Intramolecular C-H activation by dicationic Pt(II) complexes

Article abstract of DOI:10.1016/S1381-1169(02)00193-0

The dicationic complexes [(ArN=C(Me)-C(Me)=NAr)Pt(solv)₂]X₂, (Ar = 2,6-(CH₃)₂C₆H₃; 5a: solv = CH₃CN, X = CF₃SO₃^-, BF₄^-, SbF₆^-; 5b: solv = (CH₃)₂CO, X = BF₄^-, SbF₆^-) and [(CyN=C(H)-C(H)=NCy)Pt(CH₃CN)₂]X₂, (Cy = C₆H₁₁, 6: X = OTf^-, BF₄^-, PF₆^-, SbF₆^-) were synthesized from the corresponding Pt dichlorides with two equivalents of AgX. The reactions of 5a with 1-phenylpyrazole, 2-phenylpyridine, 2-vinylpyridine, and 2-(2-thienyl)pyridine afford the cyclometalated products 11-14 via intramolecular C-H activation of an sp² C-H bond of the unsaturated sidegroup. Pyridines with saturated groups at the 2-position do not undergo a similar cyclometalation reaction. In trifluoroethanol-d₃ solution, 6 undergoes cyclometalation of one of the cyclohexyl groups, an example of sp³ C-H bond activation. The latter reaction proceeds only partway to completion, implying that an equilibrium has been reached; in the case where X = OTf^--the equilibrium favors the starting dication.

Full text of DOI:10.1016/S1381-1169(02)00193-0

Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
Intramolecular C–H activation by dicationic Pt(II) complexes
Antek G. Wong-Foy, Lawrence M. Henling, Mike Day,
Jay A. Labinger∗, John E. Bercaw1
Arnold and Mabel Beckman Laboratories for Chemical Synthesis Division of Chemistry and Chemical Engineering,
California Institute of Technology, Pasadena, CA 91125, USA
Received 20 September 2001; accepted 26 December 2001
Abstract
=
=
The dicationic complexes [(ArN C(Me)–C(Me) NAr)Pt(solv)2]X2, (Ar = 2,6-(CH3)2C6H3; 5a: solv = CH3CN, X =
CF3SO3−, BF4−, SbF6−; 5b: solv = (CH3)2CO, X = BF4−, SbF6 ) and [(CyN C(H)–C(H) NCy)Pt(CH3CN)2]X2, (Cy =
C6H11, 6: X = OTf−, BF4−, PF6−, SbF6−) were synthesized from the corresponding Pt dichlorides with two equivalents
of AgX. The reactions of 5a with 1-phenylpyrazole, 2-phenylpyridine, 2-vinylpyridine, and 2-(2-thienyl)pyridine afford the
cyclometalated products 11–14 via intramolecular C–H activation of an sp2 C–H bond of the unsaturated sidegroup. Pyridines
with saturated groups at the 2-position do not undergo a similar cyclometalation reaction. In trifluoroethanol-d3 solution, 6 un-
dergoes cyclometalation of one of the cyclohexyl groups, an example of sp3 C–H bond activation. The latter reaction proceeds
only partway to completion, implying that an equilibrium has been reached; in the case where X = OTf−the equilibrium
favors the starting dication.
−
=
=
© 2002 Elsevier Science B.V. All rights reserved.
Keywords: Dicationic complexes; Activation; Cyclometalation; Platinum
1. Introduction
It is generally accepted that the C–H activation
or actual cleavage of the C–H bond governs the
overall rate and selectivity for this alkane func-
tionalization reaction (Eq. (2)). Unfortunately, the
instability of the resulting Pt(II) alkyl complex to-
wards protonolyis in protic media and its tendency to
disproportionate in aprotic media has precluded the
direct mechanistic study of this key reaction in the
actual Shilov system [5–7]. We have reported mecha-
nistic studies on the microscopic reverse of the C–H
activation reaction—the protonolysis of the Pt alkyl
bond—for a series of ligand-stabilized alkyl Pt(II)
complexes [8,9]. Insights gained from these studies
ultimately led to the development of cationic diamine
[10,11], tris(pyrazolyl)borate [12,13], and ␣-diimine
[14] supported Pt(II) complexes capable of activat-
ing alkanes under mild conditions. Key features of
The “Shilov system” represents the first example of
homogeneous alkane functionalization [1] in which
simple (or unactivated) alkanes, including methane,
are oxidized to mixtures of alcohols and alkyl chlo-
rides in aqueous solution. The process is catalytic
in [PtCl4]2− and requires stoichiometric [PtCl6]2−
(Eq. (1)) [2–4].
R–H + PtCl62− + H2O(HCl)
2−
P−tC→l4 R–OH(R–Cl) + PtCl42− + 2HCl
(1)
H2O,120 ◦C
∗
Corresponding author.
E-mail address: jal@its.caltech.edu (J.A. Labinger).
1
Co-corresponding author.
1381-1169/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S1381-1169(02)00193-0

4
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
However, nearly all previous model studies involve
a metathesis-like stoichiometry, in which one hydro-
carbyl group replaces a methyl group which departs
with the hydrogen atom as methane (Eq. (3)), We
have not yet demonstrated the exact analog of Eq. (2),
Pt-alkyl formation with proton loss, which is required
in a functioning catalyst based on Shilov chemistry.
In trying to design such a model, we noted that the
methyl groups of Pt(II) cations [L2Pt(CH3)(solv)]+
are relatively stable towards protonolysis, compared
Fig. 1. Representative aryl and alkyl ␣-diimine ligands.
these systems include an electrophilic metal center,
a non-coordinating counter-anion, a substitutionally
labile fourth ligand, and the use of a non-oxidizable
and extremely weakly coordinating solvent.
(2)
Cationic group 10 transition metal complexes of
the general type [(N–N)M(CH3)(solv)]+[WCA]−
(M = Ni, Pd; N–N = α-diimine; solv = Et2O, CH3
CN; WCA = weakly coordinating anion) have been
applied to a variety of important catalytic transfor-
mations [15].2 Pt(II) complexes [(N–N)Pt(CH3)L]+
[BF4]− (where N–N = bis (aryl)diimine; L = CH3
CN, H2O, CF3CH2OH) are thermally more ro-
bust than the corresponding Pt(II) diamine (N,N,Nꢀ,
Nꢀ-tetramethylethylenediamine) complexes, proba-
bly because of the ␣-diimines’ ability to bind to
to the neutral dimethyl or chloromethyl Pt(II) com-
plexes which are readily protonated to generate
methane [21]. For example, the methyl acetonitrile
cations [(N–N)Pt(CH3)(NCCH3)]+[BF4]− (N–N =
1, 2) can be prepared quantitatively by treating
the neutral dimethyl Pt(II) complexes even using a
slight excess of HBF4 in CH3CN [14]. More re-
markably, the methyl group in the cationic Pt(II)
complex [((C2F5)2PCH2CH2P(C2F5))Pt(Me)CO]+
resists protonolysis even in the superacid mixture
SbF5/FSO3H [22].
(3)
(4)
It is unclear whether this reduced reactivity is pri-
marily a kinetic or equilibrium effect. While the equi-
librium shown in Eq. (4) probably lies to the right in
most or all cases, we reasoned that the increased elec-
trophilicity of a dicationic Pt(II) center, as compared
to a monocationic one, would favor deprotonation,
whether from a coordinated C–H bond or a subsequent
Pt(IV) alkylhydride intermediate. This would lead to
facile exchange of H+ with D+ from the solvent and
overall incorporation of deuterium into the alkane sub-
strate, which might be detected even if the equilibrium
of Eq. (4) does lie well to the right. However, studies
on intermolecular C–H activation of arenes such as
the metal both as a moderately strong σ-donor and
weak ␲-acceptor [16,17]. Perhaps more importantly,
the versatility of the ␣-diimine synthesis permits both
the steric and electronic parameters of the ligand to be
varied relatively easily [18], making these complexes
well-suited for mechanistic studies aimed at elucidat-
ing the steric and electronic requirements, as well as
the inherent selectivity of the Pt center in the C–H ac-
tivation of alkanes under mild conditions [19,20]. Two
representative ␣-diimine ligands are shown in Fig. 1.
2
For a review of ␣-diimine ligated late transition metal olefin
polymerization complexes, see [15].

