Reductive elimination and dissociative β-hydride abstraction from Pt(IV) hydroxide and methoxide complexes

Article abstract of DOI:10.1021/om800905q

The platinum(IV) hydroxide and methoxide complexes fac-(dppbz)PtMe ₃(OR)(dppbz = o-bis(diphenylphosphino)benzene; R = H (1), CH ₃ (2)) have been prepared and characterized. Thermolysis of hydroxide 1 produces (dppbz)PtMe₂

Full text of DOI:10.1021/om800905q

Organometallics 2009, 28, 277–288
277
Reductive Elimination and Dissociative ꢀ-Hydride Abstraction from
Pt(IV) Hydroxide and Methoxide Complexes
Nicole A. Smythe,† Kyle A. Grice,† B. Scott Williams,*,‡ and Karen I. Goldberg*,†
Department of Chemistry, Box 351700, UniVersity of Washington, Seattle, Washington 98195-1700, and
The Joint Science Department of The Claremont Colleges, W. M. Keck Science Center, 925 North Mills
AVenue, Claremont, California 91711
ReceiVed September 17, 2008
The platinum(IV) hydroxide and methoxide complexes fac-(dppbz)PtMe3(OR) (dppbz ) o-bis(diphe-
nylphosphino)benzene; R ) H (1), CH3 (2)) have been prepared and characterized. Thermolysis of
hydroxide 1 produces (dppbz)PtMe2 (3) and methanol in a rare example of directly observed sp3
carbon-oxygen reductive elimination from a metal center to form an alcohol. Competitive carbon-carbon
reductive elimination to form (dppbz)PtMe(OH) (5) and ethane also occurs. In contrast, the major reaction
observed upon thermolysis of the methoxide analog 2 is neither carbon-oxygen nor carbon-carbon
reductive elimination. Instead, products expected from formal ꢀ-hydride elimination followed by
carbon-hydrogen reductive elimination are detected. Mechanistic studies suggest the operation of an
alternative mechanism to that most commonly accepted for this fundamental reaction; a dissociative
ꢀ-hydride abstraction pathway is proposed.
those of their metal-carbon analogs. This contribution focuses
on these differences in reductive elimination and ꢀ-hydride
elimination pathways when hydroxide and alkoxide groups are
involved.
In contrast to the large number of examples and mechanistic
studies of C-C reductive elimination reactions,4 there are
relatively few reported direct observations of C-O reductive
Introduction
Reductive elimination and ꢀ-hydride elimination are two
fundamental transformations that are ubiquitous in metal-
mediated processes. Consequently, significant effort has been
directed toward understanding their mechanisms.1 However,
studies of these elimination reactions have largely focused on
systems involving metal-carbon bonds; comparatively little is
known about the related reactions with metal-heteroatom
linkages. The fact that there are fewer studies of the mechanisms
of fundamental reactions of late metal hydroxides and alkoxides
relative to those of late metal alkyls may be attributable to there
being fewer monomeric hydroxide and alkoxide complexes
available for study.2 Since theoretical and experimental studies
have shown that late metal M-O bonds are at least as strong
as their alkyl analogs,3 it has been proposed that the relative
scarcity of such species may in fact stem from their access to
low energy decomposition pathways,2a,c including reductive
elimination and ꢀ-hydride elimination. Thus, the elucidation of
the mechanisms of these fundamental reactions for metal-
heteroatom species is important. Recent contributions appear
to indicate that the more polar nature of metal-heteroatom
bonds and the presence of heteroatom electron lone pairs can
result in significant differences in reaction mechanisms from
(4) Reference 1b, pp 170-175.
(5) (a) Mann, G.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 13109.
(b) Hartwig, J. F. Angew. Chem., Int. Ed. Engl. 1998, 37, 2046, and
references therein. (c) Hartwig, J. F. Acc. Chem. Res. 1998, 31, 852. (d)
Mann, G.; Incarvito, C.; Rheingold, A. L.; Hartwig, J. F. J. Am. Chem.
Soc. 1999, 121, 3224. (e) Shelby, Q.; Kataoka, N.; Mann, G.; Hartwig, J.
J. Am. Chem. Soc. 2000, 122, 10718. (f) Mann, G.; Shelby, Q.; Roy, A. H.;
Hartwig, J. F. Organometallics 2003, 22, 2775.
(6) (a) Widenhoefer, R. A.; Zhong, H. A.; Buchwald, S. L. J. Am. Chem.
Soc. 1997, 119, 6787. (b) Widenhoefer, R. A.; Buchwald, S. L J. Am. Chem.
Soc. 1998, 120, 6504, and references therein.
(7) Komiya, S.; Akai, Y.; Tanaka, K.; Yamamoto, T.; Yamamoto, A.
Organometallics 1985, 4, 1130.
(8) (a) Matsunaga, P. T.; Mavropoulos, J. C.; Hillhouse, G. L. Polyhe-
dron 1995, 14, 175. Note that the reductive elimination from Ni(II) is
promoted by oxidants, and thus Ni(III) is likely involved in the reactions.
(b) Han, R.; Hillhouse, G. L. J. Am. Chem. Soc. 1997, 119, 8135. (c) Koo,
K.; Hillhouse, G. L. Organometallics 1998, 17, 2924.
(9) Haarman, H. F.; Kaagman, J.-W. F.; Smeets, W. J. J.; Spek, A. L.;
Vrieze, K. Inorg. Chim. Acta 1998, 270, 34.
(10) Dick, A. R.; Kampf, J. W.; Sanford, M. S. J. Am. Chem. Soc. 2005,
127, 12790.
* Corresponding authors. E-mail: goldberg@chem.washington.edu;
swilliams@jsd.claremont.edu.
(11) (a) Canty, A. J.; Jin, H. J. Organomet. Chem. 1998, 565, 135. (b)
Canty, A. J.; Jin, H.; Skelton, B. W.; White, A. H. Inorg. Chem. 1998, 37,
3975. (c) Canty, A. J.; Denney, M. C.; van Koten, G.; Skelton, B. W.;
White, A. H. Organometallics 2004, 23, 5432.
† University of Washington.
‡ Joint Science Department of the Claremont Colleges.
(1) (a) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry; Univer-
sity Science Books: Sausalito, CA, 1987; pp 322-353; 386-389. (b)
Crabtree, R. H. The Organometallic Chemistry of the Transition Metals,
4th ed.; John Wiley & Sons: Hoboken, NJ, 2005; pp 55-60.
(2) (a) Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163. (b) Fulton,
J. R.; Holland, A. W.; Fox, D. J.; Bergman, R. G. Acc. Chem. Res. 2002,
35, 44. (c) Roesky, H. W.; Singh, S.; Yusuff, K. K. M.; Maguire, J. A.;
Hosmane, N. S. Chem. ReV. 2006, 106, 3813.
(3) (a) Bryndza, H. E.; Fong, L. K.; Paciello, R. A.; Tam, W.; Bercaw,
J. E. J. Am. Chem. Soc. 1987, 109, 1444. (b) Holland, P. L.; Andersen,
R. A.; Bergman, R. G.; Huang, J.; Nolan, S. P. J. Am. Chem. Soc. 1997,
199, 12800.
(12) Thompson, J. S.; Randall, S. L.; Atwood, J. D. Organometallics
1991, 10, 3906. See ref 5a for more information on this reaction.
(13) Sanford, M. S.; Groves, J. T. Angew. Chem., Int. Ed. 2004, 43,
588.
(14) (a) Vedernikov, A. N.; Binfield, S. A.; Zavalij, P. Y.; Khusnutdi-
nova, J. R. J. Am. Chem. Soc. 2006, 128, 82. (b) Khusnutdinova, J. R.;
Zavalij, P. Y.; Vedernikov, A. N. Organometallics 2007, 26, 3466. (c)
Khusnutdinova, J. R.; Newman, L. L.; Zavalij, P. Y.; Lam, Y.-F.;
Vedernikov, A. N. J. Am. Chem. Soc. 2008, 130, 2174.
(15) Williams, B. S.; Goldberg, K. I. J. Am. Chem. Soc. 2001, 123, 2576,
and references therein.
(16) Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw,
J. E. J. Organomet. Chem. 1995, 504, 75.
10.1021/om800905q CCC: $40.75
2009 American Chemical Society
Publication on Web 11/26/2008

