Structural and mechanistic investigations of the oxidation of dimethylplatinum(II) complexes by dioxygen

Article abstract of DOI:10.1021/ic0112903

The oxidation of (tmeda)Pt^II(CH₃)₂ (1, tmeda = N,N,N′,N′-tetramethylethylenediamine) to (tmeda)Pt^IV(OH)(OCH₃)(CH₃)₂ (3) by dioxygen in methanol proceeds via a two-step mechanism. The initial reaction between (tmeda)Pt(CH₃)₂ and dioxygen yields a hydroperoxoplatinum(IV) intermediate, (tmeda)Pt(OOH)(OCH₃)(CH₃)₂, (2), which reacts with a second equivalent of (tmeda)Pt(CH₃)₂ to afford the final product 3. Both 2 and 3 have been fully characterized, including X-ray crystallographic structure determinations. The effect of ligand variation on the oxidation of several dimethylplatinum(II) complexes by 2 as well as by dioxygen has been examined.

Full text of DOI:10.1021/ic0112903

Inorg. Chem. 2002, 41, 3608−3619
Structural and Mechanistic Investigations of the Oxidation of
Dimethylplatinum(II) Complexes by Dioxygen
Vsevolod V. Rostovtsev, Lawrence M. Henling, Jay A. Labinger,* and John E. Bercaw*
Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of
Technology, Pasadena, California 91125
Received December 18, 2001
The oxidation of (tmeda)PtII(CH ) (1, tmeda ) N,N,N′,N′-tetramethylethylenediamine) to (tmeda)PtIV(OH)(OCH )-
3 2
3
(CH ) (3) by dioxygen in methanol proceeds via a two-step mechanism. The initial reaction between (tmeda)Pt-
3 2
(CH ) and dioxygen yields a hydroperoxoplatinum(IV) intermediate, (tmeda)Pt(OOH)(OCH3)(CH ) (2), which reacts
3 2
3 2
with a second equivalent of (tmeda)Pt(CH ) to afford the final product 3. Both 2 and 3 have been fully characterized,
3 2
including X-ray crystallographic structure determinations. The effect of ligand variation on the oxidation of several
dimethylplatinum(II) complexes by 2 as well as by dioxygen has been examined.
Introduction
The selective oxidation of alkanes by aqueous platinum
salts was originally reported by Shilov in 1972 (eq 1).1 The
mechanism of this catalytic cycle, a topic of extensive
mechanistic studies in a number of research groups, consists
of three steps as shown in eq 2.2 Alkane activation results
in the formation of a platinum(II) alkyl complex, which is
then oxidized by [PtCl6]2-. The resulting platinum(IV) alkyl
then undergoes an SN2-type displacement to afford the
alcohol (or alkyl chloride) and regenerate the active platinum-
(II) species. The oxidation is typically carried out in water
or aqueous acetic acid at 100-120 °C. A wide range of
hydrocarbons can be converted to the corresponding alcohols
or alkyl chlorides.
One of the main drawbacks of the Shilov system is the
use of stoichiometric platinum(IV). The oxidation step has
been shown to proceed by electron transfer rather than alkyl
transfer, thus potentially allowing the use of common,
inexpensive oxidants instead of hexachloroplatinate.3 How-
ever, the oxidant must not be too strong, or all platinum(II)
will be converted to platinum(IV), destroying the catalytic
species. Ideally one would like the redox potential to lie in
a window such that the alkylplatinum(II) intermediate would
be oxidized while the remaining inorganic platinum(II) salts
would not. Unfortunately it is difficult to use electrochemical
measurements to predict the desired redox potential, because
of the irreversibility of most PtII/PtIV potentials.4 Several
oxidants have been examined, including chlorine,5 peroxy-
disulfate,6 hydrogen peroxide,7 and an electrochemical cell,8
with limited success (low turnovers and rates).
From a practical standpoint, dioxygen is undoubtedly the
most attractive reagent for catalytic alkane oxidation. Shilov
and co-workers reported that combinations of cupric or
(4) Based on irreversible one electron oxidation potentials, dimethyl-
platinum(II) compounds are almost 0.5 V easier to oxidize than
chloromethylplatinum(II) complexes, which in turn are about 0.5 V
easier to oxidize than dichloroplatinum(II) complexes. J. D. Scollard,
J. A. Labinger, J. E. Bercaw, unpublished results.
(5) Horvath, I. T.; Cook, R. A.; Millar, J. M.; Kiss, G. Organometallics
1993, 12, 8.
(6) Basickes, N.; Hogan, T. E.; Sen, A. J. Am. Chem. Soc. 1996, 118,
13111-13112.
(7) Thorn, D. L. Personal communication.
* Authors to whom correspondence should be addressed. E-mail:
bercaw@caltech.edu (J.E.B.); jal@its.caltech.edu (J.A.L.).
(1) Gol’dshleger, N. F.; Es’kova, V. V.; Shilov, A. E.; Shteinman, A. A.
Zh. Fiz. Khim. (Engl. Transl.) 1972, 46, 785.
(2) Shilov A. E.; Shul’pin, G. B. ActiVation and Catalytic Reactions of
Saturated Hydrocartbons in the Presence of Metal Complexes; Kluwer
Academic Publishers: Boston, 2000. Stahl, S. S.; Labinger, J. A.;
Bercaw, J. E. Angew. Chem., Int. Ed. 1998, 37, 2180.
(3) Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J. E.
J. Organomet. Chem. 1995, 504, 75-91.
(8) Freund, M. S.; Labinger, J. A.; Lewis, N. S.; Bercaw, J. E. J. Mol.
Catal. A: Chem. 1994, 87, L11-L15.
3608 Inorganic Chemistry, Vol. 41, No. 14, 2002
10.1021/ic0112903 CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/12/2002