A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
5
Scheme 1.
benzene and xylene in trifluoroethanol-d3 were incon-
clusive, since we were unable to rule out competing
H/D exchange catalyzed by traces of acid.
on the imine nitrogens in the axial regions above and
below the trigonal plane [32,33]. For example with
=
=
the bulkier ligand ArN C(Me)–C(Me) NAr, Ar =
2,6-(CH(CH3)2)2C6H3, the 5-coordinate olefin com-
plex A was too unstable at room temperature to be
Accordingly, we turned our attention to intramolec-
ular C–H activation to generate cyclometalated
complexes, where the favorable entropy change of
chelation would be expected to provide an additional
driving force. There is ample precedent for this be-
havior [23,24], including related platinum complexes
[25–29]. We report here C–H bond activation and
deprotonation processes for Pt(II) complexes with the
diimine auxiliary ligands 1 and 2.
=
=
isolated, whereas with ligand 2 (RN C(H)–C(H) NR,
R = cyclohexyl), A was isolated exclusively. With
=
=
ligand 1 (ArN C(Me)–C(Me) NAr, Ar = 2,6-(CH3)2
C6H3), a mixture of A and 3 was obtained. Quan-
titative conversion to the dichlorides, 3 and 4, can
be achieved by refluxing the respective 5-coordinate
olefin complexes in THF or methylene chloride.
The ␣-diimine ligands are generally unable to dis-
place COD (1,5-cyclooctadiene) from (COD)PtCl2,
NBD (norbornadiene) from (NBD)PtCl2, or SMe2
from (SMe2)2PtCl2 [31,33]. Surprisingly, treat-
ment of the substitutionally labile Pt(II) complex
[Pt(CH3CN)4][OTf]2 [34] in nitromethane with one
equivalent of an ␣-diimine did not yield the expected
bis-acetonitrile complex.
2. Results and discussion
2.1. Preparation of dicationic solvento complexes 3
and 4
The complexes examined in this study were pre-
pared from the corresponding (N–N)PtCl2 complexes
3 (N–N = 1) and 4 (N–N = 2). The dichlorides are
prepared by adding the appropriate ␣-diimine to a so-
lution of Zeise’s anion in methanol at 0 ◦C [30] or to
a THF solution of the neutral dimer [31] (Scheme 1).
The bis-solvento complexes 5a, 5b, and 6 were syn-
thesized by abstracting the chlorides with two equiv-
alents of a silver salt in a mixture of CH2Cl2/CH3CN
or (CH3)2CO (Eq. (5)).
(5)
Upon addition of the ligand, the 5-coordinate, 18
e− intermediate A precipitated from methanol so-
lution. This intermediate is generally more soluble
in non-polar halogenated solvents and ethers. The
propensity of intermediate A to liberate ethylene to
afford the corresponding dichloride complexes de-
pends on the steric bulk imparted by the substituents
The acetonitrile and acetone adducts were iso-
lated as yellow solids by addition of ether to the
reaction mixtures after filtration to remove AgCl.
In general, it was necessary to repeat the above
procedure several times to remove all the AgCl. At-
tempts at synthesizing the bis-pentafluoropyridine or
bis-tetrafluoro-4-picoline analogues in either mixtures

6
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
of CH2Cl2 and the fluorinated pyridines or in neat
fluorinated pyridines proved unfruitful due to the
insolubility of the (N–N)PtCl2/AgX mixtures.
tonitrile in 5a implies that acetonitrile is a better
ligand for Pt, as one might expect. On the other
hand, the equilibrium between 5b and the bis-aqua
complex, and previous measurement of equilibrium
constants between [(N–N)Pt(Me)(D2O)][BF4] and
[(N–N)Pt(Me)(TFE)][BF4] (TFE = trifluoroethanol)
[19]. indicate that acetone has a higher affinity for
the electrophilic Pt(II) center than water, which in
turn is a better ligand than trifluoroethanol. Thus,
a qualitative ordering of acetonitrile > acetone >
water > trifluoroethanol can be obtained for the affin-
ity of various solvents for the Pt center with ligand
1.
The 1H NMR spectra of the bis-acetonitrile and
bis-acetone complexes in weakly coordinating sol-
vents such as CD2Cl2 or CF3CD2OD (TFE-d3) are
consistent with a single C2v-symmetric species in
solution in each case. In CD2Cl2, there is a single
resonance for bound acetonitrile in 5a (X = OTf−,
BF4−, SbF6−) at 2.13 ppm; the 19F NMR spec-
trum of the OTf− salt consists of a sharp singlet at
−75 ppm, characteristic of non-coordinated OTf−
[35].3 When 5a was dissolved in more coordinat-
ing solvents such as MeOH-d4 or D2O, however,
a mixture of products was observed, the major one
being [(N–N)Pt(CH3CN)(solv)]2+X2, in which one
acetonitrile molecule had been replaced by a solvent
molecule. In contrast, a D2O solution of 5b exhibits
signals corresponding to two symmetric species,
in an approximate 5:1 ratio, as well as free ace-
tone. The major species is assigned as the bis-aqua
complex [(N–N)Pt(D2O)2]2+ which was verified
by independent synthesis. The minor species is the
C2v-symmetric bis-acetone complex 5b; its signal
grows in intensity upon addition of acetone. The
equilibrium between 5b and the bis-aqua complex
suggests that acetone has a higher affinity for the
electrophilic platinum center than water. Interest-
ingly, there was no indication of the unsymmetrical
[(N–N)Pt((CH3)2CO)(D2O)]2+X2 complex being
formed under these conditions.
With ␣-diimine 2, the only solvento complex
that could be isolated relatively cleanly was the
bis-acetonitrile adduct 6. Attempts to isolate the cor-
responding bis-acetone adducts resulted only in the
formation of sticky pale yellow oils consisting of var-
ious pr−oducts. Varying the counteranion from OTf−
−
to BF4 to SbF6 did not to help to produce any
isolable products.
2.2. Substitution reactions of 5a with
1-phenylpyrazole and 2-substituted pyridines
Addition of 1-phenylpyrazole or a 2-substituted
pyridine to a clear yellow solution of 5a results in
an immediate color change to dark yellow-brown
and (by 1H NMR) liberation of free CH3CN. At
least 95% of the resulting Pt species can be as-
signed as [(N–N)Pt(CH3CN)(L)]+ (N–N = 1, L =
1-phenylpyrazole or 2-R-pyridine), 7–10 (Eq. (6)).
The acetone ligands in 5b are more labile than
the corresponding acetonitrile ligands in 5a. In fact,
clean isolation of the complex 5b w−as possible only
in cases where the anions were BF4 or SbF6−. At-
tempted isolation of 5b with OTf− anion invariably
yielded oils that were difficult to purify. 1H NMR
spectra of the crude reaction mixtures in CD2Cl2
showed multiple products. Presumably, OTf− anion
can displace one or even both acetone molecules,
affording a mixture of dicationic, monocationic, and
neutral Pt species. The fact that OTf− anion can
easily replace bound acetone in 5b but not the ace-
(6)
Further addition of substrate (up to two equiva-
lents) does not displace the remaining bound acetoni-
trile. The increased crowding at the metal center due
to the 2-substituted heterocycles may be responsible
for the observed substitution pattern in these cases.
For example, treatment of 5a with two equivalents of
pyridine or 4-methylpicoline results in formation of
3
Johansson et al. have observed the neutral methyl triflate
=
complex (N–N)Pt(Me)(OTf )(N–N = 3,5-(CF3)2C6H3N C(Me)–
C(Me)3,5-(CF3)2C6H3N) in CD2Cl2 by 19F NMR, as indicated
by the 4J(Pt–F) coupling constant of 14.4 Hz.