278 Organometallics, Vol. 28, No. 1, 2009
Smythe et al.
elimination.5-16 Those cases involving alkyl and hydroxide
groups to produce alcohols, a potentially highly valuable product
release step in catalysis, are exceptionally rare.14,16 Notably,
several examples of C(sp3)-O reductive elimination reactions
to form methyl aryl ethers and methyl carboxylates have recently
been reported.15 In these reactions, which occur at Pt(IV)
centers, a common two-step pathway involving dissociation of
the anionic oxygen group followed by nucleophilic attack of
the oxyanion on the cationic Pt(IV) alkyl intermediate has been
proposed. A similar nucleophilic coupling has also been
proposed as the product-forming step in platinum-catalyzed
alkane oxidation reactions.16,17 In contrast to these C-O
reductive elimination reactions, C-C bond formation typically
occurs via a concerted three-centered transition state.4 Thus,
the mechanisms of reductive elimination involving metal-oxygen
bonds can be quite different than those involving metal-carbon
bonds.
Figure 1. ORTEP drawing of complexes 1 and 2, with thermal
ellipsoids at the 50% probability level. Hydrogen atoms are omitted
for clarity.
couplings from Pt(IV),15 is indicated. Given the rarity of the
formation of an alcohol by C(sp3)-O reductive elimination from
a metal center14a,b and the promise of this coupling as a product-
forming step in catalysis, this mechanistic insight is relevant to
rational catalyst design efforts.
Whereas alkyl C-O reductive elimination is a relatively rare
reaction, ꢀ-hydride elimination is accepted as a common and
predictable decomposition pathway for transition metal alkox-
ides. Notably, the reaction of alcohols and bases with transition
metal halides is a traditional method of preparing metal hydride
complexes.18 Such reactions are thought to proceed via me-
tathesis of the halide for an alkoxide group followed by
ꢀ-hydride elimination from the intermediate metal-alkoxide
species. It has been generally assumed that the latter step
proceeds by the same mechanism that has been established for
metal alkyl complexes.19 As described in organometallic
textbooks, for metal alkyl complexes, the hydride moves from
the ꢀ-carbon of the alkyl through a four-coordinate planar
transition state to an open site on the metal that is cis to the
alkyl ligand. This cis open site is a requirement for this
mechanism, and the stability of many metal alkyls with
ꢀ-hydrogens has been attributed to the lack of labile ancillary
ligands or other access to a cis open site on the metal.1 However,
a few examples of ꢀ-hydride elimination from metal alkoxide
complexes have been recently reported wherein a cis open site
does not appear to be available,20-23 suggesting that an
alternative mechanism is needed in at least some cases to
account for this fundamental reaction when metal-heteroatom
bonds are involved.
In contrast to the C-O reductive elimination reaction
observed from the hydroxide complex 1, thermolysis of the
methoxide analog 2 leads primarily to products expected from
formal ꢀ-hydride elimination followed by C-H reductive
elimination. Notably, this formal ꢀ-hydride elimination occurs
from a coordinatively saturated alkoxide complex without an
easily accessible cis open site. Evidence of alkoxide dissociation
is presented and a dissociatiVe ꢀ-hydride abstraction mechanism,
distinct from the traditional ꢀ-hydride elimination mechanism
of metal alkyls, is proposed for this reaction. This dissociative
ꢀ-hydride abstraction may be a general and unappreciated
decomposition pathway for metal alkoxides and amide species.
It is significant that both alkyl C-O reductive elimination and
formal ꢀ-hydride elimination from the Pt(IV) complexes
reported herein involve initial heterolytic cleavage of the Pt-O
bond. A similar reaction step for a metal alkyl species
(heterolytic M-C bond cleavage) would be unreasonable. Thus,
entirely different mechanisms for these fundamental reactions
of reductive elimination and ꢀ-hydride elimination, distinct from
those of metal alkyls, are possible for metal alkoxides.
Results
In this contribution, we report the synthesis, characterization,
and thermal reactivity of the monomeric platinum(IV) hydroxide
and alkoxide complexes fac-(dppbz)PtMe3(OR) (dppbz ) o-bis-
(diphenylphosphino)benzene; R ) H (1), CH3 (2)). C-O
reductive elimination was observed upon thermolysis of the
platinum(IV)-hydroxide complex 1 in C6D6. This is the first
direct observation of carbon-oxygen reductive elimination
involving an alkyl and hydroxide group to form alcohol in a
nonaqueous solvent system. Studies of this reaction indicate that
the mechanism is different than that of C-C reductive elimina-
tion. Despite hydroxide being a relatively poor leaving group
and the reaction taking place in a nonpolar solvent, dissociation
of the heteroatom group followed by nucleophilic C-O
coupling, similar to the mechanism proposed for other C(sp3)-O
Synthesis and Characterization of Platinum Complexes
1 and 2. The monomeric Pt(IV) hydroxide and methoxide
complexes, fac-(dppbz)PtMe3(OR) (R ) H (1) and CH3 (2)),
were prepared via metathesis reactions of the appropriate
potassium salts with the Pt(IV) trifluoroacetate and acetate
complexes, respectively. Reaction of fac-(dppbz)PtMe3(O2CCF3)
with excess KOH proceeds in toluene at 35 °C to form the
Pt(IV) hydroxide complex 1, which was isolated in 95% yield.
The analogous Pt(IV) methoxide complex 2 was prepared by
reaction of fac-(dppbz)PtMe3(O2CCH3) with excess KOMe in
toluene at ambient temperature and was isolated in 75% yield.
Both 1 and 2 have been characterized by X-ray crystallography,
and the ORTEP diagrams appear in Figure 1. Selected bond
lengths and angles are presented in Table 1. Complete details
of the structures are given in Supporting Information.
(17) (a) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.;
Fujii, H. Science 1998, 280, 560. (b) Stahl, S. S.; Labinger, J. A.; Bercaw,
J. E. Angew. Chem., Int. Ed. 1998, 37, 10235.
1
Both species 1 and 2 exhibit the characteristic H NMR
signals for fac-L2PtMe3(X) species. In the H NMR spectrum
1
(18) Reference 1a, p 90.
of each species, there are signals for two equivalent equatorial
and one axial Pt(IV) methyl groups, with both signals exhibiting
coupling to 195Pt and 31P. The 1H NMR spectrum for 1 in C6D6
shows a Pt(IV) hydroxide signal at -2.39 ppm, which exhibits
coupling to 195Pt (2JPt-H ) 13 Hz) as well as to two equivalent
(19) Reference 1a, p 95.
(20) Blum, O.; Milstein, D. J. Am. Chem. Soc. 1995, 117, 4582.
(21) Blum, O.; Milstein, D. J. Organomet. Chem. 2000, 479, 593–594.
(22) Ritter, J. C. M.; Bergman, R. G. J. Am. Chem. Soc. 1998, 120,
6826.
(23) Fafard, C. M.; Ozerov, O. V. Inorg. Chim. Acta 2007, 360, 286.

ꢀ-Hydride Abstraction from Pt(IV) Complexes
Organometallics, Vol. 28, No. 1, 2009 279
Table 1. Selected Bond Lengths (Å) and Angles (deg) in Structures
of 1 and 2
1
2
Pt-C1
Pt-C2
Pt-C3
Pt-P1
Pt-P2
Pt-O1
2.061(10)
2.087(10)
2.090(9)
2.352(2)
2.356(2)
2.116(7)
Pt-C1
Pt-C2
Pt-C3
Pt-P1
Pt-P2
Pt-O1
O1-C4
2.080(6)
2.072(7)
2.095(7)
2.3559(17)
2.3650(17)
2.138(4)
1.410(10)
C4-O1-Pt
P1-Pt-P2
C2-Pt-C3
O1-Pt-P1
O1-Pt-P2
O1-Pt-C2
O1-Pt-C3
O1-Pt-C1
120.2(5)
83.89(6)
88.1(3)
86.58(14)
87.43(14)
91.1(3)
P1-Pt-P2
C2-Pt-C3
O1-Pt-P1
O1-Pt-P2
O1-Pt-C2
O1-Pt-C3
O1-Pt-C1
83.98(8)
86.4(4)
88.77(19)
90.0(2)
88.3(4)
88.5(4)
175.7(4)
Figure 2. Reactant and product concentrations at 23 °C in C6D6
for reaction between 2 ([2]0 ) 6.3 mM) and water ([water]0 ) 13
mM) to form 1 and methanol. Small amounts of 3 ([3]0 ) 0.4 mM)
and fac-(dppbz)PtMe3(OCH2OCH3) (4) also formed. (1 ) 9, 2 )[,
3 )2, 4 ) b).
92.1(2)
177.7(2)
1
31P nuclei (3JP-H ) 2.7 Hz). In the H NMR spectrum of 2 in
C6D6, a methoxide signal is observed at 4.04 ppm with Pt
satellites (3JPt-H ) 24.4 Hz). Finally, in the 31P{1H} NMR
spectra, both 1 and 2 display singlets with 195Pt satellites at 15.3
the formation of 1 and methanol as the concentration of 2
decreased. Equilibrium was reached after ca. 19 h at ambient
temperature (Scheme 1, Figure 2).
The equilibrium constant (Keq ) ca. 0.3) for this reaction was
calculated from eq 1 using the equilibrium concentrations
measured by 1H NMR spectroscopy for 1, 2, and methanol and
calculated for water.26
1
and 18.2 ppm, with JPt-P ) 1161 and 1159 Hz, respectively.
The X-ray structures of 1 and 2 both show a distorted
octahedral geometry around the Pt(IV) center with the two
structures exhibiting similar angles and bond distances between
the phosphorus and platinum atoms. Notably, the axial and
equatorial platinum-methyl bond lengths are very similar in 1
(2.06 Å (ax) vs. 2.09 Å (eq)), and the same is observed for 2
(2.08 Å (ax) vs 2.07 and 2.10 Å (eq)), suggesting that the trans
influence of hydroxide and methoxide are both reasonably
similar to that of a triaryl substituted phosphine at a Pt(IV)
center.
Keq)([1][CH3OH]) ⁄ ([2][H2O])
(1)
The Pt(II) species (dppbz)PtMe2, 3,27 was detected as early
as 10 h into the reaction and accounted for 5-10% of the total
Pt species in solution at equilibrium. In addition, a small amount
(ca. 2-3%) of fac-(dppbz)PtMe3(OCH2OCH3) (4, vide infra)
was also observed. The presence of these other products limited
the accuracy to which Keq could be determined.28
In a separate experiment, the reverse reaction, that between
methanol and 1, was also demonstrated. A solution of methanol
(700 mM) and 1 (14 mM) in THF-d8 at ambient temperature
was prepared, and formation of the methoxide complex 2 (ca.
50%) was observed by 31P{1H} NMR spectroscopy within 10
min. As with the reaction of 2 and water, formation of a small
amount of the Pt(II) complex (dppbz)PtMe2 (3) (<10%) was
noted.
Thermolysis of 1. Thermolysis of 1 (4.8-9.5 mM) at 120
°C in C6D6 yielded the C-O reductive elimination products
(dppbz)PtMe2 (3) (74-93%) and methanol (34-68%), as the
major organometallic and organic products, respectively (Scheme
2, Table 2). The C-C reductive elimination products (dppbz)-
PtMe(OH) (5) (7-26%) and ethane (8-25%)29 were also
observed in these reactions. Thus, the thermolysis of 1 is similar
to that of the related fac-L2PtMe3(OR) complexes (L2 ) dppe
The Pt-O bond lengths of previously reported Pt(IV)
hydroxide complexes range from 1.94 to 2.08 Å,24 yet none of
these contain a platinum(IV) hydroxide species wherein the
hydroxo-group is trans to a methyl group. Similarly, in related
structurally characterized platinum(IV) methoxide complexes
in which the methoxide ligands are trans to hydroxide, hydro-
peroxide-, methoxide-, acetate-, or pyridine-derived ligands, the
Pt(IV)-O bonds range in length from 1.88 to 2.05 Å.14b,25
A
Pt-O bond length of 2.116(7) Å is observed in the crystal
structure of 1, and the Pt-O bond length of 2 is 2.138(4) Å.
These values are significantly longer than previously reported
Pt-O bond lengths in Pt(IV) hydroxide and methoxide com-
plexes, presumably due to the stronger trans influence of the
axial methyl group relative to those of OH, OCH3, Cl, or NR3.
Equilibrium between 1 and 2. It has been shown that a
hydroxide ligand can exchange with an alcohol at a transition
metal center to generate a metal alkoxide species and water in
a reversible reaction.2a,3a Similar reactivity was found for the
hydroxide and methoxide complexes 1 and 2 which were
observed to equilibrate with one another in the presence of
methanol or water, respectively (Scheme 1). The equilibrium
constant for this reaction was estimated by monitoring a solution
prepared by the addition of water (ca. 2 equiv) to 2 in C6D6
(6.3 mM) over 3 days at ambient temperature. Examination of
(26) The concentrations of complexes 1-3 and methanol were deter-
mined by integration of their respective 1H NMR signals against an internal
standard (toluene). The 1H NMR signal for water was broad and overlapped
with the axial methyl signals of 1 and 3 and therefore was not integrated.
The concentration of water used in the calculation of the equilibrium constant
was taken to be the initial concentration of water ([H2O]0) with the
concentration of methanol formed ([CH3OH]t) deducted from it.
(27) (a) Crumpton, D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2000,
122, 962. (b) Crumpton-Bregel, D. M.; Goldberg, K. I. J. Am. Chem. Soc.
2003, 125, 9442.
(28) Complex 3 is a product of decomposition reactions from both 1
and 2. Methanol can also be generated in these reactions.
(29) As a result of the volatility of several of the organic species
(methane, ethane, etc.), it is expected that some portion of these products
is present in the headspace of the NMR tube, and therefore integrated yields
are low estimates of the actual yields.
the solution by H and 31P{1H} NMR spectroscopy revealed
1
(24) Based on a search of Pt(IV) hydroxide crystal structures in the
Cambridge Structural Database using Conquest 1.8, Copyright CCDC 2006.
Bruno, I. J.; Cole, J. C.; Edgington, P. R.; Kessler, M.; Macrae, C. F.;
McCabe, P.; Pearson, J.; Taylor, R. Acta Crystallogr. 2002, B58, 389.
(25) (a) Lee, Y.-A.; Jung, O.-S. Angew. Chem., Int. Ed. 2001, 40, 3868.
(b) Rostovtsev, V. V.; Henling, L. M.; Labinger, J. A.; Bercaw, J. E. Inorg.
Chem. 2002, 41, 3608. (c) Dick, A. R.; Kampf, J. W.; Sanford, M. S.
Organometallics 2005, 24, 482.