Oxidation of Dimethylplatinum(II) Complexes by Dioxygen
cuprous chloride, quinones, or polyoxometalates with air can
be used instead of platinum(IV) salts.9 In a recent publication,
Sen and co-workers applied the Na2PtCl4/CuCl2/O2 system
to the oxidation of a number of alkylsulfonic acids.10 Again,
turnovers are only modest, and conditions are relatively harsh
(160-200 °C). A much more efficient platinum(II)-catalyzed
oxidation can be achieved under still more stringent condi-
tions (200 °C, oleum), utilizing bipyrimidine as a stabilizing
ligand (eq 3);11 unfortunately the product obtained, methyl
bisulfate, has little direct use, and the cost of its conversion
to a desirable product such as methanol along with that of
recycling sulfuric acid apparently precludes practical ap-
plication.
involves an axial electrophilic attack of the oxidizing agent
on the square-planar platinum(II) complex. In some cases,
precoordination of a fifth ligand prior to the oxidation step
is indicated.14b
There are only a few examples of the oxidation of
platinum(II) complexes with dioxygen. In 1977, Sarneski and
co-workers reported that [(bipy)Pt(κ2-tach)]2+, generated in
situ, was converted to [(bipy)Pt(κ3-tach)(OH2)]2+ in water
under air (eq 5).19
The possibility of achieving a mild and facile O2-oxidation
of Pt(II) to Pt(IV), perhaps by means of auxiliary ligands,
offers an intriguing route toward more efficient catalytic
systems. In 1984, Puddephatt and Monaghan reported that
several dimethylplatinum(II) complexes with diimine ligands,
such as bipyridyl, can be oxidized to dimethylplatinum(IV)
compounds by alcohols and/or water (eq 4).12 Since this
seemed at odds with the proposed last step in the Shilov
cycle, in which water reduces platinum(IV) to platinum(II),
we (and, simultaneously and independently, the group led
by Karen Goldberg) decided to reinvestigate this chemistry,
and determined that oxidation by O2 is involved; a prelimi-
nary report of those findings has been published.13
Six years later, Wieghardt and co-workers described the
aerobic oxidation of [(κ2-tacn)2Pt]2+ to [(κ3-tacn)2Pt]4+ in
water at 90 °C (eq 6).20
Most recently, Puddephatt and co-workers found that (κ2-
tacn)Pt(CH3)2 is oxidized to [(κ3-tacn)Pt(CH3)2(OH)]+ with
dioxygen (eq 7).21 It is noteworthy that all cases involve a
potentially tridentate ligand that initially is κ2-coordinated
to platinum(II).
The oxidation of platinum(II) complexes has been exten-
sively studied.14 Halogens,15 peroxides,16 platinum(IV) com-
plexes,17 and a number of other oxidizing agents18 can be
used in this reaction. The generally accepted mechanism
We report here our further mechanistic and structural
studies on the oxidation of dimethylplatinum(II) complexes
with dioxygen.
(9) Geletii, Y. V.; Shilov, A. E. Kinet. Katal. 1983, 24, 486-489. Shilov,
A. E.; Shteinman, A. A. Coord. Chem. ReV. 1977, 24, 97-143.
(10) Lin, M.; Shen, C.; Garcia-Zayas, E. A.; Sen, A. J. Am. Chem. Soc.
2001, 123, 1000-1001.
(11) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.; Fujii,
H. Science 1998, 280, 560-564.
Results and Discussion
(12) Monaghan, P. K.; Puddephatt, R. J. Organometallics 1984, 3, 444.
(13) Rostovtsev, V. V.; Labinger, J. A.; Bercaw, J. E.; Lasseter, T. L.;
Goldberg, K. I. Organometallics 1998, 17, 4530-4531.
(14) (a) Peloso, A. Coord. Chem. ReV. 1973, 10, 123-181 (b) Gossage,
R. A.; Ryabov, A. D.; Spek, A. L.; Stufkens, D. J.; van Beek, J. A.
M.; van Eldik, R.; van Koten, G. J. Am. Chem. Soc. 1999, 121, 2488-
2497.
(15) Drougge, L.; Elding, L. I. Inorg. Chem. 1985, 24, 2292-2297.
Drougge, L.; Elding, L. I. Inorg. Chem. 1988, 27, 795-798. Elding,
L. I.; Gustafson, L. Inorg. Chim. Acta 1976, 19, 165-171.
(16) Aye, K. T.; Vittal, J. J.; Puddephatt, R. J. J. Chem. Soc., Dalton Trans.
1993, 1835-1839. Dunham, S. O.; Larsen, R. D.; Abbott, E. H. Inorg.
Chem. 1993, 32, 2049-2055.
(17) Rich, R. L.; Taube, H. J. Am. Chem. Soc. 1954, 76, 2608-2611.
Mason, W. R. Coord. Chem. ReV. 1972, 7, 241-255.
(18) Berglund, J.; Voigt, R.; Fronaeus, S.; Elding, L. I. Inorg. Chem. 1994,
33, 3346-3353. Drougge, L.; Elding, L. I. Inorg. Chem. 1987, 26,
1703-1707. Hindmarsh, K.; House, D. A.; van Eldik, R. Inorg. Chim.
Acta 1998, 278, 32-42.
Oxidation of (tmeda)Pt(CH3)2. The dimethylplatinum-
(II) complexes (bipy)Pt(CH3)2 (bipy ) 2,2′-bipyridyl),
(phen)Pt(CH3)2 (phen ) 1,10-phenanthroline), and (tmeda)-
Pt(CH3)2 (1) exhibit no sign of oxidation to Pt(IV) in
methanol-d4 under an argon atmosphere. 1H NMR spectros-
copy reveals only slow deuterium incorporation into Pt-
CH3 groups, followed by the formation of methane isoto-
pomers (CH3D, CH2D2, and CHD3) along with the protonolysis
(19) Sarneski, J. E.; McPhail, A. T.; Onan, K. D.; Erickson, L. E.; Reilley,
C. N. J. Am. Chem. Soc. 1977, 99, 7376-7378.
(20) Wieghardt, K.; Koeppen, M.; Swiridoff, W.; Weiss, J. J. Chem. Soc.,
Dalton Trans. 1983, 1869-1872.
(21) Prokopchuk, E. M.; Jenkins, H. A.; Puddephatt, R. J. Organometallics
1999, 18, 2861-2866.
Inorganic Chemistry, Vol. 41, No. 14, 2002 3609

Rostovtsev et al.
product (N-N)Pt(OCH3)(CH3) (eq 8). This reaction probably
proceeds via dimethylhydridoplatinum(IV) intermediates,
formed by protonation of dimethylplatinum(II) complexes
with methanol, which undergo competing H/D exchange and
reductive elimination of methane, as previously observed in
related systems.22
In contrast, in the presence of dioxygen, (tmeda)Pt(CH3)2,
(phen)Pt(CH3)2, and (bipy)Pt(CH3)2 undergo more rapid
conversions in methanol solution. In the latter two cases,
Figure 1. EPR spectrum of the frozen blue-colored benzene solution
formed after (tmeda)Pt(CH3)2 was exposed to 1 atm of O2.
species is prevented by addition of radical scavengers (BHT,
1
the products exhibit H NMR spectra identical to those
p-methoxyphenol). The blue solutions can also be quenched
reported by Puddephatt and Monaghan for the hydroxo-
(methoxo)platinum(IV) complexes (N-N)Pt(OH)(OCH3)-
(CH3)2 (N-N ) phen, bipy).12 Exposure of a solution of
(tmeda)Pt(CH3)2 in methanol to air leads to the formation
1
with methanol; the resulting H NMR spectrum indicates a
complex mixture of platinum(IV)-containing products.
At high dioxygen pressures (0.6-1 atm) in methanol,
(tmeda)Pt(CH3)2 cleanly reacts within minutes at room
temperature, to form a mixture of 3 and an additional
dimethylplatinum(IV) species, characterized (see below) as
hydroperoxy-methoxy complex (tmeda)Pt(OOH)(OCH3)-
(CH3)2 (2, eq 10). Subsequent conversion of 2 to 3 occurs
more slowly, over hours or days (eq 11). Eventually, all
starting (tmeda)Pt(CH3)2 ends up as (tmeda)Pt(OH)(OCH3)-
(CH3)2. The dioxygen uptake was measured, using a Toepler
pump, under conditions of high (tmeda)Pt(CH3)2 and low
O2 concentrations: 1 equiv of dioxygen is consumed per 2
equiv of (tmeda)Pt(CH3)2 (eq 12).
1
of several intermediates, which can observed by H NMR
and UV-vis spectroscopies, depending upon conditions (see
below); but ultimately a single product is formed, which can
be isolated and fully characterized by 1H NMR spectroscopy,
elemental analysis, and X-ray crystallography (see below)
as the hydroxy-methoxy-platinum(IV) complex (tmeda)-
Pt(OH)(OCH3)(CH3)2 (3). In all three cases the overall
transformation is that of eq 9.
When exposed to air, a solution of (tmeda)Pt(CH3)2 (1)
in methanol turns initially pale yellow (λmax ) 480 nm) and
then blue (λmax ) 640 nm) as the yellow color fades away.
The blue color is most intense at low dioxygen pressures
and high concentrations of 1. Addition of radical scavengers,
such as 2,6-di-tert-butyl-4-methylphenol (BHT) or p-meth-
oxyphenol, immediately bleaches the blue color. If the radical
scavenger is added to the reaction mixture before addition
of dioxygen, the colored species are not observed, but there
is no effect on the rate or the outcome of the overall oxidation
to Pt(IV).
In aprotic solvents (THF, CH2Cl2, toluene) 1 reacts with
dioxygen to give intensely blue-colored solutions. Only a
small fraction (<10%) converts to the colored species, as
1H NMR spectroscopy shows most of 1 remains unreacted,
although the peaks are quite broad. These solutions are EPR
active as well (Figure 1), suggesting that the blue color is
due to a paramagnetic species, which probably consists of
several platinum atoms linked together to form a “stacked”
mixed-valent oligomer.23 Again, the formation of the colored
The relative ratio of 2 to 3 (measured by 1H NMR
spectroscopy shortly after all 1 is consumed) is strongly
dependent on the concentrations of both 1 and O2. A high
concentration of O2 (4-8.1 mM) and a low concentration
of (tmeda)Pt(CH3)2 (0.5 mM) favor the formation of 2, while
3 is favored by low [O2] (0.7-4 mM) and high [1] (50-
190 mM), as shown in Figure 2.
Isolation, Characterization, and Reactivity of (tmeda)-
Pt(OOH)(OCH3)(CH3)2 (2). The dependence of the ratio of
2/3 on O2 and (tmeda)Pt(CH3)2 concentrations can be
exploited: stirring a dilute solution of (tmeda)Pt(CH3)2 in
(22) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996,
118, 5961-5976.
(23) Matsumoto, K.; Sakai, K. AdV. Inorg. Chem. 2000, 49, 375-427.
3610 Inorganic Chemistry, Vol. 41, No. 14, 2002