A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
7
the bis-pyridine or bis-picoline adducts, while treat-
2.3. Intramolecular C–H activation of
ment of 5a with two equivalents of 2-ethylpyridine
results in only 1 pyridyl group adding to the metal
center.
1-phenylpyrazole and 2-substituted pyridines by 5a
Heating reaction mixtures containing a 1:1 or a 2:1
mixture of 1-phenylpyrazole or a 2-substituted pyri-
dine and 5a (or 5b) at 85 ◦C in acetone for 4–6 h re-
sults in deep burgundy solutions of the cyclometalated
products 11–14 (Scheme 2). In each case the 1H NMR
spectrum exhibits a characteristic doublet in the region
δ 5 to 6 with Pt satellites (3JPt–H = 26–33 Hz) that
grows in intensity over time, assigned to the proton ␣
to the Pt-bonded carbon. Cyclometalated phenyl- and
thienylpyridine complexes of platinum with other lig-
ands have been reported previously [36].
The unusually high upfield shift of this proton is
attributed to the ring current shielding of the 2,6-xylyl
groups of the diimine ligand. The ortho-proton of
the pyridyl group is likewise shifted to higher field
(∼6.5–7.0 ppm) in complexes 12–14. The crystal
structure of 12 (X = BF4−), shown in Fig. 2, clearly
indicates that the protons on C22 and C31 lie in
the shielding regions of the arenes on the diimine
Displacement of acetonitrile by 2-(2-thienyl)pyridine
in 5a could potentially lead to ␬1(N)-, ␬1(S)-, or
␬2(N,S)-bound 2-(2-thienyl)pyridine. The methyl
region of the 1H NMR spectrum exhibits seven
singlets; six for methyl groups of the diimine lig-
and (four on the aryl groups, two on the back-
bone) and one for the remaining bound acetonitrile.
The latter establishes ␬1-binding—otherwise both
acetonitriles would have been displaced—and is
also consistent with all the diimine methyl groups
being nonequivalent, assuming restricted rotation
around the Pt-(2-(2-thienyl)pyridine) bond on the
NMR time-scale. The substantial downfield shift
of the proton a to the pyridine nitrogen, at δ 9.25
(compared to δ 8.52 for free 2-thienylpyridine) im-
plies an N-bonded structure (compare 8, where
the proton ␣ to the pyridine nitrogen resonates at
δ 9.23).
Scheme 2.

8
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
acid (HX) that is generated [23,38]. Here, a 2:1 molar
ratio of substrate to Pt complex yields substantially
cleaner reactions than without the extra equivalent of
base. In principle, one should be able to replace the
second equivalent of substrate with another weak base.
This is indeed the case for the extremely hindered base,
2,6-di-t-butylpyridine; triethylamine on the other hand
gives significantly lower yields. Inorganic bases such
as Na2CO3 lead to dramatic color changes affording
deep purple or green solutions, exhibiting complex 1H
NMR spectra.
As noted above, the reaction between 5a and
2
2-(2-thienyl)pyridine could lead to a ␬ (N,S) chelate
when the second acetonitrile is displaced; but only
the cyclometalated product 14 is observed, as un-
equivocally demonstrated by a crystal structure de-
termination. The structure of 14 is shown in Fig. 3,
with selected bond lengths and bond angles reported
in Table 2.
Fig. 2. Labeled diamond drawing of the cationic portion of 12
(X = BF4−) with ellipsoids at 50% probability.
Surprisingly, reactions of 5a or 5b with 2-substituted
pyridines containing sp3 C–H bonds (2-ethylpyridine,
2-n-propylpyridine, 2-i-propylpyridine, 2-t-butylpyri-
dine, and 8-methylquinoline) did not yield the
expected cyclometalated products. The 1H NMR
spectra of the reaction mixtures indicated the forma-
tion of multiple products, of which only the proto-
nated pyridines could be identified unambiguously.
ligand. Selected metrical parameters for 12 are listed
in Table 1.
The X-ray structure reveals that C–H bond activa-
tion yields a five-membered ring, the common pref-
erence. The ␣-diimine is bound unsymmetrically to
the Pt, with a moderate lengthening of the Pt–N bond
trans to the phenyl group relative to that trans to the
pyridyl group (2.092 Å versus 2.056 Å), presumably
due to higher trans influence [37].
In general, cyclometalation reactions are carried out
in the presence of a weak base to trap the equivalent of
Table 1
Selected bond lengths and bond angles for the cationic portion of
12 (X = BF−4 )
Bond lengths (Å)
Pt1–N1
Pt1–N2
Pt1–N3
Pt1–C21
2.056 (5)
2.092 (6)
2.016 (6)
2.009 (8)
Bond angles (◦)
N1–Pt1–N2
N1–Pt1–N3
N1–Pt1–C21
Pt1–N3–C31
Pt1–C21–C22
N2–Pt1–N3
N2–Pt1–C21
N3–Pt1–C21
76.6 (2)
176.5 (2)
101.7 (3)
127.2 (5)
112.0 (5)
101.0 (3)
174.8 (3)
80.9 (3)
Fig. 3. Labeled diamond drawing of the cationic portion of 14
(X = BF4−) with ellipsoids at 50% probability.

A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
9
Table 2
Selected bond lengths and bond angles for the cationic portion of
14 (X = BF−4 ).
Bond lengths (Å)
Pt1–N1
Pt1–N2
Pt1–N3
Pt1–C1
2.013 (7)
2.095 (5)
2.009 (6)
2.006 (7)
Bond angles (◦)
N1–Pt1–N2
N1–Pt1–N3
N1–Pt1–C1
Pt1–N1–C9
Pt1–C1–C2
N2–Pt1–N3
N2–Pt1–C1
N3–Pt1–C21
103.3 (3)
177.9 (3)
78.1 (3)
128.9 (6)
132.5 (7)
76.1 (2)
176.2 (3)
102.6 (3)
Similarly, the anilines 2-t-butylaniline and N,N-dime-
thyl-o-toluidine yielded only deep green/black uniden-
tifiable mixtures. Changing solvents from acetone to
2-methoxyethanol or trifluoroethanol did not affect
the outcome of these reactions.
In order to compare the reactivity of the dicationic
platinum complexes to that of the well-studied mono-
cationic systems, complex 15 ([(N–N)Pt(Me)(2-C6H5
NC5H4)][BF4]; N–N = 1), was synthesized by treat-
ment of the methyl aqua cation, [(N–N)Pt(Me)(H2O)]+
(N–N = 1) [20] with 2-phenylpyridine in acetone
(Eq. (7)). Thermolysis of this complex in acetone-d6
does not result in cyclometalation even after pro-
longed (2 days) heating above 85 ◦C.
Scheme 3.
16 has been previously synthesized by intramolecu-
lar C–H activation of the cationic methyl-acetonitrile
complex 17b or the methyl-aqua adduct 17a (bottom
of Scheme 3) and crystallographically characterized
−
as its BF4 salt [39].
Whereas 17a undergoes cyclometalation at ambient
temperature over the course of 24 h, the acetonitrile
complex 17b and the bis-acetonitrile complex 6 must
be refluxed in trifluoroethanol for the reaction to
(7)
occur. Furthermore, whereas 17a is converted quan-
titatively to 16 after refluxing in trifluoroethanol for
48 h, complete conversion of the dication 6 (X =
OTf−) to 16 is not observed. Instead, a 2.7:1 mixture
of 6:16 was obtained after 12 h. Heating for another
24 h does not change this ratio.
The reaction is highly solvent-dependent. Switching
from TFE-d3 to acetone-d6 resulted in a mixture of
unidentifiable products. Different anions also affect
2.4. Intramolecular cyclometalation of 6
To probe the viability of alkyl C–H bond activation
by these dicationic complexes, we examined the reac-
tion pattern of 6, which contains a cyclohexyl moiety
in its ␣-diimine ligand framework. Heating a solution
of 6 in trifluoroethanol-d3 at 85 ◦C does result in cy-
clometalation, albeit slowly, of one of the cyclohexyl
groups on the imine nitrogens to yield 16 (Scheme 3).