280 Organometallics, Vol. 28, No. 1, 2009
Smythe et al.
Scheme 1
Scheme 2
Table 2. Yields of Products from Thermolyses of 1 (4.8-9.5 mM) in
C6D6 at 120 °C with and without Additives of H2O (53 mM) and
CsOH · H2O (500 mM)
Table 3. Yields of Products in Thermolyses of 2 (6-7 mM) in C6D6
at 100 °C32,33
platinum product
% yield
organic product
(CH3)2O
C2H6
% yield
%
% yield
% yield
(dppbz)PtMe2 (3)
(dppbz)PtMe(OMe) (7)
94-100
0-6
ca. 1
3-7
ca. 1
product
yield
(with H2O)
(with CsOH · H2O)
29
(dppbz)PtMe2 (3)
CH3OH
74-93
34-68
7-26
87-90
51-72
10-13
10-13
93-97
72-76
3-7
CH2O
CH3OCHO
(CH3O)2CO
CH4
32-44
21-33
12-20
19-21
(dppbz)PtMe(OH) (5)
29
29
C2H6
8-25
3-7
CH3OH
or dppbz, OR ) carboxylate or aryloxide; dppe ) 1,2-
bis(diphenylphosphino)ethane) in which C-O and C-C reduc-
tive elimination reactions occurred competitively.15 Notably,
while the thermolyses of the related fac-(dppe)PtMe3(OAc) and
fac-(dppbz)PtMe3(OAr) complexes exhibited reproducible first-
order kinetics for the disappearance of the Pt(IV) complexes,
the time required for complete conversion of different samples
of 1 was observed to range from 5 h to more than 300 h at 120
°C.
thermolysis of 1. Mass balance of the Pt was confirmed in the
reaction as 3 and 5 were the only Pt-containing species in
solution upon reaction completion and the sum of the concentra-
tions of these complexes matched the starting concentration of
1.
Thermolysis of 2. It was anticipated that if carbon-oxygen
reductive elimination were to occur from 2, (dppbz)PtMe2 (3)
and dimethyl ether would be formed. Similarly, C-C reductive
elimination from 2 would generate (dppbz)PtMe(OMe) (7) and
ethane. Thermolysis reactions of 2 at 100 °C in C6D6 proceeded
to completion over 33-35 days and produced 3 as the major
organometallic product (94-100%) (Table 3). However, less
than 1% of the corresponding organic product from the C-O
coupling reaction, dimethyl ether, was detected. Instead, a
variety of other organic products, including methyl formate and
dimethyl carbonate, were observed (Table 3).32 The C-C
reductive elimination products, (dppbz)PtMe(OMe) (7) (0-6%)
and ethane (3-7%), were observed as very minor products in
this thermolysis reaction.
In the thermolyses of 1, although the Pt product of C-O
coupling, 3, was formed in high yield, the organic product,
methanol, was observed in significantly lower yield. In addition,
the 1H NMR spectra of the samples upon completion of
thermolysis showed the presence of other organic products:
methyl formate (0-3%), methane (1-7%), and minor amounts
of three unidentified organic products.30 It is also of interest
that other Pt(IV) species, fac-(dppbz)PtMe3(OMe) (2), fac-
(dppbz)PtMe3(OCH2OCH3) (4), and an unidentified fac-
(dppbz)PtMe3(X) species (6),31 were observed transiently during
thermolysis of 1. The maximum yields of 2, 4, and 6 detected
at any point in the thermolysis were 12%, 5%, and 2%,
respectively. No signals for these species remained visible in
As was the case in the thermolysis of 1, transient organo-
metallic species were observed during the reaction. Small
amounts of fac-(dppbz)PtMe3(OCH2OCH3) (4) (maximum of
7%) and fac-(dppbz)PtMe3(OH) (1) (maximum of 5%) were
detected at intermediate reaction times.
1
either the H or 31P NMR spectra at the conclusion of the
(30) The presence of three minor unidentified products was indicated
by the observation of three small singlets at 3.49 ppm (ca. 1%), 4.68 ppm
(0-5%), and 5.25 ppm (0-3%). Estimated yields of these unknown organic
products (calculated on the assumption that the integral of each singlet
represents three hydrogens) indicate that the contribution of these species
to the overall product distribution is small.
(32) In addition to the species listed in Table 3, two minor unidentified
1
species showing singlets at 3.47 (0-3%) and 3.27 (ca.1%) ppm in the H
NMR spectrum were also evident. Estimated yields of these unknown
organic products (calculated on the assumption that the integral of each
singlet represents three hydrogens) indicate that the contribution of these
species to the overall product distribution is small.
(31) The coupling patterns observed in the 1H NMR spectrum of
complex 6 are similar to those observed for other fac-(dppbz)PtMe3X
complexes. The coupling constant (1JPtP ) 1114 Hz) associated with the
phosphorus signal of this complex in the 31P{1H} NMR spectrum of this
complex is comparable to those of other Pt(IV) (1080-1200 Hz), rather
than Pt(II) (1700-1800 Hz), dppbz complexes, in which the phosphines
are trans to methyl groups. No hydroxide signal was observed for 6; see
Experimental Section for NMR characterization of 6.
(33) Note that the total yield of organic products, including the estimates
for the two unidentified species at 3.47 and 3.27, is 92 ( 9% (see refs 29
and 32). Methane is not included, as it is a byproduct of the formation of
formaldehyde, methyl formate, and dimethyl carbonate (vide infra). The
total Pt product yield is 100 ( 4%.