Oxidation of Dimethylplatinum(II) Complexes by Dioxygen
1H NMR spectroscopy at 50-60 °C gave only irreproducible
rates and apparent reaction order.
Several dimethylplatinum(II) complexes are oxidized to
the corresponding Pt(IV) complexes by hydroperoxy complex
2 (eq 15). The reaction between (tmeda)Pt(CH3)2 and 2 in
methanol cleanly affords 2 equiv of (tmeda)Pt(OH)(OCH3)-
(CH3)2. The rate of oxidation is too fast to follow conve-
niently (at -80 °C the half-life is about 5 min at normal
NMR concentrations with a 10-fold excess of 1). However,
increasing steric crowding around platinum(II) slows this
reaction, so that the oxidation of (teeda)Pt(CH3)2 (4, teeda
) N,N,N′,N′-tetraethylethylenediamine) by 2 can be moni-
1
tored by H NMR spectroscopy around 0 °C. The reaction
is first order in both platinum reagents with k2 (-9 °C) )
7.6 × 10-3 M-1‚s-1. The temperature dependence of the rate
constant was also examined, and the activation parameters
were calculated from an Eyring plot: ∆Hq ) 13.6 kcal‚mol-1,
∆Sq ) -16.5 kcal‚mol-1‚K-1. Oxidation of (dpe)Pt(CH3)2
(5, dpe ) 1,2-dipyrrolidinoethane) by 2 is slightly faster than
that of 4: at 0 °C and [2] ) 41 mM, the pseudo-first-order
rate constant kobs ) 16.7(3) × 10-4 s-1 and 9.3(3) × 10-4
s-1 for 5 and 4, respectively.
Figure 2. Dependence of 2/3 ratio (4) on concentrations of (tmeda)Pt-
(CH3)2 and O2: (a) [(tmeda)Pt(CH3)2] ) 21 mM; (b) [O2] ) 8.1 mM.
Theoretical values (0) are calculated for a consecutive mechanism as
described in the text (see Results and Discussion).
methanol (ca. 5 mM) under 1 atm of dioxygen results in
clean formation of (tmeda)Pt(OOH)(OCH3)(CH3)2 (eq 13),
isolated and characterized as hydroperoxy complex 2 by
1
elemental analysis, H and 195Pt NMR spectroscopy, and
1
X-ray crystallography (see below). The H and 195Pt NMR
data for 2 and 3 are very similar, except for the hydroperoxy
and hydroxy protons. The former resonates at δ 6.48 ppm
while the latter shifts to δ -2.5 ppm. 2 is not stable in the
solid state for long periods of time and decomposes to an
unidentified mixture of platinum(IV)-containing compounds.
Mechanism of Oxidation of Pt(II) by O2. The observa-
tion of paramagnetic blue-colored species during the course
of the reaction initially led us to propose a radical-based
mechanism for the oxidation.13 However, since these can be
inhibited by addition of radical scavengers without affecting
the rate or outcome of the oxidation, formation of a mixture
of (tmeda)Pt(OOH)(OCH3)(CH3)2 and (tmeda)Pt(OH)(OCH3)-
(CH3)2, we believe that single-electron transfer pathways are
minor side reactions.
We propose that the reduction of dioxygen occurs in two
consecutive two-electron steps. In the first step (eq 16),
(N-N)Pt(CH3)2 reacts with dioxygen to give (N-N)Pt(OOH)-
(OCH3)(CH3)2, which can be isolated from reaction mixtures
and fully characterized for N-N ) tmeda (2). In the second
step, this hydroperoxo complex oxidizes a second equivalent
of (N-N)PtMe2 to form 2 equiv of (N-N)Pt(OH)(OCH3)-
(CH3)2 (eq 17). Disproportionation of (N-N)Pt(OOH)(OCH3)-
A degassed solution of (tmeda)Pt(18O2H)(OCH3)(CH3)2 in
methanol (prepared by treating (tmeda)Pt(CH3)2 with 18O2)
was stirred under air for 30 min at room temperature. Mass-
spectroscopic analysis of a sample of gas from the reaction
headspace revealed that 18O was present only at the natural
abundance level, indicating that the formation of 2 by eq 13
is irreversible.
In methanol, 2 disproportionates to 3 and O2 over several
1
days at room temperature, with evolution of /2 equiv of
dioxygen, as measured by Toepler pump; the identity of the
gas was confirmed by mass spectroscopy (eq 14). Attempts
to follow the kinetics of the disproportionation reaction by
Inorganic Chemistry, Vol. 41, No. 14, 2002 3611

Rostovtsev et al.
(CH3)2 to (N-N)Pt(OH)(OCH3)(CH3)2 and O2 occurs at a
slower rate (eq 18). Therefore, after conversion to the
hydroxy complex is complete, the overall stoichiometry is
2 equiv of platinum(II) per 1 equiv of dioxygen, in agreement
with experiment. An analogous mechanism was proposed
for eq 17 by Puddephatt and co-workers.21
Figure 3. First-order dependence of the observed rate of oxidation of
(phen)Pt(CH3)2 (b) and (CyDABH)Pt(CH3)2 (9) on [O2]. T ) 23 °C.
A red-orange solution of 6 in methanol-d4 slowly fades
to pale yellow under dioxygen, with formation of (phen)-
Pt(OH)(OCH3)(CH3)2. At least one additional dimethylplati-
num(IV) species can be observed in the reaction mixture.
Unlike (tmeda)Pt(CH3)2, (phen)Pt(CH3)2 is unreactive under
a dioxygen atmosphere in aprotic solvents (THF, benzene,
toluene). Dioxygen also reacts with 7 in methanol-d4 to give
(CyDABH)Pt(OH)(OCH3)(CH3)2. Although several interme-
diates were observed during the course of the oxidation by
1H NMR spectroscopy, all of them eventually converted to
the final product (CyDABH)Pt(OH)(OCH3)(CH3)2.
The rates of oxidation of 6 and 7 can be monitored by
UV-vis spectroscopy at 412 and 364 nm, respectively. The
reactions are first-order in both platinum(II) and dioxygen
(Figure 3); the nonzero intercept of the line for 6 is due to
competing protonolysis to (phen)Pt(CH3)(OCH3) (and pre-
sumably methane). The second-order rate constants are only
slightly different: 0.11 and 0.08 s-1‚M-1 for (phen)Pt(CH3)2
and (CyDABH)Pt(CH3)2, respectively. Addition of a radical
scavenger (BHT) or exclusion of light does not change the
observed rate of oxidation, nor is the rate significantly
different in 1.0 M LiClO4 solution in methanol. At low
dioxygen pressures the absorbance-time graphs have a
distinct S-shape, indicating deviations from first-order be-
havior. This could conceivably result from one or more of
the following causes: autocatalysis, slow approach to equi-
librium distribution of dioxygen between gas and liquid
phases, or a multistage reaction mechanism. Neither the
addition of an aliquot of the starting material to the reaction
mixture after the first half-life nor the use of solvents
presaturated with dioxygen affects the observed rate, arguing
against autocatalysis and gas solubility problems, respec-
tively. On the basis of the results of 1H NMR studies, it seems
likely that non-first-order behavior at low O2 pressure reflects
the complex reaction mechanism.
According to the proposed mechanism, the ratio of the
platinum(IV) products (before eq 18 has proceeded signifi-
cantly) should depend on the concentrations of Pt(II) and
dioxygen, because the hydroperoxy complex and dioxygen
will compete for unreacted Pt(II). This proposal was tested
for (tmeda)Pt(CH3)2 (1). Figure 2 shows qualitative agree-
ment: at high concentrations of O2, 2 is the main product,
whereas at higher concentrations of platinum or lower
concentrations of O2, 3 becomes predominant. That the 2/3
ratio depends at all on [O2] rules out an alternative mecha-
nism in which 2 and 3 are formed in competing parallel
reactions of 1 with O2; in such a case the product ratio would
be independent of dioxygen concentration.
The proposed mechanism can also be tested quantitatively,
by kinetic simulation. Assuming the two-step sequence of
eqs 16 and 17, the ratio of the peroxy:hydroxy products (2/
3) can be predicted, from the initial concentrations and the
ratio of second-order rate constants for eqs 16 and 17, by
numerical integration of the appropriate differential equa-
tions. This calculation was performed for the conditions
corresponding to the data point of plot a (Figure 2) with
highest [O2] while the rate constant ratio k16/k17 was varied
until good agreement between the calculated and experi-
mental values of 2/3 was achieved, yielding k16/k17 ) 18.
This ratio was then used to predict 2/3 ratios at other
concentrations of O2 and 1. As shown in Figure 2, the
predictions are in good agreement with experiment, strongly
supporting the sequential mechanism.
Reactions of Other Platinum(II) Complexes with O2.
The reactions of (phen)Pt(CH3)2 (6) and (CyDABH)Pt(CH3)2
1
(7) with dioxygen were studied by H NMR spectroscopy
(eq 19).
Oxidation of a number of (R-diimine)Pt(CH3)2 complexes
was monitored by UV-vis spectroscopy to assess the
influence of steric and electronic factors on the observed rate
of oxidation. Overall, R-diimine complexes of platinum(II)
react more slowly with O2 than do diamine complexes. This
difference in rates may be correlated with the expected
electron density at the metal center. R-Diimines, as weaker
3612 Inorganic Chemistry, Vol. 41, No. 14, 2002

