cole-Hamilton, J. Chem. Soc., Chem. Commun. 1980, 38; D. C. Griffiths,
G. B. Young, Polyhedron 1983, 2, 1095; M. Hackett, J. A. Ibers, G. M. Whitesides, J. Am. Chem. Soc 1988,
110, 1436; D. C. Griffiths, G. B. Young, Organometallics 1989, 8, 875.
[21] a) S. S. Stahl, J. A. Labinger, J. E. Bercaw, J. Am. Chem. Soc. 1996, 118, 5961; b) G. Gerdes, P. Chen,
Organometallics 2003, 22, 2217; c) J. A. Labinger, J. E. Bercaw, Nature 2002, 417, 507; d) H. A. Zhong, J. A.
Labinger, J. E. Bercaw, J. Am. Chem. Soc. 2002, 124, 1378; e) L. Johansson, M. Tilset, J. A. Labinger, J. E.
Bercaw, J. Am. Chem. Soc. 2000, 122, 10846; f) J. Procelewska, A. Zahl, R. van Eldik, H. A. Zhong, J. A.

522
Helvetica Chimica Acta ± Vol. 88 (2005)
Labinger, J. E. Bercaw, Inorg. Chem. 2002, 41, 2808; g) L. Johansson, O. B. Ryan, M. Tilset, J. Am. Chem.
Soc. 1999, 121, 1974; h) L. Johansson, O. B. Ryan, C. Römming, M. Tilset J. Am. Chem. Soc. 2001, 123,
6579; i) L. Johansson, M. Tilset, J. Am. Chem. Soc. 2001, 123, 739; l) M. W. Holtcamp, L. M. Henling, M. W.
Day, J. A. Labinger, J. E. Bercaw, Inorg. Chim. Acta 1998, 270, 467; m) R. A. Periana, D. J. Taube, S.
Gamble, H. Taube, T. Satoh, H. Fujii, Science 1998, 280, 560; n) D. D.Wick, K. I. Goldberg, J. Am. Chem.
Soc. 1997, 119, 10235.
[22] V. de Felice, A. de Renzi, A. Panunzi, D. Tesauro, J. Organomet. Chem. 1995, 488, C13; G. S. Hill, L. M.
Rendina, R. J. Puddephatt, Organometallics 1995, 14, 4966; J. G. Hinman, C. R. Baar, M. C. Jennings, R. J.
Puddephatt, Organometallics 2001, 19, 563; C. M. Ong, M. C. Jennings, R. J. Puddephatt, Can. J. Chem.
2003, 81, 1; R. J. Puddephatt, Coord. Chem. Rev. 2001, 157, 219; R. J. Puddephatt, Angew. Chem., Int. Ed.
2002, 41, 261.
[23] S. S. Stahl, J. A. Labinger, J. E. Bercaw, J. Am. Chem. Soc. 1995, 117, 9371; M. W. Holtcamp, J. A. Labinger,
J. E. Bercaw, Inorg. Chim. Acta 1997, 265, 1 1 7.
[24] M. Brookhart, M. L. H. Green, J. Organomet. Chem. 1983, 250, 395; F. A. Cotton, T. LaCour, A. G.
Stanislowski, J. Am. Chem. Soc. 1974, 96, 754.
[25] R. H. Crabtree, E. M. Holt, M. Lavin, S. M. Morehouse, Inorg. Chem. 1985, 24, 1986; M. Lavin, R. H.
Crabtree, Organometallics 1989, 8, 99.
[26] G. S. Hill, R. J. Puddephatt, Organometallics 1998, 17, 1478; P. E. M. Siegbahn, R. H. Crabtree, J. Am.
Chem. Soc. 1996, 118, 4442; K. L. Bartlett, K. I. Goldberg, W. T. Borden, J. Am. Chem. Soc. 2000, 122, 1456;
K. L. Bartlett, K. I. Goldberg, W. T. Borden, Organometallics 2001, 20, 2669; H. Heiberg, L. Johansson, O.
Gropen, O. B. Ryan, O. Swang, M. Tilset, J. Am. Chem. Soc. 2000, 122, 10831; T. M. Gilbert, I. Hristov, T.
Ziegler, Organometallics 2001, 20, 1183.
[27] U. Fekl, W. Kaminsky, K. I. Goldberg, J. Am. Chem. Soc. 2001, 123, 6423; S. Reinartz, P. S. White, M.
Brookhart, J. L. Templeton, J. Am. Chem. Soc. 2001, 123, 6425.
[28] C. Eaborn, K. Kundu, A. J. Pidcock, J. Chem. Soc., Dalton Trans. 1981, 933.
[29] A. L. Van Geet, Anal. Chem. 1968, 40, 2227; A. L. Van Geet, Anal. Chem. 1970, 42, 679.
Received October 29, 2004