ꢀ-Hydride Abstraction from Pt(IV) Complexes
Organometallics, Vol. 28, No. 1, 2009 281
Scheme 3
Scheme 4
1
Reaction of 2 with Formaldehyde. The formation of species
4 in the thermolysis of 2 suggests a possible insertion reaction
of CH2O into a Pt(IV)-OCH3 bond. To investigate this proposal,
10 equiv of paraformaldehyde (p-CH2O) was added to 2 in C6D6
(7 mM). After only 15 min at ambient temperature, the platinum
species identifiable by NMR spectroscopy were 2 and 4 in
roughly equal amounts, and a trace amount of 3 (Scheme 3).
The addition of an even greater excess of p-CH2O (>33 equiv,
not all of which initially dissolved) to 2 in C6D6 at ambient
temperature resulted in the complete conversion of 2 to fac-
(dppbz)PtMe3(OCH2OCH3) (4) and several unidentified fac-
(dppbz)PtMe3(X)34 species within 15 min. The unidentified
complexes are most likely the products of multiple insertions
organic products in the H NMR spectrum (vide supra) were
not observed.36
Thermolysis of 1 in the Presence of CsOH and H2O.
Thermolysis of 1 in the presence of an excess of CsOH · H2O
(ca. 10 equiv, 50 mM) at 120 °C in C6D6 produced much higher
and more consistent yields of both 3 (93-97%) and methanol
(72-76%) than were seen in the absence of added hydroxide
(Table 2). These reactions were complete in ca. 60-70 h. C-C
reductive elimination products, 5 and ethane, were evident in
low yields (3-7%). No methyl formate was detected but very
1
small amounts of methane (2-3%) were observed in the H
NMR spectra.37
Reaction of 2 with Methanol. Upon dissolution of 2 in a
50:50 mixture of CH3OH/C6D6 at ambient temperature, fac-
(dppbz)PtMe3H (8) was formed within a few minutes, as
1
of CH2O into a Pt-O bond based on the similarity of their H
and 31P{1H} NMR spectral data to those of 4.
Species 4 was characterized by 1H and 31P{1H} NMR
spectroscopy but could not be successfully isolated from the
other insertion products. In a similar fashion, the addition of
excess p-CH2O to 1 in C6D6 produced several new Pt species,
which were assigned by NMR as fac-(dppbz)PtMe3(OCH2)xOH
complexes. The insertion of formaldehyde into a Pt-O bond is
an interesting reaction with very little precedent in the litera-
ture.35
1
observed by H NMR spectroscopy (Scheme 4).
Species 2 was completely consumed in under 20 min, but
the product 8 was unstable in solution, slowly decomposing to
form (dppbz)PtMe2 (3) and methane as previously reported.27
This reaction of 2 was significantly faster than the reaction of
2 in pure C6D6, in which complete consumption of the same
concentration of 2 typically required over 33 days at 100 °C.
Dissolution of 2 in a 50:50 mixture of CD3OD/C6D6 formed
fac-(dppbz)PtMe3D (8-d1) within a few minutes by H NMR
spectroscopy. Compound 8-d1 then reacted to form CH3D and
3 over time.
1
Thermolysis of 1 in the Presence of H2O. Thermolyses of
1 in C6D6 in the presence of 10 equiv of water (53 mM) at 120
°C in C6D6 produced consistently higher yields of C-O
reductive elimination products, 3 (87-90%), and methanol
(51-72%) than did thermolyses of 1 without added water (Table
2). The products of C-C reductive elimination, 5 and ethane,
were observed as minor products (10-13%). These thermolyses
of 1 demonstrated not only increased yields of methanol but
also increased consistency in the yields of the different products.
For example, the yield of 3 produced in the thermolyses samples
of 1 to which no water was added ranged from 74-93%. With
the addition of water, this range was narrowed to 87-90%. It
should also be noted that in “anhydrous” samples, the heating
time required for complete conversion of 1 (5.1-9.5 mM) varied
from 5 to 300 h at 120 °C, while the reaction time for 1 at 120
°C in the presence of 10 equiv of water was consistently less
than 18 h. Thus, variable amounts of adventitious water in the
“anhydrous” samples may have contributed to the lack of
uniformity in reaction time and product yields.
Thermolysis of fac-(dppbz)PtMe3(OCD3) (2-d3). In order
to demonstrate that the source of hydride in 8 was the methoxide
ligand in 2, the partially deuterated complex fac-(dppbz)-
PtMe3(OCD3)
(2-d3)
was
prepared
from
fac-
(dppbz)PtMe3(O2CCH3) and KOCD3. Upon thermolysis of 2-d3,
CH3D and ethane were observed. No other organic products
were observed in significant yield by H NMR spectroscopy.
The observation of CH3D is consistent with the formation of
1
fac-(dppbz)PtMe3D (8-d1), followed by C-D reductive elimination.
Reaction of 2 with [18-crown-6-K][OCD3]. When an excess
of [18-crown-6-K][OCD3] was added to a thermolysis reaction
of 2 in C6D6, slow exchange of OCD3 into the methoxide
position of 2 was observed via 1H NMR spectroscopy in
competition with formation of the Pt(II) product (dppbz)PtMe2
(3). After 2 h at 100 °C, ca. 10% of the methoxide complex 2
had been converted to 2-d3, while approximately 10% of the
initial quantity of 2 had been converted to (dppbz)PtMe2 (3).
A maximum of 7% yield of the methoxide complex 2 was
detected at intermediate reaction times during the thermolyses
of 1 with added water, less than was observed in the anhydrous
thermolyses (12%). Methane was still detected (3-6%), but
methyl formate and the three other signals ascribed to minor
(36) The presence of excess water in the thermolyses of 1 also slightly
increased the maximum observed yield of the unidentified transient Pt(IV)
species (6) from 2% to 5%. Notably, it was shown in independent
experiments that the addition of water to the Pt(IV) hydroxide complex 1
(5.7mM in C6D6) also resulted in the generation of species 6 at ambient
temperature (80% conversion over 4 weeks).31 However, pure samples of
6 could not be isolated, which inhibited further characterization. The very
slow formation of 6 (weeks at ambient temperature) argues against
characterization of 6 as a simple hydrogen-bonded species 1 · H2O. A sample
of 6 in THF was analyzed by ESI-MS and shows a signal at m/z ) 1478
with an isotopic pattern indicative of two platinum atoms. The molar mass
of a dimeric complex with the form of [(dppbz)PtMe3(OH) · H2O]2 would
be 1443 g/mol.
1
(34) The coupling patterns of the H NMR spectra of these complexes
are similar to those observed for other fac- (dppbz)PtMe3X complexes. The
coupling constants (1JPtP ) 1140-1160 Hz) associated with the phosphorus
signals of these complexes in the 31P{1H} NMR spectra of these complexes
are comparable to other Pt(IV) (1080-1200 Hz), rather than Pt(II) (1700-
1800 Hz), dppbz complexes, in which the phosphines are trans to methyl
groups.
(35) (a) Litz, K. E.; Banaszak Holl, M. M.; Kampf, J. W.; Carpenter,
G. B. Inorg. Chem. 1998, 37, 6461. (b) For a similar reaction at Ni(II) see
ref 8b.
(37) In addition, two other trace organic products (4.68 ppm, 0-1%;
5.25 ppm, 3-4%)30 were observed.

282 Organometallics, Vol. 28, No. 1, 2009
Smythe et al.
complex.27 In that investigation, fac-(dppbz)PtMe3H, 8, and the
related complex fac-(dppe)PtMe3H were proposed to directly
eliminate methane from the six-coordinate geometry because
of the very similar rates and activation parameters displayed
by the two reactions, in spite of the difference in rigidity between
the dppe and dppbz ligands.56
Discussion
Reductive Elimination from Pt(IV). A variety of Pt(IV)
model complexes have been observed to exhibit sp3 C-Y (Y
) hydrogen, carbon, pnictogen, chalcogen, halogen) reductive
elimination.15,38-54 Many of these reactions have been studied
in detail and their mechanisms elucidated. Reductive elimina-
tions from Pt(IV) complexes can be categorized into two basic
mechanisms: one involving concerted bond formation and one
proceeding via nucleophilic attack. These two mechanism types
generally, but not exclusively, correspond to C-C/C-H and
C-heteroatom reductive elimination, respectively.
The concerted reductive elimination of a hydrocarbon from
isolated octahedral Pt(IV) complexes via C-H or C-C reduc-
tive elimination has been shown to generally require preliminary
ligand loss to form a five-coordinate intermediate.27 Reductive
elimination of the hydrocarbon then takes place in a concerted
fashion through a three-centered (C-M-H or C-M-C) transition
state.54,55 Initial ligand loss from the six-coordinate compound
can occur via ancillary neutral ligand dissociation,51,54 dech-
elation of one “arm” of a polydentate ligand,27,39 or dissociation
of an anionic X- ligand to form a cationic five-coordinate
species.15,38,41 Recently, neutral five-coordinate Pt(IV) alkyl
species have been isolated.40 These five-coordinate complexes
undergo thermally promoted C-C reductive elimination without
preliminary ancillary ligand dissociation,40 consistent with the
idea that the formation of a five-coordinate species allows for
direct C-C coupling.
Bercaw and co-workers have demonstrated that in their Pt(IV)
system sp3 C-O bond formation proceeds via an SN2 reaction,
in which inversion of stereochemistry at carbon was observed.17b
Similarly, Groves and Sanford have reported a Rh(III) system
that undergoes sp3 C-O reductive elimination reactions in which
both SN2 hallmarks of inversion of stereochemistry and higher
reaction rates for primary than for secondary alkyl groups were
observed.13 Vedernikov et al. have also reported reductive
elimination reactions to form sp3 C-O bonds in aqueous
solution which seem to proceed via nucleophilic attack.14a,b That
the nucleophilic attack mechanism has been documented with
a variety of metals and heteroatom groups speaks to its
generality. However, a couple of recent examples indicate that
C-heteroatom coupling from high valent d6 late metals does
not occur exclusively by this nucleophilic attack mechanism.
Concerted reductive elimination of methyl iodide from a Rh(III)
complex has recently been proposed by Milstein.57 Notably, this
elimination occurs from a five- rather than a six-coordinate
species, perhaps indicating that the intimate mechanism of
coupling bears some resemblance to that of C-C reductive
elimination reactions from high valent d6 metal centers.
Furthermore, Vedernikov has recently reported a concerted
C(sp3)-O reductive elimination from Pt(IV) to form epoxides.14c
Thus, evidently both concerted and nucleophilic attack mech-
anisms are possible for C(sp3)-heteroatom bond formation from
high valent d6 metals. Overall, there are relatively few examples
of C(sp3)-heteroatom coupling reactions from d6 metal centers
and the nucleophilic attack pathway appears to be the more
common mechanism of reaction.
There is one study in the literature that reports evidence for
a direct C-H reductive elimination from an octahedral Pt(IV)
(38) Goldberg, K. I.; Yan, J.; Breitung, E. M. J. Am. Chem. Soc. 1995,
117, 6889.
(39) Arthur, K. L.; Wang, Q. L.; Bregel, D. M.; Smythe, N. A.; O’Neill,
B. A.; Goldberg, K. I.; Moloy, K. G. Organometallics 2005, 24, 4624.
(40) (a) Fekl, U.; Goldberg, K. I. J. Am. Chem. Soc. 2002, 124, 6804.
(b) Fekl, U.; Kaminsky, W.; Goldberg, K. I. J. Am. Chem. Soc. 2003, 125,
15286. (c) Kloek, S. M.; Goldberg, K. I. J. Am. Chem. Soc. 2007, 129,
3460. (d) Luedtke, A. T.; Goldberg, K. I. Inorg. Chem. 2007, 46, 8496.
(41) Pawlikowski, A. V.; Getty, A. D.; Goldberg, K. I. J. Am. Chem.
Soc. 2007, 129, 10382.
(42) Madison, B. L.; Thyme, S. B.; Keene, S.; Williams, B. S. J. Am.
Chem. Soc. 2007, 129, 9538.
(43) Canty, A. J.; Rodemann, T.; Skelton, B. W.; White, A. H.
Organometallics 2006, 25, 3996.
(44) Wik, B. J.; Ivanovic-Burmazovic, I.; Tilset, M.; Van Eldik, R. Inorg.
Chem. 2006, 459, 3613.
(45) Procelewska, J.; Zahl, A.; Liehr, G.; Van Eldik, R.; Smythe, N. A.;
Williams, B. S.; Goldberg, K. I. Inorg. Chem. 2005, 44, 7732.
(46) Jenkins, H. A.; Klempner, M. J.; Prokopchuk, E. M.; Puddephatt,
R. J. Inorg. Chim. Acta 2003, 352, 72.
(47) Jensen, M. P.; Wick, D. D.; Reinartz, S.; White, P. S.; Templeton,
J. L.; Goldberg, K. I. J. Am. Chem. Soc. 2003, 125, 8614.
(48) Vedernikov, A. N.; Caulton, K. G. Angew. Chem., Int. Ed. 2002,
41, 4102.
(49) Reinartz, S.; White, P. S.; Brookhart, M.; Templeton, J. L.
Organometallics 2000, 19, 3854.
(50) Fekl, U.; Zahl, A.; van Eldik, R. Organometallics 1999, 18, 4256.
(51) (a) Brown, M. P.; Puddephatt, R. J.; Upton, C. E. E. J. Chem. Soc.,
Dalton Trans. 1974, 2457. (b) Roy, S.; Puddephatt, R. J.; Scott, J. D.
J. Chem. Soc., Dalton Trans. 1989, 2121. (c) Hill, G. S.; Redina, L. M.;
Puddephatt, R. J. Organometallics 1995, 14, 4966. (d) Jenkins, H. A.; Yap,
G. P. A.; Puddephatt, R. J. Organometallics 1997, 16, 1946. (e) Hill, G. S.;
Puddephatt, R. J. Organometallics 1997, 16, 4522. (f) Hill, G. S.; Yap,
G. P. A.; Puddephatt, R. J. Organometallics 1999, 18, 1408.
(52) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1995,
117, 9371.
Thermal reactions of a variety of fac-(L2)PtMe3X complexes
with different X groups (L2 ) dppe, dppbz; X ) I, O2CR, OAr,
N(R)SO2Ar) have been found to yield alkyl C-heteroatom
reductive elimination products.15,38,41 Evidence has been pre-
sented indicating that all of these C-heteroatom couplings occur
by a common mechanism as shown in Scheme 5. Following
dissociation of X- and formation of a five-coordinate Pt(IV)
cation, nucleophilic attack of the X- group on the axial methyl
group of the Pt(IV) cation forms MeX and (L2)PtMe2. Notably,
as a five-coordinate Pt(IV) intermediate bearing multiple methyl
groups is formed in these systems, C-C reductive elimination
to form ethane is also viable and observed.
Mechanism of Reductive Elimination from fac-(dppbz)-
PtMe3(OH) (1). The Pt(IV) hydroxide complex 1 undergoes
C-O reductive elimination in competition with C-C reductive
elimination, behavior that is remarkably similar to that observed
for the aryloxide and carboxylate analogues fac-
(dppbz)PtMe3(OR) (OR ) p-XC6H4O, OAc, O2CCF3).15 How-
ever, the organic products observed upon thermolysis of 1 are
more varied than those observed in the thermolysis reactions
of the related carboxylate and aryloxide compounds, wherein
the products were limited to those resulting directly from
competitive C-O and C-C reductive elimination reactions.15
(53) Appleton, T. G.; Clark, H. C.; Manzer, L. E. J. Organomet. Chem.
1974, 65, 275.
(56) A computational study of models of 8 and the related dppe system
with different substituents on the phosphorous atoms proposed that
phosphine dissociation may occur concurrently with C-H reductive
elimination. Michel, C.; Laio, A.; Mohamed, F.; Krack, M.; Parrinello, M.;
Milet, A. Organometallics 2007, 26, 1241.
(54) Puddephatt, R. J. Coord. Chem. ReV. 2001, 219, 157, and references
therein.
(55) (a) Hill, G. S.; Puddephatt, R. J. Organometallics 1998, 17, 1478.
(b) Bartlett, K. L.; Goldberg, K. I.; Borden, W. T. J. Am. Chem. Soc. 2000,
122, 1456.
(57) Frech, C. M.; Milstein, D. J. Am. Chem. Soc. 2006, 128, 12434.