Oxidation of Dimethylplatinum(II) Complexes by Dioxygen
Table 1. Relative Rates for Reaction of (R-Diimine)Pt(CH3)2 with Dioxygen in Acetonitrile/Methanol Solution
σ-donors and better π-acceptors than diamines, should give
less electron-rich platinum complexes. The electronic char-
acter of a ligand may be assessed by the stretching frequency
of the corresponding carbonyl cation where lower frequency
indicates a more electron-donating ligand. Using C-O
stretching frequencies in [(N-N)Pt(CH3)(CO)]BF4 as a probe
of electron density,24 we find νCO ) 2102.0 cm-1 for N-N
) tmeda compared to 2108.4 cm-1 for N-N ) phen, and
2105-2115 cm-1 for diazabutadiene derivatives (Table 1).
Within the group of R-diimines, observed pseudo first-order
rate constants (1 atm O2) do not vary greatly with electronic
character, spanning only a small range of values from 2.7 ×
10-4 to 14.5 × 10-4 s-1. On the other hand, there is a
pronounced steric effect: complexes of bulky ligands (N,N′-
diaryldiazabutadienes with ortho substituents of the aryl
groups) react with O2 very slowly or not at all.
(CH3)2 and (CyDABH)Pt(CH3)2 are oxidized by dioxygen,
no oxidation takes place with (tmeda)Pt(CH3)Cl, (tmeda)-
PtCl2, (tmeda)Pt(C6H5)2, or (CyDABH)Pt(C6H5)2. The mixed
methyl-phenyl compounds (tmeda)Pt(C6H5)(CH3) and
(CyDABH)Pt(C6H5)(CH3) do exhibit NMR changes consis-
tent with oxidation to Pt(IV) on exposure to dioxygen in
methanol, although the products were not isolated. Thus one
methyl group may be replaced by phenyl and reaction still
occurs, but replacing both leads to complete unreactivity
toward O2. This might be due to either steric or electronic
effects.
Dimethylplatinum(II) complexes of R-diimines are also
oxidized by 2, but these oxidations are not as clean as those
described above. With 1 equiv each of (N-N)Pt(CH3)2
(N-N ) phen, 6; CyDABH, 7) and 2, equimolar amounts of
(N-N)Pt(OH)(OCH3)(CH3)2 and 3 are formed (eq 20).
However, if an excess of 2 is present, all of it is converted
to 3, even though only a stoichiometric amount of (N-N)-
Pt(OH)(OCH3)(CH3)2 is formed. This implies that some
species formed in these reactions must act as a catalyst for
Replacement of methyl groups on Pt(II) has a more
pronounced effect on reactivity. Whereas both (tmeda)Pt-
(24) Zhong, A. Z.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2002,
124, 1378-1399.
Inorganic Chemistry, Vol. 41, No. 14, 2002 3613

Rostovtsev et al.
decomposition of 2 to 3 plus O2, although this behavior has
not been further investigated.
were unable to detect hydrogen peroxide by electrochemical
methods in any of our reactions.
A plausible alternative mechanism might be suggested by
the recent finding of a photochemical reaction of Tp*Pt-
(CH3)2H (Tp* ) tris(3,5-dimethylpyrazolyl)borate), which
reacts with dioxygen in benzene to afford Tp*Pt(CH3)2-
(OOH), apparently by a radical mechanism.28 It is known
that in protic solvents such as methanol, dimethylplatinum-
(II) complexes are in equilibrium with alkylhydridoplatinum-
(IV) species;22 the latter could undergo insertion of O2 into
the PtIV-H bond to give the hydroperoxy complex. Arguing
against this is the fact that none of our oxidations shows
any effect whatsoever of exclusion of light, as well as the
absence of any indicators of a radical pathway, except for
the minor side reaction, which as discussed above may be
inhibited without affecting formation of the Pt(IV) species.
Since formation of a hydroperoxy complex involves a
proton transfer, one might expect some dependence on [H+].
However, addition of sodium methoxide (0.5 M) does not
change the rate of oxidation of (phen)Pt(CH3)2, nor is the
rate sensitive to the concentration of methanol (18-24 M)
in CH3CN/CH3OH mixed solvent. This suggests that the first
step in eq 21, formation of 8 or 9, is rate-limiting. Oxidation
cannot be followed in acidic solutions because of competing
protonolysis of the platinum-carbon bond (see above).
There are examples of oxidations that may well involve
Pt(II)-O2 intermediates. The bis(µ2-peroxo)platinum(IV)
dimer 10 is obtained in aerobic oxidation of Pt(II) in the
presence of triazacyclononane (eq 22) and is thought to form
via a complex between κ2-tacn-Pt(II) and O2 followed by
dimerization.29
All dimethylplatinum(II) complexes can be oxidized by
aqueous hydrogen peroxide as well as dioxygen; in methanol
the same products, (N-N)Pt(OH)(OCH3)(CH3)2, are obtained.
In nonprotic solvents the analogous dihydroxy complexes
(N-N)Pt(OH)2(CH3)2 are formed.16
The oxidizing capability of (tmeda)Pt(OOH)(OCH3)(CH3)2
(2) is of limited scope: besides the oxidation of dimethyl-
platinum(II) complexes, it converts triphenylphosphine to
triphenylphosphine oxide, but does not react with Me2S, Me2-
SO, cyclohexene, or styrene. Moreover, addition of strong
acid does not promote the oxidization of these same
substrates by hydroperoxy complex 2.
Interaction of Pt(II) with Dioxygen. There must be at
least two reaction pathways accessible for the reaction of
dimethylplatinum(II) complexes with dioxygen, but the
radical pathway signaled by the appearance of deeply colored
species appears to be a minor side reaction. The major
detectable product from the reaction of platinum(II) with
dioxygen is the hydroperoxy complex (N-N)Pt(OOH)-
(OCH3)(CH3)2. Most probably that would arise via proto-
nation of a dioxygen complex (eq 21). There are no
precedents for formation of stable or metastable dioxygen
Formation of the alkylperoxoplatinum(II) complex in eq 23
was proposed to occur in a concerted manner through a six-
membered transition state.30 It is interesting that eq 23
proceeds for R-diimines with bulky aryl groups (Ar ) 2,6-
diisopropylphenyl, 2,6-dimethylphenyl); the corresponding
dimethyl complexes do not react with dioxygen in methanol
(see Table 1). The latter reaction is apparently more sensitive
to steric crowding; possibly this indicates that both dioxygen
and methanol must coordinate in the transition state.
In this context it is noteworthy that in our systems,
involving strictly bidentate ligands, only neutral platinum-
(II) complexes with very strong σ-donors such as methyl
complexes of platinum(II), with either η1- (8) or η2-structures
(9). Platinum(0) reacts with dioxygen to form η2-complexes,25
as do square-planar iridium(I) complexes isoelectronic with
(N-N)Pt(CH3)2.26 Protonation of η2-peroxo complexes of
iridium(III) and platinum(II), to give binuclear µ2,η1-peroxo
complexes or hydrogen peroxide, has been reported.27 We
(27) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F. J. Am. Chem. Soc.
1990, 112, 6726-6728. Bhaduri, S.; Casella, L.; Ugo, R.; Raithby, P.
R.; Zuccaro, C.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans. 1979,
1624-1629. Hughes, G. R.; Mingos, D. M. P. Transition Met. Chem.
1978, 3, 381-382. Muto, S.; Kamiya, Y. Bull. Chem. Soc. Jpn. 1976,
49, 2587-2589.
(28) Wick, D. D.; Goldberg, K. I. J. Am. Chem. Soc. 1999, 121, 11900-
11901.
(25) Nyman, C. J.; Wymore, C. E.; Wilkinson, G. J. Chem. Soc. A 1968,
561-563.
(29) Davies, M. S.; Hambley, T. W. Inorg. Chem. 1998, 37, 5408-5409.
(30) Tom Dieck, H.; Fendesak, G.; Munz, C. Polyhedron 1991, 10, 255-
260.
(26) Lawson, H. J.; Atwood, J. D. J. Am. Chem. Soc. 1989, 111, 6223-
6227.
3614 Inorganic Chemistry, Vol. 41, No. 14, 2002