10
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
the outcome of the reaction. Both SbF6 and OTf−
anions afforded considerably cleaner reaction mixtures
equivalent of base, although some conversion is ob-
served even without extra base present.
−
than BF4−. With PF6 no cyclometalation product
Based on the earlier studies of the corresponding
mono-cationic systems [8,19,20], we would pro-
pose Scheme 4 to account for these observations.
Replacement of the second acetonitrile ligand by a
C–H bond (perhaps preceded by formation of an
␩2-␲-arene species [20]) may take place either by
direct displacement or by a solvent-assisted pathway
to give intermediate B. In the cyclometalation of
2-phenylpyridine, the ␩1-arenenium, Bꢀ, species is
another possible structure for B; such intermediates
have been implicated in a number of studies involving
Pt and Pd arene cyclometalation [27,40].
−
was observed, which may be attributed to the extensive
decomposition of the PF6 anion as indicated by 19
F
−
NMR spectroscopy.
2.5. Comments on intramolecular C–H
activation by dicationic Pt(II)
The bis-acetonitrile complex 5a reacts readily with
2-aryl- or 2-vinylpyridines to give first substitution of
one acetonitrile by pyridine and then, under relatively
mild heating (85 ◦C), the cyclometalated products
11–14. These C–H activation reactions are slower
than those of monocationic [L2Pt(CH3)S]+ com-
plexes with benzene (Eq. (3)) where S = TFE or H2O
[14,19,20]. This trend almost surely reflects the fact
that displacement of the second acetonitrile by the re-
=
acting arene C–H (or C C) bond is more difficult than
the corresponding displacement in the monocationic
systems, because ligands such as TFE and water are
much more weakly bonded to Pt(II) than acetonitrile,
and perhaps also because the solvent-Pt(II) bond is
probably stronger in a dication. Displacement of a
ligand is required, as demonstrated by the fact that
conversion of complex 15 to 12 with liberation of
methane (Eq. (7)), a reaction that should be strongly
thermodynamically favored, is not observed under
comparable conditions. Formation of 11–14 from 5
is driven to completion by the presence of a second
From B, the actual cleavage of the C–H bond
can proceed via an oxidative addition mechanism
(Scheme 4, Path I) to generate a Pt(IV) hydride which
is then deprotonated by a base in solution, or alterna-
tively via direct deprotonation of the agostic complex
(Scheme 4, Path II). Most of the evidence supports
the formation of Pt(IV) hydrides as important inter-
mediates in C–H activation, and we tend to favor the
oxidative addition route, although it is conceivable
Scheme 4.

A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
11
that the involvement of additional external base in
3. Experimental
the present case could result in a different preferred
route.
3.1. General considerations
No cyclometalated products were observed when
the bis-solvento complexes 5a or 5b were treated
with pyridines containing adjacent aliphatic or ben-
zylic C–H bonds, although intermolecular alkane
activations by [L2Pt(CH3)(H2O)]+ have been ob-
served [14,19]. This presumably reflects the fact that
displacement of acetonitrile by an sp3 C–H bond is
less favorable than by vinyl or aryl C–H bonds (vide
supra). However, an example of clean intramolecular
C–H activation of an sp3 C–H bond is observed in
the formation of 16 from 6. This reaction is slower
than the formation of 16 from the monocationic
platinum methyl-aquo complex 17a, but proceeds
at about the same rate as monocationic platinum
methyl-acetonitrile complex 17b, again implying that
(solvent) ligand displacement is a central factor in de-
termining reactivity. Note that whereas acetone is the
solvent of choice for the cyclometalations involving
5a, no cyclometalation of 6 was observed in acetone;
clean reaction to yield the cyclometalated product
was observed only in trifluoroethanol-d3.
All air and moisture sensitive compounds were ma-
nipulated using standard high-vacuum line, Schlenk
line or cannula techniques, or in a nitrogen at-
mosphere glove box as previously described [41].
Solvents were dried over sodium/benzophenone
ketyl (toluene, THF, Et2O, petroleum ether), CaH2
(CH2Cl2), Drierite (acetone), and 3 Å molecu-
lar sieves/NaHCO3 (trifluroethanol-d3). Zeise’s
salt (K[Cl3Pt(C2H4)]·H2O) and Zeise’s dimer
[(PtCl2(C2H4)]2 were purchased from Strem Chem-
icals. All silver salts were purchased from Aldrich
and stored in
a glove box. 1-Phenylpyrazole,
2-phenylpyridine, 2-vinylpyridine, and 2-ethylpyridine
were purchased from Aldrich. 2-(2-Thienyl)pyridine
was purchased from Lancaster. ␣-Diimines 1 and
2 were prepared as previously described [42,43].
NMR spectra were recorded on a General Electric
QE300 (1H, 300.1 MHz), a Varian UNITYINOVA 500
(1H, 499.853 MHz; 13C, 125.701 MHz) or a Varian
Mercury-VX 300 (1H, 300.1 MHz; 19F, 282.081 MHz;
13C, 75.46 MHz) spectrometer. All 1H NMR shifts
are relative to the residual NMR solvent and assign-
ments are numbered as shown below. X-ray structure
determinations were performed on a Bruker SMART
1000 CCD area detector under a cold stream of N2
gas. Data reduction was done with Bruker SAINT
v6.2. All structures were solved by direct methods
using SHELXS-97 and refined with SHELXL-97.
Elemental Analyses were performed at the Caltech
Elemental Analysis Facility or by Midwest Microlab,
Indianapolis, IN.
Also, whereas formation of 16 from 17 proceeds to
completion, cyclometallation of 6 only proceeds part-
way, to an equilibrium mixture in which the dicationic
complex is favored. In principle, addition of a base
stronger than solvent to take up the liberated proton
should drive the reaction further; addition of the hin-
dered base 2,6-di-t-butylpyridine does seem to shift the
equilibrium towards the cyclometallated product 16,
but there is competing decomposition. Stronger bases
such as Na2CO3 resulted in highly colored solutions
1
of which the H NMR spectra showed only decom-
position products. Nonetheless, the observation of the
cyclometalated product 16 implies that electrophilic
C–H activation at a dicationic Pt(II) center with pro-
ton liberation is possible. The fact that the equilibrium
lies in favor of the starting material and not the highly
entropically favored cyclometalated species suggests
that direct observation of a Pt alkyl via intermolecu-
lar C–H activation may be difficult, requiring indirect
methods such as H/D exchange to access whether C–H
activation has occurred, but a catalytic cycle involving
such a step, driving the unfavorable equilibrium for-
ward by coupling to a subsequent rapid reaction, is a
potentially viable strategy.
3.2. (ArN= C(Me)–C(Me)= NAr)PtCl2,
(Ar = 2,6-(CH3)2C6H3), 3
K[Cl3Pt(␩2-C2H4)]·H2O (1.0 g, 2.71 mmol) was
dissolved in methanol and cooled to 0 ◦C in an ice
bath. To this was added ligand 1 (0.873 g, 2.98 mmol)
in small portions under vigorous stirring. A dark
brown-red precipitate formed immediately upon addi-
tion. As this brown-red precipitate is the 5-coordinate
ethylene complex and is unstable with respect to
loss of ethylene at room temperature, filtering this
intermediate is unnecessary. Instead the reaction was