ꢀ-Hydride Abstraction from Pt(IV) Complexes
Organometallics, Vol. 28, No. 1, 2009 283
Scheme 5
The presence of additional products indicates that other reactions
occur concurrently with and/or subsequently to C-O and C-C
reductive elimination. Notably, other Pt(IV) species were
observed at intermediate times during the reaction. However,
the only Pt products observed at the end of the reaction were
(dppbz)PtMe2 (3) and (dppbz)PtMe(OH) (5).
determining. If the first step were rate-limiting, hydrogen
bonding would directly increase the rate of the reaction by
increasing the rate of this dissociation step. The hydrogen
bonding interaction would also be expected to decrease the
nucleophilicity of the resulting anion and thus inhibit the second
step of the reaction (C-X bond formation). However, an overall
acceleration in the C-X reductive elimination could be explained
on the basis of a larger influence of the hydrogen bonding on
the dissociation step as was observed in the C-O reductive
elimination reactions of the aryloxide analogs of 1 and 2.15
Although this C(sp3)-OH coupling from 1 may at first seem
to be a logical extension of the mechanisms observed in these
previous cases of C-O reductive elimination from Pt(IV),15 it
should be noted that dissociation of hydroxide in a nonpolar
organic solvent such as benzene is expected to be significantly
more unfavorable than dissociation of acetate or aryloxide.
Notably, the only other example of directly observed C-O
reductive elimination to produce an alcohol is a reaction in
aqueous solution wherein hydrogen bonding to the water
molecules would be expected to stabilize the hydroxide anion.14,59
The reaction of methanol with 1 to produce 2 (Scheme 1)
explains the observation of 2 (up to 12%) at intermediate times
in the thermolysis of 1 in C6D6. On the basis of the equilibrium
reaction shown in Scheme 1, it was anticipated that the
formation of 2 could be suppressed by the addition of water to
the thermolyses of 1. Indeed, thermolysis of 1 in the presence
of a large amount of water (ca. 10 equiv) led to a smaller amount
of 2 (7%) observed during the reaction and higher yields of
C-O reductive elimination products (dppbz)PtMe2 (3) (from
74-93% to 87-90%) and methanol (from 34-68% to 51-72%).
If the C-O and C-C reductive eliminations from complex
1 proceed through a mechanism of initial dissociation of the
hydroxide to form a five-coordinate ionic intermediate analo-
gously to the aryloxide and carboxylate complexes fac-
(dppbz)PtMe3(OR) (Scheme 5, X ) OR), then the addition of
hydroxide would be expected to inhibit C-C coupling. Indeed,
thermolysis of 1 in the presence of CsOH · H2O shows a
reduction in the yield of the C-C coupled products, 5 (3-7%)
and ethane (3-7%) and produces a larger yield of C-O
coupling products, 3 (93-97%) and methanol (72-76%),
relative to thermolyses of 1 in the absence of hydroxide. Similar
results were found when the thermolyses of fac-
(dppbz)PtMe3(OAc) and fac-(dppbz)PtMe3(OAr) were carried
out in the presence of added acetate and added aryloxides,
respectively; C-C reductive elimination was inhibited and high
yields of C-O coupled products were observed.15
The mechanism shown in Scheme 5 with X ) OH is
consistent with other observations as well. In the presence of
water, the conjugate acid of hydroxide, both the C-O and C-C
coupling reactions from 1 clearly proceeded faster, being
complete in less than 18 h, compared to up to 300 h without
added water. Similar behavior was observed in thermolysis of
the analogous Pt(IV) carboxylate, aryloxide, and sulfonamide
complexes, fac-(dppbz)PtMe3X (X ) O2CR, OAr, NHSO2Ar),
with addition of the conjugate acids of the X- group, HO2CR,
HOAr and NH2SO2Ar, respectively.15,41,58 Addition of the acid
accelerates the formation of the five-coordinate intermediate
cation by assisting initial X- dissociation through hydrogen
bonding. Note that acceleration could be observed regardless
of which mechanistic step in the two-step mechanism is rate-
While there seems to be a common mechanism for the
reductive elimination reactions of 1 and those of the related
carboxylate and aryloxide species, the C-O coupling reaction
from 1 adds a new level of complexity, as the product, methanol,
is not innocent in the reaction. Like water, methanol accelerates
reductive elimination from 1 by assisting the dissociation of
hydroxide through hydrogen bonding. As a consequence, the
formation of the five-coordinate cationic intermediate is facili-
tated and both C-O and C-C coupling pathways accelerate as
the reaction progresses.15 This autoaccelerative effect was also
observed in the thermolysis of fac-(dppbz)PtMe3(NHSO2Ar)
species in which the product MeNHSO2Ar can hydrogen bond
and assist dissociation of the NHSO2Ar- anion from the starting
material.41 The fact that the methanol product reacts with the
platinum hydroxide 1 to generate water and the platinum
methoxide complex 2 adds further complications, as the reaction
pathways and products resulting from thermolysis of 2 become
available (vide infra).
Reductive Elimination in the Thermolysis of fac-(dppbz)-
PtMe3(OMe) (2). Contrary to our initial expectations based on
the observed reactivity of 1, the organic products of C-O
(dimethyl ether) and C-C (ethane) reductive elimination
accounted for only a small portion of the products of thermolysis
of 2 (1% and 3-7%, respectively, Table 3, vide supra). While
(59) Note that other examples (see ref 17b) are known in which water
acts as the nucleophile. There is one other claim in the literature of direct
C-O reductive elimination involving hydroxide,12 but later experiments
by Hartwig and Mann on a related system5a have suggested that the reaction
proceeds via nucleophilic attack of the iridium hydroxide group on methyl
iodide and not a reductive elimination step.
(58) Acceleration of an O-H reductive elimination from
a
Pd(II)(O2CR)(H) complex by protic additives was recently reported by Stahl
and co-workers. Konnick, M. M.; Stahl, S. S. J. Am. Chem. Soc. 2008,
130, 5753.