Oxidation of Dimethylplatinum(II) Complexes by Dioxygen
Figure 5. Labeled Diamond view of (tmeda)Pt(OH)(OCH3)(CH3)2 (3) with
50% ellipsoids.
Figure 4. Qualitative MO diagram of a square-planar platinum(II)
complex. Effects of axial ligands and strong σ-donors on HOMO energy.
probably occurs through an analogous transition state. The
slower reaction of 2 with (N-N)Pt(CH3)2 (N-N ) phen,
CyDABH) than with (tmeda)Pt(CH3)2 probably reflects the
lower electron density at platinum in the former.
groups react with dioxygen. In contrast, with κ2-bound but
potentially tridentate ligands, even cationic platinum(II)
complexes with weaker nitrogen-based σ-donors (amines,
imines) are converted to platinum(IV) products with air (eqs
5-7). Wieghardt and co-workers propose that coordination
of the “dangling” arm brings about repulsion between the
2
lone pair ligand and the filled dz orbital (Figure 4, left side),
raising the energy of HOMO of the platinum complex (σ-
antibonding dz -based orbital) and facilitating oxidation.20 In
2
dimethylplatinum(II) complexes, strong σ-bonds raise the
energy of the HOMO, making it more susceptible to
oxidation (Figure 4, right side); additional coordination of a
nonchelating ligand (solvent methanol) may also be required,
as suggested by the steric argument in the preceding
paragraph.
Disproportionation of (tmeda)Pt(OOH)(OCH3)(CH3)2 to
dioxygen and (tmeda)Pt(OH)(OCH3)(CH3)2 has precedent in
the chemistry of platinum(IV), Tp*Pt(CH3)2(OOH),28 and
cobalt(III), [(CN)5CoOOH]3-.32 Taking into account the
substitutional inertness of octahedral platinum(IV) com-
plexes,33 it seems likely that formation of O2 is initiated by
homolytic cleavage of the O-O bond. A similar mode of
activation has been proposed for the disproportionation of
t-BuOOH to t-BuOH and O2.34
Molecular Structures of (tmeda)Pt(OH)(OCH3)(CH3)2
(3) and (tmeda)Pt(OOH)(OCH3)(CH3)2 (2). An X-ray
structure determination for (tmeda)Pt(OH)(OCH3)(CH3)2 (3)
(Figure 5) confirmed that proposed for the bipy analogue
by Puddephatt and Monaghan on the basis of spectroscopic
features.12 Bond lengths and angles around platinum are
within the range expected for Pt-O, Pt-N, and Pt-C bonds.
Selected bond lengths and angles are shown in Table 2. There
is an intermolecular hydrogen bond between O1 and O2 (at
x, y, z - 1) of 2.946(9) Å (the O1-H1A-O2 angle is
Reactivity of (tmeda)Pt(OOH)(OCH3)(CH3)2 (2). We
interpret oxidation of dimethylplatinum(II) complexes by
(tmeda)Pt(OOH)(OCH3)(CH3)2 in the same mechanistic
terms as reactions of alkyl or hydrogen peroxides with
triphenylphosphine, dimethyl sulfide, or other oxidizable
nucleophiles.31 The lone pair of the nucleophilesin this case,
2
a pair of electrons in the dz orbital of platinum(II) complexs
attacks the antibonding (σ*) molecular orbital, effecting
heterolysis of the O-O bond (11).
In the case of (N-N)Pt(OOH)(OCH3)(CH3)2, attack most
likely occurs at the â-oxygen of the hydroperoxy ligand (eq
24). The increased steric bulk of (teeda)Pt(CH3)2 vs (tmeda)-
Pt(CH3)2 destabilizes the transition state and slows down the
rate of the reaction. Oxidation of triphenylphosphines by 2
(32) Pregaglia, G.; Morelli, D.; Conti, F.; Gregorio, G.; Ugo, R. Faraday
Discuss. Chem. Soc. 1968, 110-121.
(33) No exchange was observed between (tmeda)Pt(OH)(OCH3)(CH3)2 and
H217O after several days at room temperature. (phen)Pt(OH)(OCH3)-
(CH3)2 is stable in 1 M HCl for days (see ref 12).
(31) Conte, V.; Di Furia, F.; Modena, G. In Organic Peroxides; Ando,
W., Ed.; John Wiley & Sons: Chichester, 1992; pp 559-598.
(34) Hiatt, R.; Mill, T.; Mayo, F. R. J. Org. Chem. 1968, 33, 1416-1420.
Inorganic Chemistry, Vol. 41, No. 14, 2002 3615

Rostovtsev et al.
O-O bond lengths (1.44(2) and 1.48(2) Å) are close to
values reported in the literature for Pt(IV)-peroxy species
(cf. 1.481(5) Å in Tp*Pt(OOH)(CH3)2 (Tp* ) tris(3,5-
dimethylpyrazolyl)borate),35 1.465(3) Å in 10 (see above),
and 1.394(6) Å in (phen)Pt(OOCHMe2)(CH3)2Cl).36 As
shown in the hydrogen bond table (see Supporting Informa-
tion), there is a hydrogen bond between the hydroperoxy
group of one orientation and the methoxy oxygen of the other
orientation. These close intermolecular contacts may explain
the low stability of 2 in the solid.
Figure 6. Labeled Diamond view of (tmeda)Pt(OOH)(OCH3)(CH3)2 (2)
with 50% ellipsoids.
Experimental Section
Table 2. Selected Bond Lengths and Angles for
(tmeda)Pt(OH)(OCH3)(CH3)2 (3)
General Considerations. All air and moisture sensitive com-
pounds were manipulated using standard high-vacuum line, Schlenk,
or cannula techniques, or in a drybox under a nitrogen atmosphere.
N,N,N′,N′-Tetramethylethylenediamine (tmeda) and 1,10-phenan-
throline (phen) were purchased from Aldrich. (COD)Pt(CH3)2 was
purchased from Strem Chemicals. Ultrahigh-purity oxygen was
purchased from Liquid Air Products and used without further
purification. Solvents were dried over sodium/benzophenone ketyl
(Et2O, THF, petroleum ether), 4 Å molecular seives (CH3OH,
CH3NO2), or CaH2 (CH2Cl2). The following compounds were
prepared according to literature procedures: (CH3)2Pt{µ2-(CH3)2S}2-
Pt(CH3)2, {(CH3)2S}2Pt(CH3)Cl,37 {(CH3)2S}2PtPh2,38 (tmeda)Pt-
(CH3)2, (phen)Pt(CH3)2,39 and 1,2-dipyrrolidinoethane (dpe).40 Di-
methylplatinum(II) complexes of R-diimines were prepared from
(CH3)2Pt{µ2-(CH3)2S}2Pt(CH3)2 by the method published by Bercaw
and co-workers.41 R-Diimines were prepared as described earlier.42
NMR spectra were acquired on Varian UNITYINOVA 500 (499.853
MHz for 1H), Varian Mercury-vx 300 (300.08 MHz for 1H), General
length (Å)
angle (deg)
Pt-O1
1.995(7)
2.007(8)
2.028(6)
2.229(6)
0.55(12)
1.362(13)
0.60(3)
O1-Pt-C1
C1i-Pt-C1a
O1-Pt-O2
C1-Pt-O2
O1-Pt-N
C1i-Pt-N
C1-Pt-N
O2-Pt-N
N-Pt-Ni
90.8(3)
89.6(5)
174.9(3)
92.8(3)
87.6(2)
176.1(3)
93.9(3)
88.6(2)
82.5(3)
113(10)
119.5(6)
Pt-C1
Pt-O2
Pt-N
O1-H0A
O2-C2
C4A‚‚‚C4B
C3A‚‚‚C3B
0.74(2)
Pt-O1-H0A
C2-O2-Pt
a Symmetry transformations used to generate equivalent atoms: (i) x,
3
-y + /2, z.
Table 3. Selected Bond Lengths and Angles for
(tmeda)Pt(OOH)(OCH3)(CH3)2 (2)
length (Å)
angle (deg)
1
Electric QE300 (300 MHz for H) ,and Bruker AM500 (500.13
Pt-O1A
Pt-O3A
Pt-C8A
Pt-C9A
Pt-N1A
Pt-N2A
O1A-O2A
O3A-C7A
Pt-O1B
Pt-O3B
Pt-C8B
Pt-C9B
Pt-N1B
Pt-N2B
O1B-O2B
O3B-C7B
2.006(16)
2.051(13)
2.069(18)
2.01(2)
2.276(16)
2.17(2)
1.48(2)
1.40(2)
2.015(16)
2.033(13)
2.04(2)
O1A-Pt-O3A
C8A-Pt-N2A
C9A-Pt-N1A
O2A-O1A-Pt
C7A-O3A-Pt
O1B-Pt-O3B
C9B-Pt-N1B
C8B-Pt-N2B
O2B-O1B-Pt
C7B-O3B-Pt
170.5(6)
174.1(7)
179.2(8)
112.7(10)
120.0(12)
170.8(5)
173.9(7)
177.0(8)
116.0(11)
121.9(15)
MHz for 1H) spectrometers. All 1H NMR shifts are relative to the
residual NMR solvent. 195Pt NMR shifts were referenced with
respect to a saturated aqueous solution of K2PtCl4. UV-vis spectra
were recorded on a HP8452A diode array spectrophotometer. EPR
spectra were acquired on a Bruker X-band EPR spectrometer.
Elemental analyses were carried out either at the Caltech Elemental
Analysis Facility by Mr. Fenton Harvey or by Midwest Microlab,
Indianapolis, IN.
General Procedure for the Synthesis of (N-N)Pt(CH3)2. In a
50 mL Schlenk flask, (CH3)2Pt{µ2-(CH3)2S}2Pt(CH3)2 (0.5-0.8
mmol) was dissolved in 10 mL of dry THF. The desired diamine
(2 equiv, 1.0-1.6 mmol) was added to the solution. The resulting
mixture was stirred for 8 h at room temperature. The solution was
occasionally purged with argon to remove the liberated dimethyl
sulfide. Removal of volatiles under vacuum provided the desired
compound. If necessary, the resulting complex was purified by
recrystallization from THF/pentane or toluene/pentane mixtures.
2.05(2)
2.200(16)
2.281(16)
1.44(2)
1.39(3)
172(18)°), which forms a linear chain along the O2-Pt-
O1 axis (c axis).
The determination of the X-ray structure for (tmeda)Pt-
(OOH)(OCH3)(CH3)2 (2) was especially problematic, since
this structure is both twinned and disordered. The asymmetric
unit consists of one molecule of 2, which is disordered ∼1:1
over two orientations. Its solution was possible, although the
bond distances and angles are necessarily rather imprecise.
Orientation A, shown in Figure 6, has a slightly more
reasonable structure than orientation B, which has a carbon
atom (C1B of the tmeda backbone) with distorted geometry.
This distortion is most likely due to additional disorder of
the tmeda ligand. The selected bond lengths and angles for
both orientations are shown in Table 3. All dimensions
around the central platinum atom are within the limits
expected for Pt(IV)-O, Pt(IV)-N, and Pt(IV)-C bonds. The
(35) Wick, D. D.; Goldberg, K. I. J. Am. Chem. Soc. 1999, 121, 11900-
11901.
(36) Ferguson, G.; Monaghan, P. K.; Parvez, M.; Puddephatt, R. J.
Organometallics 1985, 4, 1669-1974.
(37) Hill, G. S.; Irwin, M. J.; Levy, C. J.; Rendina, L. M.; Puddephatt, R.
J. Inorg. Synth. 1998, 32, 149-153.
(38) Rashidi, M.; Fakhroeian, Z.; Puddephatt, R. J. J. Organomet. Chem.
1991, 406, 261-267.
(39) Clark, H. C.; Manzer, L. E. J. Organomet. Chem. 1973, 59, 411-
428.
(40) Remenar, J. F.; Lucht, B. L.; Collum, D. B. J. Am. Chem. Soc. 1997,
119, 5567-5572.
(41) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 2000, 122, 10846-10855.
(42) Tom Dieck, H.; Svoboda, M.; Greiser, T. Z. Naturforsch., B: Anorg.
Chem., Org. Chem. 1981, 36B, 823-832.
3616 Inorganic Chemistry, Vol. 41, No. 14, 2002