12
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
allowed to warm to room temperature and stirred un-
til a color change to light brown was observed (3 h).
The solid was then filtered, washed with methanol
(10 ml) and Et2O (10 ml). Drying over a water aspira-
tor yielded a light brown powder. Yield: 1.5 g (100%).
The dichloride is only sparingly soluble in most
organic solvents giving rise to poorly resolved 1H
added to this residue and filtered through a pad of
Celite to yield a yellow-orange solution. The solvent
was removed in vacuo. Redissolving in acetone and
acetonitrile followed by filtering through Celite was
repeated until all the AgCl was removed. Precipita-
tion with Et2O of a 10:1 acetone:acetonitrile solu-
tion afforded the product as a bright yellow powder.
1
1
NMR spectra. H NMR (DMSO-d6, δ): 7.17 (s, 6H,
2,6-(CH3)2C6H3), 2.21 (s, 12H, 2,6-(CH3)2C6H3),
Yield: 1.01 g (63%). H NMR (CD2Cl2, δ): 7.35 (t,
3
2H, JHH = 6.00 Hz, p-H 2,6-(CH3)2C6H3), 7.31 (d,
3
=
1.74 (s, 6H, N CCH3). Analytically calculated for
4H, JHH = 6.00 Hz, m-H 2,6-(CH3)2C6H3), 2.45
=
C20H24Cl2N2Pt (found): C, 43.02 (42.97); H, 4.33
(4.40); N, 5.02 (4.97).
(s, 6H, N CCH3), 2.44 (s, 12H, 2,6-(CH3)2C6H3).
13C {1H} NMR (CD2Cl2, δ): 188.4, 141.4, 130.9,
−
130.7, 129.6, 120.7, 20.6, 18.2, 3.5, (CF3SO3 not
3.3. (CyN=C(H)–C(H)=NCy)PtCl2,
(Cy = C6H11), 4
found). 19F {1H} NMR (CD2Cl2, δ) −79.1. Ana-
lytically calculated for C26H30F6N4O6PtS2 (found):
C, 35.99 (36.58); H, 3.48 (3.63); N, 6.46 (6.35−).
Analogous procedures were used for X = BF4
and SbF6−. For X = BF4−, 3 (0.300 g, 0.357 mmol)
was combined with AgBF4 (0.214 g, 1.10 mmol).
Workup afforded a yellow powder. Yield: 0.334 g
(83.7%). 1H NMR in CD2Cl2 is the same as for
theX = OTf−. 19F {1H} NMR (CD2Cl2, δ) −152.3.
Analytically calculated for C24H30B2F8N4Pt (found):
C, 38.79 (39.01); H, 4.07 (3.9); N, 7.54 (7.08). For
X = SbF6−, 3 (0.100 g, 0.179 mmol) was com-
bined with AgSbF6 (0.123 g, 0.367 mmol). Workup
afforded a bright yellow powder. Yield: 143 mg
(66%). 1H NMR ((CD3)2CO, δ): 7.43 (br m, 6H,
K[Cl3Pt(␩2-C2H4)]·H2O (0.405 g, 1.10 mmol) was
dissolved in methanol and cooled to 0 ◦C in an ice
bath. To this was added ligand 2 (0.266 g, 1.21 mmol)
in small portions under vigorous stirring. A bright yel-
low precipitate formed upon addition. The reaction
was stirred for 0.5 h, then the resulting precipitate was
filtered and washed with cold methanol (5–10 ml) and
dried. This solid was then dissolved in THF (30 ml)
and refluxed for 1 h after which time a orange-brown
solid formed. This was cooled in an ice bath, then fil-
tered to yield a flocculent orange-brown solid. Yield:
324 mg (60%). The dichloride is only sparingly solu-
1
ble in most organic solvents. H NMR (CD2Cl2, δ):
=
2,6-(CH3)2C6H3), 2.60 (s, 6H, N CCH3), 2.50 (s,
3
=
8.40 (s, 2H, JPtH = 100 Hz, N CH), 4.69 (m, 2H,
12H, 2,6-(CH3)2C6H3), 2.34 (s, 6H, CH3CN). An-
alytically calculated for C24H30F12N4PtSb2 (found):
C, 27.69 (27.66); H, 2.90 (3.01); N, 5.38 (5.25).
NCH(CH2)5), 2.36 (m, 4H, Cy CH2), 1.16–1.91 (m,
16H, overlapping Cy CH2). Analytically calculated
for C20H24Cl2N2Pt (found): C, 34.57 (33.93); H, 4.97
(4.73); N, 5.76 (5.39).
3.5. [(ArN=C(Me)–C(Me)=NAr)Pt(Me2CO)2]X2,
(Ar = 2,6-(CH3)2C6H3; X = BF4−, SbF6−), 5b
3.4. [(ArN=C(Me)–C(Me)=NAr)Pt(CH3CN)2]X2,
(Ar = 2,6-(CH3)2C6H3; X = CF3SO3−, BF4
SbF6−), 5a
,
−
A solution of 3 (100 mg, 0.179 mmol) and AgSbF6
(123 mg, 0.367 mmol) in dry acetone (7 ml) was
stirred at room temperature while protected from light
for 3 h. The resulting mixture was filtered through a
pad of Celite and the solvent removed in vacuo to
afford a red-orange oil. This was redissolved in dry
acetone (5 ml) and filtered through a pad of Celite
followed by removal of solvent in vacuo. The above
procedure was repeated at least one more time until
all the AgCl was removed. The crude product was
purified by repeated precipitation of acetone (2 ml)
solutions with Et2O under vigorous stirring. Drying
To a stirred suspension of 3 (1.04 g, 1.86 mmol)
in CH2Cl2 (75 ml) containing CH3CN (2 ml) was
added AgOTf (0.978 g, 3.81 mmol). As the reaction
proceeded, the color of the suspended solids slowly
turned light yellow. The mixture was stirred while
protected from light at room temperature for 12 h,
after which time a dull yellow precipitate formed.
The entire mixture was then evaporated to dryness.
Acetone (10 ml) and acetonitrile (1 ml) were then