284 Organometallics, Vol. 28, No. 1, 2009
Smythe et al.
Scheme 6
Scheme 7
Scheme 8
the small contribution from these processes prevented detailed
mechanistic study of the reaction, these products are presumably
generated by a mechanism similar to the one discussed for
reductive elimination from 1, with the substitution of methoxide
for hydroxide as the nucleophile (Scheme 5, X ) OMe).
Formaldehyde was detected only in trace amounts in the
thermolyses of 2.61 However, fac-(dppbz)PtMe3(OCH2OCH3),
4 was observed at intermediate reaction times during the
thermolysis of 2 (maximum concentration of 7% of total Pt). It
was shown in an independent experiment that p-formaldehyde
reacts with 2 at room temperature to form 4, suggesting that
formaldehyde formed in the thermolysis of 2 could insert into
the Pt-O bond of 2, thereby reducing the concentration of
formaldehyde. This product of formaldehyde trapping, 4, can
also undergo formal ꢀ-hydride elimination to generate fac-
(dppbz)PtMe3H (8) and methyl formate (Scheme 7). In contrast
to formaldehyde, methyl formate does not autopolymerize and
Were C-O reductive elimination to occur from fac-
(dppbz)PtMe3(OCH2OCH3) (4, observed as an intermediate at
concentrations of up to 7% of total Pt), it would yield
(dppbz)PtMe2 (3) and dimethoxymethane, the latter of which
60
1
was observed by H NMR spectroscopy (ca. 1%). The lack
of significant quantities of products resulting from C-O
reductive elimination (dimethyl ether and dimethoxymethane)
in the thermolysis of 2 points toward the existence of one or
more lower energy pathways that lead to the other products
observed in the reaction.
1
was quantified by H NMR (32-44% yield).
A third organic product observed in the thermolysis of 2 is
dimethyl carbonate (21-33% yield). In a process similar to the
formation of methyl formate, the formation of dimethyl carbon-
ate could involve the insertion of methyl formate into the Pt-O
bond of 2 to form a new Pt(IV) species with a dimethoxymethox-
ide ligand. Subsequent ꢀ-hydride elimination would lead to 8
and dimethyl carbonate as illustrated in Scheme 8.
Formal ꢀ-Hydride Elimination from 2. In the thermolysis
of 2, products expected from formal ꢀ-hydride reactions, i.e.,
formaldehyde (and its further reaction products), methyl formate,
dimethyl carbonate, and methane, were far more plentiful than
those generated by C-C or C-O reductive elimination from
2. The initial products expected for ꢀ-hydride elimination from
2 would be formaldehyde and fac-(dppbz)PtMe3H (8) (Scheme
6).
The Pt(IV) dimethoxymethoxide intermediate in Scheme 8
was not detected in the thermolysis of 2. However, it should be
noted that attempts to synthesize Pt(IV) species analogous to 2
with larger alkoxide ligands (e.g., ethoxide, isopropoxide) were
unsuccessful, and only 3 and the organic products from formal
ꢀ-hydride elimination were observed.62
Finally, methanol was detected in 19-21% yield in the
thermolysis of 2. It is possible that methanol was formed by
the dehydrogenation of formaldehyde by the methoxide complex
(2) to form CO, methanol and the hydride complex (8). The
Thermal studies of 8 have been reported in which 8 was
observed to undergo direct C-H reductive elimination to form
methane and (dppbz)PtMe2 (3) at 25 °C.27 Therefore, species 8
would not be stable at the 100 °C temperature of the thermoly-
ses. Consistent with this analysis, along with a high yield of 3,
a significant amount of methane (12-20% in solution) was
1
observed by H NMR spectroscopy after the thermolysis of 2
in C6D6. When the thermolysis of 2-d3 was carried out under
identical conditions, CH3D was the only isotopomer of methane
observed, indicating that the source of the hydride in 8 (and
subsequently in methane) was the methoxy group of 2. The
actual amount of methane produced probably exceeded that
measured by integration of the H NMR signal (12-20%), as
some of the methane would enter the headspace of the NMR
tube.
(60) Accurate integration of the signals for dimethoxymethane (δ 3.12
(s, 6H), δ 4.38 (s, 2H)) was not possible due to the very close proximity of
other peaks.
(61) In organic solvents, formaldehyde autopolymerizes, and so large
amounts would not be expected to be observable by 1H NMR spectroscopy.
Walker, J. F. Formaldehyde: Chapman & Hall: New York, 1964.
(62) Smythe, N. A. Reactivity Studies of Platinum(IV) Hydroxide and
Methoxide Complexes and the Study of Pincer Palladium(II) Complexes
as Potential Catalysts for Olefin Epoxidation. Ph.D. Thesis, University of
Washington, Seattle, WA, 2004.
1

ꢀ-Hydride Abstraction from Pt(IV) Complexes
Organometallics, Vol. 28, No. 1, 2009 285
Figure 3. Complexes undergoing alcohol-promoted formal ꢀ-hydride elimination.
Scheme 9
Pt(IV) hydride 8 would then undergo C-H reductive elimination
to form 3 and methane. Such reactivity has previously been
reported from Pt(II) methoxide complexes.63 Methanol could
also be generated from reaction of 2 and adventitious water as
shown in Scheme 1 (vide supra).
Ozerov observed that ꢀ-hydride elimination from a Pd(II)
ethoxide species was dramatically accelerated in the presence
of ethanol and similarly proposed alcohol-assisted dissociation
of the alkoxide.23
In light of these examples, we propose that a dissociatiVe
ꢀ-hydride abstraction mechanism may be an important and
general pathway in formal ꢀ-hydride elimination reactions from
late metal alkoxides. In such a process, initial dissociation, often
promoted by hydrogen bonding from a Brønsted acid, precedes
the abstraction by the metal of a ꢀ-hydride, which then resides
in the coordination site vacated by the alkoxide. This second
step of the proposed mechanism is very similar to the Canizzarro
reaction in organic chemistry.65 Scheme 9 depicts this mecha-
nism for the conversion of methoxide 2 to hydride 8 and
formaldehyde. While the previous examples from Ir and Pd
involve alkoxide dissociation assisted by the alcohol, it is also
possible that dissociative reactions could proceed without
assistance, albeit under more forcing conditions, as was observed
for the thermolysis of 2 in C6D6. In addition, this pathway should
also be available for other anionic ligands that both are capable
of dissociation and bear ꢀ-hydrogens. Indeed, this was proposed
An Alcohol-Promoted Dissociative ꢀ-Hydride Abstraction
Mechanism. The prevalence of products expected from a formal
ꢀ-hydride elimination in the thermolysis of 2 is surprising for
a coordinatively saturated, 18-electron complex. The accepted
mechanism of ꢀ-hydride elimination, well-established for alkyl
groups1 and generally thought to be operative in similar reactions
of alkoxides and amides,64 requires an open site on the metal
center cis to the ꢀ-hydrogen-bearing ligand. Complex 2 has no
ligand that would easily dissociate to generate an open site cis
to the methoxide. While dissociation of one arm of the
bisphosphine chelate could generate such an open site, earlier
studies of reductive elimination from a related complex,
(dppbz)PtMe4 suggest that phosphine dissociation is not kineti-
cally accessible at 100 °C.27 The fact that addition of methanol
to a C6D6 solution of 2 rapidly generates the hydride complex
8 at room temperature, while the thermal reaction of 2 in C6D6
occurs over the course of hours to days at 100 °C, indicates
that methanol dramatically accelerates this formal ꢀ-hydride
elimination process. While this acceleration of ꢀ-hydride
elimination in the presence of alcohol may seem unusual, we
suspect that it is more widespread than is currently appreciated.
Two similar examples of acceleration of formal ꢀ-hydride
elimination of an alkoxide in the presence of alcohol at iridium
as
a significant decomposition route of fac-(dppbz)Pt-
Me3(N(Me)SO2Ar), in which products consistent with ꢀ-hydride
elimination were observed.41
Thus, the dissociative ꢀ-hydride abstraction mechanism either
with methanol assistance as shown in Scheme 9 or without
methanol assistance under alcohol-free conditions is proposed
as a primary reaction pathway for 2. The results of our previous
studies of the effect of conjugate acids, HX, on reaction
pathways that proceed via preliminary dissociation of X-
(reductive elimination from fac-L2PtMe3X complexes (L2 )
dppe, dppbz, X ) OH, O2CR, OAr, N(H)SO2Ar))15,41 support
the role of methanol in assisting methoxide dissociation from
2. Notably, all of those reactions also proceed, albeit more
slowly, in the absence of HX by a similar dissociative pathway.
This is also expected to be in the case for dissociative ꢀ-hydride
abstraction from 2, which requires forcing thermal conditions
in the absence of methanol. That the thermolysis of 2 at 100
°C in the presence of [18-crown-6-K][OCD3] results in forma-
have been reported,20,21 as well as one at palladium.23
A
common feature of 2 and both iridium complexes is that the
alkoxides from which the formal ꢀ-hydride elimination occurs
are saturated, 18-electron d6 metal compounds (Figure 3).
For the iridium chloro complex (Figure 3, far left), the
reaction to form the analogous hydride complex was substan-
tially accelerated by the presence of methanol. The authors also
reported that added P(CD3)3 did not incorporate into the complex
on the time scale of the reaction. Presumably to accommodate
the traditional mechanism of ꢀ-hydride elimination previously
established in alkyl systems, the authors proposed that the role
of methanol was to assist in the dissociation of the chloride to
generate an open site cis to the methoxide ligand.20 For the
phenyl complex (Figure 3, center left), the lack of dissociable
ligands and the lack of P(CD3)3 incorporation caused Blum and
Milstein to argue that methanol was assisting the dissociation
of the methoxide ligand itself.21 More recently, Fafard and
(63) Bryndza, H. E.; Calabrese, J. C.; Marsi, M.; Roe, D. C.; Tam, W.;
Bercaw, J. E. J. Am. Chem. Soc. 1986, 108, 4805.
(64) Zhao, J.; Hesslink, H.; Hartwig, J. F. J. Am. Chem. Soc. 2001, 7220.
(65) March, J. AdVanced Organic Chemistry, 4th ed., John Wiley &
Sons: New York, 1992; pp 1233-1235.