Oxidation of Dimethylplatinum(II) Complexes by Dioxygen
3
(teeda)Pt(CH3)2. The general procedure afforded the title
compound as a white solid. Yield: 0.22 mg (55%) after recrys-
m-H in Pt-C6H5), 7.22 (d, 4H, JPtH ) 67.0 Hz, o-H in PtC6H5),
3
8.81 (s, 2H, JPtH ) 36.8 Hz, Pt-NdCH).
1
2
tallization from THF. H NMR (C6D6, δ): 0.97 (s, 6H, JPtH
)
(tmeda)Pt(C6H5)(CH3). Phenyllithium (0.25 mL, 1.8 M ether
solution, 0.45 mmol) was added by syringe to a slurry of (Me2S)2-
Pt(CH3)Cl (150 mg, 0.41 mmol) in 20 mL of dry Et2O at -78 °C.
The reaction mixture was stirred for 15 min, then warmed to 0 °C,
and stirred for an additional 25 min. Tmeda (0.5 mL) was added
via syringe, and the resulting suspension was kept at room
temperature for another 15 min and then quenched with a cold
saturated aqueous solution of NH4Cl (10 mL). The organic layer
was separated, and the aqueous layer was washed with Et2O (2 ×
20 mL). The combined organic portions were dried over MgSO4,
treated with charcoal (0.1 g), and filtered. Evaporation of Et2O gave
a white powder. Yield: 30 mg (18%). 1H NMR (CDCl3, δ): 0.45
88.0 Hz, Pt-CH3), 1.08 (t, 12H, 3JHH ) 7 Hz, NCH2CH3), 1.96 (s,
4H, NCH2CH2N), 2.44 (dq, 4H, 2JHH ) 13 Hz, 3JHH ) 7 Hz, NCH2-
CH3), 2.68 (dq, 4H, 2JHH ) 13.5 Hz, 3JHH ) 7 Hz, N-CH2-CH3).
Anal. Calcd for C12H30N2Pt: C 36.26; H 7.61; N 7.05. Found: C
36.03; H 7.48; N 6.78.
(dpe)Pt(CH3)2. The general procedure afforded the title com-
pound as a white solid in 72% yield. The product was recrystallized
from toluene/PE. 1H NMR (CD3CN, δ): 0.07 (s, 6H, 2JPtH ) 86.0
3
Hz, PtCH3), 1.78 (m, 8H, N(c-C4H8)), 2.62 (s, 4H, JPtH ) 12.4
Hz, NCH2CH2N), 2.86 (m, 4H, N(c-C4H8)), 3.21 (m, 4H, N(c-
C4H8)). Anal. Calcd for C12H26N2Pt: C 36.63; H 6.66; N 7.12.
Found: C 36.40; H 6.99; N 6.87.
2
(s, 3H, JPtH ) 88.5 Hz, PtCH3), 2.54 (s, 6H, NCH3), 2.72 (s, 6H,
NCH3), 6.83 (pseudo d, 1H, p-H in PtC6H5), 6.93 (pseudo dd, 2H,
(tmeda)Pt(OH)(OCH3)(CH3)2. A solution of (tmeda)Pt(CH3)2
(341 mg, 1 mmol) in methanol (10 mL) was stirred under air for
20 h and then concentrated in vacuo. The residual solid was taken
up in CH2Cl2 (5 mL), treated with charcoal, and filtered. Addition
of petroleum ether gave pale yellow (tmeda)Pt(OH)(OCH3)(CH3)2
3
m-H in PtC6H5), 7.38 (d, 2H, JPtH ) 73 Hz, o-H in PtC6H5).
(CyDABH)Pt(C6H5)(CH3). Substituting CyDABH for tmeda in
the above procedure gave the title compound as an approximately
1
1:1 mixture with (CyDABH)Pt(C6H5)2. H NMR signals assigned
to (CyDABH)Pt(C6H5)(CH3) (CD2Cl2/CD3OD (1:1), δ): 0.9-2.2
1
(370 mg, 95%). H NMR (CD2Cl2, 200 K, δ): -2.5 (s, 1H, Pt-
2
OH), 0.99 (s, 6H, 2JPtH ) 74 Hz, PtCH3), 2.39 (m, 4H, NCH2CH2N),
(m, 10H, N-C6H11), 1.23 (s, 3H, JPtH ) 85.4 Hz, PtCH3), 3.79
3
(m, 2H, N-CH), 6.72 (pseudo t, 2H, p-H in Pt-C6H5), 6.90 (pseudo
2.56 (overlapping s, 12H, NCH3), 2.82 (s, 3H, JPtH ) 42 Hz,
3
PtOCH3). 195Pt NMR (CD3OD, δ): -727.76. This preparation of 3
did not give satisfactory elemental analyses, but an analytically pure
sample was obtained by oxidizing (tmeda)Pt(CH3)2 with a slight
excess of 30% aqueous H2O2 and following a similar workup; the
NMR spectrum was identical. Anal. Calcd for C9H28N2O3Pt
((tmeda)Pt(OH)(OCH3)(CH3)2‚H2O): C 26.56; H 6.93; N 6.88.
Found: C 26.37; H 6.55; N 6.70.
t, 4H, m-H in Pt-C6H5), 7.32 (d, 4H, JPtH ) 72.5 Hz, o-H in
3
PtC6H5), 8.85 (s, 1H, JPtH ) 30.1 Hz, Pt-NdCH, trans to the
3
methyl group), 8.94 (s, 1H, JPtH ) 37.9 Hz, Pt-NdCH, trans to
the phenyl group).
Determination of Stoichiometry in Dioxygen. A solution of
(tmeda)Pt(CH3)2 (341 mg, 1 mmol) in methanol (20 mL) was stirred
under 2 equiv of dioxygen for 2.5 days. The leftover gas (1.5 equiv)
was quantified by a Toepler pump.
Conversion of 2 to 3. A freshly prepared solution of 2 in
methanol was placed into a medium-walled glass bomb. The
solution was degassed and placed into a 65 °C oil bath for 20 h.
The evolved gas was quantified by a Toepler pump.
(tmeda)Pt(OOH)(OCH3)(CH3)2. In a drybox, (tmeda)Pt(CH3)2
(19 mg, 0.056 mmol) was placed in a 100 mL round bottom flask,
which was then fitted with a 180° Kontes valve. The flask was
removed from the box, and dry methanol (50 mL) was added by
cannula. The resulting solution was degassed at -78 °C, and the
flask was filled with dioxygen (1 atm). The reaction mixture was
stirred for 2 h. Volatiles were removed in vacuo to give the title
compound as a white solid. Yield: 21 mg (96%). 1H NMR (toluene-
General Procedure for the NMR Tube Reaction of (N-N)-
Pt(CH3)2 with Dioxygen. The appropriate platinum complex (∼5
mg, 0.015 mmol) was placed into an NMR tube equipped with a J
Young valve. Methanol-d4 (0.6 mL) was added by vacuum transfer.
The NMR tube was filled with dioxygen (1 atm) at -78 °C. The
sample was vigorously shaken as it was warmed to room temper-
ature. The reaction was usually complete within several minutes.
In some cases (see main text), the solution turned pale pink or pale
blue.
General Procedure for Monitoring the Oxidation of Di-
methylplatinum(II) Complexes with Dioxygen by UV/Vis Spec-
troscopy. In a drybox, a stock solution of a platinum complex in
acetonitrile or toluene (50 µL, 2-5 mM) was placed in a 1 cm UV
cell equipped with a magnetic stir bar, a 14/20 ground glass adapter
at 120° to the UV cell axis, and a side arm with a magnetic stir bar
(5 mL round bottom flask at 90° to the UV cell axis). The cell was
capped, taken out of the drybox, and attached to a vacuum line
through a 128.2 mL gas bulb. Dry methanol (2.5 mL) was added
via syringe into the side arm. The entire cell was cooled to -78
°C and degassed. The gas bulb was filled with the appropriate
amount of dioxygen (100-630 Torr). After dioxygen was admitted
into the UV cell at -78 °C, both the solution of platinum complex
and the methanol were stirred for 3 min as they warmed to room
temperature, to ensure that a sufficient amount of dioxygen
dissolved in the liquid phase. The methanol was then added to the
UV cell by tipping the cell. The reaction mixture was vigorously
stirred in the cell holder. Reactions were monitored for at least 3
half-lives.
2
d8, δ): 1.64 (s, 6H, JPtH ) 76 Hz, PtCH3), 1.95 (broad s, 4H,
NCH2CH2N), 2.18 (s, 6H, NCH3), 2.24 (s, 6H, NCH3), 3.27 (s,
3H, 3JPtH ) 41.5 Hz, PtOCH3), 6.48 (s, 1H, 3JPtH ) 13 Hz, PtOOH).
195Pt NMR (CD3OD, δ): -716.89. Anal. Calcd for C9H26N2O3Pt:
C 26.66; H 6.46; N 6.91. Found: C 26.67; H 6.52; N 6.90.
(tmeda)Pt(C6H5)2. (Me2S)2PtPh2 was prepared by slowly adding
a toluene/Et2O solution of PhLi (1.05 mL, 1.8 M) by syringe to a
slurry of (Me2S)2PtCl2 (0.3 g, 0.77 mmol) in Et2O (15 mL) cooled
to 0 °C. The reaction mixture, initially yellow, was stirred at 0 °C
until a beige suspension was obtained. Excess tmeda (0.2 mL, 1.3
mmol) was added by syringe, and the reaction mixture was stirred
for an additional 30 min at room temperature. The reaction mixture
was quenched with a cold saturated aqueous solution of NH4Cl
(10 mL). The organic layer was separated, and the aqueous layer
was washed with Et2O (2 × 20 mL). The combined organic portions
were dried over MgSO4, treated with 0.1 g of charcoal, and filtered.
Evaporation of Et2O gave a white powder. Yield: 30 mg (9%). 1H
NMR (CD2Cl2/CD3OD, δ): 2.46 (s, 12H, 3JPtH ) 22.2 Hz, NCH3),
3
2.69 (s, 4H, JPtH ) 11.1 Hz, NCH2CH2N), 6.61 (m, 2H, p-H in
3
Pt-C6H5), 6.75 (m, 4H, m-H in Pt-C6H5), 7.38 (m, 4H, JPtH
75.0 Hz, o-H in Pt-C6H5).
)
(CyDABH)Pt(C6H5)2 was prepared from (Me2S)2PtPh2 and
1
CyDABH according to the above procedure. H NMR (CD2Cl2/
CD3OD (1:1), δ): 0.9-2.2 (m, 10H, N-C6H11), 3.68 (m, 2H,
N-CH), 6.79 (pseudo t, 2H, p-H in Pt-C6H5), 6.96 (pseudo t, 4H,
Inorganic Chemistry, Vol. 41, No. 14, 2002 3617