A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
13
in vacuo afforded a bright yellow solid. Product is
3.7. [(ArN=C(Me)–C(Me)=NAr)Pt(κ2-(C,N)-1-
C6H4N2C3H3)]X, (Ar = 2,6-(CH3)2C6H3;
X = OTf −), 11
very hygroscopic. Yield 68 mg (35.2%). 1H NMR
3
((CD3)2CO, δ): 7.46 (t, 2H, JHH = 6.28 Hz, p-H
3
2,6-(CH3)2C6H3), 7.40 (d, 3H, JHH = 6.28 Hz,
A solution of 5a (X = OTf−) (153 mg, 177 mmol)
and 1-phenylpyrazole (46.7 ␮l, 353 mmol) in acetone
(70 ml) was refluxed for 6 h, at which point the solu-
tion was deep red. The solvent was removed in vacuo
to yield a red-orange residue. This was dissolved
in CH2Cl2 (40 ml) and washed with aqueous HOTf
(0.1 M, 2× 20 ml), water (20 ml), and then dried over
Na2SO4. Solvent was removed in vacuo and the crude
product was purified by repeated precipitation of con-
centrated acetone (2 ml) solutions with Et2O to afford
=
m-H 2,6-(CH3)2C6H3) 2.61 (s, 6H, N CCH3), 2.57
(s, 12H, 2,6-(CH3)2C6H3). Analytically calculated
for C26H36F12N2O2PtSb2 (found): C, 29.04 (27.78);
H, 3.37 (3.38); N, 2.61 (2.45). For X = BF4−, 3
(200 mg, 0.358 mmol) was combined with AgBF4
(143 mg, 0.734 mmol) in dry acetone (15 ml). Workup
as above afforded a bright yellow powder. Yield
1
132 mg (47.4%). H NMR appears to be consistent
with the formation of the bis-acetone complex, how-
ever, the product is again extremely hygroscopic and
yielded poor analysis.
1
a red-brown powder. Yield 66 mg (48%). H NMR
3
((CD3)2CO, ␦): 8.54 (d, 1H, JHH = 2.90 Hz, H3
3.6. [(CyN=C(H)–C(H)=NCy)Pt(CH3CN)2]X2,
(Cy = C6H11, X = OTf−, BF4−, PF6−, SbF6−), 6
on N2C3H3), 7.49–7.47 (br m, 5H, 2,6-(CH3)2C6H3
3
and H4 on C6H4), 7.03 (t, 1H, JHH = 8.8 Hz, H5
3
on C6H4), 6.61 (t, 1H, JHH = 8.7 Hz, H6 on C6H4),
3
6.51 (t, 1H, JHH = 2.90 Hz, H2 on N2C3H3), 5.40
The solids 4 (135 mg, 0.278 mmol) and AgOTf
(149.6 mg, 0.583 mmol) were suspended in a mixture
of CH2Cl2 (10 ml) and CH3CN (0.5 ml) and stirred
for 3 h at room temperature while protected from
light. The resulting mixture was filtered through a
pad of Celite and the solvent removed in vacuo. The
yellow-orange oil was dissolved in acetone and a
small amount of CH3CN, filtered and solvent again
was removed in vacuo. Addition of Et2O to a con-
centrated acetone solution of this mixture yields a
viscous, yellow-orange oil. All attempts to recrystal-
lize or precipitate a solid resulted in the formation of
this oil. Redissolving the oil in acetone, followed by
removal of the solvent in vacuo a number of times
afforded a foam that could be triturated with Et2O to
yield a very sticky slightly yellow solid. Prolonged
exposure to air results in the solid turning to an or-
3
(d, 1H, JHH = 2.90 Hz, H1 on N2C3H3), 5.18
3
3
(dd, 1H, JHH = 2.90 Hz, JPtH = 30 Hz, H7 on
C6H4). 13C {1H} NMR ((CD3)2CO, δ): 184.5, 180.6,
145.5, 145.2, 145.1, 139.3, 132.5, 132.3, 131.6,
1
130.0, 129.7, 127.3, 125.7, 123.8, 122.4 (q, JCF
=
323 Hz), 112.2, 108.7, 21.2, 20.0, 17.83, 17.82, (was
not able to detect 195Pt coupling to C bonded to
Pt). Analytically calculated for C30H31F3N4O3PtS
(found): C, 46.21 (46.01); H, 4.01 (4.09); N, 7.19
(6.96).
3.8. [(ArN=C(Me)–C(Me)=NAr)Pt(κ2-(C,N)-1-
C6H4NC5H4)]X, (Ar = 2,6-(CH3)2C6H3;
X = OTf −), 12
−
=
A solution of 5a (X OTf ) (63 mg, 0.073 mmol)
and 2-phenylpyridine (20.5 ml, 0.146 mmol) in ace-
tone (10 ml) was heated at 85 ◦C overnight in a
re-sealable Schlenk tube. The resulting dark red so-
lution was diluted with water (30 ml), extracted with
CH2Cl2 (2× 10 ml), then dried over Na2SO4. The
solvent was removed in vacuo and the crude product
was purified by repeated precipitation of concen-
trated acetone (2 ml) solutions with Et2O at −20 ◦C
to afford an orange-brown powder. Yield: 24 mg
(41%). 1H NMR ((CD3)2CO, δ): 7.96–8.05 (over-
lapping m, 2H, H3 and H4 on NC5H4), 7.61 (d, 1H,
3JHH = 7.61 Hz, H5 on C6H4), 7.40–7.47 (br m, 6H,
1
ange oil. Yield: 66 mg (30%) H NMR (TFE-d3 δ):
3
=
8.25 (s, 2H, JPtH = 106.4 Hz, N CH), 3.77 (m, 2H,
ipso H on CyN), 2.73 (s, 6H, Pt(NCMe)2), 2.31–1.23
(m, 20H, CH2 on CyN). 13C {1H} NMR (CD3CN,
δ): 172.85, 71.02, 32.95, 25.78, 25.71. Analytically
calculated for C20H30F6N406PtS2 (found): C, 30.19
(29.27);−H, 3.80 (3.78); N, 7.04 (6.38). Isolation of
the BF4 salt was similarly complicated by the prod-
uct precipitating as a vi−scous oil. This occurred to a
lesser ext−ent with SbF6 as the counteranion, while
with PF6 the final product precipitated as a nice
off-white solid.