286 Organometallics, Vol. 28, No. 1, 2009
Smythe et al.
tion of 2-d3 on the same time scale as the ꢀ-hydrogen abstraction
reaction to form 3 is consistent with the dissociation of
methoxide being kinetically competent as a preliminary reaction
step.
tion mechanism proceeding via hydroxide dissociation followed
by nucleophilic attack on the Pt(IV) methyl group. Thus, the
mechanism for C-O coupling is essentially identical to that
demonstrated in other C-X couplings from similar fac-
L2Pt(IV)Me3X (X ) I, O2CR, OAr, N(R)SO2Ar) complexes in
which the X anions are significantly better leaving group
candidates.15,38,41
Formation of methanol as a product of the thermolysis of 1
has two important effects on the reaction. First, the hydrogen-
bonding ability of methanol assists in dissociation of the
hydroxide ligand, increasing the rate of both C-O and C-C
elimination. Second, methanol can exchange with 1 to form the
analogous methoxide, fac-(dppbz)PtMe3(OMe) (2). This situ-
ation is similar to the behavior noted in reductive elimination
to form sulfonamides from fac-(dppbz)PtMe3(NHSO2Ar),41 in
which the hydrogen-bonding capability of HN(Me)SO2Ar
resulted in both sulfonamide anion exchange and acceleration
of elimination.
In contrast to the reactivity observed for 1, C-O reductive
elimination is a very minor reaction in the thermolysis of the
Pt(IV) methoxide complex 2. Instead, products resulting from
formal ꢀ-hydride elimination are observed. Again, this behavior
is similar to that observed for the sulfonamide complexes, in
which products of formal ꢀ-hydrogen elimination resulted from
the decomposition of fac-(dppbz)PtMe3(N(Me)SO2Ar) species.41
Both of these systems lack ancillary ligands capable of facile
dissociation. Thus, the standard textbook mechanism established
for ꢀ-hydride elimination involving a concerted four-centered
transition state seems implausible. Instead, we propose a
dissociative ꢀ-hydride abstraction mechanism, in which dis-
sociation of the methoxide or sulfonamide group is followed
by hydride abstraction by the metal from the dissociated anion.
The lack of appreciable amounts of C-O reductive elimination
products in the thermolysis of 2 indicates that ꢀ-hydride
abstraction from the dissociated methoxide is a lower energy
pathway than nucleophilic attack by the methoxide on the methyl
group. Thus, ꢀ-hydride abstraction may be a complication in
reactions for which C-X reductive elimination reactions are
desired if the X group bears ꢀ-hydrogens. As similar observa-
tions of alcohol-promoted formal ꢀ-hydride elimination of
alkoxides with no accessible open cis coordination site have
been reported in the past several years, it seems that this
alternative mechanism of dissociative ꢀ-hydride abstraction may
be quite general. Thus, just as multiple mechanisms are possible
for reductive elimination reactions (e.g., “concerted” and two-
step “dissociation and nucleophilic attack”), depending on the
nature of the metal-bound atoms that are forming the bond, more
than one common mechanism also appears viable for ꢀ-hydride
elimination reactions. Depending on the type of metal-bound
atom and the availability of a cis open site, the reaction can
proceed via a traditional ꢀ-hydride elimination involving transfer
of a hydride to a cis open site on the metal, as is seen with
metal alkyls, or via a “dissociative ꢀ-hydride abstraction”
wherein heterolytic metal-ligand cleavage is followed by
abstraction of hydride from the dissociated ligand. Thus, our
current long-held mechanistic understanding of this fundamental
reaction, originally proposed and well-established for metal alkyl
complexes, requires amendment when alkoxide or amide ligands
are involved in which case a dissociative ꢀ-hydride abstraction
pathway should also be considered.
One other possible mechanism of formal ꢀ-hydride elimina-
tion from 2 involves dissociation of methoxide from 2 to form
a cationic intermediate, which then can abstract a ꢀ-hydride from
a second molecule of 2, thereby generating 8, formaldehyde,
and the cation again. Ritter and Bergman have proposed a
similar type of behavior in ꢀ-hydride elimination from
Cp*(PMe3)Ir(OCH2R)Ph, in which the presence of iridium
cations was observed to catalyze the formation of the corre-
sponding iridium hydrides and aldehydes.22 It is difficult to fully
investigate a similar situation in this Pt(IV) system as the five-
coordinate Pt(IV) cation is not stable. In addition, the rate of
ꢀ-hydride elimination from 2 is irreproducible, likely due to
adventitious water or a variable amount of another acid. Finally,
it would be difficult to measure the rate dependence on the
concentration of 2 since it is only sparingly soluble under the
reaction conditions. Thus, a bimolecular ꢀ-hydride elimination
mechanism cannot be ruled out for 2, although it seems unlikely
given the steric bulk of 2.
Critical in the above discussion is that the Pt(IV) example
described herein and the reported Ir(III) and Pd(II) examples
all seem to involve an abstraction of an alkoxide ꢀ-hydride rather
than a traditional ꢀ-hydride elimination to a cis open site. Since
the mechanism involves preliminary dissociation of the alkoxide,
the presence of alcohol accelerates the dissociation and there-
fore the overall reaction. Ozerov’s results with a 16-electron
Pd(II) ethoxide complex indicate that this alcohol-assisted
pathway can even operate under conditions when an open cis
site is nominally available.23 Notably, a classical method for
preparing metal-hydrides is the addition of base and excess
alcohol to a metal halide complex. The presence of alcohol may
be key to accelerating a similar dissociative ꢀ-hydride abstrac-
tion in these cases. In addition, the observation of an analogous
type of dissociative ꢀ-hydride abstraction in the thermal
reactions of sulfonamide complexes fac-(dppbz)Pt-
Me3(N(Me)SO2Ar)41 suggests that this reaction pathway is
important for other heteroatomic groups as well. Indeed, it is
likely that the mechanism depicted in Scheme 9 for 2 is
representative of an underappreciated mechanism of formal
ꢀ-hydride elimination available to both coordinatively saturated
and unsaturated late metal complexes with ꢀ-hydrogen laden
heteroatom ligands. Thus, unlike their metal alkyl counterparts,
for metal alkoxides and metal amides, there are at least two
viable modes for ꢀ-hydride elimination. In addition to the
traditional route of transfer of the ꢀ-hydride to a cis open site,
the ꢀ-hydride can be abstracted by an open site generated via
dissociation of the heteroatom group itself.
Summary and Conclusions
Two fundamental organometallic reactions, reductive elimi-
nation and dissociative ꢀ-hydride abstraction, were observed
as the primary reactions resulting from thermolysis of fac-
(dppbz)PtMe3(OH) (1) and fac-(dppbz)PtMe3(OMe) (2), re-
spectively. Thermolysis of 1 produces (dppbz)PtMe2 (3) and
methanol in good yield. The C-O reductive elimination reaction
can be optimized by adding water and hydroxide to the
thermolysis, resulting in high yields (72-76%) of methanol.
This is a rare example of a directly observed alkyl C-O
reductive elimination to yield an alcohol. Despite the expected
poor leaving group ability of hydroxide in a nonpolar solvent,
the mechanistic evidence implicates a C-O reductive elimina-
Experimental Section
General Considerations. Unless otherwise noted, all transfor-
mations were carried out under a N2 atmosphere in a drybox (O2