Rostovtsev et al.
Table 4. X-ray Experimental Data
processed data set; 2095 reflections were rejected, with 33 space
group absence violations and 32 inconsistent equivalents. Refine-
ment of F2 was against all reflections. The weighted R-factor (Rw)
and goodness of fit (S) are based on F2, and conventional R-factors
(R) are based on F, with F set to zero for negative F2. The threshold
expression of F2 > 2σ(F2) is used only for calculating R-factors(gt)
and is not relevant to the choice of reflections for refinement. The
coordinates of all hydrogen atoms including the half-populated sites
were refined with Uiso’s fixed at 120% of the Ueq of the attached
atom. There are 17 peaks in the final difference map with intensities
g|1| e‚Å-3; the two largest are -2.31 e‚Å-3 (0.92 Å from Pt) and
1.62 e‚Å-3 (0.89 Å from Pt). Nine of these peaks, with intensities
up to 1.45 e‚Å-3 and -1.53 e‚Å-3, are in the vicinity of the dis-
ordered region. The largest shift/esd of 0.155 is for the x coordinate
of the half-hydrogen H3AA which oscillates approximately 0.05 Å.
Structure Determinations for (tmeda)Pt(OOH)(OCH3)(CH3)2
(2). A semi-opaque, rounded thin wedge was selected from a batch
of crystals, which were freshly grown from acetone at -35 °C,
and mounted on a glass fiber with Paratone-N oil. This sample
was the best of many chosen from a number of crystallizations;
data collected at room temperature were no better than low-
temperature data (Table 4). Five runs of data were collected with
15 s long, -0.3° wide ω-scans at five values of æ (0°, 120°, 240°,
180°, and 300°) with the detector 5 cm (nominal) distant at a θ of
-28°. The spots were broad and sometimes split. The initial cell
for data reduction was calculated from 999 centered reflections
chosen from throughout the data frames. A total of 41 reflections
were discarded in the triclinic least-squares with a reciprocal lattice
vector tolerance of 0.008 (0.005 is customary). Of the rejects, 20
had substantially nonintegral indices and presumably are from a
small fragment with a different orientation. There was no evidence
of a larger cell. For data processing with SAINT v6.02, all defaults
were used, except that a fixed box size of 2.0 × 2.0 × 1.2 was
used, periodic orientation matrix updating was disabled, the
instrument error was set to zero, no Laue class integration restraints
were used, and, for the postintegration global least squares
refinement, no constraints were applied. The profiles exhibited an
apparently considerable truncation of many reflections, especially
in the z-direction. A number of both larger and smaller box sizes
were tried but gave worse refinements of the structure. Both the
initial cell refinement and the global cell refinement gave an ω
zero ∼0.4° too negative. No decay correction was needed. The
crystal boundaries were rudely indexed as faces; these were not
good enough for a face-indexed absorption correction. A SADABS
v2.03 â correction was applied (relative correction factors: 1.000-
0.471) with g ) 0.113, 100 refinement cycles, and defaults for the
remaining parameters. Again, a number of other options such as
discarding portions of the data and using different orders of Bessel
functions did not lead to a better model refinement.
2
3
formula
fw
C9H26N2O3Pt
405.41
C9H26N2O2Pt
389.41
cryst syst
space group
a, Å
b, Å
c, Å
monoclinic
P21/n (No. 14)
6.5985(8)
13.5019(16)
14.9187(18)
90
93.408(2)
90
1326.8(3)
4
orthorhombic
Pnma (No. 62)
16.101(5)
13.245(4)
6.134(2)
90
R, deg
â, deg
γ, deg
vol, Å3
Z
90
90
1308.1(7)
4
1.977
10.71
1.10, 0.89
752
F
calc, g/cm3
2.030
10.57
1.000, 0.471
784
µ, mm-1
Tmax,min
F000
cryst shape
cryst color
cryst size, mm
T, K
irregular thin wedge
colorless
0.07 × 0.21 × 0.22
98
prismatic
colorless
0.11 × 0.13 × 0.15
84
type of diffractometer Bruker SMART 1000 ccd Nonius CAD-4
wavelength
θ range, deg
h,k,l limits
data measd
unique data
Rint
0.71073 Å Mo KR
0.71073 Å Mo KR
2.0, 25.0
1.5, 25
-7, 7; -16, 16; -17, 17 -19, 19; -15, 0; -7, 7
16295
2335
5701
1203
0.013
1026
0.074
data, Fo > 4σ(Fo)
params/restraints
R1, wR2; all data
2096
123/0
0.052, 0.100
133/0
0.046, 0.059
0.030, 0.053
1.83
R1, wR2; Fo > 4σ(Fo) 0.046, 0.099
GOF on F2
2.20
2.98, -2.73
∆Fmax,min, e Å-3
1.62, -2.31
Standard numerical integration (Euler method) using Chipmunk
BASIC software (http://www.rahul.net/rhn/basic/) was employed
for simulation of kinetics as a function of concentrations.
General Procedure for Monitoring the Oxidation of (teeda)-
Pt(CH3)2 with (tmeda)Pt(OOH)(OCH3)(CH3)2 by 1H NMR
Spectroscopy. In a drybox, a stock solution of (tmeda)Pt(OOH)-
(OCH3)(CH3)2 (0.6-0.7 mL, 40-120 mM) in CD3OD was placed
into a Wilmad NMR tube equipped with a screw cap with Teflon
insert. An aliquot of (teeda)Pt(CH3)2 solution in methanol (50 mL)
was added by syringe to the cold NMR tube (-78 °C). The NMR
tube was vigorously shaken, placed into the spectrometer, and
allowed to equilibrate to the probe temperature while the sample
was shimmed. Arrayed spectra were scaled to the first spectrum in
the series, and kinetic fits were done using kini and kind functions
built into VNMR software (version 6.1B, Varian Associates, Inc.).
Structure Determination of (tmeda)Pt(OH)(OCH3)(CH3)2 (3).
The title complex was crystallized from methanol at -78 °C. Four
crystals were examined on the diffractometer, but only one was
eventually used (Table 4). The first three crystals were discarded
due to broad diffraction peaks and severe indexing problems. The
crystals were mounted on a glass fiber with Paratone-N oil. Data
were collected with 1.5° ω-scans because of the poor crystal quality.
The individual backgrounds were replaced by a background function
of 2θ derived from weak reflections with I < 3σ(I). The GOFmerge
was 1.08 (1203 multiples); Rint was 0.013 for 94 duplicates (there
are a small number of duplicates since four equivalent sets of
reflections were collected). Ψ-Scan data were used for the
absorption correction. There was no decay. No outlying reflections
were omitted from the refinement. Weights were calculated as
No reflections were specifically omitted from the final processed
data set, but the data were truncated at a 2θ of 50° as the high-
angle spots seemed fuzzier; 3469 reflections were rejected, with
20 space group absence violations, 0 inconsistent equivalents, and
no reflections suppressed. Refinement of F2 was against all
reflections. The weighted R-factor (Rw) and goodness of fit (S) are
based on F2, and conventional R-factors (R) are based on F, with
F set to zero for negative F2. The threshold expression of F2 >
2σ(F2) is used only for calculating R-factors(gt), and is not relevant
to the choice of reflections for refinement.
2
2
1/s2(Fo ); variances (s2(Fo )) were derived from counting statistics
plus an additional term, (0.014I)2; variances of the merged data
were obtained by propagation of error plus another additional term,
(0.014 I )2. No reflections were specifically omitted from the final
The platinum atom was refined anisotropically, and all other non-
hydrogen atoms were refined isotropically. The solvents form layers
1
perpendicular to the c-axis at z ∼ 0, /2, and 1, but do not appear
to interact. All hydrogen atoms were placed at calculated positions
3618 Inorganic Chemistry, Vol. 41, No. 14, 2002