14
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
3
2,6-(CH3)2C6H3), 7.00 (t, 1H, JHH = 7.61 Hz, H6
3.10. [(ArN=C(Me)–C(Me)=NAr)Pt(κ2-(C,N)-
(2-C4H2S)NC5H4)]X, (Ar = 2,6-(CH3)2C6H3;
X = OTf −), 14
3
3
on C6H4), 6.87 (dd, 1H, JHH = 6.3 Hz, JPtH
=
34 Hz, H1 on NC5H4), 6.81 (t, 1H, JHH = 6.3 Hz,
3
H2 on NC5H4), 6.65 (t, 1H, JHH = 7.54 Hz, H7 on
3
A solution of 5a (X = OTf−) (100 mg, 0.115 mmol)
and 2-(2-thienyl)pyridine (37.2 mg, 0.230 mmol) in
acetone (25 ml) was heated to reflux for 4 h. The
solvent was then removed. The resulting dark red
residue was dissolved in CH2Cl2 (50 ml), washed
with aqueous HOTf (0.1 M, 2 ml), water (10 ml), and
then dried over Na2SO4. Solvent was removed in
vacuo and the crude product was purified by repeated
precipitation of concentrated CH2Cl2 (2 ml) solutions
with Et2O/hexanes to afford a maroon solid. Yield:
3
C6H4), 5.27 (dd, 1H, JHH = 8.1 Hz, JPtH = 39 Hz,
8
=
H on C6H4), 2.44 (s, 3H, N CCH3), 2.36 (s, 6H,
=
2,6-(CH3)2C6H3), 2.34 (s, 3H, N CCH3), 2.28 (s,
6H, 2,6-(CH3)2C6H3). 13C {1H} NMR ((CD3)2CO,
δ): 184.14, 180.47, 168.17, 148.68, 147.10, 145.58,
144.61, 142.13, 138.66, 132.16, 132.02, 131.31,
130.33, 129.97, 129.78, 129.72, 129.21, 126.81,
124.55, 122.77, 120.41, 21.44, 20.67, 17.96, 17.92.
Analytically calculated for C32H32F3N3O3PtS
(found): C, 48.6 (47.13); H, 4.08 (4.10); N, 5.31
(5.08).
1
48.3 mg (52%) H NMR ((CD3)2CO, δ): 7.88 (t, 1H,
3JHH = 7.6 Hz, H3 on NC5H4), 7.53–7.40 (over-
lapping, 7H, 2,6-(CH3)2C6H3 and H3 on NC5H4),
3.9. [(ArN=C(Me)–C(Me)=NAr)Pt(κ2-(C,N)-
(2-C2H2NC5H4)]X, (Ar = 2,6-(CH3)2C6H3;
X = OTf −), 13
3
7.24 (d, 1H, JHH = 5.0 Hz, H5 on C4H2S), 6.68
3
(t, 1H, JHH = 6.9 Hz, H2 on NC5H4), 6.52 (dd,
3
3
1H, JHH = 6.0 Hz, JPtH = 37 Hz, H1 on NC5H4),
3
3
A solution of 5a (X = OTf−) (100 mg, 0.115 mmol)
and 2-vinylpyridine (27.4 ␮l, 0.230 mmol) in ace-
tone (40 ml) was heated to reflux for 4 h. The sol-
vent was then removed in vacuo to 1/4 the volume.
This was diluted with CH2Cl2 (30 ml), washed with
aqueous HOTf (0.1 M, 2× 20 ml), water (20 ml),
and then dried over Na2SO4. Solvent was removed
in vacuo and the crude product was purified by re-
peated precipitation of concentrated acetone (2 ml)
solutions with Et2O/hexanes to afford a rust-colored
powder. Yield: 44 mg (52%) 1H NMR ((CD3)2CO,
5.57 (dd, 1H, JHH = 5.0 Hz, JPtH = 13.3 Hz,
6
=
H on C4H2S) 2.44 (s, 3H, N CCH3), 2.39 (s, 6H,
2,6-(CH3)2C6H3), 2.33 (s, 6H, 2,6-(CH3)2C6H3),
2.31 (s, 3H, N CCH3). 13C {1H} NMR ((CD3)2CO,
=
δ): 184.06, 181.07, 163.57, 148.09, 146.46, 144.75,
143.94, 143.08, 142.74, 132.08, 131.58, 131.20,
130.36, 130.04, 129.95, 127.83, 127.69, 122.4 (q,
1JCF = 323 Hz), 120.71, 119.01, 20.86, 20.62, 17.95,
17.85. Analytically calculated for C30H30F3N3O3PtS2
(found): C, 45.22 (44.99); H, 3.80 (3.89); N, 5.27
(5.09).
3
δ): 7.84 (t, 1H, JHH = 7.7 Hz, H3 on NC5H4),
7.45–7.34 (br m, 6H, 2,6-(CH3)2C6H3), 7.30 (d,
3.11. [(ArN=C(Me)–C(Me)=NAr)Pt(CH3)
1H, JHH = 7.5 Hz, H4 on NC5H4), 6.78 (t, 1H,
3
(2-C6H5NC5H4)][BF4], (Ar = 2,6-(CH3)2C6H3) 15
3JHH = 7.8 Hz, H2 on NC5H4), 6.32 (dd, 1H,
3JHH = 5.9 Hz, JPtH = 30 Hz, H1 on NC5H4),
To a solution of (ArN C(Me)–C(Me) NAr)Pt-
Me2 (Ar = 2,6-(CH3)2C6H3) (100 mg, 0.193 mmol)
in acetone (15 ml), cooled to −20 ◦C, was added
HBF4(aq) (48%, 25.2 ␮l, 0.193 mmol). Upon ad-
dition of the acid the color of the solution turned
yellow-orange from dark red. The reaction was
stirred for 15 min, after which 2-phenylpyridine
(55.2 ml, 0.386 mmol) was added. The reaction
was allowed to stir for 2 h. while warming to
room temperature. Solvent was then removed
under vacuo and resulting red-orange residue
was precipitated from acetone with Et2O. Re-
crystallization from CH2Cl2 and Et2O afforded
3
=
=
6.30 (dd, 1H, JHH = 7.9 Hz, JPtH = 89 Hz, H6 on
3
3
3
3
C2H2), 5.84 (dd, 1H, JHH = 7.9 Hz, JPtH = 82 Hz,
5
=
H on C2H2), 2.48 (s, 3H, N CCH3), 2.35 (s, 6H,
2,6-(CH3)2C6H3), 2.33 (s, 6H, 2,6-(CH3)2C6H3), 2.31
(s, 3H, N CCH3). 13C {1H} NMR ((CD3)2CO, δ):
=
182.59, 180.57, 171.88, 156.84 (d, 1JPtC = 1041 Hz),
148.47, 145.07, 144.73, 143.02, 142.85, 139.50,
130.89, 130.66, 130.13, 129.73, 129.41, 122.4 (q,
1JCF = 323 Hz), 121.67, 121.39, 20.50, 2014, 17.94,
17.85. Analytically calculated for C28H30F3O3PtS
(found): C, 45.40 (45.17); H, 4.08 (4.15); N, 5.67
(5.45).

A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
15
an orange solid. Yield: 45 mg (31%) 1H NMR
3.12. X-ray structure determinations of 12 and 14
3
(CD2Cl2, δ): 8.64 (dd, 1H, JHH
=
5.1 Hz,
3JPtH
=
49 Hz, ␣-H on NC5H4), 7.8–7.7 (m
Dark red crystals of 12 and 14 suitable for X-ray
structure determinations were grown by layering con-
centrated solutions of 12 (CH2Cl2) and 14 (acetone)
with Et2O at −40 ◦C. The position of the Pt atom in
12 was located with a Patterson map. The presence of
the Pt atom creates a pseudo-mirror plane perpendic-
ular to the b-axis and passing through the Pt position.
The positions of the remaining atoms in the difference
Fourier maps were complicated by this pseudo-mirror
but it was possible to distinguish actual peaks from
ghost peaks. Numerous absorption correction schemes
were tried but all resulted in unsatisfactory refinement
overlapping, 3H, C6H4NC5H4), 7.54–7.42 (m
overlapping, 3H, C6H4NC5H4), 7.30–7.17 (m over-
lapping, 5H, 2,6-(CH3)2C6H3), 7.0–6.95 (m over-
lapping, 2H, 1H from C6H4NC5H4 and 1H from
=
2,6-(CH3)2C6H3), 2.30 (s, 3H, N CCH3), 2.20 (s, 3H,
=
N CCH3), 2.18 (s, 3H, 2,6-(CH3)2C6H3), 1.84 (s, 3H,
2,6-(CH3)2C6H3), 1.83 (s, 3H, 2,6-(CH3)2C6H3) 1.17
(s, 3H, 2,6-(CH3)2C6H3), 0.67 (3H, 2JPtH = 77.1 Hz,
Pt(CH3)). Analytically calculated for C32H36BF4N3Pt
(found): C, 51.62 (50.05); H, 4.87 (4.80); N, 5.64
(5.31).
Table 3
X-ray crystallographic data for 12 and 14a
Complex
12
14
Empirical formula
Formula weight
Crystal habit
C
31H32N3PtBF4
C29H29N3SPtBF4
733.51
Plate
728.50
Plate
Crystal size
Crystal color
Diffractometer
Wavelength
Temperature
0.22 mm × 0.14 mm × 0.04 mm
Brown/yellow
CCD area detector
0.71073 Mo K␣
98 K
0.26 mm × 0.19 mm × 0.07 mm
Red
CCD area detector
0.71073 Mo K␣
98 K
Unit cell dimensions
a = 0.5686(14) Å
b = 11.8223 (15) Å
c = 11.9120(16) Å
α = 90◦
a = 10.6141(19) Å
b = 11.884(2) Å
c = 11.927(2) Å
α = 90◦
β = 106.1992(2)◦
γ = 90◦
β = 105.081(3)◦
γ = 90◦
Volume
Z
1429.3(3) Å3
2
1452.6(4) Å3
2
Crystal system
Space group
Density (calculated)
Theta range
h (min, max)
k (min, max)
l (min, max)
Monoclinic
P21
Monoclinic
P21
1.693 g/cm3
1.78–28.45◦
−13, 13
1.677 Mg/m3
1.77–28.48◦
−13, 14
−15, 15
−15, 15
−15, 15
−15, 15
Reflections collected
Independent reflections
Rint
27944
21549
6691
0.1045
1.448
0.0393
6740
0.0597
1.308
0.0396
GOF on F2
Final R indices [I > 2␴(I)]b
Final weighted R [Fo2]c
0.0489
0.0564
a SADABS absorption correction applied.
ꢀ
ꢂ
ꢁ
ꢁ
b R(F) =
||Fc| − |Fo|| / |Fo|.
ꢃ
ꢁ
ꢄ1/2
ꢁ
c Rw(F2) =
w(Fo2 − Fc2)2/ w(Fo2)
.