ꢀ-Hydride Abstraction from Pt(IV) Complexes
Organometallics, Vol. 28, No. 1, 2009 287
< 1 ppm, H2O < 0.5 ppm) or using standard Schlenk techniques.
THF, benzene, and toluene were distilled under vacuum from
sodium/benzophenone ketyl. Pentane was distilled from CaH2.
Methanol was triply distilled from sodium under vacuum. Deuter-
ated solvents were dried by analogous methods and transferred
under vacuum. KH was purchased as a suspension in mineral oil,
isolated on a fritted glass funnel, washed multiple times with
hexanes and pentane, dried under vacuum, and kept in the drybox.
o-Bis(diphenylphosphino)benzene (dppbz) was purchased from
Strem, treated with KH in THF, and recrystallized from THF/
pentane. Other reagents, unless specified, were used as received
from commercial suppliers. The complexes fac-(dppbz)Pt-
Me3(O2CCH3) and fac-(dppbz)PtMe3(O2CCF3) were prepared by
established literature procedures.15 [18-crown-6-K][OCD3] was
synthesized according to an established literature procedure for the
protio analog.66 KOH was produced by the reaction of potassium
metal with water (2 equiv) in toluene and left under vacuum for 3
days after removal of the toluene. Celite was dried for >48 h at
>150 °C under vacuum. Thermolysis samples were heated in a
Neslab Ex-250HT elevated temperature bath. NMR spectra were
recorded on Bruker DPX200, AF300, or DRX499 spectrometers.
mg, 0.028 mmol) was added in a drybox to a Teflon-stoppered glass
vessel and subsequently brought out of the drybox. C6D6 (10 mL)
was added by vacuum transfer, and then 4.4 mL of this solution
was removed in vacuo to concentrate the solution of 1. In the
drybox, toluene was added to the solution as an NMR standard.
For each sample, 0.6 mL (0.003 mmol, 5 mM) of this solution was
loaded into a medium-walled NMR tube (Wilmad 504-PP), which
was subsequently sealed on a vacuum line. In the reactions
involving additional H2O, this reagent (0.035 mmol; 50 µL of 0.069
M H2O in C6D6 solution) was also added to the NMR tube prior to
sealing. For the thermolysis of 1 in the presence of cesium
hydroxide, 1 (1.7 mg, 0.0024 mmol), CsOH · H2O (4.2 mg, 0.025
mmol), toluene as a NMR standard, and C6D6 (0.5 mL) were loaded
into a medium-walled NMR tube (Wilmad 504-PP) which was
subsequently sealed on a vacuum line. For thermolysis of 2 or 2-d3,
0.5 mL of a solution of 2 or 2-d3 (0.003 mM) and toluene as a
NMR standard in C6D6 were loaded into a medium-walled NMR
tube (Wilmad 504-PP) that had been fitted with a 14/20 ground
glass joint.
In all thermolyses, after loading the NMR tube, the sample was
affixed to a stopcock adaptor, cooled in a liquid N2 bath, and
evacuated under high vacuum. The tube was then flame-sealed
under active vacuum. The sealed NMR tubes were heated, and the
reactions were quenched at various time intervals by rapidly cooling
the tubes in ice water. Thermolysis tubes were stored in a laboratory
freezer (-20 °C) or in liquid N2 when not being heated or monitored
1
All coupling constants are reported in Hz. H NMR spectra were
referenced by using residual solvent peaks and are reported in ppm
downfield of tetramethylsilane. 31P{1H} NMR spectra were refer-
enced to external H3PO4 (0 ppm). GC/MS data was collected on a
Hewlett-Packard 5971 GC-mass spectrometer. Electrospray MS data
was collected on a Bruker Esquire ion trap LC-mass spectrometer.
Preparation of fac-(dppbz)PtMe3(OH) (1). A Teflon-stoppered
glass vessel was charged with fac-(dppbz)PtMe3(O2CCF3) (267 mg,
0.33 mmol), toluene (100 mL), and KOH (387 mg, 6.9 mmol).
The reaction mixture was heated to 35 °C for 2 days. The solution
was cooled, and excess KH was added to the reaction mixture,
which was left to stir for 1 h at ambient temperature. This solution
was filtered through a Teflon syringe filter, and the filtrate was
reduced in volume to 30 mL in vacuo and left to crystallize. White
crystals were harvested, and solvent was reduced under vacuum to
5 mL to yield a second crop of crystals. Combined yield from two
crystallizations ) 223 mg, 95%. 31P{1H} NMR (C6D6): δ 15.3 (s
1
by NMR spectroscopy. All H and 31P{1H} NMR spectra were
collected on a Bruker DRX499 spectrometer, averaging the integrals
for three acquisitions at a given reaction time. For each aquisition,
eight scans were collected with a 60 s recycle delay. Signals for
all species were integrated relative to the internal standard (toluene,
T1 ) 11 s). Mass balance of Pt complexes (>95%) was confirmed
by integration of Pt-CH3 signals (trans to P) against the internal
standard at the end of the reaction.
Reaction of 1 with CH3OH. In a drybox, 1 (5.0 mg, 0.0071
mmol) and an internal NMR standard (toluene, 1 µL, 0.01 mmol)
were loaded into a medium-walled J. Young NMR tube (Wilmad
504-PP). THF-d8 (0.5 mL) was vacuum transferred into the sample.
After initial 1H and 31P{1H} NMR spectra were collected, this
sample was brought back into the drybox, and methanol (0.014
1
1
w/Pt satellites, JPtP ) 1161). H NMR (C6D6): δ -2.39 (t w/Pt
3
2
satellites, 1H, JPH ) 2.7, JPtH ) 13.0, Pt-OH), 0.02 (t w/Pt
satellites, 3H, 3JPH ) 6.6, 2JPtH ) 63.8, Pt-CH3 trans to hydroxide),
1.68 (t w/Pt satellites, 6H, 3JPH ) 7.4, 2JPtH ) 60.0, Pt-CH3 cis to
hydroxide), 6.7-7.4 (m, 20H, aryl protons on dppbz ligand), 8.67
(m, 4H, aryl protons on dppbz ligand). Anal. for C33H34OP2Pt: Calcd
C, 56.33; H, 4.87; Found C, 56.38; H, 4.90.
1
mL, 0.35 mmol, ca. 48 equiv) was added. H and 31P{1H} NMR
spectra were collected within 10 min of adding methanol. The
reaction was left to react at ambient temperature and was monitored
1
for 3 days by H and 31P{1H} NMR spectroscopy.
Reaction of 2 with H2O. In a drybox, 2 (2.4 mg, 0.0033 mmol)
and an internal NMR standard (toluene, 0.5 µL, 0.005 mmol) were
loaded into a medium-walled J. Young NMR tube (Wilmad 504-
PP). C6D6 (0.5 mL) was vacuum-transferred into the sample. After
Preparation of fac-(dppbz)PtMe3(OCH3) (2). Small pieces of
potassium metal (880 mg, 22.5 mmol) and methanol (500 µL, 12.3
mmol) were added to toluene and left to stir for 1 day. The
potassium metal was then removed and fac-(dppbz)PtMe3(O2CCH3)
(130 mg, 0.160 mmol) was added. The solution was stirred for
five days and then filtered through a medium porosity fritted glass
funnel or a syringe equipped with a Teflon filter. The solvent was
removed under vacuum and the product was recrystallized from
toluene layered with pentane. Yield ) 86.2 mg, 75%. 31P{1H} NMR
(C6D6): δ 18.2 (s w/Pt satellites, 1JPtP ) 1159). 1H NMR (C6D6): δ
0.01 (t w/Pt satellites, 3H, 3JPH ) 6.6, 2JPtH ) 64.0, Pt-CH3 trans
to methoxide), 1.76 (t w/Pt satellites, 6H, 3JPH ) 7.7, 2JPtH ) 60.0,
Pt-CH3 cis to methoxide), 4.04 (s w/Pt satellites, 3H, 3JPtH ) 24.4,
Pt-OCH3), 6.7-7.4 (m, 20H, aryl protons on dppbz ligand), 8.44
(m, 4H aryl protons on dppbz ligand). Anal. for C34H36OP2Pt:
Calcd.: C, 56.90; H, 5.06; Found: C, 57.11; H, 5.09. fac-
(dppbz)PtMe3(OCD3) was prepared similarly, except for the
substitution of methanol-d4 for methanol.
1
initial H and 31P{1H} NMR spectra were collected, this sample
was brought back into the drybox, and water (0.035 mL of a 0.185
M solution in C6D6, 0.007 mmol) was added. 1H and 31P{1H} NMR
spectra were collected at least twice per hour for the first 6 h and
then at least every 15 h for the next 75 h at ambient temperature.
1H NMR spectra were integrated against the internal standard.
Reaction of 1 with H2O. In a drybox, 1 (10.0 mg, 0.014 mmol)
and an internal NMR standard (toluene, 3 µL, 0.03 mmol) were
loaded into a 2 mL volumetric flask. The solution was diluted to 2
mL with C6D6; 0.4 mL of this solution was added to a medium-
walled NMR tubes. Water (3.0 µL) was added to the sample, and
the tube was frozen and flame-sealed under active vacuum. The
sample was thawed and left at ambient temperature. The progress
of the reactions was monitored by 1H and 31P{1H} NMR spectros-
copy at various time intervals for at least 4 weeks. 1H NMR spectra
were integrated against the internal standard. Complex 6 was
observed to form over time, with a maximum of 80% yield after 4
weeks. NMR characterization of 6: 31P{1H} NMR (C6D6): δ 15.0
Treatment of Thermolysis Samples. In a typical set of
experiments for thermolysis of 1 without additives, complex 1 (20.0
(66) Kurcok, P.; Jedlinski, Z.; Kowalczuk, M. J. Org. Chem. 1993, 58,
4219.
1
1
(s w/Pt satellites, JPtP ) 1114). H NMR: (C6D6): δ 0.14 (t w/Pt

288 Organometallics, Vol. 28, No. 1, 2009
Smythe et al.
3
2
1
satellites, 3H, JPH ) 7.7, JPtH ) 73.9, Pt-CH3 trans to O), 1.75
monitored by H and 31P{1H} NMR spectroscopy. Signals for 2
disappeared within 20 min at ambient temperature and were
replaced by those of 8. Slow decomposition of 8 was observed
with signals for 3 and methane appearing over time. The analogous
experiment was also conducted with CD3OD, in which 8-d1 was
observed and CH3D and 3 were observed to form over time.
Reaction of 2 with [18-crown-6-K][OCD3]. In a drybox, 2 (2.8
mg, 0.0039 mmol), [18-crown-6-K][OCD3] (32.9 mg, 0.097 mmol),
and an internal NMR standard (hexamethylbenzene, 2.1 mg, 0.013
mmol) were loaded into a medium-walled J. Young NMR tube
(Wilmad 504-PP). The tube was evacuated under high vacuum,
and C6D6 (0.42 mL) was vacuum-transferred into the sample.
Despite the use of the crown ether, [18-crown-6-K][OCD3] did not
completely dissolve and solid [18-crown-6-K][OCD3] was observ-
3
2
(t w/Pt satellites, 6H, JPH ) 6.9, JPtH ) 56.3, Pt-CH3 trans to
P), 6.7-8.8 (m, 24H, aryl protons on dppbz ligand).
Reaction of 2 with p-CH2O. In a drybox, 2 (2.5 mg, 0.0035
mmol), p-CH2O (1.0 mg, 0.033 mmol) and benzyl ether (ca. 0.01
mmol, internal standard) were loaded into a medium-walled NMR
tube (Wilmad 504-PP). C6D6 (0.5 mL) was vacuum-transferred into
the tube. The sample was kept frozen while the tube was flame-
1
sealed under active vacuum. The sample was thawed and H and
31P{1H} NMR spectra were acquired within 15 min at ambient
temperature.
NMR
characterization
of
fac-
(dppbz)PtMe3(OCH2OCH3) (4): 31P{1H} NMR (C6D6): δ 20.1 (s
w/ Pt satellites, JPtP ) 1139). 1H NMR: δ 0.01 (3H, t w/ Pt
1
satellites, 3JPH ) 6.6, 2JPtH ) 68.4, Pt-CH3 trans to O); 1.73 (6H,
3
2
t w/ Pt satellites, JPH ) 7.5, JPtH ) 58.0, Pt-CH3 trans to P);
1
3
able in the reaction mixture. The reaction was monitored by H
3.53 (s, Pt-OCH2OCH3); 5.56 (s w/ Pt satellites, JPtH ) 18.7, Pt-
NMR spectroscopy. The NMR tube was heated at 100 °C for
varying amounts of time, and then the reaction was quenched by
rapidly cooling the tube in ice water (0 °C). A direct comparison
of the integrals for Pt-CH3 signals (trans to O) and Pt-OCH3 for
2 in the 1H NMR spectra was used to determine how much of the
methoxide ligand had exchanged with OCD3. As the reaction
progressed, signals for CH4, CH3D, and [18-crown-6-K][OCH3]
OCH2OCH3); 6.7-8.46 (m, 24H, aryl protons on dppbz ligand).
This procedure was also carried out with a larger excess of
p-CH2O (6.3 mg, 0.21 mmol) relative to 2 (4.5 mg, 0.0063 mmol)
1
in C6D6 (0.8 mL). H and 31P{1H} NMR spectra taken within 15
min at ambient temperature showed complete consumption of 2.
The 1H NMR spectrum was complex and several singlets, including
some with Pt satellites, were observed between 3-5 ppm. These
signals are assumed to correspond to OCH2 units produced by the
iterative insertion of formaldehyde into the Pt-O bond of 4 and
those of subsequent insertion products. The 31P{1H} NMR spectrum
1
(OCH3 δ ) 4.12 in C6D6) were observed in the H NMR spectra.
Acknowledgment. The authors thank the National Science
Foundation (CHE-0137394 and CHE-0719372) for support
and Dr. W. Kaminsky for X-ray structural determinations.
1
displayed signals assigned to 4 (δ 20.13 ppm, JPtP ) 1138 Hz)
and two other species: δ 20.44 ppm, 1JPtP ) 1140 Hz and δ 20.56
1
ppm, JPtP ) 1134 Hz. The approximate percentanges of total Pt
of these species were 30%, 48%, and 22%, respectively, as
Supporting Information Available: CIF files for fac-
(dppbz)PtMe3(OH) (1) and fac-(dppbz)PtMe3(OMe) (2). This
material is available free of charge via the Internet at http://
pubs.acs.org.
estimated by 31P{1H} NMR peak heights.
Reaction of 2 with CH3OH. In a drybox, 2 (2.5 mg, 0.0035
mmol), C6D6 (0.3 mL), methanol (0.3 mL), and an internal NMR
standard (toluene, 1 µL, 0.01 mmol) were loaded into a medium-
walled J. Young NMR tube (Wilmad 504-PP). The reaction was
OM800905Q