Oxidation of Dimethylplatinum(II) Complexes by Dioxygen
with Uiso’s fixed at 120% of the Uiso’s of the attached atoms. No
restraints or constraints were used. Of the 20 largest peaks in the
final difference map, 6 are greater than |2| e‚Å-3 and another 12
are greater than |1| e‚Å-3. Seven of these 18 peaks are near the
platinum atom, including the largest positive peak of 2.98 e‚Å-3
at a distance of 0.89 Å. The largest negative peak of -2.73 e‚Å-3
is 0.75 Å from N2B (anomalously small Uiso); the other big
excursions not near the platinum are 2.29 e‚Å-3 at 0.88 Å from
C8B, 1.79 e‚Å-3 at 0.67 Å from C6B, -1.66 e‚Å-3 at 1.45 Å from
O2B, 1.63 e‚Å-3 at 0.68 Å from N2B, -1.23 e‚Å-3 at 0.99 Å from
H8BC, -1.21 e‚Å-3 at 1.07 Å from O2A, 1.13 e‚Å-3 at 0.76 Å
from C4B, and 1.06 e‚Å-3 at 0.80 Å from O1B.
The 6 most disagreeable reflections all have k ) 5 with Fcalc
weak and less than Fobs, suggesting twinning. The 20 reflections
discarded in the initial unit cell determination could not be indexed
by themselves. No other effort was made to investigate this fairly
minor problem.
no evidence for a larger cell. There are no appreciable inter-
molecular platinum-hydroperoxy or hydroperoxy-hydroperoxy
interactions. As shown in the hydrogen bond table (Supporting
Information), there is a hydrogen bond between the hydroperoxy
group of one orientation and the methoxy oxygen of the other
orientation. These close intermolecular contacts may explain the
low stability of 2 in the solid.
Acknowledgment. Support by the National Science
Foundation (Grant No. CHE-9807496), Akzo-Nobel, and BP
are gratefully acknowledged. We wish to thank Drs. William
Schaefer and Richard Marsh for help with X-ray structure
determinations and Professor John Eiler for help with mass
spectrometry measurements. We also gratefully acknowledge
Flavio Grynszpan for the cover art design.
Supporting Information Available: Details of the structure
determinations of (tmeda)Pt(OH)(OCH3)(CH3)2 (3) and (tmeda)-
Pt(OOH)(OCH3)(CH3)2 (2), including atomic coordinates, bond
lengths and angles. This material is available free of charge via
the Internet at http://pubs.acs.org. Crystallographic data for com-
pounds 2 (CDC 163314) and 3 (CDC 160706) can be obtained
free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html
(or from the Cambridge Crystallographic Data Centre, 12, Union
Road, Cambridge CB2 1EZ, U.K.; fax, +44 1223 336033; or
deposit@ccdc.cam.ac.uk). Structure factors are available from the
authors via e-mail: xray@caltech.edu.
An analysis of intermolecular contacts suggests two potential
hydrogen bonds, each between a donor hydroperoxy group and an
acceptor methoxy group. One bond is between two molecules of
orientation A translated one unit along the a-axis; the other between
two molecules of orientation B, which are similarly related.
However, the intermolecular distances also reveal some impossibly
short contacts, such as 2.08 Å between O2A (x, y, z) and C5A
(x + 1, y, z). Therefore, two molecules of orientation A cannot be
adjacent along the a-axis and thus molecules of orientation A must
comprise no more than 50% of the total molecules in the crystal.
The population parameter for orientation A refined to 0.508(9), or
essentially 1:1. Thus, orientations A and B must alternate along
the a-axis. This alternation must be occasionally broken as there is
IC0112903
Inorganic Chemistry, Vol. 41, No. 14, 2002 3619