16
A.G. Wong-Foy et al. / Journal of Molecular Catalysis A: Chemical 189 (2002) 3–16
results, therefore the uncorrected data was used in the
final refinement. All large peaks in the final difference
Fourier map are near the Pt atom. Crystallographic
data are shown in Table 3.
[11] M.W. Holtcamp, J.A. Labinger, J.E. Bercaw, J. Am. Chem.
Soc. 119 (1997) 848.
[12] D.D. Wick, K.I. Goldberg, J. Am. Chem. Soc. 119 (1997)
10235.
[13] S. Reinartz, P.S. White, M. Brookhart, J.L. Templeton,
Organometallics 20 (2001) 1709.
[14] L. Johansson, O.B. Ryan, M. Tilset, J. Am. Chem. Soc. 121
(1999) 1974.
4. Supplementary material
[15] S.D. Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100
(2000) 1169.
[16] L.M. Rendina, R.J. Puddephatt, Chem. Rev. 97 (1997) 1735.
[17] G. Van Koten, K. Vrieze, Adv. Organomet. Chem. 21 (1982)
151.
X-ray crystallographic data for all new struc-
tures, including complete listings of fractional co-
ordinates, displacement parameters, distances and
angles, and structure factors, have been deposited
at the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK (e-mail:
software@chemcrys.cam.ac.uk). Copies can be ob-
tained on request free of charge, by quoting the pub-
lication citation and the deposition numbers, which
are: 12, deposition number 166553; 14, deposition
number 149652.
[18] S.D. Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100
(2000) 1169.
[19] H.A. Zhong, J.A. Labinger, J.E. Bercaw, J. Am. Chem. Soc.
124 (2002) 1378.
[20] L. Johansson, M. Tilset, J.A. Labinger, J.E. Bercaw, J. Am.
Chem. Soc. 122 (2000) 10846.
[21] M.W. Holtcamp, J.A. Labinger, J.E. Bercaw, Inorg. Chim.
Acta 265 (1997) 117.
[22] J.F. Houlis, D.M. Roddick, J. Am. Chem. Soc. 18 (1998)
2536.
[23] A.D. Ryabov, Chem. Rev. 90 (1990) 403.
[24] G. Dyker, Angew. Chem. Int. Ed. Engl. 38 (1999) 1698.
[25] B. Rybtchinski, D. Milstein, J. Am. Chem. Soc. 121 (1999)
4528.
[26] D.L. Thorn, Organometallics 17 (1998) 348.
[27] A.J. Canty, G. Van Koten, Acc. Chem. Res. 28 (1995) 406.
[28] D.C. Griffiths, G.B. Young, Organometallics 8 (1989) 875.
[29] G.M. Whitesides, Pure Appl. Chem. 53 (1981) 287.
[30] L. Maresca, G. Natile, L. Cattalini, Inorg. Chim. Acta 14
(1975) 79.
Acknowledgements
This work was supported by the National Science
Foundation (Grant No. CHE 98-07496) and by BP.
The authors wish to thank Hong A. Zhong for helpful
discussions and for her thorough proofreading in the
early stages of the manuscript.
[31] H.C. Clark, L.E. Manzer, J. Am. Chem. Soc. 95 (1973) 3812.
[32] V.G. Albano, G. Natile, A. Panunzi, Coord. Chem. Rev. 133
(1994) 67.
[33] M.E. Cucciolito, V. De Felice, A. Panunzi, A. Vitagliano,
Organometallics 13 (1989) 1180.
[34] O.F. Wendt, N.-F.K. Kaiser, L.I. Elding, J. Chem. Soc., Dalton
Trans. (1997) 4733.
[35] H. Heiberg, L. Johansson, O. Gropen, O.B. Ryan, O. Swang,
M. Tilset, J. Am. Chem. Soc. 122 (2000) 10831.
[36] M.V. Kulikova, K.P. Balashev, P.I. Kvam, J. Songstad, Russ.
J. Gen. Chem. 70 (2000) 163, and references cited therein.
[37] T.G. Appleton, H.C. Clark, L.E. Manzer, Coord. Chem. Rev.
10 (1973) 335.
References
[1] N.F. Gol’dshleger, V.V. Es’kova, A.E. Shilov, A.A. Shteinman,
Zh. Fiz. Khim. 46 (1972) 1353.
[2] A.E. Shilov, G.B. Shul’pin, Activation and Catalytic Reactions
of Saturated Hydrocarbons in the Presence of Metal
Complexes, Kluwer Academic Publishers, Boston, MA, 2000.
[3] S. Stahl, J.A. Labinger, J.E. Bercaw, Angew. Chem. Int. Ed.
37 (1998) 2181.
[38] B. Mart´ın-Matute, C. Mateo, D.J. Cárdenas, A.M. Echavarren,
[4] A.E. Shilov, G.B. Shul’pin, Chem. Rev. 97 (1997) 2879.
[5] L. Wang, S.S. Stahl, J. Labinger, J.E. Bercaw, J. Mol. Catal.
A Chem. 116 (1997) 269.
Chem. Eur. J. 7 (2001) 2341.
[39] L. Johansson, Ph.D. Thesis, University of Oslo, Oslo, Norway,
2000.
[40] J. Cámpora, J.A. López, P. Palma, P. Valerga, E. Spillner, E.
Carmona, Angew. Chem. Int. Ed. Engl. 38 (1999) 147.
[41] B.J. Burger, J.E. Bercaw, in: A.L. Wayda, M.Y. Darensbourg
(Eds.), Experimental Organometallic Chemistry, American
Chemical Society, Washington, DC, 1987, pp. 79.
[42] H. Tom Dieck, M. Svoboda, T. Greiser, Z. Naturforsch, B
Anorg. Chem., Org. Chem. 36B (1981) 823.
[43] M. Svoboda, H. Tom Dieck, C. Krueger, Y.-H. Tsay, Z.
[6] G.A. Luinstra, L. Wang, S.S. Stahl, J.A. Labinger, J.E.
Bercaw, J. Organomet. Chem. 504 (1995) 75.
[7] V.V. Zamashchikov, E.S. Rudakov, S.A. Mitchenko, S.L.
Litvinenko, Teor. Eksp. Khim. 18 (1982) 510.
[8] S.S. Stahl, J.A. Labinger, J.E. Bercaw, J. Am. Chem. Soc.
118 (1996) 5961.
[9] S.S. Stahl, J.A. Labinger, J.E. Bercaw, J. Am. Chem. Soc.
117 (1995) 9371.
[10] M.W. Holtcamp, L.M. Henling, M.W. Day, J.A. Labinger,
J.E. Bercaw, Inorg. Chim. Acta 270 (1998) 467.
Naturforsch,
814.
B Anorg. Chem. Org. Chem. 36B (1981)